Eli Lilly Outpaces Rivals as Self-Pay Patients Drive GLP-1 Surge

自费患者推动GLP-1药物需求激增，礼来大幅领先竞争对手 礼来通过直接面向消费者（DTC）平台绕开传统医保体系，并将创纪录的25%销售额重新投入自动化和AI驱动的药物研发，从而占据了美国超过70%的GLP-1新起始用药份额。 在这次深度访谈中，礼来（Eli Lilly）首席执行官Dave Ricks探讨了这家制药巨头如何将高达25%的销售额重新投入研发，并利用基于NVIDIA技术的先进超级计算机和人工智能来加速药物发现。对话详细阐述了替尔泊肽（tirzepatide）等GLP-1药物在生理和财务层面的深远影响，解释了它们在治疗更广泛的炎症性疾病和成瘾方面的潜力，同时掀起了一场消费者自费革命。Ricks还指出了美国医疗体系的结构性缺陷，详细剖析了药品福利管理机构（PBM）的反向激励机制、利用专利公开牟利的“影子仿制药”产业的兴起，以及礼来通过LillyDirect向直接面向消费者模式的转型。最后，他概述了支持礼来维持150年企业传承与高市场估值的内部人才流动机制和控制员工总数的运营理念。 礼来将医学研发规模提升至国家级水平，每年投入约140亿美元（占销售额的25%），超过了德国全年的医学研发预算。 (0:00) 该公司已与NVIDIA合作，建造制药领域规模最大的本地部署生物超级计算机，以推动专有药物研发。 (0:54) AI驱动的药物发现目前正遭遇人类生物学知识不足的瓶颈，因为只有10%到15%的人体生物学机制被充分理解到足以准确预测毒性。 (2:16) 由于碎片化的监管要求以及极少废除旧规的“监管棘轮效应”，美国临床试验中每位患者的中位成本已攀升至40,000美元。 (10:11) 传统的“标准包装”药品单位定价模式正受到先进基因疗法的挑战，后者虽然能实现一次性治愈，但在现行医保体系下缺乏费用分摊机制。 (29:51)

Eli Lilly CEO Dave Ricks argues that sustaining biopharmaceutical innovation requires pairing nation-state scale R&D and advanced computing with overhauled clinical trials, direct distribution channels, and stronger intellectual property protections. AI Supercomputing and Biological Knowledge Bottlenecks (2:16) Eli Lilly is committing an estimated $14 billion annually to R&D—roughly 25% of sales—rapidly approaching the US National Institutes of Health's $40 billion budget. To accelerate molecular discovery, Lilly is constructing an on-premise supercomputer powered by NVIDIA B300 chips to model chemical interactions. Ricks notes that AI can generate novel atomic structures to target complex proteins like GLP-1, but cautions that computation is constrained because science currently understands only 10% to 15% of human biology. According to Ricks, AI cannot yet reliably resolve target selection or human toxicity until biological understanding expands significantly past 50%. He contends that closing this gap requires continuous robotic laboratory experimentation to generate massive empirical training datasets, an initiative he argues should be led by public institutions like the NIH. Clinical Trial Economics and Decentralized Recruitment (10:11) Bringing a single drug to market costs an average of $3.5 billion to $4 billion, with over 60% concentrated in Phase III trials that routinely burn over $1 billion annually per program. In the United States, patient enrollment costs have reached a median of $40,000 per participant, and cancer trial participation sits at just 4%, compared to over 25% in Spain and Australia. Ricks attributes this lag to fragmented regulatory oversight, such as requiring separate institutional review board approvals across individual US health systems. Regulatory requirements also accumulate via a one-way "ratchet"; historical safety concerns over drugs like Vioxx and Avandia imposed mandatory pre-approval cardiovascular outcome trials that added four to five years to diabetes drug development. To bypass institutional inertia, Lilly demonstrated the viability of decentralized trials by screening over 80,000 patients for an Alzheimer's prevention study using televisits, local diagnostics, and a single principal investigator. GLP-1 Systemic Benefits and Prevention Economics (24:59) While GLP-1 receptor agonists are widely recognized for reducing caloric intake by roughly 800 calories daily, clinical evidence indicates that they provide systemic anti-inflammatory benefits independent of weight loss. Patients experience sharp decreases in C-reactive protein, a key cardiovascular risk marker, within weeks of treatment. Lilly and its peers are currently evaluating GLP-1s across secondary conditions, including sleep apnea, severe inflammatory skin diseases like hidradenitis suppurativa, and neurodegenerative disorders such as early dementia. Despite historically sluggish insurance coverage for obesity treatments, an evaluation by the Institute for Clinical and Economic Review determined that tirzepatide is cost-effective at current pricing, breaking even on total medical expenses within two years. Ricks attributes payer skepticism to cultural biases that frame weight as a lack of discipline rather than an evolutionary mismatch, noting that modern Americans consume an average of 3,600 calories daily without the ancestral genetic mechanisms required to limit intake. Reforming Distribution and Bypassing Pharmacy Middlemen (101:31) The US commercial drug market is distorted by a widening "gross-to-net" bubble where list prices rise while net manufacturer revenues remain flat. In the case of insulin, brand list prices reached $275 even though Lilly's net realization stayed between $30 and $40, because Pharmacy Benefit Managers (PBMs) demanded larger rebate spreads. When insurers resisted covering Lilly's lower-priced unbranded biosimilar, the company responded by launching LillyDirect, an online platform partnering with telehealth providers to sell therapies directly to consumers. Ricks advocates for federal policy reforms that eliminate rebates and discounts entirely, forcing drugs to clear distribution at a single transparent price. He also proposes that manufacturers establish a firm pricing floor internationally, ending the dynamic where the US market pays 80% to 100% of global drug discovery costs while other developed economies mandate deep price cuts. Capital Compression and Global Intellectual Property Competition (70:43) The broader biotech sector faces severe capital contraction: approximately half of the SPDR S&P Biotech ETF has traded below cash value, and the Inflation Reduction Act curtails pricing flexibility roughly seven years post-launch. At the same time, the transition to a "first-to-file" patent system has enabled competitors to use published patents to engineer slightly altered "clone" molecules that replicate drug function while evading patent infringement. With China's share of the global drug development pipeline expanding from near zero to 30% over the last decade, Ricks warns that US competitive advantage is at risk. To protect innovation incentives, he recommends granting 12 years of guaranteed data exclusivity to any sponsor that funds primary Phase III trials, alongside extending patent publication confidentiality from 12 months to six years. Ricks highlights that while direct-to-consumer models and high-volume metabolic therapies can fund broader R&D, current healthcare financing remains structurally incapable of handling one-time curative treatments like gene therapies. Without subscription-based payment frameworks or intellectual property reforms, capital will continue shifting away from high-risk, curative discoveries toward lower-risk incremental treatments.

礼来公司（Eli Lilly）首席执行官戴夫·里克斯（Dave Ricks）指出，现代制药业正面临研发资金规模化、临床与监管阻力加剧以及传统中介定价体制失灵的深刻转型；礼来正通过百亿美元级研发与AI超算投入、以GLP-1为核心的系统性慢病治疗突破，以及绕过传统医药流通渠道的直接面向消费者（DTC）模式，重新定义跨国药企的商业逻辑与增长边界。 研发规模化与生物学认知的AI瓶颈 (0:00) 礼来当前以每年约140亿美元的规模推进医疗研发，约占其总销售额的25%，这一支出规模已超过德国全国的医疗研发总投入，并逐步逼近美国国家卫生研究院（NIH）400亿美元的体量。为辅助高风险科研决策，里克斯在日常高管会议中实时运行Claude与Grok等AI模型查验文献，并与NVIDIA合作部署搭载B300芯片的本地超级计算机，专门针对非天然化学结构开展分子模拟与药物发现。 然而，里克斯强调AI在制药底层的应用目前受到生物学认知不足的严格制约。业界当前掌握的人体生物学机制仅占整体的10%至15%，远未达到能够系统预测复杂人体毒性所需的50%以上阈值。这使得AI难以根本性解决药物靶点选择与临床前安全性预判问题。里克斯认为，医药行业乃至政府科研机构必须依托全天候自动化机器人实验生成海量生物学数据，才可能突破这一模型训练瓶颈。 临床试验成本危机与监管棘轮效应 (10:11) 单一新药的平均研发成本已攀升至35亿至40亿美元，其中逾60%消耗在III期临床试验阶段。在美国，招募单名受试者的中位成本高达40,000美元；受制于碎片化的机构审查委员会（IRB）分散审批，美国仅有4%的癌症患者参与临床试验，远落后于西班牙和澳大利亚超过25%的参研比例，迫使礼来将绝大部分临床试验转移至海外进行。为降低招募摩擦，礼来在一项阿尔茨海默病预防性试验中试点了去中心化模式，通过远程医疗和本土化检测，仅由全美单一主要研究员（PI）便完成了逾80,000人的筛查与快速入组。 研发周期的延长同样源于历史上的“监管棘轮效应”。以2000年代万络（Vioxx）与文迪雅（Avandia）的心血管安全事件为例，后续流行病学复盘表明心血管风险主要由检出偏倚造成，但监管政策随后强制要求所有糖尿病新药上市前必须通过心血管终点试验排除风险。该要求直接使每款糖尿病新药研发耗时增加4至5年并推高数亿美元成本，客观上也构筑了排挤低成本潜在竞争者的在位者护城河。 GLP-1机制演进与慢性病干预革新 (24:59) 以替尔泊肽（tirzepatide）为代表的GLP-1/GIP受体激动剂彻底颠覆了代谢病治疗逻辑。里克斯指出，人类在漫长演化中仅进化出应对饥饿的本能，缺乏限制摄食的遗传保护机制，导致美国成年人日均热量摄入攀升至3,600卡路里；GLP-1药物通过平抑肠道神经信号，使患者日均进食自然减少约800卡路里。更关键的是，患者在显著减重之前，其C反应蛋白（CRP）等心肌梗死风险指标即出现骤降，显示出独立的抗炎效应，目前该类药物已外延至银屑病、化脓性汗腺炎、早期痴呆和关节炎症的适应症探索中。 在支付经济学方面，经常质疑药价的美国临床与经济审查研究所（ICER）测算显示，替尔泊肽和司美格鲁肽在现有自费或保险定价下可在两年内收回下游医疗总支出，达到高性价比标准。然而受制于商业保险短视机制，美国现阶段仅有约1,000万至1,200万人使用GLP-1药物（包含未经合规检验的复方配制药），相比近1亿的潜在肥胖受众，其市场渗透仍处早期。 药品定价畸变与直销去中介化探索 (41:06) 美国多支付方体系催生了严重的“毛额与净额差价泡沫”。在胰岛素争议中，名义标价一度被推高至275美元，而制药企业获得的实际净价长期维持在30至40美元，巨额中间差价全被药品福利管理机构（PBM）以返点竞价形式抽走，使未参保患者承受了虚高的账面价格。礼来曾推出降价三分之二的未授权生物类似药胰岛素，初期却遭到商业医保抵制，因其较低标价压缩了中介渠道的利润分配空间。 为突破渠道卡死与价格扭曲，礼来成立了直接面向消费者的数字处方平台LillyDirect。借助Zepbound等指征明确、依赖远程问诊即可开方的药物，LillyDirect对接Stripe等线上基础设施，年化营收已达到数十亿美元，成为全美领先的在线处方渠道之一。里克斯主张，美国实际补贴了全球80%至100%的创新药研发利润，唯有在分销全链路推行无折扣、无返点的“单一公允价格”，才能使医药消费回归透明的市场机制。 资本紧缩、地缘竞争与长期治理逻辑 (35:08) 受美国《通胀削减法案》（IRA）限价政策与研发周期膨胀的双重挤压，创新药专利变现窗口实质性压缩，导致生物科技行业出现严重资本撤退，标普生物科技ETF（XBI）成分股中一度有半数跌破账面现金。与此同时，中国在全美及全球新药研发管线中的占比从十年前的近乎为零跃升至30%。里克斯警告，现行“先申请制”使专利在递交12个月后即被公开，易被算法和海外机构用于设计微调分子结构的“克隆药”以规避专利，对此他呼吁针对完成III期试验的创新药推行统一的12年数据排他权，并将专利保密期延长至6年。 为抵御资本周期波动并维系千亿美元级市值支撑，礼来长期贯彻以内部培养为核心的组织治理。面对业务爆发式扩张，礼来对研发人员规模实行严格的人头控制，将充裕资金定向配置于研发薪酬、实验基建、尖端算力及高风险的早期临床。里克斯强调，制药业应践行类似半导体领域的“零价值市场”拓展策略，向阿尔茨海默病无症状干预等此前没有成熟药品的全新疾病领域进军，以技术断代优势化解长周期的专利悬崖压力。 里克斯展示了礼来依靠重注新机制药物、自主算力与颠覆性直销模式所取得的先发优势，但也揭示出制药巨头共同面临的宏观脆弱性：生物学基础研究的转化效率是否真能追赶上算力折旧速度、脱离医保的纯自费慢病模式在更广泛人群中能否长久维持，以及美中两国在知识产权与数据保护规则上的博弈，均将成为决定该创新循环能否持续运转的核心变量。

[0:00] Dave Ricks: at least one or two AIs running every minute of every meeting [0:54] Dave Ricks: the biggest biologically-focused supercomputer there is [2:16] Dave Ricks: we might know 10%-15% of human biology [15:44] If Taylor Swift can sell out a concert in a few seconds, why can't I fill an Alzheimer's study? [287:30] Could you imagine this? Like, we go to a restaurant tonight and someone gives us the wine list and a bottle of like a hundred dollar Napa Cabernet is like $14,000. But then the waiter says, 'Don't worry, that's not your copay.'... Four weeks later, you get a letter in the mail that starts at the top by saying, 'This is not a bill,' but it says the $14,000 and then there's a number of deductions... And then later you get an actual bill. This is healthcare pricing. [287:30] Dave Ricks: none of the numbers mean anything [287:30] Dave Ricks: they're all lies [62:44] Dave Ricks: By the way, I didn't tell you, but I've been working on this thing over here and it seems pretty interesting. [67:04] Dave Ricks: ring the alarm bell right now [106:22] GLP-1 company with a sidecar of some other stuff [111:11] Dave Ricks: there's no limit to human disease [111:11] Dave Ricks: AI begets more AI [116:20] Dave Ricks: They're paying for medicine that's going to help someone else [122:51] Dave Ricks: one less than popes in that period of time [122:51] Dave Ricks: unspoken operating system called culture

[0:00] You spend more on medical R&D than Germany does. [0:02] Yeah, we're at the nation state level. [0:03] It'll be, yeah, $14 billion this year. [0:05] Can you imagine this? [0:07] Like, we go to a restaurant tonight [0:08] and a bottle of, like, a hundred dollar Napa Cabernet [0:10] is like $14,000. [0:12] But then the waiter says, "Don't worry, [0:14] that's not your copay." [0:15] Yeah. [0:16] I have like at least one or two AIs running [0:19] every minute of every meeting I'm in [0:21] and I just am asking it science questions. [0:23] Which one do you use for science? [0:24] Either Claude or the xAI. [0:28] What would you guess the average caloric consumption [0:31] per day in America is? [0:33] 3,600 calories. [0:34] Yeah. [0:35] Isn't that incredible? [0:36] I'm the 11th CEO of the company. [0:37] That's one less than popes in that period of time. [0:43] Where did you learn to pour? [0:45] I actually learned in Ireland. [0:47] Okay, do you want me to leave it down? [0:48] You're the expert here. [0:50] Okay, oscillate this way. [0:52] That's an excellent pint. [0:53] Yeah. [0:54] Dave Ricks is CEO of Eli Lilly, [0:56] which is now a $700 billion company [0:58] and the world's most valuable pharma company. [1:00] Eli Lilly is 150 years old. [1:02] They grew up as the first company [1:03] to mass produce insulin in the 20th century. [1:06] But today most of the company's business [1:08] is in the new GLP-1 diabetes and weight loss drugs [1:11] where they've become the market leader. [1:13] Simultaneously, Eli Lilly is upending the traditional model [1:16] by selling directly to their consumers over the internet [1:18] with LillyDirect [1:19] rather than through the traditional middleman. [1:21] All right, cheers. [1:23] Cheers. Cheers. [1:23] Thanks for coming. [1:26] Good to be here. I'm very impressed [1:27] that you came and you just poured [1:28] your own pint. Poured my own pint, yeah. [1:30] Major flex. Have glass, will pour. [1:31] Exactly. Well, actually a good place to start. [1:33] Tell us about your NVIDIA announcement that you just had. [1:36] Yeah, so today at the, what's it called, [1:38] GTC conference they have, [1:40] they unveiled that we're well underway, [1:42] actually it should be done by the end of the year, [1:44] but building a supercomputer on-prem for us [1:47] really just to run proprietary drug discovery models. [1:51] We think it's the biggest [1:53] biologically-focused supercomputer there is. [1:56] And certainly the biggest pharma's done [1:59] with B300s, the latest chip set. [2:01] And, yeah, we're only constrained by power, [2:04] like everyone else. [2:05] But, yeah, we've built a bunch of tools, [2:07] we'll run them on that, [2:08] and scientists use it to sort of co-invent, co-develop. [2:13] Focus mostly on chemistry to begin with, [2:14] but we'll expand from there. [2:16] And so is the idea here you have some target, [2:19] you've had some challenges actually drugging it, [2:23] and so you give it to one of these new chemistry models [2:26] and you ask it whether it can come up with something [2:29] totally orthogonal beyond what— [2:30] Correct, yeah. A human might have tried. [2:31] So take a really good popular example [2:35] is like GLP-1. [2:37] So that's a hormone peptide that we all excrete. [2:40] It engages targets that are what we call [2:44] G protein-coupled receptors. [2:45] So they're hard to drug targets on the outside of cells. [2:49] And to try to mimic a big huge protein [2:52] with a very small chemical [2:54] is a complicated undertaking. [2:56] And by the way, do only that [2:58] and not other things that are untoward. [3:00] And so this is sort of a frontier of drug discovery [3:03] that's been tough and very empirical. [3:05] That's a hot area for this kind of technology [3:08] because these strange arrangements of atoms [3:11] don't look like other drugs that have come before, [3:14] but they do follow the principles of organic chemistry [3:16] and seem to engage these targets effectively. [3:19] I don't know of one [3:20] that's come through the machine-driven discovery process, [3:23] which is really machine plus human, [3:25] that's made it to the clinic yet, [3:26] but they're coming. [3:27] And I think that's exciting [3:29] because those have been structures that are, [3:31] they don't exist in nature, [3:33] and yet the machines are alien [3:36] and they can predict these interactions. [3:39] I'm always struck by Derek Lowe's arguments [3:43] where he's always sounding this note of caution I guess [3:48] about the optimism [3:50] and maybe what he might view as boosterism [3:53] around AI and biomedicine [3:55] where, as I see at least, his two claims [3:57] are, one, it's really hard to select the targets [4:01] and AI doesn't help you that much there. [4:02] No. [4:03] And then so much fails at, like, human toxicity. [4:07] And again, at least so far, [4:08] AI has not been all that helpful at that step. [4:11] Yes. Do you agree with him [4:13] or is he overrating, you know, these particular challenges [4:16] and maybe underrating the challenges that AI does help solve [4:19] or, you know, thoughts on that argument? [4:20] Probably we need to create the equivalent [4:24] of what got created with human language, [4:26] which is a more complete repository of biological knowledge [4:30] to train against [4:32] before the machines get a lot better. [4:34] And today, I don't know, I would estimate [4:36] we might know 10%-15% of human biology. [4:40] So the machine's not going to be good at all [4:41] until we get way above 50%. [4:45] That probably requires, [4:46] you know, robotic 24/7 experiments [4:49] just to create training data sets [4:51] and, you know, sort of this kind of big lift effort. [4:54] The kind of thing actually NIH should be doing right now, [4:56] I would think. [4:58] But that effort's not ongoing, at least in our country. [5:01] But I think if that gets going, I think we'll know more [5:03] and the machines get better at the harder big problems, [5:06] system prediction. [5:08] Patrick and I did not, [5:10] well, just didn't finish any college. [5:12] So not only don't have a, you know, formal training [5:14] in computer science, [5:15] but don't have formal training in anything. [5:16] You did not come up through the science side of Eli Lilly, [5:21] but you seem extremely comfortable with the science. [5:23] What has been your method for ingesting all this stuff, [5:26] like, and especially as you're essentially [5:27] making science decisions at the end of the day [5:29] with the top-level capital allocation decisions? [5:31] Just how do you learn? No, that's right. [5:34] I think we probably make three or four [5:35] important decisions a year [5:36] and they're all science. [5:38] I don't know. Stay curious. [5:39] Mm-hm. Read. [5:40] Read what? [5:41] I read a lot of medical journals. [5:42] I go to conferences where data is presented. [5:44] I spend time with our scientists. [5:46] Stay curious. [5:47] Yeah. Yeah. [5:48] Now I have like at least one or two AIs running [5:51] every minute of every meeting I'm in [5:53] and I just am asking it science questions. [5:55] So you found for your learning ChatGPT or whatever your— [5:58] I don't use that one for science. [6:00] ChatGPT's too verbal. Wait, so which one [6:02] do you use for science? [6:03] I tend to use either Claude [6:07] or the xAI one. [6:10] I find it more terse [6:12] and the references actually check out more often. [6:14] Sometimes the AIs produce references [6:16] and they're actually not the thing that it said [6:18] and that takes too much work to go cross reference. [6:21] So for an autodidact, presumably the emergence of LLMs [6:24] has been transformative for you. [6:26] Yeah, well, I think for learning, [6:28] that's a whole nother topic we could talk about, [6:29] but you have to sort of question [6:31] the, like, pedagogical kind of method period. [6:35] If you can just learn continuously— [6:37] It's mastery learning for everyone. [6:37] Yeah, yeah, exactly. [6:38] So you take advantage of that. [6:39] But, you know, early in my career, [6:40] I started in our business development M&A group [6:43] and I spent my whole time with scientists [6:44] looking at little companies and projects in other companies [6:47] and trying to understand what they were worth. [6:50] Well, then you have to understand what they do [6:52] and, you know, I found that part of the industry, [6:53] I didn't expect that when I came to Lilly. [6:56] I came to the company accidentally by the way. [6:58] But when I found that, I was like, "Wow." [7:00] I loved it. [7:00] This is so interesting. [7:02] And then I had a moment [7:03] where one of the projects I worked on [7:05] became a medicine in the US [7:07] and my mother was diagnosed with a condition [7:09] and she got put on it. [7:12] And so then that's the magic. [7:13] It's like, okay, you can work on things that change people, [7:16] but the people you care about, that's— [7:18] You saw the full [7:19] end-to-end impact. That's what purpose is. [7:20] Yeah, exactly. [7:21] From lab, exactly. [7:23] So special. [7:23] Your four big decisions a year [7:25] that are kind of grounded in science, [7:26] how quantitative versus qualitative do these end up being? [7:30] Are you Rick Rubin, where it's all taste based [7:32] and you just like the feel of this direction? [7:34] Or are you Billy Beane where it's a "Moneyball" type, [7:37] you know, the ROI pencils and— [7:39] I think the system does a lot of the Billy Beane. [7:42] I think that's a change at Lilly [7:45] that's made us more successful. [7:46] I think we've actually put together a decision process [7:50] that's quite a bit more rigorous than it used to be [7:53] and that leads to fewer bad decisions. [7:55] That's good. [7:56] So that's sort of like the bumpers on the bowling alley [7:59] that you put up. [8:01] But then within that, [8:03] whether it's a strike or a single pin, [8:06] that's a little bit of the judgment and taste. [8:09] And there, though, you know, wisdom of crowds, [8:10] I think we have a great leadership team [8:12] and we all come with equal voice and sort of debate. [8:16] We actually have a rule to like never decide in one meeting. [8:18] So you're asking about the day, [8:19] but we, like, come back to it, [8:21] think about what others said [8:22] and kind of push it again. [8:24] And are you deciding? [8:26] Ultimately yes. [8:27] Nothing happens unless I say go, [8:28] and if I don't like it, then it definitely doesn't go. [8:31] But people will often persuade me [8:33] and I definitely change my mind. [8:35] And some of these are projects within the company. [8:37] So what's the structure of the industry? [8:39] We have huge expenditures on R&D, [8:42] I think more than any other sector, [8:43] as a percent of revenue, [8:44] we'll spend almost 25% of sales this year. [8:47] I like your way— On R&D. [8:48] I like your way of putting it [8:49] that you spend more on medical R&D than Germany does. [8:53] Yes, yeah, we're at the nation state level. [8:55] It'll be, yeah, $14 billion this year. [8:57] Total NIH, [8:58] which is the biggest thing on earth that spends money, [9:01] is 40. [9:02] So that's— It's up there. [9:04] It's getting close, yeah. [9:05] But some of those are projects [9:07] we've been working on it for a while [9:08] and now we have a data set, [9:09] now we need to make a decision to go to the final stage. [9:11] The final stage of testing, [9:13] the average drug costs $3.5-4 billion to make. [9:15] More than 60% of that is the last step. [9:17] So that call is the big one. [9:20] The earlier ones of, like, go forward, [9:22] okay, it's a lot of small things that add up. [9:24] You can waste a lot of money if you do that poorly, [9:26] but, you know, there's a portfolio, so that's unlikely. [9:29] But usually we're carrying five to 10 projects [9:33] in the latest phase and those are— [9:35] And so you're saying like the phase III trial— [9:36] Phase III, exactly. That's the question. [9:38] Yeah, that's the question. [9:39] What to test it in, how to test it, [9:41] what's the design of that, [9:42] go,/no-go, against that criteria. [9:45] And that stage, yeah, is going to be burning [9:49] a billion plus a year. [9:52] So it's a big investment. [9:54] Per program. [9:55] Yeah, and, of course, the returns [9:57] on most drugs that make it through that [9:59] are not positive. [10:01] So it's not just can you get through that, [10:03] but will you produce something useful enough [10:05] to create excess value for society [10:07] but also the company [10:09] to keep the whole thing running? [10:10] That's the exercise. [10:11] Okay, so to this point, the dynamics [10:15] and the funding of clinical trials determine [10:17] so much of the portfolio dynamics for you. [10:21] I think anyone who comes across [10:22] these clinical trial figures and mechanics asks themselves, [10:27] "Why?" How could this be? [10:28] Yes. [10:29] So I looked at— It's a great question. [10:31] I looked at the numbers. [10:32] So apparently the median clinical trial enrollee, [10:37] it now costs $40,000. [10:38] You know, the median US wage is $60,000. [10:42] So we're talking two thirds. [10:43] Why and why couldn't it be a 10th or a hundredth [10:48] of what it is? [10:50] Yeah, brilliant question, [10:51] and one we've spent a lot of time working on. [10:54] We've done a lot of things [10:55] to improve the drug development process. [10:56] So taking a systems approach. [10:58] And I think one of the reasons [10:59] Lilly has probably the highest return on investment in R&D [11:03] in the industry [11:04] is because not the picking of winners and losers, [11:08] but actually the process by which we run it. [11:11] I think that's at least as valuable as what we've done. [11:14] And we can come back to that if you want. [11:15] But the piece we've really not moved [11:17] is the enrollment of clinical trials, [11:21] this is going to sound super arcane when I go through it, [11:23] and the cost, which is escalating about 7%-8% [11:28] over the last decade. [11:29] That's about the same as the healthcare system. [11:31] And that's not an accident. [11:32] When people go, "Why does a trial cost so much?" [11:35] Well, we're taking the sickest slice [11:38] of the healthcare system that are costing the most [11:41] and we're ingesting them, [11:43] we're taking them out of the healthcare system [11:44] and putting them in a clinical trial. [11:46] Typically we pay for all care. [11:47] So we are literally running the healthcare system [11:49] for those individuals. [11:52] And that is, in some ways, [11:55] for control, [11:57] because you want to have the best standard of care, [11:58] so your experiment is properly conducted [12:01] and it's not just left to the whims [12:02] of hundreds of individual doctors [12:04] and people in Ireland versus the US [12:06] getting different background therapies. [12:08] So you standardize that, that costs money [12:11] because you're sort of leveling up a lot of things. [12:13] But then also, in some ways, [12:17] you're paying a premium [12:19] to both get the treating physicians [12:22] and having a great care to get the patient. [12:26] We don't offer them remuneration, [12:27] but they get great care, [12:28] an inducement to be in the study [12:31] because you're subjecting yourself quite often, [12:33] not all the case, [12:34] but to something other than the standard of care, [12:37] either placebo or this, [12:39] or in more specialized care, [12:40] often it's standard care plus X [12:43] where X could actually be doing harm, not good. [12:45] So people have to go into that in a blinded way. [12:48] And I guess the consideration is you'll get the best care. [12:51] Of the $40,000, how much of that should I look at [12:54] as inducement and encouragement for the patient [12:57] and how much should I look at it [12:58] as the cost of doing things [13:00] given the regulatory apparatus that exists? [13:02] The patient part is like the level up part [13:04] and I would say 20%-30% of the cost of studies [13:08] typically would be this. [13:09] So you're buying the best standard of care. [13:12] You're not getting something less. [13:14] That's medicine costs, [13:15] you're getting more testing, you're getting more visits. [13:17] And then there is a premium that goes to institutions, [13:21] not usually to the physician, the institution, [13:23] to pay for the time of everybody involved in it [13:26] plus something. [13:27] We read a lot about it in the NIH cuts, [13:29] the 60% Harvard markup or whatever. [13:33] There's something like that in all clinical trials too. [13:35] Overhead, coverage, whatnot. [13:37] But it's paying for things that aren't in the trial. [13:40] US healthcare is famously the most expensive in the world. [13:42] Yes. Do you run trials [13:43] outside the US? [13:44] Yeah, actually most. [13:46] Most of it, yeah. So, yeah. [13:47] I mean, we want to actually do more in the US. [13:50] This is a problem, I think, for our country. [13:52] Like, take cancer care where you think, [13:54] "Okay, what's the one thing [13:56] the US system is really good at?" [13:57] Like if I had cancer, I'd come to the US. [13:59] That's definitely true. [14:01] But only 4% of patients who have cancer in the US [14:03] are in clinical trials, [14:05] whereas in Spain and Australia, it's over 25%. [14:08] And some of that is because they've optimized the system [14:12] so it's easier to run and then enroll, [14:14] which I'd like to get to, people in the trials. [14:17] But some of it is also that the background of care [14:20] isn't as good. [14:21] So that level up inducement is better for the patient [14:25] and the physician. [14:25] Here, the standard's pretty good. [14:27] So people are like, "Meh, do I want to do something [14:29] where there's extra visits and travel time?" [14:32] There's another problem in the US, [14:33] which is we have really good standards of care [14:36] but also quite different performing systems [14:38] and we often want to place our trials [14:40] in the best performing systems that are famous, [14:43] like MD Anderson or the Brigham, [14:45] and those are the most congested with trials [14:48] and therefore they're the slowest and most expensive. [14:50] So there's a bit of a competition for place [14:53] that goes on as well. [14:54] But, overall, I would say, [14:56] like, in our diabetes and cardiovascular trials, [14:58] many, many more patients are in our trials [15:01] outside the US than in [15:03] and that really shouldn't be other than cost of the system, [15:07] and to some degree, the tuning of the system, [15:09] like I mentioned with Spain and Australia, [15:11] toward doing more clinical trials. [15:12] For instance, like, here in the US, [15:15] everywhere, you get ethics clearance, we call IRB. [15:18] The US has a decentralized system, [15:20] so you have to go to every system you're doing a study in. [15:23] Some countries, like Australia, have a single system. [15:26] So you just have one stop [15:27] and then the whole country is available to recruit, [15:30] those types of things. [15:31] You said you want to talk about enrollment? [15:32] Yeah, yeah, it's fascinating. [15:34] So drug development time in the industry [15:36] is about 10 years in the clinic, [15:38] a little less right now. [15:39] We're running a little less than seven at Lilly. [15:41] So that's the optimization I spoke about. [15:44] But actually, half of that seven is we have a protocol open. [15:48] That means it's an experiment we want to run. [15:51] We have sites trained, [15:52] they're waiting for patients to walk in their door [15:54] and to propose, "Would you like to be in the study," [15:56] but we don't have enough people in the study. [15:58] So you're in the serial process, diffuse serial process, [16:02] waiting for people to show up [16:03] and you think, "Wow, that seems like [16:05] we could do better than that. [16:07] If Taylor Swift can sell out a concert in a few seconds, [16:09] why can't I fill an Alzheimer's study? [16:11] There seem to be lots of patients." [16:13] But that's healthcare. [16:15] It's very tough. [16:17] We've done some interesting things recently [16:19] to work around that. [16:20] One thing that's an idea that partially works now [16:23] is culling existing databases [16:26] and contacting patients. [16:27] Proactive outreach. Right, [16:28] where you have, like, their lab values from, [16:31] where before there wasn't a treatment, [16:32] now there is one being studied, [16:33] would you like to be a part of it? [16:34] That's something we're doing now [16:35] with our Lp program. [16:37] That's a cholesterol subtype [16:39] where there was nothing to do about it. [16:40] A lot of people have had it tested and it's high. [16:43] You could say, "Hey, you're high. [16:44] Would you like to do something now?" [16:45] But it's still a lot to be done there [16:47] and the data that's sitting in electronic health records [16:50] in our country [16:51] is very poorly organized. [16:52] So it would be good to optimize that. [16:55] I think the other [16:56] is actually just go directly to the patients. [16:58] So who has the most interest? [16:59] It's usually the patient. [17:00] And then physicians and their institution [17:02] may not be in the trial [17:03] or they might not be interested [17:04] in spending much time on this and people— [17:06] That's kind of what I want. [17:07] I want to get an email, as you say, [17:11] you know, the system knows, you know, my health data [17:14] and what conditions I have and so forth, [17:16] and be told that a package will be arriving tomorrow [17:20] with a drug. [17:21] I can take the drug if I want to, [17:22] if I want to participate in this trial. [17:23] You can include whatever disclosures. [17:26] And, you know, some nice person [17:27] will come every month or whatever and take— [17:29] Or just telephonically visit you. [17:30] Yeah. Take my vitals, exactly, [17:32] and measure my blood pressure and what have you. [17:34] Are all these intermediaries in the systems [17:35] in, you know, the hospitals and so forth, [17:38] are they required intrinsically [17:40] for the kinds of trials you want to run? [17:41] Obviously varies a little bit on the condition or the drug. [17:43] Depends on the disease. [17:44] Yeah, yeah. Right, right. [17:45] Well, I guess, yeah, so how much of it [17:47] is there in fact intrinsically required [17:48] given, you know, the characteristics of the condition [17:50] and how much is it this is how things are done? [17:54] Yeah, what you described is actually a great vision [17:56] for where we want to go. [17:57] We've executed one of these at scale, [18:00] which is fully enrolled, [18:00] which was our Alzheimer's prevention study. [18:03] It's a more complicated medicine. [18:04] It's an infused medicine. [18:06] But we ran this with one investigator in the United States [18:09] and we screened over 80,000 people. [18:12] By the way, it's the fastest accrued [18:14] Alzheimer's study in history, [18:15] even though it's pre-Alzheimer's. [18:17] It's people with the amyloid precursor protein, [18:20] but not dementia. [18:21] It's fully enrolled. [18:23] We've treated people, [18:23] actually no one's left on treatment. [18:25] We're just watching them now, [18:26] because the treatment's a nine-month course [18:27] to deplete amyloid [18:28] and see if that can prevent the symptoms. [18:31] So that was a very successful trial. [18:33] Just what you said. [18:33] They got instructions to be in the study. [18:36] There was a televisit. [18:37] They got some diagnostic tests, blood-based, [18:40] that went in and said, [18:41] "Okay, this is sort of an indicator [18:44] you might have high amyloid," [18:44] then you can go to get a PET scan, and if that was positive, [18:47] you could be enrolled in the study. [18:49] Pretty successful. [18:50] So we'd like to replicate that. [18:51] I think one very interesting thing in the future of medicine [18:54] is that I think we will have a lot more [18:57] preventative medicines in the future. [18:59] And I think this type of study in particular [19:01] is well suited to prevention [19:03] because you have sort of the people [19:06] who are worried about their wellness, [19:07] so they're motivated. [19:08] They have means. [19:09] They're in the middle of their life. [19:10] They're working. [19:12] They don't have complications of comorbidities and so forth. [19:14] They want to be in the study, [19:16] and I think they would like to prevent [19:17] terrible conditions like Alzheimer's. [19:19] So that's an exciting new chapter we can push. [19:23] So you know Paul Janssen? [19:24] Yeah. Yeah, yeah. [19:26] So- [19:26] Explain who Paul Janssen is. Come on. [19:28] Paul Janssen, as I understand it, [19:29] was behind the discovery, invention, what have you, [19:32] of more medications than any other single— [19:35] I think a Belgian guy. [19:36] Yeah, exactly, yes, yes. [19:37] I think it's— Who invented a number of— [19:39] 79 or 80 approved— The MVP of medicine. [19:42] Michael Jordan of the NIH. Beyond MVP. [19:44] Yes. [19:45] Okay, so amazing guy. [19:47] When some outsider comes [19:49] to the clinical trial process and system [19:52] and just the development pipeline overall, [19:54] maybe they naively think, "Wow, this seems so torturous, [19:58] so expensive, so bureaucratic," what have you. [20:00] But that's how it's gotta be. [20:01] But that's how it's gotta be. [20:02] You know, or if they think it can be otherwise, [20:06] you might think that they're naive, right? [20:07] There's a video interview with Janssen from, [20:10] I think it's from the '90s, [20:11] it's from, you know, quite a while ago, [20:12] I mean, he's dead now, [20:13] where he's recounting the history of his career. [20:17] He started the company in 1953. [21:27] We could go go back to doing it the way we used to [21:29] and it's kind of a, [21:30] it's a societal choice to make it so bureaucratic. [21:33] I guess it's an explicit and implicit one. [21:37] The explicit part is through time, there have been accidents [21:41] and nothing is perfect. [21:42] We probably have 2,000 manmade approved medicines [21:47] versus natural products or vitamins or other things [21:49] and maybe 400 unique mechanisms. [21:53] So there's clustering. [21:56] Within those, there have been problems [21:58] and there's also been problems [21:59] that turned out not to be problems. [22:01] And so our detection ability is flawed. [22:03] Because of that, I think each time that occurs, [22:07] there was intervention in the system, [22:09] which is sort of a global consensus, [22:11] but mostly the developed economies [22:13] kind of harmonize their systems [22:14] either directly or indirectly [22:16] to say, "Oh, no, let's require more information [22:19] or rebalance the risk-benefit." [22:21] We've had this ratchet. [22:22] Have we gone too far? [22:23] I think that it's a function [22:25] of what the technology is at the moment. [22:27] And I think in past times, yes. [22:30] You can take the 2000s in the US [22:33] where there were two big controversial drug approvals [22:37] that were later retracted, [22:39] the Vioxx situation with Merck [22:42] and then Avastin from GSK. [22:44] These were both drugs that were for different uses, [22:47] pain and diabetes, [22:48] but through a detection requirement that the agencies, [22:51] because now we have electronic records, [22:53] we can look at things, [22:55] picked up what they thought was a trace of risk for both, [22:58] cardiovascular risks, [23:00] and intervened with labeling and escalation [23:02] until finally both companies actually removed the products [23:05] from the market, [23:06] withstood billions of dollars in product liability suits, [23:09] only to find later under a different analysis [23:12] that there was nothing to be seen there. [23:14] Both of them. [23:15] And I think there's like an ascertainment bias problem [23:19] with these studies. [23:20] There's also who was looking at this data. [23:23] But that caused a 10-year chill in drug development. [23:26] And the Avandia one we know well, [23:29] we work in diabetes, [23:30] actually caused a policy change. [23:31] And the policy change [23:32] was you must rule out cardiovascular risk [23:35] prior to market entry. [23:37] And as you may know, some conditions like diabetes [23:39] have a more continuous variable you're measuring [23:42] and so studies can be short and cheaper, glucose levels. [23:46] Other studies like cardiovascular event studies [23:48] are not a continuous variable, [23:49] it's a binary variable, [23:51] and you have to wait for natural history to occur [23:54] to pile up enough variables [23:55] to have a statistical difference. [23:57] Those are four to five-year undertakings. [24:00] So there, you just bought four or five years of extra time [24:03] before you could get any new diabetes medication. [24:05] We got better at doing them, but that was expensive. [24:08] Now, has that... [24:09] That's the explicit. [24:10] The implicit is the regulatory problem. [24:13] There must be a name for this problem [24:14] some smart person's given it, [24:15] but regulatories are added but never taken away. [24:18] So the regulation is still there. [24:20] Now, by happy accident, [24:22] we are all now really pleased [24:24] with, like, incretins like our tirzepatide to run them [24:26] because they frequently demonstrate [24:28] massive benefit on cardiovascular [24:30] and, in some ways, it creates a barrier to entry [24:33] for the next low-cost Chinese program or whatever [24:36] 'cause it's this big expensive thing you have to— [24:38] As with many regulatory— [24:39] Yeah, exactly. Right. [24:41] So is it right? [24:42] No, we're imperfect as people [24:45] and certainly as decision-makers at a collective level. [24:47] I would also say the technology for seeing early signals [24:52] has changed and improved, including computer technology, [24:57] and it's probably worth a reassessment. [24:59] Paying for prevention, you were going to ask. [25:01] Yeah, let's talk about that [25:02] because say with, you know, GLPs in the weight loss context, [25:08] they economically pay off over a very long time horizon, [25:13] but if you're looking at a short time horizon [25:14] of an insurer or an employer, they don't necessarily, [25:17] and so that's created this challenge for reimbursement [25:19] where, you know, not as many people [25:21] reimburse GLPs for weight loss [25:23] as you think would be rational. [25:24] Yep. [25:26] That just will always be the case with prevention. [25:28] And so how do you actually develop drugs [25:31] that are commercializable and reimbursable? [25:34] Yeah, well, in the obesity case, [25:36] I'll take a little bit issue with your first assertion [25:38] and then add two other problems. [25:40] The data actually is becoming more clear [25:43] that within actually a two-year timeframe, [25:44] and I hope at Stripe, you reimburse these medicines [25:46] for your patients, or for your employees. [25:48] Within two years, [25:49] you can break even basically on total medical costs. [25:52] So there's this group called ICER, [25:53] which is funded by someone who hates our industry [25:55] and the insurance companies, [25:56] and they analyze all new drugs [25:58] and usually seeking to prove that they're not worth it. [26:01] That's sort of their mission in life. [26:02] They just analyzed our medicine, [26:04] tirzepatide and semaglutide, [26:06] and they said, "Actually, they're both cost effective [26:08] at current pricing." [26:09] In fact, Zepbound, or tirzepatide, was, [26:12] the threshold they have [26:13] is to save a hundred thousand dollars per person per year [26:16] in downstream health costs, [26:18] and it was twice as effective as that [26:20] at the current pricing. [26:21] And the current pricing isn't going to stay, [26:22] let's be honest. [26:23] There'll be more competition. [26:25] The government wants to lower our prices. [26:27] So, you know, I think we're in a good place there. [26:29] Now, the two other problems [26:31] are there's sort of this incumbency problem in healthcare, [26:35] like many things, but particularly in healthcare, [26:37] where the last thing in is scrutinized the most [26:41] and the base stack of services and products we use [26:45] is never revisited. [26:46] It becomes standard of care. [26:48] But displacing that in most therapeutic spaces [26:51] and in the healthcare system in general [26:52] is extremely difficult. [26:53] I think we suffer from that here. [26:55] If the first medicine we had to treat metabolic conditions [26:59] was tirzepatide in 1972, [27:01] I have no doubt it would be reimbursed everywhere [27:04] and broadly used in the system. [27:06] But they get the ratcheting effects. [27:07] But then you're just stacking on top of it [27:09] and it's difficult to remove benefits, [27:11] it's easy to deny new ones, [27:13] and that's true in government-funded systems [27:15] but also, you know, big insurers. [27:17] I think the other thing that's going on with this one [27:20] and why we're spending so much energy [27:22] exploring, you know, real indications [27:25] for comorbid diseases that go with obesity, [27:27] which is so far pretty successful, [27:29] is that the idea of just treating someone [27:32] who's overweight or obese without any other illness, [27:36] to many people I think exposes a bias we have [27:39] about that particular condition. [27:41] That if it wasn't something you could see, [27:43] you might not have. [27:44] But I think we are conditioned [27:46] to think of someone who's overweight [27:48] as someone who's not disciplined. [27:50] The data does not show that actually. [27:52] Like our ancestors roaming the plains of whatever, [27:56] the tundra of Ireland, [27:58] walking across the ice bridge from Norway, [28:01] were in a background of starvation [28:04] and there are very few humans on earth [28:06] that have a genetic background [28:08] that has any limit on food consumption. [28:11] It's irrational. [28:12] It's a wasted piece of code. [28:13] It did no good. [28:15] Now, today, in today's environment, [28:16] we're in the flip, the complete flip, [28:18] especially here in the US, [28:21] where there's food everywhere we walk. [28:23] I came across your stat, [28:24] what would you guess the average caloric consumption [28:27] per day in America is? [28:28] 3,600 calories. [28:30] Yeah. [28:31] Yeah. Isn't that incredible? [28:32] Yeah, that's a really incredible. [28:33] And here's an interesting stat. [28:34] When you're on our medicine, [28:36] how many fewer calories do you consume on average? [28:39] On one of the GLP-1s? Yeah. [28:41] You don't need to swing it that much [28:43] to cause meaningful— [28:44] 800 calories a day. [28:45] Oh my God. 800? [28:45] Which is almost a meal. [28:47] Yeah. If you go pull up [28:48] to In-N-Out Burger— [28:49] That's second breakfast right there. [28:50] It's second breakfast. Exactly. [28:52] So that's why people lose weight so successfully. [28:54] No wonder all the food companies are so worried. [28:55] And the trick, yeah, yeah. [28:56] And the trick is people lose the weight [28:58] and they don't feel miserable, right? [29:00] So here's the thing about being obese is people, [29:03] when you start to gain a little bit of weight, [29:05] your set point sort of readjusts. [29:07] This is the missing code we have. [29:09] And there's only one direction, which is up is better. [29:13] And the more up you have [29:14] actually the more hunger it creates. [29:16] Hyperinsulinemia, which is a hunger-stimulating hormone. [29:19] And it sort of starts to overwhelm [29:21] the counterregulatory system, [29:22] which is incretins, GLP-1, GIP, [29:25] the ones we are making medicines around, [29:28] and you're out of balance and there's no going back. [29:31] And interestingly, even when people lose weight, [29:34] that balance still seems to be off, [29:37] which is why if you've ever gone on a crash diet, [29:40] you feel like shit constantly. [29:42] You want to hurt people. [29:43] You're angry. [29:45] And on these medicines, that doesn't happen. [29:47] Yes. [29:48] Which is the miracle. [29:50] People feel good and lose weight. [29:51] If you have a medicine that is recurring [29:54] and it, you know, presents some income stream [29:57] for Eli Lilly, [29:58] now, maybe nothing is truly recurring [30:00] in the sense that, you know, all patent protection ends, [30:03] but nonetheless, there's something on an ongoing basis [30:07] and I guess there are various ways to extend that, [30:09] then some genetic medicine comes along. [30:10] It's one time. [30:11] Yep, one and done. Exactly. [30:13] Is it in practice possible to charge enough upfront [30:18] such that as a company looking at its portfolio— [30:21] Pays back the R&D? [30:22] Yeah, you are in fact neutral as to which it is? [30:25] Because from first principles, for the patient, [30:27] it's way better to do that— I think you asked [30:28] about a value perception problem, [30:30] and I think we need to overcome that. [30:32] We're doing that by studying [30:34] and all these other conditions [30:34] people recognize as conditions [30:36] and then we'll insure. [30:38] And because obesity is sort of this master switch [30:40] to all these things, [30:41] that's an achievable thing. [30:41] It just costs a lot of R&D. [30:42] I'm talking about perception. [30:43] I think he's talking about reality. [30:44] Yeah, you're talking about actually pricing, [30:46] which is, why is it that the industry's evolved [30:49] to have a unit pricing model? [30:51] It's back to like a shrinkwrap software world, right? [30:54] Where you're basically just shipping a box [30:56] and all your value has to be captured [30:58] upon that invoice. [30:59] Right. Yes. [31:00] That is how we price all medicines. [31:01] You're currently, you know, in the SaaS model [31:03] and, you know, tech people know that SaaS is way better [31:05] than the shrinkwrap software business model. [31:07] Yes. And genetic medicines [31:09] are shrinkwrapped software. Exactly. [31:11] Exactly. You know, [31:11] wouldn't you be crazy to go back [31:13] even though it's better for the patient? [31:14] Yeah, so I think, [31:16] so we have some genetic medicines coming [31:18] and we're thinking actively about this. [31:19] For instance, we have a medicine development [31:21] that will knock down your LDL if it's safe enough [31:24] in a one and done PCSK9 edit in your liver [31:28] and presumably that will last the rest of your life [31:30] and your LDLs will be between 20 and 40 forever. [31:33] It looks like an amazing drug. [31:34] Yeah. [31:35] Of course, there's problems with these delivery systems, [31:37] we have to rule out safety, but let's just say it works. [31:39] How would one price that? [31:40] Because you're displacing a medicine [31:43] that costs, I don't know, $8,000-$9,000 per year. [31:45] Right. [31:46] We need to innovate that pricing model. [31:48] Why haven't we? [31:49] It's mostly because the consumption side [31:51] has no capability to do this. [31:54] Particularly governments have built all, [31:56] back to the regulatory incumbency problem, [31:59] built all this stack of rules [32:02] around the idea that I buy one unit, I pay X. [32:05] Whereas here you buy one unit [32:07] and we want money over time. [32:09] What is that? [32:10] But it's conceivable that one could create [32:12] like a licensing concept, [32:13] stealing from the SaaS model, [32:15] where you say, "We'll do the procedure for free, [32:17] and as long as it's working for you, [32:20] you will deposit X amount in our bank account [32:23] and you're getting the value [32:25] and we're getting paid for our research. [32:26] If it doesn't work," [32:28] you know, so that invokes a warranty as well. [32:30] That's an interesting idea [32:31] and one we're thinking about for these more common, [32:34] because so far gene therapy is mostly for uncommon things [32:38] where they've just charged it and someone's paid, [32:40] but for common gene therapy to really be unlocked, [32:42] this has to be solved. [32:45] It strikes me that we're describing, [32:47] you know, we're discussing how people's lives are affected [32:51] by all these treatments [32:52] and what pharma companies can produce, [32:55] which are themselves downstream [32:56] of what pharma companies can afford to invest in, [32:59] which there but for the grace of God go we, [33:03] you know, the patent time horizon is an arbitrary number, [33:06] that, you know, we have ended up with... [33:09] I sometimes think about, you know, off-label use [33:11] is, you know, very valuable in the US system. [33:14] You can imagine another universe where we hadn't ended up [33:17] with off-label use being permissible and things like this. [33:20] Do you think we need to spend more time [33:22] trying to discuss and meta edit [33:27] the R&D system and incentive system that we have [33:31] because it just has such a huge effect [33:34] on people's quality life? [33:35] Well, yeah, thanks for the question. [33:37] I love to talk about this. [33:39] So I think a lot about this [33:40] and I think if your point of view [33:43] is that we want more new medicines, [33:45] like, that would be [33:46] a better outcome— I'm from that world. [33:47] Yeah, yeah, and I am too. [33:49] Then I think there's definitely many flaws [33:51] with the current system. [33:52] Strangely, most of the discussions I have about this [33:56] are actually we have, [33:59] they don't say it out loud, but there's enough new medicine, [34:02] and where we really have a problem is affording it. [34:05] Now, interesting fact, in the US, [34:07] the most expensive healthcare system in the world, [34:08] we spend 10 cents on the dollar on medicine. [34:10] The other 90 cents go to everything else [34:12] that medicine's trying to prevent. [34:14] Yep. [34:15] Go back to 1965— [34:16] I think it's less than 10 cents. [34:18] Branded medicines are eight. [34:19] 2% is generics, which is 90% of the volume. [34:22] That's an even better deal. [34:23] We have to get to generics. Yeah, yeah. [34:24] But go back to 1965, [34:26] Medicare and Medicaid were invented. [34:28] We've gained, I believe, eight life years [34:30] of life expectancy since then, [34:32] and most studies would say five or six of those [34:34] are due to medicine. [34:35] Yet think of the cumulative expenditures [34:38] by taxpayers since that time. [34:39] It's not even close. [34:41] We should be saying, "Who can we give money to [34:43] to do more research? [34:45] Because this is clearly a better way to get through life." [34:48] The direct way is the NIH. [34:50] We could talk about that if you want. [34:52] That has limitations because institutional government. [34:54] But the private market self funds, [34:56] either through capital markets or through our R&D line, [34:59] and there would be a lot more funding [35:01] if we had an idea of price stability [35:03] or a longer return period. [35:06] That is definitely true. [35:08] The patent system is what it is because of former rules. [35:13] Moving it out in time seems exceedingly difficult [35:16] in this climate, [35:17] but actually— Despite the fact, [35:18] of course, it's shrunk [35:19] because of the longer approval timelines. [35:21] It's de facto shrunk. [35:23] And then, actually, the Biden administration passed a rule [35:26] in the Inflation Reduction Act [35:28] to actually have government price intervention in the US [35:31] at five years plus two. [35:33] So basically around seven years, [35:36] you lose that ability to recoup investment [35:38] in the same way. [35:39] Government price intervention always works out well. [35:41] Yeah, right. [35:42] It doesn't produce surpluses, let's put it that way. [35:45] You don't get more medicines that way. [35:47] So it's actually, it is collapsing, [35:49] I think, in investors' mind. [35:50] And you can see that in the capital markets. [35:52] If you look at the large cap pharmas, [35:55] not Lilly, but the other ones, [35:57] the multiple is the most compressed it's been in 20 years. [35:59] If you look at biotech, the XBI, [36:02] I think half of the XBI is trading below cash. [36:05] Yeah. [36:06] And then if you look at— Because of this. [36:08] Venture, half of the rounds last year were down rounds. [36:11] This is not a positive environment. [36:12] Would extending the patent duration actually work? [36:15] Because you referenced earlier this dynamic [36:17] where, you know, there are, [36:18] especially with biologics, [36:19] there are now so many opportunities [36:22] for copycat molecules and therapeutics and so forth. [36:25] And so, like, does it matter less [36:28] what happens with patent windows [36:29] because what actually matters [36:31] is the competitive ecosystem, the ability for fast follow? [36:34] Yeah, I think we end up [36:36] with two competitive ecosystems. [36:37] You have the on-patent one, [36:39] and here I think history would show [36:41] actually within a 10-year period, [36:43] which is typical recruitment time. [36:44] We've solved clinical trials, you know, separately. [36:46] So they're now— Okay, so it could be, [36:47] that's a way to get there, right? [36:48] Is we could simplify the regulatory framework [36:50] and have longer return periods [36:52] and increase returns to investors [36:53] and get more investment. [36:54] That's actually a real idea. [36:56] But, typically, in classes [36:58] by sort of the horse race and accidents along the way, [37:02] it's pretty uncommon you end up with one medicine. [37:04] Often you'll get many. [37:06] We can talk about GLP-1s for a minute. [37:07] We only have two right now, [37:08] but there's probably 80 [37:10] in clinical pipelines right now globally. [37:12] We have 11 others, [37:13] but there's probably 70 others not coming from Lilly. [37:17] So there will be tons of competition. [37:19] But the history shows that, [37:21] back to this medicine incumbency, [37:23] once two or three sort of get in the works of things, [37:26] unless you're kind of different, [37:28] nobody really uses it [37:30] and pricing strategies have not worked. [37:33] Now, they don't work [37:34] until there's actually a biosimilar or generic event [37:37] because here, it's not a, [37:39] "Hey, I'm a hundred dollars and you're 90." [37:42] Typically, a generic event, [37:43] you'll lose 97% of your pricing the day your patent expires. [37:48] So this is a fantastic deal for society, [37:51] but a terrible situation for an inventor. [37:54] And if you came along late hoping to induce competition, [37:57] you maybe even were half off the originator. [38:00] Now you're half off is 90% lower. [38:05] So there's no return on that at all. [38:06] Okay, so in order to stimulate [38:09] and to, you know, catalyze more R&D, [38:12] you know, one thing we could do [38:13] is we could extend the 20-year window. [38:16] What else can we do? [38:17] Either you can get it quicker to market [38:18] or extend the market. [38:19] I think pricing for value is a good idea to consider. [38:23] So, today, particularly in the United States [38:26] and in many ex-US markets, [38:28] I would point out a few of the, [38:29] you know, the Commonwealth markets are different. [38:31] They've tried to implement at a price for value scheme. [38:33] Beause they're single payer? [38:34] Well, most are single payer outside the US, [38:36] but because they chose that path [38:38] instead of a negotiated outcome or something. [38:40] But the US, we have a multi-payer model, [38:43] but it's devolved to the situation [38:44] where actually it's a very commercial kind of thing [38:48] where there's a price point a manufacturer launches at, [38:51] really nobody pays that price. [38:53] There are then many, many price points below that. [38:57] The lowest is defined by law. [38:58] It's Medicaid. [38:59] Actually the law is called Medicaid Best Price. [39:01] So state Medicaid, [39:02] they spend 5% of their dollar, [39:04] five cents a nickel on medicine, not 10 cents, [39:07] because they get lower pricing per unit. [39:09] Big insurers like UnitedHealthcare, et cetera, [39:11] get a very good deal as well, approximating the government. [39:15] And then smaller insurers and smaller employers [39:17] get a worse and worse deal. [39:18] That's the way we do it. [39:21] What it means is that manufacturers compete [39:24] mostly not on value, [39:26] but on the pricing offering, [39:28] on sort of the difference between the spread [39:31] between the list price and whatever that person got. [39:34] Making it even worse, [39:35] a number of intermediaries in that system that bulk buy [39:39] take their returns on the percent off list. [39:44] So the higher the list, the better they do. [39:46] And I think that's a terrible incentive. [39:47] These PBMs [39:48] and there's some group purchasing organizations like this. [39:52] That should go away. [39:53] And I think health is different than other commodities. [39:56] It's probably has a much more important social role [40:01] and deciding that the little guy gets the worst deal [40:05] and the big guy gets the best deal [40:07] to me feels unethical. [40:09] So I would be for a system [40:11] that there is one price point. [40:15] People can say yes or no to that. [40:17] That's one way to have value. [40:19] As an employer, you could say, "That's not worth it. [40:21] We're not going to pay for that. [40:22] And this one is worth it." [40:23] That could be informed by really independent intermediaries [40:26] who study these things, [40:27] look at all the claims records, [40:29] look at how people do on the medicine, [40:30] weigh the risks and benefits, [40:31] and produce pricing. [40:33] That happens in lots of other markets. [40:36] Bond pricing, like, lots of people do this for a living, [40:38] just not in medicine. [40:39] And I think that could be a useful tool in the US system [40:43] so that if you produced a truly surprising [40:45] and positive clinical trial result, [40:47] you could actually charge more, [40:49] and that would induce other people to say, [40:50] "Oh, let me go for higher risk, more valuable indications [40:55] instead of just do the base that gets you in the door, [40:58] now negotiate with a commercial team to drive more return. [41:01] Oh, the patent clock's running out. [41:02] Let's go to the next medicine." [41:03] I think that's not a great system right now. [41:06] There's the much discussed, [41:08] like, maybe the top discussed topic in pharma [41:11] that people know about generally [41:13] is pharma pricing [41:14] and the disparity between the US and internationally [41:16] where, you know, all the cost is in R&D, [41:19] the cost of actually producing the drugs is fairly low. [41:22] And so single-payer healthcare systems internationally [41:26] pay very low prices [41:27] and so the R&D cost is borne by the US. [41:29] And the biggest problem is not only, like, at the margin, [41:33] maybe you have fewer drugs developed [41:36] because this phenomenon [41:38] because, you know, you have fewer returns. [41:40] I think honestly the biggest problem [41:42] is the social issues it creates in the US [41:47] where it turns people against pharma [41:49] and, you know, the insulin price disparities [41:51] between the US and Canada and things like that. [41:54] Which no longer exist by the way [41:55] because we fixed that, but yes. [41:56] But that was like the hot topic for— [41:58] Yeah, it was a hot topic. [41:59] For such a long time. And it needed to be fixed. [42:00] And that's a classic example [42:01] of this commercial environment I spoke about. [42:03] I mean, our actual net on insulin really hasn't changed. [42:06] It's like $30 or $40, [42:08] but the list price got up to $275. [42:09] Why? We were competing on the spread. [42:12] And so that just drove this huge, [42:13] and so the individual— When you say the spread, [42:14] what do you mean? [42:16] Okay, so insulin, [42:18] so the latest versions were launched in the '90s and 2000s, [42:21] but they got quite along in their lifecycle because— [42:25] As in close to the end of the— [42:26] Well, they were past their patent window actually, [42:28] but there were no competitors. [42:30] Why? Beause net pricing was pretty low. [42:33] How could it be so low? [42:33] Well, the incumbent players, [42:35] mostly Novo and Lilly, [42:36] we can come back to that on GLPs as well, same players, [42:39] had a lot of CapEx in the ground. [42:41] And to start a new insulin company made no sense [42:45] at the net prices we were achieving. [42:47] Yet at the same time, the public viewed this [42:48] as this outrageous price gouging [42:50] because list prices, [42:52] if we were getting about $40 a month of therapy, [42:54] were like $270. [42:56] And so we— [42:57] So who's getting the 235? [42:59] Yeah, so middle actors. [43:02] And so big PBMs like UnitedHealthcare runs and CVS [43:06] and Express Scripts were offering to employers and others, [43:10] the government as well, [43:12] "We will create an auction, [43:13] and in this auction, we will get a take [43:15] on the percent we save you off the list price [43:19] and you'll get a lower price than you could on your own. [43:22] And we will create an auction by..." [43:23] and this is actually a highly interchangeable class. [43:27] They're not exactly the same substance, [43:29] but they are pretty close. [43:30] And so they could do this more easily [43:32] and they'll say, "We'll just pick one, [43:34] and every January, manufacturer, mail us your best deal," [43:39] and the best deals that tended to win, [43:40] we learned through time, [43:41] were those that had the biggest spread [43:42] between the high list price and a low net price. [43:45] So we competed on this. [43:46] What did we do? [43:47] We kept raising the list price [43:49] and modestly lowering our net price. [43:51] That was how the market evolved. [43:52] And after 10 years, you had this huge bubble, [43:55] gross to net bubble, [43:56] and who was paying? [43:57] Okay, so you weren't— No real payers. [43:59] But the person who walked in the pharmacy with no insurance, [44:02] they had to pay that. [44:03] That's outrageous. [44:04] That's what I mean. That should not exist. [44:06] We were able to disarm that through a number of actions. [44:09] But the critical first one was we went to the government [44:12] and we said, "We don't want this problem anymore. [44:14] We're an innovative company" 'Cause it looks bad for you. [44:16] It looks terrible. [44:17] And it's also producing these unfair outcomes. [44:19] We're going to, [44:20] because no generic has applied [44:22] for a copy of our medicine or biosimilar, [44:24] we will create our own. [44:25] So we launched our own biosimilar. [44:28] It says Lilly on the bottle. [44:29] It says Insulin Lispro, which is the generic name. [44:32] And we priced it really cheap, [44:33] like a third of the regular product. [44:36] Similar net price actually, [44:37] but quite a bit less. [44:39] Interesting fact, that launched, [44:42] all these insurance companies and middle people called me [44:44] and said, "Why'd you do this?" [44:45] I said, "Well, because we're trying [44:47] to lower insulin prices." [44:48] They said, "Don't. [44:51] This is a threat to our model." [44:52] It's like, "I don't care." [44:53] Like, we have a higher calling. [44:55] And in the first year, [44:57] no formularies covered this. [44:59] So it was really only for that cash payer. [45:00] No insurance company picked it up [45:02] even though it was dramatically cheaper. [45:04] Now it's about half of the volume, [45:06] but still half not, [45:08] because that model of this margin spread model [45:12] is still there. [45:13] But we largely have defanged that problem [45:15] by introducing a copy of our own medicine. [45:18] You know, I think we can get into differences [45:20] between the US healthcare system [45:22] and the rest of the world [45:23] where the US has a very vibrant private healthcare system, [45:27] but it's kind of weirdly unpopular [45:28] at least in certain parts [45:30] of the, you know, political discussion. [45:31] But there's a choice. [45:32] Yeah, so but taking about R&D. No, it's amazing. [45:34] Actually let me just answer that. So it is true. [45:36] If you went out and said, [45:37] "Hey, I want to back some biotechs," [45:39] and they sent you their business plan, [45:41] 80% to 100% of the revenue and return they'll pitch you on [45:44] is the US. [45:46] Meaning there is no return outside the US [45:49] if you start at the point of origin of the idea. [45:51] Now, once we get to the market with a product, [45:54] it's not sensical to not market it in these countries [45:59] at whatever price you can get [46:00] because your R&D is paid for on the US launch. [46:02] So here you're just margin gathering— [46:04] But it's the free-rider problem. [46:05] But it's the free-rider problem. [46:06] But to John's point, [46:07] this seems increasingly politically untenable. [46:09] Americans are— I agree. [46:10] Americans are waking up to this. [46:12] We should get rid of it. [46:14] It's actually not good for our industry either [46:16] because you get a skewing [46:18] in addition to the problem, the social problem. [46:19] So what does everyone do? [46:20] They tune the R&D model to the US healthcare problems [46:23] when actually we're 5% of the world population. [46:26] So shouldn't we tune it to the global health problems [46:29] and reward the global health problems? [46:31] Yeah, well, we're, you know, 25% of GDP, [46:33] but nonetheless— Yeah, okay. [46:35] Okay, 25% would be a much better improvement [46:37] over 90. When you say [46:38] we should get rid of it, we should solve it... [46:39] Yeah. How? [46:40] Yeah, so I pitched this idea [46:42] to this administration actually, [46:44] which I call the one fair price. [46:46] So— Good branding. [46:47] Yeah, yeah, yeah. [46:49] We're learning. [46:50] I need a hat. One fair price. [46:52] OFP. That's it, yeah. [46:54] But the idea would be that manufacturers introduce [46:57] at the price they want. [46:58] They are restricted by only a couple things. [47:01] One is that they will need to introduce it [47:04] in other developed economies [47:06] in a price band that's sensical [47:08] to the GDP of those countries. [47:10] GDP per capita. [47:12] Because the ability to pay [47:14] I think should largely be borne by more wealthy nations. [47:17] Okay, so you're saying— That's where the surplus is. [47:18] You introduce a drug [47:19] that costs $100 in the US. [47:21] You're saying it should cost, [47:21] you know, on the order of, you know, $70 in— [47:24] So, say, Britain, $70 in the UK or whatever. [47:26] Yeah, it's 30% less GDP per capita, [47:28] we would introduce it at 70. [47:30] Okay. [47:31] Those countries can say yes or no, [47:32] but we would basically sign a compact [47:34] that would say, "That's our deal. [47:36] We think it's worth a hundred." [47:37] We can not sell it there, [47:39] but not because we're lowering it below 70. [47:43] We have to charge them what we think it's worth. [47:45] And you could do that today. [47:47] We could. [47:47] So, you know, why not just do it? [47:50] Why do you need a compact? That's fix number one. [47:51] Fix number two is that the reimbursement system in the US, [47:54] starting with the US government itself, [47:56] would need to get rid of all discounts and rebates [47:58] so that the product moves through the channel physically [48:01] at one price [48:03] and is reimbursed at that same price. [48:05] You have to select that price. [48:06] And here you're not price discriminating anymore. [48:09] You have to sort of look at all the equities around that [48:11] and say, "This is the fair price that I select [48:14] and I'm going to live with that," [48:16] just like other commodities and things we buy every day. [48:19] And there's no skimming of that number. [48:22] And with that then, [48:24] I think you would have two good outcomes. [48:26] You would have a fair decision [48:30] about who pays for the R&D. [48:31] Presumably companies would look at the global opportunity [48:34] or at least the developed countries [48:35] and set a price that might be a little lower [48:36] in the US than normal [48:37] because they want to sell to Europe [48:40] because there's more volume available. [48:41] And if they price too high, because it's indexed, [48:44] they would not be able to do that. [48:44] What you're describing is an instantiation [48:47] of what I view as the general phenomenon. [48:49] One of the biggest shortcomings [48:51] of the US healthcare system in my view [48:54] and one of the biggest critiques you can have [48:55] is none of the numbers mean anything. [48:57] Like, just a number that you see, they're all lies. [49:00] And that kind of has to lead to market failure essentially. [49:04] Yeah, could you imagine this? [49:05] Like, we go to a restaurant tonight [49:06] and someone gives us the wine list [49:08] and a bottle of like a hundred dollar Napa Cabernet [49:11] is like $14,000. [49:13] But then the waiter says, "Don't worry, [49:15] that's not your copay." [49:16] Yeah. [49:17] And so what's my copay? [49:18] "I can't tell you." [49:19] We enjoy the wine. [49:20] We have a nice dinner. [49:22] Four weeks later, you get a letter in the mail [49:24] that starts at the top by saying, "This is not a bill," [49:28] but it says the $14,000 [49:30] and then there's a number of deductions. [49:32] And it says, "This is not a bill. Don't pay this." [49:34] And then later you get an actual bill. [49:37] This is healthcare pricing. But it feels like you could [49:39] pull on this thread quite a bit [49:40] and, you know, the next admin should, [49:43] of just the numbers should mean something. [49:45] Like, the FTC does this a lot. [49:47] You know, they say that for consumers, [49:49] numbers should be trusted [49:50] and yet we kind of let the healthcare system off the hook. [49:51] Yeah, there's no pre posting of pricing. [49:54] Now, with drugs, there is. [49:55] So one of the things we get criticism, I push back, [49:58] is, like, "Well, because you can know [49:59] a list price of a drug." [50:00] Actually most every, the other 90%, [50:03] you mostly can't know the price. [50:04] Well, we've introduced regulations [50:06] in the last couple of years [50:07] mandating some degree of transparency here [50:09] for healthcare. Yes. [50:10] Total failure. Have those worked? [50:11] Yeah. [50:13] Look up in your region who has actually complied. [50:15] Compliance is terrible. [50:17] Most major hospital systems have not complied. [50:19] Or if they have, they put on a website somewhere [50:22] a coded database that is impossible to interpret [50:25] with ICD-10 codes and price points [50:29] that consumers cannot digest. [50:30] It's like "The Hitchhiker's Guide [50:31] to the Galaxy." It's a non-searchable— [50:32] Beware of the Leopard. [50:33] You know, file, flat file, [50:36] with everything they have. [50:37] That doesn't work. It's malicious compliance. [50:40] Yeah, or facial or whatever. [50:42] And so that's not working and we need to have that. [50:44] I've actually gone to like an imaging center [50:46] and I asked, like, "What's this cost?" [50:49] And the person gets irritated with me. [50:51] Like, "Why are you asking that?" [50:52] I'm like, "I don't know." [50:53] Like... [50:55] I generally ask that [50:56] before I consume things. I want to know what it costs. [50:58] I had a situation, [50:59] my physician is like attached to a hospital. [51:02] Here's another terrible thing about US healthcare [51:04] is that the federal rules require payment differences [51:07] based on site of care. [51:09] So if you're an outpatient clinic, [51:11] which is theoretically cheaper, [51:12] the reimbursement is lower, [51:14] not because it's less valuable, [51:16] but because it's cheaper to deliver. [51:18] We punish that. [51:19] We reward where it's more expensive in a hospital complex. [51:22] So my doctor's there, [51:23] he's like, "Hey, you can get a blood draw downstairs. [51:25] Why don't you go do that now and then come back up?" [51:27] So I went down there, they drew my blood. [51:29] Five minutes later, I went to his office, [51:30] the result was there, [51:31] and I'm like, "Oh, I should have asked what that costs." [51:34] So on my way out, I asked. They're like, "That was $650." [51:37] I'm like, "For a blood draw? [51:38] Like, that's insane." [51:39] And they ran it in their own lab there. [51:42] No one asks. [51:43] And I think that's a major, major pricing problem [51:46] we have in healthcare. [51:47] And then, you know, I think people are insulated [51:49] from those costs, [51:50] but in strange ways. [51:51] Some things are deductible, some are not. [51:53] And so it's really very difficult [51:55] to make informed consumer economic decisions in health [51:59] and we need to improve that. [52:00] The other thing I observe about the US healthcare system, [52:02] so Patrick and I both grew up in Ireland, [52:03] moved here for college, [52:05] and the US has a very vibrant private healthcare system, [52:11] which is different from kind of many other countries [52:13] which have, you know, public government-run [52:16] and funded healthcare systems. [52:18] And what I notice is people just have a weird reaction [52:23] to private healthcare. [52:24] Private hospitals people think are weird [52:26] despite the fact [52:27] that, you know, you have much shorter wait times in the US [52:29] than you have in many other countries [52:30] because we have more kind of hospitals [52:32] and all these new private outpatient [52:34] kind of specialty clinics. [52:36] But the biggest one is when people come to the US, [52:38] they're shocked by pharma advertising. [52:40] And, you know, they put on a sports game [52:42] and just, like, the break comes and it's all pharma ads. [52:44] And my understanding is that there's, again, a significant [52:46] pro-social defense of this, [52:48] which is many of these drugs are shown [52:50] through all the extensive trials [52:51] that, you know, we make you guys do [52:53] to have a significant health benefit, [52:56] and then it leads to them being prescribed more by doctors [52:59] because people actually ask their doctors, [53:00] they do in fact do the thing, [53:01] they ask their doctor about them, [53:02] and it leads to more usage. [53:04] But yet people just find the whole thing weird. [53:07] You know, private, for-profit healthcare. [53:09] And so do you have a view on where this goes? [53:11] What we can do about it? [53:12] Yeah. [53:13] Well, not having grown up in a system like Ireland, [53:17] but I lived in Canada for a while, [53:18] for six years, [53:19] so I was treated in that [53:21] and I've seen it, yeah, I find it weird the other way. [53:23] Totally. Because that's [53:25] your conditioning. [53:26] I think there's good and bad to both. [53:29] I think actually, we were talking about prevention earlier, [53:31] to some degree primary care, [53:33] what I experienced in that country [53:35] was pretty good quality of care, very standardized, [53:40] which has a confidence boosting thing [53:43] when it's the same for everybody. [53:45] That also, in specialty care, is implemented [53:48] but actually to kind of a negative result, [53:51] because take, like, diagnostics. [53:55] The US and China have something like 70% [53:58] of all diagnostic capacity in the world. [54:01] That's crazy. Yeah. [54:02] But your chance of finding a tumor or something [54:05] is much, much higher in those two countries [54:06] than Ireland or certainly the UK. [54:09] And that's not a good outcome. Why? [54:13] Because I think they're focused [54:14] on cost of delivery and evenness [54:17] instead of exceptional care. [54:19] So I think we've moved on that axis [54:21] of, like, let's offer something that could be the best [54:24] and charge for it, [54:25] yet for common conditions, [54:28] we've also moved on that axis unnecessarily. [54:31] And so here you end up [54:32] with hard, you know, oak-floored primary care offices [54:36] and beautiful drapery and furniture. [54:39] That's the basis of competition [54:40] instead of actually, it's quite a simple thing, [54:42] "You need your flu vaccine. [54:45] Like, just get in and get out." [54:46] And there's a third thing coming, which you touched on, [54:48] which is prevention and self-care. [54:50] And I actually think that if we think [54:52] of the funding mechanism in our country, [54:54] but also in Europe, [54:55] and Europe's system has problems going forward, [54:58] it was really built on an acute care model [55:00] when most illness and death was accidents, [55:02] you know, things we couldn't solve. [55:05] It was basically treat them as best you can [55:07] and people were going to expire. [55:09] That's what a hospital did. [55:10] We've, of course, now evolved well beyond that. [55:11] I think that's 30% of US cost. [55:13] Now 70% is primary care [55:15] and sort of, you know, the chronic disease. [55:18] And those institutions and those funding models [55:20] are really poorly suited to that, [55:23] and particularly so if behaviors have an input into that. [55:29] Don't we want people before they get the disease [55:31] to modify their behavior? [55:32] Well, how do you charge for that? [55:33] And so, you know, there's a little bit [55:35] of a selection problem [55:38] that, you know, the best healthcare systems [55:40] get the worst behaved people [55:42] because the coverage is better. [55:44] So I think it is time for a rethink of the whole thing. [55:47] And I would think of those three different things [55:49] and try to solve for them differently. [55:51] Right now, we pretty much have one answer, [55:53] and the Europe answer has produced [55:55] kind of an institutional rationing model [55:58] that seems very fair [55:59] but actually produces poor outcomes for acute conditions. [56:02] And the US, which is very expensive, [56:05] also is unfair [56:07] but produces good outcomes for acute conditions. [56:09] Probably just the same for everything else, [56:11] but costs too much. [56:13] That's I think needs to be addressed. [56:15] In the future, [56:16] and here, medicines plus information [56:18] I think can play a really big role in disease prevention. [56:21] In the past, we haven't been able to make... [56:23] the age old problem in medicine [56:25] is this thing we call therapeutic index. [56:27] That's the difference between a dose which is harmful [56:31] and one which is helpful. [56:32] And a therapeutic index that's small is difficult. [56:36] You have to very precisely dose [56:37] and people have differences, [56:39] so it requires a lot of attention. [56:41] But over time, the techniques we make drugs with, [56:44] that therapeutic index, the TI we call it, is expanding [56:47] and expanding non-linearly. [56:49] Because of that— [56:51] Sorry, why is it expanding? [56:52] Yeah, because of new drug technologies. [56:54] Two main ones. [56:55] One is, going to the root of disease, [56:58] whether it be genetic or RNA blocking medicine. [57:01] So a lot of diseases have excess protein. [57:03] We can now really pristinely block [57:06] RNA production of the protein [57:08] and the disease, without a lot of side effects, goes away. [57:12] And by the way, these medicines happily [57:14] also have sort of this catalytic effect. [57:16] So they last a really long time. [57:19] I mentioned Lp earlier. [57:20] So that's a kind of cholesterol that's untreatable today. [57:23] We're developing a medicine [57:23] that will be a once or twice a year treatment for this [57:27] and the side effects look totally benign. [57:30] That's a really wide therapeutic index, right? [57:33] So now when you have that, [57:34] you can think, "Well, my trials I can run faster [57:37] because I don't have worries about treatment. [57:39] They can be cheaper. [57:41] I can charge less and get it to more people at scale. [57:45] And I actually don't really need a healthcare system." [57:49] And here, maybe back to the GLP-1s, [57:51] that's giving us a little bit of glimpse of these. [57:53] These are more invasive than what I just described, [57:56] but pretty safe. [57:57] People know how to treat themselves. [57:59] You certainly know if you have overweight or obesity. [58:01] You don't need a doctor to tell you that. [58:02] And platforms like our direct platform have really taken off [58:06] because it's self-paid, [58:08] but people skip all this other morass [58:10] and getting the, "This is not a bill," piece of paper. [58:12] They're just like, "Here's my visa card number. [58:15] Yeah, charge me 500 bucks, [58:16] but my problem's getting solved." [58:18] I think for prevention, [58:20] that's an intriguing future, direct to consumer. [58:23] We're here in the great city of South San Francisco, [58:27] the home of Genentech [58:30] and to some significant extent of the US biotech sector. [58:34] And we're talking a whole bunch [58:37] here in this conversation [58:38] about fostering and inducing [58:40] and creating adequate incentives for R&D. [58:42] I think the, [58:44] I mean, to some extent, [58:44] you can bucket biotech and pharma separately. [58:47] Obviously the lines blur, [58:48] but, you know, there's kind of two poles. [58:50] And I think it's very striking the extent to which, [58:53] as far as I can see at least, [58:55] the introduction of new medicines, [58:58] you know, new molecular entities, whatever, [59:00] is increasingly dominated by biotechs. [59:03] And you would know the numbers better than me, [59:04] but I think that around two thirds, [59:06] both of the revenue [59:07] and also of the, just by count, [59:10] the introductions themselves, [59:11] are attributable to biotechs rather than to pharma. [59:14] And so I guess I'm just curious [59:16] how you think about this landscape. [59:18] I mean, maybe there's some view where the role of, [59:22] in an extreme, one role of pharma [59:25] would be to be a kind of private equity portfolio manager [59:30] where you take stock of the landscape [59:32] and you look at who's doing well, [59:33] whose approach you believe in, et cetera, [59:36] and you have the risky innovation be done [59:38] by earlier stage entities, [59:40] by venture capitalists, what have you, [59:42] and then you bet on winners, you go and scale them, [59:45] and you distribute them. [59:46] One model. [59:47] Traditional model is actually, [59:48] no, you have all these internal R&D capabilities [59:51] and you're vertically integrated [59:53] and you have economies of scale [59:54] and so on and so forth. [59:55] Are we shifting there? [59:57] You know, where between those poles ought we be? [60:00] How is it changing? [60:01] Just thoughts on that whole landscape? [60:03] Yeah, well, I think there's three models [60:05] that have emerged. [60:06] One is the biotech that grows up. [60:08] Another is the outsourced early model you're describing [60:12] where we just say, "We're good at clinical trials. [60:14] Everything before that, [60:15] just gobble it up as companies mature." [60:17] You know, because the capital market is so terrible [60:19] as I described, [60:20] it's a liquidity event for the investors. [60:22] They go back and try something earlier. [60:23] And then sort of the hybrid of fully integrated plus. [60:28] I don't think anyone's really pursuing [60:29] just fully integrated anymore. [60:31] I think your head's in the sand if you're doing that. [60:34] We're running the third one. [60:35] The other two are reliable— That hybrid? [60:37] Yeah. And why do we make that choice? [60:39] I think we observe a few things. [60:41] First of all, while it's true that the origin of, [60:44] I think it's a little more than half of medicines [60:46] approved in the last 10 years have come from biotech, [60:49] hardly any of those traveled all the way through biotech [60:52] because there are, as we talked about earlier, [60:55] huge checks to write and risks to take [60:58] and the biotech investor base [60:59] is not interested in those risks, [61:00] those very binary large checks. [61:03] Whereas we can absorb them, [61:04] we can run a portfolio across many of those. [61:07] There's also scale economies in clinical trials. [61:10] There is no doubt in my mind we are faster, more robust, [61:14] probably cheaper than actually every biotech out there [61:17] trying to do their own early phase clinical trials [61:20] and manufacturing and distribution globally. [61:22] So those things benefit scale. [61:24] What doesn't? [61:25] It is discovery. [61:26] It's the early phase. [61:27] I think that's a more diffuse undertaking. [61:30] What we've done, starting with my predecessor [61:32] maybe 15 years ago, [61:33] is we started putting diverse, [61:35] spreading out our labs. [61:37] Scale's bad. [61:38] And I think our idea was 300 or 400 people, about right, [61:44] Allow for some deviation. [61:45] As in you think the— Put some outposts out there. [61:47] The act of invention has diseconomies of scale. [61:50] Yeah, so we started in San Diego. [61:52] We built a monoclonal antibody biotech hub there. [61:55] It's produced like a third of the medicines we've made [61:57] since we started. [61:58] Hugely successful. [61:59] There's 400 employees. [62:00] So it's like a biotech, but it has some benefits. [62:04] They don't have to spend any time with venture capitalists [62:06] or raising money or fussing over CapEx [62:09] and, you know, ups and downs in the markets. [62:12] Why is it good to be small? [62:13] Why is it good to have the 400-person center [62:15] in San Diego— It's interesting, you know? [62:16] And not the Indiana super center— [62:17] I don't know. [62:18] I think it's good to big on clinical trials [62:20] because you're doing global reach. [62:21] It's a scaled operation, manufacturing, distribution. [62:24] Discovery, I don't know. [62:26] Now, we have people look across, [62:29] and I think there is a benefit in being in a tribe, [62:33] but I think it's also good to have some independence. [62:36] There's a long list of pretty compelling inventions [62:40] that came out of Lilly [62:41] that were not sanctioned projects. [62:43] And so that tells you something, [62:44] that curiosity and sort of the scientific endeavor— [62:46] Weren't sanctioned in what way? [62:48] Meaning it had a budget [62:50] and had a number and a name and whatever and it was, [62:53] or pointed at, like, "Oh, here's a target. [62:56] Go put a team around that and come up with a medicine." [62:58] That does work, [62:59] but sometimes someone just says, [63:01] "By the way, I didn't tell you, [63:03] but I've been working on this thing over here [63:05] and it seems pretty interesting." [63:06] And then we fund it. [63:07] Have you guys formalized that, like, Google 20% time [63:09] where are some people are just doing curiosity or? [63:11] It depends on the area, [63:13] but we have this allowable variation idea where— [63:17] Is that the phrase, allowable variation? [63:19] No, that's my word. [63:21] But basically you just turn away and see what happens. [63:25] And we don't manage the budgets down to the nickel [63:28] at these sites. [63:29] And if you get good people, [63:30] it's like, "Let them do their thing, [63:32] let 'em cook, [63:33] and let's see what happens." [63:34] And that's certainly worked. [63:35] We built a big site here— The let them cook school [63:37] of pharma innovation. [63:37] Yeah. [63:38] Yeah, it's a thing. [63:39] I can see this as a book. [63:42] A lot of pharma innovation is iterative, right? [63:44] So the other problem is people quit too soon. [63:48] So back to GLP-1s, this story's been going, [63:51] we launched the first GLP-1 twice a day injection in 2006. [63:55] People say, "Oh, Dave, when did you know, [63:57] like, Zepbound would be huge?" [63:58] I'm like, "I don't know, 2016, 2017?" [64:01] It was a long time ago. [64:03] We've been on this story for a long time [64:04] and it was protein engineering [64:06] and making better molecules that could be dosed higher [64:08] that led us to get to more weight loss effects [64:11] and then harnessing two mechanisms. [64:12] So that's just, that's not like light bulb, [64:15] single scientist in a dark lab, [64:17] like, "I've got it." [64:19] It's teams just grinding on a theme. [64:21] And if we had directed that, [64:23] like, "Oh, make it dual acting. [64:24] It'll be better," [64:25] we wouldn't have gotten the right answer. [64:27] This is all, like, your internal innovation essentially. [64:28] Yes, so that's running. [64:30] Where do the biotechs appear? [64:31] And why do we do this? One, it is productive for us [64:33] and I do think we've built a capability [64:35] to, frankly, make molecules with more drug-like properties. [64:40] So by the time you get to that expensive escalation [64:43] of clinical trials, [64:44] the drug's behaving like a drug. [64:46] Some of that we talked about earlier with AI [64:48] and other tools we can equip. [64:50] So these things tend to work, these things don't. [64:52] It's pretty empirical still. [64:53] We don't know all of biology. [64:55] Some of it's just knowhow [64:55] of the chemists and the scientists. [64:57] But I think when we buy biotechs, [65:00] often we love the target idea. [65:02] We love the novelty of what they did. [65:04] It's just rather imperfect. [65:06] It's an 80/20 they've done. [65:08] And we will often take that invention cycle back [65:11] and do a whole other invention loop to refine it further [65:14] so that it is what we call like a big pharma asset. [65:15] So what you're buying [65:18] is the proof that there's something here, [65:20] but not necessarily [65:20] the specific solution. Yeah, exactly. [65:22] And if they haven't disclosed it [65:23] and they've got a lead, that's very interesting. [65:25] But we actively cultivate [65:27] kind of a proximate satellite group of companies [65:31] that are doing this independent of us [65:32] but we have ownership stakes in. [65:34] And then, of course, we have a watch list, [65:37] ever-growing because of China, [65:39] of entities we don't have an interest in [65:42] but are doing interesting things [65:43] and we want to, [65:44] we have all their events mapped out [65:45] and we're watching pretty much [65:47] every clinical or preclinical readout, [65:49] every patent posting in the industry, [65:51] we have a way to monitor. [65:57] For most of its 150-year history, [65:59] Eli Lilly's relationship with patients [66:01] has been through pharmacies and other intermediaries, [66:03] but that's changing. [66:04] As Dave mentions, with LillyDirect, [66:06] patients can buy treatments, like Mounjaro and Zepbound, [66:09] directly from the company [66:10] and get them delivered right to their door. [66:12] And they're not alone, [66:14] at Stripe, we're seeing a new playbook emerge [66:15] where established businesses who used [66:17] to sell via complex distribution chains [66:20] are now building direct digital relationships [66:22] with their customers. [66:24] And with Stripe, that's much more likely to succeed. [66:27] It's happening across all industries, [66:29] not just in healthcare. [66:30] Media companies, like FOX Sports, [66:32] they used to reach viewers [66:33] only through traditional cable bundles, [66:35] but now they've built [66:36] a global direct-to-customer streaming platform on Stripe. [66:39] Or Ford, they're selling cars [66:40] and trucks online through Stripe, [66:42] and not just through dealerships. [66:44] Some companies, they're creating entirely new products [66:47] to take directly to their customers. [66:48] So Fender, the guitar company, [66:50] they created a tuition marketplace that has enabled millions [66:52] of guitar classes already. [66:54] Large enterprises are creating better direct offerings [66:56] than ever before with Stripe. [66:58] Learn more at stripe.com/enterprise. [67:04] So we just mentioned [67:05] we're in South San Francisco in the Bay Area. [67:07] The Bay Area, of course, [67:08] used to have a vibrant electronics manufacturing industry [67:11] and— Yep, it left. [67:12] Yeah, companies like— [67:14] There's still the APX sign. [67:15] Yeah, exactly. And, you know, [67:16] some of the old— Yes, but— [67:17] Fairchild. Exactly. [67:19] But you mention, you know, Cypress Semiconductor [67:21] and, you know, people don't really know [67:22] what you're talking about these days. [67:23] The biotech sector started here in the '70s, I guess. [67:26] Yeah, give or take. And is, exactly, [67:27] and is still today reasonably vibrant. [67:30] Although, to your point, [67:31] it's had a tough maybe couple of years. [67:34] The share of the global drug pipeline [67:36] that was Chinese 10 years ago [67:38] was I think in the small single digits. [67:39] Approaching zero. Yeah, exactly. Right. [67:41] And now I think it's high 20s, [67:43] approaching a third. A third. [67:44] It's 30%. Okay. It's amazing. [67:46] Now is in fact 30%. [67:47] Where does this go? [67:49] And is there a US biotech sector in 20 years? [67:54] Or like electronics manufacturing, [67:56] does the whole thing just go to China? [67:58] I wouldn't predict that outcome, [67:59] but I think we should ring the alarm bell right now. [68:03] Well, I mean, is it bad? [68:05] It's not great, I would say, [68:09] I think for two reasons. One— But if they can do it [68:12] better and faster and cheaper [68:13] and we get the drugs, [68:14] isn't that awesome? [68:15] That is good, especially if the last part's true. [68:18] But I think the economy around biopharma [68:22] I think has some unique properties [68:24] that should make us want to own it. [68:25] One is it is in the knowledge economy. [68:29] To me is sort of the pinnacle— [68:31] Paradigmatic knowledge economy. [68:32] It's the premier league of knowledge economy. [68:36] You guys fund and know people... [68:37] I mean, the diversity of skill sets needed to do it well [68:42] at the highest level of their game is extreme. [68:45] I don't think it exists, [68:46] maybe rocketry, maybe there's a few other things like this, [68:49] but it requires a lot of talent. [68:52] So it tells you how you're doing I think in a way. [68:55] Can you integrate these, can you train people, [68:58] attract them from abroad, like you folks, [68:59] or train them here [69:01] and put them together in a way that produces new value? [69:03] Okay, so there's a kind of shadow [69:05] passing over the country if biotech— [69:07] And it's rewarded. [69:09] I mean, these are well-paying jobs. [69:11] There's a big economic footprint that goes with that. [69:13] And increasingly we're putting manufacturing near it, [69:16] so it actually has a trickle down [69:17] that's pretty significant as well. [69:19] And then it invokes also security concerns. [69:25] And you can imagine if we had the COVID pandemic, [69:29] in that case, basically 80% [69:31] of the medicines and vaccines that worked [69:34] were produced in the United States. [69:35] China produced some of those things. [69:37] None of them really worked. [69:38] We didn't import them. [69:39] But imagine if that was flipped. [69:41] And we're in— I mean, the EU saw this [69:44] with all the battles, yeah. An extorted position. [69:45] Yeah, the EU had a huge controversy, [69:47] and they're friendly and we had open trading, [69:49] but they have had a reckoning as well. [69:53] And there's some things [69:53] you just want to have a national competency in. [69:55] So, anyway, I have a theory about the industry. [69:57] I think there are truly novel concepts [70:00] that require a little more time and work [70:03] and it is yet to be seen [70:06] whether China has perfected that [70:08] in a way where they can create their own local system. [70:11] What they've certainly shown now, though, [70:13] is the iterative derivative, [70:16] which is a big part of the substrate of the industry, [70:19] they are refining [70:20] and becoming experts in very, very quickly. [70:23] I think that's not all bad. [70:24] There'll be more competition. [70:26] You know, there's an effect in China [70:28] where their own sort of price competition [70:30] defeats their own industries, [70:32] and you're kind of seeing that in biotech as well, [70:34] the race to the bottom on pricing. [70:36] But I think we want a national competency here. [70:39] And I think this has been a hub. [70:40] Boston is still a big hub. [70:42] We want to keep those. [70:43] But there's this new phenomenon where, [70:46] my understanding, please correct me, [70:48] is that, you know, traditionally you develop a molecule, [70:50] you patent the molecule. [70:51] Now there's the growth of these clones [70:54] where you can have the molecule [70:55] be trivially different enough for patent purposes, [70:58] but still the same action [70:59] and there's a huge amount of that [71:01] coming out of the Chinese biotech sector. [71:03] Doesn't that— But I think also [71:05] some real, like, I mean [71:07] that seems real and good [71:08] and novel— Also real innovation. [71:09] That's an interesting one, yeah. [71:10] But don't the clones effectively erode the patent system [71:14] and, you know, we would think shortening [71:15] the patent life to 15 years would be bad, and/or 10 years. [71:19] And this is shortening the patent life effectively. [71:21] Yeah, I think that practice is what I'm talking about [71:23] with this sort of refine and replicate [71:26] and our own patent system forces this, right? [71:28] So we have a, in 2011, the US changed the patent system [71:32] to first to file versus first to invent. [71:34] You used to be able to invent and sit on your patent [71:36] and all you had to do is prove, [71:38] it was messy court cases 'cause your lab notebook said, [71:40] "Oh, it was October of 2017 [71:42] and mine says 2018. [71:43] Well, which was what," right? [71:45] But now you rush to file. [71:47] And the government I think has 12 months [71:49] that they sit on that inspection [71:51] and then the patent, what is a patent? [71:54] It's a degree to publish your finding, right? [71:57] To make it a public good in return for that monopoly. [72:00] But if the monopoly is debased by 30 Chinese biotechs [72:06] who feed that patent into a computer, [72:08] the computer then can imagine chemical structures [72:11] that have one or two atom differences [72:13] that don't fit within the patent, [72:15] and then make that substance, test it, works just the same, [72:18] you've created basically a shadow generic industry [72:20] and undermined the patent system itself. [72:22] I don't think that's a great thing. [72:24] So was the shadow generic industry [72:26] basically launched by first to file [72:28] because we published the instruction manuals? [72:30] Yeah. I hadn't realized that. [72:31] And do you think, is it your view if you just switch back? [72:33] No, what I would do is, [72:36] if we want a sort of like an America-first patent system, [72:39] or Europe first, [72:41] we should do two things. [72:42] We should create a belt and suspenders so that, [72:45] some patents are tricky to write. [72:48] So there might've been an IP space that's been mined [72:51] and people are around the idea, [72:52] but you had an insight that actually unlocked the truth [72:57] and there's like a thin strip, [73:00] but you have to carefully navigate this concept [73:02] called prior art [73:03] and get that patent to stick. [73:05] People tend to constrain it more [73:07] and you leave room at the edges. [73:09] Should we make the basis a reward to investors, [73:13] the patent filer, the patent writer? [73:16] That's really the most valuable step in that case. [73:18] I don't think so. [73:19] So a belt and suspenders would be, say, [73:21] independent of a patent, [73:22] we'll give you 12 years [73:24] if you produce primary data on this product [73:26] all the way through phase III. [73:29] That's a $3 billion ticket. [73:30] These copycats aren't going to do it. [73:32] They're certainly not going to do it in US data. [73:36] And that'll be more or less constrained in markets [73:38] that don't care about that issue [73:40] and don't have this data exclusivity provision. [73:43] This already exists in biologics by the way, [73:44] so all we have to do is, in small molecules, [73:46] and this problem we're talking about [73:47] is mostly a small molecule problem. [73:49] The other thing we should do is say, [73:50] "If you are in a league of nations that really respect IP, [73:58] those nations will extend this confidentiality period [74:00] beyond the patent inspection period, [74:03] that society will get the recipe well in time [74:06] for a generic company to copy it later, [74:08] but why does it have to be 12 months?" [74:11] That's a super short time in drug development. [74:13] How about six years? [74:15] Then that product is well into the clinic, [74:17] the copycat nation can spin up in China, [74:21] but it's not going to make a difference. [74:22] So I think that those two solutions together [74:24] would be what I'd recommend. [74:26] And the electronics manufacturing industry went to China [74:30] for pretty straightforward reasons. [74:32] You know, it's— [74:33] Cost. Exactly. [74:35] It's very labor intensive [74:37] and I guess raw material intensive [74:39] and, you know, just lots of reasons, [74:40] it's cheaper in China. [74:41] But the software production industry has not. [74:43] That is still here. [74:45] To your point, biotech is an extremely knowledge [74:48] and tacit expertise intensive space. [74:53] Why is that one going to China? [74:55] Yeah, well, I'm not sure it's going, [74:58] that's why I say I'm not sure it's, [75:00] the story's over yet. Well, it's relevant. [75:00] We're not eclipsed, [75:02] but, you know, a significant share of it has gone to China. [75:04] I mean, China has a robust software industry. [75:07] We have a two internet world [75:11] and they have their, [75:12] I don't know how well Stripe does in China, [75:13] but I would guess not very well. [75:15] Well, not domestically. [75:17] Yeah, okay, well, or Adobe. [75:19] I'm on the board of that company [75:20] and I can tell you they have di minimis sales in China. [75:23] But a lot of usage. [75:25] Well, and there are tools domestically [75:26] that have been built around that [75:28] that have their own economy. [75:30] That could be a state that this is driving into [75:33] where there's sort of medicines that grow up in the world [75:36] where there's perhaps more transparency, [75:38] where the regulatory systems are more confidence instilling, [75:42] and we perhaps reform our patent system to protect that [75:46] and China has their own version, [75:47] and there's good reasons they would want to do that [75:49] because if we own all the medicines [75:51] and there is another crisis, [75:53] I'm sure they would be deeply uncomfortable. [75:54] In fact, they didn't even approve, [75:56] they had rights to the BioNTech-Pfizer vaccine [76:01] for almost the entirety of their pandemic. [76:03] They never approved it. [76:04] Why? [76:05] I think it was a national pride issue. [76:07] Didn't want to be beholden. [76:08] It was a German invention actually. [76:10] But, yeah, so, again, it intertwines with, [76:14] you know, health emergencies are social crises [76:17] that politically are difficult to navigate, as we saw, [76:20] and having that competency is important. [76:22] So I totally respect their desire to build this. [76:25] We have brilliant Chinese scientists in the world [76:28] and many of them work at our company [76:30] and were trained in the United States. [76:32] That's all good. [76:33] I just think we don't want it to move all offshore. [76:35] I don't know when exactly the generics industry, [76:38] like, really rose to be such a large share [76:42] of consumed pharmaceuticals today, [76:44] but I'm very curious about the quality control [76:47] and the attendant regulatory apparatus around it [76:50] where, I mean, as we've been discussing, [76:52] there's such scrupulous and stringency [76:55] around clinical trials for new molecules [76:58] and introductions and so forth. [77:00] My understanding is that for generics, [77:02] a lot of the external validation certification happens [77:06] at the manufacturing plant level, [77:08] not at the individual drug level. [77:09] And that for the individual drugs, [77:11] it is substantially a case of self-certification [77:15] and, you know, presentation of one's own data [77:17] as opposed to external audits. [77:19] And there have been many cases [77:21] of documented fraud and malfeasance here. [77:25] The largest generic manufacturer in India, [77:28] I forget its name, [77:29] paid a half a billion dollar fine in 2013 for— [77:31] Yeah, Cipla. [77:32] Yeah, for rampant fraud and falsification, [77:34] abuse of the biosimilarity analysis and so forth, [77:38] which obviously is bad, [77:40] but, like, the apparatus and the FDA regime is such [77:45] that there are very obvious incentives for that to happen. [77:49] And then I'm very struck by how anecdotally and online, [77:53] there are so many reports [77:54] of people switching from brand name medication [77:58] to generic medication, [78:00] finding them to be very different, [78:02] you know, subjectively and experientially, [78:05] switching back to the branded pharmaceutical, [78:09] and, you know, things going back to normal as it were. [78:12] And so there's some some kind of subjective version [78:14] of, you know, the generic is, in fact, [78:15] not, you know, directly substitutable. [78:18] And so I guess, I mean, I know nothing about this domain. [78:21] This is all just observed from afar. [78:22] I guess I'm curious for your thoughts here. [78:24] Like, does the generic industry actually work [78:27] as well as we think it does? [78:28] How much fraud is there in actuality? [78:30] You know, when your kids or family or friends or whatever [78:33] are taking a generic, [78:35] you know, do you advise them to exercise some caution? [78:38] Thoughts in this whole space? [78:39] I think on the whole, [78:40] the generic environment in the US, [78:42] which is the most developed [78:44] in the sense of percent of medicines consumed [78:47] and the cheapest in the OECD, [78:51] has been largely a positive outcome [78:53] because it's made effective medicines abundant [78:56] at very, very low cost. [78:57] When you say the generic industry in the US, [78:59] do you mean those manufactured in the US [79:00] or consumed in the US? There are no generics [79:01] manufactured in the US really. [79:03] So those that are licensed in the US and sold here. [79:06] That said, and I say that [79:08] because, you know, take an invention like statins [79:12] or HIV drugs or we invented Prozac, [79:16] still the standard of care in treating depression, [79:19] it's like three cents a day. [79:21] I don't know any product you can buy for three cents a day, [79:23] but, you know, that's an incredible value for the system. [79:28] And back to the return on investment thing, [79:30] somehow we have not articulated this very well, [79:33] but the fact that we spent money [79:34] in the '80s researching Prozac [79:36] and still today [79:37] there's tens of millions of people benefiting, [79:39] that's a public surplus [79:41] that should make us want more of those inventions. [79:43] But, anyway, digression. [79:45] So in the '80s also, [79:46] there was a change in the policy in the US [79:48] which was a trade to basically make it easier [79:52] for generics to enter the market after patent expiry [79:55] in exchange for less patent litigation. [79:59] So there was a structured path to litigation. [80:01] It wasn't just a free for all. [80:02] It used to be in the '70s, '60s, [80:04] the day you launched, you'd be in court, [80:07] and by nature, those cases are you have to defend all comers [80:11] and if one gets through, you lose. [80:12] So it's a very asymmetrical problem and it wasn't good. [80:16] So we traded that for certainty in a time window [80:20] and a structured path to patent litigation. [80:23] But the day after the last valid patent expired, [80:27] generic could enter freely [80:30] and could get on the market with a clearer bar, [80:35] a lower bar perhaps. [80:37] There's two deviations that can occur. [80:39] One is, it's the so-called 5% rule [80:42] where plus or minus 5% of the active ingredient [80:45] and some dosages in some people [80:47] are more sensitive than that [80:48] and I think some of those people have an effect. [80:50] Also, there's this, [80:52] particularly in a dry product, in a pill, [80:54] there's excipients, which are the other ingredients. [80:57] Most of what you're taking [80:58] is not actually the active ingredient. [81:00] Some of them are buffering compounds, [81:02] some of them affect absorption rate. [81:05] So those two things combined do lead to different effects. [81:08] And there is no requirement [81:10] for small molecule chemical medicine [81:13] to show proof of efficacy of any kind. [81:16] So you can do pretty simple laboratory experiments [81:20] and absorption experiments in a small number of people. [81:22] This is due to like PK [81:23] but not efficacy. Yeah, exactly. [81:25] And that is what they all do. [81:26] None of them do efficacy. [81:28] So that makes them cheap [81:29] and it makes them, plenty of them coming. [81:32] But it has this side effect. [81:34] I think it might be useful [81:36] to have a way to flag medicines [81:38] that before they go generic [81:39] have this known dosing sensitivity. [81:43] You know, again, back to the less binary regulatory, [81:47] you know, a little more judgment where there's a dial. [81:49] I think the manufacturing problems you mentioned are real. [81:52] In the quest for low cost, it all moved offshore. [81:55] These are basically chemical plants [81:57] and in the prior iteration of that technology and our EPA, [82:01] it became non-economic to make these medicines in the US [82:04] or Ireland for that matter. [82:06] There are some eastern European companies, [82:09] there's a big Israeli company, [82:11] there's several Indian companies, [82:13] and many Chinese companies that are in this business. [82:15] That's where these drugs come from. [82:17] I think that is not so stable either. [82:20] And we probably should pay a little bit more for generics. [82:23] You sometimes read about injectable generics in particular [82:26] that run short. [82:27] That's a more complicated manufacturing process, [82:30] so if you do it cheaply, you run into more problems. [82:33] We probably should pay a little premium for resilience. [82:36] Right now, that's not the situation. [82:42] So GLPs obviously were initially researched, [82:47] the initial R&D was done for diabetes. [82:49] Yes. [82:50] And then it was noticed during the clinical trials [82:53] that people were losing weight. [82:54] And so now I think the big societal discussion, [82:58] like Patrick and I were talking about, [82:59] "The Information" is a Silicon Valley tech publication [83:01] and they ran a survey about GLP-1s [83:03] and half of readers of "The Information" are on GLP-1s. [83:06] In this part of the country. Exactly. [83:08] As you can tell, it's not a normal region. [83:10] Anyway, so there's the weight loss— [83:11] But maybe in as much as Silicon Valley [83:13] is some kind of harbinger of things to come. [83:15] Could be, early adopters. Potentially. [83:16] So there's this big— Tech forward. [83:17] Weight loss discussion [83:18] and, you know, that's the, [83:19] Ozempic is obviously, you know, a different brand name [83:21] for a different drug, [83:22] but people are very familiar with it. [83:24] But now it seems like we are starting to figure out [83:27] there are all these other potential benefits of GLPs [83:32] that do not seem to be fully explained by the weight loss. [83:36] And so there's the cardiovascular benefits, [83:38] there's potential Alzheimer's links, there's fertility, [83:41] there are all these things [83:42] that aren't just, "You lost weight." [83:45] What do you think is going on there? [83:46] Explain. [83:47] Yes. So first of all, [83:48] it's different from what you just described as this accident [83:51] because actually we knew [83:52] and we were involved with the, [83:54] we launched the first GLP-1 in the world. [83:55] You're saying this is the triumph of rational design. [83:59] It's not totally rational, [84:00] but I can tell you that because the, [84:02] so GLP-1, let's go all the way back, [84:04] it's a super family of things we call incretins. [84:06] These are hormones that signal our brain [84:10] and other tissues from our gut. [84:12] We always think about our brain being in charge. [84:14] That's not how we work. [84:16] And it's back to this basic system of survival, [84:19] which is nutrition— John's very aware [84:21] of this. [84:21] Nutrition. It's a joint oversight. [84:23] When you eat a meal, which hopefully we'll do later, [84:26] our gut actually communicates with the rest of the body, [84:28] "Hey, food's on board. [84:30] You don't have to eat as much." [84:31] Maybe you get a satiation signal. [84:34] Your fat cells are told to absorb free fatty acids. [84:37] Your liver kicks into gear [84:38] to release glycogen and other things. [84:41] So all that process is, [84:42] insulin is released to absorb the nutrients, [84:45] is kicked off by incretins. [84:46] This was discovered in 1971 basically, [84:50] called the incretin effect. [84:51] It was a scientist who noticed [84:53] that if you feed someone an equivalent amount of pure sugar [84:57] versus stick it in their veins or arteries, [85:01] that their insulin levels, their metabolic processes, [85:05] rev up a lot faster. [85:07] Interesting. It's called the incretin effect. [85:09] So that the local process of ingestion through the gut [85:13] created this other thing. [85:14] That scientist and others then isolated GLP-1 [85:17] and another one called GIP, actually isolated first. [85:20] GIP and GLP make up tirzepatide, [85:22] which is Zepbound and Mounjaro. [85:23] GLP makes up semaglutide, which is Ozempic, [85:26] and was also exenatide. [85:29] The problem with these hormones is our own versions [85:32] have a half-life of like five minutes. [85:34] So they don't make very good drugs. [85:36] If you take that protein and sequence it, [85:38] people did this, [85:40] you'd have to walk around with an infusion all day. [85:42] So the longer-lasting action was the invention. [85:45] Was the invention. [85:46] And the first one that was, [85:47] before we really knew [85:48] how to do protein engineering systematically, [85:50] which we can do now, [85:52] it was found in nature actually, [85:54] famously, in the saliva of a Gila monster. [85:58] So randomly some zoologist was testing [86:00] the interesting properties of Gila monsters and noticed, [86:03] and profiled all these proteins, [86:05] and one of them was a mimetic of human GLP-1. [86:07] And, you know, did a literature search and found that. [86:10] Another scientist found that paper [86:12] and said, "That's interesting. [86:13] Let's test that compound," [86:14] a few different amino acids, [86:16] and sure enough, it lasted about four hours half-life. [86:19] So we can make it a twice a day injection. [86:22] We made that into a medicine for diabetes. [86:25] And on the cover of our annual report in 2006, [86:27] there's a woman who was one of the first patients [86:29] with a quote, says, "My diabetes is under control [86:31] and my friends say I'm losing a little weight." [86:34] That was 2006. [86:36] So this overnight phenomenon [86:37] of Ozempic and everything else, old story. [86:40] Why didn't we do it then? [86:41] Well, we needed to get the dosages higher. [86:44] And it turns out that this mechanism, [86:46] which is common for a lot of hormones, [86:48] has a threshold effect for efficacy. [86:52] You have to get above a certain level in your blood, [86:54] and certainly to lose weight and really suppress appetite, [86:56] have to get that number up, [86:58] but a peak-to-trough effect on side effects. [87:01] So the up-down part causes the nausea, [87:04] but the absolute level causes the effect. [87:07] So how do you separate those things? [87:08] You need a flat long-acting. [87:10] It wasn't just convenience. [87:12] So we made a once a week GLP-1 called dulaglutide. [87:15] We stuck the protein, the native sequence, [87:17] to the backbone of basically a monoclonal antibody [87:20] to extend its life. [87:22] Novo Nordisk did a similar thing. [87:23] That became semaglutide and Ozempic. [87:25] And then we put the original two hormones together, [87:28] GIP/GLP, in tirzepatide, [87:30] which is better than those in terms of weight loss [87:33] and A1C control and everything else. [87:35] And so with those tools, we're now exploring this terrain [87:38] of what is linked to this pathway. [87:41] I would say most of what we know and have proven [87:43] is actually right on the obesity target. [87:46] So if you think overweight has these untoward, [87:50] chronic overweight, untoward effects, [87:52] our ancestors weren't chronically overweight. [87:53] They were chronically starving. [87:55] So we didn't worry about this, but now we worry about it. [87:58] Which are diabetes, type II diabetes, [88:00] not the type I form of children, [88:03] cardiovascular health, atherosclerosis, [88:05] stroke, MI, peripheral artery disease, [88:09] kidney and liver diseases, fatty liver diseases. [88:13] These are all sort of right on that target. [88:15] Adjacent to that are other conditions [88:17] we think of more unrelated [88:19] but actually have a big, [88:20] not a perfect Venn diagram, but close enough. [88:23] One is like sleep apnea. [88:24] So like 70% of people with sleep apnea [88:28] actually have overweight or obesity. [88:30] Polycystic ovarian disease is a, [88:32] young women get this and they don't ovulate, [88:34] they can't have babies. [88:35] So that's a fertility problem. [88:37] So the circle widens. [88:40] I think the two, [88:41] and there are more of those [88:43] which are being studied and looked at. [88:44] The two interesting things [88:46] which are more incidental to the mechanism [88:48] but are definitely on the mechanism [88:50] are the brain things and the inflammation things. [88:53] So if you look at people's blood values [88:56] for cholesterol or glucose, [88:58] over the course of three to six, eight months, [89:01] on the medicine, [89:02] they fall pretty straight line with weight. [89:05] But inflammation markers drop precipitously early. [89:09] There's a marker called CRP, C-reactive protein, [89:12] which is a marker for heart attack risk. [89:14] In weeks, that starts to drop, really like 60%-70%. [89:19] Why is that happening? [89:20] It's dislocated from the drug effect. [89:22] But probably the stress of effectively overeating, [89:27] not by our modern definition [89:28] but by our ancestral definition, [89:30] is causing inflammation [89:32] and reducing that by having more fasting basically, [89:36] lower calorie levels, [89:39] reduces inflammation. [89:40] We have a study reading out toward the end of this year [89:41] in chronic knee pain. [89:43] It's going to work. [89:45] And that's a weird thing is, [89:47] "Well, wait a minute, GLP-1, chronic," [89:48] but if you follow that logic train, [89:50] it actually makes good sense. [89:52] Also inflammation causing the chronic knee pain. [89:54] Inflammation plus, [89:55] and we'll measure the inflammation in the joint, [89:57] plus there's a mechanical loading with being overweight. [90:00] If you carry around a backpack of 40 pounds extra every day, [90:03] your knees will hurt more. [90:04] So, sorry— But then the brain ones, [90:05] sorry, yeah, go ahead. Well, I was just going to, [90:06] so you're just saying [90:07] that being over-sated causes inflammation, [90:11] which then causes [90:12] all these ostensibly unrelated downstream issues. [90:17] Yeah, and you see this in chronic inflammatory diseases. [90:19] So think of, well, the signature one is a skin disease [90:22] called hidradenitis suppurativa. [90:25] A terrible name. [90:26] But people get basically boils, [90:29] and it's almost completely correlated [90:31] with excess body weight. [90:33] And we have very expensive inflammatory drugs [90:36] that have fancy targets [90:37] and are monoclonal antibodies you inject [90:40] and they cost $4,000 a month [90:42] or you can just lose weight. [90:43] And so people are using Zepbound our drug in this [90:46] and they don't have this condition anymore. [90:48] Another one which has high correlation [90:50] is psoriasis actually. [90:52] So we're doing a study [90:53] with our psoriasis drug Taltz and Zepbound. [90:55] That will read out this fall [90:57] and I'm certain will show a boost in efficacy [91:00] with the weight loss. [91:01] And that is not because of the weight loss per se. [91:04] It's the inflammatory effects. [91:05] So it's pretty interesting. [91:07] And there's a lot of these adult diseases [91:08] that are inflammatory in their root. [91:12] Of course, we'll study RI. [91:13] We have studies going in Crohn's and colitis as well. [91:15] So the brain stuff. [91:17] So clearly these drugs go to the brain [91:20] and they signal the brain. [91:21] I think the scientists say the major mechanism of that [91:25] is actually not in the brain, [91:26] but there's a part of your brain stem [91:28] exposed to the blood system, like your ganglion root, [91:31] and it is detecting these hormones on purpose, [91:35] which are satiety, [91:36] but the signal being communicated to your nerve cells [91:39] isn't as pristine as, like, "Stop eating pure sugar," [91:42] which is what GLP-1 should be saying, [91:44] or GIP, which is more of a lipid mechanism. [91:47] It's just saying, "You are satiated." [91:49] That signal gets translated into downregulating dopamine [91:54] and the desire for dopamine. [91:56] And so like cigarette smoking drops [91:58] precipitously. effects. [91:59] Yeah, right. [92:00] And opioid use disorder will test, [92:02] alcohol drops precipitously. Shopping, everything. [92:04] Yeah, your cheeky pint. [92:06] It'd have to be a zero alcohol pint. [92:10] Shopping, gambling. [92:11] These things have been observed anecdotally. [92:13] These studies are being spun up and worked on. [92:15] The category of brain, [92:17] which may be related to this axis [92:19] or there's another theory of, which is sort of, [92:21] I know people out here like to do podcasts about, [92:24] "Hey, I take microdoses of Zepbound [92:27] and I feel like I can code longer," or whatever, right? [92:31] Back to your pieces of experimental hacking your body. [92:35] There is a— [92:36] I'm sure there are people in Silicon Valley [92:37] experimenting abundantly with this. [92:38] There is a theory of this [92:40] that, actually, the glucose lowering mechanism, [92:42] and your brain only eats glucose, [92:44] it doesn't eat any other substance, [92:46] it's unique tissue that way, [92:48] of having ketonic, kind of low glucose chronically [92:54] actually improves brain acuity. [92:57] And so people are probably experiencing this. [92:59] They might be lean already [93:00] and taking a GLP-1 and getting leaner, [93:02] but they're actually, [93:04] they feel their brain functioning in a sharper way. [93:07] There's an interesting study that Novo Nordisk is doing [93:09] with their oral form of semaglutide [93:11] that'll read out pretty soon [93:13] in patients who have early dementia. [93:16] That may work and it may work for the same reason. [93:19] It may also work because you reduce strokes, [93:21] which is back to the cardiovascular axis. [93:24] But, yeah, I think we've stumbled upon here [93:27] kind of a broad footprint, a broad impact zone, [93:31] of what these types of medicines can do. [93:34] And there's many more coming. [93:35] So neurodegenerative disease, Crohn's disease, [93:38] psoriasis, cardiovascular disease. [93:40] Joint disease. Joint disease, exactly. [93:41] All the diseases. [93:43] What fraction of the population, [93:46] let's say the population over 35, [93:48] will be on a GLP-1 in 15 years? [93:53] Well, today in the US, [93:54] we probably have 10 million people, maybe 12, [93:57] if we include the compounded market, [94:00] the non-approved drugs. I was going to ask about that. [94:05] That's the official company line. [94:06] Exactly. Well, yeah. [94:08] But it's really a fraction of the adult population, [94:10] and even if you just take obesity, [94:12] it should be a hundred million. [94:14] We have a long, long way to go. [94:15] And I think we talked [94:16] at the beginning of the conversation about coverage. [94:17] That's, of course, both a real cost burden on people [94:21] but also kind of an endorsement thing. [94:24] And physicians are busy [94:25] and don't have time to write all the forms [94:27] to get it covered. [94:28] So, I mean, the popularity of the direct channel [94:31] where people with some means, [94:32] I mean, $500 a month is a big ask. [94:34] It's a car payment. [94:35] But a fair swath of the US can afford that. [94:39] Actually, the number one prescribed form [94:42] of these medications is Zepbound self-buy. [94:45] We sell more than that than our insured business. [94:47] Wow. New patient starts, [94:49] and more than all of Wegovy. [94:50] That's very interesting. Yeah. [94:52] So that's why I was, [94:53] the concept of like, "Okay, for preventative, [94:55] maybe it should be on us. [94:57] And how do we just make that cost effective and easier?" [95:01] And certainly there's a cost of your time [95:03] and shopping online and having a telehealth appointment [95:06] is much more convenient. [95:07] So the number has to go way up. [95:10] The oral project I mentioned is a key part of that [95:12] because we've literally already made billions of doses [95:16] and we are capacity constrained in some sense [95:18] on the injectable systems. [95:20] Unfortunately, there's not a good learning curve left. [95:24] We've sort of built the scaled plants. [95:26] We just have to build more of them. [95:28] We've built six or seven of these mega plants [95:30] that produce hundreds of millions of these injection systems [95:35] and we're only treating 10 or 12 million Americans, [95:38] maybe 20, 30 globally. [95:40] So to get to half a billion people globally, [95:43] that's not the path. The orals have to work. [95:44] We can't keep stamping out these. [95:46] It'll take too long. [95:46] The orals have to work, they have to be approved. [95:49] They're not going to be as good [95:50] as these multi-acting injectable hormones, [95:53] but we can probably stratify people. [95:54] If you need to lose a lot of weight, okay, start there. [95:56] And maintenance with the oral is going to be a key segment. [95:59] I would guess by the time we go generic, [96:02] it will be a large proportion of adults. [96:05] Statins got to 40 million people branded. [96:08] It's a little more than that now. [96:09] Has to be north of that. [96:10] It's interesting that all the growth in statins happened [96:12] even while it was branded. [96:14] You know, you'd expect a big bump after it went generic. [96:16] Well, you asked about consumer advertising [96:18] and I think actually, [96:20] people hate the commercial part of our business. [96:22] I have to admit, like, sometimes I dislike it. [96:25] I never watch TV except when I'm traveling [96:27] and I'll flip on CNBC in the morning while I'm getting ready [96:30] or especially on the West Coast [96:31] 'cause I'm up at like 6 or 5:30 [96:33] and I'm like, "Are you kidding me? [96:34] Like, how many ads are we running, and everyone else?" [96:38] And often it's the same drug class like this. [96:41] This cannot be productive. [96:43] I have to say, 70% of our spend now is not on linear TV. [96:47] So mostly advertising is more served up [96:49] and in your search sequence [96:51] and increasingly we're interested [96:53] in, like, generative AI optimization. [96:57] But still you do it. Why do you do it? [97:00] Because it works. [97:01] Like, it's still a productive spend. [97:03] But also promotion to physicians, [97:05] which consumers don't see, [97:06] is a big part of what we do. [97:08] And we do studies and we disseminate them [97:10] and we run education programs. [97:12] Mostly if left to their own, [97:14] and there's been studies on this [97:15] of medical inventions that are not promoted. [97:18] Yeah, they won't actually be adopted. [97:19] It's about a 16-year path to full adoption. [97:22] With the medicine promoted, it's half that. [97:25] We have an internal goal to halve that again, [97:27] to get to four years to full, whatever it is, get to it [97:31] on a global scale. [97:32] I think it's ambitious but serves a purpose. [97:33] LLMs presumably help here. [97:35] Could be, yeah. [97:36] If we can get in the, [97:37] if we can get through the problem [97:40] of convoluting the facts. [97:41] Yeah, you need to have education seminars [97:43] for the LLMs. For the LLMs, yeah. [97:44] If they weren't so trained on Reddit, [97:46] it might be a little better. [97:48] Let's train them on the New England Journal of Medicine. [97:49] Well, okay, you've mentioned Reddit. [97:51] And you just mentioned the microdosing of tirzepatide [97:55] in Silicon Valley. [97:57] Yeah. [97:58] Not recommending, not indicated. [97:59] Understood. To be clear. [98:00] This is not a promotional setting. [98:01] Do not ask your physician. [98:03] Exactly. [98:04] But my understanding [98:06] is the avant-garde Silicon Valley denizens, [98:08] you know, the frontier is really in Chinese peptides. [98:13] Yeah, frightening. [98:13] So here we're getting into the compounding. [98:15] So, of course, there's probably always been [98:18] a segment of society [98:19] that was comfortable using unapproved things. [98:24] We have a large supplement industry in this country [98:26] that has some proximity to this MAHA thing, by the way. [98:32] Most supplements have no evidence [98:33] that they're going to help you. [98:35] I take a multivitamin, [98:37] but I really don't believe in anything else. [98:40] And I think if you're eating correctly, [98:43] you should be getting better nutrition through your food [98:46] than through supplements anyway. [98:47] And it's embedded in the price at least. [98:50] I mean, you may have to buy more expensive food. [98:52] But that same tranche or the edge of that is gone to this. [98:56] What is a Chinese peptide? [98:57] It's an unapproved medicine that's never been tested in man [99:01] and is made in a Chinese lab. [99:02] It might be what you say. [99:04] This is basically the same source [99:06] for the tirzepatide compounded that people get. [99:10] People buy that from things [99:12] that look like legitimate companies, [99:14] some are publicly traded even, [99:16] that formulate them and are violating our patent. [99:19] And maybe under FDA supervision or not these plants, [99:23] most not. [99:24] I know of one that's been inspected. [99:26] It had a nasty, what we call 483 [99:28] with lots of inspection findings. [99:31] I certainly wouldn't do that, [99:32] but, of course, I run a company that does this legitimately. [99:35] My problem with those companies is less about trying to, [99:39] I like the fact that people could shortcut [99:41] the pain in the butt of the healthcare system and go direct [99:45] and we see the phenomena [99:46] of what the internet's done to commerce [99:47] could apply to health. [99:49] I think that's net a good thing. [99:50] What I don't like is they're stealing my IP. [99:52] Partly people got in this business, [99:55] the rule that guides this, [99:57] actually they technically should not be doing, some do. [100:00] The big ones don't. [100:01] They say, "Oh, we're not following that rule. [100:02] We're following a different rule, which is customization." [100:05] All these patients who need tirzepatide, [100:07] even though you can buy six different dosage forms, [100:11] they need a dose in between these six, [100:13] or, oh, the efficacy is boosted by vitamin XYZ. [100:18] By the way, we recently sent to the FDA [100:20] studies of these vitamin combinations [100:22] that show they actually augment the peptide. [100:25] So they're making a new drug never approved. [100:28] Not a good idea. [100:29] And then you have the folks you're talking about [100:31] who are served by an industry [100:34] started when I think steroids became a big deal [100:37] and the bodybuilding craze. [100:39] They're all based in like Long Beach these places. [100:43] And it's like Peptides USA, [100:46] which is the opposite of what it is, right? [100:47] It's Chinese peptides. [100:49] And they'll sell things to you [100:50] that say, "Not for human use." [100:51] Literally, that's how they protect themselves legally. [100:54] And you're injecting, [100:55] you're putting saline in [100:56] and you're putting this white powder in your body [100:58] that says, "Not for human use." [100:59] Really a terrible idea. [101:01] I know some people find success with it. [101:02] At some point through that process, [101:04] you would suggest there are several clues [101:06] as to how ill-advised it is. [101:07] Yeah, and it's not going to end well. [101:09] And there are people who've had chronic kidney failure [101:11] and permanent liver damage. [101:13] And I wouldn't do it. [101:16] The difference between that [101:17] and today buying a real thing for $500 [101:19] seems like a relatively large risk [101:22] for the cost savings we're achieving. [101:23] And I'm telling you today, [101:24] we're going to bring these prices down, [101:27] either through your insurance coverage, [101:28] which is expanding every day, [101:29] or just through competition, for sure. [101:31] And your direct stuff. [101:32] I mean, we've talked about it once or twice. [101:33] You're here in South San Francisco, [101:35] the headquarters of not only, [101:37] or, you know, the place of Genentech, [101:39] but also payments innovation. [101:40] And so you— Yes, here we are. [101:42] We're working together on the Eli Direct stuff. [101:45] Maybe you can talk a bit about that. [101:46] Yeah, I mean, talk about accidental experiments. [101:49] So this came out, back to the insulin story, [101:52] the person who was running our US business, [101:54] commercial business directly, retired. [101:56] And I didn't have a suitable candidate internally. [101:58] So I thought it'd be a good idea [101:59] to go back to that job I used to do [102:00] and be the CEO at the same time. [102:02] The team— Oh, this is very inspiring. [102:03] The team in the US [102:05] didn't find this inspiring or a good idea, [102:08] but I was, you know, kind of the business was changing [102:11] and the rise of the consumer, [102:15] communication on digital channels. [102:17] We were pretty old school at that time [102:18] and I wanted to modernize it. [102:20] And so I dug in and one of the ideas that came out of that [102:22] was, "Hey, why don't we stand up our own pharmacy [102:25] and sell directly to patients?" [102:27] And then people were like, [102:28] "Well, first of all, the existing pharmacies, [102:30] and there's only like three of them, [102:31] will hate us. [102:33] And so that seems like a bad idea. [102:35] And secondly, we know nothing about running a pharmacy [102:37] so we're going to make mistakes and hurt people." [102:40] So we kind of parked it. [102:42] A few months later, though, we were in this process [102:46] I described in long form earlier, [102:47] but in short form of, like, deescalating this insulin bubble [102:52] and getting to sort of true pricing in the market. [102:54] And one of the fights we were having [102:56] was with a large PBM company [102:59] that also owns a pharmacy chain. [103:01] And we were worried that they actually, [103:04] a different pharmacy chain actually threatened [103:05] not to carry our low-price insulin [103:07] 'cause they couldn't make enough money on it. [103:09] So I just looked at him [103:11] and said, "This is what this was for, [103:12] this idea that's been on the shelf. [103:14] We cannot be beholden to this. [103:15] We have to have a route to market ourselves," [103:18] which we had not had [103:19] since the company was founded as a pharmacy, [103:22] actually a freestanding pharmacy before any regulation. [103:25] So it was scary, [103:26] but we cobbled it together with partners [103:28] and now we do more of it ourselves. [103:30] And the first idea was, like, make sure people [103:32] who need insulin to survive [103:33] can get it at the lowest price. [103:35] We sold a little bit, but not much. [103:37] Then we launched our migraine medicine, [103:39] which was having trouble getting insurance coverage, [103:40] sold a little bit more. [103:41] Then we launched Zepbound [103:43] and we said, "Ah, this feels like the killer app [103:45] for a direct-to-patient experience [103:47] because the diagnosis step is dead easy." [103:49] Everybody knows, [103:50] everyone knows the biomarker tool in their bathroom. [103:53] It's called the scale. [103:54] They can know if the drug's working, [103:56] and we can offer telehealth post-pandemic at scale, [103:59] it was out there, a third party, [104:01] to advise patients if that was the right choice [104:03] or if the Novo product was the right choice or nothing. [104:06] And so we put that together [104:07] and, boom, this thing really grew pretty quickly. [104:10] Today, we'll annualize, [104:13] you know, in the billions of dollars. [104:15] I think it's the largest prescription platform online [104:20] in terms of revenue. [104:21] We run it on Stripe I think. [104:22] And going north from here. [104:24] And it's an interesting example [104:25] of how across so many different sectors [104:28] and, you know, across every company scale and stage, [104:31] there is this interesting way [104:36] in which all these ostensibly different models [104:38] and different businesses are discovering— [104:40] They leap, right? They leap across. [104:42] Well, they're discovering the value [104:43] of having a direct relationship with the end customer. [104:46] And, of course, I'm on the board of Adobe, [104:47] a software company, [104:48] which is, this is now in the final stage of evolution [104:52] in the software industry, [104:53] but I found it fascinating when I first arrived [104:55] that 90% of the revenue was off a website that they ran. [104:59] I thought that was, "What? [105:01] This is a great business." [105:02] A very low-cost business. [105:04] Very low-cost route to market, global. [105:06] We're very in favor of selling stuff on websites. [105:08] Yeah, exactly. I'm sure you are. [105:09] You had this insight even before for me. [105:11] And so, yeah, you know, that leap is happening. [105:14] Now, healthcare has been a hard problem for tech, I think, [105:17] which is interesting, [105:18] because you have all this bricks and mortar mess. [105:21] You have a lot of state level insurance regulation. [105:23] You'd think FinTech would've gone after this a lot sooner [105:26] 'cause it's 20% of the financial economy in the country. [105:28] Well, I think this podcast makes clear [105:30] that there are nuances to the sector. [105:32] Exactly. [105:33] I guess so, yeah. [105:34] But, you know, I think people with longer money [105:37] are starting to do it. [105:39] We actually ourselves just moved our PBM [105:42] from a traditional one [105:43] to this sort of new tech, fintech-y PBM [105:47] and we're switching all our lives [105:48] and we want more transparency, better data reporting, [105:51] and the interoperability problem that is why PBMs rose, [105:54] which was that you had a card, [105:56] like a physical card [105:57] that had your insurance number on it in 1992 [105:59] and you went into a pharmacy in Phoenix [106:01] versus one in LA [106:02] and you couldn't get your prescription filled, [106:04] they solved that problem with old tech. [106:07] That's a pretty easy problem [106:08] to solve these days with technology. [106:10] But what they built on the back of that [106:12] was a system of negotiating and capture, rent taking, [106:17] that's not so popular anymore [106:19] and we can disintermediate them easily. [106:22] We've been having this conversation for quite some time [106:25] and we haven't asked you— [106:26] With one beer. [106:28] Well, we can rectify that, [106:30] but we haven't asked you maybe one of the first questions [106:34] that we ought to have asked you, [106:35] which is Eli Lilly is the largest pharma company [106:38] in the world. [106:40] Why? [106:41] Well, in simple terms, [106:42] we are kind of a rare situation right now [106:47] in that our growth rate is high [106:50] and our profitability is expanding [106:52] and we are in an early cycle of this invention. [106:56] I think Wall Street believes— [106:57] This being? GLP-1s, [106:59] which is driving probably 80% [107:00] of the economic value of the company. [107:02] Our market cap is about— You think Eli Lilly [107:04] is a GLP-1 company [107:06] with a sidecar— Sidecar. [107:09] Of some other of some other stuff? [107:11] Yeah, that's probably trading like other, [107:12] okay, so in our sector today, [107:14] let's pick a company like Bristol Myers or Pfizer. [107:17] These are big companies [107:18] with revenues not so different from ours [107:21] and we compete with them in these other spaces. [107:23] Their market cap's between $100-$200 billion. [107:26] We're trading about $800 billion, [107:28] and that difference is the GLP-1 phenomena. [107:32] I think Wall Street also believes [107:33] that our R&D productivity has been higher. [107:37] So every dollar we put through the income statement for R&D [107:40] or through an acquisition, [107:41] we get a little bit of a premium, a management premium on. [107:45] I think most of the sector is treated the opposite way, [107:48] which is that that's actually probably going [107:49] to destroy value in some way. [107:53] And I think the other thing that's out there [107:57] is this belief that perhaps, [107:59] for those that are really long our stock, [108:01] our belief that perhaps this cycle could be different, [108:05] this cycle starting with GLP-1s, [108:07] but that you could create, [108:09] back to the route to market and the consumer, [108:11] much more of a self-pay branded business [108:15] that has staying power beyond— [108:18] Franchise value. The patent cycle. [108:19] Franchise value. Thank you. [108:21] And I think so far, the evidence is pointing that way. [108:24] Have we fully evolved to a mature version of that? No. [108:27] Have we created like an ecosystem around ourselves [108:30] like Apple has done? [108:31] No, no. [108:31] Those are all opportunities for us, [108:33] but you can kind of see them. [108:35] And self-care is an innate desire, [108:37] and I don't think the payment system's [108:39] going to fully cover all this, [108:41] but a lot of people are willing to pay [108:42] and it's not just the US. [108:43] It's a global phenomenon. [108:45] You said GLPs are one of the biggest drivers [108:47] of the business. [108:48] Eli Lilly's growing at about 30% right now revenues? [108:52] 40% year-to-date. Oh, wow. [108:53] We'll have earnings on, [108:54] yeah, actually here's a fun fact. [108:56] There's three scaled large cap companies [108:59] that have a rule of 80. [109:02] Can you name them? [109:05] Stripe's doing pretty well, but— [109:06] NVIDIA must be— [109:08] NVIDIA is the highest. [109:09] I think they're over 90. [109:11] Margin plus growth. [109:13] CoreWeave? [109:14] I think that's, [109:15] considering you're talking the CEO of Eli Lilly, please. [109:18] So top hundred market small caps out there. [109:20] Okay, okay, yeah, yeah, yeah. [109:22] So it's based quite close to where we're sitting. [109:25] Genentech? Broadcom. [109:27] Of course. Yeah, of course. [109:29] So the hardware guys and AI are killing it. [109:31] But I think what's interesting to me, [109:32] they're trading at multiples above ours, [109:35] there's a belief that their cycle [109:36] is somehow longer than ours. [109:38] And I think tirzepatide's US patent is late '30s, [109:42] orforglipron, the oral, beyond that. [109:44] So, yeah, that's my pitch to investors. [109:47] But we're in that club as well. [109:50] Will we stay there forever? [109:50] Obviously no, [109:51] but, you know, I think that's one of the reasons [109:54] we're worth more. Well, I was going to ask [109:55] some questions about that. [109:56] So first off, Novo is growing in the teens. [109:58] For two companies with GLP offerings that are working, [110:01] why are those growth rates so different? [110:03] We're taking most of the growth in the market. [110:06] Okay, so just [110:07] your product's working better. It's a share. [110:08] Right now in the US, across all forms of GLP-1, [110:12] on new patient capture, [110:13] we're basically 70%-75% right now. [110:17] So almost three to one. [110:20] And then if you, it's a high carryover business. [110:23] I think we're 60/40 on the total. [110:26] And so we're just getting most of the growth. [110:29] What would you guess Eli Lilly's P/E is? [110:32] Forward P/E? [110:34] I was using trailing, but. Most of the time, yeah. [110:37] 15? [110:38] 50. 5-0. Oh my god. [110:41] So what I was going to ask is, [110:44] you know, we're talking here about— [110:45] The sector's like 12. [110:46] So, yeah. No, exactly. [110:47] Yeah, you're correct. [110:48] Like very close. [110:50] It was a very good guess. [110:51] I was thinking high end of the sector. [110:53] Your priors were good [110:54] because that is correct for the rest. [110:55] But where I was going with this [110:56] is if you were to listen to this podcast, [110:59] I think you maybe come away thinking, [111:00] "Wow, pharma is hard. [111:02] Like, good god." You know, there's so many things [111:04] and, you know, things roll off patent [111:05] and we have the Chinese competitors and things like that. [111:07] So what is it that investors have confidence in? [111:11] Well, I think the track record of success. [111:13] We've been on a growth curve for 12 years or so. [111:16] It's certainly gone a little more hyperbolic lately. [111:19] But I think that builds confidence. [111:22] I would hope some management piece. [111:25] But also, you know, the ability [111:27] to predict where to move. [111:32] And I think if you say, "Okay, what's your recipe?" [111:34] It's an R&D business. [111:37] Everything else is around the edges. [111:38] So you have to create something better for people [111:41] that improves their health. [111:42] If you can do that, you're going to win. [111:44] Policy, this and that, the commercial strategies, [111:46] that's the 20. [111:47] The 80 is this. [111:48] And I think we do three things better than others. [111:51] One we talked about already, which is cycle time. [111:53] It's a basic concept, [111:54] but if you can make software faster than someone else, [111:57] you're going to win. [111:58] And the same in the drug business. [112:00] The second is prediction of where to tack the investment [112:05] and allocating a meaningful part [112:08] to ideas that may not be obvious today, [112:11] but actually are big problems without markets. [112:15] And we're drawn to those. [112:17] That is the third box— [112:18] As in without preexisting markets. [112:20] Yeah, I mean, there's illnesses, [112:22] but they're not medicines. [112:24] And I think a lot of companies don't work that way. [112:27] They look at, "Okay, where's our payment? [112:28] Where can I recoup my investment?" [112:30] Versus where is there a problem? [112:33] And maybe our situation [112:34] allows us to— Jensen Huang at NVIDIA [112:35] talks about how he loves $0 markets [112:38] and you're describing some [112:39] of the same things. Exactly. [112:40] Blue ocean things that are, [112:42] you know, there's no limit to human disease. [112:43] And actually the longer we help people live, [112:45] the more diseased they'll be. [112:47] So in a way, it's alike AI in that way [112:49] where it's like AI begets more AI. [112:52] It's just this growing machine. [112:54] And then I think discipline of the allocation [112:57] between the types of R&D that are extending the franchise, [113:02] these moonshots we just talked about [113:03] that could really actually create a new GLP-1-like category. [113:08] You know, we're doing this study now [113:09] that I think will be quite interesting [113:10] that is going to potentially show [113:12] you can slow Alzheimer's before it starts. [113:14] That's the kind of thing that could be a mega market. [113:18] And those have to not just help a lot of people, [113:20] but they have to save a lot of money for healthcare systems [113:23] in order to generate, I think, shared value. [113:26] And then you have to do the discipline [113:28] around the edges of, you know, the next clinical trial [113:31] for a cancer drug that's already working. [113:33] I think being multimodal in that [113:36] and really kind of balancing your bets [113:38] has been a key for our success. [113:40] Capital allocation I guess in a sense. [113:42] When people are debating drug pricing, [113:46] and they debate it a lot, [113:48] obviously the argument made [113:51] by people in the pharma industry, [113:53] which to be clear, I think both of us believe [113:56] has a lot of legitimacy, [113:57] is, "Well, if the returns aren't favorable, [114:00] we're not going to pursue the investments." [114:02] You'll never know if it was too expensive. [114:04] Exactly. Yeah. [114:05] Since society wants more drugs and not fewer drugs [114:07] and there are many diseases that have not been cured, [114:10] et cetera, et cetera. [114:11] And we've, of course, discussed these dynamics [114:13] extensively here just now. [114:15] In as much as GLP-1s represent [114:17] this enormous advancement and improvement [114:21] in Eli Lilly's just fundamental financials, [114:24] I mean, it's economically equivalent [114:27] to a 2x, 3x, 4x increase [114:30] in the realized drug price [114:32] of every drug across the board, [114:36] does that mean that people should now expect Eli Lilly [114:39] to be far more able to fund broad-based drug R&D [114:43] than it was in the past, [114:44] such that their estimate [114:45] for the prospective drug development [114:48] and, you know, bounties of drug discovery [114:50] over the next 10, 20, 30 years should be much larger [114:54] than the pre GLP-1 world? [114:57] I think so. [114:58] Although, we have to prove we can. [115:00] So we're going to try. [115:02] I have a belief that if you're not generating, [115:05] to generate double digit growth in the sector, [115:07] you need to invest at 20%-25% of sales in R&D. [115:10] That's sort of a positive return R&D stack. [115:14] And we're doing that. And you plan to hold [115:16] that ratio as revenue. Yeah, and this year, [115:18] we'll sell 60-some-odd billion. [115:21] So that's how you get to the 14 billion, [115:23] and next year, that'll grow by double digits, [115:26] and so we should try. But do you think, [115:28] if revenue goes to 120, you know— [115:30] I would try to spend 20% of that, [115:32] which which would then approximate the NIH. [115:35] And I think that's a frightening thing [115:37] because, well, how many ideas are out there? [115:40] So I think, of course, we should do more [115:42] of what we're good at. [115:43] We should be bounded by our own capabilities, [115:47] and we don't know everything about every disease, [115:49] we don't know everything about every modality, [115:51] but for the diseases we know the modalities, [115:52] we can fund more [115:54] and pursue a healthy portion of these bigger bets [115:58] on zero-value markets. [116:01] We can expand the franchise. [116:02] We can do the incremental things we have to do at scale, [116:05] and do them earlier actually. [116:06] I think that's quite an important thing. [116:09] Often we do serial clinical trials, [116:11] and by the end of the product lifestyle, [116:13] you get the final indication. [116:14] We're trying to stack them all into the beginning. [116:16] So that's expensive, but we're in a position to do that. [116:18] It's not risky actually. [116:20] So you're saying anyone who's on this, [116:21] you know, who's teetering on the brink [116:24] of purchasing a GLP-1, [116:26] especially in a self-pay mode or something like that, [116:28] and they worry maybe it's self-indulgent to do so, [116:31] you know, they should just go to the gym [116:32] or, you know, exercise willpower [116:34] and so forth. They're paying for medicine [116:35] that's going to help someone else. [116:35] You're saying that the purchase of this GLP-1 [116:38] is also a kind of subsidy— Yeah, I mean, [116:39] back to the pricing. For cancer R&D. [116:41] And just to add on the R&D, [116:42] the other thing we're doing at some scale [116:43] is actually trying to create a ecosystem [116:46] around us of invention [116:49] that we can aid in a real way, [116:50] not just own part of it. [116:52] We have lots of deployed corporate venture. [116:54] But we've built these things called Catalyze360 [116:58] or the Gateway Labs here in South San Francisco actually [117:01] where we host scale-ups, not startups. [117:03] And by host, I mean we offer our services. [117:06] So rather than hire some consultant [117:09] who retired five years ago [117:10] to help you with a particular problem, [117:12] we'll give you someone working on it right now. [117:14] And so it's sort of a loosely coupled model [117:17] without buying them. [117:19] Often the entrepreneurs leave, [117:21] we cultivate them and then maybe buy them if they're good. [117:24] You have to apply to get in. [117:25] So it's a competitive process. [117:27] We have I think seven or eight of these now [117:29] around the country [117:29] and even two in China and building one in London. [117:33] And we'll create a virtual version of this. [117:35] And that TuneLab tool, that AI tool I talked about earlier [117:38] is embedded in that as well. [117:39] So that's another way we can spend money in R&D [117:41] with other people [117:43] and kind of use other brains to develop what we're doing. [117:47] But I have to caveat, [117:49] this may not work actually, right? [117:51] There could be a frontier in what's possible [117:55] and we found it [117:56] and beyond that it's just waste. [117:59] I don't rule that out, [118:00] and I think it's an important caveat. [118:02] We'll know those signals in the next couple years [118:04] 'cause our scale is getting to be bigger [118:06] than anyone's ever done. [118:07] Most of those experiments that have been run ended badly. [118:10] People mostly bought other companies at too high a price [118:14] and then the drug didn't work out. [118:16] We're not doing that. [118:17] But, you know, it'll be interesting to see. [118:20] And then, you know, if we quit, [118:22] we'll turn into like an Apple. [118:24] We'll just start buying back massive amounts of shares [118:26] and return cash to shareholders who invested [118:28] to create this surplus. [118:30] But it's important what you're saying about people buying [118:33] is that a quarter of every dollar you spend [118:36] is going to a research lab or a clinical trial [118:40] for a medicine you might not need [118:41] or for someone you don't know. [118:43] But that's the system by which we create new medicines, [118:46] and we'll try to use that wisely. [118:48] I don't take that responsibility lightly. [118:50] That's someone's money. [118:51] But maybe they'll have Alzheimer's someday [118:54] and we'll have a solution for that, [118:55] or maybe someone they know. [118:56] But that's the virtuous circle we try to drive. [119:00] Is it meaningful to talk about what fraction of R&D [119:02] is towards specific treatments? [119:04] Is, like, focused vertical R&D [119:07] versus I presume you do a lot of horizontal R&D. [119:09] Yeah, platforms. [119:10] Yeah, exactly, platforms. [119:11] Because presumably it's even harder [119:13] to reason out the platforms and the payoffs there. [119:14] Well, there you need some scale. [119:16] There's a lot of platform companies [119:18] that get funded that are biotech. [119:19] Yeah, we have them in San Francisco. [119:20] And they are usually exploring a new platform [119:24] that's quite helpful. [119:25] That's important work they do. [119:26] And often we'll partner with them early [119:28] and try to develop that capability ourselves. [119:32] Basically, in our business, [119:34] there's two kinds of questions on early phase R&D. [119:37] One of them is this, is there a new platform [119:39] that can unlock targets we already know about [119:41] in new ways or in better ways [119:43] that create a whole field of drugs? [119:45] If you think of Genentech, [119:47] like, that was a company that exploded [119:49] based on monoclonal antibody technology [119:50] by tricking cells to make a human antibody [119:53] that solved disease. [119:55] 30-year run of spectacular new medicines. [119:58] Or Gilead Sciences nearby, [120:00] which really started on this idea of virology [120:02] and new virology chemistry and small molecules. [120:05] So we want to be there at the early stages [120:08] because it is like a catching a wave thing. [120:11] If you're late, you miss it all. [120:13] And so that's a kind of investing we do [120:15] and that's a more scaled project. [120:17] And then the other kind is, like, picking targets [120:20] and looking in the broad space of biologic discovery [120:24] and say, "Okay, of the thousand things uncovered this year, [120:27] these 15 we think could be highly relevant [120:31] and we're going to put a team around those." [120:33] This isn't the Skunk Works kind of allowable deviation [120:37] or whatever I said earlier. [120:38] But it's a purposeful thing to say, "Let's drug hunt here. [120:42] Let's use the tools we have and assault those targets [120:45] with multiple ones of those tools [120:47] and see if we can get a drug out of it." [120:48] And sometimes we've actually come out [120:49] with like a small molecule and an antibody [120:53] and an siRNA. [120:53] Like we talked about Lp earlier. [120:56] That was the case there. [120:57] We discarded the antibody, [120:58] there is one that's been developed, [121:00] and we went after the small molecule and the siRNA. [121:02] They're both in phase III. [121:04] And so that was a case of very purposeful find a target, [121:07] attack the target, get a medicine to market. [121:10] So it's both. [121:12] And we need to hedge it that way I think. [121:15] And then, of course, we watch the outside [121:16] and sometimes we miss those two signals [121:19] and we end up buying companies later in their cycle [121:21] and saying, "Hey, we can add value through clinical trials [121:23] or manufacturing scale or something else." [121:26] Last question. [121:28] Eli Lilly is more than a hundred years old and— [121:31] 150 in May. Oh, wow. [121:32] Okay, so coming up on your 150th birthday. [121:35] And I noticed that often [121:39] very tenured, successful companies [121:42] are quite serious about, and good at, [121:45] internal succession planning. [121:47] I think about, you know, Royal Dutch Shell [121:48] or, you know, companies like that, [121:50] and Eli Lilly. [121:51] You joined in what year? [121:53] 1996. [121:54] Right, you joined in 1996. [121:55] Not as a hired CEO. [121:57] No, I was a BD, M&A. [121:59] Exactly. New hire. [122:00] And you were rotated across, you know, the business. [122:03] You ran China, you ran the US business. [122:04] Development. Exactly. [122:06] All these kind of roles. [122:07] What do you think Stripe, Silicon Valley companies, [122:10] should learn from Eli Lilly [122:12] and, you know, companies like Eli Lilly, [122:14] but that's where you have experience, [122:16] about talent planning and talent development? [122:18] Yeah, fabulous question. [122:19] I do notice differences, [122:21] at least on the board I'm on and observing other companies. [122:25] Some of that might be just the clock speed of the industry [122:27] and the technology [122:29] and some of it might be the newness of companies, [122:31] because if you haven't really seen the cycles play out, [122:35] hard to kind of see the value. [122:37] You got a problem, solve it, [122:39] you know, work on the next thing. [122:40] But probably even in your company, [122:42] which has been around long enough now, [122:43] I'm sure you have people [122:44] who are, you know, single-digit hires, [122:47] you know, first few people, [122:48] who've really been excellent [122:50] and what they're doing now [122:51] is nothing like what they started doing [122:53] and you should examine, like, what are those, why? [122:56] Was it the experience path they took? [122:57] Was it innate traits they have? [122:59] Combination of those things? [123:01] You know, we've had 150 years of that [123:04] and I'm the 11th CEO of the company. [123:08] That's one less than popes in that period of time. [123:12] So it is a special honor actually, [123:15] and it's not a lifetime appointment. [123:17] I can be fired any day. [123:19] But the first four were family members [123:21] and then we've had a lot of long-running successful [123:24] and only one external. [123:26] Really? Yeah. [123:27] And I think that's part of the success of the company [123:32] is that in scaled companies, [123:33] and we've scaled for a while, [123:35] you know, not one person cannot possibly [123:38] really lead the whole thing. [123:40] You have to know the role you have to play [123:42] and you have to have others around you that can do it. [123:45] By creating that environment, [123:46] giving up some of that, [123:48] you actually grow people [123:50] and you grow people in a special way. [123:53] A way that they know how to operate [123:55] in the unspoken operating system called culture. [123:58] And so they're more effective more quickly in new roles. [124:01] And they just know the domain, right? [124:03] 'Cause there's so much to know. [124:04] They know the domain. [124:05] They also know the human domain [124:07] of how to solve problems without committees. [124:10] And, like, one of my things now as we grow so fast [124:13] is, like, keep headcount flat. [124:15] And that's makes me very unpopular [124:19] 'cause people are like, "What? [124:20] Like, how do I get this work done?" [124:22] We are growing headcount in manufacturing. [124:23] That's a unit operation business. [124:25] Are you growing in R&D as you grow the spend? [124:27] Slightly but, [124:28] so we're growing R&D high teens, low 20s. [124:32] We're growing headcount in R&D single digits. [124:34] So where does the money go? [124:36] So the money goes to projects. [124:38] Yeah, and salary. [124:40] Trials and tribulations? [124:41] I believe in paying people well. [124:41] Yeah, so there's salary growth, [124:43] but clinical trials, new equipment, new laboratories. [124:47] Supercomputers from NVIDIA, that's expensive. [124:50] So we're really trying to keep that goodness [124:53] that can come from having most of the succession internal. [124:57] What I noticed when I took over, though, [125:00] was maybe that took over too much. [125:02] There is a balance. [125:03] I have tried to bring in at the leadership level [125:06] and other ranks, compete jobs externally, [125:09] and bring in outside voices that are the minority voice [125:13] and have to have those innate traits [125:16] and kind of a culture fit that can work, [125:19] but it stimulates you in ways that sometimes you're like, [125:21] "Yeah, that felt a little uncomfortable [125:24] 'cause they came at it in a different way [125:25] or said it in a way that doesn't connect [125:28] with how we normally, [125:29] but actually they're making a good point." [125:31] So that's the blend we've tried to find. [125:34] And I look at my career [125:36] and probably four or five times I was put in a job [125:38] I had no business being in, [125:40] but somebody thought I could learn it [125:42] and that the output of that [125:43] would be good performance at the end, not at the beginning, [125:46] and then a better long-term thing for the company. [125:49] I'm so grateful for that [125:50] because that's like ultimate risk taking on people [125:53] and I would not be here without those successive jobs [125:56] where I was like, I never would've gotten them [125:57] if I applied externally, [125:59] but the company gave them to me. [126:00] And as a CEO, you have to do a lot of things [126:03] that are very horizontal [126:05] and I touch things that I had no experience in, [126:08] but I learned by graciously reading [126:11] and doing and solving real problems [126:13] that has made me more successful in this job. [126:16] A lot of companies say they're long-term oriented, [126:17] but I feel like this is a particular example [126:19] of revealed preference. [126:21] Yeah. [126:22] David, thank you. Awesome conversation. [126:23] Thanks for the beer. Yeah, great. [126:24] Thank you very much. Yeah.