Adam May谈$ABVX的超预期数据及随后股价崩跌
- Abivax($ABVX)的3期维持期数据不仅达到最乐观情景,还超出了此前无人建模的情景。50mg剂量的临床缓解差值为40%,25mg为39%,对比Rinvoq的39%/30%;在胃肠科医生最看重的终点——内镜缓解——上,obefazimod的38%差值超过Rinvoq 18%的2倍。Adam May表示:“这一结果甚至没有出现在这些数据任何情景分析的刻度上。”
- 6月1日盘后,股价从约130美元飙至175–185美元,随后因一张显示50mg组7例癌症、安慰剂组1例的表格跌破100美元,XBI也一度跌至约68美元;May称这一反应“明显错误”。其中1例是结肠异型增生(“按定义不是癌症”,很可能是编码错误),4例是非黑色素瘤皮肤癌,通常通过一次约15分钟的局部麻醉手术即可治愈,剩下1例前列腺癌和1例乳腺癌——“毫无疑问是统计噪音”。
- 将运行了7年的2期数据(1例癌症)与3期数据合并,并保守地把2期病例归入50mg组后,发生率为每100患者年0.59例,对比UC患者约0.5例的背景发生率。药物不存在致癌机制——obefazimod不具基因毒性,且重度感染在安慰剂组比50mg组更常见,因此“毫无疑问”不是免疫抑制剂;两例真正的癌症发生在不同器官,也不存在病例聚集。
- May将崩盘归因于管理层而非分子本身:公司披露的癌症病例细节超出行业先例要求,还“天真地”以为市场会按他们的方式解读。股价暴跌后,公司把Part 2安全性数据从原定10月提前至6月底前公布,涉及约500–600名额外3期患者,其中数百人已接受50mg治疗1年;按背景发生率,这一人群出现约3例新癌症属于正常范围,更多病例则可能“再次吓坏市场”。
- 在约100美元(约70亿美元)市值水平,May认为市场已经计入黑框警告;但即便是最坏情景——黑框警告、只做UC——这仍是一款数十亿美元级药物,因为Rinvoq在IBD领域峰值销售额约40亿美元,身背5项黑框警告且需要实验室监测。最乐观情景是没有黑框警告,并在2027年年中拿下克罗恩病,峰值销售额为80–100亿美元,“几乎超出一家药企收购另一家公司的能力”。
- May认为存在多个竞购者,并相信有数家药企可能感兴趣。Walker给出的乐观区间是短期完全摊薄估值约160–180亿美元(每股175–200美元);如果克罗恩病数据先出来,股价可能上探至高200美元区间,甚至约300美元。他引用了Cidara 110%的收购溢价,以及Abivax CEO在一次糟糕的读数后6周卖掉CinCor的先例。
- Nektar在没有任何新消息的情况下,从92美元增发价一路漂到50多美元,May对此“无话可说”,但指出3项压制因素:在Eli Lilly陪审团审判于夏末或秋季落槌前,专业投资者都在观望;竞争对手Q32 Bio即将公布受到市场热捧的开放标签脱发数据,但其IL-7靶向机制带来类似SCID的安全性担忧;Sanofi也尚未放弃amlitelimab。Walker表示,Sanofi仍将amlitelimab定位为数十亿美元级药物,尽管他认为该药最终会在特应性皮炎适应症上获得黑框警告。被问到二选一时,May选择Abivax,理由是“我对当前价格错误的确定性更高”。
1. 基本盘:疗效才是问题,安全性原本被视为已完成去风险
- 上一期回顾显示,诱导期数据(8周疗效)已经达标——撇开跨试验比较的限制,这可能是溃疡性结肠炎历史研究中疗效第3高的药物——剩下的维持期读数(再观察44周)主要回答疗效是否持久以及安全性问题。回头看,May当时的提醒格外醒目:“即使观察1年,对于判断一种药物的安全性来说也相当短。”
- 试验开始前,安全性看起来已经基本完成去风险:已揭盲的数据安全监察委员会(DSMB)分别审阅了试验80%和90%的数据,没有发现新的信号。正如Walker所说,DSMB意外叫停试验是“每一位生物科技投资者的噩梦”。
- 一切比较都围绕Rinvoq展开。这款JAK抑制剂的临床缓解差值为39%/30%,但黑框警告覆盖癌症、严重且致命的感染、血栓、心脏病发作/卒中和重大不良心血管事件(MACE),还需要实验室监测;Walker打趣说,“除了人为因素,它几乎涵盖了所有能要你命的方式”。市场对Abivax的共识是20–30%差值,May猜25%,而“3字头”是最乐观情景。
2. 四字头:疗效超出最疯狂的多头情景
- 结果显示,50mg剂量的临床缓解差值为40%,25mg为39%,而对应剂量的Rinvoq为39%和30%——“此前没人认为这是可行的结果”。
- 内镜缓解是胃肠科医生最看重的终点,因为结肠镜显示完全没有病变是客观结果,而问卷驱动的缓解终点“噪音更大”。Abivax的38%差值超过Rinvoq 18%的2倍。“这一结果尤其超出最疯狂的多头梦想……甚至没有出现在任何情景分析的刻度上。”
- Walker从普通投资者角度解释了这笔交易的分量:未经控制的UC最终可能走到结肠切除术。May没有粉饰后果——腹部开一个新孔,余生挂着结肠造口袋。一款能让结肠完全愈合的药“会改变你的生活,也会延长你的生命”。
3. 随后崩盘:7例癌症将股价从185美元打到100美元下方
- 6月1日16:05发布的公告称“未发现新的安全性信号”;股票恢复交易后从约130美元冲至175–185美元,随后跌到160美元、150美元,穿过110美元并跌破100美元。May最初看到公告中的病例时,觉得它们经得起常识检验——“大型3期研究里经常会出现癌症病例”——直到不断收到消息,得知50mg组有7例癌症、安慰剂组有1例。
- 抛售剧烈到拖累整个板块:XBI次日下跌约4–4.5%,一度跌至约68美元——“已经不只是向下过冲”。
- May认为管理层在披露方式上“处理得非常糟糕”:其他UC药物公司,即便产品带有癌症黑框警告,也不会以如此细的颗粒度通过新闻稿公布癌症病例。“他们想当然地认为市场能迅速解读”专家一眼就能看出的结论:这不是癌症风险。
4. 7例变2例:异型增生、皮肤癌与皮肤科医生的判断
- 第1例是结肠异型增生——“按定义不是癌症”,而且是炎症结肠中常见的癌前病变——很可能因编码错误被归入肿瘤类目;更糟的是,脚注还原样写成“结肠癌。完全错误”。
- 另外4例是非黑色素瘤皮肤癌。作为皮肤科医生,May“每天要处理10次”:美国每年约有500万例,比发生率第二高的癌症多15–20倍,几乎总能通过一次约15分钟、局部麻醉下的手术治愈。FDA甚至把非黑色素瘤皮肤癌作为独立终点追踪药物的癌症风险。Walker也提供了现实中的参照:他父亲去年在鼻子底部切掉一个,“没什么大不了”。
- 这样剩下2例“真正的”癌症——1例前列腺癌、1例乳腺癌,正是全球最常见的2种癌症——而且来自一个接近200人的组。Walker说:“实际上就是各1例。”May认为,“这毫无疑问是统计噪音”,如果一开始就把7例拆解成2例,可能完全不会发生这次崩盘。
5. 没有机制、没有聚集:市场跳过了患者年发生率计算
- FDA考察的两条致癌路径在这里都对不上:obefazimod早年就经过研究,已证明不具基因毒性;重度感染在安慰剂组比50mg组更常见,因此“毫无疑问,这款药不是免疫抑制剂”。病例也不存在聚集,致癌药物可能集中诱发某一类肿瘤,而乳腺癌和前列腺癌来自“完全不同的器官”。
- 直接比较用药组和安慰剂组本身就是方法论错误:试验的统计效力是为缓解差值而非罕见事件设计的,而且不良事件的捕捉极不对称。约80%的用药组患者完成试验,安慰剂组则大量退出;按Walker给出的数字,完成OB50试验的是约160人,安慰剂组仅66人。如果一名退出试验的患者在6个月后被诊断出肺癌,“我们永远不会知道”。
- 正确的比较对象是背景发生率:UC患者每100患者年约出现0.5例癌症。反直觉的是,最大的安全性数据库来自运行了7年的2期研究——数百个患者年里出现1例癌症。将其与3期合并,并保守地把2期病例归入50mg组后,发生率为每100患者年0.59例,“完全在正常范围内”。但“周一晚上盘后,大家显然没准备好深入看这些数据”。
6. Walker反问:市场已经消化一周,为何还没重估?
- Walker明确提出了空头逻辑:这是“生物医药圈讨论最多的名字”,卖方研究铺天盖地,银行在消化2天后已经下调评级,而股价现在是100美元而不是130美元——“你真的要在这里寻找超额收益吗?”May的回答是:“如果我们相信市场有效,现在谁也不会坐在这个播客里”,而且“每当市场面对数据的深度分析时,我总是一次又一次惊讶于它能忽略多少东西”。
- May给出的依据是:诱导期数据公布前,他曾发表计算,认为这项试验“接近你能得到的10倍上行保证”;数万人读过那篇分析,股价却连续48小时下跌,随后数据公布,带来约1,000%的上行。Walker的反驳是,公司的体量现在大得多,“盯着你的人也多得多,我的朋友”。
- Walker引用Jefferies的观点:“好药不总是好股票”,而且JAK差异化优势已经“收窄”,这引发了May最强烈的反驳:数据公布前,仅仅追平Rinvoq就是最乐观情景,所以“这让人感觉他们成了股价走势的囚徒,而不是在真正评估基本面”。
7. 即便最坏是黑框警告,仍是数十亿美元级药物
- 这笔交易的核心只剩2个变量:是否获得黑框警告——“警告与注意事项”说明本质上等同于药品标签上什么都没有——以及克罗恩病能否在2027年年中读数中成功。最坏情景是黑框警告、只做UC;Rinvoq已经证明市场空间存在,尽管疗效更差、带有5项致命风险黑框警告且需要实验室监测,IBD峰值销售额仍约40亿美元。用100亿美元收购一款峰值销售额20亿美元的药,“至少能收回成本”。
- 最乐观情景是克罗恩病成功且标签干净:UC峰值销售额40–50亿美元,克罗恩病再贡献40–50亿美元,合计“一款80–100亿美元的药……更像进入合并领域,而不是单纯收购”。UC疗效的幅度本身也支持上调克罗恩病的成功概率。
- May认为,在100美元股价上,“市场已经计入黑框警告”,即使假设最终确有黑框警告,也完全有人有理由按每股130–150美元收购。Walker的交易结构直觉是:“这是最适合设置CVR的一笔交易”——分别挂钩克罗恩病数据和标签结果。
8. 第二部分:数周内补充500–600名患者安全性数据,以及增发问题
- Part 2覆盖约1,300名启动试验、但未纳入主要维持期分析的患者中的约550–600人,其中数百人已经接受50mg治疗1年,因此其主要价值在于补充安全性数据。数据原定在10月的学术会议公布,崩盘次日被提前至6月底前:“这看起来不像是在藏数据,他们只是确实没想到大家会担心。”
- 按该人群的背景发生率,1年内出现约3例癌症是基准值。即使新增3例,也会拉低当前的患者年发生率;如果少于3例,发生率还会进一步下降,更多病例则可能“再次吓坏市场”。
- Walker认为,需要上报的癌症病例会滚动报送FDA——“如果第1周出现癌症病例,第2周前就依法必须报告”——而他“拿枪顶着脑袋猜”的答案是,管理层大概已经知道病例数量。May表示无法确认,但也猜测他们知道。
- 面对5亿美元现金储备——对于这种规模的试验来说“等于没有”——以及市场对增发的担忧,May指出管理层此前已表示会在维持期数据公布后融资,因此他预计如果股价回到约150美元,公司就会增发。增发不会扼杀并购;在RNA领域的先例中,公司曾在并购传闻期间增发,后来以更高溢价出售。一次增发最多只会让并购暂时搁置几周。
9. Nektar无消息下从92美元一路漂到50多美元
- 两位嘉宾对4月后的上涨记忆不同:Walker描述为约45美元涨到100美元,May记得是从70多美元涨到接近110美元;但两人都同意,92美元增发后,Nektar在没有任何更新的情况下跌到了50多美元。May说:“我有点无法解释这轮抛售为何能持续这么久、跌得这么重”,尤其是XBI并不疲弱;不过,针对同一资产,Nektar“向来都比一个拥有光鲜管理层的新IPO应有的交易价格低”。
- 可能的3项压制因素是:Eli Lilly诉讼,预计在夏末或初秋解决,法律结果押注“不在”生技专业投资者的能力圈内;Q32 Bio即将公布开放标签脱发数据,该公司有“炒热数据、最后令人失望”的历史,而盲法数据已经弱于Nektar,5月底还完成了一笔受到市场热捧的5,500万美元融资,RA Capital既是Nektar的重要股东,也是Q32的支持者;以及Sanofi尚未终止amlitelimab,尽管OX40L类药物已经出现卡波西肉瘤病例。Walker指出,Sanofi仍将amlitelimab定位为数十亿美元级药物,尽管他认为该药在特应性皮炎适应症上会获得黑框警告。
- May对Q32的安全性担忧来自作用机制:该药靶向IL-7,而携带IL-7受体突变的人类会患上SCID,多数情况下会死亡——“我对这款药的安全性有担忧……但从长期看,我并不担心它成为竞争对手”。对于Lilly诉讼,两人都认为陪审团面前的叙事——“邪恶的大药企偷走了这款药,把它搞砸了,还不肯还回来”——让和解几乎难以回避,但“这只股票的交易方式,完全不像市场认为他们能从诉讼中得到什么”。
10. 结论:基于确定性选择Abivax,多竞购者结局
- 被问到哪家公司被错定价得更严重时,May选择Abivax;他定义的不是上行空间最大,而是“基于客观数据,我对当前价格错误的确定性有多高”。Walker给出的近期区间是完全摊薄估值约160–180亿美元(每股175–200美元,6个月内约100%上行);如果安全性信号被排除、克罗恩病这张牌兑现,他认为股价可上探至高200美元区间,甚至达到300美元的并购价位。这些都是最乐观情景,但Walker认为下行空间相当有限,因为3期已经完成,最坏情景仍对应数十亿美元的UC销售额。Nektar长期可能上涨3–5倍,但Abivax的安全边际加上短期并购概率,“真的很难放弃”。
- 对于修复市场情绪,May不等股价走势自行回暖:“归根结底,这是一个多个竞购者参与的情景,价格将由谁给出最高报价决定……市场价格波动可能只是噪音。”他“完全相信”药企的业务拓展团队和监管团队会做出同样的癌症发生率计算。两人讨论的可比案例是Cidara:在股价大幅上涨后,因多个竞购者表达兴趣,最终仍以110%的溢价被收购。
- Walker最喜欢的一条背景信息是:Abivax约65岁的CEO此前曾在一次令人失望的3期读数后6周卖掉CinCor,交易附带CVR且溢价很高。Walker说:“闪电可能会击中2次”;May回忆称,当时的溢价非常高。
完整逐字稿
You're about to listen to Yet Another Value Podcast with your host, Andrew Walker.
Today I'm happy to have Adam May back for the second time. Adam is kind of the GOAT of biotech investing; I don't think it's crazy to say that. If you listened to the first podcast we did a couple of months ago, it got absolutely rave reviews—one of the best-rated podcasts we've ever done. People love hearing Adam because he has a deep knowledge base and is a lot of fun to talk to.
Today we're going to follow up. We're recording this Monday afternoon, June 8. Abivax, one of the companies we talked about a lot in that episode, released data last week. If this is a 1-hour-15-minute podcast, we spend the first 55 minutes talking about Abivax, and then we dive into Nektar a little bit at the end. I think you're going to learn a lot.
Abivax is a fascinating situation right now. They reported—without spoiling what's about to happen—blowout data, but there were some safety concerns, and the stock is in a really interesting place. If Adam has a read—and I think he does; I believe it—it could be very interesting. For disclosure, I'm long a little bit of Abivax and a little bit of Nektar. There's a full disclaimer at the end of the episode and in the show notes.
This podcast is sponsored by AlphaSense, and more specifically my upcoming AI webinar with AlphaSense. Look, the AI landscape is crazy. If you're an investor, it's crowded. It's confusing. Everyone's telling you to adopt AI, but nobody is telling you how—what tools to use, how to adopt AI, whether you should be focused on using it as a superpowered Google, whether you should be building your own tools, how you get used to it, all this sort of stuff. I personally find it's a lot of experimentation. It's a lot of fun, but it's really confusing and scary. So anyway, I told AlphaSense about my problems and they organized a webinar to try to help out. I'll be sitting down with Dave Wang of Wall Street Prompts and Ben Collins of AlphaSense to break down the modern AI stack for investors, including horizontal platforms like OpenAI and Claude, AI workflows, and finding specific intelligence tools—where each one can actually fit and help in a real research process. So, if you're trying to get better at AI and develop AI-enabled workflows, you're not going to want to miss this webinar. Join us—we're going to record it next week, around June 18th, and it'll be going live June 25th. There'll be a link to register in the show notes, and please feel free to send any questions you have on using AI, whether it's general tools or AI-specific tools like AlphaSense. So thanks, AlphaSense, and I'll see you for the webinar soon.
All right. Hello and welcome to Yet Another Value Podcast. I'm your host, Andrew Walker. With me today, I'm happy to have on for the second time the man, the myth, the legend, Adam May. Just a reminder: nothing on the spot is investing advice. Full disclaimer at the end of the podcast and in the show notes.
Adam, how's it going?
Going great. Glad to be back. Thanks for having me.
I'm super excited. Let me tell you how excited in 1 second. Obviously, people were thrilled when I said you were coming back on. I think the Twitter comment has like 100 comments at this point, but somebody sent me an email that said, “This is like having Victor Wembanyama on the podcast directly after an NBA Finals game.” I was like, “Wow, high praise.”
Probably a little too high, but I guess I'll take it.
It's been crazy. I'm wearing my Knicks shirt today, for those who can't see. We have Game 3 tonight. But the proximate reason for having you come on is that, in addition to the unbelievable reviews of the first podcast, there's actually been quite a bit of news for both Abivax—the ticker there is ABVX—and Nektar Therapeutics.
I think the more dramatic, interesting, and timely one is Abivax. I'm happy to toss it over to you and chat about it, or I can provide background—however you want to go. Why don't we start with Abivax?
Abivax has a lot to digest, and I probably could ramble on uninterrupted for an hour on this. Definitely interrupt me if I start doing that and ask questions that come to mind.
When we last talked about Abivax, we were discussing how I had approached predicting the data they showed in what was called an induction trial. That means the first readout of a phase 3 study in a disease called ulcerative colitis. It looked at the short-term efficacy of the drug.
That was the bigger, riskier catalyst because it basically says either this drug is active or it's not. It was very successful—absolutely very active during the induction phase. With the caveat that this is a cross-trial comparison, it was probably the 3rd-most efficacious drug that had ever been studied in ulcerative colitis. That represented huge upside.
The second big catalyst that would follow was the maintenance trial. The induction phase measured efficacy after only 8 weeks of treatment, and then the study would add 44 more weeks of treatment during the maintenance phase to see whether the drug works for a year and, if so, how much better or worse the responses get over that period of time.
The second big thing you can see from a maintenance trial is safety. Treating somebody with a drug for 8 weeks can reveal a big, rapid side-effect or safety problem, but generally speaking, if there's a problem with a drug from a safety standpoint, you're not going to find out in 8 weeks. You need longer. Frankly, even a year is a pretty short period of time to determine the safety of a drug.
When we last talked, we were waiting on the maintenance data update. Safety had largely been derisked at that point because there was something called a data safety monitoring board, or DSMB. That's usually a group of independent physicians who look at the data from the phase 3 trial as it's running.
The DSMB is unblinded, unlike everybody else in the world. Even the company's management could not have told you what the response rates were in the different arms; they were completely blinded to the results of the trial as it was running. But the data safety monitoring board can look at unblinded data to see who is on the drug and whether there are atypical side effects that appear to be occurring on the drug arm versus the other arm. If they see that, they stop a trial early or pause it and review changes, increase monitoring, or whatever is necessary.
It's every biotech investor's nightmare. You wake up on a Monday, not expecting it, and the news says, “The unblinded safety board found liver toxicity, cholesterol off the charts, or something, and we stopped dosing.” Then you say, “Oh, God, this drug is…”
It's just every way it can kill you except human causes, right? It won't cause a gunshot wound, but pretty much everything else it can do.
With very rare exceptions, you do not want to hear from the DSMB as the trial is running. What we did hear from management at the point when we talked was that 80% of the trial had already been reviewed by the DSMB, with no new safety signals. A few weeks after we talked, 90% had been reviewed, again with no new safety signals. Safety was very largely derisked.
The big question investors were asking was how effective the drug was going to be. There are 2 endpoints that really matter: the one we talked about and the primary endpoint. The primary endpoint—the one the FDA will use to assess efficacy and decide whether to approve or not approve the drug—is clinical remission delta. That's how many patients end up in clinical remission versus placebo.
The plausible range people were looking at for Abivax was a 20% to 30% placebo-adjusted clinical remission delta. On the last podcast, I said, “I'm going to guess 25%.” Twenty-five percent is a pretty good number. That would put Abivax ahead of most of the other biologic treatments—these safe, effective treatments—for ulcerative colitis.
It would have put Abivax behind a class called JAK inhibitors, namely a drug called Rinvoq, which had produced this otherworldly clinical remission delta of 39% in its phase 3 trial at the high dose and 30% at the low dose. There's a clustering of really safe drugs around 20%, 22%, and 25%, and then a huge gap in efficacy with Rinvoq, this JAK inhibitor, at 30% to 40%.
The caveat for Rinvoq is that it has a huge black-box warning for multiple safety problems: cancer, severe or fatal infections, blood clots, heart attacks and strokes, and what we call MACE, or major adverse cardiovascular events. All of those are on the black-box warning. It is incredibly effective, but very unsafe.
It's just every way it can kill you except for human causes, right? It won't cause a gunshot wound, but pretty much everything else it can do.
Yeah. And on top of that, those are all the safety warnings you have to tell a patient about. I mean, if it's in a black box, you've got to discuss it with a patient when you're prescribing a drug.
Black-box being kind of like the highest form of warning that the FDA can assign to a drug.
On top of that, there's lab monitoring requirements for the drug. So, it can cause lymphopenia, meaning it can make your white count drop very low. It can cause abnormal cholesterol and lipid elevation. So you have to have blood draws on top of all these safety problems. It’s a very clunky prescribing and user experience. Although it’s very effective, it’s very dangerous and hard to prescribe and use because you have all these hoops to jump through.
The blue-sky bull cases for Abivax in the maintenance readout were projected to be a clinical remission delta starting with a 3. If you could get it into a 3-handle on the clinical remission delta, you would be right up there with Rinvoq, with the 30% to 39% range from the high dose to the low dose. That would be standing alone above everybody else if you could get to that level.
What actually happened was a 4-handle for Abivax on the high dose: 40% at the high dose and, miraculously, 39% at the low dose. So the low and high doses matched and exceeded Rinvoq on clinical remission delta, which, as far as the people I talked to—and I’ve probably talked about this stock as much as almost anybody else in the world—that was not something people were thinking was feasible.
But to take it a step further, on the efficacy blowing out expectations, the second endpoint—it’s not the primary endpoint, but if you talk to a lot of GI doctors who prescribe these drugs, for many of them their favorite endpoint, the endpoint they rely on the most to assess efficacy, is actually endoscopic remission.
That basically is an endpoint where the GI doctor is performing a colonoscopy on an ulcerative colitis patient and says, “I see 0 evidence of any disease in this patient.” Meaning, I could not tell you if this colonoscopy came from a perfectly healthy 20-year-old or a 40-year-old with severe ulcerative colitis, because it’s been completely healed.
GI doctors like that endpoint because it’s regarded as more objective than clinical remission. The other primary endpoint involves some kind of patient questionnaire-type responses: How many bowel movements a day? Those are subjective answers that can be a little noisier and more variable.
So this endoscopic remission delta, although not the technical primary endpoint of the study, could be viewed as an equally big deal as far as the commercial uptake of a drug goes in ulcerative colitis.
Can I add one thing, just because I’m a journalist and you’re so far beyond me as a specialist? I would guess most of the audience was saying, I mean, ulcerative colitis is no joke: diarrhea, bloody diarrhea, urgency, all this sort of stuff. One of the treatments is—we talked about an alternative treatment, an alternative drug—just removing the colon, right? I only say that because if you are taking this drug and you have a healthy colon, it has completely cleared—
Your life—it changes your life, lengthens your life. Yeah. So, what you’re talking about, to frame that, we’re not just talking about cutting out a diseased colon. This may be a little graphic compared to what is normally on your podcast as a finance podcast, but when you cut out somebody’s colon, you have to create a new hole in the stomach where your processed food will exit into a plastic pouch that you carry around on your stomach, called a colostomy bag, for the rest of your life. The location where that used to exit is stitched shut.
It is a beyond-life-changing experience to have uncontrollable ulcerative colitis, and this is something that happens to many patients with it. So, absolutely no joke.
On the endoscopic remission discussion, we covered that Abivax met the unexpectedly high bar of Rinvoq on clinical remission. But on the endoscopic remission delta—which again is this endpoint that GI doctors favor—it crushed Rinvoq. This is the same sort of dynamic for Rinvoq, where it was head and shoulders above the existing competition to begin with.
The high dose of Rinvoq had an 18% endoscopic remission delta. Abivax’s high-dose endoscopic remission delta was 38%, more than double the endoscopic remission delta of this drug that had previously been the undisputed champion of long-term efficacy in ulcerative colitis. That result in particular was beyond the wildest bull-case dreams.
I can be sometimes a little critical of sell-side projections where it’s like, “Oh, what if it comes in incredibly, incredibly good?” And it’s like, “We don’t even need to talk about that because it’s never going to happen.” This outcome was not even listed on the scale for any of the scenario analyses of these data.
That data gets press-released Monday after hours, Monday, June 1, at around 4:05 p.m. The press release also says, “No new safety signals identified.” So everybody’s celebrating. I mean, this is the greatest high-visibility biotech readout ever.
The stock goes from 130 to 180 in half a second.
That’s right. So it was halted on these data. As far as a sustained stock price goes, it had been trading at an all-time high. It had briefly ticked up into the 140s at one point in the past on an M&A rumor, but the stock had been trading around 130 to 132, and that’s about as high as it had ever been able to sustain.
So from essentially all-time highs, the stock unhalted and went to the 175 to 185 range. People are celebrating. All of a sudden, a few minutes later, the stock is selling off. It’s down to 160, 150, and it starts accelerating: 110. Then it breaks below triple digits. People are asking around, “What is going on?”
I had reviewed everything that was available in the press release, which included the main efficacy data, the statement that there were no new safety findings, and a brief safety table that laid out that there were no headache problems, no pancreatitis, and no cardiac fibrosis. These were all things that had been talked about as possible safety problems ahead of time. All clear there.
It listed a few cases of cancer. That totally passed the sniff test to me. I didn’t think anything of it. You see cases of cancer in large phase 3 studies all the time.
I was waiting for the slides with the conference call to come out so I could pore through them and try to find why it was selling off. I’m looking, I’m looking, I’m looking—I can’t find anything. Then I start hearing people messaging me and posting on social media about 7 cancer cases in the high-dose, 50-mg arm of the drug versus 1 cancer case on placebo.
It does turn out that a perceived cancer risk is the reason the stock really—I mean, crashed. I think “crash” is probably a fair word to use for how much it sold off.
I don’t even know if “crash” is a fair word, because I follow a lot of the pharma companies, and Abivax again goes from 130 to 180 after hours. I mean, it opens the next day around 80, right? I think every non-megacap pharma company is down 5% the next day, just as the swoosh of margin calls—and it crashed the whole pharma sector for a day.
The XBI, the index that tracks it, was down around 4% or 4.5% that day. It went as low as 68-something that day. So, 185 after hours down to roughly 68. We’ll talk about this more, but that is beyond an overshoot to the downside. It all stemmed from this perceived cancer-risk signal.
I’ve written about this and published this stuff online at this point, but I think there’s objective, data-based reason to say that this concern that there is a safety risk of cancer being caused by this drug, obefazimod, is demonstrably false. There are a lot of ways to attack that issue.
But to first of all frame this, I think Abivax management severely mishandled the way that they presented the data, and they presented it in a way that led people to perceive a problem that wasn’t there. They themselves, as experts in this field, were able to review the data and see very clearly—as I feel like I can see—that there is no cancer risk with this drug, and they took that for granted. They took for granted that the market would be able to rapidly interpret that same data or come to that same conclusion.
In reality, they didn’t even need to show the cancer cases in the way that they did. If you go back and look at topline phase 3 press releases from other companies with ulcerative colitis drugs, even ones that do have black-box warnings for causing cancer, they’re not press-releasing the cancer cases at the level of detail that Abivax did.
In one way, it was sort of being overly open about the data that they had. They could have very easily shown it in a different way. I don’t know—you wouldn’t call it hiding the data—but there’s precedent to say that it did not need to be disclosed in the way that it was. That’s one problem.
There are 2 other problems with the way they presented it, and they’re actually really, very, very glaring. One is a huge issue. We have 7 cases of cancer that we need to explain on the 50-mg drug arm versus 1 in the placebo arm.
First of all, 1 of those cases was something called colonic dysplasia, which is by definition not cancer. Dysplasia is a term for a precancer. It is something that’s very common in ulcerative colitis patients because they have inflamed colons, and they have a higher risk of developing cancer in their colon over their lifetime.
But the reason the term exists is to distinguish it from cancer. They listed it under the cancer cases and, in the footnote, specifically referred to it as colon cancer. Completely wrong. It turns out that was probably what they would refer to as a coding error. Basically, some of these data sets are automatically compiled based on predetermined categories.
And the colonic dysplasia probably fell under a neoplasm category that includes cancers but also benign stuff like cysts and lipomas. They just failed to take it out. But beyond failing to take it out of the list of cancers, they also specifically referred to it verbatim as colon cancer, which is just factually incorrect. So from the get-go, 1 case can be completely eliminated. Now we're down to 6 versus 1. 4 of those cancer cases were something called non-melanoma skin cancer.
Mhm. That is a totally different category than what colloquially most of us would call or think of as cancer—meaning something that you get diagnosed with, you tell your family to sit down, and tell them the bad news. Non-melanoma skin cancer is 15 to 20 times more common than the second most common type of cancer.
My dad had it last year, and it was like, “Hey, not a big deal. We're going to go.” It was on the bottom of his nose. They cut it out, and it wasn't a big procedure. You can tell me if I'm wrong. Sorry to jump in.
100%. No, that's exact. Probably everybody listening knows somebody, even within their family, who has had non-melanoma skin cancer. There's probably 5 million cases of this per year in the United States, and it's almost always cured by a roughly 15-minute procedure performed under local anesthesia.
So you numb somebody up with a shot, and then you cut it out. Sometimes you just cauterize; sometimes you stitch them up. But in any case, they drive themselves there, they drive themselves home, and they go back to work afterward. You are cured of cancer that day.
Correct.
So some of the less experienced specialists looking at the name clearly thought that these squamous cell carcinomas and basal cell carcinomas that fell under the category of non-melanoma skin cancer were, for lack of a better term, legitimate or real cancer. I don't want to diminish it too much because it is—you know, I'm a dermatologist, so it's kind of my career to treat those. I don't want to overly diminish it, but it really is not a big deal. That's what I'm telling patients 10 times a day whenever I diagnose it—10 times a day, and cure it 10 times a day.
So you could eliminate 1 of the 7 because it was literally not cancer; it was as simple as them putting almost a typo in the table. You can eliminate 4 more because they were non-melanoma skin cancers, and the FDA literally measures cancer risk with drugs using an endpoint called non-melanoma skin cancer. So it was a mistake to really even talk about those 5 types of cancer. Then you're left with 2 quote-unquote real cancers in the 50 mg arm, 0 in the 25 mg arm, and 0 on placebo. Those 2 cancers in the 50 mg arm were the 2 most common types of cancer in the world besides these pseudo-cancers.
Prostate and breast.
And so, right off the bat, just distilling this from 7 down to 2 cases of cancer probably would have prevented this entire crash from happening in the first place. Adam, can I just—you said 2, and I want to be clear. It's 1 case of prostate cancer and 1 case of breast cancer, right? This is an N of almost 200, and you have 1 in 200. I'm sure people can see where I'm going. If I had this big trial of 200 and said 1 person got prostate cancer, you'd probably be like, “Oh, well, we should investigate,” but that might have just been 1 person who drew bad luck. It's not like you had 1% of 5,000 or 500,000, and you say, “Oh, that's kind of...” It's 1. It's literally 1.
Exactly. This is, in my opinion, statistical noise without a doubt. There are a few ways that the FDA is going to look at this to clarify that, and I think confirm that. One is that there is no smoking-gun mechanism for how obefazimod would be causing cancer. There are classically 2 ways in which the FDA considers a drug to be carcinogenic, meaning that it causes cancer.
One is that it causes DNA mutations. This was well studied years ago, and the data have been public for a long time. Obefazimod does not cause DNA mutations; it's not what they call genotoxic. So, no smoking gun there. The second is the one that we weren't quite sure about until this data came out, which is immunosuppression.
A drug can increase your risk for cancer because it turns down your immune system to the point that it stops seeking early cancers and killing them. The way you can tell whether or not a drug is immunosuppressive is typically by whether or not it increases your infection risk. Severe infections in this study were more common on placebo than on the 50 mg dose. So unequivocally, the drug is not immunosuppressive. From the very beginning, there's no suspicious mechanism by which we would expect that this drug could be causing cancer. So that's point 1.
Speaking to the idea that this is just statistical noise, if you follow a bunch of people for a year, you're going to have some people who have cancer, no doubt. One other big point that the FDA will look at in cases of cancer is whether there is what they call clustering of a specific type of cancer. A lot of drugs that do have a cancer risk will cause a very specific type of cancer because they're working by some specific mechanism. So some drugs will be found to just increase your risk of lymphoma, or some will increase your risk of lung cancer, and things like that.
So the fact that these were 2 completely distinct organs, breasts and prostate, there's no clustering. And then you just need to look into the background rate of what you expect the rate of new cancer development to be in a population like this. And so that's kind of the point that you're getting at with this being most likely statistical noise. If you took 1 case from the drug arm and put it in the placebo arm, then it's completely balanced with 0 signal.
The expected rate of new cancer development in ulcerative colitis patients is roughly 0.5 per 100 patient-years. And so here's another error that people made in trying to interpret this data. They're comparing the number of cancers in the drug arm to the number in the placebo arm. This study is not by any stretch statistically powered to detect differences in a very rare outcome. It's powered to detect clinical remission delta, which happens in 50% of patients on the drug. It's an extremely common outcome.
If you wanted to tell a difference in cancer risk, you would have to have thousands of patients in this trial because development of cancer is a very rare outcome. You have to have a huge study to detect differences. So you actually don't need to compare to the placebo rate. What you need to compare to is the background rate. 1, and 2, for precedent purposes, you want to compare to the rates that other drugs that got approved in ulcerative colitis without a black-box warning saw in their trials. What rates did they see in their trials?
The expected background rate is about 0.5 per 100 patient-years. Patient-year is very important because it gets at what we were talking about earlier, like in the induction period where the study was only 8 weeks long. If you follow 100 patients for only 8 weeks, that's actually a very short, small number of patient-years, and I don't—
I know I'm speaking in terms that are kind of foreign to people, probably, but this is important.
I was just going to build on—I mean, I think I know where you're going because I saw all your tweets. But if I'm just looking at obefazimod 50 mg, let's just round it to 200 people on obefazimod 50 mg and 200 people in the placebo group. Those are roughly the right numbers.
I think what you're driving to is that only 66 people finished the placebo over the 44 weeks, and 160 people finished the obefazimod 50 mg arm. So even if you're saying, “Oh, we had 2 cases in the obefazimod 50 mg arm and 0 cases in the obefazimod 25 mg—or, sorry, in the placebo,” you say, “Oh, well, it's a lot more likely in the obefazimod 50 mg.” The answer might be, “Yeah, but you had 4 times the sheer time from the obefazimod 50 mg because more people are on that thing and they're not dropping out.” So if you start looking at background rates, it might actually be less cancer. Who the heck knows. Am I drawing that correctly?
Yeah, you are. So what you're speaking to is what we would call adverse-event capture. What happened in this trial was that almost all of the patients who were on placebo dropped out of the trial because they were severe ulcerative colitis patients who were not getting any treatment, and they couldn't tolerate it. It's literally dangerous for some of these people not to be on treatment, so they had to drop out of the study.
When you drop out of the study, you're not being followed anymore for adverse events. So if 1 of the over 100 patients who dropped out of the placebo arm early on in the study, 6 months later, got diagnosed with lung cancer, we'll never know. We won't know because they dropped out of the trial. And so adverse-event capture on the drug arms was much higher because about 80% of those patients completed the study. They were actually being followed. They were actually being assessed for potential cancers.
These are all ancillary points. In the long run, what really matters is: Does the signal that we have at 50 mg, the high dose, look like we're seeing a higher rate of cancer with those cases that we have versus what we should expect in severe ulcerative colitis patients? The answer, in my opinion, is objectively no. We're not.
So another big mistake the market made was only looking at the phase 3 data. We have hundreds of patient-years. We actually have the largest safety database for this drug coming from the phase 2 study, which is counterintuitive because it had fewer patients in it.
But that phase 2 study has been running for 7 years versus 1 year of this phase 3 trial. So, as far as the number of patient-years of adverse-event capture that we have, the phase 2 study is actually a much larger safety data set. I think a lot of the initial reaction to the phase 3 data was based on a very quick, rough calculation using only the phase 3 data and forgetting the fact that we have this big phase 2 study, which had only 1 case of cancer over hundreds more patient-years.
So, when you take those 3 cases—1 from phase 2 and 2 from phase 3—and make the assumption that the phase 2 case was from a 50-milligram dose, which we don't know, the rate is 0.59 per 100 patient-years. That is, 0.59 cancers occur for every 100 years that a patient is on this drug so far. That is compared to what the literature tells us: The background rate is about 0.5.
So, we're perfectly in range with what you should expect these patients to have as far as their development of cancer goes. But that takes some work to get those numbers, and certainly people were not prepared to look at that in depth after hours on Monday night, when the stock was trading down 60% or whatever.
I want to give one more bull case that I heard, and then I want to provide some pushback because I think it'll be useful for people. The other bull case I had was from a friend who was buying this after hours at $80, and he was working off the 7 number. He was like, “Okay, cool. There's a 3% to 5% increased chance of cancer. Go talk to a gastroenterologist about having UC and see if they'd favor a 3% to 5% increased chance of cancer versus being deep into UC, and see which one they would pick.”
They were like, “This is going to be a blockbuster even if you black-box a 3% to 5% increase in cancer.” So that's the last bull case where I'm supporting you. I'll pause there, and then I want to hit you with some bear cases.
Yeah. So that's a great point, and I'll echo it by saying Rinvoq, the drug whose efficacy obefazimod just beat in maintenance—remember, it has cancer as a black-box warning and has 4 other potentially fatal adverse events listed in its black-box warning that this will lack. It has lab monitoring, worse efficacy, more black-box warnings, and lab-monitoring requirements. It is probably going to do peak sales in IBD of around $4 billion across ulcerative colitis and Crohn's.
So there's no argument in my mind that, even in the worst-case scenario—which I now think we can objectively show is unlikely, where the FDA gives this a single black-box warning for cancer—this isn't going to be a multibillion-dollar drug. When this stock got down into double digits, it was pricing in worse than a worst-case scenario, basically.
Let me provide some pushback. The first pushback, people would say, is: “Look, Adam's a super-sharp guy. Drool comes out of his mouth when he talks science, but Adam's a super-sharp guy. The market has had a lot of time to digest these results. There's a lot of sell-side research. It's not like the old days, back when it was at $10. There's tons of sell-side research, and it's probably the most talked-about name in the biopharma sphere right now.”
There are lots of event-driven people, and people think it's going to get taken out. But the market has gone from $130 to $100. People are scared, and people have had more than a week to digest this news. Adam sounds super bullish. He sounds like this is a small-potatoes issue that will get ignored—probably no black-box warning, and people will just look at the statistics. The market is not saying that.
I saw a lot of banks that downgraded the stock, not the moment after this happened, but 2 days later, when they had time to digest it. So I think a lot of people, myself included, when I say, “Hey, is this really where you're going to find edge?” Andrew will look at this and say, “The market's had a lot of time to look at this. It went from $130 to $100. They're worried about this.” I'll follow up with the 7 report, but I think you see where I'm going with that.
Yeah, absolutely. Neither of us would be on this podcast right now, or doing what we do to make money, if we believed in efficient markets. Philosophically, we could go there, but I am repeatedly amazed at the things that the market is able to ignore when it comes to deep analysis of data.
These are not obvious insights. It takes work to calculate these numbers and come to these conclusions, and I just think people either haven't done it or are worried that more cases of cancer are going to come out. We can talk about the part 2 data release, where that's actually probably going to be the case—that there are more cases of cancer—but it has to be below a certain rate. There are lots of caveats to add there.
I can't tell you why it doesn't trade back to at least the $130 range where it was before these data came out. But in some sense, I'm glad that it doesn't, because it speaks to the ability to have differentiated analysis.
I would also point out that we talked about my assessment of the induction data on the last podcast, and the prediction that one could make with the data from this company's corporate deck was essentially mathematical proof that the study was going to hit and have 1,000% upside. I published those data publicly, and you can follow the number of views these things get. It had tens of thousands to 100,000 views on this thesis, where I laid out the math number by number, showing that this was guaranteed to hit—essentially as close to a guarantee as you can ever get to a 10x upside.
Over the next 2 days, the stock traded down after tens of thousands of people read that diligence. Then, after hours, at hour 48 since I published it, the data came out, it hit, and 1,000% upside was the result. So it wouldn't be the first time with this very stock that the data are out there, that the data are very objectively clear in my opinion, and that the market is pricing them wrong.
I don't 100% disagree with you, but I would say that since then, it is much bigger now, right? Much bigger. So there are a lot more eyeballs, and there are a lot more eyeballs on you, my friend. People are looking at your tweets—
Fair. Totally fair.
I think another thing—another worry I have as well—is this: ABVX at $100 per share is about a $7 billion company, right? With an M&A premium—we'll talk about M&A in a second—everyone thinks it's going to M&A. Let's just call it a $10 billion bite size to make the numbers really easy.
I think the worry people have is, “Hey, if you get a black box or you've got this worry about cancer, this is a huge check, right? Even for big pharma, $10 billion is big.” People are kind of worried and saying, “Oh, if you get black-boxed, it's just not as neat. It's not as clean. It's not as much of an M&A-driven story.”
I don't know. Maybe we'll know by early next year or mid-next year if it's going to have a black box or not. So maybe you just need to wait for that resolution, and that's the big catalyst. But I think people are kind of nervous about that. Does the FDA just say, “All right, whatever. It's black-boxed. We'll just put a black-box warning on cancer”? I mean, there were some cancer things. I'm not sure, but I think that's the other thing people are worried about.
Yeah. No, I think that's the thing people are worried about: Is this a black-box warning or not? There's one other thing that they could do, which is adding a warnings-and-precautions note on the label. That is essentially the same as not having anything on the label at all, as far as I'm concerned.
So it really is a black-box warning or no black-box warning, and that's going to be the question. That's definitely the fair thing to point out.
Let me reframe my question. I guess, to me—with, you know, I'm just a dumb-dumb—with 5 black boxes on the competitor versus 1 black-box warning here, and better data here, right? If there is a black box, it still seems like, at this price, if it's going to be a $2 billion, $3 billion, or $4 billion drug, there's still upside here. And if there's no black box, it's sky to the moon.
I guess my question to you is: Is the bet at this point all about black box versus no black box, or can this work even if it's just the 1 black-box warning? For some reason, they do put the cancer warning on it.
At $100, I think that the M&A premium for a buyer who is assuming a black box would make this a buy. I guess that's a convoluted way of saying it. At $100, I think that you're pricing in a black-box warning, and I think that somebody could reasonably buy this for $130 to $150 a share, at least taking on the high odds favoring a black box.
There are 2 variables, actually. It's not just the black box. The second variable is whether this drug will work in Crohn's disease, which there's going to be a readout for in mid-2027.
Yep.
And so there is a huge skew between the absolute worst case and the absolute best case. The absolute worst case is that this does not work in Crohn's and has a black-box warning, meaning you're launching only in ulcerative colitis with a black box.
In that case, I think the proof of concept we have with Rinvoq very clearly shows it would still be a multibillion-dollar product. So even if a pharma were to buy it for $10 billion and it gets $2 billion in peak sales, you're at least going to make your money back by doing that, or it's not going to be a massive failure.
But the blue-sky case here would be a hit in Crohn's and no black-box warning, in which case I think you're talking about $4 billion to $5 billion in peak sales in ulcerative colitis alone and $4 billion to $5 billion in peak sales in Crohn's alone. So, an $8 billion to $10 billion peak-sales drug—that is a number that almost exceeds a pharmaceutical company's ability to buy a company with a drug of that value. You're starting to talk about a drug that, if it were getting that sort of peak-sales projection, would have to go it alone because you're getting into more of a merger territory than a buyout territory.
And so there's a very nonzero probability of that blue-sky case playing out. I think that the probability of success in Crohn's has increased because of the magnitude of efficacy that we saw in ulcerative colitis. It's not that the extreme efficacy we just saw speaks only to increased sales in ulcerative colitis; it speaks to adjusting the probability of success upward for Crohn's as well. So there's a huge skew here, and the M&A premium, if there is going to be one, will be done on a risk-adjusted basis for both of those opportunities. Right? So if the low end is a multibillion-dollar blockbuster drug in UC alone with a black-box warning, you're still looking at a pretty cheap valuation at $100 a share.
This is a perfect one for CVRs as well, right? A CVR based on the Crohn's readout, and a CVR based on whether they get a black-box warning or not, depending on when they sell. Let me give you one. I don't read a lot of sell-side research, but as I was preparing for this call and this podcast, I got a sell-side note from Jefferies that I thought was interesting. It said—and I'm just going to read a few quotes—“We'd posit that a good drug does not always equal a good stock. We see this as a situation where the outlook for the drug is arguably better than the near-term outlook for the stock.”
So those are just interesting things that hit on it. But the place I really want to focus is: “We see that the differentiation versus JAKs has narrowed, which makes the stock setup ex-M&A difficult.” It goes on to detail how the differentiation versus Rinvoq has changed based on these results. When I read that, versus both what I see on their slides and what I hear from you, it's a little different, right? Because I think you think this is best in class. We just said 4 better than 4 over 3 to start. And they're saying, “Oh, it's not that different.” So what are they saying that's leading them to say, “Hey, the differentiator is not that differentiated?”
I think they must be talking about safety, because that's really the only thing they could be talking about. I think it would be interesting—I haven't read their work prior to the readout—but if you go back and read their work prior to the readout, I would be interested to see what probability they assigned to equivalent efficacy to Rinvoq. People were not expecting this to even match, let alone exceed, Rinvoq on efficacy.
So to say that it wasn't differentiated ahead of the data—we would have considered a blue-sky scenario to match Rinvoq's efficacy. Broadly, that's what people were saying: If you could achieve JAK-like efficacy, that would be the blue-sky case. So I don't know. My take from the very brief comments you made there is that they're being prisoners of the stock-price movement and not actually evaluating the fundamentals that we're seeing here.
Yeah, look, I don't disagree, but again, I am not a science person. The one thing I worry about here is that I've got a lot of you and a few other bulls here, but this is, again, the most popular drug, I would say, in biotech, just because the skew is so high and everybody's doing it. And I'm like, hey, I feel like I get from 130 to 80 when people are saying, “Oh, cancer.” But at this point, to me, it seems pretty obvious that it's not a huge—I don't want to say huge—risk, but it seems pretty obvious that this is very small. Again, one prostate, one breast cancer, and even if it's skin cancer, the stock just hasn't gone up.
Maybe I also will say I've heard that the management team is having private meetings with investors and telling them a bunch of stuff. And I hate when I hear, “Oh, there are private meetings and I'm not inside of the information,” but neither here nor there.
Let me ask: The last time you were on the podcast, I think it was right before they hired a chief commercial officer. Both you and I were saying, look, everyone thinks this is a takeout. They do the maintenance data and then there's a takeout. Now they're in this interesting spot. I said CVR because we're going to know if it's going to get a black-box warning or not in the next 12 months, and then in mid-2027 we're going to have the Crohn's readout.
They're in this weird spot where hopefully they have FDA approval in the short to medium term and Crohn's in the medium term, but they've got this black-box overhang. They do have the chief commercial officer, and everyone thinks it's an M&A target, and you've got this weird thing. So what is your kind of read—or what would your hope be—on the M&A side? Do you want them to wait until they clear up the black box? Do you want them to pursue commercialization? Do you think they should just hit the bid now? How do you think about this?
Well, I'm glad that they hired a chief commercial officer. We talked about that last time. I think it would be silly not to have done that. What do I hope they do? Well, there's something we haven't talked about that's actually really, really key, and that's—you may have heard people referring to Part 2 of the phase 3 data set.
Oh, yeah. That was on my question. Absolutely. Let's go there.
That has the potential to completely get rid of what I think is already a relatively clear lack of a cancer signal, but that could really put this narrative to bed. Basically, Part 1 is what they planned to release all along, which is the primary efficacy analysis in maintenance. It consisted of patients who responded during induction. That was only about a third of the patients in the induction phase because you also had patients on placebo who didn't respond at all.
Roughly 1,300 patients started in this phase 3 trial. Roughly 600 have been reported on already: 200 on 25 milligrams, 200 on 50 milligrams, and 200 on placebo. So we've got—I think the number is going to depend on how many people dropped out—conservatively, maybe 550 to 600 patients in the phase 3 study that we do not have data on yet because they're in a totally different arm.
It gets a little bit convoluted explaining why they're not part of the primary efficacy analysis. And really, for all intents and purposes, for what this data set is going to be used for, which is safety, it doesn't matter. What matters is that there are several hundred patients who have been on 50 milligrams for a year that we do not have data on yet. The efficacy in that population doesn't matter much. What's going to matter is safety.
Originally, when management press-released and discussed these data, thinking naively that nobody was going to be worried by the cancer signal, they were telling us that these Part 2 data were going to come in October at a conference. Totally normal; nothing wrong with that. But when the stock is crashing 60% after hours and you say, “Hey, the data set that has extra safety sample size in it, which could exculpate this cancer risk—we're going to give that to you in 4 months. Don't worry about it,” people were freaking out, thinking they were hiding something.
And then the very next day they came out saying, “Well, okay, we will move that data up to before the end of June.” So now we're going to get this Part 2 data sooner. It does not seem like they were hiding that data. They just legitimately did not think people were going to be worried about a cancer signal.
Within the next, I guess, 3 weeks or so, we're going to get those extra patients' worth of data. So the question that we have to ask is how many cases of cancer are going to be in that study across the 25- and 50-milligram doses. We should expect, at background rates, roughly 3 cases of cancer to be identified over the course of 1 year in that number of ulcerative colitis patients.
If there are 3 cases of cancer, this could change depending on which treatment arms the cases actually occur in. It probably would be better if they were in 25 milligrams instead of 50. Even if there are 3 new cases of cancer in Part 2, that would further lower the current rate, per patient-year per 100 patient-years, of cancer that Abivax is seeing. If it's fewer than that, it'll reduce it even more and totally clear up this cancer risk.
It could be higher and spook the market again. A lot could come down to this Part 2 dataset, which should be released within the next few weeks. In a perfect world, I think you release that data, prove that there is no cancer risk—or further cement that there is no cancer risk on top of the data that I think we already have that show that—and the stock recovers to the $150 range, maybe. Certainly, I personally don't see why it couldn't go to where it was headed in the after-hours before people got spooked by this cancer signal, which is the $170s or $180s. But if we're going to accept that there's just some sort of scar left behind from that drop that just can't be healed, then sure, maybe it only goes to the $130 or $150 range, and then I would like for M&A to happen after that. That would probably be my ideal.
Let me ask a few questions about this. Number 1, the event-driven side of me says—and I am 99% sure I know the answer—but management says they're going to put the safety data out in October. The stock drops, and they pull the safety data forward to June. Do they already know the safety data, or is this still blinded?
It's a good question. I don't know. I don't think anybody knows. To me, it stands to reason that they would know because the primary purpose of that dataset is safety. So, if they were to have analyzed anything from that dataset already, it probably would have been the rates of rare adverse events in that dataset. In a way, you could say it might be a good thing if they know that data because they were so eager to pull it forward. I don't know, and I can't confirm it. I don't know how long it takes for them to collect and clean these types of data. I think people would probably be surprised by how much complexity there is in compiling this sort of information.
Gun to my head, do you think they've seen the data or not? I probably would say yes, but I'd be pretty nervous. I was trying to ask the question by dumbing it down, but I have an idea. I don't know the answer here. My guess would actually be that they did not because they were waiting until October. But my guess would also be that they, like you, are convinced that this does not cause cancer and they've seen other stuff other than what we've seen. I think they were just like, “Fuck, it's down 60%, and we'll talk,” and I think they're just like, “All right, unblind it. We'll do this. We've got huge confidence.”
Yeah.
It's possible in their stock, but that was my read. They've just got huge confidence, but I don't think they know.
No.
You can give some more color to this to say that there's a better chance that they do know. At least they would know the raw numbers, because cancer cases are reported to the FDA on a rolling basis during these studies by the DSMB, right? So, it's not like you wait until the very end and then pour through 1,000 patients and try to find cancer cases. If there was a cancer case in week 1, that was reported—it is mandated to be reported to the FDA by week 2 of the study. So these cases, if there are cases of cancer that there should be in Part 2, should have been identified on a rolling basis as the study was going, and those numbers should have been immediately ready to furnish to management from the data safety monitoring board.
I would guess they know, but I don't know.
Hell, we might even both be right. They might not know the overall safety numbers, but because of what you're saying, they might know the cancer numbers. The market's signing off on cancer, and they're like, “Oh, we can show there's no cancer risk here.” For reportable events like cancer, I would think that they would know.
Right?
Let me ask about M&A. We've talked about M&A a few times. I think one of the reasons the stock bounced—and I hate to keep using the stock bounce, but the stock goes from 180 to 80 or 70 and has come back up to 100—is that people were worried. This is a biotech that had $500 million of cash, which sounds like a lot, but when you're running these big trials, $500 million is nothing for this.
Oh, yeah. The night the stock was crashing, the bulls' bear case—and I actually think this is kind of a silly bear case—but people were terrified that they were going to have to price a secondary at $60. Management has been telling people, apparently—I mean, I've seen this through sell-side; I'm sure people have better access—that they don't need a secondary in the near term. I think part of the reason the stock jumped is because, oh, they're not going to have to blast a secondary into weakness. I guess my question is: do you think it's, “Hey, they report the safety numbers and then they do the secondary,” or is there the read that, “Hey, they're not doing the secondary, so game on”? Every biotech investor, when there's positive news, says, “If we don't have a raise within 24 hours, they're about to sell themselves.” So it could be either.
I kind of hate that, but, yeah, you're definitely predicting the Twitter commentary. No doubt. So, that'll be the question. If they present the Part 2 safety data and the stock goes back to 150, my assumption would be that they are going to raise. They said multiple times consistently before the maintenance data release that they were going to raise money after the maintenance data were released. So my assumption is that if the stock gets back to where they thought it should have been initially, they would raise money like they said they would.
People in my experience who aren't experienced buy-side specialist investors freak out whenever there's a secondary because they say, “Oh, now M&A's off the table.” In reality, M&A's off the table for a couple of weeks. RNA last year, they were in rumors and they did a secondary, and everyone said, “Oh, that must be off the table.” Then they sold, and if you read the background, they sold at a bigger premium after doing a secondary, which means the main buyer was paying even more because they're paying off more shares. It happens all the time. It's very simple leverage. You need to have enough cash to continue.
With the read being that there's an active sales process ongoing.
I think we've talked a lot about Abivax. I do want to, if we've got 5 or 10 minutes, quickly go to Nektar. The last time we talked, they came out with, if I remember, strong results in mid-April. The stock ran from $45 to $100, followed by a big secondary at $92 to cash the balance sheet up, and the stock, on no real news that I'm aware of, has just drip, drip, dripped back down to—I think it's in the mid-$50s today. I'd love to hear how you're thinking about it. Any new news or anything, while I've got you?
Yeah. So, when we were on last time, we were talking about, basically, a maintenance phase of a study for a disease called alopecia areata, an autoimmune type of hair loss. The hope for Nektar was that they were going to show that longer-term treatment could deepen responses in alopecia areata. The stock went into that maintenance update, I think, in the mid-$70s and traded up to almost $110 on the positive results, in which they showed that they did have deepening of responses with longer-term treatment.
The stock was in triple digits. They raised cash at $92 at pretty strong pricing, and since then the stock has gone from that $92 deal price to the mid-$50s, like you say, on no updates. I'm a little bit at a loss for words as to why this sell-off has been so sustained and severe. Nektar has always traded at a discount to what a fresh new IPO with a flashy management team would trade at if they had this same data set and asset in their possession. So that, in some ways, is not new or surprising to me.
But I do have to say that the Nektar sell-off has gone much further than I would have imagined, especially given the fact that the XBI, the index that sort of tracks this sector, has not done badly. It hasn't done poorly over the last few months as Nektar has sold off. I don't know. I've looked around for ways to explain why it is staying weak. There are a few that you can think about.
We did talk about, at least briefly on the last podcast, that Nektar has this ongoing litigation against Eli Lilly, where they're suing Eli Lilly for having held on to this drug and misanalyzed data for this drug—which is true. That did happen. We're probably going to have a resolution on that sometime from late summer to early fall to find out whether or not Nektar is going to get paid damages.
I do get the sense that some biotech specialists are avoiding the stock until that is cleared because it's basically out of their mandate, right? To evaluate the legal case is not something that a biotech specialist typically would be involved in. So I think there's at least some small contribution from people who may normally be interested in the stock waiting on the sidelines until this catalyst that is way out of their wheelhouse plays out. That's one thing.
The alopecia areata data, I thought, were encouraging. There's another small biotech that's been rallying lately called Q32 Bio that has a drug that's going—
Yeah.
Yeah. They’re going to be reading out data on alopecia areata in the very near term, actually, and they’ve been hyping up expectations for that. The company has a bit of a history of hyping up data and then disappointing. They did the same thing with a completely failed atopic dermatitis study with this same drug.
The alopecia areata data that Q32 Bio is going to be presenting are open-label. They already have some blinded data that are a little bit worse on efficacy than Nektar’s. There’s definitely a chance that if you just glance at the open-label data Q32 Bio is going to release, you might think it looks better. So, I wonder if there’s some overhang from potential competition. Besides the fact that the trial is open-label, I think that drug has some significant safety concerns attached to it. So, I’m not very worried about it, but the market may be.
I think the Nektar selloff started in earnest in late May, and you mentioned Q32 Bio. I’m just putting the 2 together as you say it. In late May, Q32 Bio did a $55 million raise from a pretty buzzy—
Yeah. RA Capital is one of Nektar’s biggest holders, so they’re believing in both—
And I mean, this $55 million sounds big, but it’s kind of a drop in the hat to them once you split it up. But I do wonder if part of it is people saying, “Hey, you’ve got these specialized biotech funds investing into this buzzy thing. Maybe there’s some substance to that sizzle over there, and maybe there’s not as much of a pot of gold as we thought over here.”
Yeah. I have to say that drug from Q32 Bio targets—one of its targets is a receptor for a cytokine called IL-7. If you have a human who has a mutation in the IL-7 receptor, they develop a severe form of immunodeficiency. It’s called SCID, and it’s fatal in most people. So, I have concerns about the safety of that drug, which actually echoes what happened. I think we talked about this again on the last podcast with that OX40L drug class, where you could say, “This drug blocks OX40L,” and when you look at humans who have mutated OX40L, they get a type of rare cancer. So, I have safety concerns about that drug.
I do think it might be contributing to the overhang for Nektar, but I think in the long run, I’m not concerned about it as a competitor. Speaking of OX40L, that’s one other thing that could be weighing on Nektar a little bit. Last time we talked about this OX40L class, and recently that class of drugs had been showing a bunch of these rare cases of cancer called Kaposi sarcoma. And as we talked about on the last podcast, I was postulating that Sanofi was about to discontinue that drug—
And Amlitelimab, because of those cancer cases. Since then, a Sanofi earnings call has happened where they did not yet discontinue that drug. They are still projecting it to be a multibillion-dollar drug, despite the fact that it’s going to get a black-box warning, in my opinion, in atopic dermatitis and eczema. So, it’s a very different risk-reward discussion than ulcerative colitis, like we were having earlier.
Other people may have been making the same assumption I was, saying, “Hey, Nektar’s biggest competitor, Amlitelimab, is about to get canned, and that’ll be great.” When it didn’t—at least yet—that may have led to some of the weakness. So, there are a few different little things that could be contributing to it, but in the long run, I don’t think anything has really changed since we talked, besides the fact that it looks like the alopecia areata signal is legitimate.
No, it’s just interesting because, again, I’m a generalist. As you said on the first podcast, my issue with Nektar before the readout was always that I couldn’t believe a large pharma would, when you read the lawsuit, mishandle a drug that could be a blockbuster this badly. But—
But it’s just interesting, the lawsuit angle, because you still have that lawsuit, and the legal system is much different than what is right and just. But it just kind of seems like, based on all the data we’re seeing, Eli Lilly mishandled a blockbuster that would be on the market by now if Nektar had done this. I don’t know how there aren’t hundreds of millions in damages there. It’s like, okay, I understand you’re a biotech, you’re not focused on that, but you still have the blockbuster drug. I don’t know.
And that’s the thing: this is due for a jury trial. That case is due for a jury trial. Exactly what you just said is going to play out in front of a jury if they let it get there, which is—
Big, bad Big Pharma—
—stole this drug, messed it up, wouldn’t give it back, and harmed this tiny little biotech. It’s going to play so well in front of a jury. I cannot imagine that Eli Lilly doesn’t settle. But this stock is not trading as if they’re going to get anything out of that lawsuit, so hopefully that surprises to the upside.
You know the Valley of Disillusionment, where you start learning about something, you think you’re an expert, and then you’re like, “Oh, God, I know nothing,” and then you kind of— I’m on the bottom of the Valley of Disillusionment. Two years ago, I would have been like, “Legal situations? I’m an expert.” And I’m like, “I know nothing, man.” But if I know anything, I know this: Big Pharma does not want to go to jury trials on botched handling of drugs that can help millions of people.
Right. Last question, and then we’ll obviously have you on for Part 3 because you’re the MJ of biotech. We’ve got to do the three-peat. Between Avidity, Abivax, and Nektar, which do you think is more mispriced right now? Or you can go off the board and choose something different if you want.
No, I think it’s Abivax. And I don’t know—people will interpret “the most mispriced stock” in different ways. A lot of the time, people interpret that as the stock with the most upside. But it’s not that. It’s the degree of certainty that I have that the current price is wrong based on objective data.
Yep. And for me, it is Abivax. I think the probable near-term range would probably be $16 to $18 billion. Actually, the fully diluted market cap—the fully diluted share count—is a little higher than what I think you cited. That would probably be like a $175 to $200 stock. So, the 100% upside in the near term—meaning the next 6 months—is the upside case, the blue-sky case.
I think if they let it go longer, clear up this safety signal that I am comfortable myself clearing up, but I understand that the market has not joined in on that sentiment yet, and if they let the Crohn’s data card flip, you could see it in the high hundreds, maybe even a $300 stock, depending on the strength of the Crohn’s data, at an M&A price. So, 6 months, maybe 100% upside; 1 year, maybe 200% upside—both being blue-sky scenarios over those time spans, probably. But I think the probability of those cases playing out is high, and I think the probability of downside is quite low because, as we’ve talked about, the worst-case scenario now that the phase 3 data are in is that, if it has a black-box warning, we already know it has outstanding efficacy and is going to be a multibillion-dollar drug in ulcerative colitis alone.
Certainly, Nektar could have higher upside over the long term. They could 3x, 4x, or 5x, maybe, depending on how the phase 3 trials play out. But given the margin of safety that you have on Abivax, and I think the near-term probability of it going up a lot because of M&A, I think it’s really hard to pass up.
No, look, what’s really interesting to me is you and everybody come to conclusions, but if this is a billion- or multibillion-dollar drug with a black-box warning and only UC, you’re kind of buying it at that. If they sold it on a multiple, you’re kind of buying it there. So, you’re like, “Hey, I’m pricing this on the downside case, and I’m getting all the upside for free.” And as you said, the phase 3 is out. We know this works. We know this works really darn well in this indication.
So, it’s not like when it was at 10 and you had analyzed that it was going to work, but we were waiting to see whether the drug worked. We know it works. Now it’s just a question of no black box or black box, more safety data, and whether it works in Crohn’s next year. And the thing I love about this is the CEO is, what, 65 or 66? He already retired before he started running this company, and he retired after he sold his last company following a disappointing data set—6 weeks after a disappointing data set.
Just as an interesting parallel for people to keep in mind, his last company was CinCor. They had a letdown in a phase 3 trial readout; 6 weeks later, the company was sold with a CVR. So, lightning could strike twice in that—
If I remember, it was a very big premium, too.
Oh, yeah. Yeah, yeah. They did just fine on that. And, you know, we talked—
We talked a lot about this perceived need for Abivax to overcome the market sentiment around the cancer risk to see upside. And I think this is a case where I’m not as concerned about the market coming around, because I think ultimately this is a multiple-bidder scenario, where the price is going to be determined by who has the best bid, not by what the stock price is or what the premium to the 50-day moving average is going to be. I firmly believe that there will be multiple pharmaceutical companies interested in adding this to their portfolios at these prices.
And you can look at some of the companies Merck bought in the past year. I mean, it—
What’s the one that was the flu drug? It was trading around $40, and it just—yeah, it—
It was a 110% premium—
—to the market after a huge rally.
But once you get into a bidding-war situation, you've got those strategics—
That's actually the appropriate example. Yes. Cidara had been going up on the open market, and then it was bought out at a 110% premium because there were multiple bidders. That's actually what I foresee happening with Abivax, meaning that the market-price fluctuations are probably noise.
I fully trust pharmaceutical regulatory-affairs and business-development teams to calculate the same sort of data that I have already calculated myself and come to the conclusion that there is not a cancer risk with this drug. I think that the prices they offer for it will reflect that.
Well, people are here to listen to you, not me. The other thing is that I talked to 2 doctors who were buying this after hours, and they were like, “Dude, if this is black-boxed...” Again, as I said earlier, go talk to a gastroenterologist. A small increase in cancer risk versus what this—I mean, we talked about removing the colon. At 40% success, there's a 40% chance we're not going to have to remove someone's colon.
There's a reason that Rinvoq, with 5 fatal black-box warnings, is a multibillion-dollar drug in these—
No, Rinvoq does have a few other indications as well, though, right?
That's right, but this is banking on just ulcerative colitis. It's a multibillion-dollar drug. Yeah.
Adam, this has been awesome. People know where to find you on Twitter and everything. We'll have you on for Part 3 because you're the MJ of biotech. We've got to do the three-peat in the near future, but I appreciate you coming on. I appreciate all the work here, and we will chat soon.
Sounds great. Thanks.
A quick disclaimer, nothing on this podcast should be considered investment advice. Guests or the hosts may have positions in any of the stocks mentioned during this podcast. Please do your own work and consult a financial adviser. Thanks.