Episode 166 - December 12, 2025

Summary

• The House-passed FY2026 NDAA frames biotech as a national-security asset alongside AI, quantum, and hypersonics — with new money and new constraints. Incubate’s John Stanford flagged a new DoD biotech office and an SBIR-like funding stream for synbio, biomanufacturing, and AI drug discovery, while a “watered down version of the BIOSECURE Act” is “on its path to becoming law,” enforced through two companies-of-concern lists (DoD’s 1260H and a subjective White House/OMB list where CDMOs/CROs may appear and lobby to stay off). A “reverse CFIUS” outbound-investment regime excludes biotech “yet,” and SBIR/STTR has no current path to restart; the next opportunity is the January 30 government-funding bill.
• The financing thaw looks real — an almost-$2B financing day (Terns raised nearly $750M against a roughly $650M target, Wave surpassed $400M) with XBI up about 35%–36% YTD — but the IPO window remains the missing confirmation. Chris Garabedian called these “the best signs we’ve seen in years” and wants 10–15 IPOs in H1 2026. Sam Fazeli said most raises followed good data and added Kymera, MoonLake, and Dyne. Sam had guessed 15 IPOs for the year against Josh’s 50; Yaron Werber said his later-stage private-company list is more Phase 2/3-heavy than ever, but might contain only about 20.
• ASH’s BTK battle left BeiGene challenged but not dethroned: Lilly’s Jaypirca relapsed/refractory data “really did not beat at all” and IRC-assessed numbers “looked dramatically worse” than the abstract, while frontline looked good but extremely early. Yaron’s verdict: “you could argue BeiGene is no longer the absolute leader with no competition at all,” but it has “a monster pipeline” — including a BTK degrader now head-to-head against Lilly. “This is going to be fireworks for a while.”
• In the BCMA wars, Gilead/Arcellx’s anito-cel “looks considerably cleaner” and clinicians say it’s “probably going to be the preferred drug” when it reaches market next year — but a fourth-line label versus CARVYKTI’s second-line label makes off-label use “extremely hard.” Sam Fazeli’s standout conference data was J&J’s MajesTEC-3: teclistamab plus daratumumab with 36-month OS of 83% versus 65% — bispecific data “that could stand up to the CAR-Ts” — while EsoBiotec’s in-vivo BCMA CAR-T with four-of-four MRD negativity stole the innovation spotlight.
• Dyne’s exon-51 DMD skipper made close to 3% of normal dystrophin at six months — “about 10x” standard of care — yet the stock traded up but somewhat sideways amid investor confusion about “philosophically what exactly does this FDA stand for.” Paul’s key logic: Exondys 51 hasn’t failed a confirmatory study, so it shouldn’t be pulled — and if it isn’t, there’s no obvious path for the FDA to reject a drug clearly making more protein that looks safe. Chris agreed on approval odds but “still struggles” that no strong clinical dataset has correlated to the dystrophin biomarker.
• Obesity news is now daily: Pfizer paid Fosun’s YaoPharma $150M upfront (up to $1.94B) for a Phase 1 oral GLP-1 with no data yet, and Wave’s single-dose WVE-007 readout cut visceral fat 9% with a lean-mass gain — but total body weight didn’t fall. Adam Feuerstein dubbed it a “gym-bro drug”; Grace Colón countered with the headline “WVE-007 early obesity data has license to thrill,” while Sam said Lilly’s retatrutide triple-agonist data “came out actually better than I thought.”
• The panel’s 2026 optimism scores — Chris 7.5, Sam 8, Paul 7.7 — aligned with the new BioPharma Sentiment Index at 78/100, where investors are most bullish, followed by biotech professionals, pharma, and lastly academics, “obviously because of the NIH cuts.”

Deep dive

1. The NDAA makes biotech a national-security asset — with strings attached

• John Stanford’s headline: the House-passed NDAA “isn’t a bill just for warships and planes” — biotech has now joined quantum, hypersonics, and AI as fields “viewed as a national security asset.” The DoD gets a new biotech office and authority to build an SBIR-like program subsidizing synthetic biology, biomanufacturing, AI-designed molecules, and infectious disease; for the first time even the intelligence community (CIA, NSA, ODNI) gets biotech liaisons with “enormous pools of money to begin to draw down on.”
• The constraint side: a “watered down version of the BIOSECURE Act” is on its path to becoming law. It would bar federal agencies from buying from “biotechnology companies of concern,” with flow-through problems for recipients of federal money that use listed companies. Two lists to watch — DoD’s existing 1260H list and a new, “more subjective” White House/OMB list where CDMOs and CROs may appear and get a chance to argue their way off. “Those companies of concern are probably going to lobby the White House pretty hard.”
• The sleeper provision: a “reverse CFIUS” outbound-investment regime for AI, quantum, and hypersonics. Biotech “is not covered — yet,” but the White House can add fields over time, meaning funds might someday have to notify Washington of a foreign investment “and at times ask for their permission.” Stanford: “this is a huge deal that we have to pay attention to.”
• Answering Sam on NIH, Stanford dispelled the catastrophe narrative: cuts to the roughly $45B spend are “pretty nominal,” though “in some areas they are existential,” with priority-setting in transition; DoD priorities will be reflexive to “what is China investing in,” with data protection and synthetic DNA/RNA obvious focuses. SBIR/STTR stays paused over a logjam “between only a couple members of Congress” about foreign-funded “SBIR mills”; the next chance to break it is the January 30 government-funding bill.

2. The financing thaw is real; the IPO window is the missing proof

• Grace Colón’s setup: one day this week hit almost $2B in financing — Terns went out for roughly $650M and raised nearly $750M, while Wave surpassed $400M — with XBI up roughly 35%–36% year to date. Chris: “these are the best signs we’ve seen in years” — rising valuations, more biotechs above a $1B market cap, and fewer negative-enterprise-value companies — but the real tell would be “10 to 15 IPOs in the first and second quarter of 2026.”
• Sam’s caveat: most of these raises came on the back of good data — he added Kymera, MoonLake, and Dyne to the list — which is healthy and “hard to argue with,” but “still very very different than opening an IPO window.”
• The standing bet from last week: Sam said 15 IPOs next year; Josh said 50, leaving Sam feeling “stupid.” Yaron said his later-stage private-company list is “more of a phase two, and sometimes phase three, profile than it’s ever been,” versus 2020 when IND acceptance was a value-creating event in an S-1. “But I don’t know if there’s 50 of those… it might be 20.”

3. ASH battle one: Jaypirca mixed, BeiGene keeps the crown but loses the walkover

• Yaron on the $12B BTK market: Brukinsa is now the global leader on a quarterly basis, both in the U.S. and ex-U.S., after clean head-to-head superiority over Imbruvica. Lilly’s noncovalent Jaypirca — whose abstract “looked like shock and awe” — delivered “a little bit for everybody”: relapsed/refractory data “really did not beat at all,” and the paper’s IRC-assessed data “looked dramatically worse” than the investigator-assessed numbers Lilly had put in the abstract.
• The frontline data looked really good but extremely early — and most of the survival edge over the historical BR arm reflects comparator toxicity: of 10 deaths, 9 were toxicity-related and only 1 was disease progression. Sentiment holds that physicians will still start with covalent inhibitors (you can sequence Lilly’s drug after Brukinsa or Calquence, but not clearly the reverse), and the touted safety differences “ended up being a little marginal.”
• Bottom line, per Yaron: “you could argue BeiGene is no longer the absolute leader with no competition at all,” but it has “a monster pipeline” in CLL — a BCL2 inhibitor combination and a BTK degrader now being tested head-to-head against Lilly’s drug. “This is going to be fireworks for a while.”

4. BCMA wars: anito-cel’s safety versus CARVYKTI’s label — and in-vivo CAR-T arrives

• Yaron called ASH “a little bit of a tough meeting” for J&J/Legend: CARVYKTI’s efficacy is tremendous and capacity constraints end early next year, but Paul Matteis highlighted 1%–3% parkinsonism/neurotoxicity and a roughly 0.5% rate of immune-mediated enterocolitis that are absent with Gilead/Arcellx’s anito-cel, which “looks considerably cleaner.” Clinician feedback is that when anito-cel reaches market it will “probably be the preferred drug” because of its safety, except its fourth-line label makes off-label use “extremely hard” without data or coverage. CARVYKTI should therefore keep growing in earlier lines and in high-risk patients.
• Sam flagged AstraZeneca’s dual BCMA/CD19 CAR-T, AZD0120, from the Gracell acquisition: the CD19 arm targets progenitor cells for potentially deeper, longer responses, and 36-month China-based data showed 89% OS in newly diagnosed high-risk and frail transplant-ineligible patients — striking enough that the session chair asked whether it could displace autologous stem-cell transplant. Sam’s gut: “it’s going to do better and… this therapy is going to come earlier.”
• Yaron’s star of the show: EsoBiotec’s in-vivo BCMA CAR-T — no leukapheresis, no conditioning, “literally you just put an IV in the patient and give them the lentivirus and the fusogen” — produced four of four rapid MRD-negative responses, essentially no ICANS, and very little CRS. J&J has a broad collaboration with EsoBiotec but does not have rights to this particular drug; Yaron’s caution was that “you essentially have a gene therapy here,” and durability is unknown.

5. MajesTEC-3 was the standout — and “chemo is really going away”

• Sam called J&J’s MajesTEC-3 the conference’s standout: teclistamab plus daratumumab versus dara-based triplets, with roughly 290 patients per arm, delivered “phenomenal” 36-month OS of 83% versus 65%, with 36-month PFS of 30% against CARTITUDE-4’s 26% at 30 months in a similar population — “something here that could stand up to the CAR-Ts.” Sam said it was up for FDA approval in 2026, if he was not mistaken.
• The concern sits in the OS curve: early infection deaths from chronic B-cell depletion, many during COVID. Physicians consulted think decent IVIG, now also available subcutaneously, can manage it. Sam added the sequencing wrinkle that bispecific-first patients have a lower probability of responding well to a later CAR-T because of exhausted T cells, absent a long treatment gap.
• Yaron emphasized Genmab/AbbVie’s Epkinly: updates out to three years in DLBCL, subcutaneous and increasingly outpatient, point to “the best regimen out there” ahead of second- and first-line readouts next year — while Roche will counter with Columvi stacked on Polivy-R-CHP, “a cocktail of five, six drugs.” “Chemo is really going away, thank God… standing ovation at ASH. It’s hard not to get a little emotional as to how good things are going.”
• Terns’ TERN-701 allosteric CML inhibitor: Sam’s tell — “the fundraising tells you the story.” Once daily, with or without food, with no DLTs or pancreatic toxicity, it produced 43% MMR after prior asciminib and 50% in prior-TKI patients in the 85-patient CARDINAL trial. His open question: can Terns commercialize CML alone — “or will they even have the option, because the data is maybe so good that someone else might get interested?”

6. ESMO Asia and bladder: CTLA-4 redemption, crowded new targets, CG versus J&J

• Yaron on BioNTech’s “gentler,” more pH-mediated CTLA-4 in phase 3 second-line lung cancer: after a partial FDA hold over suspected toxicity in nonsquamous disease, the squamous group in Part A looked “definitively better on efficacy so far than docetaxel” — the old drug “everybody’s going to beat and then they all get knocked out one after the next.” Still, “you expect there is going to be a price to pay on more tox”; “it’s nice to see something finally working in that segment.”
• Sam’s roundup: GSK’s B7-H3 ADC posted a 33% ORR in Chinese nonsquamous NSCLC — broadly similar to other early programs including BNT324 and AMG 994, though muted in squamous and oncogene-driven cohorts — and Zai Lab’s DLL3/CD3 trispecific ZL-1310 set the bar in the China-only small-cell cohort, with the caveat that Chinese lung cancer is often not smoking-driven and is associated mostly with pollution, making cross-trial comparisons especially difficult.
• In non-muscle-invasive bladder cancer, CG Oncology’s BOND-003 against J&J’s TAR-200, the gemcitabine-eluting “pretzel” in SunRISe-1, showed similar six-month event-free survival, with CG’s side-effect profile “a little bit better.” They are completely different modalities, so physician preference will decide the split.

7. Dyne’s DMD win collides with an FDA nobody can read

• Paul: Dyne’s exon-51 skipper, using a transferrin-receptor muscle-delivery platform, made close to 3% of normal dystrophin at six months — “about 10x what you see with the standard of care” — plus compelling separation on functional endpoints versus placebo, better than expectations. In the FDA environment of the past decade, “an absolute no-brainer that this would be highly likely to get approved”; instead, the stock traded up but somewhat sideways amid investor confusion about “philosophically what exactly does this FDA stand for” — bullish Marty comments, confusing actions, and comments from Dr. Prasad.
• Chris, front-row veteran of the Sarepta saga: Sarepta set a low bar, especially with Exondys 51, which Dyne clearly cleared on the dystrophin biomarker — but he “still struggles” that no strong clinical dataset has correlated to the biomarker, partly because of sample size and partly because of the effect size achievable in six months or two years.
• Paul’s load-bearing argument on Sarepta contagion: Exondys 51 hasn’t failed a confirmatory study, so it shouldn’t be pulled for a different exon skipper’s failure — “and if it’s not pulled from the market, what would you see as the path for the FDA to reject something that is clearly making more protein and looks like it’s safe?” Chris concurred, recalling how FDA once tried to extrapolate Prosensa’s and PTC’s failures onto Sarepta — the meta-question being whether this FDA invents new criteria “that would throw everybody off.”

8. Obesity’s daily churn, and a 7.5–8 read on 2026

• Pfizer paid Fosun subsidiary YaoPharma $150M upfront, up to $1.94B in milestones plus tiered royalties, for a Phase 1 oral small-molecule GLP-1 with no data yet. Sam “tried to squeeze some data out” and said data might arrive by the middle of the year, wondering whether the program could eventually implicate the outbound-investment regime Stanford described.
• Wave’s single-dose WVE-007 data — 9% visceral-fat reduction, nearly 5% total-body-fat reduction, and a lean-mass increase, with the stock up a couple hundred percent — split the commentariat: Adam Feuerstein called it a “gym-bro drug” and noted total body weight didn’t fall, while Grace countered with the headline “WVE-007 early obesity data has license to thrill.” Paul’s synthesis: it directly addresses a possible GLP-1 muscle deficiency, “but what is the actual regulatory path here?” — and is the small early dataset replicable?
• Sam on Lilly’s retatrutide: the triple-agonist data “came out actually better than I thought”; much of the dropout confusion was patients “losing too much weight,” which was already known, and Lilly has positioned it for very obese patients.
• Grace’s closing poll, “don’t condition it”: Chris 7.5, Sam 8, and Paul “7.7 just to split the difference.” The new BioPharma Sentiment Index was 78/100, aligned with the panel’s outlook; investors were most bullish, followed by biotech professionals, pharma, and lastly academics, “obviously because of the NIH cuts.”