Episode 163 -- November 14, 2025

Summary

• Rick Pazdur’s appointment as CDER director is the stabilizing hire the industry wanted after George Tidmarsh’s departure. Maraganore called the over-25-year FDA oncology veteran “a breath of fresh air,” while Werber’s watch items are an old-school but progressive regulatory approach, no ODAC setbacks, and whether Pazdur and CBER chief Vinay Prasad can move past their public disagreements.
• Makary and Prasad’s NEJM “plausible mechanism pathway” is promising but vague, with its implementation apparently at odds with the FDA’s uniQure position. The proposal could let bespoke gene- and cell-therapy developers use single-patient expanded access to generate data supporting marketing approval, with a possible platform extension to broader mutation sets and eventually antibodies or small molecules. Armstrong flags this year’s recurring pattern of headline policies whose execution runs counter to them.
• Biotech M&A has turned into competitive bidding. Lundbeck’s $2.6B bid for Avadel, a total offer of $23 including $2 of sales-milestone CVRs, tops Alkermes’ agreed $2.1B offer of $18.50 plus a $1.50 IH-approval CVR; Fadia expects Alkermes to consider a counter. The Financial Times also reported another bidder remained involved in the Merck–Cidara deal until the last minute.
• Armstrong’s Pfizer–Metsera reconstruction shows how “deeply nasty and at times personal” the fight became — and offered a partial first glimpse at Trump-administration pharma antitrust. The FTC called Metsera on November 7 threatening to sue, but focused on the unusual deal structure rather than anticompetitiveness; Armstrong speculates the process may have been affected by Pfizer and Bourla’s relationship with the administration. Former Pfizer R&D chief Mikael Dolsten pulled his Novo board nomination at the last second after Pfizer reportedly raised objections.
• Alkermes’ VIBRANCE-2 delivered the first robust orexin-agonist dataset in narcolepsy type 2, supporting the class beyond NT1. Placebo-adjusted MWT gains of 9.3/6.7/6.7 minutes trail Centessa’s more-than-10-minute result, but Fadia cautions the cross-trial comparison because of sample size, heterogeneity, and differing timepoints. ESS reached about 10 points at the higher doses, with on-target insomnia and urinary urgency and fewer visual disturbances than in Alkermes’ NT1 data. Maraganore expects multiple companies to reach market, with differentiation potentially coming from secondary endpoints, dosing flexibility, time to market, and payer dynamics.
• In-vivo editing sits at a crossroads: strong CRISPR ANGPTL3 data at AHA versus an Intellia clinical hold after a phase 3 patient death. A single infusion cut ANGPTL3 73%, triglycerides 55%, LDL nearly 50%, and ApoB 33%; those taking a PCSK9 inhibitor saw more than an 80% LDL reduction. The dataset also included one death reported as unrelated. Separately, an elderly MAGNITUDE patient died after grade 4 liver injury, and with Amvuttra dosed quarterly and Ionis/AstraZeneca’s monthly autoinjector reading out next year, Werber asks whether CRISPR-Cas9 is necessary. Korro separately discontinued its AATD RNA-editing program.
• Cogent’s PEAK win in second-line GIST — 16.5-month mPFS versus about 9 months for sunitinib alone — shows the funding window reopening. In a field that has been “a graveyard for targeted oncology” (including Deciphera’s INTRIGUE failure), the stock rose about 2.5 times and Cogent raised $200M in equity plus a $200M convertible offering. Neurocrine’s NBI-770 MDD miss, by contrast, leaves that NMDA-subunit mechanism “probably less attractive.”
• The politics-and-public beat: an almost entirely closed MAHA Summit and Dave Ricks’ consumer-facing podcast turn. DeAngelis reports the invite-only DC summit, with a cheered Neuralink panel and Dr. Oz saying oral GLP-1s could hit the market “as early as March,” notably devoted little attention to vaccines. The group praised Ricks’ Cheeky Pint appearance as a smart way to reach average consumers, while Maraganore said the industry has “done a horrible job communicating what we do and why it matters.”

Deep dive

1. Pazdur to CDER: the stabilizing hire the industry wanted

• Maraganore’s framing: Pazdur agreeing to become CDER director is “a breath of fresh air” after Tidmarsh’s departure — which appeared related in part to actions involving Kevin Tang and a lawsuit — and reopened the wound around industry concerns about FDA stability. Pazdur has been at the agency for over 25 years, is a clinician, and is highly respected for his oncology work.
• The caveat: CBER chief Vinay Prasad has publicly disagreed with Pazdur, including just the day before the discussion. Maraganore hopes that disagreement can give way to rapprochement and that Marty Makary, Prasad, and Pazdur will work closely and collaboratively.
• Werber’s two hopes: an “old school but progressive — I mean from a regulatory perspective — kind of approach” that moderates the changes, and that ODAC “doesn’t suffer any setbacks at all.” The tail risk: “it would be devastating if a year from now Pazdur raised his hand and said it’s time for me to move on — which I doubt.”

2. The “plausible mechanism pathway”: ambitious, vague, and apparently at odds with uniQure

• Werber’s read of the Makary–Prasad NEJM two-pager, partly inspired by Baby KJ’s N-of-1 CRISPR therapy for a urea-cycle disorder: it initially targets bespoke gene- and cell-therapy products with a biologically defined pathogenic cause, a specific mutation or target characterized by natural history, and demonstrated target engagement. Animal or non-animal models and invasive or preferably noninvasive confirmation could be used.
• The FDA could consider the patient’s own natural history as a control, with a clinical outcome consistent with improvement in progressive disease or a disease-free course in a waxing-and-waning disease. The goal is for single-patient expanded access to generate enough data to support marketing authorization, potentially through a series of consecutive successful patients.
• The proposal could support accelerated or regular approval, with postmarketing safety and risk-mitigation commitments. It is intended eventually to broaden to antibodies and small molecules and to serve as a platform for related personalized products. The paper’s example is a disease with 150 mutations and common symptomatic manifestations potentially supporting approval across the mutations, but the meaning of “a few patients” and “platform” remains unclear.
• Werber’s catch: “essentially in parallel they decided to immediately go against it in the uniQure situation relating to Huntington’s.” He said elements of the pathway could arguably apply there, yet the FDA appeared to move against what had been agreed with the previous team. Maraganore adds that the scope will remain open until formal guidance is written, citing N-Lorem and other N-of-1 efforts.
• Fadia’s investor angle: the concept is positive for innovation, but inappropriate use could produce safety or efficacy failures that set it back — and she wants to know how payers would view the evidence and how pricing would be determined.
• Armstrong’s pushback: a recurring pattern this year in which “there’s a headline policy that people really seem to like… and then the actual execution from a regulatory standpoint seems to run counter to that over and over again,” creating “meaningful concern that we might not get to a really stable, consistent path forward.”

3. Inside the MAHA Summit: almost no media, little vaccine discussion, beef-tallow chips

• DeAngelis reports that hundreds of people, including members of the regulatory bodies at NIH and FDA, J.D. Vance, RFK Jr., Secretary Kennedy, and biotech figures, were expected at the invite-only inaugural summit Wednesday at Washington’s Waldorf Astoria. Politico first disclosed the gathering, whose agenda presented it as an exclusive networking event.
• At CRISPR, Samarth Kulkarni appeared with Alexis Borisy and George Yancopoulos; Baby KJ and sickle cell were discussed. A Neuralink panel reportedly drew cheers. Dr. Oz said oral GLP-1s could hit the market “as early as March,” while Lilly’s orforglipron had recently received one of the commissioner’s vouchers for expedited review.
• The notable absence: vaccines “really weren’t much of a topic.” The day focused more on food, lifestyle, general health trends, and brain-computer interfaces, and attendees received beef-tallow potato crisps.
• It was closed to essentially all media, including Fox News. The only broadcast session was Vice President Vance’s fireside chat, which CNN covered through the pool system; a STAT colleague observed the venue from its lobby.

4. Orexins deliver in NT2 — and Avadel becomes a bidding war

• Fadia on Alkermes’ VIBRANCE-2: the first robust NT2 dataset, with 93 patients, provides strong evidence that orexin agonists can deliver meaningful benefit beyond NT1 in a heterogeneous population and supports the class’s potential in larger neurologic populations such as MS and Parkinson’s.
• ALKS 2680 produced placebo-adjusted MWT changes of 9.3, 6.7, and 6.7 minutes at 10, 14, and 18 mg at week 6. That looks lighter than Centessa’s more-than-10-minute result, but Fadia cautions against cross-trial conclusions: Alkermes enrolled 93 patients versus 10 for Centessa, and the studies used different reported timepoints and treatment durations. Heterogeneity in NT2 and unknown tachyphylaxis remain important uncertainties.
• ESS, which Fadia called a potentially more relevant regulatory endpoint, especially in Europe, fell to about 10 points at the two higher doses, the range considered normal and comparable to Centessa’s week-2 result. Safety was consistent with the target mechanism: insomnia and urinary urgency were the main adverse events, while visual disturbances were less frequent than in Alkermes’ NT1 data. Higher doses and split dosing remain possible.
• Maraganore expects multiple companies to reach market. Differentiation could come from cognition or nighttime-sleep endpoints, dosing flexibility, time to market, and payer dynamics.
• Lundbeck’s $2.6B interloper bid for Avadel exceeds Alkermes’ agreed $2.1B offer: Lundbeck’s total $23 offer includes $2 of CVRs tied to 2027 and 2030 sales milestones, versus Alkermes’ $18.50 upfront plus a $1.50 CVR tied to approval of idiopathic hypersomnia. Fadia expects Alkermes to consider at least a counteroffer.
• On FTC risk for Lundbeck versus Alkermes, Fadia said it is possible but noted that the products have completely different mechanisms, scheduled versus non-scheduled status, and nighttime versus daytime use. With additional generic Xyrem versions expected early next year, she was unsure how much pricing power either oxybate or orexin would provide.

5. M&A turns nasty: the Metsera postmortem and Merck’s $9.2B Cidara buy

• Armstrong on his Endpoints boardroom reconstruction of Pfizer–Metsera: “two big companies that really needed to find their vibe back” — Pfizer searching for a post-pandemic move after an obesity failure, while Novo had lost its early obesity lead and crown. The fight became “deeply nasty and at times personal”: former Pfizer R&D chief Mikael Dolsten pulled his own Novo board nomination at the last second, with Armstrong’s understanding that Pfizer raised objections.
• The antitrust glimpse: the FTC called Metsera on November 7, the final day of the transaction process, and said, “if you try to do this deal, we’re going to sue.” Armstrong said the agency appeared focused on the unusual offer structure rather than the anticompetitive aspect, and speculated that the outcome may have been affected by Pfizer and Bourla’s close relationship with the administration.
• Werber on Merck–Cidara: the $9.2B deal is about three times Cidara’s recent market cap and gives Merck CD388, a breakthrough-designated, late-stage, long-acting, strain-agnostic influenza antiviral. The neuraminidase inhibitor is linked to a proprietary Fc fragment to create a once-per-season preventative for influenza A and B. Phase 2 showed 60%–76% protection against lab-confirmed flu from a single shot versus roughly 40%–50% for typical seasonal vaccines; ANCHOR phase 3 data are expected in the first quarter of next year. The deal also helps Merck build around the 2028 Keytruda loss of exclusivity.
• DeAngelis adds that Merck research chief Dean Li told STAT’s podcast that the company is pursuing an assortment of deals and products rather than one replacement blockbuster, including efforts such as its oral PCSK9 program.
• The small-cap version: Day One, with about a $1B market cap, agreed to buy Mersana largely for its B7-H4 ADC, building beyond Ojemda, its oral brain-penetrant RAF kinase inhibitor for pediatric low-grade glioma. The low-upfront, milestone-heavy structure lets Day One access the asset without a large initial payment.
• Fadia said the anchor is not the crowded ovarian/endometrial field but ACC, a rare salivary-gland cancer where B7-H4 expression is about 94% and the reported response rate was 55.6%.

6. Gene editing’s overhang: a MAGNITUDE death versus striking ANGPTL3 data

• Werber on CRISPR’s CTX310 Cas9 program at AHA: a single IV infusion in 15 patients on maximal background therapy cut ANGPTL3 by 73%, triglycerides by 55%, LDL by nearly 50%, and ApoB by 33%; participants taking a PCSK9 inhibitor saw more than an 80% LDL reduction.
• Safety appeared acceptable in the early dataset: no treatment-related serious adverse events, no grade-3 liver-enzyme elevations, grade-2 reversible infusion reactions in 20%, and one grade-2 transaminitis that resolved by day 14. One death was reported as not treatment-related. The FDA is calling for 15 years of in-vivo genome-editing surveillance.
• The Intellia counterweight: across more than 650 MAGNITUDE-1 and MAGNITUDE-2 patients, fewer than 1% had previously had grade-4 liver-enzyme elevations, and MAGNITUDE-2’s roughly 47 patients had none. In late September, an elderly man in his 80s in MAGNITUDE-1 was hospitalized with grade-4 transaminitis and increased bilirubin; the event was announced October 27, the program went on clinical hold October 29, and the patient died around November 5.
• With Amvuttra offering quarterly dosing and Ionis/AstraZeneca’s eplontersen using a monthly autoinjector in CARDIO-TTRansform, expected to read out probably in Q3 next year, Werber asks whether a CRISPR-Cas9 therapy is necessary. Maraganore calls the situation “a big overhang.”
• Werber said Korro’s reversible ADAR/LNP RNA-editing AATD program missed the thresholds in REWRITE and was discontinued amid competition from Wave, AIRNA, and Beam. He called it the right but difficult decision; Maraganore noted that the stock was penalized and expressed hope for Korro’s urea-cycle-disorder program.
• Armstrong’s closing meta-point: cholesterol has progressed from statins to long-acting PCSK9 biologics to gene therapies and potentially Merck’s oral small-molecule PCSK9, setting up a commercial and scientific battle in which “efficacy doesn’t always have to come with more complexity.”

7. Cogent’s GIST standout, Neurocrine’s MDD miss, and an industry learning to talk

• Fadia on Cogent’s PEAK phase 3: first-line Gleevec has a PFS of almost 19 months, while second-line Sutent has a little over 8 months. Bezuclastinib covers exons 17 and 18, complementing sunitinib’s exon 13/14 coverage. The combination posted 16.5-month mPFS versus about 9 months for sunitinib and a 45.6% versus 25% ORR in second-line GIST.
• The result is notable in a field that has “been a graveyard for targeted oncology programs,” including Deciphera’s failed phase 3 INTRIGUE study of Qinlock. Fadia estimated a $4B-plus opportunity with about 3,000 second-line patients. Cogent’s market cap rose about 2.5 times, followed by a $200M equity offering and a $200M convertible offering; Maraganore called it a sign of markets coming back alongside the XBI.
• Neurocrine’s NBI-770, an NR2B negative allosteric modulator intended to spare PV-positive interneurons and avoid ketamine’s dissociative effect, missed its primary endpoint in a 73-patient phase 2 MDD study. Fadia said it may have failed to trigger the circuit-level changes involving AMPA, mTOR, and BDNF needed for robust antidepressant activity, though the small multi-arm study may also have been underpowered for the day-five effect it was designed to detect. She called the mechanism’s outlook “probably less attractive.”
• Maraganore reflected on James Watson’s death at 97. Watson was the last surviving key scientist associated with the discovery of DNA’s structure, alongside Crick, Franklin, Wilkins, and Linus Pauling, but his scientific career was clouded by unacceptable bigoted and prejudiced comments. Watson’s advice to Maraganore about RNAi was: “John, just go fast.”
• The group praised Dave Ricks’ appearance on John Collison’s Cheeky Pint. DeAngelis highlighted Ricks’ line that if Taylor Swift can sell out a certain number of arenas, the industry can enroll a corresponding number of trials. Citing Catherine Friedman of GV, she described healthcare as becoming more consumerized and said it was smart for Ricks to address average consumers rather than industry peers.
• Fadia also recommended the interview, particularly Ricks’ discussion of transparency in pricing without undermining research incentives. Maraganore said the industry has “done a horrible job communicating what we do, why it matters, and why it’s good.”