Summary

The Sarepta/Elevidys saga escalated from restructuring to a full shipment halt in under a week: after reports emerged of a third death — this one in a four-patient limb-girdle muscular dystrophy study — Sarepta held a Friday “damage control” call. The FDA then asked Sarepta to voluntarily stop all Elevidys shipments; the company refused Friday night, then relented by Monday. Brian Skorney noted that, in the industry’s view, when the FDA asks for a voluntary action, it is “not necessarily a request… but more or less a demand.”
The path back for Elevidys is being litigated through leaks, with new information breaking almost every 6 hours: a senior FDA official first framed re-approval as a near-insurmountable hurdle, then reporting suggested a softer ask — a better safety profile via a different manufacturing process or dose in 10–12 patients. Skorney thinks the outcome “is going to be a political decision” and a messy public-opinion fight à la Avastin in breast cancer, with DMD advocacy groups “very vocal and very powerful.”
The hosts split on the removal itself: Sam Fazeli, speaking as a parent, asked whether “the parents [should] not be given the choice,” while Paul Matteis credited the FDA for moving fast and argued Sarepta “has benefited from a confusingly lower regulatory standard dating back to Exondys 51” — making this an outlier, not a read-through to other rare-disease names. Eric Schmidt’s stark investor takeaway: a company headed for a $4–5B top line became a potential bankruptcy risk in weeks.
Replimune’s RP1 CRL in refractory melanoma is the more tradeable FDA signal: a low-30s ORR, about half complete responses, against single-digit standard-of-care response rates, plus breakthrough designation and mid- and late-cycle meetings that went well, still couldn’t overcome an apparent higher-up objection to a single-arm study. Fazeli “would have probably approved it and let the physicians decide,” noting the FDA “could have told them that a year ago” — and Schmidt believes that without accelerated approval, “this drug’s never coming to market. It’s going to die.”
Matteis’ governing framework for the sector: “I still feel like we’re really figuring out who this FDA actually is.” Makary’s biological-plausibility talk sounded like “Peter Marks 3.0,” while other recent agency actions — including the DMD cardiac-component decision and Replimune’s CRL — make Matteis more cautious. He asked of the antidepressants-in-pregnancy session whether it was “an open discussion or… a preset conclusion to just question the orthodoxy,” and said small biotech needs visible approval “wins” to keep cost of capital from rising.
New CDER director George Tidmarsh is the first of the four HHS/FDA leaders with real industry experience (four companies including La Jolla, Horizon, Coulter, and Threshold), which Fazeli reads as a positive — though he shares the Prasad/Makary perspective on issues including COVID, and Matteis wonders “how long they can play nicely” given Laura Loomer publicly attacking Prasad’s appointment and GOP pro-flexibility rhetoric clashing with regulatory rigor.
On the clinical tape: Alkermes hit nearly all endpoints in narcolepsy type 1 but sold off on withheld safety detail about mild visual disturbances — Matteis’ aphorism: “opacity is often discounted even more than a mildly negative reality” — with full data at World Sleep in September; Abivax’s oral miR-124 enhancer stunned skeptics with 15–18% remission in phase 3 UC, ran over 500% to a $4B+ cap, and raised about $700M. Blenrep got a PDUFA extension rather than a CRL despite a 7-1 ODAC loss, and GH Research’s inhaled DMT remains stuck on a rat lung-toxicity clinical hold.
Macro turned constructive: China’s share of Fazeli’s proprietary drug-licensing dataset jumped to 14% from a 9% average last year — about 90% biologics and roughly 40% of $1B+ deal dollars — while pharma’s direct-to-consumer moves, including Novartis’ CEO calling DTC “the core of all his discussions with HHS” and AstraZeneca’s $50B U.S. investment, may let the industry “escape the wrath of tariffs” and MFN. Schmidt’s caveat on the optimism: “Be careful. You’re going to wake up in the morning… we have to eat humble pie.”

Deep dive

1. Sarepta’s “voluntary” halt that wasn’t

Skorney’s recap of the sequence: two commercial deaths on Elevidys this year, both in nonambulatory DMD patients; a restructuring announcement Wednesday; and unanswered questions about why the limb-girdle programs were stopped — then Thursday-night reports of a death in a four-patient limb-girdle study, followed by a Friday “damage control conference call” where Skorney “asked pretty starkly… where their threshold for materiality is.” Despite physician and parent hesitation, Elevidys was still generating hundreds of millions of dollars per quarter, while investors worried about debt payments over the next two years and partnership milestones. Sarepta had stopped shipping to nonambulatory patients but continued shipping to ambulatory patients. There is “certainly a disagreement” on the Street and within the company about what constitutes a material update.
Then the escalation: the FDA asked Sarepta to voluntarily halt all Elevidys shipments, including to ambulatory patients. Sarepta refused Friday night; by Monday it relented. Skorney noted that, in the industry’s view, when the FDA asks you to do something voluntarily, it is “not necessarily a request… but more or less a demand.”
The return path is whipsawing through leaks: a person “high up in the FDA” first positioned re-approval as a near-insurmountable hurdle; not long after, reporting suggested a softer ask — demonstrate a better safety profile “through a different manufacturing process, a different dosage in 10 to 12 patients” and perhaps establish a path back to market. The live debate is what the real hurdle will be, particularly with Prasad, an outspoken pre-appointment critic of the Elevidys approval and of Peter Marks’ decision to approve it over FDA staff objections.

2. Pull the drug or preserve parental choice — the hosts split

Fazeli, speaking as a parent (“my heart goes to any parent who had this hope”), still lands on the harsh regulatory frame: the FDA’s job is risk-benefit, the approval rested on “perhaps not the strongest” efficacy signals, and these are Grade 5 events — deaths. If the safety issue is provably confined to nonambulatory patients, maybe it stays on the market; but “should the parents not be given the choice of making that decision for their offspring if it doesn’t affect anybody else?”
Matteis’ contrarian credit to the agency: it “acted relatively quickly” amid STAT reporting questioning whether Sarepta had shared all safety data. His long-held view is that Sarepta “has benefited from a confusingly lower regulatory standard dating back to Exondys 51,” without ultimately doing confirmatory work there, so he “just [doesn’t] really know how [he] would extrapolate this into anything else.”
Skorney’s forecast — the load-bearing quote of the episode: whether Elevidys returns “is going to be a political decision.” He calls Prasad “one of the smartest people I’ve ever met” but “a much more stringent regulator than a lot of us would be,” who “is pushing on something that I think he may regret opening up” — invoking the Avastin breast-cancer withdrawal and Makena fights. “What the fate is, I think, has yet to be written.”
Schmidt on the leak channel: initially critical of a “senior FDA official… spilling the beans to the media before patients or their families,” he now partly understands it — after Sarepta refused the request, the FDA may have had to use “the court of public opinion” to get the drug off the market. Understandable, “but I don’t think it’s ideal.”

3. Tidmarsh at CDER: the first insider among the outsiders

Fazeli’s read on the new CDER director: unlike Bhattacharya, Makary, and Prasad, George Tidmarsh is an industry veteran — four companies by his count: La Jolla, Horizon, Coulter, and Threshold Pharmaceuticals — replacing Jacqueline Corrigan-Curay, who had been acting director since Patrizia Cavazzoni stepped down in January. A positive: “somebody who understands what it takes” for smaller companies to get drugs through. The thing to watch: whether CDER and CBER finally align, since historically “some had a lower bar… and others were really holding drug companies to account for everything.”
Schmidt’s caveat: Tidmarsh shares the same outsider-critic perspective as the other leaders — he reportedly met Makary at Bhattacharya’s anti-COVID conference — and “it’s maybe a little bit easier sometimes to criticize than to do.” Now that they are the establishment, can they still practice transparency and open debate while running the FDA?
Matteis on the coalition’s fragility: uniformity from Trump to RFK Jr. exists mainly on COVID; beyond it “there’s not that much uniformity in world health care view” — witness Laura Loomer going “all out” against Prasad’s appointment, and Makary’s claim that “biological plausibility” could support ultra-orphan approvals sitting at odds with the more rigorous clinical framework associated with Prasad’s Ending Medical Reversal. “I just wonder how long they can play nicely.”

4. “We’re still figuring out who this FDA actually is”

Matteis’ paradox list: Makary’s plausibility talk was “Peter Marks 3.0,” and rare-disease companies keep getting favorable agreements such as small pivotal open-label trials, while other recent agency actions include the DMD cardiac-component decision and last week’s CRL. His conclusion: “my FDA meter on the degree of flexibility… is changing every week. I’m leaning maybe a little bit more cautious.”
His test case for good-faith discourse: the antidepressants-in-pregnancy panel. Questioning default medical decisions without great randomized-controlled-trial data is “a laudable goal… in a vacuum,” but the panel included several people with a reputation for opposing antidepressants — “was that an open discussion or was that a discussion geared around seemingly a preset conclusion to just question the orthodoxy?”
Schmidt’s structural worry: decisions increasingly come down to a single leader rather than review staff, and a report that morning showed rank-and-file reviewers “heading for the exits” — cause and effect unclear, but a trend he is “very mindful of.”

5. Replimune’s CRL: flexibility promised, then withdrawn

The setup as Schmidt tells it: RP1 in second-line-plus melanoma, where standard-of-care response rates are in the single digits; the drug posted a low-30s ORR — roughly half complete responses, per Fazeli — with durable immunotherapy-style control, breakthrough designation, and mid- and late-cycle meetings that went well. The CRL, apparently driven by someone higher up at the agency — presumed to be Prasad — said a single-arm study could not support approval in oncology despite the unmet need.
Fazeli’s verdict: “I would have probably approved it and let the physicians decide how to use it” — oncologists “find the right patients.” The standard of care had shifted with the approval of Amtagvi, but that product is not off-the-shelf and does not have the same complete-response rate. If a control arm was required, “they could have told them that a year ago.”
Schmidt’s blunt terminal-value call: Replimune lacks resources for a randomized trial, so absent a successful accelerated-approval appeal, “this drug’s never coming to market. It’s going to die” — and KOLs who believe the drug works never got a panel where they could weigh in.
Matteis’ market translation: companies keep insisting “nothing has changed,” but “we need to see, for lack of a better word, some wins in this arena” to keep small biotech believing the FDA is open for business. Schmidt: uncertainty means higher cost of capital. Fazeli’s counterpoint: at least one other single-arm story involving a changed standard of care, Carvykti, is trading fine — investors are not reflexively extrapolating the outcome to every single-arm study.

6. Blenrep gets a PDUFA extension; GH Research’s baffling hold

Fazeli on Blenrep: a “bruising” 7-1 ODAC vote against in one trial, yet the FDA extended the PDUFA date instead of issuing the obvious CRL. The objections — ocular toxicity at the proposed dose and only 5% U.S. patients — leave room to return with lower doses, perhaps 1.9 mg, longer dosing intervals, and DREAMM-14 data. Myeloma doctors “deal with seriously toxic agents” routinely, the European Union just approved it, and “maybe Prasad here with his hematology hat on has been the opposite influence.”
Matteis on GH Research: inhaled DMT for treatment-resistant depression with an “unprecedentedly large effect size” in a European phase 2 randomized controlled trial, still stuck on a U.S. clinical hold over a rat lung-toxicity finding — even after rerunning the rat model, running a dog model that did not show the finding at the FDA’s request, and citing a meta-analysis showing the issue is usually rat-specific. His unease: “is the nonclinical team communicating with the clinical folks?” — and the investor dilemma: “do we even really know what else the FDA needs to hear?”

7. Alkermes punished for opacity; Abivax rewarded despite mystery mechanism

Matteis on the Alkermes orexin readout in narcolepsy type 1: the drug hit almost every endpoint in a study that was not well-powered — success was expected given great phase 1 data, multiple positive trials from Takeda, and strong biology — yet the stock sold off because the company withheld color on the class’s “mild visual disturbances” until the full data and safety update at World Sleep in September. His diagnosis: “opacity is often discounted even more than a mildly negative reality.” His work on driving-study requirements and comparable labels suggests no real regulatory consequence if the issue stays mild; versus oxybates, the class looks “very, very differentiated.”
Schmidt on Abivax (ABVX): an oral miR-124 enhancer with a “very strange, maybe even unclear” anti-inflammatory mechanism that many had discounted delivered solid phase 3 ulcerative-colitis data — 15–18% clinical remission rates comparing “relatively favorably” to injectable IL-23s such as Skyrizi and Tremfya — sending the stock up over 500% to a $4B+ cap with a roughly $700M raise overnight.

8. Prime’s repricing, China’s 14%, and pharma’s DTC détente

On Prime Medicine repricing underwater options — IPO’d at $17 and now around $4 — Skorney sees a “catch-22”: retention is real, but so is the value of letting the sector’s excess correct. Schmidt defended repricing for employees — “you can’t double-punish yourself. You can’t have one of your key drugs not work and then have all your key people leave” — but questioned repricing board options because boards may have greater fiduciary duties to shareholders. Matteis noted that exclusions for boards and senior management are sometimes used, though Prime chose not to exclude them.
Fazeli’s proprietary licensing data: China’s share of drug licenses jumped to 14% from a 9% average last year, near nothing a few years back; about 90% of China-sourced deals are biologics versus roughly 50% ex-China, and China deals took roughly a 40% share of $1B+ deal dollars.
The DTC pivot: Bristol Myers Squibb and Pfizer did it — or announced it — with Eliquis; Novartis’ CEO called it “the core of all his discussions with HHS”; and Roche’s CEO said cutting out middlemen could take drug prices down about 50%, against the president’s demand for “500, 600, even a thousand percent.” Fazeli’s deadpan: “he could give them all away.” With AstraZeneca’s $50B five-year U.S. investment, “maybe pharma companies are going to escape the wrath of tariffs” and MFN — though Schmidt warned: “Be careful. You’re going to wake up in the morning. We have to eat humble pie.”

Episode 150 - August 1, 2025 Hidden Gems' Chris Waller Judges Scientific Thesis