Episode 144 - June 6, 2025

Summary

• Sanofi’s $9.1B acquisition of Blueprint Medicines is the year’s second-largest biotech deal after Intra-Cellular Therapies, and Eric Schmidt frames it as the model outcome: an $18 IPO ten years ago becoming an “almost 10-bagger” through systemic mastocytosis execution. Sam Fazeli says most of the low-value CVRs ($2 development, $4 regulatory) appear achievable, and a consensus ~$2B drug by 2030, if that estimate is right, makes $9.1B “not ridiculously stretched.” Brad Loncar calls it an unusually mature deal for a Sanofi known for paying heavily for risky assets.
• Summit/Akeso’s HARMONi readout addressed the biggest bear case—China-to-US translation—yet the stock sold off on a missed OS endpoint (p=0.057, alpha spent, so it cannot become statistically significant). Management said the geographic subsets were consistent. Eric argues an OS hazard ratio in the 0.8 range plus a 0.52 PFS HR seems like a drug that could be used; Sam was “agog” at the reaction and says bears are “just looking for excuses,” like in the early Krazati days. He also says the key value was Western read-across, since a Johnson & Johnson regimen is already approved in this post-EGFR setting and other approaches are coming. Brad’s counter: with no conclusive hypothesis for why the bispecific beats a VEGF+PD-1 combo—which Roche tried and failed—and nobody hitting OS yet, “there’s a legitimate risk that’s still on the table.”
• Bristol’s partnership with BioNTech (about $3.5B “pretty much guaranteed” through 2028) takes some risk out of the PD-L1×VEGF question without answering it, per Sam, and Eric warns Summit will remain “extraordinarily volatile” as more data emerge. Sam’s read of Pfizer-partnered 3SBio’s ASCO data: ORR softened only modestly to 67.5% at n=34, responses appeared to deepen over time rather than reflect only a first-scan effect, but 17% grade-3 hypertension versus ~5% for SMT112 shows “these molecules are different. They’re not all doing the same thing.”
• Vera’s positive phase 3 atacicept data in IgA nephropathy (~low-40s placebo-adjusted proteinuria reduction, filing toward year-end) was immediately shadowed by Otsuka’s sibeprenlimab, but Eric disputes STAT News’s claim that Otsuka’s data are better and estimates the market at probably $15B+. His key point: patients and doctors care about stabilizing kidney-function decline—data Vera has and Otsuka so far does not—and “there is no usual one-winner-take-all phenomenon in biotech.” Brad flags Vera’s competitiveness as putting it in a strong position for potential interest from larger pharma.
• Brad’s emerging-technology call is protein degradation: Kymera’s STAT6 degrader showed 90% degradation from 1.5mg doses upward, with clean safety and T2/Th2 biomarker reductions (TARC, eotaxin-3) in healthy volunteers—a potential once-daily oral Dupixent competitor, with atopic-dermatitis patient data due in Q4. Nurix struck a STAT6 deal with Sanofi the same day; Sam calls degradation “a cornerstone of future drug development” for opening targets not amenable to enzyme inhibition, even after Arvinas’s subset-only phase 3 disappointed the market.
• Sam said AstraZeneca was likely the ASCO winner, with a seventh straight plenary—MATTERHORN’s “unprecedented” event-free survival in gastric cancer, first-line DESTINY-Breast09, and Gilead’s Trodelvy posting a 0.65 PFS HR in first-line PD-L1+ TNBC (ASCENT-04). Eric’s broader theme: innovation “sneaks up on you”—pancreatic response rates once celebrated at 7%–12% now reach ~25% with Bristol’s PRMT5, and Immatics’ PRAME TCR therapy showed an apparently durable 55% response rate in refractory melanoma.
• The contrarian TIGIT story: while Roche, BeiGene and iTeos have exited, AstraZeneca is running 10 phase 3s across ~8,000 patients and Arcus persists—and both programs are Fc-silent, which both companies believe is “day and night” different. Rilvegostomig combo data showed 71% ORR in PD-L1-high lung cancer; readouts land in 2026–27. In head and neck, Eric argues the Street has given Merus (~$4B+ valuation) winner-take-all status over Bicara (trading near cash) even though physicians see both as “off the charts,” at roughly 3× Keytruda’s response rate in a $3B–$4B market.
• Policy signals appeared more positive—Makary’s pro-innovation comments, RFK’s nod to the UPenn base-editing baby, and FDA CEO listening sessions—but Eric’s caveat is sharp: “talk is cheap,” staffing concerns remain, MFN drug pricing is “still not in a good place,” and Sam adds that the Section 232 tariff investigation still hangs over the sector.

Deep dive

1. Sanofi–Blueprint: the way biotech success stories should play out

• Eric’s framing of the $9.1B takeout is the section’s thesis: he took Blueprint public about ten years ago at an $18 IPO price, when “systemic mastocytosis sounded like Greek to me,” and sees the company as an “almost a 10-bagger over 10 years.” It is the second-largest deal of the year after Intra-Cellular Therapies, on a modest ~27% premium.
• Sam’s pushback on the perennial lament—via Adam Feuerstein—that M&A kills future Amgens: “we seem, in the biotech world, to want our cake and eat it.” Small deals “don’t count,” while big ones “could have been an Amgen.” For Sanofi, the deal fits rare disease and immunology ambitions with a possible $2B drug by 2030, if consensus is right, though he says M&A deals of this size generally will not be game changers for large pharma. He assumes most of the $2 development and $4 regulatory CVRs are achievable given their low levels.
• Brad’s angle: given Sanofi’s reputation for paying a lot for risky things that have blown up, this is a mature, well-timed strike while valuations across even good companies remain depressed.

2. Summit/Akeso: the translation box is addressed, the OS box isn’t

• HARMONi, a second-line EGFR+ lung-cancer study, was the first ivonescimab dataset with a US population alongside China. Eric says management was “quite adamant” that the geographic subsets were consistent, addressing the largest bear case, though the claim still needs to be borne out in broader data. The stock fell because OS missed at p=0.057 with all alpha spent, which may prevent an FDA filing in this relatively small market.
• Sam viewed the trial’s main value as China-to-West read-across: a Johnson & Johnson regimen is already approved for post-EGFR patients without an OS signal, while AstraZeneca’s TROP2 drug and other approaches are coming. Eric’s puzzle remains: “if you have a hazard ratio for survival anywhere in the 0.8 range… it seems like a drug that’s gonna be used. Please tell me what I’m missing.”
• Brad’s answer is the unresolved mechanism problem: Roche studied VEGF and PD-1 as separate agents in combination and never succeeded, and the old hypothesis that VEGF shrinks tumors initially but “can accelerate mets over the long term” lingers. He’s “in the camp that this is all going to succeed in the end, but we’re not there yet.”
• Sam heard bears claim the US subset was too immature to move the hazard ratio and wasn’t buying it: recruitment ran long enough that US patients could already have contributed substantially to PFS, even if OS events remain immature. Doing the “magic maths” on 0.52 overall versus ~0.47 in China might put the US HR around ~0.62—“why is that bad?” The excuse-making reminded him of the early Krazati days.

3. BioNTech–Bristol and 3SBio: the bispecifics are not all the same molecule

• Sam’s worry about BioNTech’s asset was always the PD-L1 half: poll the conference and everyone says PD-1 beat PD-L1 historically (Keytruda/Opdivo versus durvalumab/atezolizumab), but goes “all hypothetical” on PD-L1×VEGF. The Bristol deal—roughly $3.5B, which Sam characterized as pretty much guaranteed through 2028—“takes some of that risk away” without clarifying it, and creates a possible logic for combining PD-L1 as a targeting agent with a separate PD-1.
• Eric’s synthesis of why big pharma keeps showing interest: these bispecifics show properties one would not expect from a simple VEGF-plus-PD-1 combination—less of the expected systemic VEGF side effects, activity in squamous tumors, head-to-head activity versus pembrolizumab, and activity in second-line EGFR+ patients. “There’s a lot going on here… buckle up,” because Summit will stay “extraordinarily volatile,” but “over time, data will win.”
• Sam’s analyst discipline on Pfizer-partnered 3SBio: ORR softened from roughly 70% to 67.5% as the dataset grew from 24 to 34 patients, disease control stayed near 100%, and spider plots appeared to show responses deepening over time. He took that as evidence against a mere “first scan effect,” while acknowledging the interpretation could be challenged. The catch: 17% grade-3 hypertension versus ~5% for SMT112. “These molecules are different. They’re not all doing the same thing.”

4. Vera vs Otsuka in IgAN: proteinuria headlines miss what nephrologists want

• Eric describes Vera’s “round trip” week: phase 3 atacicept, a BAFF/APRIL inhibitor, delivered placebo-adjusted proteinuria reductions “kind of in the low forties” in a disease whose standard care has largely been symptomatic or nondisease-modifying, with filing planned toward year-end. Otsuka’s sibeprenlimab, an APRIL-only inhibitor, then reported numerically larger proteinuria reductions at EULAR.
• He takes direct issue with STAT News calling Otsuka’s data better: baseline variables confound the comparison, and “what kidney patients care about… isn’t whether you have lower protein in the urine”; it is slowing the downward progression of kidney function—data Vera has and Otsuka so far does not. Eric thinks the opportunity is probably $15B-plus and the space is big enough for two players.
• Brad, careful about M&A talk, still connects the dots: in the hands of a much bigger pharma, this drug in this market looks compelling, and nephrology has been an active deal space over the last couple of years.

5. Protein degradation is Brad’s “first inning” technology bet

• His case: biotech always needs a new technology to get excited about, and after Arvinas’s first-ever PROTAC phase 3 was deemed disappointing—successful, but only for about 40% of that particular breast-cancer group—Kymera’s STAT6 data landed better. Kymera reported 90% degradation from 1.5mg upward, out to 200mg, with clean safety and reductions in T2/Th2 biomarkers including TARC and eotaxin-3 in healthy volunteers. The concept is a once-daily oral Dupixent competitor; phase 1b atopic-dermatitis patient data comes in Q4, with asthma next.
• Translation to inflammatory disease remains untested because the Kymera data were from healthy volunteers, though the biomarker results were encouraging. Sam’s endorsement is structural: degradation “gives you a whole new set of targets” not amenable to standard enzyme inhibition—“a cornerstone of future drug development,” assuming molecule design and pharmacology come right. The same day, Nurix struck its own STAT6 deal with Sanofi, with BTK data due at EHA after BeiGene’s at ASH.

6. ASCO verdict: no blockbuster headline, but innovation “sneaks up on you”

• Eric’s reframing of a “down year”: there were no game-changing surprises, but standards have quietly transformed—pancreatic and melanoma response rates once celebrated at “7%, 12%” now draw less excitement at 20%–40%. His two emerging stories are PRMT5, pioneered by Bristol and to some extent Amgen, with Bristol showing ~25% response in pancreatic cancer and Tango potentially emerging; and Immatics’ PRAME-directed TCR therapy at a 55% response rate in refractory melanoma, which appeared durable with follow-up beyond a year. The KOLs they spoke with were highly enthusiastic.
• Sam said it felt like another AstraZeneca ASCO—seven straight years with a plenary, though “no standing ovation”: Imfinzi’s MATTERHORN gastric data showed a “really unprecedented” event-free-survival benefit; first-line HER2+ Enhertu data came from DESTINY-Breast09; and Gilead’s Trodelvy produced a 0.65 PFS hazard ratio in first-line PD-L1+ TNBC (ASCENT-04). His line worth keeping: “there are now people with myeloma or breast cancer who live long enough to die of something else. If you can’t call that cure, I dunno what you’re looking for.”
• Brad’s meta-point on Gilead’s biggest ASCO in years: the over-$20B Trodelvy acquisition required clinical adjustments and enormous sustained investment that “a smaller company could not have done”—an argument for scale, and credit to a program he admits “maybe us included” once criticized.

7. TIGIT’s Fc-silent second act, and Arcus’s pivot to HIF-2α

• Brad’s setup: after Roche’s SKYSCRAPER miss, BeiGene’s exit and iTeos literally closing shop, AstraZeneca and Arcus remain full speed ahead—and both TIGIT programs are Fc-silent, which both companies believe makes them “day and night different.” Sam’s simple-person logic: why would you want antibody-directed cytotoxicity against the T cell you’re trying to activate?
• AstraZeneca is “putting their money where their mouth is”: 10 phase 3s, ~8,000 patients—five NSCLC, two biliary tract, and one each in HCC, endometrial and gastric cancer—only one of them monotherapy. Susan Galbraith has cited an interferon-gamma signature signal from SITC. Rilvegostomig combination data, likely from TROPION-Lung04, showed 71% ORR in PD-L1-high and 40% in PD-L1-negative disease, with the reported cuts surpassing the monotherapy results they had seen. TROP2-related adverse events included 53% stomatitis. Trials read out in 2026–27; Sam corrected himself that GEMINI is rilvegostomig plus chemotherapy in HER2-negative BTC, with a 31% ORR.
• On Arcus’s HIF-2α, Brad’s positioning call is that two years ago everyone would have named TIGIT as Arcus’s lead; now HIF-2α—a validated target already on the market, where Arcus’s data “looks slightly better than Merck’s right now”—is clearly the main program.

8. Merus vs Bicara, and a policy week of good words awaiting follow-through

• Eric’s market-structure complaint in frontline head and neck cancer, a $3B–$4B market: physicians see both Merus’s EGFR×LGR5 program and Bicara’s ficerafusp alfa (EGFR×TGF-β) as “pretty much off the charts,” with 50%–60% response rates on top of Keytruda, about three times the response rate expected from Keytruda alone. Yet the Street has given Merus, valued around $4B-plus, winner-take-all status while Bicara trades only a little above cash.
• Brad’s bear probe: Merus appears active in both HPV-positive and HPV-negative disease, while Bicara selected the HPV-negative group. Eric’s rebuttal is that experts increasingly view the two diseases almost as different histologies, and Bicara’s TGF-β mechanism gives a biologic rationale for selecting HPV-negative patients—“turning what could be a negative into a positive.”
• Brad’s aside on leadership: of his 15 ASCO interviews, Bicara CEO Claire Mazumdar’s was his favorite—showing up on video amid a beaten-down stock and hostile analyst comparisons is “what really true leaders do.”
• The policy close: Eric heard encouraging FDA signals all week—Makary’s pro-innovation comments, the cell-and-gene meeting, RFK’s comment on the UPenn base-editing baby, and CEO listening sessions rolling out—but “talk is cheap,” staffing and resourcing concerns persist, and John Crowley said “we’re still not in a good place with MFN.” Sam’s final overhang is the Section 232 tariff investigation, which “might take several months.”