Welcome back to This Week in Biotech by Biotech Blueprint, edition 116, covering biotech and pharma news from August 28 to September 3, 2026.
Biotech’s next frontier
Genomic medicine is at an inflection point. What comes next has even the skeptics feeling hopeful.
🎙️ Biotech Blueprint brings you weekly video updates on the latest biotech and pharma news, plus in-depth podcast interviews with industry leaders. You can find us on YouTube, Spotify, and Apple Podcasts. If you want to be featured on our podcast, please get in touch.
📩 For consulting inquiries, please email me.
👥 If you haven’t already, connect with me on LinkedIn.
🛒 And make sure to check out our MERCH STORE.
VIDEO SUMMARY
THIS WEEK’S KEY TAKEAWAYS 🔑
You may remember the Replimune Tudriqev drama from this July and August. The company had already gotten two complete response letters in the past, but this time the drug was approved by the FDA. On Wednesday, internal FDA memos were released and showed that Asha Das, director of the biologics center’s office of clinical evaluation, overruled her review team to give Tudriqev accelerated approval on August 6 even though the reviewers wanted to reject it for the third time. Last week, I reported on Revolution Medicines’ Rasonque receiving approval in just 35 days under the Commissioner’s National Priority Voucher. Two different stories, but both show regulatory outcomes drifting away from the evidence and toward whoever holds seniority. This is pretty hard to underwrite.
The second theme this week is China. GSK paid Hutchmed $110 million upfront for a preclinical EGFR-KRAS conjugate, Roche paid Simcere $75 million upfront for a trispecific T cell engager, and Akeso’s ivonescimab beat Keytruda on overall survival in a China-only Phase 3. In the same week, the acting heads of CDER and CBER plus the directors of the device center and the Oncology Center of Excellence published a commitment to expand foreign site oversight and add inspectors at the agency’s China offices, after reporting surfaced three deaths in Chinese trials that were never disclosed when they happened. Big pharma keeps buying Chinese assets, and the FDA is now openly acknowledging it can't inspect most of the trials behind them.
The third theme is what companies are willing to tell you. Akeso and Summit announced an overall survival win against the biggest drug in oncology and published no hazard ratio, no median survival and no enrollment number. Novartis announced two positive Phase 3 multiple sclerosis trials with no annualized relapse rate. Ultragenyx announced a Phase 3 failure and disclosed no effect sizes either, which is the newer development. The practice has spread from wins to losses. The exception this week was AbbVie, which put out a full CERVINO table with a hazard ratio, a confidence interval, a p-value and grade by grade safety.
Finally, Trump’s most favored nation pricing reached the mid cap biotechs. Nine companies signed, and the discounts land in Medicaid in exchange for protection in Medicare and from tariffs. Medicaid is where prices were already lowest.
Have a great long weekend and I will be back next week with This Week in Biotech #117.
BIOTECH/PHARMA NEWS 🧬
🔹 Newly released agency documents show that Asha Das overrode her review staff in granting accelerated approval to Tudriqev plus nivolumab on Aug. 6, four days past the PDUFA date and eight days after a 10-3 favorable advisory committee vote. The review team had recommended rejection, which would have been the third complete response letter in 13 months after July 2025 and April 2026, both over whether the single arm IGNYTE data could show the oncolytic virus contributed anything on top of nivolumab. The approved label already encodes the reviewers’ objection. The efficacy population is 91 patients with at least one noninjected lesion, response rate 24.2%, median duration 14.1 months, not the 33.6% and 24.8 months that advisory committee voted on. What’s interesting and new here is the documentation of a formal internal disagreement resolved by seniority.
🔹 GSK licensed Hutchmed’s HMPL-A830 for $110M upfront and up to $1.19B in milestones and royalties, taking global rights outside mainland China, Hong Kong, Macau and Taiwan for an antibody-targeted therapeutic conjugate that links an EGFR antibody to a KRAS-blocking small molecule payload, entering the clinic this year. Roche followed with $75M upfront to Simcere Zaiming for a preclinical trispecific T cell engager, its second China deal in two weeks. Simcere has now signed five licensing deals since early 2025 worth up to $5.66B. Days later, four FDA center leaders published an editorial committing to more Bioresearch Monitoring inspectors in China, remote assessments, reviewer training to flag problem sites, and public disclosure when human-subject concerns arise. Their instruction to sponsors was blunt: treat the inability to inspect or access data as a material factor in regulatory strategy. China ran under 8% of global trials in 2010 and passed the United States on annual registrations in 2020.
🔹 Reporting this week detailed three deaths in Chinese investigator-initiated trials that were not disclosed when they occurred. A boy died in Aug 2025 after a HuidaGene CRISPR therapy for Duchenne muscular dystrophy, with public updates on the trial halted for more than a year. A 6 year old girl died after brain-targeted gene editing at Xinhua Hospital, and her death was never announced. The investigators published their results without it. And a man in his 50s with systemic sclerosis died in March 2026 after RiboX’s in vivo CAR-T, five months before the FDA cleared that program in August. Representatives John Moolenaar and Ben Cline have asked the agency to reject China-generated data unless sites are audited annually. There’s a scientific thread underneath the regulatory one. Two of these deaths involved viral delivery of gene editors, an approach US researchers had largely moved away from over exactly these safety concerns. The regulatory question is whether an investigator-initiated trial in a country the FDA can’t readily inspect should feed a US filing at all.
🔹 The White House announced agreements with Alcon, Astellas, BeOne Medicines, BridgeBio, CSL, Kyowa Kirin, Sun Pharma, Teva and UCB on August 31, with Incyte signing separately. Each aligns outpatient prices for state Medicaid programs to the lowest rate charged abroad through the CMS GENEROUS model, which expires in 6 years. In exchange the companies are exempt from most favored nation pricing in Medicare and from Section 232 pharmaceutical tariffs, and they will list on TrumpRx. Collectively they committed at least $19.6B to US manufacturing, and the administration projects more than $600B in savings. The math is less generous than that. Medicaid has had mandated steep statutory rebates since 1990, Medicaid drug spending was $54B in 2024 against Medicare’s $163B, and RBC’s Brian Abrahams notes Medicaid is under 2% of sales for BridgeBio’s Attruby. Manufacturers traded a discount in their smallest channel for protection in their largest one.
🔹 The FDA approved Mimrylo (rusfertide) on August 28 for erythrocytosis in adults with polycythemia vera, the first hepcidin mimetic of any kind. It’s a once weekly subcutaneous peptide that restrains red cell production through iron regulation rather than suppressing the marrow. In the Phase 3 VERIFY trial, 293 phlebotomy-dependent patients were randomized to rusfertide or placebo on top of standard care, and 76.9% of rusfertide patients achieved a clinical response during weeks 20 through 32. Protagonist discovered the molecule and ran it through Phase 3. Takeda holds global rights. The day before, Lisraya (brepocitinib) became the first oral therapy approved for dermatomyositis, a JAK1 and TYK2 inhibitor from Priovant, on the strength of VALOR, a 241-patient trial of 15 mg, 30 mg and placebo that read out in the New England Journal of Medicine in March and added skin endpoints in JAMA Dermatology last month. Two rare diseases, two mechanisms nobody had approved before, and no advisory committee for either.
🔹 Ultragenyx’s intrathecal antisense oligonucleotide apazunersen, designed to unsilence the paternal UBE3A allele in Angelman syndrome, missed its primary endpoint of change from baseline in the Bayley-4 cognitive raw score, and its key secondary of net response on the Multidomain Responder Index. Ultragenyx reported no differences between treated and control groups on the index or on mean changes across its five component endpoints, and released no effect sizes at all. Safety matched Phase 1/2. Shares fell 44.8%. Consensus had carried peak sales above $1.8B. Leerink’s Joseph Schwartz said the company’s language “leaves very little room for an optimistic interpretation” and that the investment case has shifted from pipeline to commercial execution. His revised peak estimate for UX111 in Sanfilippo syndrome is $120-$240M. Ultragenyx is weighing significant expense reductions and still targets profitability in 2027. Ionis fell 4% in sympathy. The Phase 1/2 signal here was strong enough to convince everyone, and that’s the lesson. Open label neurodevelopmental gains don’t survive a sham control.
🔹 A prespecified interim analysis of Akeso and Summit’s HARMONi-2 trial met the key secondary endpoint of overall survival against pembrolizumab monotherapy in first-line lung cancer patients with PD-L1 tumor proportion score of at least 1%, per the independent data monitoring committee. No hazard ratio, no median survival and no enrollment figure were released. UBS’s David Dai estimated roughly a 24% relative reduction in risk of death, which is an analyst’s guess rather than data. Summit rose as much as 18%, Akeso closed up 3% in Hong Kong. This is the survival result the class needed after the 2024 progression-free survival win left the question open and HARMONi-6 drew arguments about how Chinese patient characteristics translate. It’s still a China-only trial, and US approval rests on the global HARMONi program, whose survival data are expected in 2028. Beating Keytruda head to head anywhere is a genuine event. Doing it without publishing a single number is this week in miniature.
🔹 AbbVie’s B cell maturation antigen T cell engager etentamig cut the risk of progression or death by 60% versus investigator’s choice of standard therapy in 393 patients with a median of three prior lines of relapsed/refractory multiple myeloma. Progression free survival hazard ratio 0.40 (95% CI 0.29 to 0.54, p<0.0001), response rate 74.0% versus 45.7% (p<0.0001), 12 month overall survival 87.9% versus 72.0% (HR 0.48, nominal p=0.0012) at 11.4 months median follow up. Cytokine release syndrome occurred in 28.3%, almost all grade 1, with no grade 3 or higher events and 0.9% grade 1 neurotoxicity. Grade 3/4 infections ran higher than control at 27.7% versus 19.2%, but fatal infections were lower at 1.5% versus 3.1%, and discontinuations were 3.6% versus 9.6%. AbbVie will discuss next steps with regulators.
Have a great rest of your week and thanks for reading Biotech Blueprint!
DISCLAIMER: This publication is for informational and educational purposes only and does not constitute investment, legal, medical, or tax advice, a solicitation, or an offer to buy/sell any security. Information is believed reliable but no warranty is made as to accuracy or completeness; views may change without notice. Do your own research and consult qualified professionals.
DISCLOSURE & CONFLICTS: The author may hold positions in securities mentioned and may change positions at any time without notice. No compensation is received from companies mentioned, and there are no known material business relationships unless explicitly stated. Content may reference clinical data and regulatory events; always consult primary sources and a licensed clinician for medical decisions. Past performance is not indicative of future results.
LIABILITY: Use of this content is at your own risk. The author assumes no responsibility for any losses arising from reliance on this material.
Front cover image source: Madeline Monroe/C&EN/Shutterstock
Hi. If interested.
2 gems in china
ImmuneOnco
Laekna
Both strong cash position, very interesting pipeline, loads of data H2