After leukemia treatments failed, Alyssa Tapley, at age 13, was told she would die. Then, doctors tried an experimental gene-edited therapy. She became the first human to try the treatment made possible by U.S. federal funding. Now 16, she’s cancer-free and planning her future. cbsn.ws/4ou0OMt

We're launching Challenge 1 of the Anthropic x Adaptyv Protein Design Competition on @proteinbase: design a conditional binder to EGFR, a key cancer target. The goal: bind EGFR under the acidic conditions of the tumor environment, but not at the normal pH around healthy cells. Moreover, the proteins must bind to both the human and the mouse variant of EGFR to support easier preclinical testing

ALLO CAR-T is shrinking lung cancer brain metastases. That same product, MT027, is cleared for U.S. Phase 2 in recurrent GBM. @KiraGenBio is building for broader, lasting control: dual targeting + edits designed to resist tumor suppression. ALLO CNS CAR-T is HERE!!!🚀

Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family. It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5.

#China #immunotherapy #CAR_T 🦉 A brain response can lose to cancer elsewhere. MT027 is delivered into the cerebrospinal fluid, bringing B7-H3 CAR-T cells close to brain metastases without relying on delivery through the bloodstream. Two of the three patients with measurable disease had partial responses, each with about 80% tumor shrinkage. That is a signal for local treatment, but the denominator matters: five patients were treated, and the other two were not included in the response analysis after overall disease progression or changes in clinical condition. The next test is whether intracranial control can be sustained alongside control of cancer elsewhere. These early data cannot establish that local CAR-T offers a survival advantage—or that it works better in the brain than a systemic B7-H3 therapy. #T_MAXIMUM_Pharmaceutical 👉prnewswire.com/news-releases…

#immunotherapy #CAR_T #oncology 🦉 In GBM, CAR-T failure may start with the patient’s own T cells. CAR-T engineering usually assumes the cell is a workable chassis and the receptor is the part that needs improvement. This paper complicates that assumption. CAR-T cells made from recurrent GBM patients showed greater pre-treatment exhaustion and produced substantially less survival benefit than the same CD133-targeting strategy built from healthy-donor T cells. The defect was visible before the cells ever encountered tumor antigen. That matters because GBM may be damaging the raw material used to manufacture its own therapy. Peripheral immune dysfunction is therefore not just another obstacle CAR-T encounters after infusion; it may already be embedded in the final product. Adding more sophisticated receptors or exhaustion-resistant circuitry cannot necessarily erase the biological history of the starting T cell. The authors’ allogeneic, TRAC-edited CD133 CAR-T offers one way around that problem, but the evidence remains preclinical and the patient comparison was small. NSG mice also cannot model host rejection of donor cells. Still, the conceptual shift is useful: in GBM, the next CAR-T improvement may need to begin before CAR design—with the choice of the cell itself. pmc.ncbi.nlm.nih.gov/article…

Same thing @2048vc We never passed saying too early

#immunotherapy #CAR_T @FDA #China 🦉 FDA cleared IASO208, a CD20 in vivo CAR-T, for a U.S. Phase 1b after just 11 patients in a Chinese IIT. Skipping leukapheresis and lymphodepletion is the real test; efficacy and durability remain largely unproven. prnewswire.com/news-releases…

CAR-T therapies engineer patients’ immune cells to kill cancer and they have achieved remarkable results in blood cancers that resist other treatments. Solid tumors (e.g. liver, breast cancer) have proved much harder -- until now. This new report of one patient in a Phase I trial offers a very promising example of progress: a three-year-old with chemotherapy-resistant metastatic liver cancer achieved complete remission lasting at least a year -- a truly breakthrough result considering the severity of this cancer. The history of how this results came about shows how CAR-T progress depends on repeated cycles of engineering, testing in patients and learning from the results. Most of this iterative learning happens in Phase I trials like the current one, which shows how important such studies are for driving medical progress forward. One additional difficulty for solid tumors is that they suppress immune activity, which in turn makes it difficult for CAR-T cells to sustain an attack. So to extend CAR-T’s success in blood cancers to solid tumors, researchers had to help the engineered cells survive in this more hostile territory. In 2020, the team’s earlier clinical work across 4 Phase I trials tested CAR-T cells targeting glypican-3, a protein on certain solid tumors. In 12 patients, these cells produced no objective tumor responses. Another 12 received cells engineered to also produce IL-15, a naturally occurring immune signaling protein that helps T cells survive and multiply. Although 4 patients responded, none achieved complete remission. But the responses were promising and revealed that there was value in giving CAR-Ts their own survival signals. So the researchers doubled down on this lesson. In this new Phase I study, built on the learnings from the previous one, researchers engineered the expression of two different cytokines – IL-15 and IL-21. They also added an emergency off switch that allows doctors to destroy the engineered cells if necessary to reduce toxicity. Remarkably, a first infusion shrank the cancer of the 3 year old patient and a after a second, the remaining tumor disappeared. A year later the patient remained disease-free. There is an obvious caveat here: one patient cannot definitively establish how reliably the treatment works. But the results are striking enough to offer real hope that this would work in at least a subset of patients. I believe this development beautifully illustrates why Phase I trials matter: they let researchers observe how engineered cells behave inside patients and carry those lessons into further development. I am committed to making these studies easier to run in order to accelerate this cycle of iterative learning.

The response to this opinion on social media has been a festival of reflexive praise. However, the article is intellectually unserious. It takes a small secondary issue and inflates it into the central obstacle to curing cancer and improving human health, while ignoring the two fundamental limitations dominating modern medicine: 1.) We still do not understand human biology well enough to design drugs that reliably work. 2.) We have a health-care system that cannot afford the drugs that occasionally do. Faster Phase I trials will mostly allow us to discover faster that yet another supposedly promising cure has failed. These trials primarily determine safety, activity, and dosage. They do not establish that a drug improves health. Likely much more than 90 percent of oncology drugs entering Phase I never become approved drugs. Even candidates that get to Phase III, the final major testing stage, frequently fail because they do not work. That is not a paperwork problem. It reflects our primitive understanding of health and disease and our inability to predict how interventions will behave in the human body. The most important role for AI is not generating more tiny, uncontrolled Phase I datasets. It is building better models of human biology that predict pharmacology, toxicity and efficacy before patients are exposed. The goal should be to make drug development less dependent on enormously expensive, high-failure-rate trial and error, which is what clinical trials are. The second limitation is economic. We already have cancer and other drugs costing more than $100K per year, sometimes approaching $500K, while producing modest benefits for relatively small numbers of patients. Many patients would require multiple such drugs over their lifetimes. There is no plausible way to scale that cost across the population. Producing more experimental drugs without solving the scientific and economic problems is not health-care abundance. It is an abundance of low-probability experiments followed by an abundance of drugs that patients and society cannot afford. By all means, eliminate pointless bureaucracy. Academic medical centers charge too much for trials, pile on overhead and spend months negotiating budgets while creating far too many administrative obstacles. I have personally had clinical researchers approach me asking for roughly $800K to run a Phase I trial based on my research. The FDA was not stopping them. They were looking for someone to finance a low-probability experiment that would generate research money, pad their CVs and help them win a promotion in a system that rewards scientists for how much money they bring in. Faster Phase I trials do not solve the central problems. Our science remains inadequate and our health-care system remains broken.

ArsenalBio has probably been the company to watch in the CAR-T field. Top notch founders and incredible smart team, which ended up not being enough. Despite people telling me I’m crazy or that this tech will not work, I’ve been a believer in in vivo cell reprogramming for a few years now, and it looks like ArsenalBio may have concluded that in vivo CAR-T is probably the way to go. fiercebiotech.com/biotech/ar…