Roche scraps off-the-shelf CAR-T from Poseida deal worth up to $1.5B

Why it matters: According to Poseida’s 2023 announcement, RG6540 was designed to address two limitations of existing CAR-T treatments: the weeks required to manufacture therapies from patients’ own immune cells, and the ability of some B-cell cancers to escape CD19-targeted treatments by losing that protein. Using ready-made donor cells and targeting both CD19 and CD20 could address those problems, although RG6540 had not demonstrated those advantages in clinical trials.

• Gilead’s approved autologous CAR-T therapy Yescarta, made from each patient’s own cells, provides a benchmark for the potential time savings: a 2025 study found a median 30.6 days or more between collecting a patient’s cells and infusion. The wait was potentially associated with worse clinical outcomes. An off-the-shelf alternative would aim to shorten that interval without sacrificing safety or efficacy.
• The cut raises questions about the deal’s remaining payments. Roche’s 2025 accounts put the maximum contingent payout at $472 million (€420 million), with a fair value of $142 million (€127 million) at closing. One-quarter of the potential payment per share depends on a qualifying CD19/CD20 or BCMA/CD19 CAR-T starting a pivotal trial in an autoimmune disease by December 2034.

Zoom in: Roche removed RG6540 from its Phase 1 pipeline on October 22, alongside six other experimental drugs, without disclosing a reason.

• A Phase 1 lupus trial had already been withdrawn in August. Roche confirmed to BioSpace that the B-cell cancer trial was terminated, while the multiple sclerosis study remains listed as active but not recruiting.
• Schinecker told Fierce Biotech that Roche had stopped developing the molecule, not abandoned CD19/CD20 CAR-T. He said the acquisition had produced better candidates, although none is identified in Roche’s published clinical pipeline.

What may have happened: Pavan Bachireddy, a prominent researcher at MD Anderson Cancer Center in Houston who studies how off-the-shelf CAR-T cells fare in patients, said he couldn’t speak to Roche’s reasons. Asked more broadly what might explain such a cut, he pointed to the field’s known hurdle: patients’ immune systems can reject donor cells, cutting short how long they last.

• “We need to figure out immune rejection to enable product persistence,” he told European Biotechnology. “Once that happens, the ceiling is quite high for off-the-shelf CAR therapies in oncology.”

Reality check: A European rival is further along with the two-target idea. Germany’s Miltenyi Biomedicine asked the EMA in May to approve zamto-cel, which hits CD19 and CD20 but is made from each patient’s own cells, for adults whose large B-cell lymphoma resisted first-line treatment or returned within two years of starting it.

What’s next: Roche has yet to identify its replacement candidates, show what makes them superior, or establish publicly whether they qualify for the acquisition milestones.