Chimeric antigen receptor (CAR) T-cell therapy has delivered extraordinary results and has been hailed as a major clinical advance for difficult-to-treat hematological malignancies. Excitement for CD19-targeted CAR T-cell therapy has been bubbling for two years, but on July 7, 2016, the U.S. FDA delivered a major blow by putting a clinical hold on a Phase II (ROCKET) trial of Juno Therapeutics JCAR015 for adult acute lymphoblastic leukemia (ALL). Less than one week later on July 12, 2016, the hold was lifted and the trial resumed. Should we be worried about what this means for the future of CAR T-cell therapy? And what, if any, impact could this have on CAR T-cell therapy development?
Three patient deaths resulting from cerebral edema have been reported. All three patients received an intense preconditioning treatment comprising cyclophosphamide + fludarabine. Preconditioning chemotherapy aims to reduce tumor burden and reduce immunosuppressive cells, and enhance efficacy of CAR T-cell therapy. Fludarabine was added to the treatment protocol in Q2 2016 after early clinical data indicated improved efficacy with cyclophosphamide + fludarabine. Notably, the majority of patients in the ROCKET trial have received cyclophosphamide only with no treatment-related deaths.
Fludarabine has a history of demonstrating neurotoxicity, but Juno Therapeutics reports that the toxicity seen in the ROCKET study may be from the fludarabine-containing preconditioning regimen plus JCAR015 at its dose, rather than fludarabine per se. The FDA has allowed the ROCKET trial to be resumed using cyclophosphamide preconditioning only.
The impact for Juno Therapeutics’ CAR T-cell platform
- Efficacy without fludarabine: It is unknown to what extent dropping fludarabine will impact efficacy; however, cyclophosphamide only may still achieve substantial efficacy. Phase I data (from the 2016 ASCO meeting) for JCAR015 in relapsed/refractory (R/R) ALL patients treated with cyclophosphamide only, showed complete response (CR) to be 75-91% depending on the extent of disease; six-month overall survival (OS) was 57-73%.
- Risk/benefit ratio: CAR T-cell therapy is a high risk approach, and a clinical trial hold resulting from patient deaths is serious, but the FDA’s swift response to allow the ROCKET trial to resume is perhaps indicative of a positive risk to benefit ratio. CRs demonstrated by this drug class are unprecedented. For patients with R/R ALL treatment options are exhausted and prognosis is poor. There is an urgent need for new (even high risk) therapies.
- Other CAR T-cell platform candidates: JCAR014 and JCAR017, are also being tested in combination with cyclophosphamide + fludarabine; severe neurotoxicity has been observed but no unexpected events reported. JCAR014 and JCAR015 use the 4-1BB costimulatory domain whereas JCAR015 uses the CD28 costimulatory domain and this difference could, in part, explain the toxicity observed in the ROCKET study. JCAR014 and JCAR015 studies are continuing.
Repercussions for other CAR T-cell therapy developers
- Shifting stock price: Juno Therapeutics shares plummeted by more than 30% following the hold on the ROCKET study; share price was up by 20% once the hold was lifted. Shock waves were felt by other CAR T-cell players (e.g., Kite Pharma, Cellectis) as they too saw there share prices dip.
- Kite Pharma’s KTE-C19: Kite Pharma’s lead CAR T-cell candidate is being evaluated in the Phase I/II (ZUMA-1) trial with cyclophosphamide + fludarabine as the preconditioning regimen. Similar to JCAR015, KTE-C19 also uses CD28 as the costimulatory domain. Development is ongoing. On the day the clinical hold was announced for JCAR015, Kite Pharma announced that enrollment to the ZUMA-1 study had completed.
- Targeting different diseases: Kite is in pursuit of diffuse large B-cell carcinoma (DLBCL) an aggressive form of Non-Hodgkin’s lymphoma (NHL). Note: KTE-C19 is also being tested in adult R/R ALL which could sound alarm bells. The lead indication for Novartis’s CAR T-cell therapy, CTL019, is pediatric ALL; in contrast to JCAR015 and KTE-C19, CTL019 utilizes the 4-1BB costimulatory domain.
The brief clinical hold on JCAR015 has undoubtedly created a stir and caused jitters among investors, but there is no evidence to suggest that the patient deaths on the ROCKET trial were a class effect, nor is there compelling early clinical data to indicate that JCAR015 plus cyclophosphamide preconditioning regimen (i.e., the FDA agreed modified protocol) will negative impact the risk to benefit ratio to such a magnitude that will jeopardize FDA approval.
It is unknown whether and to what extent the CD28 costimulatory domain is indicative of severe neurotoxicity, and this raises potential uncertainly for other CAR T-cell therapies with the same domain. However, I think it’s too soon and unjust to extrapolate recent events to other CAR T-cell therapies. There is still so much we have to learn about CAR T-cell therapies.
Timing of regulatory submissions and the first FDA approval of CAR T-cell therapy is the million-dollar question. We could still see CAR T-cell therapies enter the market in 2017. Prior to the clinical hold for JCAR015 in June 2016, Decision Resources Group reported that Juno Therapeutics and Kite Pharma were the front runners with Novartis erring on the side of caution with more conservative timelines for regulatory submission. The clinical hold on JCAR015 was brief and the modified protocol is not expected to negatively impact recruitment; while there may be a small knock-on delay for regulatory submission and approval this is not expected to be significant. Indeed, there are other factors, such as getting a commercial manufacturing plant up and running (and have passed FDA inspection) that could come into play and impact approval and launch timings.
Watch the CAR T-cell therapy space with Decision Resources Group. See our recent blog: The CAR T-Cell Drive to Market and content on upcoming offerings on the ALL Disease Landscape & Forecast and update to the NHL and CLL Disease Landscape & Forecast