ASCO18: Celgene makes its case as CAR-T rival in lymphoma
- Celgene's experimental cell therapy JCAR017, or liso-cel, could prove competitive in lymphoma with already on-market treatments from Gilead and Novartis, suggest updated clinical results presented Sunday at the annual meeting of the American Society of Clinical Oncology.
- Importantly, data from 102 evaluable patients offer further evidence liso-cel may be less risky, with comparatively lower rates of severe cytokine release syndrome (CRS) and neurotoxicity — two common side effects of treatment with CAR-T.
- Investors will likely focus on the response rate at six months, which has become a key marker to gauge the efficacy of CAR-T. On that measure, liso-cel led to responses in 49% of the 37 patients with relapsed or refractory B-cell non-Hodgkin lymphoma who met the dose and disease criteria for Celgene's pivotal study. The complete response rate for that group was 46%.
Succeeding with liso-cel is crucial for Celgene, which spent $9 billion in January to acquire the drug and its maker Juno Therapeutics. While buying Juno gave Celgene a broader CAR-T pipeline, liso-cel is the nearest to market and already figures heavily into the company's future revenue forecasts.
The data presented at ASCO offers some reassurance that investment was not misplaced.
"Liso-cel is clearly competitive with [Gilead's Yescarta] and [Novartis' Kymriah] in terms of efficacy, durability of response and safety," said Caron Jacobson, medical director of the Immune Effector Cell Therapy program at the Dana-Farber Cancer Institute, in remarks Sunday following presentation of the data.
Yet the way Celgene is cutting the data from its study, called TRANSCEND, means the response rates highlighted in its release are from a small number of patients.
Among all 102 evaluable patients in the full data set — which includes multiple dose levels and a broader range of lymphoma patients — the overall and complete response rates checked in at 40% and 34%, respectively.
For the purposes of its pivotal study, Celgene is focusing on patients in a narrower "core" group who received the higher dose of 100 million CAR-T cells. Results from 73 patients in the core group were reported, 37 of whom were given the selected dose.
If the response rates seen in the core group hold up over more patients, liso-cel would be broadly competitive with the efficacy demonstrated by Yescarta (axicabtagene ciloleceul) and Kymriah (tisagenlecleucel).
"[Liso-cel] still has the smallest efficacy database ... but is up significantly from the last data cutoff of [14 patients]. Thus, data is pointing more in the direction of being a best-in-class CART on efficacy in DLBCL in our view," wrote Evercore ISI analyst Salim Syed in a June 3 note to investors.
CAR-T therapy efficacy in lymphoma
|TRANSCEND, full||TRANSCEND, core||ZUMA-1||JULIET|
|Best ORR (# of pts.)||75% (102)||80% (73)||82% (101)||53% (81)|
|Best CR (# of pts.)||55% (102)||59% (73)||54% (101)||40% (81)|
|ORR, 6 months (# of pts.)||40% (102)||47% (73)||41% (101)||37% (46)|
|CR, 6 months (# of pts.)||34% (102)||41% (73)||36% (101)||30% (46)|
SOURCE: ASH 2017 presentations, New England Journal of Medicine, companies
Quick comparisons of response rates, however, can mask important differences in how studies of the three CAR-T therapies were conducted — a point Dana-Farber's Jacobson noted.
"We always want to caution on comparing across studies, but I think we should be particularly cautious in this context," she said, citing the use of bridging chemotherapy in Novartis' trial as well as different levels of disease burden among patients studied.
Treatment with liso-cel led to severe CRS in only 1% of patients and to severe neurotoxicity in 13% — lower than what was observed in trials of Yescarta and Kymriah. Across the full data set, more than half of patients given Celgene's CAR-T didn't experience either CRS or neurotoxicity.
That safety profile gave Juno, and now Celgene, the confidence to continue moving forward with testing liso-cel as an outpatient treatment.
"If there is one product of the three that can go outpatient, it is JCAR017," said Wim Souverijns, head of global marketing in hematology and oncology at Celgene, in an interview.
Delivering CAR-T as an outpatient treatment could be particularly important for hospitals, lowering administration costs and freeing up bed space.
In the non-pivotal cohort of Celgene's study, eight patients were treated in the outpatient setting. Novartis is also exploring whether patients need to remain in the hospital following CAR-T treatment: A full quarter of participants in the Swiss pharma's JULIET study were treated as outpatients.
Celgene, though, is confident that liso-cel can still compete even though it will be coming from behind.
"There is potential to really lead this market even if you're third — if your data are holding up and if you do outshine competitors from a service perspective," Souverijns said.
Follow Ned Pagliarulo on Twitter