SAN FRANCISCO – The CD19-directed 4-1BB chimeric antigen receptor (CAR) T-cell product JCAR017 demonstrated increased CAR T-cell expansion and persistence, and higher durability of response at higher dose levels – with manageable toxicities – in a pivotal phase 1 trial of relapsed/refractory B-cell non-Hodgkin lymphoma.
However, preliminary modeling data suggest that a therapeutic window exists for the CAR T-cell expansion, which means that development of strategies for pushing patients into that window could enhance efficacy and limit toxicity associated with JCAR017, Tanya Siddiqi, MD, of City of Hope Comprehensive Cancer Center, Duarte, Calif., reported at the ASCO-SITC Clinical Immuno-Oncology Symposium.
TRANSCEND NHL 001 is a multicenter, seamless design, pivotal trial, which started as a phase 1 first-in-human study of JCAR017, a defined composition CAR T-cell product also known as lisocabtagene maraleucel and administered at precise doses of CD4+ and CD8+ CAR T cells. Dose-finding and dose-expansion cohorts have been investigated, and currently the pivotal diffuse large B-cell lymphoma (DLBCL) cohort is enrolling, Dr. Saddiqi said, noting that dose level 2 (1 x 108 cells given as a single dose) was selected for that cohort.
The current findings are based on the TRANSCEND core population – a set of patients selected from the dose-finding and dose-expansion cohorts. This population includes patients with DLBCL not otherwise specified, transformed follicular lymphoma, or high-grade double- or triple-hit lymphomas, she said, explaining that she and her colleagues looked at prelymphodepletion baseline patient characteristics and biomarkers to assess how they related to outcomes and toxicities. This precise dosing of JCAR017 reduces variability, enabling the identification of potential patient factors associated with clinical outcomes, she said.
Response rates among 27 patients who received dose level 2 were high, with an 81% overall response rate and a 63% complete response rate.
“Patients with complete remission seemed to have more of a durable response,” she said. “In the core population at dose level 2, 50% with 6 months of follow-up seemed to remain in complete response, so there’s a dose-response effect in these patients.”
It doesn’t appear that the dose affects development of cytokine release syndrome (CRS) or neurotoxicity (NT), as the rates of these (30% and 20%, respectively) did not differ by dose level or schedule, but certain baseline features, such as a lactate dehydrogenase (LDH) level above 500 U/L and tumor burden measured as the sum of the product of diameters of 50 cm or greater, do appear to affect the development of these toxicities.
For example, 10 of 13 (77%) of patients with both of those baseline characteristics developed CRS, and 7 of 13 (54%) developed NT; for those without these characteristics, the odds of developing CRS and NT were significantly lower, she said.
Of note, higher Cmax, or peak expansion of CAR T cells in vivo, was seen at dose level 2, and exposure (area under the curve) was also higher, as was expected at that dose level, Dr. Siddiqi said.
“It also appears that there is a trend of patients with higher tumor burden to have higher expansion of their CAR T cells in vivo,” she said, noting that some of those patients were “superexpanders” with very high expansion of CAR T cells. “Similarly there were certain cytokines at the prelymphodepletion time point that seem to be higher … in patients who also go on to have higher expansion of their CAR T cells.”
These included interleukin-7, IL-15, MIP (macrophage inflammatory protein)–1alpha, and tumor necrosis factor (TNF)–alpha.
“Patients with certain higher inflammatory cytokines and biomarkers of inflammation also were noted to have higher events of CRS and neurotoxicity, so not just higher tumor burden, but also higher inflammatory state at baseline seems to affect patients in terms of getting any grade CRS or any grade neurotoxicity,” she said.
Those associated with CRS included ferritin, C-reactive protein (CRP), IL-10, IL-15, IL-16, TNF-alpha, and MIP-1beta levels, and those associated with NT included ferritin, CRP, d-dimer, IL-6, IL-15, TNF-alpha, and MIP-1alpha levels.
“Interestingly, there’s an inverse relationship between the expression of some of these biomarkers and inflammatory markers and the durability of response at 3 months,” she said.
Patients with higher tumor burden, LDH, and other markers either had no response at 3 months, or they had a very rapid response at the 1-month mark and then lost their response by 3 months, she noted.
“One of the theories could be that these patients with higher tumor burden, higher inflammatory state – they have such a high peak expansion rapidly after receiving their CAR T cells that potentially those CAR T cells may be getting exhausted or dying off very quickly, and therefore patients can lose their response,” she said.
When these data are considered together and modeled, there seems to be a therapeutic window where patients who have optimal expansion of CAR T cells in vivo may have lesser toxicity, higher overall response rates, and better durability of response, she said.
That is, on one end of the spectrum, there are patients with lower CAR T-cell expansion who have lower toxicities, but who also have lower overall response rates and lower durability of response, and on the other end, there are patients with superhigh expansion of CAR T cells, who have higher response rates, but also higher rates of toxicity, and who lose their response quickly.
“So if we can identify patients who would be in this window of optimal target expansion of CAR T cells, and if we could find strategies or mechanisms to move patients who are on either end of that spectrum into this window, we may be able to get patients with better efficacy and lower toxicities, and there could be combination strategies that could help get us there,” she said.
TRANSCEND NHL 001 is sponsored by Juno Therapeutics. Dr. Siddiqi reported serving as a consultant or adviser for Juno Therapeutics, and as a member of the speakers bureau for Pharmacyclics/Janssen and Seattle Genetics. Research funding was provided to her institution by Juno Therapeutics and several other companies.
sworcester@frontlinemedcom.com
SOURCE: Siddiqi T et al. ASCO-SITC, abstract 122.