Dramatic Improvement in NHL Outcomes With CAR T-Cell Versus Standard Therapy

Sattva S. Neelapu, MD

Treatment with the autologous anti-CD19 CAR T-cell therapy axicabtagene ciloleucel (axi-cel) significantly improved outcomes in refractory non-Hodgkin lymphoma (NHL) compared with standard therapies, a standardized analyses of 2 clinical trials showed.

Three-fourths of patients treated with axi-cel achieved objective responses as compared with 20% of patients who received standard therapy. More than half of the axi-cel cohort had complete responses versus 6% of conventionally treated patients.

Standardized survival analyses yielded a 6-month overall survival of 77% with axi-cell and 35% with conventional treatment for refractory NHL, Sattva Neelapu, MD, reported at the 2017 Society of Hematologic Oncology annual meeting.

“Standardized analyses suggest that treatment with axi-cel in the ZUMA-1 trial yielded 8 to 10 times higher odds ratios for overall response and complete response, respectively, relative to currently available therapies,” Neelapu, associate professor of lymphoma and myeloma at MD Anderson Cancer Center, concluded in a poster presentation.

“The estimated hazard ratio suggests an overall 77% reduction in the risk of death for patients treated with axi-cell in ZUMA-1 versus those treated with currently available therapies in the SCHOLAR-1 trial. Sensitivity analyses with ECOG performance status as a covariate yielded consistent results.”

The ZUMA-1 study was the first multicenter pivotal trial of axi-cel (formerly KTE-C19) in patients with refractory, aggressive NHL. Initial results, reported earlier this year at the AACR annual meeting (abstract CT019), showed an 82% objective response rate, including complete responses in 54% of 101 patients.

Based on data from ZUMA-1, the FDA granted a priority review to axi-cel in May 2017 for transplant-ineligible patients with relapsed or refractory NHL. Under the Prescription Drug User Fee Act, the FDA is scheduled to make its decision by November 29, 2017.

SCHOLAR-1 was a pooled analysis of data for 635 patients with refractory, aggressive NHL treated with currently available therapies. The results showed low response rates and a 2-year survival of 20% (Blood. 2017;doi:10.1182/blood.2017.03.769620 [epub]).

Neelapu and colleagues analyzed the ZUMA-1 data within the context of outcomes achieved with available therapies used in SCHOLAR-1, performing a comparative analysis that adjusted for imbalances in key covariates.

The studies had balanced distribution of sex, disease type, and the proportion of patients with disease refractory to 2 consecutive lines of therapy. ZUMA-1 included more patients with 3 or more prior lines of therapy (69% vs 26%), with advanced (stage III-IV) disease (85% vs 67%), and less favorable IPI scores (≥3, 48% vs 35%). More patients in ZUMA-1 had transformed follicular lymphoma or primary mediastinal large B-cell lymphoma (24% vs 4%), and more patients in SCHOLAR-1 had primary refractory disease (20% vs 2%).

The patient population for SCHOLAR-1 comprised 2 phase III trials conducted at centers in the United States and Canada. Patients received available therapies for refractory NHL. Maximum follow-up for survival ranged from 7.6 to 14.8 years across participating centers.

Patients in ZUMA-1 received the target dose of axi-cel (2 x 10⁶ cells/kg) after low-dose conditioning with cyclophosphamide and fludarabine for 3 days. The modified intention-to-treat population involved 101 patients who received axi-cel. The trial had a median survival follow-up of 8.7 months.

Standardized analyses included prespecified covariates selected for weighting: refractory disease status and whether a patient proceeded to stem-cell transplant after determination of refractory status. A sensitivity analysis incorporated ECOG performance status as a covariate for standardization.

The primary analyses showed an objective response rate of 82% and complete responses in 54% of patients in the ZUMA-1 study, compared with 26% and 7% in SCHOLAR-1.

The standardized comparison yielded an objective response rate of 74% and complete responses in 51% of patients versus 20% and 6% in SCHOLAR-1. The results for overall response rate translated into a standardized difference of 0.55 and an odds ratio of 7.97 in favor of ZUMA-1 (P <.0001). The results for complete response resulted in a standardized difference of 0.45 and an odds ratio of 10.2 in favor of ZUMA-1 (P <.0001).

The median overall survival had yet to be reached for the ZUMA-1 cohort in either the primary analysis or standardized comparison. Medians for the SCHOLAR-1 study population were 6.3 and 3.9 months, respectively. The 6-month survival in the primary analyses was 80% in ZUMA-1 and 53% in SCHOLAR-1. The standardized comparison yielded 3-month survival of 93% for ZUMA-1 and 60% for SCHOLAR-1, and 6-month survival of 77% and 35% for ZUMA-1 and SCHOLAR-1, respectively (HR, 0.23; P <.0001).

The sensitivity analysis yielded response rates of 82% and 34% for ZUMA-1 and SCHOLAR-1, respectively, which translated into an odds ratio of 12.9 in favor of ZUMA-1 (P <.0001). Complete response rates were 54% in ZUMA-1 and 15% in SCHOLAR-1, resulting in an odds ratio of 11.1 in favor of ZUMA-1 (P <.0001).

Finally, the sensitivity analysis of survival showed that the median had yet to be reached in ZUMA-1 compared with 5.8 months in SCHOLAR 1. The 3- and 6-month survival was 95% and 80% in ZUMA-1 versus 74% and 49% in SCHOLAR-1. The differences represented a 66% reduction in the hazard ratio in favor of ZUMA-1 (P <.0001).

“While adjustments are made for imbalances in a limited number of prognostic factors and comparisons are not based on a randomized study, axi-cel represents a significantly improved treatment option for patients with refractory, aggressive NHL compared with currently available therapies as used in the SCHOLAR-1 study,” Neelapu concluded.

Neelapu SS, Locke FL, Bartlett NLm et al. SCHOLAR-1 versus ZUMA-1: A standardized comparison of outcomes in patients with refractory, aggressive non-hodgkin lymphoma (rNHL). Presented at 2017 SOHO Annual Meeting. Abstract NHL-023.