Survival after anti-CD19 CAR T-cell therapy for B-cell non-Hodgkin lymphoma and CLL with 10 years of follow-up Open Access

TO THE EDITOR:

Large B-cell lymphoma (LBCL), the most prevalent subtype of non-Hodgkin lymphoma (NHL), has a poor prognosis for patients who experience relapse.¹ The poor outcomes of patients with relapsed or refractory LBCL present an unmet need for effective treatment options, as does chronic lymphocytic leukemia (CLL) that is refractory to prior therapies.² Chimeric antigen receptor (CAR) T-cell therapy was developed to address this therapeutic gap. A phase 1 clinical trial conducted at the National Cancer Institute Surgery Branch investigated the safety and efficacy of anti-CD19 CAR T-cells using the FMC63-28Z CAR in the treatment of NHL and CLL.^3-5 FMC63-28Z–expressing CAR T-cells were the earliest CAR therapy to show antigen-specific activity in humans, and FMC63-28Z CAR T-cells were developed into axicabtagene ciloleucel (axi-cel), the first CAR T-cell therapy for lymphoma.^6,7 This initial trial of FMC63-28Z CAR T-cells demonstrated early evidence of responses and durability in B-cell malignancies, even in chemotherapy-refractory patients.^4,5,8 Previous publication of 5-year follow-up data of axi-cel for LBCL demonstrated an estimated 5-year event-free survival (EFS) of 30.3% and overall survival (OS) of 42.6%.^7,9 We report long-term outcomes after FMC63-28Z anti-CD19 CAR T-cells.

Patient data are derived from a single-center phase 1 clinical trial of anti-CD19 CAR T-cell therapy (NCT00924326), as previously described,⁸ or from a gene therapy long-term follow-up protocol (NCT00923026; Table 1; supplemental Methods). The baseline malignancy burden for patients with LBCL in this cohort is similar to that of the ZUMA-1 cohort, with a median sum of the product of the diameters of 36.56 cm² vs 37.23 cm², respectively (Table 1).⁹ Patients with LBCL had received a median of 4 lines of prior therapy, similar to the ZUMA-1 cohort^7,10 (Table 1).

Achieving complete remission (CR) was not significantly associated with malignancy subtype, baseline sum of the product of the diameters, number of lines of prior therapy, prior autologous stem cell transplant, or chemotherapy refractoriness. Patients achieving CR were older, but this was not statistically significant (Mann-Whitney U test, P = .0522.). As previously reported,⁸ patients who achieved CR had higher peak blood levels of CAR-positive cells than those who did not (P = .0041).

The median follow-up time for EFS was 5.53 years, with a median EFS of 5.01 (95% confidence interval [CI], 0.699 to not reached [NR]). There were no significant differences in EFS between malignancy subtypes (Kaplan-Meier log rank, P = .941; Figure 1A). The 5-year EFS was 52.2% (95% CI, 39.1-69.5) for all patients and 52.1% (95% CI, 36.2-75.0) for patients with LBCL. The 10-year EFS for all patients was 45.2% (95% CI, 31.7-64.3). The 5-year EFS for patients achieving a CR was 87.0% (95% CI, 74.2-100; Figure 1B; supplemental Results). The median follow-up time for OS was 10.2 years; the median OS was NR (95% CI, 7.31 to NR). The 10-year OS was 58.1% (95% CI, 44.6-75.8) for all patients and 50.7% (95% CI, 34.7-74.3) for patients with LBCL (supplemental Results). There were no significant differences in OS between subtypes (log rank, P = .396; Figure 1C). The 10-year OS for patients achieving CR was 81.9% (95% CI, 67.3- 99.7; Figure 1D; supplemental Results).

Although the 5-year EFS for patients on this trial with LBCL is numerically higher than that observed in the long-term follow-up of the ZUMA-1 clinical trial of axi-cel for LBCL (52.1% vs 30.3%),⁹ the CI for the EFS estimate in our study is broad and overlaps with the CI reported in ZUMA-1 (95% CI, 36.2-75.0 vs 21.5-39.6),⁹ making comparisons difficult. Any potential differences in efficacy between the trials may be attributable to subtle differences between the patient populations, possibly due to careful selection of patients treated initially in the early CAR T-cell era. Because much of this trial was conducted before current CAR T-cell toxicity management protocols were established, only 1 patient on the trial received corticosteroids; only 4 patients received tocilizumab. High cumulative corticosteroid doses have been associated with shortened progression-free survival.¹¹ However, due to the limited number of patients and confounding factors, no firm conclusions can be drawn regarding the effect of toxicity management in this trial on malignancy outcomes.

An univariable analysis of EFS identified older age as having a lower hazard for an event (hazard ratio [HR], 0.929; 95% CI, 0.892-0.968; P = .000414). Forward selection multivariable analysis of EFS identified older age as still having a lower HR when controlling for other variables; no other significant associations were identified (age HR, 0.920; 95% CI, 0.865-0.9785; P = .00799). In the univariable and multivariable Cox proportional hazards regression analysis for OS, there were no significantly associated baseline patient factors.

Consistent with other reports,^9,12 increasing age is a factor that is associated with improved EFS but not OS. This effect appears to be particularly relevant with axi-cel.¹² This association persists in our analysis when controlling for other covariates. The underlying mechanism for this phenomenon remains unclear, but it is hypothesized that younger patients may have particularly biologically aggressive tumors or there may be a selection bias for older patients with higher performance status to be offered CAR T-cell therapy.¹² 

Furthermore, consistent with other reports,^13-15 the most common cause of mortality was progression of primary malignancy, followed by infections and second malignancies (supplemental Table 1). This underscores the importance of developing CAR T-cell therapies with improved efficacy, as well as the necessity for vigilance in identifying potentially preventable adverse events such as infections in the posttreatment population. Long-term surveillance for second malignancies is also essential.

Seven patients (16.3%) experienced 8 second solid tumor malignancies, excluding nonmelanoma skin cancers. Four patients (9.3%) had new hematologic malignancies, including 3 cases of myelodysplastic syndrome and 1 new B-cell neoplasm, although it was not determined whether this neoplasm was clonally related to the primary B-cell malignancy (Figure 1E; supplemental Table 2). In total, 25.6% of patients experienced new second malignancies during a median follow-up of 8.05 years (Figure 1F). There were no T-cell malignancies. At 3 years after treatment, the cumulative incidences of second solid tumor and hematologic malignancies were 6.1% and 3.3%, respectively. Patients who developed second malignancies were older at the time of CAR T-cell infusion than those who did not (Mann-Whitney U test, P = .0359). Patients who received ≥4 prior lines of therapy had a higher proportion of individuals developing second malignancies than those who received ≤3 lines (Fisher exact, P = .0014). Patients who developed second malignancies did not have significant differences in the rate of prior autologous stem cell transplant or in peak blood levels of CAR-positive cells.

Individuals with NHL are known to have an increased risk of second solid and hematologic malignancies compared to the general population.^16,17 Our analysis shows rates of myeloid malignancies comparable to other reports of patients after CAR T-cell therapy.¹⁸ Rates of solid tumor malignancies are numerically higher than some reports,^19,20 but this may reflect the long follow-up and high OS rate in these patients, thus leading to survival bias in this group. The 3-year cumulative incidence of solid tumor and hematologic malignancies in our trial is similar to a recent report of patients who received anti-CD19 CAR T-cell therapy for lymphoma.²¹ An association between older age and a higher likelihood of second malignancies has been previously reported after CAR T-cell therapy.^18,21 Additionally, more heavily pretreated patients had a greater risk of developing second malignancies. Increased exposure to cytotoxic therapy is a known risk factor for subsequent development of second myeloid malignancies.²² It is unknown whether lymphocyte-depleting chemotherapy, long-term B-cell depletion due to CAR T cells, or inflammatory states such as cytokine release syndrome could contribute to the development of second malignancies. Use of CAR T-cell therapy earlier in a patient's treatment course, before exposure to multiple cytotoxic agents, could potentially prevent the development of secondary myeloid malignancies.

This study is, to our knowledge, the longest follow-up report of patients who have received CAR T-cell therapy for lymphoma. Results underscore the necessity of ongoing surveillance for second malignancies in ongoing survivors. This study also demonstrates that anti-CD19 CAR T-cell therapy provides durable remissions for B-cell lymphoid malignancies and particularly excellent outcomes for those who achieve CR.

The clinical trial reported here was approved by the institutional review board of the National Institutes of Health.

Acknowledgments: The authors thank all investigators, research nurses, patient care coordinators, and the National Institutes of Health (NIH) Clinical Center personnel, who assisted in the care of these patients. The authors also thank the staff involved with the production of the chimeric antigen receptor T cells. The authors especially thank the patients and their families.

The research reported here was supported by the intramural research program of the NIH Clinical Center, the National Cancer Institute (NCI) Center for Cancer Research, and by a research agreement between the NCI and Kite, a Gilead Company. The Surgery Branch of the NCI holds Cooperative Research and Development Agreements with Iovance and Neogene.

Contribution: J.N.K. and S.A.R. conceived and designed the study; L.M., S.L.G., J.C.Y., S.A.R., and J.N.K. provided patient care and acquired clinical data; D.A.N. performed laboratory analysis for quantification of chimeric antigen receptor–positive cells; T.G., J.N.K., and J.N.B. analyzed and interpreted data and wrote the manuscript; and all authors reviewed and edited the manuscript.

Conflict-of-interest disclosure: S.L.G. reports in-kind support (interleukin-2) from Iovance. J.N.K. reports royalties from Kite, a Gilead Company, Celgene/Bristol Myers Squibb, and Kyverna Therapeutics, Inc; and research funding from Kite, a Gilead Company and Celgene/Bristol Myers Squibb. The remaining authors declare no competing financial interests.

J.N.B. reports current employment with AstraZeneca PLC.

Correspondence: Jennifer N. Brudno, AstraZeneca, 1 Medimmune Way, Gaithersburg, MD 20878; email: jbrudno@gmail.com.