CD19 chimeric antigen receptor (CAR) T cells can induce prolonged remissions and potentially cure a significant proportion of patients with relapsed/refractory large B-cell lymphomas. However, some patients may die of causes unrelated to lymphoma after CAR T-cell therapy. To date, little is known about the nonrelapse mortality (NRM) after CAR T-cell therapy. Using the French DESCAR-T registry, we analyzed the incidence and causes of NRM and identified risk factors of NRM. We report on 957 patients who received standard-of-care axicabtagene ciloleucel (n = 598) or tisagenlecleucel (n = 359) between July 2018 and April 2022, in 27 French centers. With a median follow-up of 12.4 months, overall NRM occurred in 48 patients (5.0% of all patients): early (before day 28 after infusion) in 9 patients (0.9% of all patients and 19% of overall NRM), and late (on/after day 28 after infusion) in 39 patients (4.1% of all patients and 81% of overall NRM). Causes of overall NRM were distributed as follows: 56% infections (29% with non–COVID-19 and 27% with COVID-19), 10% cytokine release syndromes, 6% stroke, 6% cerebral hemorrhage, 6% second malignancies, 4% immune effector cell associated neurotoxicities, and 10% deaths from other causes. We report risk factors of early NRM and overall NRM. In multivariate analysis, both diabetes and elevated ferritin level at lymphodepletion were associated with an increased risk of overall NRM. Our results may help physicians in patient selection and management in order to reduce the NRM after CAR T-cell therapy.

In the POLARIX study (ClinicalTrials.gov identifier: NCT03274492 ), polatuzumab vedotin plus rituximab, cyclophosphamide, doxorubicin, and prednisone (Pola-R-CHP) showed a significant progression-free survival (PFS) benefit versus rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) in patients with previously untreated intermediate- or high-risk diffuse large B-cell lymphoma (DLBCL; median follow-up: 28 months). In this 5-year update, sustained PFS benefits favoring Pola-R-CHP were observed. In the global intention-to-treat population (N = 879; median follow-up: 64.1 months), Pola-R-CHP demonstrated a significant PFS benefit over R-CHOP (hazard ratio [HR], 0.77 [95% CI, 0.62 to 0.97]), with 5-year PFS rates of 64.9% (95% CI, 59.8 to 70.0) and 59.1% (95% CI, 53.9 to 64.3), respectively. Although not statistically significant, overall survival analysis showed a HR of 0.85 (95% CI, 0.63 to 1.15) at the 5-year data cut compared with 0.94 (95% CI, 0.67 to 1.33) at the 2-year data cut. In the expanded population, 46 and 62 patients had lymphoma-related deaths in the Pola-R-CHP and R-CHOP arms, respectively. Exploratory analyses showed favorable 5-year survival rates with Pola-R-CHP in high-risk subgroups, including activated B-cell DLBCL and International Prognostic Index score 3-5. Long-term tolerability was similar between treatment arms. Findings confirm Pola-R-CHP represents a standard of care for frontline treatment of DLBCL.

Bcl-2 phosphorylation at serine-70 (S70pBcl2) confers resistance against drug-induced apoptosis. Nevertheless, its specific mechanism in driving drug-resistance remains unclear. We present evidence that S70pBcl2 promotes cancer cell survival by acting as a redox sensor and modulator to prevent oxidative stress-induced DNA damage and execution. Increased S70pBcl2 levels are inversely correlated with DNA damage in chronic lymphocytic leukemia (CLL) and lymphoma patient-derived primary cells as well as in reactive oxygen species (ROS)- or chemotherapeutic drug-treated cell lines. Bioinformatic analyses suggest that S70pBcl2 is associated with lower median overall survival in lymphoma patients. Empirically, sustained expression of the redox-sensitive S70pBcl2 prevents oxidative stress-induced DNA damage and cell death by suppressing mitochondrial ROS production. Using cell lines and lymphoma primary cells, we further demonstrate that S70pBcl2 reduces the interaction of Bcl-2 with the mitochondrial complex-IV subunit-5A, thereby reducing mitochondrial complex-IV activity, respiration and ROS production. Notably, targeting S70pBcl2 with the phosphatase activator, FTY720, is accompanied by an enhanced drug-induced DNA damage and cell death in CLL primary cells. Collectively, we provide a novel facet of the anti-apoptotic Bcl-2 by demonstrating that its phosphorylation at serine-70 functions as a redox sensor to prevent drug-induced oxidative stress-mediated DNA damage and execution with potential therapeutic implications.

CD19-directed CAR T-cells have been remarkably successful in treating patients with relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL) and transformed follicular lymphoma (t-FL). In this cohort study, we treated 60 patients with axicabtagene ciloleucel or tisagenlecleucel. Complete and partial metabolic responses (CMR/PMR) were obtained in 40% and 23% of patients, respectively. After 6.9 months of median follow-up, median progression-free survival (mPFS) and overall survival (mOS) were estimated at 3.1 and 12.3 months, respectively. Statistical analyses revealed that CMR, PFS, and OS were all significantly associated with age-adjusted international prognostic index (aaIPI, p < 0.05). T-cell subset phenotypes in the apheresis product tended to correlate with PFS. Within the final product, increased percentages of both CD4 and CD8 CAR⁺ effector memory cells (p = 0.02 and 0.01) were significantly associated with CMR. Furthermore, higher CMR/PMR rates were observed in patients with a higher maximal in vivo expansion of CAR T-cells (p = 0.05) and lower expression of the LAG3 and Tim3 markers of exhaustion phenotype (p = 0.01 and p = 0.04). Thus, we find that aaIPI at the time of infusion, phenotype of the CAR T product, in vivo CAR T-cell expansion, and low levels of LAG3/Tim3 are associated with the efficacy of CAR T-cell therapy in DLBCL patients.

Conventional therapies for patients with T-cell prolymphocytic leukemia (T-PLL) such as cytotoxic chemotherapy and alemtuzumab have limited efficacy and considerable toxicity. Several novel agent classes have demonstrated preclinical activity in T-PLL, including inhibitors of the JAK/STAT and TCR pathways, as well as histone deacetylase (HDAC) inhibitors. Recently, the BCL-2 inhibitor venetoclax also showed some clinical activity in T-PLL. We sought to characterize functional apoptotic dependencies in T-PLL to identify novel combination therapy in this disease. Twenty-four primary T-PLL patient samples were studied using BH3 profiling, a functional assay to assess the propensity of a cell to undergo apoptosis ('priming') and the relative dependence of a cell on different anti-apoptotic proteins. Primary T-PLL cells had a relatively low level of priming for apoptosis, and predominantly depended on BCL-2 and MCL-1 for survival. Selective pharmacologic inhibition of BCL-2 or MCL-1 induced cell death in primary T-PLL cells. Targeting JAK/STAT pathway with the JAK1/2 inhibitor ruxolitinib or HDAC with belinostat both independently increased dependence on BCL-2 but not MCL-1, thereby sensitizing T-PLL cells to venetoclax. Based on these results, we treated two patients with refractory T-PLL with the combination of venetoclax and ruxolitinib. We observed a deep response in the JAK3-mutated T-PLL and a stabilization of the unmutated disease. Our functional, precision medicine-based approach identified inhibitors of HDAC and the JAK/STAT pathway as promising combination partners for venetoclax, warranting further exploration of such combinations clinically in T-PLL.

We present the results of a phase 2 study evaluating the combination of obinutuzumab + idelalisib in relapsed/refractory (R/R) Waldenström macroglobulinemia (WM). The goal was to determine the safety and efficacy of a fixed-duration chemotherapy-free treatment. During the induction phase, patients received idelalisib + obinutuzumab for 6 cycles, followed by a maintenance phase with idelalisib alone for ≤2 years. Forty-eight patients with R/R WM were treated with the induction combination, and 27 patients participated in the maintenance phase. The best responses, reached after a median of 6.5 months (interquartile range, 3.4-7.1; range, 2.6-22.1 months), were very good partial response in 5 patients, partial response in 27 patients, and minor response in 3 patients, leading to overall response rate and major response rate estimates of 71.4% (95% confidence interval [CI], 56.7-83.4) and 65.3% (95% CI, 50.4-78.3), respectively. With a median follow-up of 25.9 months, median progression-free survival was 25.4 months (95% CI, 15.7-29.0). Univariate analysis focusing on molecular screening found no significant impact of CXCR4 genotypes on responses and survivals but a deleterious impact of TP53 mutations on survival. Although there was no grade 5 toxicity, 26 patients were removed from the study because of side effects; the most frequent were neutropenia (9.4%), diarrhea (8.6%), and liver toxicity (9.3%). The combination of idelalisib + obinutuzumab is effective in R/R WM. Nonetheless, the apparent lack of impact of genotype on outcome could give new meaning to targeting of the phosphatidylinositol 3-kinase pathway in WM. This trial was registered at www.clinicaltrials.gov as #NCT02962401.