HLA-DR expression on monocytes and outcome of anti-CD19 CAR T-cell therapy for large B-cell lymphoma.

Despite their unprecedented success in relapsed/refractory (R/R) large B-cell lymphoma (LBCL), anti-CD19 CAR T cells are associated with significant toxicity, and more than half of patients relapse. As monocytes emerged as key players in CAR therapy, we sought to evaluate the evolution of HLA-DR expression on monocytes (mHLA-DR) before and after commercial anti-CD19 CAR T-cell infusion in a large cohort (n = 103) of patients with R/R LBCL and its association with adverse events and treatment response. Cy-Flu-based lymphodepletion (LD) upregulated mHLA-DR in 79% of the cases, whereas in 2l% of cases (15 patients), the mHLA-DR level decreased after LD, and this decrease was associated with poorer outcome. Low mHLA-DR at day minus 7 (D-7) (<13 500 antibodies per cell) before CAR T-cell infusion correlated with older age, poorer performance status, higher tumor burden, and elevated inflammatory markers. With a median follow-up of 7.4 months, patients with low mHLA-DR D-7 exhibited a poorer duration of response and survival than the higher mHLA-DR D-7 group. For toxicity management, tocilizumab was more frequently used in the low-mHLA-DR D-7 group. These data suggest that monocyte dysregulation before LD, characterized by the downregulation of mHLA-DR, correlates with an inflammatory and immunosuppressive tumor environment and is associated with failure of anti-CD19 CAR T cells in patients with R/R LBCL. Modulation of these myeloid cells represents a promising field for improving CAR therapy.

Correcting the bias of the net benefit estimator due to right-censored observations.

Generalized pairwise comparisons (GPCs) are a statistical method used in randomized clinical trials to simultaneously analyze several prioritized outcomes. This procedure estimates the net benefit (Δ). Δ may be interpreted as the probability for a random patient in the treatment group to have a better overall outcome than a random patient in the control group, minus the probability of the opposite situation. However, the presence of right censoring introduces uninformative pairs that will typically bias the estimate of Δ toward 0. We propose a correction to GPCs that estimates the contribution of each uninformative pair based on the average contribution of the informative pairs. The correction can be applied to the analysis of several prioritized outcomes. We perform a simulation study to evaluate the bias associated with this correction. When only one time-to-event outcome was generated, the corrected estimates were unbiased except in the presence of very heavy censoring. The correction had no effect on the power or type-1 error of the tests based on the Δ. Finally, we illustrate the impact of the correction using data from two randomized trials. The illustrative datasets showed that the correction had limited impact when the proportion of censored observations was around 20% and was most useful when this proportion was close to 70%. Overall, we propose an estimator for the net benefit that is minimally affected by censoring under the assumption that uninformative pairs are exchangeable with informative pairs.