Relapse characterization in diffuse large b cell lymphoma patients undergoing commercial car‐t cell therapy: experience from a single centre

CAR-T cell therapy is approved for the treatment of adults with relapsed or refractory (R/R) diffuse large B-cell lymphoma (DLBCL). However, 40-60% of patients relapse after therapy and characterization of relapse has been poorly reported. To describe relapse characteristics in biopsy samples in patients with DLBCL receiving CAR-T cell therapy in our center. All consecutive patients diagnosed with R/R DLBCL who were infused with commercial CAR-T cells from June 2019 to December 2020 were included. CAR-T cell expansion in PB was monitored by multiparameter flow cytometry (FC) through detection of labeled CD19 in T cells (Human CD19 Protein®). PET-CT scanner evaluation was performed on days +30, +90 and +180. In cases where PET-CT detected progression or indeterminate response, core-needle biopsy of the accessible lesions was performed. Study of biopsy included: histological analysis; FC immune phenotyping including CAR-T cell presence and detection of exhausted CD3+ populations (LAG3, TIM3, PD1). DNA was purified from biopsy using GeneRead DNA FFPE Kit (Qiagen). Mutations of CD19 (exon2-5) were analyzed by Sanger sequencing. 30 patients with R/R lymphoma were treated: 21 with axi-cel 9 with tisa-cel. Median age at infusion was 57 years (r: 22-79); 16 (53%) were female. Median follow up-was 9.5 months (r: 3-18). PFS at 6 months was 42% and OS 72%. CAR-T cell expansion in PB was detected in 29 patients (97%). Complete and overall response rate at day 30 (CR and ORR) were 39% and 75%, respectively (28 evaluable patients), and 41% and 55% at day 100, respectively (27 evaluable patients). At day 180, 10 out of 14 evaluable patients (38.5%) showed CR. A total of 15 patients experienced relapse or progression. Biopsy of the accessible lesions was performed in 10 (66%) (Table 1). Relapsed was confirmed in all cases except in one in whom sample was insufficient. Three patients who were CD19+ by IHQ on the pre-infusion biopsy showed CD19- at relapse. CAR-T cell in PB at relapse was detected in all except one patient that did not expand CAR-T after 2 infusions and 60% of patients presented CAR-T in the biopsy. TME (tumor microenvironment) TCD3+ lymphocytes were >20% in 80% of the cases and no patient had <5%. In tumor sample, FC failed to identify cellularity in 37% of patients; in the rest of samples, percentage of exhausted TCD3 presented an average of 80% (range 64-84). CD19 study revealed mutation in one patient (p.V279L) who presented a CD19+ relapse. In our series of relapsed lymphomas after CAR-T anti-CD19 treatment, CAR-T cell was present in all patients in PB but only 60% in tissue. CD19 loss and mutations in DNA are possible relapse mechanisms. Exhausted TCD3 lymphocytes were abundant in TME, which would support its relevant role preventing a good function of the carts in situ. More complex biopsies analyses are necessary to better understand relapse mechanisms. No conflicts of interests pertinent to the abstract.