Abstract
Background:
CD19-directed CAR-T cell therapy has shown promising efficacy in relapsed/refractory (R/R) B-cell malignancies in clinical trials resulting in the approval and commercialization of two products (tisagenlecleucel/Tisa-cel and axicabtagene ciloleucel/Axi-cel) for R/R diffuse large B cell lymphoma (DLBCL) and primary mediastinal large B cell lymphoma (PMBCL). However, relapses occur in 60-65% of patients (pts) and thus a better understanding of the early determinants of response is critical to improve long-term survival in the real-world scenario.
Aims of the study: To assess whether CAR-T cell expansion after infusion represents a crucial determinant to sustain effective anti-tumor responses to both Tisa-cel and Axi-celTo evaluate differences in CAR-T cell kinetics due to the use of CD28 or 4-1BB costimulatory moleculesTo identify immune phenotypic features of infusion products accounting for CAR-T cell expansion and survival probability
Methods:
We analyzed samples from 43 pts [29 DLBCL, 8 high grade B-cell lymphoma (HGBCL) and 6 PMBCL] treated with Axi-cel (n=22) and Tisa-cel (n=21) at the Fondazione IRCCS Istituto Nazionale Tumori prospectively collected between November 2019 and April 2021. CAR-T cells were monitored in the peripheral blood (PB) on days 0, 4, 7, 10, 14, 21, 28 and monthly post infusion by flow cytometry (FCM). Cells were stained with CD19 CAR Detection Reagent (Miltenyi), CD3, CD4, CD8, CD45, CD14, CD45RO, CD62L, CD197, CD279, CD223 and CD366. Residual cells obtained from washings of 32 infused commercial CAR-T bags (10 Tisa-cel and 22 Axi-cel) were also analyzed by FCM. Data were acquired on a BD FACSCanto II (BD Biosciences) and a MACSQuant® Analyzer MQ10 (Miltenyi) and analyzed using FlowJo software, version 10.
Results:
The median time to maximal expansion of CAR-T cells was at day 10 post infusion with no differences between Axi-cel and Tisa-cel [median concentration at day 10 (C 10) 25 for Axi-cel vs 26 CAR-T cells/µl for Tisa-cel; p, ns], nor among the different histologies (median C 10 33 for DLBCL vs 19 for HGBCL vs 18 CAR-T cells/µl for PMBCL; p, ns). On the contrary, CAR-T peak concentration (C max) was higher in responders at 3 months post infusion (RE, n=28) (defined as pts achieving complete or partial response by PET/CT) than in non responders (NR, n=13) (median C max 87 in RE vs 26 in NR CAR-T cells/µl; p<0.01; Fig 1A). Consistently, the magnitude of CAR-T cell expansion in the first 30 days was higher in RE than in NR [median area under the curve (AUC 0-30) 189 vs 50; p<0.005; Fig 1B]. Circulating CAR-T cells were enriched in subpopulations representing naïve T cells (CD8+ T N; CD45RO−/CD62L+) in RE (median 0.4% in RE vs 0.04% in NR, p<0.05) while NR had significantly higher levels of effector memory T cells (CD8+ T EM; CD45RO+/CD62L+) (median 26.5% in RE vs 66.2% in NR, p<0.05). Additionally, the extent of CAR-T cell expansion predicted the progression free survival (PFS), but not the overall survival (OS), irrespective of the product used (Fig 2, p<0.05) and the overall survival was improved by salvage treatment with bispecifc antibodies. Finally, we evaluated whether CAR-T cell expansion was influenced by the immune phenotypic attributes of the infused products. A significant enrichment of central memory populations (CD8+ T CM; CD45RO−/CCR7+/CD62L+) among CAR-T cells within the infusion products of pts with longer PFS was documented, as compared with those with shorter PFS (CD8+ T CM; median 15.2% vs 3.1%; p<0.005).
Conclusion:
To the best of our knowledge, this is the first study assessing the clinical utility of early CAR-T cell monitoring in lymphoma pts receiving both commercial anti-CD19 CAR-T cell therapies. We provide evidence that in pts treated with Axi-cel and Tisa-cel: i) the in vivo kinetics of the CAR-T cell products are similar, consistent with the fact that no differences in the outcome of Axi-cel and Tisa-cel treated pts were detected; ii) CAR-T cell expansion is critical for efficacy and predicts the PFS; iii) circulating CAR-T cells in responders have a naïve phenotype; iv) a memory signature in the CAR-T cell product before infusion is associated with in vivo expansion and survival.
Figure 1 Figure 1.
Disclosures
Chiappella: Celgene Bristol Myers Squibb: Other: lecture fee, advisory board; Incyte: Other: lecture fee; Novartis: Other: lecture fee; Astrazeneca: Other: lecture fee; Servier: Other: lecture fee; Takeda: Other: advisory board; Gilead Sciences: Other: lecture fee, advisory board; Clinigen: Other: lecture fee, advisory board; Roche: Other: lecture fee, advisory board; Janssen: Other: lecture fee, advisory board. Corradini: AbbVie, ADC Theraputics, Amgen, Celgene, Daiichi Sankyo, Gilead/Kite, GSK, Incyte, Janssen, KyowaKirin, Nerviano Medical Science, Novartis, Roche, Sanofi, Takeda: Consultancy; AbbVie, ADC Theraputics, Amgen, Celgene, Daiichi Sankyo, Gilead/Kite, GSK, Incyte, Janssen, KyowaKirin, Nerviano Medical Science, Novartis, Roche, Sanofi, Takeda: Honoraria; KiowaKirin; Incyte; Daiichi Sankyo; Janssen; F. Hoffman-La Roche; Kite; Servier: Consultancy; Novartis; Gilead; Celgene: Consultancy, Other: Travel and accommodations; BMS: Other: Travel and accommodation; Sanofi: Consultancy, Honoraria; Amgen; Takeda; AbbVie: Consultancy, Honoraria, Other: Travel and accommodations; Incyte: Consultancy; Novartis, Janssen, Celgene, BMS, Takeda, Gilead/Kite, Amgen, AbbVie: Other: travel and accomodations.
TCR engagement negatively affects CD8 but not CD4 CAR T cell expansion and leukemic clearance
