Outcomes of Relapsed Refractory Double Hit Lymphoma: A Multicenter Observational Cohort

Introduction: In pre-chimeric antigen receptor T cell therapy (pre-CART) era, retrospective studies of relapsed/refractory double hit lymphoma (R/R DHL), with MYC and BCL2 and/or BCL6 rearrangement, show inferior survival compared to non-DHL DLBCL. Large study of relapse pattern, chemorefractoriness and outcomes in CART era is lacking. Here we examine treatment patterns, response and outcomes of R/R DHL in a large observational cohort including patients (pts) treated in CART era.

Methods: In this retrospective multicenter observational study, adult pts (pts) with DHL R/R after anthracycline-based therapy, were enrolled. Demographic and clinical variables were obtained by retrospective electronic health records review. Study objectives were overall response rates (ORR), complete response rates (CR), progression free survival (PFS) and overall survival (OS). Time to event endpoints were calculated from date of treatment.

Results: From 1st January 2011-30^th December 2020, 223 adult pts of 12 participating institutions, met eligibility criteria (Table 1). 104 (47%) were classified as high-grade B cell lymphoma (HGBCL), 87 (39) and were classified as diffuse large B cell lymphoma (DLBCL) and 32 (14%) were HGBCL transformed from follicular lymphoma (tFL). 125 (56%) were refractory to frontline therapy (PRD) and 51 (23%) had relapsed within 1 year of frontline therapy (ER).

178 (78%) pts were treated with 2^nd line curative-intent salvage regimen. ORR and CR were 79 (46%) and 40 (23%), respectively. Pts with PRD had significantly lower ORR and CR of 33% (N=35, p<0.001) and 14% (N=15, p<.001) respectively. 42 (24%) pts underwent autologous stem cell transplant (ASCT). 39 did not undergo ASCT for disease progression (24), treatment related death (2), briding to CART in partial response (PR, 3), unknown reason (10). In univariable analyses, age (OR: 1.03 CI₉₅: 1.01-1.07, p=0.002) and PRD (OR: 0.3 CI₉₅: 0.1-0.8, p<0.001) were associated with lower odds of ORR. Presence of PRD (OR:0.2 CI₉₅: 0.05-0.5, p<0.001) and ER (OR:0.3 CI₉₅: 0.08-0.8, p<0.001) was associated with significantly lower odds of CR. Type of Salvage therapy was not associated with response.

2 yr PFS and OS of all 223 pts were 22% (CI₉₅: 8.2-35.8) and 36.3% (CI95: 26.3-46.3) respectively. In univariable analyses, high LDH (HR: 2.0 CI₉₅: 1.3-3.0,p<0.01), presence of BCL2 rearrangement (HR: 2.3 CI₉₅: 1.3-4.2,p=0.007), EN disease (HR: 1.5 CI₉₅: 1.0-2.2,p=0.04), and receipt of noncellular therapy (nonCT) (HR: 5.1 CI₉₅: 3.8-10.6,p<0.001) were associated with inferior PFS. High LDH (HR: 1.7 CI₉₅: 1.1-2.1,p=0.02), receipt of nonCT (HR: 5.1 CI₉₅: 2.5-10.1, p<0.001) were associated with inferior OS.

Pts who underwent ASCT had 2 yr PFS and OS of 46% (CI₉₅: 28.8-63.2) and 57.8% (CI₉₅: 43.8-71.8) respectively. Pre-ASCT response was not associated with PFS or OS.

73 (32%) received CART (14 in 2^nd line, 35 in 3^rd line and 24 in 4^th line); 72 had progressed on curative-intent chemoimmunotherapy and 69 were refractory. 42 (57%) pts had CR, 11 (15%) had PR and 15 (21%) had progressive disease (PD) to CART. 2 yr PFS and OS were 29.6% (CI₉₅: 7.8-51.4) and 52.4% (CI₉₅: 38.4-66.4) respectively. Post-CART CR had significantly higher PFS compared to PR (HR: 2.5 CI₉₅: 1.1-5.6, p=0.03) and PD (HR: 6.4, CI₉₅:3.2-13.1, P<0.001). Post-CART PD (HR: 2.66, CI₉₅:1.11-6.4, P<0.001) was significantly associated with adverse OS but not PR (HR: 1.7, CI₉₅:0.6-4.6, P=0.2).

108 (68%) did not receive CART or ASCT in 2^nd or later line (nonCT); 87 had curative intent therapy and 21 had palliative-intent therapy. ORR was 33% (N=30), CR was 17% (N=19), 2 yr PFS and OS of 7.3% (CI₉₅: 0-35.8%) and 18% (CI₉₅: 0-39) respectively.

In pts progressive on 2nd line curative-intent salvage therapy, when adjusted for adverse features, CART (N=59) had significantly higher ORR (OR: 5.6 CI₉₅: 2.1-16.1, p<0.001), CR (OR: 5.3 CI₉₅: 2-16, p<0.001), PFS (HR 5.6 CI₉₅: 2.8-10.9, p<0.0001) and OS (HR: 5.1 CI₉₅: 2.3-11, p<0.001) compared to nonCT (N=87).

Conclusions: R/R DHL have low and short-lasting response to curative intent salvage therapy. PRD is less likely to respond to further chemotherapy. Minority of pts with chemosensitive disease who can be bridged to ASCT have durable outcomes. Outcomes of CART in chemoresistant pts are significantly better than nonCT and comparable to chemosensitive pts undergoing ASCT. CART can overcome chemorefractoriness of DHL.