Background: Richter Syndrome diffuse large B-cell lymphoma type (DLBCL-RS) is clinically and biologically different from de novo DLBCL, although the phenotypic profile comparison between both entities has scarcely been reported. Finding a minor monoclonal B-cell lymphoma (MBL) chronic lymphocytic leukemia (CLL) type population at de novo DLBCL diagnosis is not common and its biological and prognostic significance is unknown. Our aim is to characterize and compare DLBCL cases with prior or concomitant CLL (DLBCL-RS), de novo DLBCL with concurrent MBL CLL-type (DLBCL-MBL) and de novo DLBCL without MBL, focusing on the DLBCL phenotypic profile.

Methods: Retrospective multicenter study including three groups of patients: 1) DLBCL-RS (N=63), 2) DLBCL-MBL (N=25), and 3) de novo DLBCL without MBL (N=28). Multiparametric flow cytometry (FCM) and/or immunohistochemistry were assessed to characterize the DLBCL phenotypic profile in each group. Marker-selection was made according to WHO, ICC, EuroFlow, ERIC, and ESCCA recommendations. Comparisons among DLBCL phenotype (13 markers studied: CD19, CD20, CD5, CD79b, CD23, CD43, CD200, CD38, CD10, surface light chain, Ki67, P53, and Myc) between groups were made according to chi-square or Fisher tests. The prognostic impact of each marker on overall survival (OS) in the DLBCL-RS and DLBCL-MBL groups was analyzed according to univariate hazard ratio (HR) by Cox regression model. Clonal relationship information was also reported.

Results: Median age at DLBCL-RS diagnosis was 63 years (IQR 56-73) and 62% were males. Four cases were concomitantly diagnosed with CLL and DLBCL-RS; the median time to transformation in the remaining 59/63 was 78 months (IQR 33-127). Clonal relationship between DLBCL and CLL was demonstrated in 28/31 cases with available information. Among DLBCL-MBL cases, median age was 74 (IQR 63-79) and 64% were males. In all 25/25 cases the MBL CLL-type was detected in bone marrow (BM) by FCM while staging de novo DLBCL, with a median infiltration of 0.8% (IQR 0.1-3.3); no patient met criteria for CLL diagnosis. Histological assessment found BM involvement by MBL in 3/25 patients. Clonal relationship between DLBCL and MBL was demonstrated in all 7/7 cases with available information. CD5 was positive in 66% DLBCL-RS and 37% DLBCL-MBL, respectively (P=0.03), while only in 11% DLBCL without MBL (P <0.001 and P=0.04 against DLBCL-RS and DLBCL-MBL, respectively). The CD79b expression was less frequent in DLBCL-RS (36%) and DLBCL-MBL (25%) (P=0.7) when compared with DLBCL without MBL (93%) (P <0.001 and P=0.02). DLBCL-RS was CD43+ in 77% cases, CD200+ in 63% and Myc overexpressed in 46%, equivalent to DLBCL-MBL with 67% CD43+ (P=0.6), 75% CD200+ (P=0.7) and 50% Myc overexpressed (P=0.8); DLBCL without MBL was CD43+ in 30% (P=0.001 and P=0.1), CD200+ in 33% (P=0.07 and P=0.2) and Myc overexpressed in 26% (P=0.1 and P=0.1). Positive CD23 was found in 61% DLBCL-RS, higher than DLBCL-MBL (16%, P=0.004) and DLBCL without MBL (10%, P <0.001). The rate of CD10 expression was lower in DLBCL-RS (12%) than in DLBCL-MBL (42%, P=0.003) and DLBCL without MBL (57%, P <0.001). CD19, CD20, CD38, Ki67 and P53 expression did not significantly differ between the three DLBCL groups. In 25/32 DLBCL-RS cases the expressed light-chain was the same between the CLL and the DLBCL (κ 22/25 and λ 3/25), in 3/32 it was negative in CLL, in 2/32 negative in DLBCL, and 2/32 switched the light-chain (1 from κ to λ and 1 from λ to κ). In all DLBCL-MBL cases with light-chain positivity by DLBCL and MBL (8/13), the restriction was the same by the two populations (κ 6/8 and λ 2/8); in the remaining 5/13, the MBL and the DLBCL were negative for surface light chain expression in 2/5 and 3/5 cases, respectively. One-year OS was 41% in DLBCL-RS and 70% in DLBCL-MBL. In DLBCL-RS, CD20 (HR 0.3 CI95% 0.1-0.8; P=0.02) and CD43 (HR 13.7 CI95% 1.7-110; P=0.014) were associated with OS. In DLBCL-MBL the CD5 positivity was related with worse OS (HR 5 CI95% 1.2-20; P=0.02).

Conclusion: De novo DLCBL with concurrent MBL CLL-like exhibits a phenotypic profile similar to DLBCL-RS in some relevant makers, and different from de novo DLBCL without MBL. This fact, added to the same light chain restriction and clonal relationship between DLBCL and MBL, suggests that DLBCL-MBL may be considered as Richter transformation, although its prognostic impact is still to be determined. Our next step is to validate this data in a prospective cohort.

Disclosures

Romero:Kyowa Kirin: Membership on an entity's Board of Directors or advisory committees; Kite: Membership on an entity's Board of Directors or advisory committees; Incyte: Membership on an entity's Board of Directors or advisory committees; Roche: Membership on an entity's Board of Directors or advisory committees. Piris-Villaespesa:BluePrint: Consultancy, Research Funding, Speakers Bureau; Novartis: Consultancy, Speakers Bureau; GSK: Consultancy, Speakers Bureau. Garcia Vela:Janssen-Cilag, S.A.: Honoraria.

This content is only available as a PDF.
2024
Sign in via your Institution