Background:MYC and BCL6 rearrangements (r) in high-grade B-cell lymphoma (HGBL) are generally associated with aggressive disease and poor outcomes. However, there are conflicting reports regarding the prognostic significance. Patients (pts) with a reciprocal t(3;8) BCL6::MYCr (t(3;8)+) were not found to have adverse outcome. In contrast, pts with MYC and BCL6r involving other translocation partners (TP; t(3;8)-) seem to have an inferior prognosis, though the underlying mechanism for this discrepancy remains unclear.

Aim: Analysis aimed at uncovering the molecular differences that could explain the prognostic discrepancies between t(3;8)- and t(3;8)+ pts as well as analysis of TP of t(3;8)-.

Patients and methods: All samples sent for diagnosis between 08/2005 – 04/2025 were used for detection of concurrent MYCr and BCL6r, and n=63 pts were identified. Whole-genome sequencing (WGS) was performed for 41 patients (median coverage 100x), whole transcriptome sequencing (WTS) for 39 cases. Clinical data was available for 44 pts. SPSS (version 19.0.0) software was used for statistical and for overall survival (OS) analysis (Kaplan-Meier).

Results: 12/63 (19%) pts with MYCr and BCL6r were t(3;8)+, 51/63 (81%) were t(3;8)-. An additional BCL2r was detected in all t(3;8)+ pts (“triple-hit”), but only in 31/51 (61%) t(3;8)- pts. Of note, all BCL2r were caused by t(14;18)(q32;q21)/IGH::BCL2 translocation. In t(3;8)- pts, the most frequent TP of MYC was IGH (30/51), followed by IGL (9/51) and IGK (4/51). Thus in 81% an IG locus was identified as TP of MYC. BCL6 was also frequently rearranged with IGH (18/51), but IGKr and IGLr were rare (all IG loci: 45%), whereas CIITA (16p13) and RHOH (4p14) were identified as TP in 8/51 and 4/51 of cases, respectively. Pts with t(3;8)+ were younger and showed a higher frequency of females compared to t(3;8)- pts (median 64 vs. 69 years; 58% vs. 39% females). Median OS of the total cohort was 11 months; however OS was better for t(3;8)+ compared to t(3;8)- pts, corroborating previous studies (n.r. vs. 10 months). Further, when the t(3;8)- cohort was split into cases with IG as TP (for MYC and/or BCL6, n=44) and cases with other TP (n=7), the negative impact on OS was only found in cases with IG partners (median n.r. vs. 6 months). Hence, for further analysis, the t(3;8)+ and other TP cases were combined (t(3;8)+/other, n=19) and compared to t(3;8)- cases with IG as TP (t(3;8)-/IG, n=44): OS differed significantly between these subgroups (n.r. vs. 6 months, p=0.019). The t(3;8)+/other pts showed a significant association to BCL2r when compared to t(3;8)-/IG (19/19 = 100% vs. 25/44 = 57%, p<0.001). However, the presence of BCL2 did not influence OS, as t(3;8)-/IG pts with or without BCL2r showed equally poor OS. WTS analysis of the subgroups did not show differences in expression of BCL2, BCL6, MYC or other differentially expressed genes. In the total cohort, molecular mutations (mut) were frequently found in CREBBP (n=17, 27%), KMT2D (n=18, 29%) and MYC (n=15, 24%), whereas TP53 mut were infrequent (n=7, 11%). MYC mut were almost exclusively found in t(3;8)-/IG pts (14/44 vs. 1/19, 32% vs. 5%, p=0.026), whereas CREBBP (16% vs. 53%, p=0.005) and KMT2D (21% vs. 47%, p=0.038) mut were significantly associated to t(3;8)+/other. No influence on OS was found for the presence of CREBBP, KMT2D or TP53 mut. However, MYC mut showed a negative impact on OS in the total cohort (17 vs. 2 months; p=0.015) and even within the t(3;8)-/IG cohort (10 vs. 1 month; p=0.043). In multivariate COX regression analyses, both MYC mut and t(3;8)-/IG showed a negative impact on OS (MYC mut: HR=2.1, p=0.043; t(3;8)-/IG: HR=3.2, p=0.032).

Conclusions: HGBL with BCL6r/MYCr can be separated into two groups dependent on the TP: t(3;8)+/other is characterized by (i) better survival, (ii) association with BCL2r and (iii) association with CREBBP and KMT2D mut. By contrast, t(3;8)-/IG shows (i) inferior survival and (ii) association with MYC mut. Particularly, the combination of IG loci as TP and MYC mut leads to a dismal outcome of ~1 month OS, whereas the presence of concurrent BCL2r (“triple-hit”) does not influence OS. Thus, the TP and the presence of MYC mut determines OS of patients with MYCr and BCL6r. Therefore, CBA and/or WGS is required for correct classification. Due to the impact on OS and the specific molecular characteristics, t(3;8)-/IG cases might be considered as separate entity.

This content is only available as a PDF.
2025
Sign in via your Institution