NOT ALL IGHV3-21 CLL ARE EQUAL: SUBSET #2 DISPLAYS A DISTINCTIVE CLINICOBIOLOGICAL PROFILE WITH REMARKABLE SIMILARITIES TO SUBSET #169, ITS CLOSE IMMUNOGENETIC RELATIVE

ABSSUB-3789

Background: In CLL, subsets with stereotyped B-cell receptor immunoglobulins (BcR IG) represent a remarkable one-third of all cases. Intriguingly, CLL subset patients expressing certain stereotyped BcR IG have high intra-subset homogeneity regarding biological and clinical features as well as outcome. CLL subset #2, the largest subset overall, carries IGHV3-21/IGLV3-21 IG that virtually always bear somatic hypermutations (SHM), ranging from minimal or borderline (most frequently) to extensive (rarely).

Aims: Among 8593 CLL patients with available IG and clinicobiological information, we identified 254 (3%) cases classified as subset #2 that we characterized in detail.

Methods: Patient characteristics were as follows: 60% males; median age: 64; Binet stages B-C: 53%; SHM status: 61% IGHV-mutated (<98% germline identity, M-CLL), 39% IGHV-unmutated (U-CLL; only 2/254 cases with no SHM); FISH aberrations: isolated del(13q): 56% (47% in M-CLL vs 70% in U-CLL, p=0.02) | trisomy 12: 4.7% | del(11q): 23% (23% in M-CLL vs 26% in U-CLL, p=0.8) | del(17p): 4.7%.

Results: No differences in time-to-first-treatment (TTFT) were found between U-CLL vs M-CLL subset #2 cases (19 vs 23 months, p=0.6). Interestingly, however, among the M-CLL, the presence of del(11q) was associated with significantly shorter TTFT (13 vs 29 months, p=0.03); no such differences were seen among U-CLL subset #2 cases, regarding the incidence of del(11q). We next compared subset #2 cases to 183 non-subset #2 IGHV3-21 (non#2/3-21) CLL cases and noted a significantly lower (p=0.002) SHM load with 53% of the non#2/3-21 cases (97/183) being U-CLL. Non#2/3-21 CLL also had significantly longer TTFT compared to subset #2 CLL (60 vs 19 months, p=0.001). This difference arose mainly from the superior outcome of non#2/3-21 cases with mutated IGs (median TTFT 153 months) since the TTFT of non#2/3-21 U-CLL, although longer, did not differ significantly from subset #2 (30 vs 19 months, p=0.4). No other major differences were found (genomic aberrations included). Within our cohort we also identified a minor group of 19 cases (0.2% of the series), termed subset #169, that express the IGHV3-48/IGLV3-21 gene combination, carry VH CDR3 identical in length and similar in composition to those of subset #2 IG and display an analogous SHM profile. Given the overall high identity (97%) between the IGHV3-21 and IGHV3-48 genes, we explored whether subsets #2 and #169, whose immunogenetic signatures are clearly related, share similar clinicobiological features and outcomes. Indeed, this turned out to be the case, since their FISH genomic profiles were similar, especially with regards to del(11q) (38% in subset #169) and del(17p) (absent in subset #169); as was their median TTFT (p=1). Of note, in an ongoing study from our group, subsets #2 and #169 display an almost identical (~45%) frequency of SF3B1 mutations.

Summary/Conclusion: Altogether, we demonstrate that IGHV3-21 CLL should not be considered as homogeneous, with subset #2 emerging as uniformly aggressive, thus contasting non#2/3-21 patients whose prognosis clearly depends on SHM status like the rest CLL. We also argue that studying BcR stereotypy is relevant for improved understanding and eventual management of CLL through the identification of subsets with distinct features that can be expected to benefit from the implementation of specific treatments. Finally, we highlight for the first time the biological and clinical links between subsets with related immune signatures, raising the intriguing possibility that a higher-order organization of subsets based on their immunogenetic features will also be highly relevant.

Keywords: Chronic lymphocytic leukemia, IGH

http://www.ehaweb.org/