B CELL RECEPTOR STEREOTYPY MAKES A CLINICAL DIFFERENCE IN CLL: REVELATIONS FROM A MULTI-INSTITUTIONAL SERIES OF 4615 CASES

Background:

In CLL, ~30% of all cases, carry quasi-identical, stereotyped VH CDR3 sequences within their B cell receptors (BcR) and can be clustered to different subsets defined by distinct VH CDR3 sequence motifs. Increasing evidence suggests that patients belonging to the same subset may share similar clinico-biological features and, perhaps, outcome. Nevertheless, even the largest VH CDR3-defined subsets account for only up to 3% of the cases with available IGHV-D-J sequence information, clearly indicating that for meaningful conclusions to be reached, reliable subset assignment and large patient cohorts are imperative.

Aims:

We investigated the clinical implications of BcR stereotypy identified in a series of 4615 patients with CLL, consolidated in the context of a multi-institutional collaboration.

Methods:

Subset identification was based on our previously reported approach (Blood 2012;119:4467), and associations were sought for major subsets which collectively accounted for 727 cases, namely: (i) subsets with unmutated IGHV genes (U-CLL): #1, n=129; #3, n=47; #5, n=24; #6, n=47; #7, n=78; #8, n=33; #31, n=19; #59, n=19; (ii) subsets with mutated IGHV genes (M-CLL): #4, n=61; #16, n=19; #77, n=23; #148, n=49; #201, n=21; and, (iii) subset #2 (IGHV3-21, variable mutational status), n=158.

Results:

Regarding age at diagnosis, (i) M-CLL subsets #4, #16 and #201 (all utilizing the IGHV4-34 gene), #77, and #148 as well as U-CLL subsets #3, #5 and #7 (all utilizing the IGHV1-69 gene) and #31 contained significantly younger patients (P<0.001) compared to subsets #1, #2, #6, #8 and #59. While, these findings confirm and significantly extend previous reports by us and others about subset #4 they provide novel information about all other subsets. Significantly (p<0.001) different gender distributions were identified, with male:female ratios ranging from greater than 4.0 for U-CLL subsets #3 and #5 (both IGHV1- 69) and #31 (IGHV3-48) to less than 0.8 for U-CLL subset #8 (IGHV4-39) and M-CLL subset #201 (IGHV4-34). Among the most populated subsets, strikingly different (P<0.01) profiles of genomic aberrations (hierarchical model) were identified, exemplified by: (i) diminished frequency of del(17p) in subset #2 versus all others, especially subsets #1 and #8; (ii) significantly higher frequency of del(11q) in subsets #3, #5 and #7 (all IGHV1-69, range 37% to 44.4%) versus all others; (iii) significantly higher frequency of trisomy 12 in subset #8 (69.2%) versus all others; (iv) homogeneous profile of subset #4 that is characterized by high frequency of del(13q) and very low frequency of other aberrations extending to complete absence of del(17p). We assessed time-to-firsttreatment (TTFT) and noted significantly shorter TTFT in subset #2 vs. M-CLL subsets vs. U-CLL subsets and also vs. U-CLL as a whole (p<0.01), thus corroborating previous observations about the adverse prognosis of subset #2. Among U-CLL subsets in general, significantly worse outcomes were found for subsets #31 and #59 (P=0.06 and P=0.005, respectively). Focusing on U-CLL subsets utilizing the IGHV1-69 gene (subsets #3, 5, 6, 7, 59), significantly shorter TTFT was identified for subset #59 (P=0.03). M-CLL subsets in general had significantly longer TTFT compared to U-CLL subsets (p<0.05). Among M-CLL subsets, a trend for longer TTFT was seen for subset #148 (P=0.06).

Summary / Conclusion:

In conclusion, in the largest series analyzed thus far, we provide results strongly supporting the concept that the molecular classification of CLL into stereotyped subsets is biologically and clinically relevant.

Abstrakta v časopise Haematologica 2013, Suppl1

Online Program