Patients with chronic lymphocytic leukemia (CLL) cells that use IGHV3-21 typically have aggressive disease similar to that of patients with CLL cells that use unmutated IGHV. However, the IGHV3-21 genes expressed in CLL often have less than 98% sequence homology with the known germline IGHV3-21 and hence are considered mutated. However, many “mutated” IGHV3-21 in CLL have higher homology with the germline IGHV3-21 (e.g. >97%) than do most other IGHV in CLL that also are considered mutated (mean percent homology = 93 ± 4% (SD), N= 1,260). As such, some argue that 97% be used as the threshold for defining IGHV mutation status. To test whether this definition is valid for IGHV3-21 we examined the mutation status of the rearranged and/or expressed Ig light chain V genes. We examined 63 (2.6%) IgVH3-21 cases from a large cohort (N=2,457) of CLL patients evaluated by the CRC. Forty of 63 (63.5%) IGHV3-21 cases used ? and 23 (36.5%) used ? light chains. While the ?-expressing cases had perfect correlation in mutational status between IgH and IgL, 12 of 40 (30%) ?-expressing cases were discordant in that they had IGHV3-21 with 94–97.6% germline sequence homology and IGLV that were all unmutated. Nine of these 12 (75%) also displayed a strikingly homologous and short IgH third complementary determining region (CDR) motif (DANGMDV) and light chains encoded by IGLV3-21. To study the differentiation history of such cases, we examined for productively-rearranged, and formerly-expressed, IGKV alleles and non-functional IGKV alleles that had undergone rearrangement with the kappa deleting element (KDE). Analysis for IGKV-KDE rearrangements was performed on 7 of the 9 IGHV3-21/IGLV3-21 cases. Two cases had two non-functional IGKV alleles and five (71%) had one non-functional IGKV allele and one productively-rearranged IGKV allele that apparently was aborted via receptor editing. All 5 productively rearranged IGKV alleles had somatic mutations with germline-homology ranging from only 93.6% to 97.6% (median 96.3%). Non-conservative mutations were clustered in the CDRs, arguing that they were selected in an apparent antigen-driven response. All 9 non-functional IGKV alleles had 100% homology with known germline IGKV genes. For comparison, we examined for IGKV-KDE rearrangements in 4 cases of ?-expressing CLL cells that used the unmutated IGHV1-69 gene and IGLV3-09. All such cases each had two non-functional IGKV alleles that had 100% homology with known IGKV rearranged with KDE, indicating that none of these cases experienced IGKV receptor editing, a significantly lower incidence than that noted for cases using IGHV3-21/IGLV3-21 (P<0.01). These results indicate that despite sharing association with more aggressive disease, most CLL cases that use IGHV3-21/IGLV3-21 are distinctive from cases that use unmutated IGHV1-69 in their B cell differentiation history in that they are derived from B cells that formerly expressed ? light chains that had undergone somatic mutation and selection in an antigen-driven immune response prior to undergoing IGKV receptor editing. Despite expressing IGHV with >97% homology to the known germline IGHV3-21 and IGLV with >98% homology to known germline IGLV, these cases have experienced somatic mutation and apparent germinal center maturation prior to leukemogenesis.

Author notes

Disclosure: No relevant conflicts of interest to declare.