Characterization and Prognostic Impact of Multiple Productive IGHV Rearrangements in CLL

Background:

In chronic lymphocytic leukemia (CLL) one of the strongest prognostic factors is IGHV mutational status. Infrequently, patients present not only with a single IGHV rearrangement but with multiple productive rearrangements. In about 2% of all CLL patients analyzed on cDNA level multiple rearrangements display the same mutational status and are categorized accordingly following ERIC recommendations. In another 1% rearrangements with discordant IGHV mutational status are detected and preclude a definite risk assignment. Only limited data exist on these rare subgroups.

Aim:

To characterize treatment-naive CLL patients with multiple productive IGHV rearrangements and determine the impact on prognosis.

Patients and Methods:

Out of 8,016 treatment-naive CLL patients between 2005 and 2015 and with data on IGHV mutational status we identified 204 (3%) with multiple productive rearrangements. IGHV mutational status was analyzed on cDNA and in all cases according to ERIC recommendations. IGHV mutated status (M) was defined by sequence identity <98% and unmutated status (U) by ≥98%. Chromosome banding analysis was available in 102 cases and interphase FISH with probes for 17p13, 13q14, 11q22 and centromeric region of chromosome 12 in 191. Male:female ratio was 3:1 and median age 68 years (range: 38-89). Additionally, data on SF3B1 and TP53 mutations was present in all cases. Follow-up data on time to first treatment (TTT) and overall survival (OS) was available in 105 cases with a median follow-up of 4 years. For statistical comparison we used a cohort of 1,262 untreated CLL patients with single IGHV rearrangement (median age: 67 years; range: 30-91, median follow-up: 6 years).

Results:

Out of 204 patients with multiple, productive rearrangements 199 (98%) presented with two and 5 patients (2%) with three IGHV rearrangements. Concordant IGHV mutated status (MM) was present in 120 cases (59%), whereas concordant unmutated status (UU) was seen in 34 patients (17%). In 50 cases (25%) a mixed IGHV status (UM) was detected.

We analyzed frequencies of complex karyotype by CBA, biclonality according to immunophenotype (concurrent kappa restricted and lambda restricted subpopulations) and/or CBA, TP53 disruption (TP53mut and/or del(17p)), SF3B1mut, del(11q), trisomy 12, and del(13q). Overall, a higher frequency of biclonality was detected in patients with multiple vs. single IGHV rearrangements (16% vs. 1%, p<0.001). However, association to neither MM, UU nor UM existed. MM presented with molecular and cytogenetic characteristics similar to M. Correspondingly, UU showed similar frequencies of mutations and aberrations to U, except for higher frequency of trisomy 12 in UU vs. U (42% vs. 19%, p=0.003). Interestingly, UM presented with characteristics similar to U and UU. UM was associated with TP53 disruption vs. M (16% vs. 5%, p=0.003) and vs. MM (5%, p=0.035) as well as with SF3B1mut vs. M (16% vs. 5%, p=0.008). Furthermore, UM cases showed high frequency of del(11q) vs. M (29% vs. 3%, p<0.001) and vs. MM (1%, p<0.001) and less frequently del(13q) sole vs. M (41% vs. 60%, p=0.011) and MM (41% vs. 69%, p=0.001).

No significantly differences in TTT were observed between MM and M (median: 13 vs. 14 years) and between UU and U (6 vs. 4 years), respectively. However, the difference between MM vs. UU (p=0.022) and M vs. U (p<0.001) was significant. The UM subgroup presented with a TTT (median: 4 years) similar to U and UU, whereas it was significantly shorter vs. M (p=0.003) and MM (p=0.006), respectively. A similar picture emerged for survival. 5-year OS of MM was not different vs. M (94% vs. 90%) but vs. U (78%, p=0.001). The statistical analysis of OS in UU was hampered by low case numbers. UM presented again with similar 5-year OS vs. U (81% vs. 78%, n.s.) and significantly worse OS vs. M (90%, p=0.049) and vs. MM (94%, p=0.014).

Conclusions: (1) Patients with multiple productive IGHV rearrangements and concordant IGHV status show similar prognosis and characteristics to patients with single rearrangement with the respective IGHV status. (2) Cases with mixed IGHV status show similar prognosis to patients with IGHV unmutated status and accordingly are characterized by high frequencies of adverse prognostic factors like TP53 disruption, SF3B1mut, and del(11q), whereas del(13q) sole is less frequent.