Mutations and translocations associated with venetoclax resistance in chronic lymphocytic leukemia

Abstract

The mechanisms underlying acquired resistance to venetoclax (VEN) remain elusive. The BCL-2 family of proteins encompasses a complex network of anti-apoptotic and pro-apoptotic regulators that together govern cell fate decisions. While venetoclax selectively targets BCL2, resistance mechanisms often involve compensatory changes in other family members. For instance, MCL1 amplification and increased expression of BCL-XL have been observed in both preclinical and clinical settings, contributing to treatment failure. Recent data from the CLL13 trial suggested that a highly complex karyotype (CK) with five or more abnormalities (CK ≥ 5) at baseline independently predicted worse progression-free survival following VEN-based treatment. Using whole exome sequencing, our group has found that clonal evolution during VEN treatment selects for 8p deletion +/− gain of MCL1 in some resistant chronic lymphocytic leukemia (CLL) cases [1]. BCL2 G101V mutation is also associated with resistance but can appear before clinical progression at very low variant allele frequencies [2, 3]. The optimal threshold for significance and the clinical utility of measuring G101V remains to be established, and the extent to which BCL2 mutations impact CLL progression during VEN treatment is not yet fully understood. In this study, we evaluated clinical next-generation sequencing (NGS) data, with BCL2 G101V mutation measured by droplet digital PCR (ddPCR), during continuous VEN treatment in CLL. In addition to the G101V mutation, we assessed a broader spectrum of BCL2 mutations using targeted NGS (see Supplementary Methods for Amplicon-based Targeted NGS). Ethical approval was obtained from the Dana-Farber Cancer Institute Institutional Review Board, and all patients provided written informed consent prior to sample and data collection.

We retrospectively reviewed 2850 treatment periods in our database and extracted 234 VEN treatment periods for 207 patients (Supplementary Fig. S1). We ultimately collected 33 patients with available clinical NGS data during continuous VEN treatment; 19 patients with clinical NGS data at disease progression, designated as Group PD, and 14 patients with clinical NGS data during treatment without progression, designated as Group NP. All had banked DNA samples for ddPCR. We retrospectively evaluated a total of 69 sequential clinical NGS results, as well as 85 ddPCR results for BCL2 G101V mutation, from the 33 patients in our cohort. At baseline, before the initial period of VEN treatment, both deletion 17p (del(17p)) and TP53 mutations were found to be enriched in the PD group (P = 0.02 and P = 0.046, respectively) (Supplementary Table S1). Other fluorescence in situ hybridization (FISH) abnormalities, such as del(11q), del(13q), and trisomy 12, as well as gender, age at diagnosis, and IGHV mutation status were not different between the two groups before the initial period of VEN treatment. The number of treatments before VEN was also not different between the groups (no prior treatment, one, two or more, PD vs. NP: 0.0%, 21.1%, 78.9% vs. 7.1%, 28.6%, 64.3%). One patient in the PD group who received 4th line treatment prior to VEN showed primary refractoriness to the drug. The proportion of CK ≥ 5 at baseline was higher, albeit not significantly so, in the PD group (PD vs. NP, 35.7% vs. 9.1%, p = 0.18). The median observation period, measured in days from VEN initiation to treatment response evaluation, including the clinical NGS test, was 537 days in the PD group and 589.5 days in the NP group, respectively (not different).