Somatic mutations in cancer-related genes in chronic myeloid leukemia patients receiving asciminib Free

Background: Identification of possible factors of resistance to tyrosine kinase inhibitors (TKIs) in chronic myeloid leukemia (CML) patients (pts) is a pressing problem.Somatic mutations (SM) in some cancer-related genes may be associated with therapy failure in CML pts. The allosteric inhibitor asciminib (ASC) is effective in CML pts with previous treatment failure. Few studies showed a possible correlation between ASC therapy failure and SM (Shanmuganathan N., 2024), and this data are of particular interest.

Aim: To assess the detection rate and spectrum of SM in CML pts with treatment failure and to study their impact on efficacy of ASC therapy.

Methods: Adult (≥18 years) CML pts with the lack of efficacy or intolerance to ≥2 prior TKI lines or T315I-positive pts after ≥1 prior TKI were enrolled in a Managed Access Program to receive ASC therapy (n=68). For the mutational analysis 20 pts having samples at the ASC initiation and being resistant to prior TKI therapy were selected, while pts with no samples and prior intolerance only were excluded. In SM analysis group (n=20) median age was 56 (range 40-81) year, males 45%. Median CML duration before ASC initiation was 5 years (2-22), 11 (55%) pts had ≥4 prior TKIs, 8 (40%) were ponatinib (PON) pretreated. BCR::ABL1≥10% at baseline was in 15 (75%) pts, no molecular response-2 (MR2; BCR::ABL1≥1%) ever achieved before ASC was in 13 (65%) pts. Nineteen pts had CP (chronic phase), 5/19 had CP2 due to additional chromosomal abnormalities (ACAs) at baseline. One pt had accelerated phase at ASC initiation. Two dosing regimen were used: 40 mg BID in BCR::ABL1^T315I-negative (n=11) and 200 mg BID in BCR::ABL1^T315I-positive pts (n=9). Non-T315I mutations were G250E, F359V (40 mg BID) and F317L (200 mg BID) in 3 pts.

We analyzed MR2 and major molecular response (MMR, BCR::ABL1<0.1%) achievement by cumulative incidence function; evaluated factors of MMR achievement in univariate analysis; made a survival analysis (Kaplan-Meier method, intention-to-treat principle) depending on the presence of SM.

The DNA samples at baseline were analyzed in a targeted NGS panel of 18 genes, associated with myeloid neoplasms (ASXL1, DNMT3A, FLT3, IDH1, IDH2, NPM1, RUNX1, SF3B1, SRSF2, TET2, TP53, U2AF2, KIT, WT1, CEBPA, ZRSR2, JAK2, GATA2), and ABL1. We assessed SM dynamics in 15 pts who had samples collected at additional timepoints: n=4 at diagnosis, n=6 during prior TKI therapy, n=5 on ASC therapy for >1 year.

Results: The median follow-up from the start of ASC therapy was 3.5 years (range 0.8–5).

SM in 5 non-ABL1 cancer-related genes were detected in 10/20 pts (50%) with a median of variant allele frequency 28% (range 6-50%). ASXL1 mutations were found in 7 (35%) pts, DNMT3A in 4 (20%), RUNX1, CEBPA, TET2 – in 1 pt each (5%). ABL1 mutations in DNA were detected in 8/20 (40%) pts. In 5/20 (25%) pts ABL1 mutation co-existed with another: the most frequent combination was ABL1+ASXL1 (n=4; 20%) including one pt with ABL1+ASXL1+DNMT3A; 1 pt had ABL1+DNMT3A. SM presence didn't differ by age, sex, ELTS risk group, ACA and BCR::ABL1 mutations presence, phase, number of prior TKIs, CML duration, best level of BCR::ABL1 >1% or <1% on previous treatment and prior PON failure. SM were more common for pts with BCR::ABL1 level >10% vs <10% (p=0.026).

The 3-year overall survival and 3-year progression free survival in pts with and without SM was 67% vs 100% (p=0.06) and 55% vs 100% (p=0.024), respectively. Three-year probability of MR2 and MMR was lower in pts with SM: 20% vs 70% (p=0.018), 0% vs 50% (p=0.004), respectively.

Dynamic analysis showed that SM may persist during TKI therapy (n=7, including n=2 from diagnosis) and may appear during TKI therapy (n=3). Dynamic BCR::ABL1 mutation analysis revealed persistence of T315I, F359V, F317L mutations in 7 pts, disappearing (T315I, E255K, G250E) - in 4 pts and the appearance ofA337T (n=1) and F311T+M244V (n=1) BCR::ABL1 mutations.

Conclusion: SM in cancer-related genes are often detected in resistant CML pts (50%). The most frequently mutated gene was ASXL1 (35%), that could co-exist with BCR::ABL1 mutations (20%). ASC therapy was less effective in pts with SM studied in a small group; therefore, new prospective studies in larger groups are needed to investigate the influence of other factors. SM may persist or appear in dynamic; such clonal evolution may be a precursor or a by-stander of disease progression.