SF3B1 mutations are associated with shorter time to first treatment in chronic lymphocytic leukemia

Abstract

Abstract Background: Chronic lymphocytic leukemia (CLL) is a clinically heterogeneous hematologic malignancy. The identification of risk factors has enhanced prognostic stratification. Next-generation sequencing (NGS) has enabled the comprehensive characterization of the genomic landscape of CLL. This study aims to investigate the clinical and prognostic impact of SF3B1 mutations, focusing on their association with time to first treatment (TTFT), overall survival (OS), and correlation with established risk factors. Methods This is a retrospective analysis of the DFCI CLL Database, based on patients (pts) with available NGS data. A total of 88 genes were assayed by NGS using the NEBNext kit (New England Biolabs) at Brigham and Women's Hospital, a Clinical Laboratory Improvement Amendments (CLIA) certified laboratory. For SF3B1, exons 12 to 18 were sequenced.Complex karyotype was defined as ≥3 chromosomal abnormalities.Kaplan–Meier method and log-rank tests were used to assess time-to-event outcomes. Cox regression was used for univariate and multivariate analyses. Results This study analyzed 1,569 pts who underwent prognostic evaluation by NGS before treatment initiation. The SF3B1 mutation was identified in 186 pts (11.8%). Male individuals were more prevalent in the SF3B1-mutated group compared to the wild-type group (68.3% vs. 58%, p=0.007). The median age at diagnosis did not differ significantly between the groups (61 years vs 62 years). Rai staging at diagnosis showed higher rates of intermediate (60.2%) and high-risk (12.4%) in SF3B1-mutated pts compared to intermediate (51.9%) and high-risk (4.8%) rates in the non-mutated (p<0.001). Beta-2-microglobulin (B2M) levels above 3.5 mg/L were observed in 28% of the mutated cohort (p < 0.001). The prevalence of unmutated IGHV was higher in the SF3B1-mutated group (64%) compared to the wild-type group (40.1%) (p< 0.001). Similarly, del(11q) was more frequent in mutated pts (20.7% vs. 9.2%, p< 0.001). No statistically significant differences were found between groups in del(17p) (p=0.809) and complex karyotype (p=0.316). When specifically analyzing the SF3B1 mutation, 6% of all pts with mutated SF3B1 had more than one SF3B1mutation. The most frequently detected hotspot was p.K700, observed in 33% of the mutated cases. Other recurrent hotspots included p.G742 (7%) and p.G740 (7%). Although no direct association was observed between SF3B1 and any other gene included in the NGS, there was a trend toward co-occurrencewith ATM (Odds Ratio (OR 1.8, FDR = 0.067), GNB1 (OR 5.09, FDR = 0.067), and XPO1 (OR 2.8, FDR = 0.067). The median follow-up time for the entire cohort was 50 months (range 44.6–54.3). The median TTFT was significantly shorter in the SF3B1-mutated group (57.2 months, 95% CI: 46.5–82.1) compared to wild-type group (95.4 months, 95% CI: 84.2–101.8). 34% of pts in wild-type group required first line treatment, compared to 59% in SF3B1 cohort. In multivariate analysis, the SF3B1 mutation was identified as an independent risk factor for shorter TTFT (HR 1.39, 1.11–1.73, p=0.004). Other factors also associated with this outcome included unmutated IGHV status (HR 3.29, 2.67–4.05, p<0.001), del(11q) (HR 1.29, 1.01–1.65, p=0.041), complex karyotype (HR 1.39, 1.10–1.76, p=0.006), and del(17p) and/or TP53 mutation (HR 1.37, 1.08–1.74, p=0.010). Considering pts treated in the first line setting within SF3B1-mutated group, 28.1% received a BCL-2 inhibitor combined with anti-CD20 monoclonal antibody (anti-CD20 mAbs), and 28.1% received a BTK inhibitor (BTKi) +- anti-CD20 mAbs. A subset of pts was treated in clinical trial, 19% underwent triple combination therapy: 13.6% were administered a BCL-2 inhibitor, anti-CD20 mAbs, and a BTKi, while 5.4% were treated with a phosphoinositide 3-kinase inhibitor, anti-CD20 mAbs, and a BTKi. Additionally, 6 pts (5.4%) received chemotherapy combined with a BTKi. Only 9% were given chemotherapy alone. OS did not differ between the cohorts with SF3B1 mutations and the wild-type; the 4-year OS rates were 98.3% (95% CI: 96.1-100%) and 97.5% (95% CI: 96.5-98.6%), respectively. Conclusion The SF3B1 mutation in pts with CLL is associated with higher Rai stage, elevated B2M, unmutated IGHV status, and association with del(11q), and a shorter time between diagnosis and first treatment, but did not lead to a difference in overall survival in the era of targeted therapies.