Characterization of BTK mutation pathologic status and association with known prognostic mutations (TP53, ATM, BIRC3, SF3B1, NOTCH1) in CLL: An AACR Project GENIE Analysis

Background

The advent of targeted BTK inhibition revolutionized treatment for CLL; however, the development of resistance mutations following treatment initiation diminishes long-term efficacy. We present primary analysis of the AACR GENIE v15 clinicogenomic database characterizing the landscape of BTK mutations stratified by pathologic status and assessing for co-association between known prognostic gene mutations.

Methods

1871 CLL samples with 1478 unique patient IDs were identified. Data for BTK, TP53, ATM, BIRC3, SF3B1, and NOTCH1 mutations were isolated. The OncoKB mutational database was utilized to classify registry BTK mutations as ‘inconclusive’, ‘likely oncogenic’, ‘resistance’, or ‘not characterized’. Independence analysis (odds ratio (OR), chi-squared test of independence) was performed to assess for pairwise association of co-mutation prevalence.

Results

Sample associated demographic characteristics were: Gender (Male 61.8%, 38.2% Female), Race (White 86.0%, Black 4.17%, Asian 2.78%, Other/Unknown 7.06%), Ethnicity (Non-hispanic 91.0%, Hispanic 3.00%, Other/Unknown 6.04%). The median age at sequencing was 65. 104 BTK mutations were identified and characterized as: 76 ‘resistance associated’, 22 ‘likely oncogenic’, 2 ‘inconclusive’, and 4 ‘not characterized’; the most common resulting amino acid changes were Cys481Ser (51.9%), Leu528Trp (10.6%), Cys481Arg (9.6%), and Val416Leu (9.6%). Statistically significant co-mutational associations included: BTK and TP53 (OR 6.40, 95% CI 3.93–10.4, p<0.001), TP53 and SF3B1 (OR 2.75, 95% CI 2.03-3.73, p<0.001), BTK and ATM (OR 2.55, 95% CI 1.44 – 4.35, p = 0.003), BTK and SF3B1 (OR 2.50, 95% CI 1.43 – 4.35, p = 0.002).

Conclusions

Our analysis illustrates the prevalence of pathologic BTK mutations in a cohort of real-world CLL patients undergoing next generation sequencing. A significant association of prevalence between pathologic BTK mutations with known prognostic TP53 mutations was identified. Further co-mutational association was noted with previously identified prognostic genes (SF3B1, ATM). This informs collective understanding of mutational prevalence in a real-world mixed pre and post treatment cohort.