2003 - Genomic Evolution and Resistance to Pirtobrutinib in Covalent BTK-Inhibitor Pre-treated Chronic Lymphocytic Leukemia Patients: Results from the Phase I/II BRUIN Study

Background:
Pirtobrutinib, a highly selective, non-covalent (reversible) BTK-inhibitor (BTKi), demonstrated broad efficacy in patients (pts) with chronic lymphocytic leukemia (CLL)/SLL who were resistant to cBTKi. Mechanisms of resistance to pirtobrutinib have not been systematically analyzed to date.

Aim:
To explore the genomic evolution of pirtobrutinib resistance in cBTKi pre-treated CLL/SLL pts.

Methods:
cBTKi pre-treated CLL pts treated with pirtobrutinib monotherapy in the phase 1/2 BRUIN trial (NCT03740529) who subsequently developed disease progression (PD) were included in this analysis. Targeted next-generation sequencing (NGS) of all exons of 74 relevant genes was centrally performed on peripheral blood mononuclear cells collected at baseline and at or near PD. Somatic mutations were reported with a limit of detection (LoD) of 5% variant allele frequency (VAF). Manual inspection of BTK exons with a LoD of 1% was performed to detect whether BTK mutations identified at PD were present at baseline at < 5% VAF.

Results:
As of 29 July 2022, 49 cBTKi pre-treated CLL pts who progressed on pirtobrutinib had paired NGS data available at baseline and PD. In this group, the median age was 69 y (range, 36-86), median number of prior lines was 4 (range, 1-10), and 41 pts (84%) had discontinued prior cBTKi due to PD. Pts received one or more of the following cBTKi: ibrutinib (n=44, 90%), acalabrutinib (n=10, 20%), or zanubrutinib (n=1, 2%). The ORR to pirtobrutinib (including PR-L) was 80%.

The most common alterations at baseline were mutations in BTK (51%), TP53 (49%), ATM (27%), NOTCH1 (20%), SF3B1 (18%), and PLCG2 (10%). Among the 25 pts with ≥1 BTK mutation detected at baseline, BTK mutations included C481S (n=23), C481R (n=4), C481Y (n=2), C481F (n=1), T474I (n=1). BTK C481 VAF decrease or complete clearance was observed at PD in nearly all pts (92%, 22/24, median VAF decrease =100%). NGS of samples at PD showed 71% (35/49) of pts acquired ≥1 mutation, with 55% (27/49) acquiring ≥1 BTK mutation. Among these 27 pts, a total of 36 acquired BTK mutations were identified; the most common were gatekeeper mutations (T474I/F/L/Y, 17/49, 35%), kinase-impaired (L528W, 9/49, 18%), and variants of unknown significance (VUS) proximal to the ATP-binding pocket (6/49, 12%; including V416L (n=2), A428D (n=2), D539G/H (n=1), Y545N (n=1)) (Figure). Manual inspection for acquired BTK at PD in the corresponding baseline samples revealed that 9 mutations (in 8 pts) pre-existed at baseline at low VAFs (1-4%). These included 6 gatekeeper T474I/L, 2 kinase impaired L528W, and 1 VUS A428D. These pts responded to pirtobrutinib (6/8, 75% ORR) and had received prior ibrutinib (n=5) and acalabrutinib (n=4). The most commonly acquired non-BTK mutations were TP53 (7/49, 14%) and PLCG2 (4/49, 8%).

Conclusions:
Pts who progressed on pirtobrutinib showed clearance of BTK C481 clones and the emergence or outgrowth of non-C481 clones, particularly gatekeeper T474 and kinase-impaired L528W mutations, as well as other VUS. Many acquired BTK mutations were shown to pre-exist at baseline at low VAF, reflecting emergence on prior cBTKi. Importantly, these baseline kinase domain BTK mutations did not preclude pirtobrutinib efficacy. Approximately half of pts did not acquire BTK mutations and 29% did not acquire any mutations in this targeted panel, suggesting alternate resistance mechanisms. Whether similar patterns of resistance would manifest if pirtobrutinib was utilized in earlier lines of therapy or prior to cBTKi treatment remains uncertain.