Efficacy of Therapies Following Venetoclax Discontinuation in CLL: Focus on B-Cell Receptor Signal Transduction Inhibitors and Cellular Therapies

Introduction: Venetoclax (VEN) based therapy has become a standard of care in front line and relapsed-refractory (R/R) CLL based on favorable efficacy and toxicity. Whereas prospective data regarding activity of therapies following ibrutinib (IBR) or idelalisib (IDE) are available in the settings of progression (VEN, non-covalent BTKi) and intolerance (acalabrutinib), how best to manage patients (pts) who discontinue (dc) VEN remains a key unanswered question. With the increased use of VEN in early lines of therapy (LOT; CLL 14, MURANO), the activity of BTK inhibitors (BTKi) and cellular therapies following VEN becomes a critical issue. No prospective study has addressed this question, and currently reported VEN clinical trials have limited information about subsequent treatments. While recent data describe VEN resistance mechanisms (Guieze 2018, Blombery 2019), the impact of VEN resistance on efficacy of post VEN therapies is unknown. To address this gap, we conducted an international study to identify a large cohort of pts who dc VEN and have been subsequently treated. Methods: We conducted an IRB approved multicenter (31 US, EU, South American sites, in partnership with UK CLL Forum and CORE registry), retrospective cohort study of CLL pts who dc VEN for any reason. We examined demographics, dc reasons, responses, survival, adverse events (AEs) and activity of post VEN therapies. Primary endpoints were overall response rate (ORR) and progression free survival (PFS) for the post VEN treatments stratified by treatment type (BTKi, PI3Ki and cellular therapy: CAR-T or alloHSCT). ORR was defined by iwCLL criteria and PFS was defined from VEN dc to disease progression (PD), death, or last follow up for next treatment. Pts were further stratified by BTKi (resistant / intolerant) and PI3Ki exposure prior to VEN. PFS-2 was defined as time from VEN start to tumor progression on IBR or death from any cause. Results: 326 CLL pts who dc VEN in the front line (4%) and R/R settings (96%) were identified. The cohort was 69% male, 87% white, median (med) age 66 (38-91) at VEN start, 27% treated with VEN based combinations (n=88, med 6 cycles anti-CD20 abs). Pre VEN prognostic features: 82% IGHV unmutated (n tested=166), 47% del17p (n=306), 45% TP53 mut (n=217), 39% complex karyotype (n=273), 23% BTK mut (n=79), 18% NOTCH1 mut (n=103), 10% PLCγ2 mut (n=74). Pts received med 3 therapies (0-11) prior to VEN; 40% were BTKi naïve (n=130), 60% were BTKi exposed (196) and 81% were IDE naïve (n=263). Most common reasons for VEN dc were PD (38%), AE (20%), Richter\u27s transformation (RT, 14%), 8% pt preference, and HSCT 5%. Of 326 pts who dc VEN, 188 (58%) were treated with a subsequent LOT, 61 are alive and untreated and 77 died prior to a subsequent LOT. Post VEN sequencing analyses focused on BTKi, PI3Ki and cellular therapy (CAR-T or alloHSCT) activities following VEN dc (Table1). ORR to BTKi was 84% (n=44) vs. 54% (n=30, p\u3c.001 for ORR) in BTKi naïve vs. exposed patients (estimated med PFS 32 months (M) for BTKi naïve, 4 M in BTKi resistant, not reached in BTKi intolerant; Figure 1AB). ORR to PI3Ki was 47% in PI3Ki naïve pts following VEN, though responses were not durable (med PFS 5 M; Figure 1C). 66% responded to CAR-T post VEN (n=18), med PFS 9 M; med PFS was not reached for 19 pts who underwent alloHSCT post VEN (Figure 1D). Med PFS-2 for pts treated with VEN followed by IBR was not reached with med follow up 22 M (24 M PFS 78%, Figure 2). Med PFS for RT pts treated post VEN was 5 M (variable therapies). Conclusions: In the largest experience of therapies following VEN dc in CLL, we demonstrated that therapy selection following VEN requires consideration of prior novel agent exposure and reasons for discontinuation. For BTKi naïve pts, selection of a covalently binding BTKi results in high ORR and durable remissions. PFS-2 data provide reassurance for using VEN prior to IBR. For BTKi exposed pts, BTK inhibition is not effective in the setting of BTKi resistance but should be considered if prior BTKi intolerance. PI3K inhibition following VEN does not appear to result in durable remissions even in PI3Ki naïve pts, suggesting possible overlap in resistance mechanisms (BTK or VEN with PI3K). We conclude that BTKi in naïve or previously responsive pts and alloHSCT following VEN appear to be the most effective strategies with durable responses. These data suggest that a number of effective regimens exist for post VEN pts, providing support for VEN use earlier in the course of CLL