MRD-guided ibrutinib plus venetoclax improves outcomes in CLL patients with TP53, ATM, or NOTCH1 aberrations compared to ibrutinib and FCR: Results from the Phase III NCRI FLAIR trial

Abstract

Abstract Introduction: We have previously reported improved survival in patients (pts) with recurrent gene mutations following treatment with ibrutinib and rituximab (IR) compared to fludarabine, cyclophosphamide and rituximab (FCR). In this study we assessed the impact of baseline gene aberrations on Progression-Free Survival (PFS) in patients receiving MRD-guided ibrutinib plus venetoclax (I+V) compared to those treated (tx) with ibrutinib (I & IR) or FCR in the FLAIR trial. We also report on treatment outcomes based on Immunoglobulin Heavy Chain Variable region (IGHV) Somatic Hypermutation (SHM) and stereotyped subset #2 (S#2) status. Method: FLAIR (ISRCTN01844152) is an ongoing phase III, multicentre, randomised, controlled, open label trial comparing IR and FCR in patients with untreated CLL requiring therapy according to IWCLL criteria. It was subsequently adapted to compare I+V and I alone with FCR. Chromosomal abnormalities were detected using Fluorescent In Situ Hybridization (FISH) with targeted probes for chromosomes 11q and 17p. Pts with >20% 17p deletion (del) were excluded from the trial. Extracted DNA was sequenced using Illumina MiSeq and analysed using an in-house pipeline. Amplicon based targeted sequencing of 33 recurrently mutated (mut) genes was performed in parallel. Detected variants were reported down to minimum variant allele fraction of 3% at 100X coverage. SHM status was determined by PCR amplification of IGHV-IGHD-IGHJ gene rearrangements using IGHV leader/FR1 primers. Bidirectional Sanger sequencing was analysed using IMGT V-Quest and ARResT/AssignSubsets. Cox's proportional hazards was used to estimate Hazard Ratios (HR). Log-rank test was used to estimate p-value where HR was zero (zero progression events in one or more treatment arms). Results: 1172 patients were randomly assigned to receive I+V (n=260), I&IR (n=263 & 386 resp.) or FCR (n=263). Somatic gene mutations were assessed in 1161/1172 (99.1%) pts at baseline and mutations were detected in 715/1161 (61.6%). The frequency of these mutations in pts ranged from 0.2-16.7% with mutations in SF3B1 (16.7%), ATM (14%), NOTCH1 (11.6%), RPS15 (5.1%), MYD88 (5%), TP53 (4.7%), BIRC3 (4.7%), POT1 (4.5%), and BRAF (4.2%) being the most frequent. Del(11q) was detected by FISH in 187/1150 pts (16.3%). IGHV SHM status was available for 1077/1172: 620 (57.6%) IGHV Unmutated (IGHV-UM; >98% of nucleotide identity to germline), 457 (42.4%) IGHV Mutated (IGHV-M). 77/1077 pts were assigned to CLL S#2 (7.1%, n=42 IGHV-M, n=35 IGHV-UM). A significant improvement in PFS was observed between TP53mut pts tx with I+V compared to I&IR and FCR (p=0.04 and 0.009 resp.). The 5yr PFS for TP53mut pts was 100% for I+V (n=11), 70% for I&IR (n=31) and 62.3% for FCR (n=12). Similar improvements in PFS were also observed for del(11q), ATMmut, SF3B1mutand NOTCH1mut pts tx with I+V compared to I&IR and FCR (p=0.001 and <0.001; HR:0.1 p=0.02 and HR:0.03 p<0.001; HR:0.21 p=0.03 and HR:0.07 p<0.001; HR:0.22 p=0.04 and HR:0.09 p=0.002 resp.). For RPS15mut pts a significant improvement in the PFS was seen between pts tx with I+V compared to FCR (p=0.001) but not I&IR (p=0.24). The 5yr PFS for pts tx with I+V, I&IR or FCR was 100%, 82% and 43.5% for del(11q); 97.6%, I&IR 79.7% and 39.2% for ATMmut; 94.1%, 79.5% and 48.3% for SF3B1mut; 97.0%, 75.2% and 54.6% for NOTCH1mut and 100%, 90.2% and 31.7% for RPS15mutpts resp. For IGHV-UM pts (excl. S#2) a significant improvement in PFS was observed when tx with I+V compared to I&IR and FCR (HR:0.20 p<0.001 and HR:0.07 p<0.001 resp.). The 5yr PFS was 94.9% for I+V, 79.3% for IR&I and 49.7% for FCR. A similar improvement was also observed for IGHV-M pts (excl. S#2) tx with I+V compared to FCR but not IR (HR:0.37 p=0.007 and HR:0.64 p=0.188 resp.). The 5yr PFS was 90.1% for I+V, 84.7% for I&IR and 75.2% for FCR. For S#2 pts tx with I+V PFS was significantly improved compared to FCR but not I&IR (p=0.002 and 0.119 resp.). The 5yr PFS was 100% for I+V, 88.9% for I&IR and 52.7% for FCR.Conclusions: Our results from theFLAIR trial demonstrate that targeted treatment is highly effective at mitigating the poor outcome previously associated with unmutated IGHV-UM SHM status, CLL S#2 and recurrent gene aberrations when treated with chemoimmunotherapy. Notably, we report exceptional responses following MRD-guided I+V treatment compared to both the Ibrutinib and FCR arms of the trial, especially in pts with TP53, ATM or NOTCH1 aberrations.