Genomic aberrations, IGHV mutation status and mutations in genes such as TP53 are established prognostic factors in CLL in the context of chemoimmunotherapy. Their role is less-well established when using chemo-free regimens such as obinutuzumab (GA-101) plus venetoclax (Ven-G).
This study evaluates the prognostic impact of genetic risk factors on obinutuzumab+chlorambucil (G-Clb) vs. obinutuzumab+Venetoclax (Ven-G).
We assessed the incidence and impact of genetic factors in the phase 3 CLL14 trial comparing G-Clb vs.Ven-G in patients with CIRS>6 or creatinine clearance < 70 ml/min. Genomic aberrations were assessed by FISH, IGHV by sequencing with a threshold of 98% homology and mutations via amplicon-based targeted NGS for TP53, NOTCH1, SF3B1, ATM, MYD88, FBXW7, POT1, BIRC3, XPO1, NFKBIE, EGR2 and RPS15 for variants with allele fraction ≥10%. Of the intention to treat population (n = 432) FISH/IGHV/NGS was assessable in 418/408/421 cases. PFS and OS were estimated via Kaplan-Meier method and compared by non-stratified log-rank test. Hazard ratios (HRs) were calculated via Cox proportional hazards regression modeling.
The incidence of genomic aberrations considering the hierarchical model were del(17p) 7%, del(11q) 18%, +(12q) 18% and del(13q) 35%. IGHV was unmutated in 61% of patients. The incidence of gene mutations was NOTCH1 23% (exon 34 and/or 3'UTR), SF3B1 16%, ATM 13%, TP53 10%, XPO1 6%, RPS15 5%, POT1 5%, BRAF 4%, BIRC3 4%, NFKBIE 4%, EGR2 4%, MYD88 2%, FBXW7 1%. High coincidence was found for del(17p) and TP53mut in 77% of all del(17p) cases and 57% of all TP53mut cases.
None of the parameters impaired overall response rate (ORR) to Ven-G at treatment completion. For G-Clb, the ORR rate was lower in patients with del(17p) (36% vs. 73%), del(11q) (58% vs. 74%), TP53mut (58% vs. 74%), ATMmut (56% vs. 75%) and BIRC3mut (33% vs. 74%).
At a median follow-up of 29 months, there were 107 events for PFS and 37 for OS in the intention to treat population. Del(17p) was the only genomic abnormality with impact on PFS in G-Clb (HR 4.6, p < 0.001) and Ven-G (HR 4.4, p = 0.001; Fig 1a). Similarly, TP53 mutations affected PFS in both treatment arms (G-Clb HR 2.7, p = 0.001; Ven-G HR 3.1, p = 0.01). However, TP53mut without del(17p) was not associated with shorter PFS. None of the other evaluated factors affected Ven-G efficacy, while for G-Clb del(11q) (HR 2.3, p = 0.002), BIRC3 (HR 4.0, p = 0.001), NOTCH1 (HR 1.8, p = 0.03) and unmutated IGHV (HR 3.4, p < 0.001) were adverse factors.
In several genetic subgroups including del(17p), del(11q), TP53mut, NOTCH1mut, SF3B1mut, ATMmut Ven-G was superior to G-Clb in PFS. Regarding IGHV, only patients with unmutated status had a significant PFS benefit from Ven-G in comparison to G-Clb (IGHVunm HR 0.2, p < 0.001; IGHVmut HR 0.6, p = 0.29; Fig.1b). Multivariable testing for interaction between treatment and IGHV mutational status was significant (p = 0.03) indicating unmutated IGHV as a predictive factor for increased benefit from Ven-G. OS was lower with del(17p) in both treatment arms (G-Clb: HR 11.0, p < 0.001; Ven-G: HR 3.4, p = 0.03) and with TP53mut, BRAFmut and IGHVunmut in the G-Clb arm (HR 5.5, p = 0.002; HR 6.6, p < 0.01; HR 5.4, p = 0.03), while none of the other factors were significantly associated with OS.
Prognostic value of genomic aberrations, IGHV and gene mutations were confirmed for G-Clb, while with Ven-G only del(17p) and TP53mut were associated with short PFS and only del(17p) with short OS. Unmutated IGHV was identified as a predictive factor characterizing a group of patients with particular benefit from Ven-G.