Clonal Dynamics after Venetoclax-Obinutuzumab Therapy: Novel Insights from the Randomized, Phase 3 CLL14 Trial

Background

Measurement of minimal residual disease (MRD) is an established and sensitive prognostic tool to assess the depth of response during and after treatment of chronic lymphocytic leukemia (CLL) and to understand disease dynamics after treatment. Retracing these kinetics is paramount to understand which group of patients is at risk of relapsing despite initial MRD response. To this end, we here provide an analysis of clonal growth patterns in patients treated within the CLL14 trial.

Methods

In total, 432 patients with previously untreated CLL and co-existing conditions were randomized to receive chlorambucil or venetoclax (216 patients per arm) until completion of cycle 12 (à 28 days), and in combination with obinutuzumab for the first 6 cycles. The primary endpoint was progression-free survival (PFS), MRD was a secondary endpoint. MRD was analyzed by ASO-PCR and, as reported here, by next-generation sequencing (Adaptive clonoSEQ Assay, cut-off: 10^-4, 10^-5 and 10^-6). The limit-of-quantification of the clonoSEQ Assay is less than 10^-6. Samples from peripheral blood (PB) are collected every 3 to 6 months until 9 years from last patient enrolled; for this analysis, the longest follow-up MRD sample was from month 24 after treatment completion.

A patient-specific clonal growth rate model was formed: MRD level at time t is defined as MRD(t) = λ₀ x 10^μt, where λ₀ is the baseline MRD (i.e. MRD at end of treatment (EoT)), and μ is the growth rate parameter. For the time from MRD status λ₀ to new MRD state λ₁ , t_change was assumed (1/μ) x log₁₀ (λ₁/λ₀). An exponential regression model was fitted to each patient to estimate each respective λ₀ and μ, by which the patients’ MRD clone doubling time was calculated. Data are reported in relation to the intention-to-treat population; p-values are exploratory only. For modelling, only patients with at least two MRD assessments after EoT and who had not experienced PD before EoT were included.

Results

Two months after treatment completion (follow-up month 3), 40% (7%) of patients in the Ven-Obi arm (Clb-Obi arm) had uMRD levels <10^-6, 26% (13%) ≥ 10^-6 and <10^-5, 8% (14%) ≥ 10^-5 and <10^-4, 5% (21%) ≥ 10^-4 and <10^-2, and 3% (26%) ≥10^-2 (Figure A).

Most patients (56%) who had undetectable MRD (uMRD) levels at EoT already had uMRD levels at cycle 7, i.e. after completing Ven-Obi therapy. In 25% of the Ven-Obi treated patients, MRD response deepened after continuing with 6 cycles of venetoclax monotherapy (Figure B).

In a PFS landmark analysis after EoT, patients in the Ven-Obi arm with MRD levels ≤10^-5 had a 2-year PFS after EoT of approximately (approx.) 93%, while patients with detectable MRD >10^-2 had a 2-year PFS of approx. 37% (Figure D).

Patient-specific clonal growth rates were estimated for 123 patients in the Ven-Obi arm and 143 patients in the Clb-Obi arm (Figure C). Of note, 38 patients in the Ven-Obi arm and 4 patients in the Clb-Obi arm had MRD results below the assays limit of quantification (LOQ), indicating the deepest response beyond the assay’s sensitivity. These patients were not included as a growth rate could not be accurately assessed for those very-deep responders.

The average clonal growth rate in the Ven-Obi arm (n=123) was µ = 0.0034 (95% CI: 0.0030-0.0038), translating into an MRD doubling time of approx. 89 days (95% CI: 79-102) and an increase of the MRD clone size by 1 with log10 scale within approx. 296 days (95% CI: 263-337) (Figure E). For patients in the Clb-Obi arm (n=143), µ was 0.0042 (95% CI: 0.0038-0.0047), corresponding to a doubling time of approx. 71 days (95% CI: 65-79), and an increase of the MRD clone size by 1 with log10 scale within approx. 237 days (95% CI: 215-263). The average growth rate was lower in the Ven-Obi arm compared to the Clb-Obi arm (p=0.0057).

Analyses of growth rate patterns within various biological and clinical risk groups as well as genetic subgroups are currently being conducted and will be presented at the meeting.

Conclusions

This analysis demonstrates that individual clonal growth rates can be used to estimate the MRD doubling time after a fixed-duration treatment. Clonal growth was lower after Ven-Obi than after Clb-Obi, indicating more effective MRD eradication and clonal growth modulation with Ven-Obi. In a considerable subgroup of Ven-Obi treated patients, no clonal growth was measurable during observation, indicating deepest remissions.