Long-Term Remissions with MRD-Guided Acalabrutinib, Venetoclax and Obinutuzumab in Relapsed/Refractory CLL: Follow-up Efficacy and Circulating Tumor DNA Analysis of the CLL2-Baag Trial

Background

The phase 2 CLL2-BAAG trial tested a triple combination of acalabrutinib (acala), venetoclax (ven) and obinutuzumab (obi) after an optional debulking with bendamustine in patients (pts) with relapsed/refractory (r/r) CLL. In its primary endpoint analysis, 75.6% of pts achieved undetectable MRD (uMRD) in peripheral blood (PB) at final restaging (RE) after approximately 6 months of the triplet (Cramer, Lancet Haematol. 2022). We now report efficacy data and circulating tumor DNA (ctDNA) analyses with additional follow-up.

Methods

In the induction phase, obi was started in cycle 1 (days 1, 8, 15) and given 4-weekly in cycles 2-6. Acala was added in cycle 2 and ven in cycle 3. Maintenance treatment with continuous acala and ven and 3-monthly obi was administered until achievement of a deep remission with uMRD <10^-4 in PB or for up to 24 months (mo). MRD was measured centrally by flow cytometry (FCM) in PB and by digital droplet PCR (ddPCR) of patient-specific VDJ rearrangements and CLL-related mutations in blood plasma. uMRD was defined as <1 CLL cell/10,000 leukocytes (<10^-4) and no detected ctDNA.

Results

In total, 46 pts were included in the trial, 1 pt had to be excluded from the analysis due to a retrospectively noted violation of exclusion criteria. The median number of previous treatments was 1 (range 1–4) and 18 pts (40%) had received a BTK inhibitor (BTKi) and/or ven (BTKi: 8, ven: 7, both: 3) prior to inclusion, 14/44 pts (31.8%) had del(17p) and/or TP53 mutations, 34 (75.6%) had unmutated IGHV.

With a median observation time of 34.4 mo (range 12.0-39.2), all 45 pts are off study treatment. The median treatment duration was 14.7 mo (range 6.1-32.9), 2 pts (4.4%) discontinued treatment during the first 6 induction cycles, both due to adverse events (AEs). Following the MRD- and response-guided approach, 25 pts (55.6%) discontinued therapy according to confirmed uMRD and 9 pts (20.0%) completed the maximum of 8 maintenance cycles due to persisting MRD and/or lack of a complete response.

As reported, the uMRD rate in PB at RE was 75.6% with 10 pts showing detectable MRD ≥10^-4. In the course of maintenance, 7 of these 10 pts (70%) achieved uMRD, 1 pt had a Richter’s transformation (RT) and 2 pts still had detectable MRD after 8 cycles of maintenance (Figure 1A).

With all pts off-treatment, uMRD <10^-4 in PB was achieved in 42 of 45 pts (93.3%) at any time point, including 17 of 18 pts (94.4%) previously exposed to ven and/or a BTKi and 13 of 14 pts (92.9%) with TP53 aberrations. The median time to uMRD in PB was 5.4 months from the start of study treatment.

Median progression-free survival (PFS) was not reached, the estimated 30-mo PFS rate for the whole analysis population was 88.2% (Figure 1B). A similar PFS was shown for pts with TP53 aberrations (30-mo PFS 85.7%) and pts previously exposed to BTKi and/or ven (30-mo PFS 93.3%). The estimated overall survival at 30 mo was 100%. One patient died of COVID-19 18 mo after the end of study treatment.

In total, 564 paired FCM/ctDNA samples were available. Overall, 17 MRD recurrences (after uMRD by both methods) occurred. Of these, 11 were first detected by ctDNA, and 3 by FCM (+3 with concordant detection, Figure 1C). Among those with earlier ctDNA detection, more pts appeared to have high-risk genetics (unmutated IGHV in 10/11 pts, complex karyotype in 4/8 evaluated pts), del(11q) and prior exposure to ven/BTKi compared to those with earlier detection by FCM, and MRD recurrences seemed to occur earlier (mean, 7.1 mo after end of treatment). Notably, in 2 of 3 CLL-type progressions (PD, both nodal with normal lymphocyte counts), ctDNA was detected 6 mo/14 mo before MRD by FCM and 7 mo/17 mo before clinical PD, in the third pt ctDNA and FCM MRD recurred at the same time. In a pt with RT (DLBCL), patient-specific VDJ ctDNA was detected at diagnosis while FCM MRD remained negative.

At the time of the data cut (02/2023) 50 serious AEs (SAEs) were reported, 33 were of CTC G3, 6 of CTC G4 and 1 CTC G5. The most common SAEs were COVID-19 (n=8), other pneumonias (n=5), infusion-related reactions (n=5) and neutropenias (n=3).

Conclusions

Time-limited MRD-guided acala, ven and obi led to an estimated 30-mo PFS rate of 88.2% in pts with r/r CLL. Remissions deepen with ongoing maintenance and almost all pts achieve uMRD in PB including pts with TP53 aberrations and those treated with BTKi and/or ven in prior lines. The addition of ctDNA-based analyses to conventional MRD assessment by FCM seems to improve early detection of relapses.