First Prospective Data on Impact of Minimal Residual Disease on Long-Term Clinical Outcomes after Venetoclax Plus Rituximab Versus Bendamustine Plus Rituximab: Phase III MURANO Study

Introduction

Although CLL minimal residual disease (MRD) status is used in contemporary clinical trials aimed at maximizing response or determining treatment duration, its role as a predictive factor for PFS has only been established following chemoimmunotherapy. In contrast, whether MRD is a valuable tool for treatment choice in the era of novel targeted agents for CLL is unknown. Unlike kinase inhibitors, the BCL2 inhibitor venetoclax does result in undetectable MRD (uMRD). MURANO demonstrated significant PFS benefit for venetoclax + rituximab (VenR) given for a fixed duration vs bendamustine + rituximab (BR) in relapsed/refractory (R/R) CLL. Here we present analysis of peripheral blood (PB) MRD kinetics in relation to PFS in MURANO with long follow up, when all pts have completed therapy.

Methods

Pts were randomized to VenR (Ven 400mg/d for 2 yrs + R for first 6 mo) or BR (6 mo). Response was assessed clinically using complete blood count and physical exam at follow-up visits. PB MRD was analysed centrally by ASO-PCR and/or flow cytometry at Cycle 4, end of combination therapy (EOCT; mo 9) and every 3 mo thereafter until 3 yrs, then every 6 mo. As strong concordance between PB and bone marrow (BM) MRD with VenR was previously shown (Hillmen et al. ASCO 2018), we focus on PB MRD. Pts were categorized into uMRD (<1 CLL cell per 10,000 leukocytes [<10^-4]), intermediate (int)-MRD+ (≥10^-4 - <10^-2), and high-MRD+ (≥10^-2) status. Results are reported for the intention-to-treat (ITT) population.

Results

As of May 8 2018, median follow-up was 36.0 mo. VenR pts achieved high PB uMRD rates at EOCT (62% vs 13% for BR). The same pattern was observed through mo 24, when Ven single-agent treatment was scheduled to cease: 48% uMRD in VenR vs 2% in BR, and 16% int-MRD+ in VenR vs 7% in BR (Table 1). By then, 18% pts in the VenR arm had progressed/died/withdrew from study vs 66% in BR. Consistently high uMRD rates were observed in all VenR subgroups, including pts with high-risk cytogenetics and molecular factors: del(17p) and/or TP53 mutated: 57% at EOCT and 36% at mo 24; IGHV unmutated: 61% at EOCT and 51% at mo 24.

Landmark analysis of PFS by MRD status in PB at the EOCT response visit showed that, regardless of treatment arm, uMRD status was associated with longer PFS. Within MRD+ pts, the int-MRD+ group had improved PFS over high-MRD+ pts (Fig 1). Due to the very low rate of uMRD in the BR arm, all further analyses were performed in VenR pts; PFS curves overlapped for pts with uMRD, regardless of whether in CR or PR by investigator assessment, while inferior PFS for MRD+ PR/nPR pts becomes manifest from mo 18 after EOCT (Fig 2). MRD+ pts in CR (all int-MRD+) did as well as uMRD, but numbers are very small.

130 pts in VenR arm completed 2 yrs Ven treatment without PD; among them: 83 (64%) were uMRD, 23 (18%) were int-MRD+, 14 (10%) were high-MRD+ and 10 (8%) had missing data at mo 24. With 9.9 mo median follow-up from Ven completion, amongst uMRD pts at mo 24, the majority remained uMRD (57/83, 69%) and 26/83 (31%) converted to confirmed MRD+ (2 serial assay positive). Conversions were mainly to int-MRD+ (21/26, 81%), all of whom remain PFS event-free; 19% (5/26) converted to high-MRD+, 2 had PD per iwCLL criteria (Fig 3). Of pts converting from uMRD to MRD+ after cessation of Ven at mo 24, 40% (10/25) carried del(17p) and/or TP53 mutation, vs 23% (12/53) among those who did not convert. Of int-MRD+ pts at mo 24, 9/23 (39%) became high-MRD+ (2 had PD); there was 1 PD among the remaining 14. PD was seen in 11/14 (79%) of pts who were high-MRD+ at mo 24.

Conclusion

The higher PB uMRD rate observed at EOCT in the VenR arm vs BR arm was maintained at completion of Ven treatment, including in high-risk subgroups, consistent with maintained PFS benefit seen with longer follow up. VenR pts who achieved uMRD or int-MRD+ had durable PFS. MRD status powerfully identified distinct outcomes for pts with PR/nPR; longer follow-up is needed to determine the impact of MRD status among CR/CRi. In the first yr post-completion of fixed-duration VenR treatment, majority of uMRD pts remained uMRD; reemergence of MRD+, mainly at int levels, is seen in a minority of pts, and has shown a low rate of progression to clinical PD to date. These data are the first to demonstrate the value of PB MRD as a predictive marker of clinical outcome for the fixed-duration chemotherapy-free regimen, VenR. Our data corroborate that uMRD and int-MRD+ rate is a meaningful endpoint and desirable goal of CLL therapies, including targeted agents.

View largeDownload slide

View largeDownload slide

Close modal