Analysis of Minimal Residual Disease in Follicular Lymphoma Patients in Gadolin, a Phase III Study of Obinutuzumab Plus Bendamustine Versus Bendamustine in Relapsed/Refractory Indolent Non-Hodgkin Lymphoma

Background

Minimal residual disease (MRD) response after first-line treatment of follicular lymphoma (FL) is likely to predict clinical course. The prognostic relevance of MRD in relapsed/refractory (r/r) FL remains unclear. We report MRD assessment with respect to clinical outcomes in r/r FL pts in GADOLIN (NCT01059630). GADOLIN is an open-label, multicenter, randomized phase III study of pts with r/r indolent NHL (refractory to rituximab) to investigate the efficacy and safety of obinutuzumab (G) plus bendamustine (B) followed by G maintenance vs B alone. A significant improvement in PFS in the G-B arm was reported (Sehn L, et al. ASCO 2015).

Methods

MRD was analyzed by t(14;18) and/or Ig variable domain allele-specific RQ-PCR in pts with a clonal marker detectable at diagnosis in peripheral blood (PB) or bone marrow (BM) by consensus PCR. (Assessment of Ig rearrangements allows detection of a lymphoma marker in pts with a t(14;18) breakpoint not detected by generic PCR and avoids false positive signals, as a low level of the translocation can be detected in some healthy individuals.) Assays were designed with a sensitivity of 10^-5 and accepted for MRD assessment when a sensitivity of ≤10^-4 was reached. Results were evaluated according to ESG-MRD criteria. MRD status was analyzed at interim staging (C4, D1) and after end of induction (EOI), and defined as negative (–ve) if RQ-PCR and subsequent nested PCR produced a –ve result, i.e. achieving an MRD response. PFS was measured from the EOI date.

Results

321 of 396 randomized pts were diagnosed with FL. Baseline samples (PB and/or BM) were available for 285 of the 304 FL pts who had completed induction at the clinical cut-off date (1 Sep 2014; FL-ITT population). A clonal marker was detected in 183 (64%) of these pts; 128/183 (70%) had a RQ-PCR assay fulfilling the sensitivity criteria. EOI samples were available for 93 pts (biomarker-evaluable population) and 64 had a PB sample at mid-induction to assess MRD-response kinetics. The distribution of age, stage, and FLIPI in the biomarker-evaluable population was similar to the non-evaluable population of the B arm, while there was an enrichment of younger age, advanced-stage disease, and high risk FLIPI in the biomarker-evaluable population of the G-B arm.

MRD response was analyzed in 93 pts at EOI and was significantly higher in pts receiving G-B induction, with 42/51 (82%) achieving MRD –ve status compared with 18/42 pts (43%) in the B arm (p<0.0001, Chi-Squared). At mid-induction, 30/39 (77%) pts in the G-B arm achieved early MRD –ve status vs 10/25 (40%) in the B arm (p=0.0029;Table). MRD response was associated with clinical CR; 2/33 (6%) MRD positive (+ve) pts vs 17/60 (28%) MRD –ve pts achieved a CR. Moreover, 39/63 pts with partial remission were MRD –ve.

Pts in the G-B arm who were MRD –ve at EOI had an improved PFS at 24 mo post-EOI (74%; median PFS not reached) compared with the B arm (21%; median PFS, 7.6 mo). PFS for MRD non-responders was comparably poor in both treatment arms; all pts progressed before 24 mo with a post-EOI median PFS of 5.4 mo (G-B) and 3.0 mo (B;Figure).

Discussion

Our results suggest that G significantly contributes to the depth of response to B vs B alone during induction treatment and support the notion that MRD status at EOI treatment is a sensitive marker of efficacy in the setting of r/r FL. MRD response identifies a prognostically favorable group of pts that appear to benefit from treatment with G-B at relapse. The improved PFS outcome suggests that these pts also benefit from G maintenance. Pts without an MRD response had a very poor prognosis, irrespective of treatment arm. Future analyses will assess MRD kinetics during maintenance and follow-up.

Table. MRD Status by Treatment Arm at EOI and mid induction

Figure. PFS from the Date of the EOI Sample, by Treatment Arm and MRD Status