Impact of Minimal Residual Disease Analysis in the Era of Rituximab Maintenance in Follicular Lymphoma: Data from "FOLL12" Phase III Trial of the Fondazione Italiana Linfomi

Introduction. Minimal residual disease (MRD) analysis has been largely used in follicular lymphoma (FL) for outcome prediction. However, despite the broad employment of rituximab maintenance for long-term disease control, there are few published data describing the prognostic impact of MRD monitoring during anti-CD20 maintenance or pre-emptive rituximab strategies. Therefore, we performed a wide and comprehensive MRD analysis at several predefined time points in the phase III, “FOLL12” trial [Luminari JCO 2022], sponsored by the Fondazione Italiana Linfomi (FIL). This is the largest MRD-driven clinical trial ever conducted in FL comparing conventional rituximab maintenance (after R-CHOP or BR) vs a combined PET/MRD response-adapted post-induction approach. We here present for the first time the MRD results stratified by post-induction treatment arm. Methods. Peripheral blood (PB) and bone marrow (BM) samples were centralized at the four Italian Euro-MRD certified laboratories of the FIL MRD Network. MRD was assessed with consensus primers on IGH::BCL2 rearrangements by RQ-PCR at end of induction (EOI) and every six months thereafter during either two years rituximab maintenance (reference arm, REF) or a response-adapted approach, offering up to three cycles of four, weekly pre-emptive rituximab infusions only in case of MRD positivity (experimental arm, EXP), till month 24. Results. High MRD negativity rates were obtained at EOI in both treatment arms (REF vs EXP: 92% vs 88% p=0.255), but the MRD kinetics changed overtime during the 24 months after EOI: if 60% of patients steadily maintained MRD negativity, 27% experienced at least one MRD recurrence (being MRD negative at EOI) while 13% converted at least once from MRD positive to MRD negative; importantly, only 1% of patients steadily maintained MRD positivity overtime without clinical relapse. Actually, focusing on the time points after EOI MRD negativity rates decreased in patients of the EXP arm vs patients receiving standard rituximab maintenance, with an overall higher incidence of MRD positivity in EXP vs REF during the 2 years after EOI (relative risk of MRD positivity: 1.78, 95% CI 1.30-2.45, p<0.001). In particular, patients in REF were characterized by lower rates of MRD recurrence vs EXP (18% vs 36%, p=0.001), accounting for an overall doubled risk ratio of MRD recurrence in EXP vs REF (RR 2.03, 95%CI 1.31-3.16, p=0.001). Consequently, sustained MRD negativity at 12 months after EOI was more frequent in REF than in EXP: 90% (111/124) vs 78% (103/132), p=0.017. On the other hand, the effect of either therapeutic strategy in inducing MRD negativization in MRD positive cases was limited and similar: 14% of patients in REF vs 12% in EXP, respectively (RR 1.15, 95% CI 0.61-2.17, p=0.671). Interestingly, among MRD positive patients, no difference in MRD quantitative burden was noted between the two study arms. MRD positivity at EOI was predictive of a worse PFS in EXP arm (5-yr PFS 26% vs 61%, HR 3.05, 1.78-5.22, p<0.001), while MRD positive patients at EOI receiving maintenance rituximab were not strongly associated with adverse prognosis at present clinical update, if compared to MRD negative ones (5-yr PFS 82% vs 76% p=0.542). However, the persistence or reappearance of an MRD positive signal in PB during the second year after EOI (time points +12, +18 and +24 months) predicted a worse PFS, adjusted by treatment arms (HR 2.58, p=0.003). Finally, it is interesting to note that MRD negative patients in REF experienced a statistically significant better outcome than MRD negative patients in EXP (5-yr PFS 76% vs 61% p=0.002), highlighting that rituximab maintenance benefits also patients with sustained MRD negativity, by preserving this favorable status overtime and delaying the occurrence of clinical relapse. Conclusions. These data suggest that a regular MRD monitoring in PB is needed to better understand the complex MRD kinetics in an indolent disease such as FL, in order to early identify patients at risk of relapse. Notably, rituximab maintenance in FL improved patients’ outcome independently from their MRD status at EOI, mainly by halving the risk of MRD recurrence overtime, thus favoring a sustained MRD negativity and eventually delaying the occurrence of clinical relapse.