Predictive Power of Early, Sequential Minimal Residual Disease and Fluorine-18-Fluorodeoxyglucose Positron Emission Tomography Monitoring in Young Patients with Mantle Cell Lymphoma in the Lyma Trial: A Lysa Study

PURPOSE Minimal residual disease (MRD) can predict outcomes in patients with mantle cell lymphoma (MCL) but data are limited for younger patients undergoing ASCT (autologous-stem-cell-transplantation) and rituximab maintenance (RM) and no data are available on the clinical value of combining MRD with positron-emission-tomography (PET). PATIENTS AND METHODS. Long term follow-up data from a phase III trial in newly-diagnosed MCL patients of < 66 years (LYMA; NCI NCT00921414, Sarkozy C et al, J Clin Oncol, 2023) were examined to determine the relationship between MRD, PET and progression-free survival (PFS), overall survival (OS) and progression-of-disease before 24 months (POD24). Pre- and post-ASCT timepoints were considered and analyses were conducted in both the RM and observation (Obs) arms. MRD was evaluated by quantitative polymerase chain reaction (qPCR; minimal sensitivity 10-4) of clonal immunoglobulin gene rearrangements, in peripheral blood (PB) and / or bone marrow (BM). MRD status was interpreted according to EURO-MRD guidelines. RESULTS: MRD assays were established in 223 of 299 enrolled patients (74.6%). The MRD patient set was broadly similar to the non-MRD set except for a higher incidence of bone marrow involvement and Ann Arbor stage II/III patients in the former. Non-evaluable MRD was mostly due to no diagnostic or follow-up samples, no MRD target or MRD assay failure. Overall, MRD measured in either PB and / or BM was negative in 123/195 patients (63%), pre-ASCT, and 155/192 patients (81%), post-ASCT. MRD negativity rates were higher in PB at pre-ASCT (77%, 144 of 187 MRD-evaluable patients) and post-ASCT (94%, 181/193 patients) compared to BM at either time-point [pre-ASCT: 64% BM-MRD negative, 115/180 patients; post-ASCT: 81% BM-MRD negative, 137/170 patients]. Pre-ASCT MRD status in PB (but not BM), was predictive of PFS (hazard ratio [HR]: 2.43 [95% CI, 1.53 to 3.86]; P=0.0001), OS (HR, 3.31 [95% CI, 1.87 to 5.85]; P<0.0001) and POD24 (OR, 5.41 [95% CI, 2.15 to 13.64]; P=0.0003). Post-ASCT, PB-MRD status (but not BM-MRD) was predictive of PFS (HR, 2.24 [95% CI, 1.01 to 4.95]; P=0.0414) (but not OS or POD24). When analyzed by treatment arm, patients in RM who were PB-MRD negative at pre-ASCT showed longer PFS and OS compared to those who were PB-MRD positive (HR, 2.86 [95% CI, 1.10 to 7.44]; P=0.0241 and HR, 2.94 [95% CI, 1.03 to 8.38]; 0.0361, respectively). Patients in the observation arm who were PB-MRD positive at pre-ASCT fared poorly when compared to patients who were PB-MRD negative with a shorter PFS (HR, 2.45 [95% CI, 1.24 to 4.83]; P=0.0079) and OS (HR, 3.96 [95% CI, 1.36 to 11.54]; P=0.0066). Insufficient events were available for post-ASCT analyses. However, RM provided superior PFS (HR, 0.44 [95% CI, 0.21 to 0.89]; P=0.0186) (but not OS) for PB-MRD-negative patients at pre-ASCT, while at post-ASCT, PB-MRD negative patients under RM showed improved PFS (HR, 0.32 [95% CI, 0.16 to 0.62]; P=0.001) and OS (HR, 0.45 [95% CI, 0.22 to 0.92]; P=0.0243). In view of the stronger predictive power of PB-MRD, over BM-MRD, we next asked whether integration of PET and MRD could improve outcome prediction. Superior outcome was associated to negativity for both PB-MRD and PET at either pre-ASCT (108/171 of patients with available PET/MRD data, 63%: PFS (HR, 2.14 [95% CI, 1.34 to 3.41]; P=0.0011), OS (HR, 2.195 [95% CI, 1.195 to 4.034]; P=0.0095), POD24 (OR, 2.98 [1.17-7.62]; P=0.0223) or post-ASCT (120/147 patients with available PET/MRD data, 81.6%: PFS (HR, 2.4 [95% CI, 1.28 to 4.50]; P=0.0048) or OS (HR, 3.02 [95% CI, 1.40 to 6.52]; P=0.0032) compared to PB-MRD and / or PET positive patients at either timepoint. When analyzed by treatment arm (RM or Obs), PET/PB-MRD status at pre- or post-ASCT did not impact outcome. However, RM maintained impact on PFS (but not OS) in post-ASCT (but not pre-ASCT), double-negative PB-MRD/PET patients (HR, 0.32 [95% CI, 0.14 to 0.69]; P=0.0025). CONCLUSION: Early sequential PB-MRD monitoring at the pre- and post-ASCT treatment phase is a powerful strategy for early clinical outcome prediction in MCL in the ASCT RM setting and has potential as a surrogate endpoint for clinical trials. Our results do not support de-escalation of RM upon MRD/PET negativity either pre- or post-ASCT. Combining MRD with PET offers improved predictive power and warrants evaluation as a new platform for response-adapted treatment in MCL in a constantly changing therapeutic environment.