PS1257 LYMPHOCYTE TO MONOCYTE RATIO (LMR) PREDICTS PROGRESSION‐FREE SURVIVAL INDEPENDENTLY FROM FLIPI AND IS USEFUL IN IDENTIFYING PATIENTS WITH EARLY DISEASE PROGRESSION

Background: Despite the overall good prognosis of follicular lymphoma (FL), approximately 20 to 30% of the cases experience early progression after first line therapy and suffer adverse survival (Casulo et al., 2015; Shi et al., 2017). Several efforts have been made to identify pre‐treatment factors useful to assess prognosis, however currently available parameters do not allow to accurately stratify patients before treatment. Lymphocyte to monocyte ratio (LMR) has been previously identified as an useful prognostic markers in diffuse large B‐cell lymphoma treated with Rituximab‐based chemotherapy (Rambaldi et al., 2013). Aims: In this study we aimed to evaluate the potential role of LMR in identifying patients with FL experiencing early progression after first line therapy. Methods: 831 patients with a diagnosis of grade 1–3a FL were identified in the Bergamo Lymphoid Cancer Registry (NCT03131531) between 1983 to 2017. Cases managed with watchful observation after diagnosis (N = 145) or receiving local therapy (N = 121) or without adequate clinical information or follow‐up data (N = 117) were excluded from the analysis. The remaining patients (N = 448) with stage II‐IV disease and receiving first line chemotherapy at diagnosis were included in the study. The majority of the patients were treated with Rituximab‐based therapy (72%). Chemotherapy regimens included CHOP (N = 298), CVP (N = 76), Bendamustine (N = 22), Chlorambucil (N = 34) or others (N = 18). Rituximab maintenance was administered to 61 patients. Patients with early progression were defined as those with a PFS event before 30 months from first‐line treatment start (PFS30). Receiver operating characteristic (ROC) analysis was performed to assess the optimal cut‐off of LMR to predict PFS30. Kaplan‐Meyer curves, log‐rank tests and Cox models were used for survival analyses. Results: In the study population, 26% of the patients experienced early progression and were characterized by a shorter median OS compared to the other group (6.3 vs 17 years, P  < 0.0001, Figure 1A). The optimal cut‐off of LMR to predict PFS30 was identified, with patients with LMR ≤2 or >5.2 (N = 176) being at higher risk for a PFS event compared to those with LMR >2 and ≤5.2 (N = 272) (Figure 1B). In multivariate analysis for PFS, LMR retained its significance independently from FLIPI (P = 0.0001). The simple addition of LMR to standard FLIPI permitted to refine patient prognosis in terms of PFS (Figure 1C) and OS (Figure 1D), with three groups identified i.e. low risk (FLIPI low and favorable LMR), intermediate risk (FLIPI low and unfavorable LMR; FLIPI int or high and favorable LMR) and high risk (FLIPI int or high and unfavorable LMR). Specifically, 57 (36%) patients with intermediate FLIPI were upgraded in the high risk category and 80 (54%) high risk FLIPI were downgraded into the intermediate one. Patients in the new high risk category were characterized by a 42% probability to have a PFS event before 30 months, compared to the 22% and 10% of the intermediate and low risk group, respectively. LMR was significantly associated with OS in univariate analysis (P = 0.047), but not in multivariate analysis with FLIPI (P = 0.11). The combined LMR and FLIPI index predicted survival with a 30‐months OS of 96%, 94% and 85% in the low, intermediate and high risk category, respectively (P = 0.0032). Summary/Conclusion: LMR is a significant prognostic factor for PFS in FL, independent from FLIPI and closely related to the use of Rituximab. The combination of LMR with currently available prognostic markers can be useful in identifying high patient before the start of first line treatment.