PS1251 CLINICO‐BIOLOGICAL CHARACTERISTICS AND TREATMENT OUTCOMES FOR BLASTOID MANTLE CELL LYMPHOMA PATIENTS INCLUDED IN CLINICAL TRIALS. A LYSA STUDY

Background: Blastoid Mantle Cell Lymphoma (B‐MCL), including blastic and pleomorphic variants, is a rare morphological subtype of MCL, accounting for about 10–15% of all cases, with a more adverse prognosis compared to the classical form. Few data are currently available in the literature focusing specifically on this rare subtype of MCL. Aims: The aim of our study was to perform a retrospective analysis of all B‐MCL from the LYSA cohort included in 5 prospective clinical trials conducted from 2004 to 2012 in order to compare their clinico‐biological characteristics and outcomes to classical variants. Methods: Individual data from patients enrolled in 5 studies were analysed, respectively RiPAD+C (Houot R. et al, Ann Oncol 2011), RiBVD (Gressin R. et al, Haematologica 2018), MCL Younger (Hermine O. et al, Lancet 2016), MCL Elderly (Kluin‐Nelemans H.C. et al, NEJM 2012) and LYMA (Le Gouill S. et al, NEJM 2017) trials. A centralised pathological review of all cases was performed at the time of inclusion in each clinical trial, but an additional central review, according to 2017 WHO classification, was specifically achieved by expert pathologists from the LYSA for all B‐MCL. Univariate analysis of Progression‐Free Survival (PFS) and Overall Survival (OS) were performed and a stratified multivariate Cox model was used to assess the correlation between histologic subtypes and outcomes. Results: Individual data from 773 patients were collected. Among them, 59 had a B‐MCL subtype confirmed by central review (53 pleomorphic and 6 blastic variants) and 714 had a classical subtype. The two cohorts were similar in age (median: 61.2 y; range: 27–83), ECOG status, B symptoms, bone marrow and extra‐nodal involvement, MIPI risk group (low: 39%; intermediate: 30%; high: 31%) and haematological parameters (Hb level, leukocytes and lymphocyte counts). B‐MCL patients differed significantly in Ann Arbor stage (stage III‐IV: 86% in B‐MCL vs 96%, p = 0.002), bulky disease (31% in B‐MCL vs 19%, p = 0.040), LDH level (LDH > normal limit: 63% in B‐MCL vs 40% p  < 0.001) and Ki67% (Ki67 > 30%: 87% in B‐MCL vs 29%, p  < 0.001). With a median follow‐up of 60 months, univariate PFS was significantly lower for B‐MCL with a median of 28 mo (95% CI = 15‐NR) compared to 59 mo (95% CI = 54–72) for classical variants (p = 0.048). Univariate OS was also significantly lower for B‐MCL patients with a median of 65 mo (95% CI: 29‐NR) vs 109 mo (95% CI: 94‐NR) (p = 0.005). In multivariate analysis stratified by studies, B‐MCL subtype (HR (95% CI) for PFS 1.50 (1.03–2.19), OS 1.82 (1.21–2.74)) elevated LDH level (HR (95% CI) for PFS 1.62 (1.3–2.03), OS 1.79 (1.36–2.35)) and intermediate/high MIPI scores (HR (95% CI) for intermediate MIPI: PFS 1.64 (1.21–2.22), OS 1.82 (1.24–2.67); high MIPI: PFS 2.71 (1.96–3.74), OS 3.56 (2.4–5.29)) were significantly associated with a worse PFS and OS. Summary/Conclusion: This study describes one of the largest cohorts of well‐characterized B‐MCL patients included in 5 prospective clinical trials. All pathological diagnosis of B‐MCL were centrally reviewed and strictly classified according to the 2017 WHO classification. B‐MCL variant was associated with high LDH level, Ki67 > 30% and bulky disease. PFS and OS were significantly lower for B‐MCL patients compared to patients with a classical form.