P53 and P16 Protein Expression Have Prognostic Value in Mantle Cell Lymphomas in the Lyma Trial in the Lysa Group

Introduction: Mantle Cell Lymphomas (MCL) are rare non-Hodgkin lymphomas with often poor outcome. Currently, definitive criteria predicting time to treatment failure and overall survival are based on clinical factors included the mantle cell lymphoma prognostic index (MIPI) and the proliferation index assessed by Ki67 expression. In order to find other predictive parameters we tested 4 immunohistochemical (IHC) markers usually correlated with a more aggressive behavior in tumors in a large series of MCL samples from patients enrolled in LYMA trial and treated with 4 R-DHAP then Rituximab-BEAM and autograft.

Methods: The LYMA trial includes 300 patients for whom we obtained specimen suitable for Tissue Micro Arrays (TMA) for 166 patients. On these formalin fixed paraffin embedded diagnostic TMA, immunohistochemistry was performed using p53, Myc, p16 and transferrin receptor (CD71) antibodies . These markers were scored by quantifying the percentage of cells stained on each TMA spot. IHC results were then compared with clinical data base of the patients and with Ki67 expression. For each variable, different cut-offs were studied on progression free survival (PFS) and overall survival (OS).

Results: Patients characteristics of the 166 patients were similar to those of the total population of the LYMA trial: age ≤60y (72%), male sex (78%), stage IV (84%), low MIPI score predominant and classic histological type (81%). In univariate analysis, p53 protein expression with a cut-off >30% increased significantly the risk of poor PFS (HR=3.03 [1.66,5.54], p=0.001) as p16 expression with a cut-off > 10% (HR=2.90 [1.56, 5.40], p=0.003). These 2 markers were was also significant for OS (HR=2.49 [1.25, 4.98], p=0.006) for p53 and (HR=3.32 [1.65, 6.68], p=0.002) for p16. In multivariate analysis, p53 and p16 markers remained significant for PFS ajusted or not for the MIPI whereas for OS if we ajust models on MIPI, only P16 was a significant pronostic factor (p=0.0001). If we combined both markers p53 and p16 we could determine 3 groups of patients: low risk group if p16 is <10% and p53<30%, intermediate risk: when one of the 2 risk markers was over the cut-off determined and high risk group when both p53 and p16 markers were over the cut-off determined. In the intermediate risk and high risk groups, the PFS is worse (p=0.009 and ≤0.001 respectively) and in the high risk group, the OS is very poor (HR=6.12 [2.60,14.41] p<0.001). These findings were independant to the Ki67 expression which was significant for both PFS and OS in uni and multivariate analysis in our series. Myc and CD71 markers were not significant on PFS and OS. A correlation was found between the MCL histological subtype classic or blastoid and p53 and p16 markers: blastoid type was more often observed in MCL cases expressing p53 >30% and p16 >10% (p=0.093 and p=0.034 respectively).

Conclusion: In 166 patients treated in the LYMA trial, we found that patients with high p53 expression >30% and high p16 expression >10% had a significant shorter PFS and poor OS. It could be then recommended to incorporate p53 and p16 IHC in routine diagnostic practice. Find new novel therapies for these patients presenting with high risk should be one of the next challenges for MCL.