Validation of Mantle Cell International Prognostic Index (MIPI) in Mantle Cell Lymphoma (MCL) in a Retrospective Single Institution Series

Introduction: The clinical course of MCL is characterized by a continuous pattern of relapse and a poor long term outcome with a median Overall Survival (OS) of four years and a 15% of long term survivors. Recently a new clinical prognostic score (MIPI), including performance status, age, LDH level and leukocyte count has been reported. This score allows a more reliable estimation of individual clinical course. We retrospectively applied the MIPI score to patients with MCL.

Patients and methods: Between 1999 and 2007, 40 patients with MCL diagnosed and treated in a single institution entered into the study. Clinical characteristics were as follows: median age 56 years (range 37–81), 80% male; 82% stage IV; 78% bone marrow involvement and 15% MCL with blastoid variant. First line treatments were: high dose chemoimmunotherapy including Rituximab (R) with autologous stem cell transplantation (R-HDC) in 26 patients and Rituximab-CHOP like chemotherapy (R-CHOP) in 14. Crude Kaplan-Meier OS and progression-free survival (PFS) curves were estimated both overall and stratified by MIPI and International Prognostic Index (IPI) score. Differences between curves were tested using the 2-tailed log-rank test. In order to quantify the predictive discrimination of MIPI and IPI scores, univariate logistic models (with death and progression event as binary outcomes) were fitted and the area under the receiver operating characteristic (ROC) curves (c index) was estimated.

Results: According to MIPI score 17 patients (43%) were at low risk (LR, score 0–3), 13 patients (32%) at intermediate risk (IR, score 4–5) and 10 patients (25%) at high risk (HR, score >5). According to IPI score 14 patients (35%) were at low risk (LR), 16 patients (40%) at low-intermediate risk (LIR) and 10 patients (25%) at intermediate-high and high risk (IH-HR). At the end of the treatment, 30 patients achieved a CR, five a PR and five did not respond. Relapses occurred in 17 patients and seven of them died of lymphoma. With a median follow-up (FU) of 29 months, OS was 85% (95% CI: 66%–93%); with a median FU of 21 months, PFS was 70% (95% CI: 51%–83%). Twenty-nine months OS rates for MIPI score were: LR 100%, IR 81%, HR 66% respectively (p=.07) and for IPI score were: LR 92%, LIR 94%, IH-HR 65% respectively (p=.09). Twenty-one months PFS rates for MIPI score were: LR 92%, IR 59%, HR 45% respectively (p=.006) and for IPI score were: LR 73%, LIR 87%, IH-HR 44% respectively (p=.09). MIPI score was more predictive than IPI score for the death event and for the progression event: the c index was 74% and 73% for MIPI compared to 72% and 69% for IPI respectively. In a subgroup analysis performed on 26 R-HDC patients, OS and PFS rates stratified for MIPI were: for OS, LR 100% vs IR 80% vs HR 69% (p=.4) and for PFS, LR 91% vs IR 80% vs HR 57% (p=.04) respectively.

Discussion: in our retrospective series of patients, MIPI prognostic score discriminates among patients with different PFS. Relapses remain the most important issue for all patients affected by MCL, namely in HR group according to MIPI. New therapeutic strategies are warranted to improve the prognosis of MCL.