LMO2 Protein Expression Predicts Survival in Patients with Diffuse Large B-Cell Lymphoma Treated with Immunochemotherapy (RCHOP): A Multicenter Validation Study

Background: Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous entity which necessitates the characterization of prognostic markers to better define risk groups. Previously, we showed that the expression of LMO2 mRNA was a strong predictor of superior outcome in patients with DLBCL in a multivariate model of six genes (Lossos et al, NEJM 2004). Subsequently, using a novel monoclonal anti-LMO2 antibody, we showed that LMO2 protein expression predicted outcome in DLBCL patients treated with anthracyline-based chemotherapy with and without rituximab (Natkunam et al, JCO 2008). For validation of our prior findings we analyzed an independent cohort of DLBCL patients treated with RCHOP that had not been included in our previous study.

Methods: 106 patients with de novo DLBCL treated with RCHOP and followed for clinical outcome were included. Immunohistochemistry (IHC) for LMO2 was performed on tissue microarrays containing cores of biopsies obtained at initial diagnosis. Specimens expressing LMO2 in more than 30% of neoplastic cells were defined as positive. LMO2 expression was correlated with the international prognostic index (IPI), overall survival (OS) and progression free survival (PFS).

Results: The median age of the study population was 58.5 years (24–85). The distribution of patients according to the IPI was 47% low risk (0–1 factors), 24% low intermediate (2 factors), 14% high intermediate (3 factors) and 15% high risk disease (>=4 factors). 70 patients (66%) were LMO2 positive (compared to 55% in our original study). 18 patients (17%) have died and the median survival has not been reached. At a median follow-up of 32 months the 2 year OS for LMO2 positive versus negative cases was 91% (95% confidence interval [CI] 0.84, 0.98) versus 70.8% (95% CI 0.558, 0.875) respectively, p=0.04. No significant difference was observed in the 2 year PFS between LMO2 positive versus negative cases. A multivariate Cox regression analysis that included IPI scores and LMO2 expression as dependent variables for OS demonstrated a trend for LMO2 (p=0.07) and significance for the IPI (p=0.046) as independent predictors for OS. For PFS only the IPI remained statistically significant (p=0.03).

Conclusions: This study validates the prognostic impact of LMO2 protein expression in the RCHOP treatment era in an independent cohort of DLBCL patients. Despite the higher proportion of LMO2 positive cases in the current cohort, a relatively short follow-up and few adverse events, LMO2 protein expression was associated with improved OS. Continued clinical follow-up of this cohort is ongoing to fully assess the impact on outcome. Assessment of LMO2 protein expression by routine IHC in biopsy samples of newly diagnosed DLBCL may help identify patients with different outcomes. Clinical trials will be necessary to assess the optimal therapy for patients based on LMO2 status.