A gene expression inflammatory signature specifically predicts multiple myeloma evolution and patients survival

Abstract

Multiple myeloma (MM) is closely dependent on cross-talk between malignant plasma cells and cellular components of the inflammatory/immunosuppressive bone marrow milieu, which promotes disease progression, drug resistance, neo-angiogenesis, bone destruction and immune-impairment. We investigated the relevance of inflammatory genes in predicting disease evolution and patient survival. A bioinformatics study by Ingenuity Pathway Analysis on gene expression profiling dataset of monoclonal gammopathy of undetermined significance, smoldering and symptomatic-MM, identified inflammatory and cytokine/chemokine pathways as the most progressively affected during disease evolution. We then selected 20 candidate genes involved in B-cell inflammation and we investigated their role in predicting clinical outcome, through univariate and multivariate analyses (log-rank test, logistic regression and Cox-regression model). We defined an 8-genes signature (IL8, IL10, IL17A, CCL3, CCL5, VEGFA, EBI3 and NOS2) identifying each condition (MGUS/smoldering/symptomatic-MM) with 84% accuracy. Moreover, six genes (IFNG, IL2, LTA, CCL2, VEGFA, CCL3) were found independently correlated with patients' survival. Patients whose MM cells expressed high levels of Th1 cytokines (IFNG/LTA/IL2/CCL2) and low levels of CCL3 and VEGFA, experienced the longest survival. On these six genes, we built a prognostic risk score that was validated in three additional independent datasets. In this study, we provide proof-of-concept that inflammation has a critical role in MM patient progression and survival. The inflammatory-gene prognostic signature validated in different datasets clearly indicates novel opportunities for personalized anti-MM treatment.

Figure 1

(a) Comparison analysis to investigate the main affected canonical pathway during evolution from MGUS to MM. Color intensity represents the degree of significance of pathway modulation in each disease condition. (b) Probe fluorescence intensity of the eight genes that resulted significantly associated with MGUS, sMM and MM condition after multinomial logistic regression analysis The range and interval of all axes was automatically determined to evidence differences in fluorescence distribution between different conditions. (c) Percentages of patients correctly classified according to the 8-genes model. (d) ROC curves built to evaluate the accuracy of the 8-genes model.

Figure 2

Algorithm for prognostic score identification.

Figure 3

(a) On the left, heatmap reporting probe fluorescence intensity of six selected genes for each patient evaluated in accordance with its survival, prognostic score and PG. On the right, Kaplan–Meier curves reporting patients' survival according to their PG. Median survival and Hazard ratio values are reported below the curves. (b) Evaluation of correlation between PGs, the two variables forming the international staging system (albumin and B2-microglobulin) and CRP (c-reactive protein).

Figure 4

Validation of the PGs (based on the six variables prognostic score) in three different datasets: (a) The HR group included subjects with a higher risk to experience an event-free survival and OS lower than 24 months. (b, c) The HR group identified patients with the lowest OS.