Plasma Cell Myeloma: Can Histologic Grade Predict the Molecular Subsets?
OF Ikpatt, JD Shaughnessy, BP Nair, ZN Singh. University of Arkansas for Medical Sciences, Little Rock, AR
Background: Many studies emphasize the role of morphology in multiple myeloma (MM). For e.g. plasmablastic MM is correlated with shorter survival and the small lymphocyte-like MM reportedly better. Lack of definitive cytological and numerical criteria in the different morphological grading systems coupled with morphological heterogeneity in MM limits reproducibility of such grading. With the development of a prognostically predictive molecular classification based on gene-expression signatures, the relevance of morphological subtyping in MM is presently unclear. Correlation of the phenotypic grading with molecular subgroups has not yet been performed. We evaluated two histological grading systems to see if they can predict the molecular subsets of MM.
Design: BM biopsies from fifty eight untreated MM patients with previously determined molecular subsets generated from unsupervised hierarchic clustering of gene expression profiles (GEP) (Affymetrix U133 1559 gene microarray platform), were independently evaluated by an experienced pathologist blinded to the GEP data. Grading done according to the Bartl and Greipp schema was correlated with molecular subgroups.
Results: The distribution of Bartl and Greipp histological grades among the 7 molecular subsets of patients is summarized in the table.
|MORPHOLOGICAL||MOLECULAR SUBSETS / RISK CATEGORY|
|GREIPP||CD: 6 low||HY: 9 low||LB: 11 low||MY: 4 low||MF: 11 high||MS: 11 high||PR: 6 high|