Gene Expression Profiling (GEP) of Myeloma (MM) Cells To Predict Attainment (near) Complete Response to Primary Therapy with Thalidomide-Dexamethasone (Thali-Dex) for Newly Diagnosed MM.

In recent years, treatment paradigm of targeting the “soil” bone marrow (BM) microenvironment as a means of interfering with the growth of the MM “seed” has provided the rationale for investigational clinical trials of novel agents combined with old drugs in an attempt to maximize tumour response. Thalidomide, which represents an effective treatment strategy for relapsed/refractory MM, actually represents a standard of care also for newly diagnosed MM patients. We have recently demonstrated that thali-dex combination as front-line therapy in preparation for autologous transplantation is superior to VAD in terms of increased rate of response (≥ partial response: 76%) and magnitude of tumour reduction. In particular, the probability to attain at least a very good partial response (VGPR) was 19%, including 13% of patients in complete response (CR) or near CR (nCR). In the present study we adopted a GEP strategy in an attempt to identify a signature able to predict the probability to attain ≥ nCR to combined thali-dex as upfront therapy for patients with newly diagnosed MM. CD138+ samples obtained at diagnosis from 32 patients enrolled in the “Bologna 2002” clinical trial were used throughout the study; all patients were evaluable for response to thali-dex in preparation for autologous transplantation. GEP was performed using the Affymetrix HG133 Plus microarray platform. The Affymetrix output (CEL files) was imported into Genespring 7.3 (Agilent technologies) microarray analysis software, where data files were normalized across chips using GCRMA and to the 50^th percentile, followed by per gene normalization to median. Genes differently expressed in subgroups of patients were selected by an ANOVA analysis. Criteria of response were those established by Bladè et al, with the addition of a VGPR and nCR categories. Overall, six of the 32 patients (19%) obtained at least a nCR to thali-dex, whereas the remaining 26 patients either achieved a partial response or did not respond. We identified a gene signature of 162 genes, able to significantly distinguish patients with ≥ nCR from the others (p=0.05). We then adopted a Nearest-Neighbours (NN) classifier (using 3 first neighbours), with a “Leave-one-out Cross Validation” procedure, to identify a list of ten genes able to predict the ≥ nCR in our series of patients (see table). Of interest, the gene list encompasses CCND2, one of the most important cell cycle regulator known to be involved in tumour progression in MM patients and the anti-apoptotic gene CFLAR, both down-regulated in nCR patients. These results could be the first step to adopt microfluidic cards, in an attempt to select at diagnosis patients who will respond very favourably to a particular treatment strategy. Supported by Università di Bologna, Progetti di Ricerca ex-60% (M.C.); Ministero dell’Università e Ricerca Scientifica (MIUR), progetto FIRB, RBAU012E9A_001 (M.C.); and Fondazione Carisbo.