In Myeloma, the Prognostic Impact of Hyperdiploidy Is Mainly Related to the Gain of Chromosome 5

Multiple myeloma is characterized by multiple chromosomal abnormalities, and virtually 100% of the patients are cytogenetically abnormal. Among the few recurrent changes described, hyperdiploidy is one of the most frequent. In cytogenetic series, hyperdiploidy is reported in 50 to 60% of the cases. It has been associated with prolonged survival in some series, whereas other series concluded to a neutral prognostic impact. These conflicting data might be related to the bias of cytogenetic analyses, which are abnormal in only 10 to 30% of the cases. In order to circumvent the pitfall of the low proliferation of plasma cells, we conducted a study based on DNA microarray. A series of 192 patients presenting multiple myeloma at diagnosis and treated homogeneously in tandem transplantation IFM trials were analyzed using the 500K Affymetrix SNP-arrays. In agreement with cytogenetic series, hyperdiploidy (> 47 chromosomes) was observed in 99 patients (53%). As previously reported, the most frequently gained chromosomes were chromosomes 3, 5, 7, 9, 11, 15, 19, 21, and X in female patients. We did perform a prognostic analysis. Hyperdiploid patients displayed a clear survival advantage as compared with non-hyperdiploid patients: p=0.006 and p=0.0008 for event free survival (EFS) and overall survival (OS), respectively. When adding all the chromosomal abnormalities observed on the SNP-arrays, 3 chromosomal lesions were independently associated with prognosis : gain of 1q, loss of 12p, and gain of chromosome 5. If serum b2- microglobulin (sb2m) level and t(4;14) were entered in the model, only sb2m, del(12p) and gain of chromosome 5, kept an independent prognostic value. These data suggest that most of the prognostic impact of hyperdiploidy could be linked to the gain of chromosome 5 (p=0.0001 and p<0.0001, for EFS and OS, respectively). In order to test this hypothesis, we did compare the survival of hyperdiploid patients with gain of chromosome 5 to that of hyperdiploid patients lacking chromosome 5 gain. As expected, patients with chromosome 5 gain displayed a significantly longer EFS and OS than those lacking this gain (p=0.005 and p=0.001 for EFS and OS, respectively). In conclusion, this study suggests that most of the prognostic value of hyperdiploidy is actually borne by chromosome 5.