Cell lines of hyperdiploid myeloma, are we there yet?

Abstract

We read with great interest the recent paper describing the establishment of stroma-dependent myeloma cell lines, in particular two from hyperdiploid multiple myeloma (H-MM) patients (Li et al, 2007). As there are currently no bona fide H-MM cell lines, the author’s efforts in establishing these two hyperdiploid cell lines are extremely laudable. However, closer inspection of the genetic and molecular characteristics of the two hyperdiploid cell lines suggests a number of atypical features. Firstly, both hyperdiploid cell lines have informative karyotype and have the transcription profile of the proliferation (PR) group, based on a recently reported molecular classification (Zhan et al, 2006). Previous studies have shown that informative cytogenetics can be obtained from only about 30% of all patients (Fonseca et al, 2004) and patients with an informative karyotype have a significantly higher plasma cell labelling index, a marker for proliferation (Rajkumar et al, 1999). In the study by Zhan et al (2006), the PR group comprised tumours of different genetic subtypes that are proliferative and associated with the worst survival. Second, in the LD line, besides the typical trisomies of chromosomes 5 and 9, the karyotype and spectral karyotyping results also showed trisomies of chromosomes 6, 8 and 20. In a previous analysis of 469 informative hyperdiploid karyotypes, these trisomies were seen in 20%, 10% and 12% of cases respectively. Furthermore, in this analysis, these trisomies were associated with more complex karyotypes, suggesting that they occur late during clonal evolution (Chng et al, 2006). In an array comparative genomic hybridization analysis of 103 cases of H-MM, where the majority of cases do not have informative karyotypes, trisomy of chromosomes 6, 8 and 20 were seen in 14%, 7% and 7% respectively (W. J. Chng, unpublished observations). In comparison, trisomies of chromosome 3, 5, 7, 9, 11, 15 and 19, were each seen in excess of 50% of cases when studied by both methods. Third, in the CF cell line, gene expression level of TP53 and RB1 was very low (in the lowest 5–10 percentile of expression in the total therapy II dataset), and may well have deletion of these genes. In fact, in a previous study from the same group, low expression (lowest 10 percentile) of TP53 was associated with 17p13 deletion and a significantly worse prognosis (Xiong et al, 2006). Overall, these features suggest that the H-MM cases from which these cell lines were developed are atypical, clonally advanced and more aggressive subtypes of H-MM. Indeed, recent studies have identified molecular heterogeneity within H-MM and identified subtypes with poorer prognosis (Carrasco et al, 2006; Chng et al, 2007). Therefore, we still have some way to go before developing cell lines from typical cases of hyperdiploid MM. This will be extremely important in advancing biological and therapeutic studies in this most common genetic subtype of MM.