Interphase FISH-Defined Amplification of Chromosome 1q21 (AMP1q21) Identify High-Risk Subsets among Patients with Multiple Myeloma (MM) Lacking Metaphase Cytogenetic Abnormalities (CA).

Abstract

Background: The presence of CA (esp. hypodiploidy and del13) defines a “malignant” myeloma entity with a very poor prognosis. Gene expression profiling (GEP) studies in 351 patients receiving Total Therapy 2 (TT2) identified overexpression of CKS1B mapping to 1q21 as an adverse feature for EFS and OS (Shaughnessy et al., ASH 2004). The aim of this study was to determine whether more readily applicable FISH detection of amp1q21, available in 476 patients, had independent and potentially equally important prognostic implications as CA and GEP-based CKS1B expression levels.Patients and Methods: The details of the development of the amp1q21 index are contained in (Shaughnessy et al, submitted, Nature Medicine). The amp1q21 index is a weighted average of the proportions of cells with 3 and >3 copies of the CKS1B gene, ranging from 0 through 100. The risk of event and death increases as the amp1q21 index increases; for simplicity in reporting this relationship, patients tumor cells exhibited diploid, triploid, and > triploid using cut points selected to maximize the overall difference in OS between these three groups.Results: Of 668 patients enrolled in TT2, 476 had FISH data, with no amp1q21 in 58%, trisomy in 31% and > trisomy in 11%. With median follow-up 3 years, 187 had an event and 131 died. Median EFS was 63 mo with an estimated 59% OS at 72 mo. Three-year estimates of EFS and OS progressively decreased with 1q21 amplification level: from 80%/89% with no amp1q21 to 48%/69% with trisomy to 43%/54% with at > trisomy (p<0.001/p<0.001). On multivariate analysis that included standard prognostic factors, ISS stage and CA, trisomy and > trisomy of 1q21 retained independent adverse implications for both EFS (HR 2.08, p<0.001) and OS (HR 2.15; p<0.001). In the no CA group of 327 patients, 3-yr EFS/OS progressively declined from 82%/92% in the 213 patients with no amp1q21 to 63%/79% in 87 with trisomy, to 40%/58% in 27 with > trisomyy (p<0.001/p<0.001). Among the 144 patients (31%) with CA, 3-yr EFS/OS were 72%/77% among 61 with no amplification and similar at 33%/51% in the remaining 83 with >= trisomy (p<0.001/p=0.003). Using an 18 month landmark analysis, CR significantly improved subsequent 3-yr EFS/OS among the 130 with amp1q21 (at least trisomy) from 42%/51% to 65%/76% (p=0.01/p<0.001) - similar to those without amp1q21 (in whom CR was not yet important).Conclusion: A gene dose-dependent adverse effect on both EFS and OS was observed for amp1q21. The 2/3 of patients without CA (and historically exhibiting much better prognosis than those with CA) could be separated into 3 distinct subgroups based on level of amp1q21. Conversely, among patients with CA, absence of amp1q21 implied EFS and OS levels similar to those without CA but with trisomy. Finally, CR was critical to overcoming the negative consequences of amp1q21. Collectively, our data strongly support the routine application of FISH for amp1q21 analysis in clinical trials of MM.