Risk stratification by SKY92+ISS outperforms ifish markers t(4;14) and del(17p) in multiple myeloma

Van Vliet M., Kuiper R., Ubels J., Dumee B., Bosman L., Van Beers E., De Best L., Waage A., Zweegman S., Morgan G., Goldschmidt H.

Research output: Contribution to journalArticleAcademicpeer-review


Background: Multiple Myeloma (MM) is a disease with a heterogeneous genetic makeup across patients, and differing survival. Various prognostic markers have been identified, which may be employed in risk stratified treatment paradigms. A large comparison has unveiled that the combination of SKY92 + ISS is the best prognostic marker for identifying high risk and low risk patients. However, a recurring question is how this marker model relates to the currently established iFISH abnormalities such as t(4;14), and del(17), which have earlier been associated with poor overall survival (OS). Here, we demonstrate that iFISH is outperformed by the SKY92 + ISS marker combination for risk stratification. Aims: To show the prognostic relevance of iFISH markers in perspective of the SKY92 + ISS risk stratification model for prognosis of MM patients. Methods: Four datasets were gathered, HOVON-65/GMMG-HD4 (n=329), HOVON-87/NMSG-18 (n=143), MRC-IX (n=246), and a Czech cohort (EMTAB- 1038, n=66). Pooled, this provided n=559 MM patients for which GEP data, iFISH annotation, Overall Survival (OS), and ISS were available. The pooled dataset was split into three risk strata as previously proposed (Kuiper et al., 2015): SKY92 standard risk (SR) and ISS1 (low risk), SKY92 Standard Risk and ISS2/3 (intermediate risk), SKY92 high risk (HR). Subsequently, each of these three risk strata were further refined into two groups: those that had either iFISH t(4;14) and/or del(17p) characteristics and those without. The Cox proportional Hazards model was applied to those two groups in each of the three strata separately, to calculate hazard ratios and associated p-values. Since groups would become too small, splitting the groups into all possible FISH subgroups was not performed, i.e. positive for both t(4;14) and del(17p), and those positive for one and negative for the other. Results: The four datasets were pooled and then split into three risk strata, providing n=140 low risk (25.0%, SKY92 SR and ISS1), n=289 intermediate risk (51.7%, SKY92 SR and ISS2/3), and n=130 high risk (23.3%, SKY92 HR) MM patients, see Figure 1A. Out of the 140 low risk patients 28 (20%) had a t(4;14) and/or del17, whose survival was equivalent to those without iFISH aberrations (HR=2.4, p=0.054, see Figure 1B). Moreover, their survival was much longer than those in the SKY92 HR group, signifying that they would be at risk for overtreatment when iFISH would be used for risk stratification. Similarly, 55 (19%) out of 289 intermediate risk patients were found to harbor either iFISH t(4;14) or del(17p), but with no impact on their survival (hazard ratio=1.3, p=0.17, see Figure 1C). Finally, the group of SKY92 high risk patients were clearly overrepresented for t(4;14) and del(17p), with 65 out of 130 cases (50%). However, those 50% of patients that did not bear a t(4;14) and/or del(17p) aberration (i.e. non high risk by iFISH) had the same poor OS as those with them (hazard ratio=1.0, p=0.82, see Figure 1D). As a consequence, applying the current prognostic models using iFISH, those 50% HR patients not detected by iFISH would be at risk of undertreatment. (Figure presented) Summary/Conclusions: Risk stratification based on SKY92 and ISS is a powerful tool to identify both high risk MM patients and low risk MM patients. Patients identified to be iFISH t(4;14) and/or del(17p) are only high risk when they co-occur with SKY92 HR. Moreover, when they are SKY92 SR, their prognosis is better defined by their ISS stage.
Original languageEnglish
Pages (from-to)86
Number of pages1
Issue numberSupplement 1
Publication statusPublished - 2016

Cite this