The Low-Bone (LB) Subtype of Multiple Myeloma Exhibits an Inferior Outcome in African Americans

Background: African Americans (AA) have an earlier onset of multiple myeloma (MM) and twice the mortality rate compared to Caucasian Americans (CA) in United States. We have defined seven molecular subtypes referred to as CD-1, CD-2, LB, HY, MS, MF, and PR in newly diagnosed MM (NDMM). Four of these (CD-1, CD-2, LB, and HY) have shown a good prognosis, while other three subtypes (MF, MS, and PR) exhibit a poor prognosis. In this study, we seek to investigate MM subtypes that are associated with inferior overall survival (OS) in AA patients.

Materials and methods: A total of 1,706 NDMM with Affymetrix microarray data were included in this study. This cohort of MM patients contains 1,458 Caucasian Americans (CA) (85.5%) and 208 African Americans (AA) (12.2%). Patients were treated on Total Therapy Protocols (TTPs) at the University of Arkansas for Medical Sciences (UAMS). The median follow-up is 7.71 years and ranged 26.61 years. Each patient sample was classified into one of seven subtypes or an Unknown subtype which was previously called the Myeloid (MY) subtype as well as high-risk or low-risk group defined by GEP70. Kaplan-Meier curves and log-rank test were performed for survival analysis.

Results: No significant difference in OS was observed between AA and CA MM patients. Significant survival differences were observed in both AA and CA MM patients according to GEP70, but there was no difference between AA and CA MM patients in either high-risk group or low-risk group. The 7 molecular subgroups were identified based on 559 NDMMs with a median follow-up 3 years published in Blood 2006. Four of MM subtypes (CD-1, CD-2, LB, and HY) in 1,458 CA MM patients showed better outcomes than the other three subtypes (MF, MS, and PR) with a follow-up for 15 years for both OS and event-free survival (EFS), while the survival difference disappeared after 15-year follow-up. However, subtype survival of the 208 AA MM patients was different from CA. We found that the low-risk LB group was a high-risk subtype in AA NDMMs with a similar OS as the PR subtype, while the high-risk subtypes of MF and MS showed a better OS similar to the CD-1, CD-2, HY, and MY subtypes. We did not find a significant difference for EFS in AA myeloma subtypes. To further confirm the survival difference of LB subtype between AA and CA NDMMs, we performed Kaplan-Meier curves and log-rank tests. The LB subtype in AA patients showed worse OS and EFS compared to CA patients. Because CST6 is the most significant gene highly expressed in the LB subtype in both AA and CA MM patients, we further compared survival among four groups (AA-LB-CST6^high, AA-LB-CST6^low, CA-LB-CST6^high, and CA-LB-CST6^low) of MM patients. Intriguingly, MM patients characterized with CA-LB-CST6^high showed the best OS, while those MM patients with AA-LB-CST6^high showed the worst OS.

Conclusions: We confirmed MM molecular subtypes and risk groups using a much larger sample size and a much longer follow-up. However, clearly different outcomes of MM subtypes were identified between AA and CA MM patients. The LB subtype showed a worse outcome in AA than in CA MM patients. We also found that the LB subtype of AA MM patients with high-CST6 expression had the worst OS, suggesting that CST6 may be a biomarker negatively influencing AA MM patient survival.