Sequential Improvements in the Outcome of Autologous Stem Cell Transplantation for Multiple Myeloma over Time

Introduction

Recently a number of clinical trials have reported on a comparison between autologous stem cell transplantation (ASCT) for multiple myeloma (MM) versus conventional dose therapies. These studies have shown that even in the era of "novel agents", early ASCT is associated with significant improvements in progression free survival (PFS). Further a recent population based study has suggested that having received ASCT is one of the most important prognostic factors governing long-term outcomes for patients with MM. These observations prompted us to examine the long-term outcomes achievable in a series of ASCT patients where novel agents were incorporated into patient management and how these outcomes have changed over time.

Methods

To investigate long-term survival, we followed patients with newly diagnosed MM who received an ASCT at the University of Arkansas for Medical Sciences (UAMS) Myeloma Institute, from 1989 to 2018. Patients with smoldering myeloma, and those who did not receive an ASCT were excluded from the study. In addition to ASCT, patients who were enrolled on a clinical trial received experimental chemotherapeutic regimens determined by the clinical trials protocol being implemented at the time. All other patients received standard therapies appropriate at the time of diagnosis. Based on the year of their first ASCT, patients were separated into time periods that roughly corresponded to the dates of different protocols when novel agents were being tested at UAMS (1997 or earlier [reference group], 1998-2003, 2004-2008, 2009-2013, and 2014 or later). Kaplan-Meier curves with log-rank tests, and Cox proportional hazards models adjusting for age, sex and race/ethnicity were used to evaluate survival during the different time periods. Because the demographics of the UAMS patients have changed over time (with older and non-White patients becoming more common in more recent years), we also explored approaches to correct for temporal changes in both MM patients at UAMS and within the general US population, including a Fine-Gray competing risk analysis, and a relative survival analysis using SEER life tables to account for population differences in mortality by age, year, sex, and race/ethnicity.

Results

A total of 4,326 MM patients met the eligibility criteria for this study, with 2,465 patients enrolled on a clinical trial. The median follow-up time was 10.4 years, with median overall survival of 6.9 years. In the adjusted Cox proportional hazards model, patients who had their first ASCT in 2014 or later had significantly improved survival compared to those who had their first ASCT in 1997 or earlier (HR, 0.32; 95% CI, 0.25-0.41). Among patients who had a valid cause of death (n=2961), Fine-Gray competing risk analyses confirmed this finding (HR, 0.10; 95%CI, 0.05-0.19). After using SEER life tables to correct for differences in normal mortality, we find less relative excess risk (RER) for MM death with each successive time period compared to the "1997 or earlier" reference group: 1998-2003 (RER, 0.67; 95%CI, 0.57-0.78), 2004-2008 (RER, 0.56; 95% CI, 0.48-0.66), 2009-2013 (RER, 0.56; 95% CI, 0.47-0.66), 2014 or later (RER, 0.31; 95% CI, 0.22-0.42). These results suggest that as chemotherapeutics and patient management strategies have changed over time, the MM survival has steadily and consistently improved at UAMS.

Conclusions

Our results suggest that after correcting for population differences in mortality by age, sex, year, and race/ethnicity, MM survival in UAMS patients has improved over time. These trends could be due to a number of factors, including improvements in MM therapies, risk stratification models, management of older patients, and increased understanding of MM biology and meaningful prognostic factors. Additional analyses to examine these traits, and Cure-Rate survival models to determine the likelihood of achieving long-term disease-free survivorship, are currently ongoing.

View largeDownload slide

View largeDownload slide

Close modal