Renal impairment at presentation in multiple myeloma continues to be associated with poor survival

We were interested to read the recent publication by Murphy et al (2014), which showed that, despite the recent well-recognized improvements in overall survival (OS) in multiple myeloma, patients in their centre presenting with dialysis-dependent renal failure continue to have poor outlook, with no improvement in OS between 1995 and 2012. These findings were of particular interest to us as we recently collected data from the Greater Glasgow and Clyde area of Scotland, to assess survival trends in patients with multiple myeloma and renal impairment. We identified 105 patients with a recorded diagnosis of myeloma and a creatinine of >130 μmol/l at presentation to our renal service. The patients were split into three time cohorts for analysis: 1990–99 (n = 27), 2000–04 (n = 29) and 2005–10 (n = 49). These cohorts were chosen to reflect the introduction of Thalidomide in 2000 and Bortezomib in 2005 into our clinical practice. Kaplan–Meier survival curves were generated and the log rank test used to analyse the data. There was no significant difference with regards to sex and mean presenting serum creatinine in the three cohorts. There was, however a trend towards a higher mean age (65·5, 70·0 and 70·0 years, P = 0·13) and percentage requiring renal replacement therapy (RRT) at presentation (5%, 13% and 15%; P = 0·1) in the later two cohorts. We were also surprised to find that there was no improvement in overall patient survival in our centre. We found a median actuarial survival of 47·4 months for patients treated in 1990–99, 16 months for 2000–04 and 17·7 months for 2005–10 (log rank P = 0·24) (Fig 1). We found that patients who required RRT at presentation had a very significantly worse prognosis than those who did not require RRT: median survival of 10·5 months compared to 33·6 months (P ≤ 0·001). We also found no difference in OS in the three patient cohorts when those requiring RRT were excluded: median survival of 35·1 months in 1990–99, 18·7 months in 2000–04 and 33·6 months in 2005–10 (P = 0·72) (Fig 2). These results remain difficult to explain. Some of the patients in the earliest cohort showed unexpectedly long survival while the later two cohorts show survival duration comparable with other published studies assessing the impact of renal impairment on myeloma outcome (Eleutherakis-Papaiakovou et al, 2007). We wonder if there was selection bias in the earliest cohort based on performance status rather than age. When comparing the 2000–04 and 2005–10 cohorts, there was a trend towards improvement in OS when patients requiring RRT were excluded. Bortezomib only became available in our clinical practice in 2005 and therefore it may be that our analysis is too early to capture the recognized good outcome of Bortezomib in this specific patient group. Future analysis may hopefully confirm this. In conclusion, our centre's experience shows that patients with multiple myeloma and renal impairment at presentation continue to have poor outcome. In support of Murphy et al (2014), we too are yet to see significant improvements in OS for myeloma patients requiring RRT despite the introduction of novel therapies. AL, CG and RS were involved in study design. AL collected the data. AL, CG and RS analysed the data. AL wrote the first draft of the letter and CG and RS contributed to the finalization of the letter and approved its submission. There are no conflicts of interest to report.