A clinical prediction model for outcome and therapy delivery in transplant-ineligible patients with myeloma (UK Myeloma Research Alliance Risk Profile): a development and validation study. Issue 3 (March 2019)

• Record Type: Journal Article
• Authors: Cook, Gordon; Royle, Kara-Louise; Pawlyn, Charlotte; Hockaday, Anna; Shah, Vallari; Kaiser, Martin F; Brown, Sarah R; Gregory, Walter M; Child, J Anthony; Davies, Faith E; Morgan, Gareth J; Cairns, David A; Jackson, Graham H
• Is Part Of: Lancet. Volume 6:Issue 3(2019)
• Page Start: e154
• Page End: e166
• Publication Date: 2019-03
• Subjects: Hematology -- Periodicals; Blood -- Diseases -- Periodicals; 616.15005
• DOI: 10.1016/S2352-3026(18)30220-5
• Languages: English
• ISSNs: 2352-3026

Abstract

Summary: Background: Tolerability of treatments for multiple myeloma can depend on the characteristics of the patient being treated. We aimed to develop and validate a risk profile, using routinely collected data, that could predict overall survival in patients with multiple myeloma who were ineligible for stem-cell transplantation. Methods: We used patient data from two randomised controlled trials done in patients with newly diagnosed multiple myeloma who were ineligible for stem-cell transplantation (the NCRI Myeloma XI study [NCRI-XI, n=1852] and the MRC Myeloma IX study [MRC-IX, n=520]), to develop the UK Myeloma Research Alliance Risk Profile (MRP) for overall survival. We used multivariable Cox regression with a least absolute shrinkage and selection operator penalty term. Multiple imputation by chained equations was used to account for missing data in the development and internal validation of the model. The MRP was internally validated in NCRI-XI and externally validated in MRC-IX. The D-statistic was estimated in the developed model and used to internally and externally validate the model according to prespecified criteria. Findings: The MRP included WHO performance status, International Staging System, age, and C-reactive protein concentration as prognostic variables. The MRP was prognostic of overall survival and was successfully internally validated in NCRI-XI and externally validated in MRC-IX (D-statistic NCRI-XI: 0·840 [95% CI 0·718–0·963] and MRC-IX: 0·654 [0·497–0·811]). The MRP groups defining low-risk, medium-risk, and high-risk patients were associated with progression-free survival and early mortality. A decrease in the percentage of protocol dose delivered and quality of life at baseline were associated with increased risk. The MRP groups remained prognostic in patients exposed to different therapeutic combinations and in patients with genetic high-risk disease defined according to both the UK and International Myeloma Working Group definitions. Interpretation: We have developed and externally validated a risk profile for overall survival containing widely available clinical parameters. This risk profile could aid decision making in patients with multiple myeloma ineligible for stem-cell transplantation, but further external validation is required. Funding: Medical Research Council, Novartis, Schering Health Care, Chugai, Pharmion, Celgene, Ortho Biotech, Cancer Research UK, Celgene, Merck Sharp & Dohme, and Amgen.