Charlson score and competing mortality

Abstract

We read with great interest the study by Daskivich et al determining the benefit of aggressive treatment in a huge sample of elderly men with early-stage prostate cancer.1 The authors concluded that men with a Charlson score of ≥3 did not benefit from aggressive prostate cancer treatment. Limiting the claim-based assignment of the Charlson score to a 12-month period prior to the prostate cancer diagnosis1 most likely caught preferably severe conditions (indicated by a percentage of 76% of patients having a Charlson score of 0,1 which is in contrast with lower rates even in samples from patients who underwent radical prostatectomy, who are known to be preselected by good health status2, 3). This special feature could partly explain the strikingly high 10-year competing mortality rates noted in this study (71% for a Charlson score of 2 and 84% for a Charlson score ≥3).1 These data should be interpreted with caution. Populations with a more complete documentation of less severe Charlson score conditions may have lower competing mortality rates in the individual risk classes, particularly when the involved patients have been filtered by health status assessment during initial treatment decision-making. In our sample of patients selected for radical prostatectomy (947 men aged ≥66 years who were treated between 1992-2005), we calculated the following 10-year competing mortality rates: 21% for a Charlson score of 2 (201 patients; 95% confidence interval, 13%-30%) and 40% for a Charlson score ≥3 (62 patients; 95% confidence interval, 24%-55%). When considering these relatively low competing mortality rates, compared with the figures observed by Daskivich et al,1 it appears conceivable that appropriately selected elderly patients may benefit from aggressive treatment, even those with a Charlson score of ≥3. No specific funding was disclosed. Dr. Froehner has received meeting participation support from Janssen and Pfizer for work performed outside of the current study. Dr. Wirth has acted as a paid lecturer and consultant for Bayer and Janssen-Cilag; a paid consultant for Merck, Dendreon, Farco-Pharma, Ferring, Ipsen, and Teva; and as a paid lecturer for Sanofi-Aventis, Apogepha, Orion, and Pfizer for work performed outside of the current study. Michael Froehner, MD Department of Urology University Hospital “Carl Gustav Carus” Technical University of Dresden Dresden, Germany Rainer Koch, PhD Department of Medical Statistics and Biometry University Hospital “Carl Gustav Carus” Technical University of Dresden Dresden, Germany Manfred P. Wirth, MD Department of Urology University Hospital “Carl Gustav Carus” Technical University of Dresden Dresden, Germany