Distribution and association of cancer with mortality in end-stage renal disease patients receiving dialysis

Abstract

Abstract Background and aims Cancer in end-stage renal disease (ESRD) patients is an important comorbidity to be taken into consideration while planning for renal replacement therapy (RRT) options due to its associated increased mortality. This study aims to investigate the natural history and association of cancer with all-cause mortality in an ESRD population receiving dialysis. Method The study was conducted on 1271 ESRD patients receiving dialysis between January 2012 and December 2017. A comparative analysis was carried out between 119 patients with and 1152 without cancer history at entry into this study (baseline). A 1:2 (119 cancer: 238 no cancer) propensity score matched sample of 357 patients was also used for analysis. Cox-regression analysis was used to study the strength of the association between cancer and all-cause mortality. Kaplan–Meier (KM) analysis was used to demonstrate the difference in cumulative survival between the groups. A competing risk analysis was also carried out to calculate the probability of competing events (death, transplant and incident cancer). Results At baseline, 10.1% of the cohort had a history of cancer (current and past) with the annual incident rate being 1.3%. Urological cancers were the leading site of cancer. The median age of our cohort was 63 years with a predominance of males (63%) and Caucasians (79%). The majority (69%) of the cohort were receiving haemodialysis. 47% had a history of diabetes with 88% being hypertensive. During a median follow-up of 28 months, the proportion of deaths observed was similar between the groups in the matched sample (cancer 49.6 versus no-cancer 52.1%, p value 0.77). In a univariable Cox-regression model, there was no significant association between cancer and all-cause mortality (HR 1.28; 95% CI 0.97–1.67; p = 0.07). The KM estimates showed similar observations in the cumulative survival between the groups (matched sample log-rank, p value 0.85). In competing risk analysis, the cumulative probability of death at 5 years was non-significantly higher in the cancer group (cancer group 64% vs no cancer group 51%, p value 0.16). Conclusions In our real-world multi-morbid dialysis cohort of 119 cancer patients, baseline cancer history did not prove to be an independent risk factor for all-cause mortality in the first 5 years of follow-up, suggesting the need for a case-by-case approach in provision of RRT options, including transplantation.