The incidence of renal cell carcinoma (RCC) is higher in patients with end-stage renal disease (ESRD), with a relatively recent study reporting a standardized incidence ratio (SIR) of 4.03 (95% CI, 3.88-4.19).1 The underlying pathophysiology to explain this increased incidence is likely multifactorial and includes increased oxidative stress, impaired immunity, excess free radicals, and reduced DNA repair.2 Patients undergoing hemodialysis (HD) may develop different variations of RCC, including bilateral, multicentric and non–clear cell tumors.3
There are currently mixed data regarding the clinical outcomes of patients with RCC who are concomitantly on HD, with some reports indicating a better prognosis based on an earlier diagnosis being made.4,5 In contrast, other studies have shown no differences in cancer-specific survival (CSS) when comparing patients with RCC on HD to those not receiving HD.6 Recently, a group led by Hayami et al conducted a retrospective study to further investigate the prognostic role of HD in patients with RCC and published their findings in Seminars in Dialysis.3
The authors performed a retrospective chart review of 388 patients who underwent a radical or partial nephrectomy for RCC at a single hospital in Japan from 2005 to 2013. In it, they found that 66 of 388 (17%) of patients were found to be on HD, of which the median duration of dialysis was 168 months. The most common causes of ESRD on HD were chronic glomerulonephritis (59%), autosomal dominant polycystic kidney disease (20%), diabetic nephropathy (8%), and miscellaneous causes (17%). Forty-one percent of the patients on HD had multicentric tumors with different histologic types.
When comparing the HD and non-HD groups, there were no significant differences between sex and age of patients at surgery. An incidental diagnosis was made more frequently in the non-HD group compared with the HD group (78% vs 65%, P =.014). HD patients had a shorter follow-up compared with non-HD patients (42.5 months vs 54.5 months, P =.021).
With respect to histologic subtype, clear cell was the most common in both groups, however, it was significantly higher in the non-HD vs HD group (89% vs 53%, P <.001). Papillary, clear cell papillary, and acquired cystic disease (ACD)-associated subtypes were all significantly more common in the HD group. Patients in the HD group were more likely to be classified as TNM stage I compared with the non-HD group (86% vs 74%, P =.045). Similarly, patients in the HD group were significantly more likely to have Fuhrman nuclear grade 3 and 4 compared with non-HD patients. In general, patients on HD were more likely to have smaller (median 3 cm vs 4 cm, P =.002), multicentric (41% vs 1.2%, P <.001), and bilateral tumors (14% vs 0.6%, P <.001). More patients in the HD group were found to have a sarcomatoid component (11% vs 3%, P =.002).