Risk of secondary malignancy following radiation therapy for prostate cancer

Abstract

We investigated whether prostate cancer patients treated with external beam radiation therapy (EBRT) have a higher cumulative incidence of secondary cancer compared with patients treated with radical prostatectomy (RP). We used state-wide linked data from South Australia to follow men with prostate cancer diagnosed from 2002 to 2019. The cumulative incidence of overall and site-specific secondary cancers between 5 and 15 years after treatment was estimated. Fine-Gray competing risk analyses were performed with additional sensitivity analyses to test different scenarios. A total of 7625 patients were included (54% underwent RP and 46% EBRT). Characteristics of the two groups differed significantly, with the EBRT group being older (71 vs. 64 years), having higher comorbidity burden and being more likely to die during follow-up than the RP group. Fifteen-year cumulative incidence for all secondary cancers was 27.4% and 22.3% in EBRT and RP groups, respectively. In the adjusted models, patients in the EBRT group had a significantly higher risk of genitourinary (adjusted subhazard ratio (aSHR), 2.29; 95%CI 1.16-4.51) and lung (aSHR, 1.93; 95%CI 1.05-3.56) cancers compared with patients in the RP group. However, there was no statistically significant difference between the two groups for risk of any secondary cancer, gastro-intestinal, skin or haematologic cancers. No statistically significant differences in overall risk of secondary cancer were observed in any of the sensitivity analyses and patterns for risk at specific cancer sites were relatively consistent across different age restriction and latency/time-lag scenarios. In conclusion, the increased risk of genitourinary and lung cancers among men undergoing EBRT may relate partly to treatment effects and partly to unmeasured residual confounding.

Figure 1

Participant selection procedures. RP, radical proctectomy; EBRT, external beam radiation therapy; brachy, brachytherapy (high dose rate and low dose rate combined). *Men in both SA-PCCOC and SACR datasets and have prostate cancer (C61) as their primary cancer diagnosis (diagnosed from 2002 to 2019). § only men with metastatic prostate cancer at diagnosis excluded and this doesn’t account men with nodal involvement. ‡ Includes 782 men who had received EBRT after RP. Further sensitivity analyses were done by including these men as a different treatment category in the model. ¥ Men who received brachytherapy were not included in the main analyses, they were kept for further sensitivity analyses. 72 men (43 who had RP and 29 who had EBRT) with ICD-10 codes C77, C78, C79 & C80 as their second cancer were not classified as having a second cancer since these cancers are likely to be metastases from their prostate cancer. They were included in the analyses as ‘no second cancer’.

Figure 2

Cumulative incidence of overall and site specific second cancers (unadjusted).

Figure 3

Comparisons of adjusted cumulative incidences of secondary cancers after EBRT vs. RP. RP, radical proctectomy; EBRT, external beam radiation therapy.