Prostate cancer incidence and survival in relation to prostate cancer as second cancer in relatives

Abstract

Objectives: To investigate if the risk of prostate cancer (PC) differs based on the order of primary PC diagnosed in first-degree relatives (FDRs) given possibly different risk factors for PC as first primary cancer (PCa-1) and second primary cancer (PCa-2).

Subjects and methods: In this Swedish nationwide cohort, PC diagnosis was followed for among 149,985 men with one FDR affected by PCa-1, 10,972 with one FDR affected by PCa-2 and 2,896,561 without any FDRs affected by cancer in a maximum of 57 years. PC patients were further followed for death due to PC since diagnosis. Relative risk (RR) of PC was estimated with Poisson regression and hazard ratio (HR) with Cox proportional hazard model.

Results: Compared to men without any FDRs affected by cancer, the RRs of PC in men with one FDR affected by PCa-1 and PCa-2 were 2.12 (95% confidence interval [CI]: 2.07-2.17) and 1.69 (1.54-1.85), respectively. The risk in men with one FDR affected by PCa-2 was significantly lower than those with one FDR affected by PCa-1 after additionally adjusting for family relationship (father-son and brothers) and age at diagnosis of PC in FDR (RR _{PCa-2 vs PCa-1} , 0.85, 95% CI, 0.78-0.94). PC patients with a family history of PCa-2 were more likely to be detected at late-stage and less likely to be diagnosed by screening, compared to those with a family history of PCa-1. Patients whose PC was diagnosed after the diagnosis of PCa-1 in FDRs had a better survival than those without a family history of cancer (HR, 0.88, 95% CI, 0.80-0.97), but no such association was observed among patients with a family history of PCa-2.

Conclusion: Our study indicates a discrepancy between PC risks associated with a family history of PCa-1 and PC-2 and the reason behind it may be multifactorial.

Keywords: cancer screening; cumulative incidence; familial clustering; multiple primary cancer; prognosis.