Characteristics Associated with the Use of Diagnostic Prostate Biopsy and Biopsy Outcomes in Australian Men

Modifiable risk factors for subsequent lethal prostate cancer among men with an initially negative prostate biopsy

BACKGROUND

Previously suggested modifiable risk factors for prostate cancer could have resulted from detection bias because diagnosis requires a biopsy. We investigated modifiable risk factors for a subsequent cancer diagnosis among men with an initially negative prostate biopsy.

METHODS

In total, 10,396 participants of the Health Professionals Follow-up Study with an initial negative prostate biopsy after 1994 were followed for incident prostate cancer until 2017. Potential risk factors were based on previous studies in the general population. Outcomes included localised, advanced, and lethal prostate cancer.

RESULTS

With 1851 prostate cancer cases (168 lethal) diagnosed over 23 years of follow-up, the 20-year risk of any prostate cancer diagnosis was 18.5% (95% CI: 17.7-19.3). Higher BMI and lower alcohol intake tended to be associated with lower rates of localised disease. Coffee, lycopene intake and statin use tended to be associated with lower rates of lethal prostate cancer. Results for other risk factors were less precise but compatible with and of similar direction as for men in the overall cohort.

CONCLUSIONS

Risk factors for future prostate cancer among men with a negative biopsy were generally consistent with those for the general population, supporting their validity given reduced detection bias, and could be actionable, if confirmed.

Family history, obesity, urological factors and diabetic medications and their associations with risk of prostate cancer diagnosis in a large prospective study

Background

Prostate cancer (PC) aetiology is unclear. PC risk was examined in relation to several factors in a large population-based prospective study.

Methods

Male participants were from Sax Institute’s 45 and Up Study (Australia) recruited between 2006 and 2009. Questionnaire and linked administrative health data from the Centre for Health Record Linkage and Services Australia were used to identify incident PC, healthcare utilisations, Prostate Specific Antigen (PSA) testing reimbursements and dispensing of metformin and benign prostatic hyperplasia (BPH) prescriptions. Multivariable Cox and Joint Cox regression analyses were used to examine associations by cancer spread, adjusting for various confounders.

Results

Of 107,706 eligible men, 4257 developed incident PC up to end 2013. Risk of PC diagnosis increased with: PC family history (versus no family history of cancer; HR_adjusted = 1.36; 95% CI:1.21–1.52); father and brother(s) diagnosed with PC (versus cancer-free family history; HR_adjusted = 2.20; 95% CI:1.61–2.99); severe lower-urinary-tract symptoms (versus mild; HR_adjusted = 1.77; 95% CI:1.53–2.04) and vasectomy (versus none; HR_adjusted = 1.08; 95% CI:1.00–1.16). PC risk decreased with dispensed prescriptions (versus none) for BPH (HR_adjusted = 0.76; 95% CI:0.69–0.85) and metformin (HR_adjusted = 0.57; 95% CI:0.48–0.68). Advanced PC risk increased with vasectomy (HR_adjusted = 1.28; 95% CI:1.06–1.55) and being obese (versus normal weight; HR_adjusted = 1.31; 95% CI:1.01–1.69).

Conclusion

Vasectomy and obesity are associated with an increased risk of advanced PC. The reduced risk of localised and advanced PC associated with BPH, and diabetes prescriptions warrants investigation.