Background

To evaluate PSA levels and kinetic cutoffs to predict positive bone scans for men with non-metastatic castration-resistant prostate cancer (CRPC) from the Shared Equal Access Regional Cancer Hospital (SEARCH) cohort.

Methods

Retrospective analysis of 531 bone scans of 312 clinically CRPC patients with no known metastases at baseline treated with a variety of primary treatment types in the SEARCH database. The association of patients' demographics, pathological features, PSA levels and kinetics with risk of a positive scan was tested using generalized estimating equations.

Results

A total of 149 (28%) scans were positive. Positive scans were associated with younger age (odds ratio (OR)=0.98; P=0.014), higher Gleason scores (relative to Gleason 2-6, Gleason 3+4: OR=2.03, P=0.035; Gleason 4+3 and 8-10: OR=1.76, P=0.059), higher prescan PSA (OR=2.11; P<0.001), shorter prescan PSA doubling time (PSADT; OR=0.53; P<0.001), higher PSA velocity (OR=1.74; P<0.001) and more remote scan year (OR=0.92; P=0.004). Scan positivity was 6, 14, 29 and 57% for men with PSA<5, 5-14.9, 15-49.9 and ⩾ 50 ng ml(-1), respectively (P-trend <0.001). Men with PSADT ⩾ 15, 9-14.9, 3-8.9 and <3 months had a scan positivity of 11, 22, 34 and 47%, correspondingly (P-trend <0.001). Tables were constructed using PSA and PSADT to predict the likelihood of a positive bone scan.

Conclusions

PSA levels and kinetics were associated with positive bone scans. We developed tables to predict the risk of positive bone scans by PSA and PSADT. Combining PSA levels and kinetics may help select patients with CRPC for bone scans.

Purpose We investigated imaging practice patterns in men with nonmetastatic (M0) castration resistant prostate cancer. Materials and Methods We analyzed data on 247 patients with documented M0 CRPC from the SEARCH database. Patients were selected regardless of primary treatment modality and all had a negative bone scan after a castration resistant prostate cancer diagnosis. Cox models were used to test associations of time to a second imaging test with several demographic and clinical factors. Results During a median followup of 29.0 months (IQR 12.9-43.5) after a post-castration resistant prostate cancer bone scan was negative, 190 patients (77%) underwent a second imaging test. On univariable analysis patients with higher prostate specific antigen (HR 1.13, p = 0.016), shorter prostate specific antigen doubling time (HR 0.79, p <0.001) and faster prostate specific antigen velocity (HR 1.01, p <0.001) were more likely to undergo a second imaging test. Treatment center was also a significant predictor of a second imaging test (p = 0.010). No other factor was a significant predictor. Results were similar on multivariable analysis. It was estimated that approximately 20% of men with a prostate specific antigen doubling time of less than 3 months did not undergo an imaging test in the first year after a post-castration resistant prostate cancer negative bone scan. However, 50% of patients with prostate specific antigen doubling time 15 months or greater underwent a second imaging test in the first year. Conclusions Clinicians use some known predictors of positive imaging tests to determine which patients with M0 castration resistant prostate cancer undergo a second imaging test. However, there may be under imaging in those at high risk and over imaging in those at low risk. Further studies are needed to identify risk factors for metastasis and form clear imaging guidelines in patients with M0 castration resistant prostate cancer.