Prognostic value of a computer-aided diagnosis system involving bone scans among men treated with docetaxel for metastatic castration-resistant prostate cancer

Abstract

Background: The bone scan index (BSI), which is obtained using a computer-aided bone scan evaluation system, is anticipated to become an objective and quantitative clinical tool for evaluating bone metastases in prostate cancer. Here, we assessed the usefulness of the BSI as a prognostic factor in patients with metastatic castration-resistant prostate cancer (mCRPC) treated using docetaxel.

Methods: We analyzed 41 patients who received docetaxel for mCRPC. The Bonenavi system was used as the calculation program for the BSI. The utility of the BSI as a predictor of overall survival (OS) after docetaxel was evaluated. The Cox proportional hazards model was used to investigate the association between clinical variables obtained at docetaxel treatment, namely PSA, patient age, liver metastasis, local therapy, hemoglobin (Hb), lactase dehydrogenase (LDH), albumin (Alb), PSA doubling time, and BSI and OS.

Results: The median OS after docetaxel therapy was 17.7 months. Death occurred in 22 (53.7%) patients; all deaths were caused by prostate cancer. In multivariate analysis, three factors were identified as significant independent prognostic biomarkers for OS after docetaxel; these were liver metastases (yes vs no; HR, 3.681; p = 0.026), Alb (<3.9 vs ≥ 3.9; HR, 3.776; p = 0.020), and BSI (>1% vs ≤ 1%; HR, 3.356; p = 0.037). We evaluated the discriminatory ability of our models including or excluding the BSI by quantifying the c-index. The BSI improved the c-index from 0.758 to 0.769 for OS after docetaxel. CRPC patients with a BSI >1 had a significantly shorter OS than patients with a BSI ≤ 1 (p = 0.029).

Conclusions: The BSI, liver metastases and Alb were independent prognostic factors for OS after docetaxel. The BSI might be a useful tool for risk stratification of mCRPC patients undergoing docetaxel treatment.

Fig. 1

Kaplan–Meier curve for overall survival (OS) after docetaxel induction therapy. The median OS after docetaxel was 17.7 months

Fig. 2

Correlation of EOD classification and the BSI. Box plots indicate the first and third quartiles. The band inside the box shows the median. Lines extending vertically from the boxes (whiskers) indicate variability outside the upper and lower quartiles. There was a significant correlation between EOD and the BSI (r = 0.693). BSI; bone scan index, EOD; extent of disease on bone scan

Fig. 3

Correlation of ALP levels and the BSI. There was a strong correlation between serum ALP levels and the BSI (r = 0.643). ALP; alkaline phosphatase, BSI; bone scan index

Fig. 4

Kaplan–Meier curve for overall survival (OS) after docetaxel induction therapy according to the bone scan index (BSI). The blue line indicates survival for patients with a BSI ≤1 (n = 18) and the red line indicates survival for patients with a BSI >1 (n = 23). The median OS of patients with a BSI ≤1 and a BSI >1 was 26.5 and 15.4 months, respectively (p = 0.029)

Fig. 5

Kaplan–Meier curve for overall survival (OS) after docetaxel induction therapy according to risk group. We stratified the patients into two cohorts with low risk (0–1 risk factor present) and high risk (2–3 risk factors present). The blue line indicates survival for patients with low risk (n = 25) and the red line indicates survival for patients with high risk (n = 16). The median OS of patients with low risk and high risk was 26.5 and 11.5 months, respectively (p < 0.000). The risk factors in each risk group were shown in Table 3