Predictive value of bone scan index using computer-aided diagnosis system for bone scans in patients receiving first-line hormone therapy for metastatic hormone-sensitive prostate cancer.

2016 ◽  
Vol 34 (2_suppl) ◽  
pp. 225-225
Author(s):  
Yasuhide Miyoshi ◽  
Masato Yasui ◽  
Shuko Yoneyama ◽  
Koichi Uemura ◽  
Takashi Kawahara ◽  
...  

225 Background: Recently, the CHAARTED and STAMPEDE studies showed a survival benefit for docetaxel when started with androgen deprivation therapy (ADT) in men with newly diagnosed metastatic hormone-sensitive prostate cancer (mHSPC). While GETUG-AFU 15 failed to demonstrate a survival benefit of early chemotherapy. New biomarker for select the candidate for early chemotherapy in mHSPC is warranted. The objective of this study is to evaluate the bone scan index (BSI) using computer-aided diagnosis system for bone scans for predictive factor in patients receiving ADT as first-line hormone therapy for mHSPC. Methods: We identified consecutive 85 mHSPC patients treated with maximum androgen blockade (MAB) as first-line hormone therapy. We analyzed the correlations between progression-free survival (PFS) of MAB and clinicopathological characteristics, including patients’ age, initial PSA levels, Gleason scores, clinical TNM stage, hemoglobin (Hb), lactase dehydrogenase (LDH), c-reactive protein (CRP), and bone scan index (BSI). Statistical analyses were assessed using cox proportional hazards regression models. Results: The median patients’ age was 73 and the median follow-up duration was 11.3months. The median initial PSA value was 270 ng/ml. Median BSI was 2.7 % (range: 0.0-14.6). Clinical or PSA progression occurred in 55 (64.7%) patients. The median time to progression was 12.9 months. In multivariate analysis, three significant risk factors for PFS were identified; patients’ age ( > 73 years old vs ≤ 73; HR 0.53, p = 0.038), initial PSA levels ( > 270 ng/mL vs ≤ 270; HR 0.53, p = 0.038), and BSI ( > 2.7 vs ≤ 2.7; HR 3.0, p < 0.000). We stratified the patients into two cohorts with low risk (0-1 risk factor present) and high risk (2-3 risk factors present). We found a significant difference in PFS among risk groups (median PFS 15.3 months vs 8.5, p < 0.000). Conclusions: Patients’ age, initial PSA levels, and bone scan index were the significant predictive factors for MAB as first-line hormone therapy in patients with mHSPC. These findings might support the decision-making of induction of early chemotherapy for mHSPC.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. e17507-e17507
Author(s):  
Vipal P. Durkal ◽  
Nicholas George Nickols ◽  
Matthew Rettig

e17507 Background: Prostate cancer commonly metastasizes to the bone and is associated with reduced survival, pathologic fractures and bone pain. The assessment of bone lesions is made with the technetium Tc99m(99mTc) bone scan, which relies on the subjective interpretation of radiologists and has a wide interobserver variability. There is an unmet need for a more objective and quantifiable measurement tool. Progenics Pharmaceuticals has introduced an automated bone scan index (aBSI), which employs artificial intelligence to quantify skeletal tumor burden. The automated bone scan index has been prospectively validated and is reproducible in large Phase III studies. The aBSI was validated by our study in the Veteran population at the West LA VA Medical Center. Methods: The first positive technetium 99 Tc99m bone scans of veterans diagnosed with metastatic, castration-sensitive prostate cancer were evaluated. Since 2011, a total of 107 evaluable patient bone scans were studied (n = 107). Patients with visceral metastases were excluded to evaluate only those with skeletal metastases. An automated bone scan index (aBSI) was generated for each scan using the Progenics Pharmaceuticals’ artificial intelligence platform. Multivariate analysis of aBSI with overall survival, prostate cancer specific survival, time from diagnosis to first positive bone scan, age at diagnosis, ethnicity, and Gleason score was assessed. Results: The study demonstrated a wide range of aBSI values (Range 0-16.84). Values calculated above the Median aBSI value (1.0) were prognostic for Overall Survival (p = 0.0009) and Prostate Cancer-Specific Survival (p = 0.0011). Patients in the highest quartile of aBSI values (range 5.2-16.84) showed a statistically significant Prostate Cancer-Specific Mortality (p = 0.0300) when compared to the lowest two quartiles (Range 0-1.07). The time from diagnosis to the first positive Tc99m bone scan statistically correlated with aBSI values (p = 0.0016). Multivariate analysis using Cox regression was utilized in the final statistical analysis of prostate cancer-specific mortality and overall survival. Conclusions: The automated Bone Scan Index provides a quantifiable and validated artificial intelligence biomarker to address an unmet need among metastatic prostate cancer patients. This tool was validated among Veterans, a pertinent population that is commonly affected by metastatic prostate cancer.


2017 ◽  
Vol 99 (4) ◽  
pp. 400-405 ◽  
Author(s):  
Yasuhide Miyoshi ◽  
Shuko Yoneyama ◽  
Takashi Kawahara ◽  
Yusuke Hattori ◽  
Jun-ichi Teranishi ◽  
...  

2015 ◽  
Vol 117 (5) ◽  
pp. 748-753 ◽  
Author(s):  
Mads H. Poulsen ◽  
Janne Rasmussen ◽  
Lars Edenbrandt ◽  
Poul F. Høilund-Carlsen ◽  
Oke Gerke ◽  
...  

2012 ◽  
Vol 110 (11b) ◽  
pp. E628-E634 ◽  
Author(s):  
Yozo Mitsui ◽  
Hiroaki Shiina ◽  
Yasushi Yamamoto ◽  
Masuo Haramoto ◽  
Naoko Arichi ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document