Bone scan index: A new biomarker of bone metastasis in patients with prostate cancer

Transcription

1 International Journal of Urology (2017) 24, doi: /iju Review Article Bone scan index: A new biomarker of bone metastasis in patients with prostate cancer Kenichi Nakajima, 1 Lars Edenbrandt 2 and Atsushi Mizokami 3 1 Department of Nuclear Medicine, Kanazawa University, Kanazawa, Japan, 2 Department of Clinical Physiology and Nuclear Medicine, University of Gothenburg, Gothenburg, Sweden, and 3 Department of Urology, Kanazawa University, Kanazawa, Japan Abbreviations & Acronyms ANN = artificial neural network BSI = bone scan index CRPC = castration-resistant prostate cancer EOD = extent of disease MDP = methylenediphosphonate PSA = prostate-specific antigen QOL = quality of life Correspondence: Kenichi Nakajima M.D., Department of Nuclear Medicine, Kanazawa University, 13-1 Takara-machi, Kanazawa , Japan. nakajima@med. kanazawa-u.ac.jp This is an open access article under the terms of the Creative Commons Attribution- NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is noncommercial and no modifications or adaptations are made. Received 10 March 2017; accepted 1 May Online publication 26 May 2017 Abstract: Bone scintigraphy is one of the first-line imaging modalities for the screening and follow up of bone metastasis in patients with prostate cancer. The amount (%) of bone metastasis can be calculated using a bone scan index thanks to recent advances in quantitative bone scintigraphy. Since an artificial neural network was applied for hot-spot characterization and quantitation, the bone scan index has become a simple, reproducible and practical means of quantifying bone metastasis. The bone scan index is presently considered as an imaging biomarker of bone metastasis. The present article summarizes the principles and application of bone scan index using dedicated software (EXINI bone in Europe and North America; BONENAVI in Japan), and the advantages and cautions of using the bone scan index. The bone scan index could serve as a practical marker with which to monitor disease progression and treatment effects in multicenter studies, and to manage prostate and other types of cancer in the clinical setting. Key words: artificial neural network, biomarker, bone scan, quantitation, scintigraphy. Introduction Although recent advances in oncological treatment have helped to improve the survival of patients with prostate cancer, the appropriate management of bone metastasis has become increasingly important. The occurrence of bone metastasis worsens the QOL of cancer patients due to events associated with the skeleton, such as bone pain, pathological fractures, hypercalcemia and overall survival. Early diagnosis of bone metastasis and appropriate surrogate markers linked to relevant outcomes have therefore been investigated. 1 The BSI was developed as a marker of the total amount of bone metastasis using whole-body scintigraphy with 99m Tc-MDP. 2,3 The BSI is now considered as an index, or imaging biomarker, of the extent of whole-body bone metastasis. The present article reviews the purpose of the BSI, how it is calculated, how it can be interpreted compared with conventional criteria, and the key benefit of using it for prognostic applications and to monitor treatment. Need for quantitation Bone scintigraphy is a classical imaging modality. However, it continues to be the first choice for surveying the entire body at once, and it can indicate the need for additional imaging when bone metastasis is suspected. 4 As the nature of bone metastasis is osteoblastic in patients with prostate cancer, bone scintigraphy can visualize abnormalities as hot spots or localized accumulation. Although hot spots are also common in non-metastatic conditions, such as trauma, degenerative bone changes and infections, it still has convenient characteristics for screening when indications are selected and interpretation is appropriate. However, therapeutic effects might not be easily determined by visual assessment of the number and size of hot spots. The criteria of the Prostate Cancer Clinical Trials Working Group 2 include counting the number of bone metastases and recognizing that two or more metastatic lesions compared with a previous imaging assessment are signs of exacerbation. 5 EOD analysis classifies <6, 6 20, and >20 hot spots as grades 1, 2 and 3, respectively, and involvement of >75% of the ribs, vertebrae and pelvic bones (super scan) as grade However, counting hot spots could still be subjective, and thus a more objective quantitative methodology has been sought from the viewpoint of clinical and investigative purposes The Authors. International Journal of Urology published by John Wiley & Sons Australia, Ltd on behalf of the Japanese Urological Association.

2 Bone scan index as a biomarker of bone metastasis History of BSI Investigators at Memorial Sloan-Kettering Cancer Center (New York, NY, USA) developed the BSI during 1977 as a promising means of measuring bone metastasis. 2 The initial method was semiautomatic, but the manual steps were laborious and the metric was not widely applied. A Swedish group then developed a method based on an ANN to completely automate the analysis of whole-body bone scans. The ANN is an artificial intelligence that is trained using examples derived from large databases, 9 and it was initially used as a tool to support clinical decisions due to having acceptable and reproducible diagnostic accuracy. 10 The tool was further developed to calculate BSI. The steps for the automated computer analysis somewhat follow the actual diagnostic steps that are taken by humans, namely the detection, characterization and judgment of severity of abnormalities. The software is now available as EXINI bone (EXINI Diagnostics, Lund, Sweden). The BSI was also introduced to Japan as BONENAVI software (FUJIFILM RI Pharma, Tokyo, Japan). 11 After a database project using Japanese patients with prostate, breast and other types of cancer, its diagnostic ability was enhanced and adjusted to the Japanese practice of bone scintigraphy. 12 In addition to diagnostic measurements of the total amount of bone metastases, the BSI is also useful for prognostic purposes. For example, risk for survival among patients with prostate cancer classified as BSI <1.4%, % and >5.1% could be clearly stratified. 13 Thus, the prognostic value of the BSI is promising, and it is being applied more frequently. BSI based on ANN: Steps for calculating BSI The BSI is automatically processed in seven steps as follows (Fig. 1) Segmentation of whole-body image: A whole-body bone template consisting of 12 regions including the skull, ribs, vertebra, pelvis and extremities is constructed based on the average bone architecture for each sex determined from standard scintigrams. 2. Detection of hot spots: Potentially abnormal regions are identified based on a specific algorithm for detecting hot spots that incorporates the count threshold as well as surrounding structures and physiological distribution. 3. Standardization: Normal bone counts are standardized so that bone images with different timing are shown in the same count density format. As technologists can easily adjust the range of displayed counts, this normalization step is convenient for comparing serial data in the same patient. 4. Hot-spot quantitation: Hot spots are quantified in all possible abnormal regions and segments. 5. Hot-spot classification: Features of hot spots, such as size, shape, intensity and distribution, are determined, and ANN analysis classifies all hot areas into high ( 0.5) and low (<0.5) probabilities of metastasis. These areas are visualized as red and blue areas, respectively, on whole-body images. 6. Calculation of BSI: The fraction of the skeleton of each abnormal area with a high probability (>0.5) is determined, and then the BSI is calculated as the sum of all such fractions. As the ANN system is trained to identify bone metastasis, BSI is an indicator of the total amount of bones with metastases in the ideal calculation setting. 7. Display format: The three major indices are the probability of abnormalities (ANN range 0 1), the total amount of bone metastasis (BSI [%]) and the number of hot spots. Serial changes in the BSI and number of hot spots can be shown as graphs that are convenient for understanding the clinical course of patients during therapy. (a) Original (b) Segmentation (c) Quantification ANN = 1.00, BSI = 6.2 % HS number = 56 Fig. 1 Steps in EXINI bone/bonenavi software. (a) Anterior and posterior views of the original bone scintigram. (b) After appropriate segmentation of whole body scintigram, (c) hot spots (HS) are detected and characterized. The final output is probability of abnormalities (ANN), BSI and (c) number of HS. Probability of metastasis is 1.00 (100%), BSI is very high (6.2%) and 56 HS are evident in this patient The Authors. International Journal of Urology published by John Wiley & Sons Australia, Ltd on behalf of the Japanese Urological Association. 669

3 K NAKAJIMA ET AL. BSI and extent of bone metastasis In principle, the BSI can be automatically calculated without technologist interaction. 14 However, visualized red (high probability) and blue (low probability) regions that are considered inappropriate can be reclassified. Although processing usually works well, urinary contamination, known traumatic regions and apparent misjudgment, for example, can cause regions to become colored. Although the ANN is trained to recognized metastatic patterns, such minor misjudgment is unavoidable. However, even in the face of minor false positive findings, the automated BSI is sufficiently useful and practical for serial follow-up studies during treatment. The present authors believe that to modify all interpretations in a series of studies is impractical, except for apparent artifactual mistakes in judgment. Furthermore, an ANN value with a probability approximately 0.5 could be on either side of the red and blue areas depending on minor changes during treatment, but this is not a misjudgment of metastasis. One of the key features of the BSI is that the extent of bone metastasis can be objectively grasped on bone scintigrams (Fig. 2). The BSI is 0% in patients without metastasis, and it increases with the amount of defined metastases. Multiple metastatic hot spots usually correspond to BSI >1%. Multiple hot spots distributed in whole-body images of bones or super scan usually have a BSI of 5% to 10%. The understanding of BSI ranges and bone imaging patterns are convenient to standardize the perception of disease severity and therapeutic changes when patients are referred to or from other hospitals. Analytical validation with a revised algorithm showed the linearity of automated BSI with simulated phantom BSI including ranges of BSI >10%. 15 BSI for diagnosis and treatment The BSI can evaluate therapeutic courses including hormonal therapy and chemotherapy, and could thus serve as a biomarker of bone metastasis instead of various chemical markers of bone and tumor markers. The BSI closely correlates with the bone metabolic markers, pyridinoline crosslinked carboxy-terminal telopeptide of type I collagen, bone alkaline phosphatase and tartrate-resistant acid phosphatase- 5b, and the correlation is closest with log (bone alkaline phosphatase). 16 In contrast, the correlation with PSA is only fair. The EOD can be graded based on the number of hot spots as described above. However, the BSI can be more effective than such crude visual grading, and reflect changes in therapeutic responses more appropriately than visual interpretation. Figure 3 shows a patient with prostate cancer. Anti-androgen therapy including bicalutamide significantly decreased bone metastasis while PSA values increased during the first year. Therapy with radium-223 (Xofigo; Bayer Yakuhin, Tokyo, Japan) significantly decreased the BSI during the second year, showing that the treatment of bone metastasis was effective. BSI for prognosis A promising application of BSI is prognostic evaluation. A study of patients with CRPC found that a doubled BSI results in a 1.9-fold increase in the risk of death. 17 Bivariate analysis showed that a change in BSI adjusted for PSA is prognostic at 3 and 6 months of treatment, whereas PSA is not prognostic if adjusted for BSI. The overall survival is similarly longer (a) (b) (c) (d) Whole-body bone image: posterior view Artificial neural network analysis BSI Fig. 2 Posterior bone scintigram with 99m Tc-MDP (upper panel) and processed results with artificial neural network (lower panel). Extent of bone metastasis increases from (a) (d), which is easily understood using quantitative BSI (%) The Authors. International Journal of Urology published by John Wiley & Sons Australia, Ltd on behalf of the Japanese Urological Association.

4 Bone scan index as a biomarker of bone metastasis Year 1 Year 2 Ra-223 BSI (%) PSA (ng/ml) Fig. 3 Follow-up studies of a man aged in his 60s with prostate cancer. After Ra-223 therapy, BSI (%) clearly decreased, showing a good response in this patient. Serial values of PSA are shown. for patients with CRPC and a reduced automatic BSI than for other patients. 18 Whereas the Gleason score and BSI are associated with the survival of patients with high-risk prostate cancer undergoing primary hormonal therapy, clinical stage and PSA are not prognostic. 19 These findings showed that the BSI might play roles in predicting outcomes and selecting appropriate treatment strategies. Perspectives in Japan The diagnostic accuracy of BONENAVI software was investigated on its introduction in Japan, and initial training proceeded using a single-center database. 20 A Japanese bone atlas that fit the population was also added. 21 To further enhance its diagnostic ability, the ANN system was re-trained in a multicenter database project using a new set of 1532 patients. 12 The training databases included 451 and 624 patients with prostate and breast cancer, respectively. The diagnostic accuracy of the re-trained ANN system was significantly better in the validation group (n = 503), compared with the training data in Sweden and the initial training data in Japan. The area under the receiver operating characteristic curve was 0.96 (sensitivity 94%, specificity 88%) for prostate cancer, and 0.92 for breast cancer, showing good diagnostic accuracy (Fig. 4). Although high diagnostic accuracy is a good basis for the accurate determination of metastatic cancer, a more attractive use of BSI would be for monitoring treatment and prognosis. A Japanese multicenter registry of patients with prostate cancer (PROSTAT-BSI study: Prostate cancer Registry Of STandard hormonal And chemo Therapy using Bone Scan Index) started to evaluate the role of BSI during hormonal treatment and chemotherapy. By the end of 2016, 237 patients were registered from 36 hospitals based on collaborations among Departments of Urology, Radiology and Nuclear Medicine in Japan. The role of BSI in the management of patients with prostate cancer will be analyzed after a followup period of 3 years. A new phase 2 clinical study is under investigation in patients with CRPC (JapicCTI ) in Japan to assess the efficacy and safety of TAS-115 (vascular endothelial growth factor receptor-targeted tyrosine kinase inhibitor; Taiho Pharmaceutical, Tokyo, Japan). 22 The primary outcome of that study uses BSI calculated with BONE- NAVI software. BSI could be a good surrogate marker for evaluating the effects of potentially useful new drugs for bone metastasis. Recent literature related to BSI and prognosis in patients with prostate cancer is summarized in Table 1. Although several threshold values have been used for survival analysis, such as BSI of 1% and 5%, in general, BSI of >1% resulted in progression, shorter overall survival or higher incidence of cancer death. An increase and decrease in BSI during treatment are also related to worsening and improvement in prognosis, respectively. European and North American perspectives Investigators in Sweden and at the Memorial Sloan-Kettering Cancer Center have collaborated to qualify BSI as an 2017 The Authors. International Journal of Urology published by John Wiley & Sons Australia, Ltd on behalf of the Japanese Urological Association. 671

5 K NAKAJIMA ET AL. (a) Prostate cancer 1.0 BN2 Sensitivity AUC EB: 0.94 BN1: BN2: EB BNI (b) Breast cancer 1.0 BN2 Sensitivity AUC EB: 0.86 BN1: 0.91 BN2: Specificity 1-Specificity EB BNI Fig. 4 Receiver operating characteristic analyses. Curves were generated from patients with (a) prostate and (b) breast cancer. The artificial neural network of the software was trained using databases in Sweden (EXINI bone; EB), Japan (BONENAVI 1; BN1) and revised in Japan (BN2). The area under the receiver operating characteristic curve (AUC) is equally high in patients with prostate cancer, but improved with BN2 in patients with breast cancer. 12 Table 1 Application of BSI in patients with prostate cancer Author and reference number Year No. patients Patients or study setting BSI cut-off and prognosis Sabbatini et al Androgen independent prostate cancer from a randomized trial Meirelles et al Progressing prostate cancer with bone metastasis BSI of <1.4%, % and >5.1% (median survivals of 18.3, 15.5 and 8.1 months, respectively) BSI >1.27 and <1.27 (median survival of 14.7 and 28.2 months, respectively) Dennis et al Metastatic CRPC in four clinical trials Doubling in BSI, 1.9-fold increase in risk of death; change in BSI, prognostic at 3 and 6 months on treatment Mitsui et al CRPC patients with taxane-based Reduction in BSI showing longer overall survival chemotherapy Ulmert et al Prostate cancer cases in two large population-based cohorts BSI groups of 0, and >1.00 correlated prostate cancer death Kaboteh et al High-risk prostate cancer patients receiving primary hormonal therapy 5-year probability of survival: 55% for patients without metastases, 42% for BSI <1, 31% for BSI 1 5 and 0% for BSI >5 Kaboteh et al Prostate cancer patients with chemotherapy 2-year survival for patients with increasing and decreasing BSI from baseline to follow-up scans: 18% and 57%, respectively Reza et al Prostate cancer patients during androgen deprivation therapy Armstrong Randomized double-blind trial of tasquinimod et al. 26 in men with metastatic CRPC Uemura et al Patients with metastatic CRPC treated with docetaxel Miyoshi et al Hormone-naive prostate cancer patients with bone metastases Poulsen et al Patients with prostate cancer awaiting initiation of androgen-deprivation therapy BSI groups of 0, and >1.00 correlated prostate cancer death (5-year survival rates of 80%, 60% and 25%, respectively) BSI worsening at 12 weeks, prognostic for progression-free survival; patients with BSI >1.0 showing reduced survival Patients with BSI >1 showing shorter overall survival than patients with BSI 1 Median OS: not reached in patients with BSI 1.9, 34.8 months in patients with BSI >1.9 Patients with BSI 1, significantly shorter overall survival than patients with BSI <1 imaging biomarker. This work has comprised the pre-analytical, analytical and clinical validation of the automated, objective method of calculating BSI. The effect of variability in image acquisition on the BSI calculation was analyzed during the pre-analytical validation. 30 The results showed that the BSI is dependent on the quality of anterior and posterior images defined as counts, but not on vendor-specific gamma cameras. It is therefore recommended that the scan speed be adjusted to ensure that the total count is 1.5 million per image; that is, to adhere to the guidelines of the Society of Nuclear Medicine and Molecular Imaging. The analytical validation showed that BSI is an accurate, precise and reproducible method for quantitatively assessing tumor burden. 15 The pre-analytical and analytical studies incorporated both simulated bone scans and studies of actual patients. The advantage of the simulated phantom studies is that tumor burden is known and can be predefined. The third step in the validation process clinically validates BSI as a prognostic imaging biomarker for overall survival, radiographic progression-free survival and symptomatic skeleton-related events. Bone images from a phase 2 clinical trial were analyzed, and the results showed that BSI and changes in BSI during treatment are independently associated with The Authors. International Journal of Urology published by John Wiley & Sons Australia, Ltd on behalf of the Japanese Urological Association.

6 Bone scan index as a biomarker of bone metastasis overall survival in men with metastatic CRPC. 26 Images from a phase 3 trial are currently under evaluation. The latest technical development is the introduction of a web-based service for calculating BSI for the North American market that has resulted in a validated imaging biomarker that is now available for clinical applications. Conclusion The new imaging biomarker BSI is a quantitative index of the total amount of bone metastasis determined based on the application of a neural network trained using a large number of learning databases. The development of novel drugs and internal radiation therapy for extending a better QOL requires an optimal surrogate marker of bone metastasis. Although the worldwide use of BSI is increasing, appropriate understanding, and the effective use of a computer-assisted diagnostic and management system will be required. Acknowledgments The authors thank the researchers and technologists for multicenter collaborations to develop BONENAVI software in Japan, and the editorial assistance of Norma Foster. Conflict of interest K Nakajima has collaborative research with FUJIFILM RI Pharma, Japan. L Edenbrandt is employed part-time by EXINI Diagnostics AB, Sweden. References 1 Scher HI, Morris MJ, Larson S, Heller G. Validation and clinical utility of prostate cancer biomarkers. Nat. Rev. Clin. Oncol. 2013; 10: Erdi YE, Humm JL, Imbriaco M, Yeung H, Larson SM. Quantitative bone metastases analysis based on image segmentation. J. Nucl. Med. 1997; 38: Imbriaco M, Larson SM, Yeung HW et al. A new parameter for measuring metastatic bone involvement by prostate cancer: the bone scan index. Clin. Cancer Res. 1998; 4: Van den Wyngaert T, Strobel K, Kampen WU et al. The EANM practice guidelines for bone scintigraphy. Eur. J. Nucl. Med. Mol. Imaging 2016; 43: Scher HI, Halabi S, Tannock I et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J. Clin. Oncol. 2008; 26: Noguchi M, Kikuchi H, Ishibashi M, Noda S. Percentage of the positive area of bone metastasis is an independent predictor of disease death in advanced prostate cancer. Br. J. Cancer 2003; 88: Soloway MS, Hardeman SW, Hickey D et al. Stratification of patients with metastatic prostate cancer based on extent of disease on initial bone scan. Cancer 1988; 61: Jorgensen T, Muller C, Kaalhus O, Danielsen HE, Tveter KJ. Extent of disease based on initial bone scan: important prognostic predictor for patients with metastatic prostatic cancer. Experience from the Scandinavian Prostatic Cancer Group Study No. 2 (SPCG-2). Eur. Urol. 1995; 28: Sadik M, Jakobsson D, Olofsson F, Ohlsson M, Suurkula M, Edenbrandt L. A new computer-based decision-support system for the interpretation of bone scans. Nucl. Med. Commun. 2006; 27: Sadik M, Hamadeh I, Nordblom P et al. Computer-assisted interpretation of planar whole-body bone scans. J. Nucl. Med. 2008; 49: Takahashi Y, Yoshimura M, Suzuki K et al. Assessment of bone scans in advanced prostate carcinoma using fully automated and semi-automated bone scan index methods. Ann. Nucl. Med. 2012; 26: Nakajima K, Nakajima Y, Horikoshi H et al. Enhanced diagnostic accuracy for quantitative bone scan using an artificial neural network system: a Japanese multi-center database project. EJNMMI Res. 2013; 3: Sabbatini P, Larson SM, Kremer A et al. Prognostic significance of extent of disease in bone in patients with androgen-independent prostate cancer. J. Clin. Oncol. 1999; 17: Ulmert D, Kaboteh R, Fox JJ et al. A novel automated platform for quantifying the extent of skeletal tumour involvement in prostate cancer patients using the bone scan index. Eur. Urol. 2012; 62: Anand A, Morris MJ, Kaboteh R et al. Analytic validation of the automated bone scan index as an imaging biomarker to standardize quantitative changes in bone scans of patients with metastatic prostate cancer. J. Nucl. Med. 2016; 57: Wakabayashi H, Nakajima K, Mizokami A et al. Bone scintigraphy as a new imaging biomarker: the relationship between bone scan index and bone metabolic markers in prostate cancer patients with bone metastases. Ann. Nucl. Med. 2013; 27: Dennis ER, Jia X, Mezheritskiy IS et al. Bone scan index: a quantitative treatment response biomarker for castration-resistant metastatic prostate cancer. J. Clin. Oncol. 2012; 30: Mitsui Y, Shiina H, Yamamoto Y et al. Prediction of survival benefit using an automated bone scan index in patients with castration-resistant prostate cancer. BJU Int. 2012; 110: E Kaboteh R, Damber JE, Gjertsson P et al. Bone scan index: a prognostic imaging biomarker for high-risk prostate cancer patients receiving primary hormonal therapy. EJNMMI Res. 2013; 3: Horikoshi H, Kikuchi A, Onoguchi M, Sjostrand K, Edenbrandt L. Computer-aided diagnosis system for bone scintigrams from Japanese patients: importance of training database. Ann. Nucl. Med. 2012; 26: Kikuchi A, Onoguchi M, Horikoshi H, Sjostrand K, Edenbrandt L. Automated segmentation of the skeleton in whole-body bone scans: influence of difference in atlas. Nucl. Med. Commun. 2012; 33: Watanabe K, Hirata M, Tominari T et al. The MET/vascular endothelial growth factor receptor (VEGFR)-targeted tyrosine kinase inhibitor also attenuates FMS-dependent osteoclast differentiation and bone destruction induced by prostate cancer. J. Biol. Chem. 2016; 291: Meirelles GS, Schoder H, Ravizzini GC et al. Prognostic value of baseline [18F] fluorodeoxyglucose positron emission tomography and 99mTc-MDP bone scan in progressing metastatic prostate cancer. Clin. Cancer Res. 2010; 16: Kaboteh R, Gjertsson P, Leek H et al. Progression of bone metastases in patients with prostate cancer automated detection of new lesions and calculation of bone scan index. EJNMMI Res. 2013; 3: Reza M, Bjartell A, Ohlsson M et al. Bone scan index as a prognostic imaging biomarker during androgen deprivation therapy. EJNMMI Res. 2014; 4: Armstrong AJ, Kaboteh R, Carducci MA et al. Assessment of the bone scan index in a randomized placebo-controlled trial of tasquinimod in men with metastatic castration-resistant prostate cancer (mcrpc). Urol. Oncol. 2014; 32: Uemura K, Miyoshi Y, Kawahara T et al. Prognostic value of a computeraided diagnosis system involving bone scans among men treated with docetaxel for metastatic castration-resistant prostate cancer. BMC Cancer 2016; 16: Miyoshi Y, Yoneyama S, Kawahara T et al. Prognostic value of the bone scan index using a computer-aided diagnosis system for bone scans in hormone-naive prostate cancer patients with bone metastases. BMC Cancer 2016; 16: Poulsen MH, Rasmussen J, Edenbrandt L et al. Bone scan index predicts outcome in patients with metastatic hormone-sensitive prostate cancer. BJU Int. 2016; 117: Anand A, Morris MJ, Kaboteh R et al. A preanalytic validation study of automated bone scan index: effect on accuracy and reproducibility due to the procedural variabilities in bone scan image acquisition. J. Nucl. Med. 2016; 57: The Authors. International Journal of Urology published by John Wiley & Sons Australia, Ltd on behalf of the Japanese Urological Association. 673