Comparison of Diagnostic Performance Between Manual Diagnosis Following PROMISE V2 and aPROMISE Utilizing Ga/F-PSMA PET-CT

Comparison of Diagnostic Performance Between Manual Diagnosis Following PROMISE V2 and aPROMISE Utilizing Ga/F-PSMA PET-CT | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Comparison of Diagnostic Performance Between Manual Diagnosis Following PROMISE V2 and aPROMISE Utilizing Ga/F-PSMA PET-CT Yuki Enei, Takafumi Yanagisawa, Atsuya Okada, Hidetoshi Kuruma, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6360294/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 15 Jul, 2025 Read the published version in Annals of Nuclear Medicine → Version 1 posted 4 You are reading this latest preprint version Abstract Backgrounds Automated PROMISE (aPROMISE), which is an artificial intelligence-supported software for prostate-specific membrane antigen (PSMA) PET-CT based on PROMISE V2, has demonstrated diagnostic utility with better correspondence rates compared to manual diagnosis. However, previous studies have consistently utilized F-PSMA PET-CT. Therefore, we investigated the diagnostic utility of aPROMISE using both F- and Ga-PSMA PET-CT of Japanese patients with metastatic prostate cancer (mPCa). Materials and Methods We retrospectively evaluated 21 PSMA PET-CT images (Ga-PSMA PET-CT: n = 12, F-PSMA PET-CT: n = 9) from 21 patients with mPCa. A single, well-experienced nuclear radiologist performed manual diagnosis following PROMISE V2 and subsequently performed aPROMISE-assisted diagnosis to assess miTNM and details of metastatic sites. We compared the diagnostic time and correspondence rates of miTNM diagnosis between manual and aPROMISE-assisted diagnoses. Additionally, we investigated the differences in diagnostic performance between the two radioisotopes. Results aPROMISE-assisted diagnosis was significantly associated with shorter median diagnostic time compared to manual diagnosis (427 seconds [IQR: 370–834] vs. 1,114 seconds [IQR: 922–1291], p < 0.001). The time reduction with aPROMISE-assisted diagnosis was particularly notable when using Ga-PSMA PET-CT. aPROMISE had high diagnostic accuracy with 100% sensitivity for miT, M1a, and M1b stages. Notably, for M1b stages, aPROMISE achieved 100% sensitivity and specificity, regardless of the type of radioisotope used. However, aPROMISE missed five visceral metastases (2 adrenal and 3 liver), resulting in lower sensitivity for miM1c stage (63%). In addition to detecting metastatic sites, aPROMISE successfully provided detailed metrics, including the number of metastatic lesions, total metastatic volume, and SUV mean. Conclusions aPROMISE-assisted diagnosis significantly reduces diagnostic time and achieves high accuracy compared to manual diagnosis, regardless of the type of radioisotopes used. While aPROMISE successfully detects bone metastases, its limitations in detecting visceral metastases need to be addressed. This study supports the utility of aPROMISE in Japanese patients with mPCa and underscore the need for further validation in larger cohorts. PSMA PET-CT metastatic prostate cancer artificial intelligence diagnosis PROMISE V2 aPROMISE Figures Figure 1 Figure 2 1. Introduction Prostate-specific membrane antigen (PSMA) PET-CT has become an indispensable tool for prostate cancer (PCa) diagnosis across various clinical settings, including the detection of metastatic disease (e.g., oligometastasis) and clinically significant localized PCa [ 1 – 3 ]. Standardized diagnostic frameworks, such as the Prostate Cancer Molecular Imaging Standardized Evaluation (PROMISE) version 2.0, have been established to enhance reproducibility and generalizability in diagnostic reporting for PSMA PET-CT[ 4 ]. There is growing interest in integrating artificial intelligence (AI) into medical imaging owing to its ability to automate and leverage radiomics, uncovering features potentially imperceptible to the human eye. Recently, automated PROMISE (aPROMISE, EXINI Diagnostics AB, Lund, Sweden), an AI-supported diagnostic software based on deep learning algorisms, has gained attention and received FDA approval for its ability to streamline PSMA PET-CT interpretation[ 5 ]. Preliminary studies have shown that aPROMISE achieves high concordance rates compared to manual interpretation, underscoring its potential to enhance clinical workflows[ 4 , 6 , 7 ]. However, these studies have predominantly utilized F-PSMA PET-CT, leaving a critical gap in understanding the utility of aPROMISE for Ga-PSMA PET-CT, particularly in non-Western populations, such as Japanese patients. To address this gap, we investigated the diagnostic performance of aPROMISE in Japanese patients with metastatic PCa (mPCa) using both F- and Ga-PSMA PET-CT, aiming to validate its utility and explore potential tracer-specific variations in diagnostic outcomes. 2. Materials and Methods 2.1. Patients We retrospectively selected a total of 21 PSMA PET-CT images from Japanese 21 patients with mPCa. The patients underwent [68Ga]Ga-PSMA-11 PET-CT(n = 12) or [18F]PSMA-1007 PET-CT (n = 9) between January 2020 and April 2024. Patients provided written informed consent before examination. This study was approved by The Jikei University Institutional Review Board (36 − 003(12102)). 2.2. PROMISE V2 and aPROMISE The PROMISE criteria have been proposed as a framework for whole-body molecular imaging-based TNM staging, denoted miTNM staging, to describe the prostate cancer disease extent on PSMA PET-CT[ 4 ]. The second version of PROMISE framework (PROMISE V2), which integrates an updated miTNM staging system, modifying assessment of local disease and PSMA-expression score (quantify and standardize the level of PSMA in prostate cancer tissues) for clinical routine [ 8 ] aPROMISE is an artificial intelligence-enabled medical device software for quantitative analysis and standardized reporting of PSMA PET/CT images in prostate cancer following the PROMISE criteria [ 6 , 8 ]. aPROMISE was developed via deep learning algorithms trained on over 3,000 PSMA images to perform automated segmentation, detection, and quantification of PSMA PET/CT lesions, implementing and extending the PROMISE criteria[ 6 – 8 ] . 2.3. Diagnostic and evaluation methods To validate the diagnostic performance of aPROMISE, we compared the manual diagnosis with aPROMISE-assisted diagnosis as the way below. A single well-experienced nuclear radiologist diagnosed molecular imaging TNM (miTNM) staging and details of each image following the PROMISE V2 reporting sheets [ 8 ]: 1) Manual diagnosis following PROMISE V2 and 2) aPROMISE-assisted diagnosis. These diagnoses were randomly and separately conducted with decent interval duration and blinded patient information on each image. We compared the diagnostic time and correspondence rates of miTNM diagnosis and number of PSMA-positive lesions between the two diagnostic methods. Diagnostic time was measured by manual measurement by the radiologist from the start to the end of filling the reporting sheets. Concordance rates were evaluated under the definition that manual diagnosis by an experienced nuclear radiologist was defined as “true”. 2.4. Statistical analysis The diagnostic time (in seconds) for aPROMISE-assisted diagnosis and manual diagnosis was represented and compared using the median and interquartile range. Diagnostic concordance between the two diagnostic methods was evaluated using a 2×2 contingency table. Sensitivity and specificity were calculated based on this table. All statistical analyses were performed using R version 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria), and a p-value < 0.05 was considered statistically significant. 3. Results 3.1. Overview of manual and aPROMISE-assisted diagnosis The summary of manual diagnosis and aPROMISE-assisted diagnosis are shown in Table 1 . In miT staging, seven patients (33%) were detected with PSMA-positive lesions in manual diagnosis, while nine patients (43%) were identified using aPROMISE-assisted diagnosis. In Ga-PSMA PET-CT, five patients were consistently diagnosed as PSMA-positive using both diagnostic methods. However, in F-PSMA PET-CT, aPROMISE-assisted diagnosis identified two additional patients with PSMA-positive lesions (22% vs. 44%). The distribution of PSMA expression scores for the hottest lesion was comparable between the two diagnostic methods. Regarding the number of positive lesions, aPROMISE-assisted diagnosis identified more patients with two PSMA-positive lesions (14% vs. 56%). In miN staging, eight patients (38%), including four for N1 and four for N2, were detected with PSMA-positive lesions in manual diagnosis. On the other hand, seven patients (33%), including four for N1 and three for N2, were detected using aPROMISE-assisted diagnosis. The distribution of PSMA expression scores for the hottest lesion and the number of PSMA-positive lesions were comparable between the two diagnostic methods. In miM1a staging, ten patients (48%) were detected with PSMA-positive lesions in manual diagnosis, while 12 patients (57%) were identified using aPROMISE-assisted diagnosis. aPROMISE-assisted diagnosis was more likely to detect higher PSMA expression scores for the hottest lesion (30% with a score of 3 in manual diagnosis vs. 58% with a score of 3 in aPROMISE-assisted diagnosis) and identified a greater number of patients with multiple PSMA-positive lesions (50% vs. 67%). In miM1b staging, 17 patients (81%) were identified with PSMA-positive lesions, irrespective of diagnostic methods. While the distribution of PSMA expression scores for the hottest lesion showed a slight difference, with a greater number of score 3 lesions detected in aPROMISE-assisted diagnosis (71% vs. 88%), the distribution of the number of PSMA-positive lesions was entirely similar. In miM1c staging, aPROMISE-assisted diagnosis showed lower detection rates (n = 8, 38% in manual diagnosis vs. n = 5, 24% in aPROMISE-assisted diagnosis). aPROMISE-assisted diagnosis missed three liver lesions and two other metastases. Detailed information and case presentations on the discordance in detecting visceral metastases between the two methods are provided in a separate section, 3.4. Diagnostic disconcordance in visceral metastases . Beyond miTNM staging, aPROMISE-assisted diagnosis provided detailed evaluations of PSMA PET-CT images ( Supplementary Tables 1 ). Notably, only aPROMISE-assisted diagnosis offered precise information on the number of metastases, particularly in cases with numerous PSMA-positive lesions. Furthermore, aPROMISE routinely reported detailed metrics such as the total number and volume of PSMA-positive lesions, as well as the mean and maximum values of standardized uptake values (SUV) ( Supplementary Table 1 ). 3.2. Diagnostic time The comparison of diagnostic time measured by the radiologist between the two diagnostic methods is shown in Fig. 1 . As shown in Fig. 1A , median diagnostic time for aPROMISE-assisted diagnosis was significantly shorter (427 seconds [IQR: 370–834]) compared to manual diagnosis (1114 seconds [IQR; 922–1291], p < 0.001). In 18 patients using F-PSMA PET-CT, median diagnostic time of aPROMISE-assisted and manual diagnosis were 704 seconds (IQR: 493–1019) and 1114 seconds (IQR: 679–1123), respectively (p = 0.5, Fig. 1B ). In 24 patients using Ga-PSMA PET-CT, median diagnostic time of aPROMISE-assisted and manual diagnosis were 390 seconds (IQR: 350.25-455.25) and 1204 seconds (IQR: 984.25-1370.5), respectively (p < 0.001, Fig. 1C ). 3.3. Concordance rates Concordance rates were evaluated based on the aforementioned definition, with manual diagnosis by an experienced nuclear radiologist considered as the “true” reference. Table 2 summarizes the sensitivity and specificity of aPROMISE-assisted diagnosis compared to manual diagnosis. In miT staging, sensitivity and specificity were 100% and 86%, respectively, in the overall cohort. In miN staging, sensitivity and specificity were 88% and 100%, respectively, for the diagnosis of node positivity. However, due to discordance in identifying N1 and N2 diseases, aPROMISE-assisted diagnosis showed low concordance rates for N1 disease (25%). In miM1a staging, sensitivity and specificity were 100% and 82%, respectively, in the overall cohort. Notably, in miM1b staging, sensitivity and specificity were both 100%, regardless of the radioligand used for PSMA PET-CT. In miM1c staging, despite discordance in detecting visceral metastatic lesions, sensitivity and specificity were 63% and 100%, respectively. In summary of the differential diagnostic performance across radioligand types, sensitivity and specificity were both 100% in miT, miN, and miM1b staging with Ga-PSMA PET-CT, as well as in miM1a and miM1b staging with F-PSMA PET-CT. 3.4. Diagnostic disconcordance in visceral metastases The summary of eight patients who had visceral metastasis using manual diagnosis is shown in Table 3 . We identified only one patient (case #10) who completely matched visceral metastasis between manual and aPROMISE-assisted diagnoses. In total, aPROMISE missed five metastatic lesions (2 adrenal and 3 liver metastases), while detected two lung metastases. For example, in the case #4, multiple liver metastases were detected by manual diagnosis; however, no liver metastases were reported in aPROMISE-assisted diagnosis. (Fig. 2) 4. Discussion In this study, we compared the diagnostic times and concordance rates of manual and AI-assisted diagnostics, as well as the differential diagnostic performance of Ga-PSMA PET and F-PSMA PET. Our study confirmed the diagnostic utility of aPROMISE and revealed several important insights. First, aPROMISE-assisted diagnosis significantly reduced diagnostic time compared to manual diagnosis. Second, aPROMISE-assisted diagnosis demonstrated high diagnostic performance, particularly in detecting bone metastases, regardless of the type of radioligand used. Third, however, discrepancies were observed between the two diagnostic methods in detecting visceral metastatic lesions. Last but not least, while several previous studies have highlighted the utility of AI-assisted diagnosis in F-PSMA PET-CT imaging, no similar reports have been published for Ga-PSMA PET-CT [ 6 ]. Despite the preliminary nature of this study, it is the first to demonstrate the efficacy of aPROMISE even when using Ga-PSMA PET-CT. The application of AI-assisted diagnosis has the potential to reduce the burden on nuclear physician and radiologists while ensuring consistency in quantification for PSMA PET-CT diagnosis [ 9 , 10 ]. Nickols et al. demonstrated that aPROMISE-assisted diagnosis led to low inter-reader variability between two nuclear radiologists in PCa staging[ 6 ]. This study comprehensively showed a high level of Cohen pairwise k-agreement between the two radiologists (ranging from 0.77 to 0.90) in detecting regional lymph node and/or distant bone metastatic disease[ 6 ]. In the current study, we assessed another aspect of reducing the diagnostic burden through AI-assisted diagnosis, focusing on its ability to theoretically reduce diagnostic time. Our findings are the first to objectively demonstrate that aPROMISE-assisted diagnosis significantly shortens diagnostic time compared to manual diagnosis. Taken together, aPROMISE-assisted diagnosis addresses key limitations in PSMA PET-CT diagnosis, such as potential inter-reader variability and time-consuming nature of manual interpretation. Regarding the diagnostic accuracy of AI-assisted PSMA PET-CT diagnosis, a recent systematic review reported relatively high sensitivity (ranging from 62 to 97%) and accuracy (with an area under the curve up to 98%) for AI in detecting various types of metastatic diseases[ 10 ]. Specifically for the diagnostic capability of aPROMISE, Johnnson et al. demonstrated the high accuracy of segmentation achieved by aPROMSIE with excellent Dice scores compared to manual segmentation by experienced nuclear radiologists[ 7 ]. Furthermore, aPROMISE has shown impressive sensitivity rates for detecting potential lesions, achieving 91.5% for regional lymph nodes and 86.7% for bone lesions in metastatic PCa patients[ 7 ]. In our study, aPROMISE platform demonstrated exceptional diagnostic accuracy, achieving 100% sensitivity across miT, M1a, and M1b staging. Notably, for M1b staging, aPROMISE achieved 100% sensitivity and specificity, regardless of types of radioisotopes used. This high sensitivity underscores its clinical applicability for detecting significant pathological findings in mCRPC patients. On the other hand, aPROMISE missed some visceral metastases. To date, there has been no reports specifically focusing on the diagnostic utility of aPROMISE for detecting visceral metastases, likely due to the relatively low number of patients with visceral metastasis[ 10 ]. In our study, aPROMISE missed three liver and two adrenal metastases were missed. One possible explanation for missing liver metastases is the normal intake of PSMA in the liver, which may obscure metastatic lesions[ 11 , 12 ]. Despite its automation capabilities, aPROMISE still requires physicians to review and confirm the identification of lesions, particularly for visceral metastasis. Our findings suggest that integrating the AI-assisted diagnosis and manual interpretation is still important for more robust judgement for visceral metastases. In the argument of diagnostic performance comparisons between Ga-PSMA PET-CT and F-PSMA PET-CT, F-PSMA PET-CT has been recognized for its potential advantages in pelvic imaging owing to minimally excretion into the urinary tract, a longer half-life, production in larger quantities, and lower positron energy, which may improve spatial resolution[ 13 , 14 ]. However, a recent meta-analysis comparing the diagnostic impact of Ga-labeled PSMA and F-labeled PSMA imaging concluded that there is insufficient evidence to differentiate the radiotracers based on their clinical impact Given the liver-dominant excretion of [18F] PSMA-1007, it was observed to have higher liver uptake but lower urinary tract uptake compared to [68Ga] Ga-PSMA-11[ 14 ]. This difference could potentially affect AI-assisted diagnosis depending on the location of metastases. Due to the pilot nature of our study, we did not statistically compare the concordance rates between the radiotracers. However, our study demonstrated the comparable diagnostic concordance rates with aPROMISE-assisted diagnosis even when using Ga-PSMA PET-CT. To elucidate the potential impact of differential diagnostic performance based on tracer types, further well-designed studies with a sufficient number of patients are urgently needed. In addition to its shorter diagnostic time and high correspondence rates, aPROMISE provides a detailed and comprehensive assessment of disease burden and metrics in PSMA PET-CT imaging, which are critical for clinical decision-making in PSMA-radioligand therapy (RLT), such as 177-Lu PSMA RLT[ 10 ]. Of note, aPROMISE successfully reported the SUV mean value integrating all metastatic lesions in every case ( Supplementary Table 1 ). The SUV mean value has been recognized as a predictive factor for oncologic outcomes in patients with mCRPC undergoing PSMA-RLT[ 15 ]. Furthermore, post-hoc analyses of the TheraP and VISION trials demonstrated that whole-body SUV mean predicts treatment response in patients receiving 177Lu-PSMA-617 RLT[ 16 , 17 ]. These findings highlight the potential of specific software, such as aPROMISE, to support personalized medicine for patients with mCRPC, emphasizing its utility in routine clinical practice. Our study had several limitations. First, as a retrospective study with a limited number of patients, it restricted the ability to conduct robust statistical analysis. Second, we did not randomly select images from different tracers for PSMA-PET CT, resulting in imbalanced patient and disease demographics between the groups. Third, only one radiologist performed both the manual and aPROMISE-assisted diagnoses. Despite randomizing the order of PSMA-PET imaging and ensuring sufficient time intervals between diagnoses of the same patient, this study design might have influenced the diagnostic outcomes. 5. Conclusions aPROMISE-assisted diagnosis significantly reduced diagnostic time compared to manual diagnosis. aPROMISE-assisted diagnosis demonstrated high diagnostic accuracy, achieving 100% sensitivity for miT, M1a, M1b staging. Particularly for M1b staging, aPROMISE-assisted diagnosis achieved 100% sensitivity and specificity, regardless of the type of radioisotope used. However, aPROMISE missed several visceral metastatic lesions, specifically liver and adrenal metastasis. While our analysis confirmed the diagnostic utility of aPROMISE in Japanese patients with mPCa including both Ga-PSMA PET-CT and F-PSMA PET-CT, integrating AI-assisted diagnosis with manual interpretation should be considered for more robust judgement for visceral metastases. Further well-designed studies with a sufficient number of patients are necessary to validate our findings and clarify the differential diagnostic performance between the radioisotope types. Abbreviations PSMA: Prostate-specific membrane antigen PCa: Prostate cancer mPCa: metastatic PCa PROMISE: Prostate Cancer Molecular Imaging Standardized Evaluation AI: Artificial intelligence aPROMISE; automated PROMISE miTNM; molecular imaging TNM SUV: Standardized uptake values Declarations Conflict of interest The authors declare that they have no conflict of interest Research involving human participants and/or animals Data collection adhere to principles of the Declaration of Helsinki. This study was approved by The Jikei University Institutional Review Board (36-003(12102)). Informed consent All patients signed an informed consent agreeing to share their own anonymous information for future studies. Author contributions YE and TY: Project development, Data collection, Data analysis, Manuscript writing. KM and HK: Project development, Manuscript writing. CO, KW and NL : Data collection; Manuscript writing. AO and TK: Project development, Data analysis, Manuscript writing References Hofman MS, Emmett L, Sandhu S, et al. Overall survival with [177Lu]Lu-PSMA-617 versus cabazitaxel in metastatic castration-resistant prostate cancer (TheraP): secondary outcomes of a randomised, open-label, phase 2 trial. Lancet Oncol. 2024;25:99–107. Sartor O, de Bono J, Chi KN, et al. Lutetium-177-PSMA-617 for metastatic castration-resistant prostate cancer. N Engl J Med. 2021;385:1091–103. Morris MJ, Castellano D, Herrmann K, et al. 177Lu-PSMA-617 versus a change of androgen receptor pathway inhibitor therapy for taxane-naive patients with progressive metastatic castration-resistant prostate cancer (PSMAfore): a phase 3, randomised, controlled trial. Lancet. 2024;404:1227–39. Eiber M, Herrmann K, Calais J, et al. Prostate cancer molecular imaging standardized evaluation (PROMISE): Proposed miTNM classification for the interpretation of PSMA-ligand PET/CT. J Nucl Med. 2018;59:469–78. Calais J, Eiber M, Iagaru A, et al. Prospectively planned and independent validation of aPROMISE in a phase III CONDOR study for rapid lesion detection and standardized quantitative evaluation for 18F-DCFPyL (PSMA) imaging in prostate cancer. J Nucl Med. 2022;63:2496. Nickols N, Anand A, Johnsson K, et al. APROMISE: A novel automated PROMISE platform to standardize evaluation of tumor burden in 18F-DCFPyL images of veterans with prostate cancer. J Nucl Med. 2022;63:233–9. García Vicente AM, Lucas Lucas C, Pérez-Beteta J, et al. Analytical performance validation of aPROMISE platform for prostate tumor burden, index and dominant tumor assessment with 18F-DCFPyL PET/CT. A pilot study. Sci Rep. 2024;14:3001. Seifert R, Emmett L, Rowe SP, et al. Second version of the prostate cancer molecular imaging standardized evaluation framework including response evaluation for clinical trials (PROMISE V2). Eur Urol. 2023;83:405–12. Lindgren Belal S, Frantz S, Minarik D, et al. Applications of artificial intelligence in PSMA PET/CT for prostate cancer imaging. Semin Nucl Med. 2024;54:141–9. Liu J, Cundy TP, Woon DTS, Lawrentschuk N. A systematic review on artificial intelligence evaluating metastatic prostatic cancer and lymph nodes on PSMA PET scans. Cancers (Basel). 2024;16. https://doi.org/10.3390/cancers16030486 . Damjanovic J, Janssen J-C, Prasad V, et al. 68Ga-PSMA-PET/CT for the evaluation of liver metastases in patients with prostate cancer. Cancer Imaging. 2019;19:37. Rauscher I, Maurer T, Fendler WP, et al. (68)Ga-PSMA ligand PET/CT in patients with prostate cancer: How we review and report. Cancer Imaging. 2016;16:14. Giesel FL, Hadaschik B, Cardinale J, et al. F-18 labelled PSMA-1007: biodistribution, radiation dosimetry and histopathological validation of tumor lesions in prostate cancer patients. Eur J Nucl Med Mol Imaging. 2017;44:678–88. Buteau JP, Martin AJ, Emmett L, et al. PSMA and FDG-PET as predictive and prognostic biomarkers in patients given [177Lu]Lu-PSMA-617 versus cabazitaxel for metastatic castration-resistant prostate cancer (TheraP): a biomarker analysis from a randomised, open-label, phase 2 trial. Lancet Oncol. 2022;23:1389–97. Gafita A, Calais J, Grogan TR, et al. Nomograms to predict outcomes after 177Lu-PSMA therapy in men with metastatic castration-resistant prostate cancer: an international, multicentre, retrospective study. Lancet Oncol. 2021;22:1115–25. Huang S, Ong S, McKenzie D, et al. Comparison of 18F-based PSMA radiotracers with [68Ga]Ga-PSMA-11 in PET/CT imaging of prostate cancer-a systematic review and meta-analysis. Prostate Cancer Prostatic Dis. 2024;27:654–64. Kuo PH, Morris MJ, Hesterman J, et al. Quantitative 68Ga-PSMA-11 PET and clinical outcomes in metastatic castration-resistant prostate cancer following 177Lu-PSMA-617 (VISION trial). Radiology. 2024;312:e233460. Tables Tables 1 to 3 are available in the Supplementary Files section. Supplementary Files Table1aPRO1.jpg Table2aPRO1.jpg Table3aPRO1.jpg S.Table1aPRO1.jpg Cite Share Download PDF Status: Published Journal Publication published 15 Jul, 2025 Read the published version in Annals of Nuclear Medicine → Version 1 posted Reviewers agreed at journal 12 Apr, 2025 Reviewers invited by journal 11 Apr, 2025 Editor assigned by journal 02 Apr, 2025 First submitted to journal 02 Apr, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6360294","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":441501997,"identity":"d15ed94f-c549-4b20-83cd-15cb82af855f","order_by":0,"name":"Yuki Enei","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1UlEQVRIie2QvQrCMBCALwTiEtu1pWB9hHTpJH2WSkBXH8DBqVNxLujDnBQ66gM4KAhODpmKQxFbKbiZuAnmG47j4Ls/AIvlZxETyjgAdjlZmSmzrxUoAbjxSkirs1ocBg4vdwjLBOhGM0Ygm0eFOFI2zFKESgLZokY53eKAd4rLBQJDIEWqm+LWQSP2reIqhIeRwlnQXcSGOSDJDBQfWeznQrZPrgRO15Jrb3GQXr17k8gwlxel6mQU6T427lvKV2xX4lHx2YCwb5m8K55GsVgslr/jCekkPqUQT2TqAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0006-1124-2322","institution":"The Jikei University school of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Yuki","middleName":"","lastName":"Enei","suffix":""},{"id":441501998,"identity":"e2528ea2-d13b-43b0-b16a-60898315541f","order_by":1,"name":"Takafumi Yanagisawa","email":"","orcid":"","institution":"Department of Urology, Jikei University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Takafumi","middleName":"","lastName":"Yanagisawa","suffix":""},{"id":441501999,"identity":"7139cab7-ae66-4d74-9961-45e476575ad9","order_by":2,"name":"Atsuya Okada","email":"","orcid":"","institution":"Jinsenkai MI Clinic, Osaka, Japan","correspondingAuthor":false,"prefix":"","firstName":"Atsuya","middleName":"","lastName":"Okada","suffix":""},{"id":441502000,"identity":"57560f49-6f8c-44fb-adb8-ce4e3a27f58a","order_by":3,"name":"Hidetoshi Kuruma","email":"","orcid":"","institution":"Bashamichi Sakura Clinic","correspondingAuthor":false,"prefix":"","firstName":"Hidetoshi","middleName":"","lastName":"Kuruma","suffix":""},{"id":441502001,"identity":"5d1f7dbd-9be3-48e7-9c9c-8b1728cc386a","order_by":4,"name":"Chieko Okazaki","email":"","orcid":"","institution":"Theranostics Yokohama","correspondingAuthor":false,"prefix":"","firstName":"Chieko","middleName":"","lastName":"Okazaki","suffix":""},{"id":441502002,"identity":"8c0d67c1-480b-463e-879d-6e1e3c25512c","order_by":5,"name":"Ken Watanabe","email":"","orcid":"","institution":"Department of Radiology, Jikei University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Ken","middleName":"","lastName":"Watanabe","suffix":""},{"id":441502003,"identity":"f687b2fa-5958-4243-ab9d-311e46677cc2","order_by":6,"name":"Nat P Lenzo","email":"","orcid":"","institution":"ICON cancer centre","correspondingAuthor":false,"prefix":"","firstName":"Nat","middleName":"P","lastName":"Lenzo","suffix":""},{"id":441502004,"identity":"d76f986d-71c8-4352-b3ae-c5ecd68e8eba","order_by":7,"name":"Takahiro Kimura","email":"","orcid":"","institution":"Department of Urology, Jikei University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Takahiro","middleName":"","lastName":"Kimura","suffix":""},{"id":441502005,"identity":"fe010722-f156-4312-b686-93877d0a0dee","order_by":8,"name":"Kenta Miki","email":"","orcid":"","institution":"Department of Urology, Jikei University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Kenta","middleName":"","lastName":"Miki","suffix":""}],"badges":[],"createdAt":"2025-04-02 10:37:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6360294/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6360294/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s12149-025-02086-9","type":"published","date":"2025-07-15T15:57:27+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":80815332,"identity":"e99945d8-540f-48b8-99ea-06551cc1fbdc","added_by":"auto","created_at":"2025-04-17 10:57:26","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":253862,"visible":true,"origin":"","legend":"\u003cp\u003eThe comparison of diagnostic time between the manual and aPROMISE-assisted diagnosis (A) All modalities (B) F-PSMA PET (C) Ga-PSMA PET; Diagnostic times for each type of radioisotopes were also compared.\u003c/p\u003e","description":"","filename":"Fig1aPRO1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6360294/v1/500a375e39ff20008b71a15a.jpg"},{"id":80815340,"identity":"9764a583-857f-41c9-9f09-a4419b5f06a8","added_by":"auto","created_at":"2025-04-17 10:57:26","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":168308,"visible":true,"origin":"","legend":"\u003cp\u003eThe case #4; Disconcordance between manual and aPROMISE-assisted diagnoses:\u003c/p\u003e\n\u003cp\u003eMultiple liver metastases noted by manual diagnosis were dismissed in the report of aPROMISE\u003c/p\u003e","description":"","filename":"Fig2aPRO1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6360294/v1/4b90c481ba20213b53068170.jpg"},{"id":87219344,"identity":"ca69bc06-3faf-4a9b-8dee-681b4c4e2d60","added_by":"auto","created_at":"2025-07-21 16:04:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1028243,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6360294/v1/95442c8a-c29d-40e8-b372-b3045408c868.pdf"},{"id":80815335,"identity":"ce5bd462-76af-44d8-bcc1-6c872d37032b","added_by":"auto","created_at":"2025-04-17 10:57:26","extension":"jpg","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":324520,"visible":true,"origin":"","legend":"","description":"","filename":"Table1aPRO1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6360294/v1/cc9fdb090ba8d4356c0f5283.jpg"},{"id":80815333,"identity":"ca07a1f1-afef-4313-82e0-7ac9f09f1322","added_by":"auto","created_at":"2025-04-17 10:57:26","extension":"jpg","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":220173,"visible":true,"origin":"","legend":"","description":"","filename":"Table2aPRO1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6360294/v1/9ee95276b50208a8835e32cc.jpg"},{"id":80815337,"identity":"49e080a5-4592-49c1-a677-e2d57fb03ff2","added_by":"auto","created_at":"2025-04-17 10:57:26","extension":"jpg","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":179413,"visible":true,"origin":"","legend":"","description":"","filename":"Table3aPRO1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6360294/v1/86e3f154c8c06efe881bb871.jpg"},{"id":80815342,"identity":"40dc32ba-523f-4df2-8973-efbd34163a62","added_by":"auto","created_at":"2025-04-17 10:57:26","extension":"jpg","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":237142,"visible":true,"origin":"","legend":"","description":"","filename":"S.Table1aPRO1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6360294/v1/6d7aa5aae9034ecbe8d1c2e7.jpg"}],"financialInterests":"","formattedTitle":"Comparison of Diagnostic Performance Between Manual Diagnosis Following PROMISE V2 and aPROMISE Utilizing Ga/F-PSMA PET-CT","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eProstate-specific membrane antigen (PSMA) PET-CT has become an indispensable tool for prostate cancer (PCa) diagnosis across various clinical settings, including the detection of metastatic disease (e.g., oligometastasis) and clinically significant localized PCa [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Standardized diagnostic frameworks, such as the Prostate Cancer Molecular Imaging Standardized Evaluation (PROMISE) version 2.0, have been established to enhance reproducibility and generalizability in diagnostic reporting for PSMA PET-CT[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThere is growing interest in integrating artificial intelligence (AI) into medical imaging owing to its ability to automate and leverage radiomics, uncovering features potentially imperceptible to the human eye. Recently, automated PROMISE (aPROMISE, EXINI Diagnostics AB, Lund, Sweden), an AI-supported diagnostic software based on deep learning algorisms, has gained attention and received FDA approval for its ability to streamline PSMA PET-CT interpretation[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Preliminary studies have shown that aPROMISE achieves high concordance rates compared to manual interpretation, underscoring its potential to enhance clinical workflows[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. However, these studies have predominantly utilized F-PSMA PET-CT, leaving a critical gap in understanding the utility of aPROMISE for Ga-PSMA PET-CT, particularly in non-Western populations, such as Japanese patients.\u003c/p\u003e \u003cp\u003eTo address this gap, we investigated the diagnostic performance of aPROMISE in Japanese patients with metastatic PCa (mPCa) using both F- and Ga-PSMA PET-CT, aiming to validate its utility and explore potential tracer-specific variations in diagnostic outcomes.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Patients\u003c/h2\u003e \u003cp\u003eWe retrospectively selected a total of 21 PSMA PET-CT images from Japanese 21 patients with mPCa. The patients underwent [68Ga]Ga-PSMA-11 PET-CT(n\u0026thinsp;=\u0026thinsp;12) or [18F]PSMA-1007 PET-CT (n\u0026thinsp;=\u0026thinsp;9) between January 2020 and April 2024. Patients provided written informed consent before examination. This study was approved by The Jikei University Institutional Review Board (36\u0026thinsp;\u0026minus;\u0026thinsp;003(12102)).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. PROMISE V2 and aPROMISE\u003c/h2\u003e \u003cp\u003eThe PROMISE criteria have been proposed as a framework for whole-body molecular imaging-based TNM staging, denoted miTNM staging, to describe the prostate cancer disease extent on PSMA PET-CT[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. The second version of PROMISE framework (PROMISE V2), which integrates an updated miTNM staging system, modifying assessment of local disease and PSMA-expression score (quantify and standardize the level of PSMA in prostate cancer tissues) for clinical routine [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] aPROMISE is an artificial intelligence-enabled medical device software for quantitative analysis and standardized reporting of PSMA PET/CT images in prostate cancer following the PROMISE criteria [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. aPROMISE was developed via deep learning algorithms trained on over 3,000 PSMA images to perform automated segmentation, detection, and quantification of PSMA PET/CT lesions, implementing and extending the PROMISE criteria[\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] .\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Diagnostic and evaluation methods\u003c/h2\u003e \u003cp\u003eTo validate the diagnostic performance of aPROMISE, we compared the manual diagnosis with aPROMISE-assisted diagnosis as the way below. A single well-experienced nuclear radiologist diagnosed molecular imaging TNM (miTNM) staging and details of each image following the PROMISE V2 reporting sheets [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]: 1) Manual diagnosis following PROMISE V2 and 2) aPROMISE-assisted diagnosis. These diagnoses were randomly and separately conducted with decent interval duration and blinded patient information on each image.\u003c/p\u003e \u003cp\u003eWe compared the diagnostic time and correspondence rates of miTNM diagnosis and number of PSMA-positive lesions between the two diagnostic methods. Diagnostic time was measured by manual measurement by the radiologist from the start to the end of filling the reporting sheets. Concordance rates were evaluated under the definition that manual diagnosis by an experienced nuclear radiologist was defined as \u0026ldquo;true\u0026rdquo;.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4. Statistical analysis\u003c/h2\u003e \u003cp\u003eThe diagnostic time (in seconds) for aPROMISE-assisted diagnosis and manual diagnosis was represented and compared using the median and interquartile range. Diagnostic concordance between the two diagnostic methods was evaluated using a 2\u0026times;2 contingency table. Sensitivity and specificity were calculated based on this table. All statistical analyses were performed using R version 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria), and a p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.1. Overview of manual and aPROMISE-assisted diagnosis\u003c/h2\u003e \u003cp\u003eThe summary of manual diagnosis and aPROMISE-assisted diagnosis are shown in \u003cb\u003eTable\u0026nbsp;1\u003c/b\u003e. In miT staging, seven patients (33%) were detected with PSMA-positive lesions in manual diagnosis, while nine patients (43%) were identified using aPROMISE-assisted diagnosis. In Ga-PSMA PET-CT, five patients were consistently diagnosed as PSMA-positive using both diagnostic methods. However, in F-PSMA PET-CT, aPROMISE-assisted diagnosis identified two additional patients with PSMA-positive lesions (22% vs. 44%). The distribution of PSMA expression scores for the hottest lesion was comparable between the two diagnostic methods. Regarding the number of positive lesions, aPROMISE-assisted diagnosis identified more patients with two PSMA-positive lesions (14% vs. 56%).\u003c/p\u003e \u003cp\u003eIn miN staging, eight patients (38%), including four for N1 and four for N2, were detected with PSMA-positive lesions in manual diagnosis. On the other hand, seven patients (33%), including four for N1 and three for N2, were detected using aPROMISE-assisted diagnosis. The distribution of PSMA expression scores for the hottest lesion and the number of PSMA-positive lesions were comparable between the two diagnostic methods.\u003c/p\u003e \u003cp\u003eIn miM1a staging, ten patients (48%) were detected with PSMA-positive lesions in manual diagnosis, while 12 patients (57%) were identified using aPROMISE-assisted diagnosis. aPROMISE-assisted diagnosis was more likely to detect higher PSMA expression scores for the hottest lesion (30% with a score of 3 in manual diagnosis vs. 58% with a score of 3 in aPROMISE-assisted diagnosis) and identified a greater number of patients with multiple PSMA-positive lesions (50% vs. 67%).\u003c/p\u003e \u003cp\u003eIn miM1b staging, 17 patients (81%) were identified with PSMA-positive lesions, irrespective of diagnostic methods. While the distribution of PSMA expression scores for the hottest lesion showed a slight difference, with a greater number of score 3 lesions detected in aPROMISE-assisted diagnosis (71% vs. 88%), the distribution of the number of PSMA-positive lesions was entirely similar.\u003c/p\u003e \u003cp\u003eIn miM1c staging, aPROMISE-assisted diagnosis showed lower detection rates (n\u0026thinsp;=\u0026thinsp;8, 38% in manual diagnosis vs. n\u0026thinsp;=\u0026thinsp;5, 24% in aPROMISE-assisted diagnosis). aPROMISE-assisted diagnosis missed three liver lesions and two other metastases. Detailed information and case presentations on the discordance in detecting visceral metastases between the two methods are provided in a separate section, \u003cem\u003e3.4. Diagnostic disconcordance in visceral metastases\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eBeyond miTNM staging, aPROMISE-assisted diagnosis provided detailed evaluations of PSMA PET-CT images (\u003cb\u003eSupplementary Tables\u0026nbsp;1\u003c/b\u003e). Notably, only aPROMISE-assisted diagnosis offered precise information on the number of metastases, particularly in cases with numerous PSMA-positive lesions. Furthermore, aPROMISE routinely reported detailed metrics such as the total number and volume of PSMA-positive lesions, as well as the mean and maximum values of standardized uptake values (SUV) (\u003cb\u003eSupplementary Table\u0026nbsp;1\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.2. Diagnostic time\u003c/h2\u003e \u003cp\u003eThe comparison of diagnostic time measured by the radiologist between the two diagnostic methods is shown in \u003cb\u003eFig.\u0026nbsp;1\u003c/b\u003e. As shown in \u003cb\u003eFig.\u0026nbsp;1A\u003c/b\u003e, median diagnostic time for aPROMISE-assisted diagnosis was significantly shorter (427 seconds [IQR: 370\u0026ndash;834]) compared to manual diagnosis (1114 seconds [IQR; 922\u0026ndash;1291], p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In 18 patients using F-PSMA PET-CT, median diagnostic time of aPROMISE-assisted and manual diagnosis were 704 seconds (IQR: 493\u0026ndash;1019) and 1114 seconds (IQR: 679\u0026ndash;1123), respectively (p\u0026thinsp;=\u0026thinsp;0.5, \u003cb\u003eFig.\u0026nbsp;1B\u003c/b\u003e). In 24 patients using Ga-PSMA PET-CT, median diagnostic time of aPROMISE-assisted and manual diagnosis were 390 seconds (IQR: 350.25-455.25) and 1204 seconds (IQR: 984.25-1370.5), respectively (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, \u003cb\u003eFig.\u0026nbsp;1C\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.3. Concordance rates\u003c/h2\u003e \u003cp\u003eConcordance rates were evaluated based on the aforementioned definition, with manual diagnosis by an experienced nuclear radiologist considered as the \u0026ldquo;true\u0026rdquo; reference. \u003cb\u003eTable\u0026nbsp;2\u003c/b\u003e summarizes the sensitivity and specificity of aPROMISE-assisted diagnosis compared to manual diagnosis.\u003c/p\u003e \u003cp\u003eIn miT staging, sensitivity and specificity were 100% and 86%, respectively, in the overall cohort. In miN staging, sensitivity and specificity were 88% and 100%, respectively, for the diagnosis of node positivity. However, due to discordance in identifying N1 and N2 diseases, aPROMISE-assisted diagnosis showed low concordance rates for N1 disease (25%).\u003c/p\u003e \u003cp\u003eIn miM1a staging, sensitivity and specificity were 100% and 82%, respectively, in the overall cohort. Notably, in miM1b staging, sensitivity and specificity were both 100%, regardless of the radioligand used for PSMA PET-CT. In miM1c staging, despite discordance in detecting visceral metastatic lesions, sensitivity and specificity were 63% and 100%, respectively.\u003c/p\u003e \u003cp\u003eIn summary of the differential diagnostic performance across radioligand types, sensitivity and specificity were both 100% in miT, miN, and miM1b staging with Ga-PSMA PET-CT, as well as in miM1a and miM1b staging with F-PSMA PET-CT.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.4. Diagnostic disconcordance in visceral metastases\u003c/h2\u003e \u003cp\u003eThe summary of eight patients who had visceral metastasis using manual diagnosis is shown in \u003cb\u003eTable\u0026nbsp;3\u003c/b\u003e. We identified only one patient (case #10) who completely matched visceral metastasis between manual and aPROMISE-assisted diagnoses. In total, aPROMISE missed five metastatic lesions (2 adrenal and 3 liver metastases), while detected two lung metastases. For example, in the case #4, multiple liver metastases were detected by manual diagnosis; however, no liver metastases were reported in aPROMISE-assisted diagnosis. (Fig.\u0026nbsp;2)\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eIn this study, we compared the diagnostic times and concordance rates of manual and AI-assisted diagnostics, as well as the differential diagnostic performance of Ga-PSMA PET and F-PSMA PET. Our study confirmed the diagnostic utility of aPROMISE and revealed several important insights. First, aPROMISE-assisted diagnosis significantly reduced diagnostic time compared to manual diagnosis. Second, aPROMISE-assisted diagnosis demonstrated high diagnostic performance, particularly in detecting bone metastases, regardless of the type of radioligand used. Third, however, discrepancies were observed between the two diagnostic methods in detecting visceral metastatic lesions. Last but not least, while several previous studies have highlighted the utility of AI-assisted diagnosis in F-PSMA PET-CT imaging, no similar reports have been published for Ga-PSMA PET-CT [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Despite the preliminary nature of this study, it is the first to demonstrate the efficacy of aPROMISE even when using Ga-PSMA PET-CT.\u003c/p\u003e \u003cp\u003eThe application of AI-assisted diagnosis has the potential to reduce the burden on nuclear physician and radiologists while ensuring consistency in quantification for PSMA PET-CT diagnosis [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Nickols et al. demonstrated that aPROMISE-assisted diagnosis led to low inter-reader variability between two nuclear radiologists in PCa staging[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. This study comprehensively showed a high level of Cohen pairwise k-agreement between the two radiologists (ranging from 0.77 to 0.90) in detecting regional lymph node and/or distant bone metastatic disease[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. In the current study, we assessed another aspect of reducing the diagnostic burden through AI-assisted diagnosis, focusing on its ability to theoretically reduce diagnostic time. Our findings are the first to objectively demonstrate that aPROMISE-assisted diagnosis significantly shortens diagnostic time compared to manual diagnosis. Taken together, aPROMISE-assisted diagnosis addresses key limitations in PSMA PET-CT diagnosis, such as potential inter-reader variability and time-consuming nature of manual interpretation.\u003c/p\u003e \u003cp\u003eRegarding the diagnostic accuracy of AI-assisted PSMA PET-CT diagnosis, a recent systematic review reported relatively high sensitivity (ranging from 62 to 97%) and accuracy (with an area under the curve up to 98%) for AI in detecting various types of metastatic diseases[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Specifically for the diagnostic capability of aPROMISE, Johnnson et al. demonstrated the high accuracy of segmentation achieved by aPROMSIE with excellent Dice scores compared to manual segmentation by experienced nuclear radiologists[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Furthermore, aPROMISE has shown impressive sensitivity rates for detecting potential lesions, achieving 91.5% for regional lymph nodes and 86.7% for bone lesions in metastatic PCa patients[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. In our study, aPROMISE platform demonstrated exceptional diagnostic accuracy, achieving 100% sensitivity across miT, M1a, and M1b staging. Notably, for M1b staging, aPROMISE achieved 100% sensitivity and specificity, regardless of types of radioisotopes used. This high sensitivity underscores its clinical applicability for detecting significant pathological findings in mCRPC patients.\u003c/p\u003e \u003cp\u003eOn the other hand, aPROMISE missed some visceral metastases. To date, there has been no reports specifically focusing on the diagnostic utility of aPROMISE for detecting visceral metastases, likely due to the relatively low number of patients with visceral metastasis[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. In our study, aPROMISE missed three liver and two adrenal metastases were missed. One possible explanation for missing liver metastases is the normal intake of PSMA in the liver, which may obscure metastatic lesions[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Despite its automation capabilities, aPROMISE still requires physicians to review and confirm the identification of lesions, particularly for visceral metastasis. Our findings suggest that integrating the AI-assisted diagnosis and manual interpretation is still important for more robust judgement for visceral metastases.\u003c/p\u003e \u003cp\u003eIn the argument of diagnostic performance comparisons between Ga-PSMA PET-CT and F-PSMA PET-CT, F-PSMA PET-CT has been recognized for its potential advantages in pelvic imaging owing to minimally excretion into the urinary tract, a longer half-life, production in larger quantities, and lower positron energy, which may improve spatial resolution[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. However, a recent meta-analysis comparing the diagnostic impact of Ga-labeled PSMA and F-labeled PSMA imaging concluded that there is insufficient evidence to differentiate the radiotracers based on their clinical impact Given the liver-dominant excretion of [18F] PSMA-1007, it was observed to have higher liver uptake but lower urinary tract uptake compared to [68Ga] Ga-PSMA-11[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. This difference could potentially affect AI-assisted diagnosis depending on the location of metastases. Due to the pilot nature of our study, we did not statistically compare the concordance rates between the radiotracers. However, our study demonstrated the comparable diagnostic concordance rates with aPROMISE-assisted diagnosis even when using Ga-PSMA PET-CT. To elucidate the potential impact of differential diagnostic performance based on tracer types, further well-designed studies with a sufficient number of patients are urgently needed.\u003c/p\u003e \u003cp\u003eIn addition to its shorter diagnostic time and high correspondence rates, aPROMISE provides a detailed and comprehensive assessment of disease burden and metrics in PSMA PET-CT imaging, which are critical for clinical decision-making in PSMA-radioligand therapy (RLT), such as 177-Lu PSMA RLT[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Of note, aPROMISE successfully reported the SUV mean value integrating all metastatic lesions in every case (\u003cb\u003eSupplementary Table\u0026nbsp;1\u003c/b\u003e). The SUV mean value has been recognized as a predictive factor for oncologic outcomes in patients with mCRPC undergoing PSMA-RLT[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Furthermore, post-hoc analyses of the TheraP and VISION trials demonstrated that whole-body SUV mean predicts treatment response in patients receiving 177Lu-PSMA-617 RLT[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. These findings highlight the potential of specific software, such as aPROMISE, to support personalized medicine for patients with mCRPC, emphasizing its utility in routine clinical practice.\u003c/p\u003e \u003cp\u003eOur study had several limitations. First, as a retrospective study with a limited number of patients, it restricted the ability to conduct robust statistical analysis. Second, we did not randomly select images from different tracers for PSMA-PET CT, resulting in imbalanced patient and disease demographics between the groups. Third, only one radiologist performed both the manual and aPROMISE-assisted diagnoses. Despite randomizing the order of PSMA-PET imaging and ensuring sufficient time intervals between diagnoses of the same patient, this study design might have influenced the diagnostic outcomes.\u003c/p\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eaPROMISE-assisted diagnosis significantly reduced diagnostic time compared to manual diagnosis. aPROMISE-assisted diagnosis demonstrated high diagnostic accuracy, achieving 100% sensitivity for miT, M1a, M1b staging. Particularly for M1b staging, aPROMISE-assisted diagnosis achieved 100% sensitivity and specificity, regardless of the type of radioisotope used. However, aPROMISE missed several visceral metastatic lesions, specifically liver and adrenal metastasis. While our analysis confirmed the diagnostic utility of aPROMISE in Japanese patients with mPCa including both Ga-PSMA PET-CT and F-PSMA PET-CT, integrating AI-assisted diagnosis with manual interpretation should be considered for more robust judgement for visceral metastases. Further well-designed studies with a sufficient number of patients are necessary to validate our findings and clarify the differential diagnostic performance between the radioisotope types.\u003c/p\u003e "},{"header":"Abbreviations","content":"\u003cp\u003ePSMA: Prostate-specific membrane antigen\u003c/p\u003e\n\u003cp\u003ePCa: Prostate cancer\u003c/p\u003e\n\u003cp\u003emPCa: metastatic PCa\u003c/p\u003e\n\u003cp\u003ePROMISE: Prostate Cancer Molecular Imaging Standardized Evaluation\u003c/p\u003e\n\u003cp\u003eAI: Artificial intelligence\u003c/p\u003e\n\u003cp\u003eaPROMISE; automated PROMISE\u003c/p\u003e\n\u003cp\u003emiTNM; molecular imaging TNM\u003c/p\u003e\n\u003cp\u003eSUV: Standardized uptake values\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eConflict of interest \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no conflict of interest\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResearch involving human participants and/or animals\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData collection adhere to principles of the Declaration of Helsinki. This study was approved by The Jikei University Institutional Review Board (36-003(12102)).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInformed consent\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll patients signed an informed consent agreeing to share their own anonymous information for future studies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYE and TY: Project development, Data collection, Data analysis, Manuscript writing. KM and HK: Project development, Manuscript writing. CO, KW and NL : Data collection; Manuscript writing. AO and TK: Project development, Data analysis, Manuscript writing\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eHofman MS, Emmett L, Sandhu S, et al. Overall survival with [177Lu]Lu-PSMA-617 versus cabazitaxel in metastatic castration-resistant prostate cancer (TheraP): secondary outcomes of a randomised, open-label, phase 2 trial. Lancet Oncol. 2024;25:99\u0026ndash;107.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSartor O, de Bono J, Chi KN, et al. Lutetium-177-PSMA-617 for metastatic castration-resistant prostate cancer. N Engl J Med. 2021;385:1091\u0026ndash;103.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorris MJ, Castellano D, Herrmann K, et al. 177Lu-PSMA-617 versus a change of androgen receptor pathway inhibitor therapy for taxane-naive patients with progressive metastatic castration-resistant prostate cancer (PSMAfore): a phase 3, randomised, controlled trial. Lancet. 2024;404:1227\u0026ndash;39.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEiber M, Herrmann K, Calais J, et al. Prostate cancer molecular imaging standardized evaluation (PROMISE): Proposed miTNM classification for the interpretation of PSMA-ligand PET/CT. J Nucl Med. 2018;59:469\u0026ndash;78.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCalais J, Eiber M, Iagaru A, et al. Prospectively planned and independent validation of aPROMISE in a phase III CONDOR study for rapid lesion detection and standardized quantitative evaluation for 18F-DCFPyL (PSMA) imaging in prostate cancer. J Nucl Med. 2022;63:2496.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNickols N, Anand A, Johnsson K, et al. APROMISE: A novel automated PROMISE platform to standardize evaluation of tumor burden in 18F-DCFPyL images of veterans with prostate cancer. J Nucl Med. 2022;63:233\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarc\u0026iacute;a Vicente AM, Lucas Lucas C, P\u0026eacute;rez-Beteta J, et al. Analytical performance validation of aPROMISE platform for prostate tumor burden, index and dominant tumor assessment with 18F-DCFPyL PET/CT. A pilot study. Sci Rep. 2024;14:3001.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeifert R, Emmett L, Rowe SP, et al. Second version of the prostate cancer molecular imaging standardized evaluation framework including response evaluation for clinical trials (PROMISE V2). Eur Urol. 2023;83:405\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLindgren Belal S, Frantz S, Minarik D, et al. Applications of artificial intelligence in PSMA PET/CT for prostate cancer imaging. Semin Nucl Med. 2024;54:141\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu J, Cundy TP, Woon DTS, Lawrentschuk N. A systematic review on artificial intelligence evaluating metastatic prostatic cancer and lymph nodes on PSMA PET scans. Cancers (Basel). 2024;16. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/cancers16030486\u003c/span\u003e\u003cspan address=\"10.3390/cancers16030486\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDamjanovic J, Janssen J-C, Prasad V, et al. 68Ga-PSMA-PET/CT for the evaluation of liver metastases in patients with prostate cancer. Cancer Imaging. 2019;19:37.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRauscher I, Maurer T, Fendler WP, et al. (68)Ga-PSMA ligand PET/CT in patients with prostate cancer: How we review and report. Cancer Imaging. 2016;16:14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiesel FL, Hadaschik B, Cardinale J, et al. F-18 labelled PSMA-1007: biodistribution, radiation dosimetry and histopathological validation of tumor lesions in prostate cancer patients. Eur J Nucl Med Mol Imaging. 2017;44:678\u0026ndash;88.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eButeau JP, Martin AJ, Emmett L, et al. PSMA and FDG-PET as predictive and prognostic biomarkers in patients given [177Lu]Lu-PSMA-617 versus cabazitaxel for metastatic castration-resistant prostate cancer (TheraP): a biomarker analysis from a randomised, open-label, phase 2 trial. Lancet Oncol. 2022;23:1389\u0026ndash;97.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGafita A, Calais J, Grogan TR, et al. Nomograms to predict outcomes after 177Lu-PSMA therapy in men with metastatic castration-resistant prostate cancer: an international, multicentre, retrospective study. Lancet Oncol. 2021;22:1115\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang S, Ong S, McKenzie D, et al. Comparison of 18F-based PSMA radiotracers with [68Ga]Ga-PSMA-11 in PET/CT imaging of prostate cancer-a systematic review and meta-analysis. Prostate Cancer Prostatic Dis. 2024;27:654\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuo PH, Morris MJ, Hesterman J, et al. Quantitative 68Ga-PSMA-11 PET and clinical outcomes in metastatic castration-resistant prostate cancer following 177Lu-PSMA-617 (VISION trial). Radiology. 2024;312:e233460.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables 1 to 3 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"annals-of-nuclear-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"anme","sideBox":"Learn more about [Annals of Nuclear Medicine](http://link.springer.com/journal/12149)","snPcode":"12149","submissionUrl":"https://www.editorialmanager.com/anme/default2.aspx","title":"Annals of Nuclear Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"PSMA PET-CT, metastatic prostate cancer, artificial intelligence, diagnosis, PROMISE V2, aPROMISE","lastPublishedDoi":"10.21203/rs.3.rs-6360294/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6360294/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackgrounds\u003c/h2\u003e \u003cp\u003eAutomated PROMISE (aPROMISE), which is an artificial intelligence-supported software for prostate-specific membrane antigen (PSMA) PET-CT based on PROMISE V2, has demonstrated diagnostic utility with better correspondence rates compared to manual diagnosis. However, previous studies have consistently utilized F-PSMA PET-CT. Therefore, we investigated the diagnostic utility of aPROMISE using both F- and Ga-PSMA PET-CT of Japanese patients with metastatic prostate cancer (mPCa).\u003c/p\u003e\u003ch2\u003eMaterials and Methods\u003c/h2\u003e \u003cp\u003eWe retrospectively evaluated 21 PSMA PET-CT images (Ga-PSMA PET-CT: n\u0026thinsp;=\u0026thinsp;12, F-PSMA PET-CT: n\u0026thinsp;=\u0026thinsp;9) from 21 patients with mPCa. A single, well-experienced nuclear radiologist performed manual diagnosis following PROMISE V2 and subsequently performed aPROMISE-assisted diagnosis to assess miTNM and details of metastatic sites. We compared the diagnostic time and correspondence rates of miTNM diagnosis between manual and aPROMISE-assisted diagnoses. Additionally, we investigated the differences in diagnostic performance between the two radioisotopes.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eaPROMISE-assisted diagnosis was significantly associated with shorter median diagnostic time compared to manual diagnosis (427 seconds [IQR: 370\u0026ndash;834] vs. 1,114 seconds [IQR: 922\u0026ndash;1291], p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). The time reduction with aPROMISE-assisted diagnosis was particularly notable when using Ga-PSMA PET-CT. aPROMISE had high diagnostic accuracy with 100% sensitivity for miT, M1a, and M1b stages. Notably, for M1b stages, aPROMISE achieved 100% sensitivity and specificity, regardless of the type of radioisotope used. However, aPROMISE missed five visceral metastases (2 adrenal and 3 liver), resulting in lower sensitivity for miM1c stage (63%). In addition to detecting metastatic sites, aPROMISE successfully provided detailed metrics, including the number of metastatic lesions, total metastatic volume, and SUV mean.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eaPROMISE-assisted diagnosis significantly reduces diagnostic time and achieves high accuracy compared to manual diagnosis, regardless of the type of radioisotopes used. While aPROMISE successfully detects bone metastases, its limitations in detecting visceral metastases need to be addressed. This study supports the utility of aPROMISE in Japanese patients with mPCa and underscore the need for further validation in larger cohorts.\u003c/p\u003e","manuscriptTitle":"Comparison of Diagnostic Performance Between Manual Diagnosis Following PROMISE V2 and aPROMISE Utilizing Ga/F-PSMA PET-CT","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-17 10:57:21","doi":"10.21203/rs.3.rs-6360294/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2025-04-12T05:48:40+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-11T06:30:24+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-04-02T23:57:34+00:00","index":"","fulltext":""},{"type":"submitted","content":"Annals of Nuclear Medicine","date":"2025-04-02T06:36:53+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"annals-of-nuclear-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"anme","sideBox":"Learn more about [Annals of Nuclear Medicine](http://link.springer.com/journal/12149)","snPcode":"12149","submissionUrl":"https://www.editorialmanager.com/anme/default2.aspx","title":"Annals of Nuclear Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"72e604b9-9339-4db9-b2be-14300309d4be","owner":[],"postedDate":"April 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-07-21T16:01:06+00:00","versionOfRecord":{"articleIdentity":"rs-6360294","link":"https://doi.org/10.1007/s12149-025-02086-9","journal":{"identity":"annals-of-nuclear-medicine","isVorOnly":false,"title":"Annals of Nuclear Medicine"},"publishedOn":"2025-07-15 15:57:27","publishedOnDateReadable":"July 15th, 2025"},"versionCreatedAt":"2025-04-17 10:57:21","video":"","vorDoi":"10.1007/s12149-025-02086-9","vorDoiUrl":"https://doi.org/10.1007/s12149-025-02086-9","workflowStages":[]},"version":"v1","identity":"rs-6360294","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6360294","identity":"rs-6360294","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}