ESMO Real World Data and Digital Oncology最新文献

{"title":"Artificial intelligence for clinical trial design, conduct, and analysis: a narrative review","authors":"D.G. Knapen,&nbsp;M. van Kruchten,&nbsp;D.J.A. de Groot,&nbsp;K.E. Broekman,&nbsp;R.S.N. Fehrmann","doi":"10.1016/j.esmorw.2026.100682","DOIUrl":"10.1016/j.esmorw.2026.100682","url":null,"abstract":"<div><div>Drug development in oncology is facing rising complexity, prolonged timelines, and increasing costs, with many trials failing to reach completion or secure regulatory approval for the investigated treatment. Against this backdrop, artificial intelligence (AI) and real-world data have emerged as promising tools to improve the efficiency and quality of clinical research. This narrative review explores how AI can support clinical trial design, conduct, and analysis. AI-driven methods can streamline information gathering, optimize eligibility criteria, and predict trial success, thereby reducing costly failures. Patient recruitment and retention, often the most challenging aspects of oncology trials, may benefit from AI-supported matching algorithms, digital health technologies, and personalized interactions. During trial conduct, natural language processing and sensor-based monitoring offer opportunities to reduce administrative burden and capture real-world, patient-centred outcomes. For trial analyses, AI enhances radiology, digital pathology, pharmacometrics, and multimodal modelling, enabling more accurate prognostic and predictive insights. In addition, AI-based simulations, such as digital twins and <em>in silico</em> trials, hold potential to complement conventional trial designs. However, despite rapid technical advances, evidence supporting clinical utility remains limited. Most applications are tested retrospectively or in single-centre settings, limiting generalizability. Regulatory frameworks, including the European Union Artificial Intelligence Act and emerging ESMO standards for AI-based biomarkers, emphasize transparency, reproducibility, and prospective validation. Ultimately, the successful integration of AI into oncology trials will depend less on technical capacity than on rigorous evaluation, harmonized regulation, and adoption of shared quality standards.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100682"},"PeriodicalIF":0.0,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146077771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ovarian cancer recurrence prediction: comparing confirmatory to real-world predictors with machine learning","authors":"D. Katsimpokis ,&nbsp;A.E.C. van Odenhoven ,&nbsp;M.A.J.M. van Erp ,&nbsp;H.H.B. Wenzel ,&nbsp;M.A. van der Aa ,&nbsp;M.M.H. van Swieten ,&nbsp;H.P.M. Smedts ,&nbsp;J.M.J. Piek","doi":"10.1016/j.esmorw.2025.100666","DOIUrl":"10.1016/j.esmorw.2025.100666","url":null,"abstract":"<div><h3>Background</h3><div>Ovarian cancer is one of the deadliest cancers in women, frequently diagnosed at an advanced stage, with a 5-year survival rate of 17%-28% in advanced-stage (International Federation of Gynecology and Obstetrics IIB-IV) disease. Machine learning (ML) may provide a better tool for survival prognosis than traditional methods and could provide insight into predictive factors. This study focuses on advanced-stage ovarian cancer and contrasts expert-derived predictive factors with data-driven ones from the Netherlands Cancer Registry (NCR) to predict progression-free survival.</div></div><div><h3>Materials and methods</h3><div>A Delphi questionnaire was conducted to identify 14 predictive factors which were included in the final analysis. ML models (regularized Cox regression, random survival forests, and XGBoost) were used to compare the Delphi expert-based set of variables with a real-world data (RWD) variable set derived from the NCR. A non-regularized Cox model was used as the benchmark.</div></div><div><h3>Results</h3><div>While regularized Cox models with the RWD variable set outperformed the traditional Cox regression with the Delphi variables (c-index: 0.70 versus 0.64, respectively), XGBoost showed the best performance overall (c-index: 0.75). The most predictive factors for recurrence, not identified by Delphi, were surgery type and debulking results, post-operative chemotherapy administration, number of platinum cycles, and socioeconomic status.</div></div><div><h3>Conclusions</h3><div>Our results highlight that ML algorithms have higher predictive power compared with the traditional Cox regression. Moreover, RWD from a cancer registry identified more predictive variables than a panel of experts. Overall, these results have important implications for artificial intelligence (AI)-assisted clinical prognosis and provide insight into the differences between AI-driven and expert-based decision making in survival prediction.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100666"},"PeriodicalIF":0.0,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926165","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Histotype-specific incidence and survival of urothelial carcinoma—an analysis of the German North Rhine-Westphalia Cancer Registry","authors":"C. Darr ,&nbsp;L. Möller ,&nbsp;A. Szentkiralyi ,&nbsp;K. Claassen ,&nbsp;N. Schürger ,&nbsp;H. Reis ,&nbsp;T. Hilser ,&nbsp;A. Stang ,&nbsp;B.A. Hadaschik ,&nbsp;H. Kajüter ,&nbsp;V. Grünwald","doi":"10.1016/j.esmorw.2025.100678","DOIUrl":"10.1016/j.esmorw.2025.100678","url":null,"abstract":"<div><h3>Background</h3><div>The aim of this study was to evaluate the incidence and survival of urothelial carcinoma (UC) and non-urothelial tumor types. The primary objective was to define the incidence and survival of pure UC and non-UC tumor types.</div></div><div><h3>Materials and methods</h3><div>Malignant invasive urinary cancers of the urothelial tract diagnosed between 2008 and 2022 were identified via the North Rhine-Westphalia Cancer Registry and classified according to the 2016 World Health Organization (WHO) classification (fourth edition). Evaluation focused on pure tumor types: UC, papillary invasive UC (PIUC), squamous-cell carcinoma (SCC), adenocarcinoma (ADC), rarer tumor types and mixed histologies grouped under other specific tumor types (OSTT), and unspecified tumor types (UTT). The primary outcomes were age-standardized incidence rate with estimated annual percentage changes, and relative survival.</div></div><div><h3>Results</h3><div>UC and PIUC were the most common histology types, accounting for 57.9% and 26.2% (<em>n</em> = 73 751), respectively. The age-standardized incidence of PIUC and UTT decreased over time in both sexes, while OSTT incidence increased over time. Incidence rate declined among men with SCC and among women with ADC. Relative survival was notably poor for SCC (32.3%) and OSTT (23.8%), in contrast to a more favorable outcome for PIUC (72.0%). Decreasing survival with advancing T-stages was noted, particularly for UC, PIUC, ADC, and UTT; T1 in PIUC showed the best survival (81.0%) and T3-4 in OSTT had the poorest survival (16.4%).</div></div><div><h3>Conclusion</h3><div>Relative survival was most favorable for PIUC, UC, and ADC. In contrast, SSC and OSTT exhibited poorer survival outcomes, highlighting a pressing medical need for enhanced treatment options for these subgroups.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100678"},"PeriodicalIF":0.0,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146037817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Feasibility of Wave Health to monitor electronic patient-reported outcomes (ePROs) in prostate cancer patients","authors":"B. Gomez Mediavilla ,&nbsp;B. Congregado ,&nbsp;E. Linares Espinos ,&nbsp;I. Rivero Belenchon ,&nbsp;L. Basterretxea ,&nbsp;G. Sancho ,&nbsp;N. Lainez ,&nbsp;D. Buchser ,&nbsp;T. Leturgez ,&nbsp;A.A. Schmitz ,&nbsp;J. Pozueta ,&nbsp;I. Gurruchaga ,&nbsp;D. Cacho Lavin ,&nbsp;B. Esteban ,&nbsp;A. Hernandez ,&nbsp;M. Lashey ,&nbsp;F. Sanguedolce ,&nbsp;I. Duran","doi":"10.1016/j.esmorw.2025.100177","DOIUrl":"10.1016/j.esmorw.2025.100177","url":null,"abstract":"<div><h3>Background</h3><div>Benefits of electronic patient-reported outcomes (ePROs) are well established, but concerns exist around their applicability in certain cancer patient populations. This multicenter study aimed to evaluate the feasibility and compliance of a novel platform, Wave Health, to automate the collection of ePROs among patients undergoing systemic treatment for prostate cancer.</div></div><div><h3>Patients and methods</h3><div>The study was conducted across eight Spanish hospitals where prostate cancer patients able to use smartphones were asked to download the Wave Health app and fill out different ePRO questionnaires weekly for 12 weeks. The primary outcome was compliance with questionnaire completion.</div></div><div><h3>Results</h3><div>Eighty-seven patients were enrolled (48% aged &gt;70 years; 48% &gt;high school educated). The compliance rate using the app was 93.3% to complete 19 scheduled ePRO questionnaires during their participation, and the acceptance rate was 94.3%. Most patients described the app as intuitive and completed most ePROs (&gt;80%) ‘without help’, even though 28.7% reported a lack of technology confidence. The mean Cancer Health Literacy Test-30-Spanish (CHLT-30-DKspa) score increased from 22.3 at baseline to 23.2 at week 12. Fatigue, weakness, and poor sleep were the most common symptoms reported by patients. Most health care providers (97.6%) rated the platform as ‘very useful,’ ‘useful,’ or ‘somewhat useful.’</div></div><div><h3>Conclusion</h3><div>This study demonstrated that ePRO collection via the Wave Health platform is feasible, with high rates of compliance, acceptance, and satisfaction among prostate cancer patients under systemic treatment regardless of age, education, or level of tech confidence, and that the app is perceived as useful among health care providers.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"10 ","pages":"Article 100177"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145020671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clinical outcomes of systemic anticancer therapies in solid cancer patients with liver and kidney transplant: an observational cross-sectional study","authors":"E. Chen ,&nbsp;N. Belkaid ,&nbsp;C. Duvoux ,&nbsp;D. Sahali ,&nbsp;P. Grimbert ,&nbsp;M.-B. Matignon ,&nbsp;A. Hulin ,&nbsp;S. Babai ,&nbsp;C. Chouaïd ,&nbsp;J.B. Assié ,&nbsp;M. Carvalho ,&nbsp;C. Tournigand ,&nbsp;E. Kempf","doi":"10.1016/j.esmorw.2025.100185","DOIUrl":"10.1016/j.esmorw.2025.100185","url":null,"abstract":"<div><h3>Background</h3><div>Solid cancer patients with liver transplant (LT) or kidney transplant (KT) receiving systemic anticancer therapies (SACT) are likely to present an increased risk of treatment-related infections. We aimed to describe their clinical outcomes.</div></div><div><h3>Patients and methods</h3><div>This retrospective study included SACT-treated patients from two university medical oncology centers between 2000 and 2023, identified through automated extraction and manual record review. We excluded patients treated with immune checkpoint inhibitors. We manually collected immunosuppressant and SACT administration, patient SACT grade 3 to 5 toxicities, cause of death, and overall survival.</div></div><div><h3>Results</h3><div>Fifty-three patients were included: 39 (74%) men, 33 (62%) LT, and 20 (38%) KT; median age was 62 years (interquartile range 32-82 years); 19 (58%) and 15 (75%) LT and KT patients, respectively, had stage IV diseases. Primary cancer was liver (<em>n</em> = 16; 45%) and digestive (<em>n</em> = 10; 30%) for LT patients; digestive (<em>n</em> = 7; 35%), kidney, breast, and lung (all three <em>n</em> = 3; 15%) for KT patients. From the time before cancer diagnosis to SACT administration, the immunosuppressant regimen was changed in both subgroups. Overall, 29 (88%) LT patients and 17 (85%) KT patients were exposed at least once to standard cytotoxic drugs, and 8 (24%) LT patients and 3 (15%) KT patients to tyrosine kinase inhibitors. Thirty-three patients had grade 3-4 toxicities, 52% LT patients and 30% KT patients. Twenty-four patients had grade 5 toxicities (79% LT and 86% KT patients treatment-related infections, respectively). One-year overall survival rates were 52% for LT patients and 50% for KT patients. After treatment-related infections, death was due to tumor progression for 9 LT patients (32%) and 5 (42%) KT patients, respectively.</div></div><div><h3>Conclusions</h3><div>Treatment-related infections drive the prognosis of LT and KT patients undergoing SACT.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"10 ","pages":"Article 100185"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145220928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prognostic factors and outcomes of patients with advanced non-small-cell lung cancer treated with osimertinib: a retrospective database study☆","authors":"M. Pérol ,&nbsp;C. Chouaid ,&nbsp;P. Macouillard ,&nbsp;L. Bosquet ,&nbsp;C. Knott ,&nbsp;S. Jose ,&nbsp;J. Edwards ,&nbsp;I. Luccarini ,&nbsp;J. Sermon ,&nbsp;N.J. Perualila ,&nbsp;J. Diels ,&nbsp;J. Penton ,&nbsp;A. Bjerrum","doi":"10.1016/j.esmorw.2025.100197","DOIUrl":"10.1016/j.esmorw.2025.100197","url":null,"abstract":"<div><h3>Background</h3><div>The purpose of this study was to describe patient profiles and real-world outcomes and identify prognostic factors among patients with common epidermal growth factor receptor (<em>EGFR</em>)–mutated advanced non-small-cell lung cancer (NSCLC) treated with first-line (1L) or second-line (2L) osimertinib.</div></div><div><h3>Patients and methods</h3><div>This retrospective study used data from the Epidemiological Strategy and Medical Economics (ESME; France), Rigshospitalet/Herlev Hospital (RH; Denmark), and National Cancer Registration and Analysis Service (NCRAS; UK) databases. Progression-free survival (PFS), time to treatment discontinuation (TTD), time to next therapy (TTNT), and overall survival (OS) were analyzed using Kaplan–Meier estimates. Prognostic factors for outcomes were identified using univariate and multivariable proportional hazards regression.</div></div><div><h3>Results</h3><div>A total of 1646 patients were included (ESME, <em>n</em> = 624; RH, <em>n</em> = 133; NCRAS, <em>n</em> = 889). For patients receiving 1L osimertinib, median PFS, TTD, TTNT, and OS were 11.9, 15.0, 18.8, and 24.8 months, respectively. Corresponding values for 2L osimertinib were 7.4, 10.0, 11.9, and 17.1 months. Although distinct prognostic factors were identified for different outcomes and lines of treatment, poor Eastern Cooperative Oncology Group performance status score, presence of exon 21 L858R substitutions, and liver metastases were consistently associated with poorer outcomes across all endpoints. Of 771 patients receiving 1L osimertinib, at least 38% died before initiating a 2L treatment; 41% of patients receiving 2L osimertinib (<em>n</em> = 875) died before initiating a third-line treatment. The most common subsequent therapy was chemotherapy [1L osimertinib, 47% (97/208); 2L osimertinib, 52% (193/368)].</div></div><div><h3>Conclusion</h3><div>This retrospective real-world analysis highlights the unmet need for more effective treatment options in common <em>EGFR</em>-mutated advanced NSCLC.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"10 ","pages":"Article 100197"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145362426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated risk stratification in myelofibrosis","authors":"D.T. Kristensen ,&nbsp;T.C. El-Galaly ,&nbsp;A.S. Roug","doi":"10.1016/j.esmorw.2025.100654","DOIUrl":"10.1016/j.esmorw.2025.100654","url":null,"abstract":"","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"10 ","pages":"Article 100654"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145617882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Artificial intelligence-driven phase recognition in lung surgery: a single-centre pilot study☆","authors":"S.J.-Y. Ohtani-Kim ,&nbsp;J. Samejima ,&nbsp;M. Wakabayashi ,&nbsp;M. Tada ,&nbsp;T. Miyoshi ,&nbsp;Y. Matsumura ,&nbsp;K. Tane ,&nbsp;K. Aokage ,&nbsp;Y. Ishikawa ,&nbsp;K. Hayashi ,&nbsp;T. Ogane ,&nbsp;K. Sasaki ,&nbsp;S. Takenaka ,&nbsp;Y. Kinebuchi ,&nbsp;M. Ito ,&nbsp;M. Tsuboi","doi":"10.1016/j.esmorw.2025.100194","DOIUrl":"10.1016/j.esmorw.2025.100194","url":null,"abstract":"<div><h3>Introduction</h3><div>Effective surgical specimen management during lung resection is crucial for accurate analyses and treatment. Minimally invasive techniques complicate workflows; thus, artificial intelligence (AI)-based solutions are needed to improve their safety and efficiency. We assessed the feasibility of AI for the automated classification of surgical phases in thoracoscopic wedge resection, examining the link between classification accuracy and surgical complexity, particularly during specimen extraction.</div></div><div><h3>Patients and methods</h3><div>This single-centre retrospective observational study from Japan included 73 video recordings of video-assisted thoracic surgery lung wedge resections with extraction of a single specimen carried out from January 2021 to December 2023. A Swin Transformer AI model was used to classify five distinct surgical phases: preparatory actions, lesion identification, resection, and specimen extraction. Pre- and postprocessing techniques improved model performance across different phases. The primary outcome was AI model performance in classifying surgical phases using metrics such as accuracy, precision, recall, and F1 score.</div></div><div><h3>Results</h3><div>The modified AI model achieved an overall accuracy of 0.778, with phase-specific accuracies ranging from 0.574 to 0.911. Significant improvements were observed in critical phases for specimen management (accuracy: 0.816). Clinical factors, including the number of access ports and phase duration, were key determinants of accuracy.</div></div><div><h3>Conclusions</h3><div>Our AI-driven phase recognition model for thoracoscopic lung surgery videos shows potential for optimizing operating room workflow, enhancing real-time decision-making, and improving efficiency by automating surgical phase classification.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"10 ","pages":"Article 100194"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268341","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recommendations and definitions for time-to-event endpoints using real-world data in oncology","authors":"P.A.J. Vissers ,&nbsp;M.A.G. Elferink ,&nbsp;M.J. Bijlsma ,&nbsp;L.G. van der Geest ,&nbsp;M. Koopman ,&nbsp;H.W.M. van Laarhoven ,&nbsp;G.A.P. Nieuwenhuijzen ,&nbsp;P.S.N. van Rossum ,&nbsp;F.P.C. Sijtsma ,&nbsp;G.R. Vink ,&nbsp;J. de Vos-Geelen ,&nbsp;H.W. Wilmink ,&nbsp;J.H.W. de Wilt ,&nbsp;R.H.A. Verhoeven ,&nbsp;F.N. van Erning","doi":"10.1016/j.esmorw.2025.100649","DOIUrl":"10.1016/j.esmorw.2025.100649","url":null,"abstract":"<div><h3>Background</h3><div>Real-world data is increasingly used to assess effectiveness of treatments in patients treated in everyday clinical practice. The aim of this study is to establish definitions for time-to-event endpoints based on real-world data, and to evaluate their applicability to cancer registry data.</div></div><div><h3>Materials and methods</h3><div>A multidisciplinary panel consisting of seven epidemiologists, one statistician and seven medical specialists with expertise in gastrointestinal oncology was composed to reach consensus on the definitions of time-to-event endpoints. After obtaining the final definitions, the time-to-event endpoints were applied to a population-based cohort of patients with different types of gastrointestinal cancer from the Netherlands Cancer Registry.</div></div><div><h3>Results</h3><div>For nine time-to-event endpoints, consensus on the definitions was reached: real-world data based (RW) recurrence-free survival, recurrence rate, local recurrence rate, locoregional recurrence rate, distant recurrence rate, progression-free survival, progression rate, treatment failure rate, and overall survival. For RW disease-free survival, no consensus was reached. All time-to-event endpoints appeared suitable for calculation with Netherlands Cancer Registry data for cohorts with follow-up data based on the established definitions, with the exception of RW-treatment failure rate due to the unavailability of cause of death.</div></div><div><h3>Conclusions</h3><div>This study offers clear definitions for time-to-event endpoints based on RW data which can be applied to cancer registry data. Uniform use of these definitions in future studies will enable better interpretation of results and aid in comparison between studies, but may require additional development when adapted to other countries, cancer types or data sources.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"10 ","pages":"Article 100649"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145571606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrating radiomics and real-world data to predict immune checkpoint inhibitor efficacy in advanced non-small-cell lung cancer☆","authors":"L. Provenzano ,&nbsp;M. Favali ,&nbsp;L. Mazzeo ,&nbsp;A. Spagnoletti ,&nbsp;M. Ruggirello ,&nbsp;G. Calareso ,&nbsp;F.G. Greco ,&nbsp;R. Vigorito ,&nbsp;A. Quarta ,&nbsp;F. Calimeri ,&nbsp;M. Monteleone ,&nbsp;G. Baselli ,&nbsp;E. De Momi ,&nbsp;B. Guirges ,&nbsp;A. Di Lello ,&nbsp;A. Zec ,&nbsp;A. Ferrarin ,&nbsp;C. Giani ,&nbsp;C. Silvestri ,&nbsp;M. Occhipinti ,&nbsp;A. Prelaj","doi":"10.1016/j.esmorw.2025.100182","DOIUrl":"10.1016/j.esmorw.2025.100182","url":null,"abstract":"<div><h3>Background</h3><div>Immunotherapy (IO) revolutionized the prognosis of patients with non-small-cell lung cancer (NSCLC). However, identifying optimal candidates for this treatment remains challenging. Based on previous studies suggesting the potential power of radiomics in predicting clinical outcomes in different clinical settings, we aimed to assess its capability in predicting IO efficacy in advanced NSCLC patients.</div></div><div><h3>Materials and methods</h3><div>A total of 375 advanced NSCLC patients treated with IO-based regimens from April 2013 to May 2022 were enrolled. Primary lung lesions were segmented and radiomic features extracted. Using clinical benefit rate and overall survival status at 6 and 24 months (OS6 and OS24) as endpoints, machine learning classifiers were trained and then evaluated on a test set.</div></div><div><h3>Results</h3><div>Model achieving the highest performance predicting long-term survival (OS24) reached an accuracy of 0.71 and area under the curve of 0.79 on the test set, using the combination of radiomic features and real-world data (RWD) as input. Combining radiomics with RWD consistently allowed to outperform predictions obtained using the current standard predictive biomarker, i.e. programmed death-ligand 1 expression, for most of the outcomes.</div></div><div><h3>Conclusions</h3><div>We explored a radiomics-based signature with potential utility in predicting the prognosis of NSCLC patients undergoing IO. Further validation is required to confirm its clinical applicability and to support oncologists in making prognostic assessments.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"10 ","pages":"Article 100182"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}