ESMO Real World Data and Digital Oncology最新文献_第2页

Automated extraction of temporalized tumor evolution from oncology EMRs using natural language processing 利用自然语言处理从肿瘤电子病历中自动提取暂时肿瘤演化

ESMO Real World Data and Digital Oncology

Pub Date : 2026-01-08 DOI: 10.1016/j.esmorw.2025.100660

C. Vinot , C. Ferté , T. Gaboriaud , A. Minvielle-Sebastia , A. Dubois , R. Schwob , S. Mallah , M. Pajiep , J.M. Alliot , A. Ferreira , E. Pons-Tostivint , J. Mazieres , A. Yazigi

Background

Extracting temporally sensitive outcomes such as tumor progression from unstructured electronic medical records (EMRs) remains a major challenge in oncology. This study evaluates a solution with a domain-adapted natural language processing (NLP) pipeline designed to extract structured, temporally anchored clinical outcomes from narrative EMR data.

Patients and methods

Patients with oncogene-addicted advanced or metastatic non-small-cell lung cancer (NSCLC) treated with oral targeted therapies between January 2020 and June 2023 at a French academic hospital were included. Extracted Facts were benchmarked against expert annotations. All outputs were mapped to Observational Medical Outcome Partnership vocabularies. F1-scores were calculated for the correct Concept detection without and with their Temporality. Real-world progression-free survival (rwPFS) was estimated based on retrieved clinical outcomes.

Results

Among 1030 NSCLC patients treated between 2020 and 2023, 112 were confirmed to have advanced or metastatic disease with an oncogenic driver mutation, primarily EGFR (n = 66), ALK (n = 23), and KRAS (n = 16). The NLP pipeline achieved high accuracy in extracting clinical concepts, with an F1-score of 79.7% for tumor evolution concepts and 62.0% when temporality was included. Overall performance across all domains reached F1-scores of 76.5% for concept extraction and 63.7% with temporality. Median rwPFS was 21.9 months for EGFR-mutated, 52.4 months for ALK-translocated, and 5.0 months for KRAS-mutant tumors, in line with published benchmarks. Reviewing automatically collected data was 5.8 times faster compared with manual collection.

Conclusions

Our solution demonstrates robust performance for extracting temporally structured tumor outcomes from EMRs and supports the reconstruction of real-world endpoints in oncology.

从非结构化电子病历（emr）中提取肿瘤进展等时间敏感结果仍然是肿瘤学的主要挑战。本研究评估了一个领域适应自然语言处理（NLP）管道的解决方案，该管道旨在从叙事EMR数据中提取结构化的、暂时固定的临床结果。患者和方法纳入2020年1月至2023年6月在法国一家学术医院接受口服靶向治疗的癌基因成瘾晚期或转移性非小细胞肺癌（NSCLC）患者。根据专家注释对提取的事实进行基准测试。所有输出都映射到观察性医疗结果伙伴关系词汇表。f1分数计算了没有和有时间性的正确概念检测。真实无进展生存期（rwPFS）是根据检索到的临床结果估计的。在2020年至2023年期间接受治疗的1030例NSCLC患者中，112例被证实患有晚期或转移性疾病，并伴有癌性驱动突变，主要是EGFR (n = 66), ALK （n = 23）和KRAS （n = 16）。NLP管道在提取临床概念方面具有较高的准确性，肿瘤演化概念的f1得分为79.7%，考虑时间性的f1得分为62.0%。所有领域的整体表现在概念提取方面达到了76.5%，在时间性方面达到了63.7%。egfr突变的中位rwPFS为21.9个月，alk易位的为52.4个月，kras突变的为5.0个月，与已发表的基准一致。审查自动收集的数据比手动收集快5.8倍。结论sour解决方案在从emr中提取时间结构的肿瘤结果方面表现出强大的性能，并支持肿瘤现实世界终点的重建。

{"title":"Automated extraction of temporalized tumor evolution from oncology EMRs using natural language processing","authors":"C. Vinot , C. Ferté , T. Gaboriaud , A. Minvielle-Sebastia , A. Dubois , R. Schwob , S. Mallah , M. Pajiep , J.M. Alliot , A. Ferreira , E. Pons-Tostivint , J. Mazieres , A. Yazigi","doi":"10.1016/j.esmorw.2025.100660","DOIUrl":"10.1016/j.esmorw.2025.100660","url":null,"abstract":"<div><h3>Background</h3><div>Extracting temporally sensitive outcomes such as tumor progression from unstructured electronic medical records (EMRs) remains a major challenge in oncology. This study evaluates a solution with a domain-adapted natural language processing (NLP) pipeline designed to extract structured, temporally anchored clinical outcomes from narrative EMR data.</div></div><div><h3>Patients and methods</h3><div>Patients with oncogene-addicted advanced or metastatic non-small-cell lung cancer (NSCLC) treated with oral targeted therapies between January 2020 and June 2023 at a French academic hospital were included. Extracted Facts were benchmarked against expert annotations. All outputs were mapped to Observational Medical Outcome Partnership vocabularies. F1-scores were calculated for the correct Concept detection without and with their Temporality. Real-world progression-free survival (rwPFS) was estimated based on retrieved clinical outcomes.</div></div><div><h3>Results</h3><div>Among 1030 NSCLC patients treated between 2020 and 2023, 112 were confirmed to have advanced or metastatic disease with an oncogenic driver mutation, primarily EGFR (<em>n</em> = 66), ALK (<em>n</em> = 23), and KRAS (<em>n</em> = 16). The NLP pipeline achieved high accuracy in extracting clinical concepts, with an F1-score of 79.7% for tumor evolution concepts and 62.0% when temporality was included. Overall performance across all domains reached F1-scores of 76.5% for concept extraction and 63.7% with temporality. Median rwPFS was 21.9 months for EGFR-mutated, 52.4 months for ALK-translocated, and 5.0 months for KRAS-mutant tumors, in line with published benchmarks. Reviewing automatically collected data was 5.8 times faster compared with manual collection.</div></div><div><h3>Conclusions</h3><div>Our solution demonstrates robust performance for extracting temporally structured tumor outcomes from EMRs and supports the reconstruction of real-world endpoints in oncology.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100660"},"PeriodicalIF":0.0,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926649","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}

引用次数: 0

Ovarian cancer recurrence prediction: comparing confirmatory to real-world predictors with machine learning 卵巢癌复发预测：用机器学习比较验证性预测和真实预测

ESMO Real World Data and Digital Oncology

Pub Date : 2026-01-08 DOI: 10.1016/j.esmorw.2025.100666

D. Katsimpokis , A.E.C. van Odenhoven , M.A.J.M. van Erp , H.H.B. Wenzel , M.A. van der Aa , M.M.H. van Swieten , H.P.M. Smedts , J.M.J. Piek

Background

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.

Materials and methods

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.

Results

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.

Conclusions

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.

背景：卵巢瓦里癌是女性中最致命的癌症之一，常在晚期被诊断出来，晚期（国际妇产科联合会IIB-IV）疾病的5年生存率为17%-28%。机器学习（ML）可能提供比传统方法更好的生存预后工具，并可以提供对预测因素的洞察。这项研究的重点是晚期卵巢癌，并将专家得出的预测因素与来自荷兰癌症登记处（NCR）的数据驱动因素进行对比，以预测无进展生存期。材料与方法采用德尔菲问卷法确定14个预测因素，并纳入最终分析。使用ML模型（正则化Cox回归、随机生存森林和XGBoost）将基于Delphi专家的变量集与来自NCR的真实世界数据（RWD）变量集进行比较。采用非正则化Cox模型作为基准。结果使用RWD变量集的正则化Cox模型优于使用Delphi变量集的传统Cox回归模型（c-index分别为0.70和0.64），但XGBoost的综合性能最好（c-index为0.75）。最能预测复发的因素是手术类型和减积结果、术后化疗给药、铂疗程数和社会经济地位，但Delphi没有确定。结论与传统的Cox回归相比，ML算法具有更高的预测能力。此外，来自癌症登记处的RWD比专家小组确定了更多的预测变量。总的来说，这些结果对人工智能（AI）辅助临床预后具有重要意义，并为人工智能驱动和基于专家的生存预测决策之间的差异提供了见解。

{"title":"Ovarian cancer recurrence prediction: comparing confirmatory to real-world predictors with machine learning","authors":"D. Katsimpokis , A.E.C. van Odenhoven , M.A.J.M. van Erp , H.H.B. Wenzel , M.A. van der Aa , M.M.H. van Swieten , H.P.M. Smedts , 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":"2026-01-08","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}

引用次数: 0

Real-world outcomes following PARP inhibitor maintenance in ovarian cancer by BRCA status: a retrospective cohort study BRCA状态影响卵巢癌患者PARP抑制剂维持后的真实结果：一项回顾性队列研究

ESMO Real World Data and Digital Oncology

Pub Date : 2026-01-06 DOI: 10.1016/j.esmorw.2025.100659

K. Zucker , B. Pickwell-Smith , A. Samani , A. Sujenthiran , H. Pittell , P. Mpofu , G. Hall

Background

The introduction of poly (adenosine-diphosphate ribose) polymerase inhibitors (PARPis) has significantly improved progression-free survival for patients with ovarian cancer after response to first-line platinum-based chemotherapy. Yet, there are concerns that PARPi may compromise response to further platinum due to cross-resistance mechanisms. This study investigates the clinical effectiveness of chemotherapy post-progression on PARPi maintenance therapy after initial platinum-based treatment for ovarian cancer, regardless of BRCA mutations, using real-world data. Additionally, this study summarises results across randomised trials of PARPi as first-line maintenance.

Materials and methods

This study used a United States-based electronic health record-derived de-identified database. A retrospective descriptive analysis was conducted on patients with ovarian cancer diagnosed from 1 January 2015 onwards. Patient data were collected, including BRCA and homologous recombination deficiency (HRD) status. Time to next treatment (TTNT), treatment-free interval (TFI), and the impact of BRCA and HRD status were assessed.

Results

Among 3649 patients, 81% had known BRCA status, of whom 83% were BRCA-negative, 17% were BRCA-positive, and 19% had unknown BRCA status. The majority (80%) had unknown HRD status. Notably, 17% received first-line PARPi. Patients with BRCA mutations displayed longer TTNT and TFI when receiving first-line PARPi initially. However, after receiving subsequent treatments, BRCA-mutated and non-mutated patients demonstrated shorter TFI and TTNT intervals, suggesting a possible influence of PARPis on subsequent chemotherapy efficacy.

Conclusions

The study underscores previous concerns that initiating PARPi in the first line may impact treatment duration and intervals for subsequent therapies in patients with ovarian cancer. Future investigations should explore the interplay of PARPi maintenance in the first-line setting, HRD status, and response to subsequent platinum-based therapies.

多（腺苷-二磷酸核糖）聚合酶抑制剂（PARPis）的引入显著提高了一线铂类化疗后卵巢癌患者的无进展生存期。然而，由于交叉耐药机制，PARPi可能会损害对进一步铂的反应。本研究利用真实世界数据，探讨卵巢癌患者在初始铂基治疗后PARPi维持治疗进展后化疗的临床有效性，无论BRCA突变如何。此外，本研究总结了PARPi作为一线维持治疗的随机试验结果。材料和方法本研究使用了基于美国的电子健康记录衍生的去识别数据库。对2015年1月1日起诊断为卵巢癌的患者进行回顾性描述性分析。收集患者资料，包括BRCA和同源重组缺陷（HRD）状态。评估下一次治疗时间（TTNT）、无治疗间隔（TFI）以及BRCA和HRD状态的影响。结果3649例患者中，已知BRCA状态的占81%，其中BRCA阴性的占83%，BRCA阳性的占17%，未知BRCA状态的占19%。大多数（80%）HRD状态未知。值得注意的是，17%的患者接受了一线PARPi治疗。BRCA突变患者最初接受一线PARPi治疗时，TTNT和TFI延长。然而，在接受后续治疗后，brca突变和未突变的患者表现出更短的TFI和TTNT间隔，这表明PARPis可能影响后续化疗疗效。该研究强调了先前的担忧，即在一线启动PARPi可能会影响卵巢癌患者后续治疗的治疗时间和间隔。未来的研究应探讨PARPi维持在一线环境、HRD状态和对后续铂类治疗的反应的相互作用。

{"title":"Real-world outcomes following PARP inhibitor maintenance in ovarian cancer by BRCA status: a retrospective cohort study","authors":"K. Zucker , B. Pickwell-Smith , A. Samani , A. Sujenthiran , H. Pittell , P. Mpofu , G. Hall","doi":"10.1016/j.esmorw.2025.100659","DOIUrl":"10.1016/j.esmorw.2025.100659","url":null,"abstract":"<div><h3>Background</h3><div>The introduction of poly (adenosine-diphosphate ribose) polymerase inhibitors (PARPis) has significantly improved progression-free survival for patients with ovarian cancer after response to first-line platinum-based chemotherapy. Yet, there are concerns that PARPi may compromise response to further platinum due to cross-resistance mechanisms. This study investigates the clinical effectiveness of chemotherapy post-progression on PARPi maintenance therapy after initial platinum-based treatment for ovarian cancer, regardless of <em>BRCA</em> mutations, using real-world data. Additionally, this study summarises results across randomised trials of PARPi as first-line maintenance.</div></div><div><h3>Materials and methods</h3><div>This study used a United States-based electronic health record-derived de-identified database. A retrospective descriptive analysis was conducted on patients with ovarian cancer diagnosed from 1 January 2015 onwards. Patient data were collected, including <em>BRCA</em> and homologous recombination deficiency (HRD) status. Time to next treatment (TTNT), treatment-free interval (TFI), and the impact of <em>BRCA</em> and HRD status were assessed.</div></div><div><h3>Results</h3><div>Among 3649 patients, 81% had known <em>BRCA</em> status, of whom 83% were <em>BRCA</em>-negative, 17% were <em>BRCA</em>-positive, and 19% had unknown <em>BRCA</em> status. The majority (80%) had unknown HRD status. Notably, 17% received first-line PARPi. Patients with <em>BRCA</em> mutations displayed longer TTNT and TFI when receiving first-line PARPi initially. However, after receiving subsequent treatments, <em>BRCA</em>-mutated and non-mutated patients demonstrated shorter TFI and TTNT intervals, suggesting a possible influence of PARPis on subsequent chemotherapy efficacy.</div></div><div><h3>Conclusions</h3><div>The study underscores previous concerns that initiating PARPi in the first line may impact treatment duration and intervals for subsequent therapies in patients with ovarian cancer. Future investigations should explore the interplay of PARPi maintenance in the first-line setting, HRD status, and response to subsequent platinum-based therapies.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100659"},"PeriodicalIF":0.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926646","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}

引用次数: 0

Second-line therapies after CDK4/6 inhibitor failure in HR-positive/HER2-negative metastatic breast cancer patients: real-world data from the HERMIONE-13 study hr阳性/ her2阴性转移性乳腺癌患者CDK4/6抑制剂失效后的二线治疗：来自herone -13研究的真实数据

ESMO Real World Data and Digital Oncology

Pub Date : 2026-01-06 DOI: 10.1016/j.esmorw.2025.100665

M.E. Cazzaniga , V. Cogliati , E. Rossi , I. Paris , F. Borella , G. Moretti , O. Garrone , M. Pistelli , R. Palumbo , A. Ferro , P. Vici , U. De Giorgi , S. Madaro , L. Coltelli , M. Giordano , A.R. Gambaro , S.V.L. Nicoletti , F. Zustovich , E. Landucci , T. Gamucci , S. Capici

Background

Cyclin-dependent kinase 4/6 inhibitors (CDK4/6i) combined with endocrine therapy (ET) are the standard first-line treatment for hormone receptor (HR)-positive, human epidermal growth factor receptor 2 (HER2)-negative metastatic breast cancer (MBC). However, treatment resistance and progression remain significant challenges, and optimal second-line strategies are not well defined.

Materials and methods

HERMIONE-13 is a multicentre, observational study conducted across 18 Italian centres, including both retrospective and prospective cohorts. The study aimed to describe real-world second-line treatment patterns following progression on CDK4/6i and to identify factors influencing therapeutic decisions. Clinical outcomes, including real-world progression-free survival (rwPFS) and overall survival (OS), were also evaluated.

Results

Among 254 assessable patients, 67.3% received chemotherapy (CHT) and 32.7% received ET ± targeted therapy (TT) as second-line treatment. The most common regimens included capecitabine and everolimus plus exemestane. Multivariable analysis showed that younger age, prior fulvestrant use, and shorter CDK4/6i duration were associated with CHT choice. Median rwPFS was 5.8 months for CHT and 5.3 months for ET ± TT. Median OS was longer in the ET ± TT group (3.8 versus 2.3 years). Metronomic CHT showed promising activity with a median rwPFS of 9.7 months.

Conclusions

In the Italian real-world setting, CHT remains the predominant second-line choice after CDK4/6i failure, though ET ± TT may offer comparable or superior outcomes in selected patients. Treatment decisions are influenced by clinical history and patient characteristics. These findings underscore the need for personalized approaches and molecular profiling to guide post-CDK4/6i therapy in HR-positive/HER2-negative MBC.

细胞周期蛋白依赖性激酶4/6抑制剂（CDK4/6i）联合内分泌治疗（ET）是激素受体（HR）阳性、人表皮生长因子受体2 （HER2）阴性转移性乳腺癌（MBC）的标准一线治疗方法。然而，治疗耐药性和进展仍然是重大挑战，最佳二线策略尚未明确。材料和方法shermione -13是一项在意大利18个中心进行的多中心观察性研究，包括回顾性和前瞻性队列。该研究旨在描述现实世界中CDK4/6i进展后的二线治疗模式，并确定影响治疗决策的因素。临床结果，包括真实世界无进展生存期（rwPFS）和总生存期（OS），也进行了评估。结果254例可评估患者中，67.3%接受化疗（CHT）， 32.7%接受ET±靶向治疗（TT）作为二线治疗。最常见的方案包括卡培他滨和依维莫司加依西美坦。多变量分析显示，年龄较小、既往使用过氟维司汀、CDK4/6i持续时间较短与CHT选择相关。中位rwPFS为CHT组5.8个月，ET±TT组5.3个月。ET±TT组的中位生存期更长（3.8年比2.3年）。节律性CHT显示有希望的活性，中位rwPFS为9.7个月。结论：在意大利现实环境中，CDK4/6i失败后，CHT仍然是主要的二线选择，尽管ET±TT可能在选定的患者中提供类似或更好的结果。治疗决定受临床病史和患者特征的影响。这些发现强调了个性化方法和分子分析指导hr阳性/ her2阴性MBC后cdk4 /6i治疗的必要性。

{"title":"Second-line therapies after CDK4/6 inhibitor failure in HR-positive/HER2-negative metastatic breast cancer patients: real-world data from the HERMIONE-13 study","authors":"M.E. Cazzaniga , V. Cogliati , E. Rossi , I. Paris , F. Borella , G. Moretti , O. Garrone , M. Pistelli , R. Palumbo , A. Ferro , P. Vici , U. De Giorgi , S. Madaro , L. Coltelli , M. Giordano , A.R. Gambaro , S.V.L. Nicoletti , F. Zustovich , E. Landucci , T. Gamucci , S. Capici","doi":"10.1016/j.esmorw.2025.100665","DOIUrl":"10.1016/j.esmorw.2025.100665","url":null,"abstract":"<div><h3>Background</h3><div>Cyclin-dependent kinase 4/6 inhibitors (CDK4/6i) combined with endocrine therapy (ET) are the standard first-line treatment for hormone receptor (HR)-positive, human epidermal growth factor receptor 2 (HER2)-negative metastatic breast cancer (MBC). However, treatment resistance and progression remain significant challenges, and optimal second-line strategies are not well defined.</div></div><div><h3>Materials and methods</h3><div>HERMIONE-13 is a multicentre, observational study conducted across 18 Italian centres, including both retrospective and prospective cohorts. The study aimed to describe real-world second-line treatment patterns following progression on CDK4/6i and to identify factors influencing therapeutic decisions. Clinical outcomes, including real-world progression-free survival (rwPFS) and overall survival (OS), were also evaluated.</div></div><div><h3>Results</h3><div>Among 254 assessable patients, 67.3% received chemotherapy (CHT) and 32.7% received ET ± targeted therapy (TT) as second-line treatment. The most common regimens included capecitabine and everolimus plus exemestane. Multivariable analysis showed that younger age, prior fulvestrant use, and shorter CDK4/6i duration were associated with CHT choice. Median rwPFS was 5.8 months for CHT and 5.3 months for ET ± TT. Median OS was longer in the ET ± TT group (3.8 versus 2.3 years). Metronomic CHT showed promising activity with a median rwPFS of 9.7 months.</div></div><div><h3>Conclusions</h3><div>In the Italian real-world setting, CHT remains the predominant second-line choice after CDK4/6i failure, though ET ± TT may offer comparable or superior outcomes in selected patients. Treatment decisions are influenced by clinical history and patient characteristics. These findings underscore the need for personalized approaches and molecular profiling to guide post-CDK4/6i therapy in HR-positive/HER2-negative MBC.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100665"},"PeriodicalIF":0.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926701","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}

引用次数: 0

How artificial intelligence applied to digital pathology could guide treatment personalization in breast cancer 将人工智能应用于数字病理学如何指导乳腺癌的个性化治疗

ESMO Real World Data and Digital Oncology

Pub Date : 2026-01-06 DOI: 10.1016/j.esmorw.2025.100662

T. Ruelle , T. Grinda , L. Del Mastro , M. Lacroix-Triki , B. Pistilli , G. Gessain

Introduction

The global increase in breast cancer incidence is determining substantial augment of pathologists’ workload. Digital pathology (DP), which consists of digitizing haematoxylin–eosin (H&E)-stained slides into whole-slide images, enables new workflows and horizons. The integration of artificial intelligence (AI) with DP led to the emergence of computational pathology (CP), which has the potential of offering valuable diagnostic, prognostic and predictive tools. The aim of this narrative review is to provide an overview of the state of the art in this rapidly evolving field.

Diagnostic, predictive and prognostic application

Several AI tools demonstrated enhanced performances as compared with visual evaluation of pathologists, such as identification of invasive breast cancers, sentinel lymph node metastasis, histological grading, Ki-67, estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor 2 (HER2) and programmed death-ligand 1 (PD-L1) quantification. Due to this, most of them are currently authorized in the European market. Beyond diagnosis, CP has shown strong potential in predicting, from a simple H&E digital slide, treatment response (e.g. neoadjuvant chemotherapy, anti-HER2 therapy and endocrine therapy), immunohistochemistry status (ER, PR, HER2, PD-L1), molecular alterations (e.g. BRCA1/2 and homologous recombination deficiency status), genomic signatures (e.g. Oncotype DX) and ultimately the patient’s risk of recurrence.

Conclusion

Despite challenges such as high validation costs, initial economic investments and potential AI biases, CP holds great promise. Continued effort to address these barriers and improve AI reliability are crucial. Emerging AI tools, including agent-driven systems, point towards the potential integration of CP into routine clinical workflow practice, even though clinical validation has not yet been established.

全球乳腺癌发病率的增加决定了病理学家工作量的大幅增加。数字病理学（DP），包括数字化血红素-伊红（H&；E）染色的幻灯片成全幻灯片图像，实现了新的工作流程和视野。人工智能（AI）与DP的整合导致了计算病理学（CP）的出现，它有可能提供有价值的诊断、预后和预测工具。这篇叙述性评论的目的是对这一迅速发展的领域的现状提供一个概述。与病理学家的视觉评估相比，一些人工智能工具表现出更高的性能，如浸润性乳腺癌的识别、前哨淋巴结转移、组织学分级、Ki-67、雌激素受体（ER）、孕激素受体（PR）、人表皮生长因子2 （HER2）和程序性死亡配体1 （PD-L1）的量化。正因为如此，目前大多数都在欧洲市场获得了授权。除了诊断之外，CP在预测治疗反应（如新辅助化疗、抗HER2治疗和内分泌治疗）、免疫组织化学状态（ER、PR、HER2、PD-L1）、分子改变（如BRCA1/2和同源重组缺陷状态）、基因组特征（如Oncotype DX）以及最终患者复发风险方面显示出强大的潜力。尽管存在验证成本高、初始经济投资和潜在的人工智能偏差等挑战，但CP具有很大的前景。继续努力解决这些障碍并提高人工智能的可靠性至关重要。新兴的人工智能工具，包括代理驱动的系统，指出了将CP整合到常规临床工作流程实践的潜力，尽管临床验证尚未建立。

{"title":"How artificial intelligence applied to digital pathology could guide treatment personalization in breast cancer","authors":"T. Ruelle , T. Grinda , L. Del Mastro , M. Lacroix-Triki , B. Pistilli , G. Gessain","doi":"10.1016/j.esmorw.2025.100662","DOIUrl":"10.1016/j.esmorw.2025.100662","url":null,"abstract":"<div><h3>Introduction</h3><div>The global increase in breast cancer incidence is determining substantial augment of pathologists’ workload. Digital pathology (DP), which consists of digitizing haematoxylin–eosin (H&E)-stained slides into whole-slide images, enables new workflows and horizons. The integration of artificial intelligence (AI) with DP led to the emergence of computational pathology (CP), which has the potential of offering valuable diagnostic, prognostic and predictive tools. The aim of this narrative review is to provide an overview of the state of the art in this rapidly evolving field.</div></div><div><h3>Diagnostic, predictive and prognostic application</h3><div>Several AI tools demonstrated enhanced performances as compared with visual evaluation of pathologists, such as identification of invasive breast cancers, sentinel lymph node metastasis, histological grading, Ki-67, estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor 2 (HER2) and programmed death-ligand 1 (PD-L1) quantification. Due to this, most of them are currently authorized in the European market. Beyond diagnosis, CP has shown strong potential in predicting, from a simple H&E digital slide, treatment response (e.g. neoadjuvant chemotherapy, anti-HER2 therapy and endocrine therapy), immunohistochemistry status (ER, PR, HER2, PD-L1), molecular alterations (e.g. <em>BRCA1/2</em> and homologous recombination deficiency status), genomic signatures (e.g. Oncotype DX) and ultimately the patient’s risk of recurrence.</div></div><div><h3>Conclusion</h3><div>Despite challenges such as high validation costs, initial economic investments and potential AI biases, CP holds great promise. Continued effort to address these barriers and improve AI reliability are crucial. Emerging AI tools, including agent-driven systems, point towards the potential integration of CP into routine clinical workflow practice, even though clinical validation has not yet been established.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100662"},"PeriodicalIF":0.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926648","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}

引用次数: 0

Age and frailty in anticancer drug regulatory assessment: a comprehensive cohort study of European marketing authorisations 2012-2023 抗肿瘤药物监管评估中的年龄和虚弱：2012-2023年欧洲上市许可的综合队列研究

ESMO Real World Data and Digital Oncology

Pub Date : 2026-01-06 DOI: 10.1016/j.esmorw.2025.100663

O. Tenhunen , K. Penttinen , C. Voltz , H. Kuitunen , M. Turpeinen

Background

Prevalence of cancer increases with age, but older adults remain underrepresented in pivotal clinical trials. We analysed the representation of older adults and poor performance score patients in initial marketing authorisation (MA) of new anticancer medicines in the European Union (EU).

Materials and methods

Medicines with an anticancer indication authorised in the EU during the years 2012-2023 were analysed based on public assessment reports (EPARs). We identified original clinical studies and datasets based on the classification in the EPARs.

Results

One hundred and thirty-seven products and MAs, corresponding with 153 efficacy datasets fulfilled the inclusion criteria. Most studies had enrolled subjects in age groups 65-74, 75-84, and ≥85 years of age; however, the proportion of individuals in these age groups was low. The majority of cases analysed efficacy in patient subgroups of >65 years of age, whereas a minority of studies evaluated efficacy for groups older than 70 years. No differences were observed between time cohorts from 2012-2017 and 2018-2023. Fifty-one percent of datasets allowed inclusion of patients with an Eastern Cooperative Oncology Group performance status of ≥2, whereas the median proportion of these patients was 1.8%. Safety was specifically analysed in subjects of >65 years of age in 92% of MAs. A dedicated discussion on age and/or frailty was identified in 45% of the MAs. Age- or frailty-related Annex II conditions were found in 10 cases.

Conclusion

Older adults and patients with higher performance scores have been included in the pivotal MA data and analyses, but their proportions have remained low. An opportunity for structured communication on age-related uncertainties is identified.

癌症患病率随着年龄的增长而增加，但老年人在关键临床试验中的代表性仍然不足。我们分析了欧盟（EU）新抗癌药物初始上市许可（MA）中老年人和表现评分较差患者的代表性。材料和方法根据公共评估报告（epar）对2012-2023年欧盟批准的具有抗癌适应症的药物进行了分析。我们根据EPARs中的分类确定了原始临床研究和数据集。结果137个产品和MAs对应153个疗效数据集符合纳入标准。大多数研究纳入的受试者年龄为65-74岁、75-84岁和≥85岁；然而，这些年龄组的个体比例很低。大多数病例分析了65岁患者亚组的疗效，而少数研究评估了70岁以上患者的疗效。在2012-2017年和2018-2023年的时间队列之间没有观察到差异。51%的数据集允许纳入Eastern Cooperative Oncology Group表现状态≥2的患者，而这些患者的中位比例为1.8%。在92%的MAs中，对65岁受试者的安全性进行了专门分析。在45%的MAs中发现了关于年龄和/或虚弱的专门讨论。在10例病例中发现与年龄或虚弱有关的附件二病症。结论老年人和表现得分较高的患者已被纳入关键的MA数据和分析，但其比例仍然很低。确定了就与年龄有关的不确定性进行结构化沟通的机会。

{"title":"Age and frailty in anticancer drug regulatory assessment: a comprehensive cohort study of European marketing authorisations 2012-2023","authors":"O. Tenhunen , K. Penttinen , C. Voltz , H. Kuitunen , M. Turpeinen","doi":"10.1016/j.esmorw.2025.100663","DOIUrl":"10.1016/j.esmorw.2025.100663","url":null,"abstract":"<div><h3>Background</h3><div>Prevalence of cancer increases with age, but older adults remain underrepresented in pivotal clinical trials. We analysed the representation of older adults and poor performance score patients in initial marketing authorisation (MA) of new anticancer medicines in the European Union (EU).</div></div><div><h3>Materials and methods</h3><div>Medicines with an anticancer indication authorised in the EU during the years 2012-2023 were analysed based on public assessment reports (EPARs). We identified original clinical studies and datasets based on the classification in the EPARs.</div></div><div><h3>Results</h3><div>One hundred and thirty-seven products and MAs, corresponding with 153 efficacy datasets fulfilled the inclusion criteria. Most studies had enrolled subjects in age groups 65-74, 75-84, and ≥85 years of age; however, the proportion of individuals in these age groups was low. The majority of cases analysed efficacy in patient subgroups of >65 years of age, whereas a minority of studies evaluated efficacy for groups older than 70 years. No differences were observed between time cohorts from 2012-2017 and 2018-2023. Fifty-one percent of datasets allowed inclusion of patients with an Eastern Cooperative Oncology Group performance status of ≥2, whereas the median proportion of these patients was 1.8%. Safety was specifically analysed in subjects of >65 years of age in 92% of MAs. A dedicated discussion on age and/or frailty was identified in 45% of the MAs. Age- or frailty-related Annex II conditions were found in 10 cases.</div></div><div><h3>Conclusion</h3><div>Older adults and patients with higher performance scores have been included in the pivotal MA data and analyses, but their proportions have remained low. An opportunity for structured communication on age-related uncertainties is identified.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100663"},"PeriodicalIF":0.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926650","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}

引用次数: 0

UK real-world evidence on pembrolizumab and neoadjuvant chemotherapy for early-stage triple-negative breast cancer 派姆单抗和新辅助化疗治疗早期三阴性乳腺癌的英国真实证据

ESMO Real World Data and Digital Oncology

Pub Date : 2026-01-06 DOI: 10.1016/j.esmorw.2025.100664

J. McKeon , E. Daniels , A. Halstead , C. Machado , S. Potter , T. Strawson-Smith , A. Okines , S. Mcintosh , H. Tovey , T. Robinson

Background

The KEYNOTE-522 trial showed that use of pembrolizumab-chemotherapy for stage 2-3 triple-negative breast cancer improved pathological complete response (pCR) rates and event-free and overall survival compared with chemotherapy alone. This retrospective multicentre cohort study collated real-world-evidence (RWE) of the efficacy and safety of neoadjuvant pembrolizumab-chemotherapy.

Patients and methods

Patients treated with pembrolizumab-chemotherapy who completed pembrolizumab treatment before 1 July 2024 were included in the study. Data were collected on safety and efficacy of treatment. pCR rates were compared between subgroups using chi-square and tests for trend (where appropriate), no adjustment was made for multiplicity.

Results

Five hundred and forty-five patients were recruited from 34 UK centres (median age 50 years, range 21-79 years). Neoadjuvant treatment delays and discontinuations were reported by 59% [95% confidence interval (CI) 55% to 63%] and 47% (95% CI 42% to 51%) respectively. Immune-related adverse events (irAEs) were experienced by 65% (95% CI 61% to 69%); 20% of patients (95% CI 16% to 23%) experienced grade ≥3 irAEs. A total of 83% (95% CI 80% to 86%) and 45% (95% CI 41% to 50%) of patients had unplanned medical contacts and admissions. pCR rate was 56% (95% CI 52% to 60%). IrAEs were associated with increased pCR rates, but steroid use and pembrolizumab-chemotherapy discontinuation did not adversely affect rates of pCR.

Conclusions

This large, national RWE study demonstrates pCR rate was lower compared with that observed in KEYNOTE-522. High rates of dose reductions, discontinuations and unplanned medical contacts (with significant health care resource use) suggests that patient selection is important for this regimen. Collection of long-term outcomes will be important to better evaluate the benefit/risks of pembrolizumab, especially in certain subgroups.

KEYNOTE-522试验显示，与单独化疗相比，使用派姆单抗化疗治疗2-3期三阴性乳腺癌可提高病理完全缓解（pCR）率、无事件生存期和总生存期。这项回顾性多中心队列研究整理了新辅助派姆单抗化疗的有效性和安全性的现实证据（RWE）。患者和方法2024年7月1日前完成派姆单抗化疗的患者纳入研究。收集治疗的安全性和有效性数据。亚组间pCR率采用卡方检验和趋势检验（如适用）进行比较，未对多重性进行调整。结果从34个英国中心招募了545名患者（中位年龄50岁，21-79岁）。新辅助治疗延迟和中断的报告分别为59%[95%可信区间（CI） 55%至63%]和47% （95% CI 42%至51%）。65%的患者发生免疫相关不良事件（irae）（95%可信区间61% - 69%）；20%的患者（95% CI 16% - 23%）经历≥3级irae。共有83%（95%置信区间为80%至86%）和45%（95%置信区间为41%至50%）的患者有计划外的医疗接触和入院。pCR率为56% （95% CI 52% ~ 60%）。irae与pCR率增加相关，但类固醇使用和停止派姆单抗化疗对pCR率没有不利影响。结论：这项大型的全国性RWE研究表明，与KEYNOTE-522中观察到的pCR率相比，pCR率更低。剂量减少、停药和计划外医疗接触（大量使用卫生保健资源）的高比率表明，患者选择对该方案很重要。收集长期结果对于更好地评估派姆单抗的获益/风险非常重要，特别是在某些亚组中。

{"title":"UK real-world evidence on pembrolizumab and neoadjuvant chemotherapy for early-stage triple-negative breast cancer","authors":"J. McKeon , E. Daniels , A. Halstead , C. Machado , S. Potter , T. Strawson-Smith , A. Okines , S. Mcintosh , H. Tovey , T. Robinson","doi":"10.1016/j.esmorw.2025.100664","DOIUrl":"10.1016/j.esmorw.2025.100664","url":null,"abstract":"<div><h3>Background</h3><div>The KEYNOTE-522 trial showed that use of pembrolizumab-chemotherapy for stage 2-3 triple-negative breast cancer improved pathological complete response (pCR) rates and event-free and overall survival compared with chemotherapy alone. This retrospective multicentre cohort study collated real-world-evidence (RWE) of the efficacy and safety of neoadjuvant pembrolizumab-chemotherapy.</div></div><div><h3>Patients and methods</h3><div>Patients treated with pembrolizumab-chemotherapy who completed pembrolizumab treatment before 1 July 2024 were included in the study. Data were collected on safety and efficacy of treatment. pCR rates were compared between subgroups using chi-square and tests for trend (where appropriate), no adjustment was made for multiplicity.</div></div><div><h3>Results</h3><div>Five hundred and forty-five patients were recruited from 34 UK centres (median age 50 years, range 21-79 years). Neoadjuvant treatment delays and discontinuations were reported by 59% [95% confidence interval (CI) 55% to 63%] and 47% (95% CI 42% to 51%) respectively. Immune-related adverse events (irAEs) were experienced by 65% (95% CI 61% to 69%); 20% of patients (95% CI 16% to 23%) experienced grade ≥3 irAEs. A total of 83% (95% CI 80% to 86%) and 45% (95% CI 41% to 50%) of patients had unplanned medical contacts and admissions. pCR rate was 56% (95% CI 52% to 60%). IrAEs were associated with increased pCR rates, but steroid use and pembrolizumab-chemotherapy discontinuation did not adversely affect rates of pCR.</div></div><div><h3>Conclusions</h3><div>This large, national RWE study demonstrates pCR rate was lower compared with that observed in KEYNOTE-522. High rates of dose reductions, discontinuations and unplanned medical contacts (with significant health care resource use) suggests that patient selection is important for this regimen. Collection of long-term outcomes will be important to better evaluate the benefit/risks of pembrolizumab, especially in certain subgroups.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100664"},"PeriodicalIF":0.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926700","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}

引用次数: 0

Evaluating the statistical utility and information loss in the transformation of a real-world oncology database from CDISC-SDTM to OMOP-CDM 评估现实世界肿瘤数据库从CDISC-SDTM到OMOP-CDM转换的统计效用和信息损失

ESMO Real World Data and Digital Oncology

Pub Date : 2025-12-15 DOI: 10.1016/j.esmorw.2025.100655

A. Lambert , C. Castagne , D. Pau , J. Chmiel , A. Labarga , E. Boernert , F. Margraff , C. Bachot , L. Kaczmarek , D. Toshev , J. Martinez G. , T. Stone

Background

This project documented the conversion of a French real-world oncology study database from Clinical Data Interchange Standards Consortium (CDISC) Study Data Tabulation Model (SDTM) to the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM). The goal was to measure syntactic and semantic information loss resulting from this transformation.

Patients and methods

Source data originated from a retrospective observational study of HER2-positive early breast cancer patients. Initially in CDISC-SDTM standards, the data included 73 variables detailing patient demographics, disease characteristics, surgery, and follow-up information such as adjuvant treatment. Data conversion to OMOP-CDM utilized extract, load and transform (ELT) procedures with Observational Health Data Sciences and Informatics (OHDSI) tools and the data build tool (dbt), encountering challenges in mapping specific variables and maintaining data granularity. Information loss assessments involved statistical analyses.

Results

The source database was successfully mapped to OMOP-CDM and standardized terminologies. Statistical results from the OMOP-transformed database were consistent with those from the original SDTM database, achieving 100% concordance across all tested equality criteria for univariate, bivariate, logistic, survival, and correlation analyses at a 95% confidence interval or respective P value significance levels. Information loss (<1%) during conversion varied based on the original database’s detail and mapping approach.

Conclusions

Statistical utility of the real-world oncology dataset was maintained after transformation to OMOP-CDM, ensuring reproducibility of statistical analyses. Information loss during conversion is significantly dependent on the intrinsic characteristics and level of standardization of the source database, particularly if not originally adhering to standard vocabularies.

本项目记录了法国真实世界肿瘤研究数据库从临床数据交换标准联盟（CDISC）研究数据制表模型（SDTM）到观察性医疗结果合作伙伴关系（OMOP）公共数据模型（CDM）的转换。我们的目标是度量这种转换导致的语法和语义信息损失。患者和方法资料来源于一项her2阳性早期乳腺癌患者的回顾性观察研究。最初在CDISC-SDTM标准中，数据包括73个变量，详细描述了患者人口统计学、疾病特征、手术和辅助治疗等随访信息。数据转换到OMOP-CDM使用了带有观察健康数据科学和信息学（OHDSI）工具和数据构建工具（dbt）的提取、加载和转换（ELT）程序，在映射特定变量和维护数据粒度方面遇到了挑战。信息损失评估涉及统计分析。结果源数据库成功映射到OMOP-CDM和标准化术语。omop转换数据库的统计结果与原始SDTM数据库的统计结果一致，在95%的置信区间或各自的P值显著性水平下，单变量、双变量、logistic、生存和相关分析的所有检验的平等标准均达到100%的一致性。转换期间的信息损失（<1%）根据原始数据库的细节和映射方法而变化。结论转换为OMOP-CDM后，保持了真实肿瘤数据集的统计效用，确保了统计分析的可重复性。转换过程中的信息丢失很大程度上取决于源数据库的内在特征和标准化水平，特别是在最初不遵循标准词汇表的情况下。

{"title":"Evaluating the statistical utility and information loss in the transformation of a real-world oncology database from CDISC-SDTM to OMOP-CDM","authors":"A. Lambert , C. Castagne , D. Pau , J. Chmiel , A. Labarga , E. Boernert , F. Margraff , C. Bachot , L. Kaczmarek , D. Toshev , J. Martinez G. , T. Stone","doi":"10.1016/j.esmorw.2025.100655","DOIUrl":"10.1016/j.esmorw.2025.100655","url":null,"abstract":"<div><h3>Background</h3><div>This project documented the conversion of a French real-world oncology study database from Clinical Data Interchange Standards Consortium (CDISC) Study Data Tabulation Model (SDTM) to the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM). The goal was to measure syntactic and semantic information loss resulting from this transformation.</div></div><div><h3>Patients and methods</h3><div>Source data originated from a retrospective observational study of HER2-positive early breast cancer patients. Initially in CDISC-SDTM standards, the data included 73 variables detailing patient demographics, disease characteristics, surgery, and follow-up information such as adjuvant treatment. Data conversion to OMOP-CDM utilized extract, load and transform (ELT) procedures with Observational Health Data Sciences and Informatics (OHDSI) tools and the data build tool (dbt), encountering challenges in mapping specific variables and maintaining data granularity. Information loss assessments involved statistical analyses.</div></div><div><h3>Results</h3><div>The source database was successfully mapped to OMOP-CDM and standardized terminologies. Statistical results from the OMOP-transformed database were consistent with those from the original SDTM database, achieving 100% concordance across all tested equality criteria for univariate, bivariate, logistic, survival, and correlation analyses at a 95% confidence interval or respective <em>P</em> value significance levels. Information loss (<1%) during conversion varied based on the original database’s detail and mapping approach.</div></div><div><h3>Conclusions</h3><div>Statistical utility of the real-world oncology dataset was maintained after transformation to OMOP-CDM, ensuring reproducibility of statistical analyses. Information loss during conversion is significantly dependent on the intrinsic characteristics and level of standardization of the source database, particularly if not originally adhering to standard vocabularies.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100655"},"PeriodicalIF":0.0,"publicationDate":"2025-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791774","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}

引用次数: 0

Balancing clinical relevance, legal boundaries, and technological solutions: a case-based analysis of secondary use of electronic health records in Sweden 平衡临床相关性、法律界限和技术解决方案：瑞典电子健康记录二次使用的案例分析

ESMO Real World Data and Digital Oncology

Pub Date : 2025-12-15 DOI: 10.1016/j.esmorw.2025.100661

Z. Dóczi , A. Valachis

Background

The secondary use of electronic health records (EHRs) poses legal challenges, particularly when the responsibility for managing EHRs lies with local or regional authorities. This article presents a case-based analysis of the secondary use of EHR data in contexts where data privacy responsibilities are managed regionally in Sweden.

Methods and results

Using two distinct purposes for the secondary use of the digital tool Patient Overview Breast Cancer: (i) assessing the uptake of new treatment strategies in a real-world setting for quality assurance, and (ii) evaluating the effectiveness of these strategies in specific patient subgroups with limited evidence for research purposes, the study explored the distinctions between research and quality assurance, the legal implications of each framework, and the potential role of federated learning as a privacy-preserving technological solution.

Conclusions

Federated learning offers a promising approach to overcome legal and organizational barriers to secondary use of regional EHRs in Sweden, enabling scalable, clinically meaningful insights for cancer care. However, its effective implementation requires a unified national framework that balances personal integrity with patient safety, supported by regulatory sandboxes.

电子健康记录（EHRs）的二次使用带来了法律挑战，特别是当管理EHRs的责任属于地方或区域当局时。本文提出了一个基于案例的分析，在瑞典数据隐私责任管理区域的背景下，电子病历数据的二次使用。方法和结果使用两种不同的目的进行数字工具的二次使用(i)评估在现实环境中对质量保证的新治疗策略的采用情况，以及（ii）评估这些策略在特定患者亚组中的有效性，研究目的证据有限，该研究探讨了研究与质量保证之间的区别，每个框架的法律含义，以及联邦学习作为隐私保护技术解决方案的潜在作用。联邦学习提供了一种很有前途的方法，可以克服瑞典区域电子病历二次使用的法律和组织障碍，为癌症治疗提供可扩展的、有临床意义的见解。然而，它的有效实施需要一个统一的国家框架，在监管沙盒的支持下平衡个人诚信与患者安全。

{"title":"Balancing clinical relevance, legal boundaries, and technological solutions: a case-based analysis of secondary use of electronic health records in Sweden","authors":"Z. Dóczi , A. Valachis","doi":"10.1016/j.esmorw.2025.100661","DOIUrl":"10.1016/j.esmorw.2025.100661","url":null,"abstract":"<div><h3>Background</h3><div>The secondary use of electronic health records (EHRs) poses legal challenges, particularly when the responsibility for managing EHRs lies with local or regional authorities. This article presents a case-based analysis of the secondary use of EHR data in contexts where data privacy responsibilities are managed regionally in Sweden.</div></div><div><h3>Methods and results</h3><div>Using two distinct purposes for the secondary use of the digital tool Patient Overview Breast Cancer: (i) assessing the uptake of new treatment strategies in a real-world setting for quality assurance, and (ii) evaluating the effectiveness of these strategies in specific patient subgroups with limited evidence for research purposes, the study explored the distinctions between research and quality assurance, the legal implications of each framework, and the potential role of federated learning as a privacy-preserving technological solution.</div></div><div><h3>Conclusions</h3><div>Federated learning offers a promising approach to overcome legal and organizational barriers to secondary use of regional EHRs in Sweden, enabling scalable, clinically meaningful insights for cancer care. However, its effective implementation requires a unified national framework that balances personal integrity with patient safety, supported by regulatory sandboxes.</div></div>","PeriodicalId":100491,"journal":{"name":"ESMO Real World Data and Digital Oncology","volume":"11 ","pages":"Article 100661"},"PeriodicalIF":0.0,"publicationDate":"2025-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791775","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}

引用次数: 0

AI for clinical trials in oncology 用于肿瘤临床试验的人工智能

ESMO Real World Data and Digital Oncology

Pub Date : 2025-12-09 DOI: 10.1016/j.esmorw.2025.100658

E. Cerami , I.B. Riaz , K.L. Kehl

In this perspective, we summarize recent developments in artificial intelligence (AI) applications for oncology clinical trials, divided into four key areas: (1) AI-driven drug design; (2) trial risk assessment, or predicting whether a novel therapy candidate will be safe and efficacious; (3) trial matching—that is, retrieval of reasonable trial options for specific patients, or patients for specific trials; and (4) trial eligibility screening or pre-screening, given a candidate patient–trial match.

从这个角度来看，我们总结了人工智能（AI）在肿瘤临床试验中的应用的最新进展，分为四个关键领域：(1)人工智能驱动的药物设计；(2)试验风险评估，或预测一种新的候选疗法是否安全有效；(3)试验匹配——即为特定患者检索合理的试验选择，或为特定试验检索患者；(4)试验资格筛选或预筛选，给定候选患者-试验匹配。

引用次数: 0