JCO Clinical Cancer Informatics最新文献

Purpose: The quality of radiotherapy auto-segmentation training data, primarily derived from clinician observers, is of utmost importance. However, the factors influencing the quality of clinician-derived segmentations are poorly understood; our study aims to quantify these factors.

Methods: Organ at risk (OAR) and tumor-related segmentations provided by radiation oncologists from the Contouring Collaborative for Consensus in Radiation Oncology data set were used. Segmentations were derived from five disease sites: breast, sarcoma, head and neck (H&N), gynecologic (GYN), and GI. Segmentation quality was determined on a structure-by-structure basis by comparing the observer segmentations with an expert-derived consensus, which served as a reference standard benchmark. The Dice similarity coefficient (DSC) was primarily used as a metric for the comparisons. DSC was stratified into binary groups on the basis of structure-specific expert-derived interobserver variability (IOV) cutoffs. Generalized linear mixed-effects models using Bayesian estimation were used to investigate the association between demographic variables and the binarized DSC for each disease site. Variables with a highest density interval excluding zero were considered to substantially affect the outcome measure.

Results: Five hundred seventy-four, 110, 452, 112, and 48 segmentations were used for the breast, sarcoma, H&N, GYN, and GI cases, respectively. The median percentage of segmentations that crossed the expert DSC IOV cutoff when stratified by structure type was 55% and 31% for OARs and tumors, respectively. Regression analysis revealed that the structure being tumor-related had a substantial negative impact on binarized DSC for the breast, sarcoma, H&N, and GI cases. There were no recurring relationships between segmentation quality and demographic variables across the cases, with most variables demonstrating large standard deviations.

Conclusion: Our study highlights substantial uncertainty surrounding conventionally presumed factors influencing segmentation quality relative to benchmarks.

Using the electronic health record to address the underdiagnosis of Lynch syndrome.

Purpose: Prostate cancer (PCa) represents a highly heterogeneous disease that requires tools to assess oncologic risk and guide patient management and treatment planning. Current models are based on various clinical and pathologic parameters including Gleason grading, which suffers from a high interobserver variability. In this study, we determine whether objective machine learning (ML)-driven histopathology image analysis would aid us in better risk stratification of PCa.

Materials and methods: We propose a deep learning, histopathology image-based risk stratification model that combines clinicopathologic data along with hematoxylin and eosin- and Ki-67-stained histopathology images. We train and test our model, using a five-fold cross-validation strategy, on a data set from 502 treatment-naïve PCa patients who underwent radical prostatectomy (RP) between 2000 and 2012.

Results: We used the concordance index as a measure to evaluate the performance of various risk stratification models. Our risk stratification model on the basis of convolutional neural networks demonstrated superior performance compared with Gleason grading and the Cancer of the Prostate Risk Assessment Post-Surgical risk stratification models. Using our model, 3.9% of the low-risk patients were correctly reclassified to be high-risk and 21.3% of the high-risk patients were correctly reclassified as low-risk.

Conclusion: These findings highlight the importance of ML as an objective tool for histopathology image assessment and patient risk stratification. With further validation on large cohorts, the digital pathology risk classification we propose may be helpful in guiding administration of adjuvant therapy including radiotherapy after RP.

There has been growing interest in the use of real-world data (RWD) to address clinically and policy-relevant (research) questions that cannot be answered with data from randomized controlled trials (RCTs) alone. This is, for example, the case in rare malignancies such as sarcomas as limited patient numbers pose challenges in conducting RCTs within feasible timeliness, a manageable number of collaborators, and statistical power. This narrative review explores the potential of RWD to generate real-world evidence (RWE) in sarcoma research, elucidating its application across different phases of the patient journey, from prediagnosis to the follow-up/survivorship phase. For instance, examining electronic health records (EHRs) from general practitioners (GPs) enables the exploration of consultation frequency and presenting symptoms in primary care before a sarcoma diagnosis. In addition, alternative study designs that integrate RWD with well-designed observational RCTs may offer relevant information on the effectiveness of clinical treatments. As, especially in cases of ultrarare sarcomas, it can be an extreme challenge to perform well-powered randomized prospective studies. Therefore, it is crucial to support the adaptation of novel study designs. Regarding the follow-up/survivorship phase, examining EHR from primary and secondary care can provide valuable insights into identifying the short- and long-term effects of treatment over an extended follow-up period. The utilization of RWD also comes with several challenges, including issues related to data quality and privacy, as described in this study. Notwithstanding these challenges, this study underscores the potential of RWD to bridge, at least partially, gaps between evidence and practice and holds promise in contributing to the improvement of sarcoma care.

Primer that helps clarify large-scale clinical data sets and participant demographics for oncologists.

Purpose: Decision about the optimal timing of a treatment procedure in patients with hematologic neoplasms is critical, especially for cellular therapies (most including allogeneic hematopoietic stem-cell transplantation [HSCT]). In the absence of evidence from randomized trials, real-world observational data become beneficial to study the effect of the treatment timing. In this study, a framework to estimate the expected outcome after an intervention in a time-to-event scenario is developed, with the aim of optimizing the timing in a personalized manner.

Methods: Retrospective real-world data are leveraged to emulate a target trial for treatment timing using multistate modeling and microsimulation. This case study focuses on myelodysplastic syndromes, serving as a prototype for rare cancers characterized by a heterogeneous clinical course and complex genomic background. A cohort of 7,118 patients treated according to conventional available treatments/evidence across Europe and United States is analyzed. The primary clinical objective is to determine the ideal timing for HSCT, the only curative option for these patients.

Results: This analysis enabled us to identify the most appropriate time frames for HSCT on the basis of each patient's unique profile, defined by a combination relevant patients' characteristics.

Conclusion: The developed methodology offers a structured framework to address a relevant clinical issue in the field of hematology. It makes several valuable contributions: (1) novel insights into how to develop decision models to identify the most favorable HSCT timing, (2) evidence to inform clinical decisions in a real-world context, and (3) the incorporation of complex information into decision making. This framework can be applied to provide medical insights for clinical issues that cannot be adequately addressed through randomized clinical trials.

Purpose: Artificial intelligence (AI) models can generate scientific abstracts that are difficult to distinguish from the work of human authors. The use of AI in scientific writing and performance of AI detection tools are poorly characterized.

Methods: We extracted text from published scientific abstracts from the ASCO 2021-2023 Annual Meetings. Likelihood of AI content was evaluated by three detectors: GPTZero, Originality.ai, and Sapling. Optimal thresholds for AI content detection were selected using 100 abstracts from before 2020 as negative controls, and 100 produced by OpenAI's GPT-3 and GPT-4 models as positive controls. Logistic regression was used to evaluate the association of predicted AI content with submission year and abstract characteristics, and adjusted odds ratios (aORs) were computed.

Results: Fifteen thousand five hundred and fifty-three abstracts met inclusion criteria. Across detectors, abstracts submitted in 2023 were significantly more likely to contain AI content than those in 2021 (aOR range from 1.79 with Originality to 2.37 with Sapling). Online-only publication and lack of clinical trial number were consistently associated with AI content. With optimal thresholds, 99.5%, 96%, and 97% of GPT-3/4-generated abstracts were identified by GPTZero, Originality, and Sapling respectively, and no sampled abstracts from before 2020 were classified as AI generated by the GPTZero and Originality detectors. Correlation between detectors was low to moderate, with Spearman correlation coefficient ranging from 0.14 for Originality and Sapling to 0.47 for Sapling and GPTZero.

Conclusion: There is an increasing signal of AI content in ASCO abstracts, coinciding with the growing popularity of generative AI models.

Purpose: To comprehensively synthesize the existing evidence concerning mHealth interventions for patients with breast cancer (BC).

Design: On July 30, 2023, we searched PubMed, PsycINFO, and Google Scholar for articles using the following inclusion criteria: evaluation of mHealth interventions in patients with cancer, at least 30 participants with BC, randomized control trials or prospective pre-post studies, determinants of health (patient-reported outcomes [PROs] and quality of life [QoL]) as primary outcomes, interventions lasting at least 8 weeks, publication after January 2015. Publications were excluded if they evaluated telehealth or used web-based software for desktop devices only. The quality of the included studies was analyzed with the Cochrane Collaboration Risk of Bias Tool and the Methodological Index for Non-Randomized Studies.

Results: We included 30 studies (20 focused on BC), encompassing 5,691 patients with cancer (median 113, IQR, 135.5). Among these, 3,606 had BC (median 99, IQR, 75). All studies contained multiple interventions, including physical activity, tailored information for self-management of the disease, and symptom tracker. Interventions showed better results on self-efficacy (3/3), QoL (10/14), and physical activity (5/7). Lifestyle programs (3/3), expert consulting (4/4), and tailored information (10/11) yielded the best results. Apps with interactive support had a higher rate of positive findings, while interventions targeted to survivors showed worse results. mHealth tools were not available to the public in most of the studies (17/30).

Conclusion: mHealth interventions yielded heterogeneous results on different outcomes. Identifying lack of evidence on clinical scenarios (eg, patients undergoing systemic therapy other than chemotherapy) could aid in refining strategic planning for forthcoming research endeavors within this field.

Purpose: Dynamic operations platforms allow for cross-platform data extraction, integration, and analysis, although application of these platforms to large-scale oncology enterprises has not been described. This study presents a pipeline for automated, high-fidelity extraction, integration, and validation of cross-platform oncology data in patients undergoing treatment for rectal cancer at a single, high-volume institution.

Methods: A dynamic operations platform was used to identify patients with rectal cancer treated at MD Anderson Cancer Center between 2016 and 2022 who had magnetic resonance imaging (MRI) imaging and preoperative treatment details available in the electronic health record (EHR). Demographic, clinicopathologic, tumor mutation, radiographic, and treatment data were extracted from the EHR using a methodology adaptable to any disease site. Data accuracy was assessed by manual review. Accuracy before and after implementation of synoptic reporting was determined for MRI data.

Results: A total of 516 patients with localized rectal cancer were included. In the era after institutional adoption of synoptic reports, the dynamic operations platform extracted T (tumor) category data from the EHR with 95% accuracy compared with 87% before the use of synoptic reports, and N (lymph node) category with 88% compared with 58%. Correct extraction of pelvic sidewall adenopathy was 94% compared with 78%, and extramural vascular invasion accuracy was 99% compared with 89%. Neoadjuvant chemotherapy and radiation data were 99% accurate for patients who had synoptic data sources.

Conclusion: Using dynamic operations platforms enables automated cross-platform integration of multiparameter oncology data with high fidelity in patients undergoing multimodality treatment for rectal cancer. These pipelines can be adapted to other solid tumors and, together with standardized reporting, can increase efficiency in clinical research and the translation of actionable findings toward optimizing patient outcomes.