JCO Clinical Cancer Informatics最新文献

Purpose: Extracting inclusion and exclusion criteria in a structured, automated fashion remains a challenge to developing better search functionalities or automating systematic reviews of randomized controlled trials in oncology. The question "Did this trial enroll patients with localized disease, metastatic disease, or both?" could be used to narrow down the number of potentially relevant trials when conducting a search.

Methods: Six hundred trials from high-impact medical journals were classified depending on whether they allowed for the inclusion of patients with localized and/or metastatic disease. Five hundred trials were used to develop and validate three different models, with 100 trials being stored away for testing. The test set was also used to evaluate the performance of GPT-4o in the same task.

Results: In the test set, a rule-based system using regular expressions achieved F1 scores of 0.72 for the prediction of whether the trial allowed for the inclusion of patients with localized disease and 0.77 for metastatic disease. A transformer-based machine learning (ML) model achieved F1 scores of 0.97 and 0.88, respectively. A combined approach where the rule-based system was allowed to over-rule the ML model achieved F1 scores of 0.97 and 0.89, respectively. GPT-4o achieved F1 scores of 0.87 and 0.92, respectively.

Conclusion: Automatic classification of cancer trials with regard to the inclusion of patients with localized and/or metastatic disease is feasible. Turning the extraction of trial criteria into classification problems could, in selected cases, improve text-mining approaches in evidence-based medicine. Increasingly large language models can reduce or eliminate the need for previous training on the task at the expense of increased computational power and, in turn, cost.

Purpose: Thyroid nodules are common in the general population, and assessing their malignancy risk is the initial step in care. Surgical exploration remains the sole definitive option for indeterminate nodules. Extensive database access is crucial for improving this initial assessment. Our objective was to develop an automated process using convolutional neural networks (CNNs) to extract and structure biomedical insights from electronic health reports (EHRs) in a large thyroid pathology cohort.

Materials and methods: We randomly selected 1,500 patients with thyroid pathology from our cohort for model development and an additional 100 for testing. We then divided the cohort of 1,500 patients into training (70%) and validation (30%) sets. We used EHRs from initial surgeon visits, preanesthesia visits, ultrasound, surgery, and anatomopathology reports. We selected 42 variables of interest and had them manually annotated by a clinical expert. We developed RUBY-THYRO using six distinct CNN models from SpaCy, supplemented with keyword extraction rules and postprocessing. Evaluation against a gold standard database included calculating precision, recall, and F1 score.

Results: Performance remained consistent across the test and validation sets, with the majority of variables (30/42) achieving performance metrics exceeding 90% for all metrics in both sets. Results differed according to the variables; pathologic tumor stage score achieved 100% in precision, recall, and F1 score, versus 45%, 28%, and 32% for the number of nodules in the test set, respectively. Surgical and preanesthesia reports demonstrated particularly high performance.

Conclusion: Our study successfully implemented a CNN-based natural language processing (NLP) approach for extracting and structuring data from various EHRs in thyroid pathology. This highlights the potential of artificial intelligence-driven NLP techniques for extensive and cost-effective data extraction, paving the way for creating comprehensive, hospital-wide data warehouses.

Purpose: Toxicity to systemic cancer treatment represents a major anxiety for patients and a challenge to treatment plans. We aimed to develop machine learning algorithms for the upfront prediction of an individual's risk of experiencing treatment-relevant toxicity during the course of treatment.

Methods: Clinical records were retrieved from a single-center, consecutive cohort of patients who underwent neoadjuvant treatment for early breast cancer. We developed and validated machine learning algorithms to predict grade 3 or 4 toxicity (anemia, neutropenia, deviation of liver enzymes, nephrotoxicity, thrombopenia, electrolyte disturbance, or neuropathy). We used 10-fold cross-validation to develop two algorithms (logistic regression with elastic net penalty [GLM] and support vector machines [SVMs]). Algorithm predictions were compared with documented toxicity events and diagnostic performance was evaluated via area under the curve (AUROC).

Results: A total of 590 patients were identified, 432 in the development set and 158 in the validation set. The median age was 51 years, and 55.8% (329 of 590) experienced grade 3 or 4 toxicity. The performance improved significantly when adding referenced treatment information (referenced regimen, referenced summation dose intensity product) in addition to patient and tumor variables: GLM AUROC 0.59 versus 0.75, P = .02; SVM AUROC 0.64 versus 0.75, P = .01.

Conclusion: The individual risk of treatment-relevant toxicity can be predicted using machine learning algorithms. We demonstrate a promising way to improve efficacy and facilitate proactive toxicity management of systemic cancer treatment.

Purpose: Research on rare diseases and atypical health care demographics is often slowed by high interparticipant heterogeneity and overall scarcity of data. Synthetic data (SD) have been proposed as means for data sharing, enlargement, and diversification, by artificially generating real phenomena while obscuring the real patient data. The utility of SD is actively scrutinized in health care research, but the role of sample size for actionability of SD is insufficiently explored. We aim to understand the interplay of actionability and sample size by generating SD sets of varying sizes from gradually diminishing amounts of real individuals' data. We evaluate the actionability of SD in a highly heterogeneous and rare demographic: adolescents and young adults (AYAs) with cancer.

Methods: A population-based cross-sectional cohort study of 3,735 AYAs was subsampled at random to produce 13 training data sets of varying sample sizes. We studied four distinct generator architectures built on the open-source Synthetic Data Vault library. Each architecture was used to generate SD of varying sizes on the basis of each aforementioned training subsets. SD actionability was assessed by comparing the resulting SD with their respective real data against three metrics-veracity, utility, and privacy concealment.

Results: All examined generator architectures yielded actionable data when generating SD with sizes similar to the real data. Large SD sample size increased veracity but generally increased privacy risks. Using fewer training participants led to faster convergence in veracity, but partially exacerbated privacy concealment issues.

Conclusion: SD is a potentially promising option for data sharing and data augmentation, yet sample size plays a significant role in its actionability. SD generation should go hand-in-hand with consistent scrutiny, and sample size should be carefully considered in this process.

Purpose: This retrospective cohort study evaluated whether linked electronic health record (EHR) pharmacy data were adequately complete and timely to detect primary nonadherence to breast cancer adjuvant endocrine therapy (AET).

Materials and methods: Linked EHR pharmacy data were extracted from the EHR for patients with stage 0 to III breast cancer who had their first prescription order for AET between 2016 and 2021. Patients with the first dispense event within 90 days of the prescription were classified as having sufficient or insufficient data available for early detection of primary adherence.

Results: A total of 1,446 eligible patients had a first AET prescription order between 2016 and 2021; these orders were routed to 871 unique pharmacies, of which 856 (98.2%) were contracted with the linked EHR pharmacy database and 15 (1.8%) were not contracted. Among the 1,428 patients with a first prescription sent to a contract pharmacy, 164 (13%) had incomplete linked EHR pharmacy data refresh events to assess primary adherence. Among the 1,244 patients with at least 1 refresh event after their first prescription, 82% occurred within 90 days and were sufficiently timely for early detection of primary adherence. Overall, 32% of patients would benefit from an intervention to verify or improve primary adherence to AET.

Conclusion: Although linked EHR pharmacy data have adequate completeness of contract pharmacy data, local configurations of data refresh events tailored to medication reconciliation workflows are incomplete (13%) and insufficiently timely (32%) to fully support clinical decision support (CDS) for early detection of primary medication nonadherence. Prospective CDS interventions using linked EHR pharmacy data are possible with enhancements to the frequency and timeliness of refresh events.

Purpose: Oral adjuvant endocrine therapy (AET) reduces the risk of cancer recurrence and death for women with hormone receptor-positive (HR+) breast cancer. Because of adverse symptoms and socioecologic barriers, AET adherence rates are low. We conducted post hoc analyses of a randomized trial of a remote symptom and adherence monitoring app to evaluate characteristics associated with higher app use, satisfaction, and how app use was associated with AET adherence.

Methods: Patients prescribed AET were randomly assigned to receive one of three intervention conditions: app, app + feedback, or enhanced usual care. Baseline and 6-month follow-up surveys, app use, and pillbox-monitored AET adherence data for app and app + feedback participants were used. Logistic regression evaluated the association between sociodemographic/clinical characteristics and app utilization and satisfaction, and how app use was associated with AET adherence (>80%).

Results: Overall, 163 women with early-stage HR+ breast cancer were included; 35.0% had high app use (≥75% of weeks enrolled). No sociodemographic characteristics were associated with app use. Satisfaction with the app was higher among those who were younger (88.9% for age 31-49 years v 54.9% for age 65+ years, P < .001), identified as White (76.8% v 60.1% for Black, P = .045), had lower health literacy (85.4% v 68.2% with higher health literacy, P = .017), or were nonurban residents (85.7% v 68.6% for urban, P = .021). Most participants (90.3%) with high app use were AET-adherent compared with 66.8% for those with lower app use (P < .001).

Conclusion: Use of a remote monitoring app was similar across sociodemographic characteristics, and more frequent app use was associated with a higher likelihood of 6-month AET adherence. Encouraging women to monitor medication adherence and communicate adverse symptoms could improve AET adherence.

Purpose: This study developed and validated a novel deep learning radiomic biomarker to estimate response to immune checkpoint inhibitor (ICI) therapy in advanced non-small cell lung cancer (NSCLC) using real-world data (RWD) and clinical trial data.

Materials and methods: Retrospective RWD of 1,829 patients with advanced NSCLC treated with PD-(L)1 ICIs were collected from 10 academic and community institutions in the United States and Europe. The RWD included data sets for discovery (Data Set A-Discovery, n = 1,173) and independent test (Data Set B, n = 458). A radiomic pipeline, containing a deep learning feature extractor and a survival model, generated the computed tomography (CT) response score (CTRS) applied to the pretreatment routine CT/positron emission tomography (PET)-CT scan. An enhanced CTRS (eCTRS) also incorporated age, sex, treatment line, and lesion annotations. Performance was evaluated against progression-free survival (PFS) and overall survival (OS). Biomarker generalizability was further evaluated using a secondary analysis of a prospective clinical trial (ClinicalTrials.gov identifier: NCT02573259) evaluating the PD-1 inhibitor sasanlimab in second or later line of treatment (Data Set C, n = 54).

Results: In RWD Test Data Set B, the CTRS identified patients with a high probability of response to ICI with a PFS hazard ratio (HR) of 0.46 (95% CI, 0.26 to 0.82) and an OS HR of 0.50 (95% CI, 0.28 to 0.92) in the first-line ICI monotherapy cohort, after adjustment for baseline covariates including the PD-L1 tumor proportion score. In Clinical Trial Data Set C, the CTRS demonstrated an adjusted PFS HR of 1.03 (95% CI, 0.43 to 2.47) and an OS HR of 0.33 (95% CI, 0.14 to 0.91). The CTRS and eCTRS outperformed traditional imaging biomarkers of lesion size in PFS and OS for RWD Test Data Set B and in OS for the Clinical Trial Data Set.

Conclusion: The study developed and validated a deep learning radiomic biomarker using pretreatment routine CT/PET-CT scans to identify ICI benefit in advanced NSCLC.

Purpose: Electronic patient-reported outcome measures (ePROMs) are increasingly collected routinely in clinical practice and may be prognostic for survival in adults with advanced non-small cell lung cancer (NSCLC) in addition to clinical data. This study developed ePROM-enhanced models for predicting 1-year overall survival in patients with advanced NSCLC at the start of immunotherapy.

Methods: This is a single-center study using consecutive patients from a tertiary cancer hospital in England. Using Cox proportional hazards models, we developed one clinical factor-only model and three ePROM-enhanced models, each including one of the following factors: quality of life (as measured by EuroQoL five-dimension five-level utility score) and overall symptom burden and number of moderate-to-severe symptoms (as measured by patient-reported version of Common Terminology Criteria for Adverse Events). Predictive performance was evaluated and compared through bootstrapping internal validation, and clinical utility was determined via decision curve analysis.

Results: The clinical factor-only model contained age, histology, performance status, and neutrophile-to-lymphocyte ratio. While calibration was similar between the clinical factor-only and ePROM-enhanced models, the latter showed improved discrimination by 0.020 (95% CI, 0.011 to 0.024), 0.024 (95% CI, 0.016 to 0.031), and 0.024 (95% CI, 0.014 to 0.029) when enhanced with ePROMs on quality of life, overall symptom burden, and number of moderate-to-severe symptoms, respectively. If care decisions are to be made at risk thresholds between 25% and 75%, the ePROM-enhanced models led to higher net benefit than the clinical factor-only model and the default strategies of intervention for all and intervention for none.

Conclusion: The ePROM-enhanced models outperformed the clinical factor-only model in predicting 1-year overall survival for patients with advanced NSCLC receiving immunotherapy and showed potential clinical utility for informing decisions in this population. Future studies should focus on validating the models in external data sets.

Purpose: Breast density is a widely established independent breast cancer risk factor. With the increasing utilization of digital breast tomosynthesis (DBT) in breast cancer screening, there is an opportunity to estimate volumetric breast density (VBD) routinely. However, current available methods extrapolate VBD from two-dimensional (2D) images acquired using DBT and/or depend on the existence of raw DBT data, which is rarely archived by clinical centers because of storage constraints.

Methods: We retrospectively analyzed 1,080 nonactionable three-dimensional (3D) reconstructed DBT screening examinations acquired between 2011 and 2016. Reference tissue segmentations were generated using previously validated software that uses 3D reconstructed slices and raw 2D DBT data. We developed a deep learning (DL) model that segments dense and fatty breast tissue from background. We then applied this model to estimate %VBD and absolute dense volume (ADV) in cm³ in a separate case-control sample (180 cases and 654 controls). We created two conditional logistic regression models, relating each model-derived density measurement to likelihood of contralateral breast cancer diagnosis, adjusted for age, BMI, family history, and menopausal status.

Results: The DL model achieved unweighted and weighted Dice scores of 0.88 (standard deviation [SD] = 0.08) and 0.76 (SD = 0.15), respectively, on the held-out test set, demonstrating good agreement between the model and 3D reference segmentations. There was a significant association between the odds of breast cancer diagnosis and model-derived VBD (odds ratio [OR], 1.41 [95 % CI, 1.13 to 1.77]; P = .002), with an AUC of 0.65 (95% CI, 0.60 to 0.69). ADV was also significantly associated with breast cancer diagnosis (OR, 1.45 [95% CI, 1.22 to 1.73]; P < .001) with an AUC of 0.67 (95% CI, 0.62 to 0.71).

Conclusion: DL-derived density measures derived from 3D reconstructed DBT images are associated with breast cancer diagnosis.