Cohort Selection For Clinical Trials Research Articles

Somatic genomic mosaicism in the brain: Identified mutations provide challenges and opportunities for the clinic

Recent work by Bae etal.1 represents a major next-generation sequencing effort to identify somatic genomic mosaicism in normal and diseased human brains. Some samples displayed age-associated hypermutability, and the general possibility that somatic mutations can drive brain disease has implications for new therapeutic strategies, disease staging, biomarkers, and cohort selection for clinical trials.

Integrating additional factors into the TNM staging for cutaneous melanoma by machine learning.

BackgroundIntegrating additional factors into the TNM staging system is needed for more accurate risk classification and survival prediction for patients with cutaneous melanoma. In the present study, we introduce machine learning as a novel tool that incorporates additional prognostic factors to improve the current TNM staging system.Methods and findingsCancer-specific survival data for cutaneous melanoma with at least a 5 years follow-up were extracted from the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute and split into the training set (40,781 cases) and validation set (5,390 cases). Five factors were studied: the primary tumor (T), regional lymph nodes (N), distant metastasis (M), age (A), and sex (S). The Ensemble Algorithm for Clustering Cancer Data (EACCD) was applied to the training set to generate prognostic groups. Utilizing only T, N, and M, a basic prognostic system was built where patients were stratified into 10 prognostic groups with well-separated survival curves, similar to 10 AJCC stages. These 10 groups had a significantly higher accuracy in survival prediction than 10 stages (C-index = 0.7682 vs 0.7643; increase in C-index = 0.0039, 95% CI = (0.0032, 0.0047); p-value = 7.2×10−23). Nevertheless, a positive association remained between the EACCD grouping and the AJCC staging (Spearman’s rank correlation coefficient = 0.8316; p-value = 4.5×10−13). With additional information from A and S, a more advanced prognostic system was established using the training data that stratified patients into 10 groups and further improved the prediction accuracy (C-index = 0.7865 vs 0.7643; increase in C-index = 0.0222, 95% CI = (0.0191, 0.0254); p-value = 8.8×10−43). Both internal validation using the training set and temporal validation using the validation set showed good stratification and a high predictive accuracy of the prognostic systems.ConclusionsThe EACCD allows additional factors to be integrated into the TNM to create a prognostic system that improves patient stratification and survival prediction for cutaneous melanoma. This integration separates favorable from unfavorable clinical outcomes for patients and improves both cohort selection for clinical trials and treatment management.

Understanding Fatigue in Primary Biliary Cholangitis.

Fatigue affects 50% of primary biliary cholangitis patients and is severe in approximately 20%, significantly affecting quality of life. The pathogenesis of fatigue in primary biliary cholangitis is poorly understood. This study aimed to explore subgroups of fatigue to support targeting of selected groups in future clinical trials. Data were derived from the UK-PBC cohort. Participants completed the PBC-40, Hospital Anxiety and Depression Score, Epworth Sleepiness Scale, and Orthostatic Grading Scale for symptoms assessment. Fatigue and cognitive symptoms were regarded as clinically significant if they exceeded the previously defined cutoff for "moderate" symptom. Of 2002, patients for whom full PBC-40, fatigue, and cognitive symptom domain scores were available, 1203 (60%) had significant fatigue and 730 (36%) had significant cognitive symptoms. Among the 1203 patients with significant fatigue, 663 (55%) also had significant cognitive symptoms (termed fatigue with cognitive symptoms) with a significant linear association between the fatigue and cognitive symptom severity. "Fatigue with cognitive symptoms" patients were younger and more likely to have severe fatigue. They also experienced greater social and emotional impact. Fatigue in PBC is complex, and there has been no progress to date in identifying therapies able to improve it. One factor in slow progress may be the heterogeneity of patients describing fatigue complicating effective cohort selection for clinical trials. This study has identified potential discrete subgroups of fatigued patients with and without cognitive symptoms. The group of patients expressing "fatigue with cognitive symptoms" was homogenous and may represent a coherent cohort for clinical trials.

Abstract 2037: Association of molecular heterogeneity and actionable mutations with survival in lung cancer

Abstract Introduction: Today, the use of new technologies such as electronic medical records and testing methods are standard throughout healthcare centers. These new tools deliver an abundance of information to physicians, researchers, and the medical community. Harnessing the potential of this data can be an influential instrument to assist physicians in offering personalized medicine to patients and conduct complex research with ease. Nevertheless, the development, infrastructure, and implementation of distinct disease-specific registries for the collection, storage, and utilization of this data is crucial to benefit from this vital clinical and research information. Patients and Methods: In this study, we utilized an institutional thoracic patient registry complete with clinical and genomic data to carry out an in-depth analysis on lung adenocarcinoma patients (N=415). The patients selected and evaluated for analysis were seen at City of Hope from 2008 to 2016. The data was collected between 2016 and 2018 through retrospective chart review and compiled within the Thoracic Oncology Registry. This study was approved by the Institutional Review Board at COH under IRB 18217 and was conducted according to the Declaration of Helsinki. Data was de-identified and analyzed anonymously. Results: The lung adenocarcinoma median age was 64 (range, 22-92) years old, and the median OS was 33.29 months (95% CI, 29.77-39.48). The majority of patients were female (245/415; 59%). All the patients had metastatic disease, while the bulk of the patients (369/415; 89%) presented as Stage IV at the time of initial diagnosis. Oncogenes with the most commonly occurring mutations were found in EGFR (207/415; 50%), KRAS (97/352; 28%), and ALK rearrangement (28/377; 7%), while most frequent alterations in tumor suppressor genes were in TP53 (140/283; 49%), LRP1B (63/228; 28%), and STK11 (39/278; 14%). The top three actionable mutations identified include EGFR (177/415; 42.7%), ALK (28/377; 7.4%), and BRAF V600E (7/288; 2.4%). When survival was compared between the number of patients with actionable mutations (222/415; 53.5%) and wild-type patients (193/415; 46.5%), there was a substantial difference in median OS (39.8 months v 26.0 months, P<0.001). Comparing the TCGA and AACR Project GENIE dataset to our database, there was a surprising difference in the percentage of actionable mutations (15.0% v 22.3% v 53.5%). Conclusion: Our results showed an increased median overall survival in patients with lung adenocarcinoma that harbor actionable mutations compared to wild-type patients. Exploiting the appropriate use of clinical and genomic data collected and maintained in patient registries may better account for patient populations seen in the academic and community settings to improve cohort selection for clinical trials and survival outcomes. Citation Format: Jeremy Fricke, Isa Mambetsariev, Rebecca Pharaon, Yingyu Wang, Chen Chen, Fang Qiu, Lynette Smith, Surinder K. Batra, Wasim M. Nasser, Erminia Massarelli, Marianna Koczywas, Karen Reckamp, Ravi Salgia. Association of molecular heterogeneity and actionable mutations with survival in lung cancer [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2037.

Cohort selection for clinical trials using multiple instance learning.

Identifying patients eligible for clinical trials using electronic health records (EHRs) is a challenging task usually requiring a comprehensive analysis of information stored in multiple EHRs of a patient. The goal of this study is to investigate different methods and their effectiveness in identifying patients that meet specific eligibility selection criteria based on patients' longitudinal records. An unstructured dataset released by the n2c2 cohort selection for clinical trials track was used, each of which included 2-5 records manually annotated to thirteen pre-defined selection criteria. Unlike the other studies, we formulated the problem as a multiple instance learning (MIL) task and compared the performance with that of the rule-based and the single instance-based classifiers. Our official best run achieved an average micro-F score of 0.8765 which was ranked as one of the top ten results in the track. Further experiments demonstrated that the performance of the MIL-based classifiers consistently yield better performance than their single-instance counterparts in the criteria that require the overall comprehension of the information distributed among all of the patient's EHRs. Rule-based and single instance learning approaches exhibited better performance in criteria that don't require a consideration of several factors across records. This study demonstrated that cohort selection using longitudinal patient records can be formulated as a MIL problem. Our results exhibit that the MIL-based classifiers supplement the rule-based methods and provide better results in comparison to the single instance learning approaches.

Precision medicine and actionable alterations in lung cancer: A single institution experience.

Oncology has become more reliant on new testing methods and a greater use of electronic medical records, which provide a plethora of information available to physicians and researchers. However, to take advantage of vital clinical and research data for precision medicine, we must initially make an effort to create an infrastructure for the collection, storage, and utilization of this information with uniquely designed disease-specific registries that could support the collection of a large number of patients. In this study, we perform an in-depth analysis of a series of lung adenocarcinoma patients (n = 415) with genomic and clinical data in a recently created thoracic patient registry. Of the 415 patients with lung adenocarcinoma, 59% (n = 245) were female; the median age was 64 (range, 22-92) years with a median OS of 33.29 months (95% CI, 29.77-39.48). The most common actionable alterations were identified in EGFR (n = 177/415 [42.7%]), ALK (n = 28/377 [7.4%]), and BRAF V600E (n = 7/288 [2.4%]). There was also a discernible difference in survival for 222 patients, who had an actionable alteration, with a median OS of 39.8 months as compared to 193 wild-type patients with a median OS of 26.0 months (P<0.001). We identified an unprecedented number of actionable alterations [53.5% (222/415)], including distinct individual alteration rates, as compared with 15.0% and 22.3% in TCGA and GENIE respectively. The use of patient registries, focused genomic panels and the appropriate use of clinical guidelines in community and academic settings may influence cohort selection for clinical trials and improve survival outcomes.

Cohort Selection for Clinical Trials From Longitudinal Patient Records: Text Mining Approach.

BackgroundClinical trials are an important step in introducing new interventions into clinical practice by generating data on their safety and efficacy. Clinical trials need to ensure that participants are similar so that the findings can be attributed to the interventions studied and not to some other factors. Therefore, each clinical trial defines eligibility criteria, which describe characteristics that must be shared by the participants. Unfortunately, the complexities of eligibility criteria may not allow them to be translated directly into readily executable database queries. Instead, they may require careful analysis of the narrative sections of medical records. Manual screening of medical records is time consuming, thus negatively affecting the timeliness of the recruitment process.ObjectiveTrack 1 of the 2018 National Natural Language Processing Clinical Challenge focused on the task of cohort selection for clinical trials, aiming to answer the following question: Can natural language processing be applied to narrative medical records to identify patients who meet eligibility criteria for clinical trials? The task required the participating systems to analyze longitudinal patient records to determine if the corresponding patients met the given eligibility criteria. We aimed to describe a system developed to address this task.MethodsOur system consisted of 13 classifiers, one for each eligibility criterion. All classifiers used a bag-of-words document representation model. To prevent the loss of relevant contextual information associated with such representation, a pattern-matching approach was used to extract context-sensitive features. They were embedded back into the text as lexically distinguishable tokens, which were consequently featured in the bag-of-words representation. Supervised machine learning was chosen wherever a sufficient number of both positive and negative instances was available to learn from. A rule-based approach focusing on a small set of relevant features was chosen for the remaining criteria.ResultsThe system was evaluated using microaveraged F measure. Overall, 4 machine algorithms, including support vector machine, logistic regression, naïve Bayesian classifier, and gradient tree boosting (GTB), were evaluated on the training data using 10–fold cross-validation. Overall, GTB demonstrated the most consistent performance. Its performance peaked when oversampling was used to balance the training data. The final evaluation was performed on previously unseen test data. On average, the F measure of 89.04% was comparable to 3 of the top ranked performances in the shared task (91.11%, 90.28%, and 90.21%). With an F measure of 88.14%, we significantly outperformed these systems (81.03%, 78.50%, and 70.81%) in identifying patients with advanced coronary artery disease.ConclusionsThe holdout evaluation provides evidence that our system was able to identify eligible patients for the given clinical trial with high accuracy. Our approach demonstrates how rule-based knowledge infusion can improve the performance of machine learning algorithms even when trained on a relatively small dataset.

Cohort selection for clinical trials: n2c2 2018 shared task track 1.

Track 1 of the 2018 National NLP Clinical Challenges shared tasks focused on identifying which patients in a corpus of longitudinal medical records meet and do not meet identified selection criteria. To address this challenge, we annotated American English clinical narratives for 288 patients according to whether they met these criteria. We chose criteria from existing clinical trials that represented a variety of natural language processing tasks, including concept extraction, temporal reasoning, and inference. A total of 47 teams participated in this shared task, with 224 participants in total. The participants represented 18 countries, and the teams submitted 109 total system outputs. The best-performing system achieved a micro F1 score of 0.91 using a rule-based approach. The top 10 teams used rule-based and hybrid systems to approach the problems. Clinical narratives are open to interpretation, particularly in cases where the selection criterion may be underspecified. This leaves room for annotators to use domain knowledge and intuition in selecting patients, which may lead to error in system outputs. However, teams who consulted medical professionals while building their systems were more likely to have high recall for patients, which is preferable for patient selection systems. There is not yet a 1-size-fits-all solution for natural language processing systems approaching this task. Future research in this area can look to examining criteria requiring even more complex inferences, temporal reasoning, and domain knowledge.

Cohort selection for clinical trials using deep learning models.

ObjectiveThe goal of the 2018 n2c2 shared task on cohort selection for clinical trials (track 1) is to identify which patients meet the selection criteria for clinical trials. Cohort selection is a particularly demanding task to which natural language processing and deep learning can make a valuable contribution. Our goal is to evaluate several deep learning architectures to deal with this task.Materials and MethodsCohort selection can be formulated as a multilabeling problem whose goal is to determine which criteria are met for each patient record. We explore several deep learning architectures such as a simple convolutional neural network (CNN), a deep CNN, a recurrent neural network (RNN), and CNN-RNN hybrid architecture. Although our architectures are similar to those proposed in existing deep learning systems for text classification, our research also studies the impact of using a fully connected feedforward layer on the performance of these architectures.ResultsThe RNN and hybrid models provide the best results, though without statistical significance. The use of the fully connected feedforward layer improves the results for all the architectures, except for the hybrid architecture.ConclusionsDespite the limited size of the dataset, deep learning methods show promising results in learning useful features for the task of cohort selection. Therefore, they can be used as a previous filter for cohort selection for any clinical trial with a minimum of human intervention, thus reducing the cost and time of clinical trials significantly.

Cohort selection for clinical trials using hierarchical neural network.

Cohort selection for clinical trials is a key step for clinical research. We proposed a hierarchical neural network to determine whether a patient satisfied selection criteria or not. We designed a hierarchical neural network (denoted as CNN-Highway-LSTM or LSTM-Highway-LSTM) for the track 1 of the national natural language processing (NLP) clinical challenge (n2c2) on cohort selection for clinical trials in 2018. The neural network is composed of 5 components: (1) sentence representation using convolutional neural network (CNN) or long short-term memory (LSTM) network; (2) a highway network to adjust information flow; (3) a self-attention neural network to reweight sentences; (4) document representation using LSTM, which takes sentence representations in chronological order as input; (5) a fully connected neural network to determine whether each criterion is met or not. We compared the proposed method with its variants, including the methods only using the first component to represent documents directly and the fully connected neural network for classification (denoted as CNN-only or LSTM-only) and the methods without using the highway network (denoted as CNN-LSTM or LSTM-LSTM). The performance of all methods was measured by micro-averaged precision, recall, and F1 score. The micro-averaged F1 scores of CNN-only, LSTM-only, CNN-LSTM, LSTM-LSTM, CNN-Highway-LSTM, and LSTM-Highway-LSTM were 85.24%, 84.25%, 87.27%, 88.68%, 88.48%, and 90.21%, respectively. The highest micro-averaged F1 score is higher than our submitted 1 of 88.55%, which is 1 of the top-ranked results in the challenge. The results indicate that the proposed method is effective for cohort selection for clinical trials. Although the proposed method achieved promising results, some mistakes were caused by word ambiguity, negation, number analysis and incomplete dictionary. Moreover, imbalanced data was another challenge that needs to be tackled in the future. In this article, we proposed a hierarchical neural network for cohort selection. Experimental results show that this method is good at selecting cohort.

Clinical trial cohort selection based on multi-level rule-based natural language processing system.

Identifying patients who meet selection criteria for clinical trials is typically challenging and time-consuming. In this article, we describe our clinical natural language processing (NLP) system to automatically assess patients' eligibility based on their longitudinal medical records. This work was part of the 2018 National NLP Clinical Challenges (n2c2) Shared-Task and Workshop on Cohort Selection for Clinical Trials. The authors developed an integrated rule-based clinical NLP system which employs a generic rule-based framework plugged in with lexical-, syntactic- and meta-level, task-specific knowledge inputs. In addition, the authors also implemented and evaluated a general clinical NLP (cNLP) system which is built with the Unified Medical Language System and Unstructured Information Management Architecture. The systems were evaluated as part of the 2018 n2c2-1 challenge, and authors' rule-based system obtained an F-measure of 0.9028, ranking fourth at the challenge and had less than 1% difference from the best system. While the general cNLP system didn't achieve performance as good as the rule-based system, it did establish its own advantages and potential in extracting clinical concepts. Our results indicate that a well-designed rule-based clinical NLP system is capable of achieving good performance on cohort selection even with a small training data set. In addition, the investigation of a Unified Medical Language System-based general cNLP system suggests that a hybrid system combining these 2 approaches is promising to surpass the state-of-the-art performance.

Hybrid bag of approaches to characterize selection criteria for cohort identification.

The 2018 National NLP Clinical Challenge (2018 n2c2) focused on the task of cohort selection for clinical trials, where participating systems were tasked with analyzing longitudinal patient records to determine if the patients met or did not meet any of the 13 selection criteria. This article describes our participation in this shared task. We followed a hybrid approach combining pattern-based, knowledge-intensive, and feature weighting techniques. After preprocessing the notes using publicly available natural language processing tools, we developed individual criterion-specific components that relied on collecting knowledge resources relevant for these criteria and pattern-based and weighting approaches to identify "met" and "not met" cases. As part of the 2018 n2c2 challenge, 3 runs were submitted. The overall micro-averaged F1 on the training set was 0.9444. On the test set, the micro-averaged F1 for the 3 submitted runs were 0.9075, 0.9065, and 0.9056. The best run was placed second in the overall challenge and all 3 runs were statistically similar to the top-ranked system. A reimplemented system achieved the best overall F1 of 0.9111 on the test set. We highlight the need for a focused resource-intensive effort to address the class imbalance in the cohort selection identification task. Our hybrid approach was able to identify all selection criteria with high F1 performance on both training and test sets. Based on our participation in the 2018 n2c2 task, we conclude that there is merit in continuing a focused criterion-specific analysis and developing appropriate knowledge resources to build a quality cohort selection system.

Precision medicine and actionable alterations in lung cancer.

e20707 Background: In recent years, healthcare centers have become more reliant on the advent of new technology, testing methods and a greater use of electronic medical records, which provide a plethora of information available to physicians, researchers, and the medical community. This aggregate of information stands to be a powerful tool that can enable one to perform highly complex investigations more easily, as well as aid physicians in their efforts to provide personalized medicine to patients. However, in order to take advantage of vital clinical and research data, we must initially make an effort to create an effective infrastructure for the collection, storage, utilization, and protection of this information with uniquely designed disease-specific registries that have adequate informatics support to allow for the collection of a large number of patients. Methods: In this study, we perform an in-depth analysis of a series of lung adenocarcinoma patients (n = 415) with genomic and clinical data in a recently created thoracic patient registry. Results: Of the 415 patients in the analysis, 59% (n = 245) were female; the median age was 64 (range, 22-92) years with a median OS of 33.29 months (95% CI, 29.77-39.48). The frequency of the most commonly occurring oncogenes was 50% EGFR (n = 207/415), 28% KRAS (n = 97/352), and 7% ALK rearrangement (n = 28/377), while the most commonly occurring tumor suppressor genes consisted of TP53 (n = 140/283 [49%]), LRP1B (n = 63/228 [28%]), and STK11 (n = 39/278 [14%]). The most common actionable alterations were identified in EGFR L858R/exon 19 deletion (n = 177/415 [42.7%]), ALK rearrangement (n = 28/377 [7.4%]), ROS1 rearrangement (n = 3/257 [1.2%]), BRAF V600E (n = 7/288 [2.4%]) and MET exon 14 splice site/deletion (n = 7/287 [2.4%]). While there were no median OS differences in patients who were tested under a broad-panel vs small-panel (33.4 months vs 33.5 months; 95% CI; P = 0.36), there was a discernible difference in survival for 222 patients, who had an actionable alteration, with a median OS of 39.8 months as compared to 193 patients who were wild-type with a median OS of 26.0 months (95% CI; P &lt; 0.001). In addition, we identified an unprecedented number of patients with actionable alterations [53.5% (222/415)], including distinct individual alteration rates, as compared with 15.0% and 22.3% in TCGA and GENIE respectively. Conclusions: The use of patient registries, focused genomic panels and the appropriate use of clinical guidelines in community and academic settings may influence cohort selection for clinical trials and improve survival outcomes.

TADPOLE Challenge: Accurate Alzheimer's disease prediction through crowdsourced forecasting of future data.

The Alzheimer's Disease Prediction Of Longitudinal Evolution (TADPOLE) Challenge compares the performance of algorithms at predicting the future evolution of individuals at risk of Alzheimer's disease. TADPOLE Challenge participants train their models and algorithms on historical data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. Participants are then required to make forecasts of three key outcomes for ADNI-3 rollover participants: clinical diagnosis, Alzheimer's Disease Assessment Scale Cognitive Subdomain (ADAS-Cog 13), and total volume of the ventricles - which are then compared with future measurements. Strong points of the challenge are that the test data did not exist at the time of forecasting (it was acquired afterwards), and that it focuses on the challenging problem of cohort selection for clinical trials by identifying fast progressors. The submission phase of TADPOLE was open until 15 November 2017; since then data has been acquired until April 2019 from 219 subjects with 223 clinical visits and 150 Magnetic Resonance Imaging (MRI) scans, which was used for the evaluation of the participants' predictions. Thirty-three teams participated with a total of 92 submissions. No single submission was best at predicting all three outcomes. For diagnosis prediction, the best forecast (team Frog), which was based on gradient boosting, obtained a multiclass area under the receiver-operating curve (MAUC) of 0.931, while for ventricle prediction the best forecast (team EMC1 ), which was based on disease progression modelling and spline regression, obtained mean absolute error of 0.41% of total intracranial volume (ICV). For ADAS-Cog 13, no forecast was considerably better than the benchmark mixed effects model (BenchmarkME ), provided to participants before the submission deadline. Further analysis can help understand which input features and algorithms are most suitable for Alzheimer's disease prediction and for aiding patient stratification in clinical trials. The submission system remains open via the website: https://tadpole.grand-challenge.org/.

The clinical implications of integrating additional prognostic factors into the TNM

The management of solid tumors is governed by host and tumor factors that traditionally have incorporated TNM staging with additional pathologic, biologic, and clinical host factors. Beyond the anatomic-based TNM, increasingly new prognostic and predictive factors are being discovered that have important survival and treatment implications. However, because the TNM is based on a "bin" model, additional prognostic factors would rapidly overwhelm the current system. This communication demonstrates the clinical implications and improved patient prognosis derived from a new algorithmic model based on clustering analysis. A new algorithm is described that integrates additional factors into the TNM and calculates survival. The results indicate that additional factors can be integrated into the TNM staging system providing additional patient stratification without changing the TNM definitions. Adding prognostic factors to traditional TNM staging increases substratification of given stages and identifies and separates favorable and unfavorable clinical outcomes for specific TNM stages. Integration of additional prognostic factors into the TNM by a clustering algorithm can change the stratification of patient outcome. This may guide the clinician to select a more rational management program based on the additional factors and improve cohort selection for clinical trials.