Patient Recruitment In Clinical Trials Research Articles

Gold Coast Criteria in ALS Diagnosis: A Real-World Experience

Background: Revised El Escorial (rEEC) and Awaji criteria are currently used for diagnosing and categorizing amyotrophic lateral sclerosis (ALS). However, they are complex; their sensitivity is still not optimal for research purposes, and they present high inter-rater variability in clinical practice. To address these points, in 2019, a new set of diagnostic criteria was proposed, namely the Gold Coast criteria (GCC), characterized by a dichotomous diagnostic categorization, i.e., ALS or not ALS. Methods: In order to investigate the sensitivity, specificity, and clinical usefulness of GCC in a practical clinical setting, we retrospectively evaluated 131 patients diagnosed with ALS and 104 control subjects. ALSFRS-R score, electrophysiological tests, neuroradiological investigations, and CSF analysis were obtained. rEEC, Awaji, and GCC were applied at the first and last evaluations. Results: The sensitivity of GCC (93.1%; 96.1%) was greater than rEEC (71.8%; 87%) and Awaji criteria (77.8%; 89.3%) both at the first visit and last follow-up. The GCC’s specificity (28.8%) is lower than that of the other two criteria (rEEC 45.2%; Awaji 43.3%). Conclusions: Our study suggests that in a real-world setting, the GCC are more sensitive and have substantially lower risk of false negative diagnoses than rEEC and Awaji criteria. Although rEEC had the highest specificity, they may delay the diagnosis. Systematically using the GCC could help to achieve an earlier diagnosis and quickly refer patients to the correct management. The low specificity of GCC is likely to not significantly impact patient recruitment in clinical trials; therefore, its use might allow a faster and earlier enrollment.

Learning to match patients to clinical trials using large language models

Objective:This study investigates the use of Large Language Models (LLMs) for matching patients to clinical trials (CTs) within an information retrieval pipeline. Our objective is to enhance the process of patient-trial matching by leveraging the semantic processing capabilities of LLMs, thereby improving the effectiveness of patient recruitment for clinical trials. Methods:We employed a multi-stage retrieval pipeline integrating various methodologies, including BM25 and Transformer-based rankers, along with LLM-based methods. Our primary datasets were the TREC Clinical Trials 2021–23 track collections. We compared LLM-based approaches, focusing on methods that leverage LLMs in query formulation, filtering, relevance ranking, and re-ranking of CTs. Results:Our results indicate that LLM-based systems, particularly those involving re-ranking with a fine-tuned LLM, outperform traditional methods in terms of nDCG and Precision measures. The study demonstrates that fine-tuning LLMs enhances their ability to find eligible trials. Moreover, our LLM-based approach is competitive with state-of-the-art systems in the TREC challenges.The study shows the effectiveness of LLMs in CT matching, highlighting their potential in handling complex semantic analysis and improving patient-trial matching. However, the use of LLMs increases the computational cost and reduces efficiency. We provide a detailed analysis of effectiveness-efficiency trade-offs. Conclusion:This research demonstrates the promising role of LLMs in enhancing the patient-to-clinical trial matching process, offering a significant advancement in the automation of patient recruitment. Future work should explore optimising the balance between computational cost and retrieval effectiveness in practical applications.

Identify and mitigate bias in electronic phenotyping: A comprehensive study from computational perspective

Electronic phenotyping is a fundamental task that identifies the special group of patients, which plays an important role in precision medicine in the era of digital health. Phenotyping provides real-world evidence for other related biomedical research and clinical tasks, e.g., disease diagnosis, drug development, and clinical trials, etc. With the development of electronic health records, the performance of electronic phenotyping has been significantly boosted by advanced machine learning techniques. In the healthcare domain, precision and fairness are both essential aspects that should be taken into consideration. However, most related efforts are put into designing phenotyping models with higher accuracy. Few attention is put on the fairness perspective of phenotyping. The neglection of bias in phenotyping leads to subgroups of patients being underrepresented which will further affect the following healthcare activities such as patient recruitment in clinical trials. In this work, we are motivated to bridge this gap through a comprehensive experimental study to identify the bias existing in electronic phenotyping models and evaluate the widely-used debiasing methods’ performance on these models. We choose pneumonia and sepsis as our phenotyping target diseases. We benchmark 9 kinds of electronic phenotyping methods spanning from rule-based to data-driven methods. Meanwhile, we evaluate the performance of the 5 bias mitigation strategies covering pre-processing, in-processing, and post-processing. Through the extensive experiments, we summarize several insightful findings from the bias identified in the phenotyping and key points of the bias mitigation strategies in phenotyping.

Changes in patient recruitment parameters and their derivatives under the influence of external factors – population size and density of residence in a certain area

Background. The lack of targeted patient recruitment for clinical trials reaches 90%, which leads to failures of a trial as a whole and insufficient access to the necessary treatment or diagnostic method for patients. To find out the reasons for recruitment failures, many factors are considered, the action of which is difficult to assess due to high variability. In general, various factors are named that reduce patient recruitment, while factors that improve it are much less known.Objective: to investigate changes in the parameters and indicators of patient recruitment for clinical trials depending on the influence of external factors.Material and methods. A retrospective analysis of four international multicenter clinical trials of phases II–III was performed by 16 patient recruitment parameters and their 6 derivatives (indicators) – both widely used in the literature and newly proposed. A total of 622 patients from 70 clinical centers located in 59 cities of Russia, Ukraine, and Belarus were included in the study. The methods of descriptive statistics and typing were used. To study the influence of factors, internal and external factors were selected, external factors including population size, area and density of residence were analysed, and changes in parameters and indicators depending on the influence of each factor were examined. The area and population density were studied in inseparable connection with each other.Results. A simple classification of factors was proposed – external and internal to the clinical center where patients were recruited. The factors classified as external were analyzed depending on the change in the proposed parameters of patient recruitment for clinical trials and their relationships – indicators (derivatives). The final rate of patient recruitment and the final number of patients recruited in the population group of 1–2 million people had statistically significant (p<0.05) higher values (0.57±0.20 and 15.08±5.06, respectively) than in group with up to 1 million people – 0.14±0.05 and 3.75±1.24.Conclusion. For the first time, an extended panel of parameters and indicators that allow evaluating the influence of various factors on patient recruitment for clinical trials was proposed. The value of the proportion of parameters and indicators that had statistical differences among themselves in the group influenced by the population size factor was more than twice as large as the proportion of similar parameters and indicators in the group influenced by area and population density: 47% and 23%, respectively, which may indicate more strong influence of the first factor.

Automated Clinical Trial Cohort Definition and Evaluation with CQL and CDS-Hooks.

Patient recruitment for clinical trials faces major challenges with current methods being costly and often requiring time-consuming acquisition of medical histories and manual matching of potential subjects. Designing and implementing an Electronic Health Record (EHR) and domain-independent automation architecture using Clinical Decision Support (CDS) standards that allows researchers to effortlessly enter standardized trial criteria to retrieve eligibility statistics and integration into a clinician workflow to automatically trigger evaluation without added clinician workload. Cohort criteria are translated into the Clinical Quality Language (CQL) and integrated into Measures and CDS-Hooks for patient- and population-level evaluation. Successful application of simplified real-world trial criteria to Fast Healthcare Interoperability Resources (FHIR®) test data shows the feasibility of obtaining individual patient eligibility and trial details as well as population eligibility statistics and a list of qualifying patients. Employing CDS standards for automating cohort definition and evaluation shows promise in streamlining patient selection, aligning with increasing legislative demands for standardized healthcare data.

BRD4L cooperates with MYC to block local tumor invasion via suppression of S100A10

Targeted therapy based on BRD4 and MYC shows promise due to their well-researched oncogenic functions in cancer, but their tumor-suppressive roles are less understood. In this study, we employ a systematic approach to delete exons that encode the low-complexity domain (LCD) of BRD4L in cells by using CRISPR–Cas9. In particular, the deletion of exon 14 (BRD4–E14) results in cellular morphological changes towards spindle-shaped and loosely packed. BRD4–E14 deficient cells show increased cell migration and reduced cell adhesion. The expression of S100A10 was significantly increased in cells lacking E14. BRD4L binds with MYC via the E14-encoded region of the LCD to inhibit the expression of S100A10. In cancer tissues, there is a positive correlation between BRD4 and MYC, while both of these proteins are negatively associated with S100A10 expression. Finally, knocking out the BRD4–E14 region or MYC promotes tumor growth in vivo. Together, these data support a tumor-suppressive role of BRD4L and MYC in some contexts. This discovery emphasizes the significance of a discreetly design and precise patient recruitment in clinical trials that testing cancer therapy based BRD4 and MYC.

Optimizing Patient Recruitment in Global Clinical Trials using Nature-Inspired Metaheuristics

A common problem seen in the ineffective execution of global multicenter trials is the frequent inability to recruit a sufficient number of patients. A myriad of barriers to recruiting enough patients timely exist and may include practical limits on the number of recruiting sites imposed by the various countries, as well as administrative, cost and unanticipated issues. The Poisson-gamma recruitment model is a widely accepted statistical tool used to predict and track recruitment at different levels of a trial and make inference. An optimal recruitment plan can be designed using mathematical optimization, however, the search is complicated with multiple nonlinear objectives and constraints that arise from regulatory and budgetary considerations. We review nature-inspired metaheuristic algorithms which are powerful, flexible, general purpose optimization tools and demonstrate their capability and utility in optimizing complex recruitment designs for global clinical trials using the Poisson-gamma model as an exemplary model. This research opens up new avenues for improving the efficiency and effectiveness of patient recruitment in clinical trials, thereby potentially accelerating the development of new medical treatments.

Changing the first screening parameters for patients and their derivatives in Phase II–III clinical studies

Actuality. The search for parameters of clinical trials (CT) associated with the recruitment of patients is primarily associated with the search for parameters that can predict the recruitment — predictors of patient recruitment, since patient recruitment is an important and one of the most complex processes in clinical trials, the success of which depends on the entire study. The authors propose to use the parameters of the recruitment of patients assumed by the center, the number of exclusion criteria according to the protocol, and other simple methods that allow predicting the upcoming recruitment; however, the proposed approaches are unclaimed. The parameter of success in enrolling patients in the previous 10 or more studies proposed by the authors obviously has its drawbacks because clinical centers that did not participate in 10 successful studies are automatically excluded. However, the search for a simple parameter to predict upcoming recruitment continues; for example, the time of the first screening has its own predictor potential. We investigated changes in this parameter and its derivatives during the recruitment of patients to selected international multicenter clinical trials (IMCTs). Materials and methods. A retrospective analysis of 4 IMCTs of II–III phases was performed on the recruitment of patients and changes in the time parameters of the first screening and its derivatives. Aim. Consider changes in the time parameter of the first screening and its derivatives during patient recruitment. Statistical analysis. Descriptive statistics with typing. Results. The change in the time parameter of the first screening during the recruitment of patients for CT was considered. The time of the first screening varies depending on the type of clinical center and the experience of the principal investigator. Discussion. According to the literature, a change in the time parameter of the first screening may indicate the success of the recruitment of patients in the center. Conclusion. The changes in the time of the screening set studied by us will allow us to apply the obtained values in clinical trials.

Optimizing Patient Recruitment for Clinical Trials: A Hybrid Classification Model and Game-Theoretic Approach for Strategic Interaction

Optimizing Patient Recruitment for Clinical Trials: A Hybrid Classification Model and Game-Theoretic Approach for Strategic Interaction

How can current oncological datasets be adjusted to support the automated patient recruitment in clinical trials?

This study aims to identify necessary adjustments required in existing oncological datasets to effectively support automated patient recruitment. We extracted and categorized the inclusion and exclusion criteria from 115 oncological trials registered on ClinicalTrials.gov in 2022. These criteria were then compared with the content of the oBDS (Oncological Base Dataset version 3.0), Germany's legally mandated oncological data standard. The analysis revealed that 42.9% of generalized inclusion and exclusion criteria are typically present as data fields in the oBDS. On average, 54.6% of all criteria per trial were covered. Notably, certain criteria such as comorbidities, pregnancy status, and laboratory values frequently appeared in trial protocols but were absent in the oBDS. The omission of criteria, notably comorbidities, within the oBDS restricts its functionality to support trial recruitment. Addressing this limitation would enhance its overall effectiveness. Furthermore, the implications of these findings extend beyond Germany, suggesting potential relevance and applicability to oncological datasets globally.

The Impact of Artificial Intelligence on Optimizing Diagnosis and Treatment Plans for Rare Genetic Disorders.

Rare genetic disorders (RDs), characterized by their low prevalence and diagnostic complexities, present significant challenges to healthcare systems. This article explores the transformative impact of artificial intelligence (AI) and machine learning (ML) in addressing these challenges. It emphasizes the need for accurate and early diagnosis of RDs, often hindered by genetic and clinical heterogeneity. This article discusses how AI and ML are reshaping healthcare, providing examples of their effectiveness in disease diagnosis, prognosis, image analysis, and drug repurposing. It highlights AI's ability to efficiently analyze extensive datasets and expedite diagnosis, showcasing case studies like Face2Gene. Furthermore, the article explores how AI tailors treatment plans for RDs, leveraging ML and deep learning (DL) to create personalized therapeutic regimens. It emphasizes AI's role in drug discovery, including the identification of potential candidates for rare disease treatments. Challenges and limitations related to AI in healthcare, including ethical, legal, technical, and human aspects, are addressed. This article underscores the importance of data ethics, privacy, and algorithmic fairness, as well as the need for standardized evaluation techniques and transparency in AI research. It highlights second-generation AI systems that prioritize patient-centric care, efficient patient recruitment for clinical trials, and the significance of high-quality data. The integration of AI with telemedicine, the growth of health databases, and the potential for personalized therapeutic recommendations are identified as promising directions for the field. In summary, this article provides a comprehensive exploration of how AI and ML are revolutionizing the diagnosis and treatment of RDs, addressing challenges while considering ethical implications in this rapidly evolving healthcare landscape.

Digital Trial Solutions eScreening: Software to rank patients by their predicted clinical trial eligibility using real-world data and machine learning.

e13587 Background: The recruitment process for clinical trials is time-consuming and labor-intensive. Electronic screening tools aim to facilitate patient recruitment by focusing the recruiters’ efforts on patients most likely to be eligible. These tools are required to produce useful estimates of patient eligibility despite large amounts of missingness in Real World Data (RWD). Digital Trial Solutions (DTS) eScreening—unlike other proposed tools that use only top-level Inclusion/Exclusion (I/E) criteria—queries for all computable criteria and imputes missing data using Machine Learning (ML). Our ML models use the ConcertAI data from DTS clinical sites. Methods: DTS eScreening provides an eligibility score for each patient in a clinical trial cohort based on how many I/E criteria are fulfilled, weighted by criteria relevance as assessed by clinical experts. If information is missing in the patient record, DTS eScreening predicts the probability of passing the criterion using a binary classifier trained for each criterion. We evaluated classifier performance on a held-out test set. We evaluated the utility of DTS eScreening by collecting quantitative and qualitative data from users of DTS eScreening. Results: All evaluations were done on a multi-site clinical trial with 49 digitized criteria. The cohort was 10,156 patients with the appropriate cancer indication. Model performance: twenty-three criteria had complete data and so no ML models were built. To avoid biased model estimates, we avoided modeling ten criteria with greater than 99% missingness or 1000:1 imbalance. Seventeen LightGBM classifiers were built on the full cohort for criteria such as ECOG, stage, labs, and vitals. The model performance showed an average precision of >99% for most models, while the lowest precision was 83%. DTS eScreening utility: using our tool, clinical screeners reviewed 114 patients with the highest eligibility scores of the 10,156-patient cohort. Of the 114 patients, 38 were determined to meet all digitized eligibility criteria. The precision on reviewed patients was 33%, higher than other reports in the literature. Feedback from semi-structured interviews with 3 users indicated the tool was useful for reducing review time. A screener said: “It makes screening easier for me because the system already filters the most eligible patients at the top.”. Conclusions: DTS eScreening tool demonstrates the ability to conduct patient-trial matching using computable I/E criteria while addressing missingness in RWD. Positive initial feedback from users highlights improvements in the efficiency of patient recruitment in clinical trials. We believe our solution is scalable and can be applied beyond oncology. Future improvements include relaxing restrictions for modeling at low counts to include additional I/E criteria and incorporating unstructured data.

Deriving breast cancer chemotherapy patterns from real-world data.

e13586 Background: Real World Data (RWD) collected during routine medical practice can help in clinical trial design, planning and execution. In addition, it can provide a compelling picture of safety of medical products, and account for all potential adverse events that can be encountered in routine clinical practice. However, raw drug information about chemotherapy in cancer care is not readily understandable in the format it is collected. Analytical techniques need to be applied to extract regimen information, which includes drugs, dosage, number of cycles and cycle length. Methods: A retrospective study was performed using data on 7,798 breast cancer patients from the TriNetX Network, a federated network of de-identified, HIPAA-compliant, health data from 21 healthcare organizations across North America as of May 2022. We investigated a method built on rule-based algorithm and clustering analysis to extract regimens and their patterns of administration and align them into lines of treatment (LOT). To derive drug administration patterns, we clustered administration time periods using three features: total number of drug administrations, median number of days between drug administrations, and standard deviation of the days between drug administrations. Results: The patterns of administration that correspond to the two most common regimens for Erb-B2 receptor tyrosine kinase 2 negative group (ERBB2-) stages 1, 2 and 3 are shown. We looked at patients with Hormone Receptor (HR+) positive and Triple Negative (TN) breast cancer. Results of our analysis were in a close agreement with NCCN Guidelines. However, regimen administration patterns varied. This information can be useful to characterize patients based on the adherence to the expected chemotherapy regimens. It can also provide meaningful insight into burden of illness, such that patients who have higher variability of the drug administration might be faced with clinical problems such as drug tolerability or side effects. The variability might also signify access to care challenges. Conclusions: Understanding regimen LOT and administration patterns is central to research based on RWD, but it can also be useful for clinical trial design, execution, as well as site selection and patient recruitment for clinical trials. [Table: see text]

The Canadian Registry for Amyloidosis Research: A national multi-disciplinary registry for real-world evidence (P12-8.014)

The Canadian Registry for Amyloidosis Research: A national multi-disciplinary registry for real-world evidence (P12-8.014)

Patient Recruitment in Clinical Trials: Areas of Challenges and Success, a Practical Aspect at the Private Research Site

Patient Recruitment in Clinical Trials: Areas of Challenges and Success, a Practical Aspect at the Private Research Site

Availability of Structured Data Elements in Electronic Health Records for Supporting Patient Recruitment in Clinical Trials.

Automated identification of eligible patients for clinical trials is an evident secondary use for electronic health records (EHR) data accumulated during routine care. This task requires relevant data elements to be both available in the EHR and in a structured form. This work analyzes these data quality dimensions of EHR data elements corresponding to a selection of frequent eligibility criteria over a total of 436 patient records at 10 university hospitals within the MIRACUM consortium. Data elements from demographics, diagnosis and laboratory findings are typically structured with a completeness of 73 % to 88 % while medication as well as procedures are rather untructured with a completeness of only 44 %. The results can be used to derive suggestions for data quality improvement measures with respect to patient recruitment as well as to serve as a baseline to quantify future developments.

Should informed consent and information related to patient recruitment in clinical trials be available to the reader of scientific articles? A case study in dentistry

ABSTRACT Ethical aspects in research should be transparently reported. This study aimed to investigate whether informed consent and information related to patient recruitment in clinical studies are well reported in the scientific literature. Randomized clinical trials (RCTs) on root coverage procedures published between November 2016 and November 2021 were selected from the PubMed database. Items/questions were used to guide the extraction of data related to patient recruitment, with a focus on the detailed report of informed consent used to clarify the research to the patient. Data were extracted from the published article and the respective research protocol published in a public registry. Information related to potential selective outcome reporting (SOR) was also extracted. In total, 187 documents were initially screened and 74 reports of RCTs were included. No informed consent was published in the article. Only one research protocol provided a link to the informed consent. Deviations from reporting in the research protocol and published article were found, suggesting SOR. Informed consent and information related to patient recruitment in RCTs on root covering procedures are severely underreported. The present findings may stimulate further discussion and debate on the need for making this information publicly available.

Challenges of Systemic Therapy Investigations for Bone Sarcomas.

Bone sarcoma is a rare component of malignant solid tumors that accounts for only ~0.2% of malignancies. Bone sarcomas present various histological types, and genomic mutations differ markedly by the histological types. Although there are vast mutations in various bone sarcomas, most of them are non-actionable, and even potential targetable mutations that are actionable targets in other malignancies have not shown the appropriate responses in clinical trials for bone sarcomas. Investigations of new systemic therapy, including molecular targeted therapies for bone sarcomas, have thus not progressed like those for other solid tumors. Another problem is that high rates of pediatric/adolescent and young adult patients have bone sarcomas such as osteosarcoma, and patient recruitment for clinical trials (especially randomized trials) is challenging. For pediatric patients, evaluations of tolerability and appropriate dose modifications of new drugs are needed, as their findings could provide the threshold for investigating new drugs for bone sarcomas. To solve these problems, improvements in registry systems, real world data, and pediatric extrapolation have been attempted. We review the issues regarding targeted drug investigations for bone sarcomas, focusing on the current clinical evidence and efforts to resolve these issues.

Ixazomib Based Maintenance Therapy for the Patients with Newly Diagnosed Multiple Myeloma in Real-World Practice

BackgroundThe Phase III TOURMALINE-MM4 Trial demonstrated that the oral proteasome inhibitor, Ixazomib maintenance significantly improved progression-free survival (PFS) in patients with newly diagnosed multiple myeloma (NDMM) not undergoing ASCT. However, patient recruitment for clinical trials is highly selective so the extent to which study populations represent real-world patients is unclear. Currently, data validating the use of ixazomib maintenance in clinical practice in Chinese landscape is scarce.Methodswe conducted a retrospective, observational study of NDMM patients meeting International Myeloma Working Group (IMWG) criteria, who received ixazomib-based maintenance therapy. Data was collected from 26 medical centers in southern China. Enrollment was stratified by risk stage, transplant status and maintenance regimens. The characteristics, toxicities and survival outcomes were recorded.ResultsWe included 105 NDMM patients who received ixazomib-based maintenance treatment. Mean age at diagnosis was 59.82 ± 9.80 years, with 62.9% male patients. MM International Staging System (ISS), stage 3 was the most common with 49.04% of patients; 62.86% had an Eastern Cooperative Oncology Group performance score ≥2, and 81.9% with concomitant diseases. By the end of over 70 months follow-up since initial diagnosis, median PFS was 69.45 (95%CI: 31.42-107.48) months, and median OS was not achieved yet. We explored potential influencing factors of progression at the post-maintenance evaluation with a crude model using logistic regression. Difference in PFS was not statistically significant by status at pre-maintenance response, transplantation, maintenance therapy regimens (single/two/three-agent) and dexamethasone in maintenance regimens (p >0.05). Eight patients (7.62%) had grade≥3 treatment-emergent adverse events (TEAEs); 5.7% discontinued treatment because of TEAEs. Common any-grade TEAEs included fatigue (18.1%), diarrhea (16.19%), and peripheral neuropathy (11.43%).ConclusionAlthough there are certain differences in maintenance treatment options for newly diagnosed multiple myeloma patients in real-world practice, the results of our retrospective study support that the ixazomib-based maintenance therapy was effective with acceptable toxicity and single-agent ixazomib maintenance provided enough clinical benefit compared to two or three agents.Keywords: multiple myeloma, ixazomib, maintenance therapy, real-world practice [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Adaptive Learning of Drug Quality and Optimization of Patient Recruitment for Clinical Trials with Dropouts

Problem definition: Clinical trials are crucial to new drug development. This study investigates optimal patient enrollment in clinical trials with interim analyses, which are analyses of treatment responses from patients at intermediate points. Our model considers uncertainties in patient enrollment and drug treatment effectiveness. We consider the benefits of completing a trial early and the cost of accelerating a trial by maximizing the net present value of drug cumulative profit. Academic/practical relevance: Clinical trials frequently account for the largest cost in drug development, and patient enrollment is an important problem in trial management. Our study develops a dynamic program, accurately capturing the dynamics of the problem, to optimize patient enrollment while learning the treatment effectiveness of an investigated drug. Methodology: The model explicitly captures both the physical state (enrolled patients) and belief states about the effectiveness of the investigated drug and a standard treatment drug. Using Bayesian updates and dynamic programming, we establish monotonicity of the value function in state variables and characterize an optimal enrollment policy. We also introduce, for the first time, the use of backward approximate dynamic programming (ADP) for this problem class. We illustrate the findings using a clinical trial program from a leading firm. Our study performs sensitivity analyses of the input parameters on the optimal enrollment policy. Results: The value function is monotonic in cumulative patient enrollment and the average responses of treatment for the investigated drug and standard treatment drug. The optimal enrollment policy is nondecreasing in the average response from patients using the investigated drug and is nonincreasing in cumulative patient enrollment in periods between two successive interim analyses. The forward ADP algorithm (or backward ADP algorithm) exploiting the monotonicity of the value function reduced the run time from 1.5 months using the exact method to a day (or 20 minutes) within 4% of the exact method. Through an application to a leading firm’s clinical trial program, the study demonstrates that the firm can have a sizable gain of drug profit following the optimal policy that our model provides. Managerial implications: We developed a new model for improving the management of clinical trials. Our study provides insights of an optimal policy and insights into the sensitivity of value function to the dropout rate and prior probability distribution. A firm can have a sizable gain in the drug’s profit by managing its trials using the optimal policies and the properties of value function. We illustrated that firms can use the ADP algorithms to develop their patient enrollment strategies.