Find out how we can help you succeed!
Learn More
What to Send: PM related information that would assist PMs with Leadership, Strategy and Technology. The information can be a short description with the details at the included link. Do not provide advertising related materials.
Where to Send: Submit items of interest to Editor@pminj.org .Review: The information will be reviewed for relevant PM content for the PM community prior to posting.
PMINJ is not responsible for the content or quality of any posted materials.
As the field of clinical trial management evolves, the role of AI becomes increasingly significant. Currently, semi-automated cloud-based systems manage patient screening, recruitment and retention, drug dispensation, toxicity monitoring, safety, efficacy, and data capture, which is still governed by manual entry. However, with the rapid advances in Artificial Intelligence, Machine Learning, and Data Mining, this article focuses on the key areas of clinical trial management safety and efficacy, followed by data capture and how AI disrupts the standards.
Many academic institutions and companies are focused on bringing new innovative solutions to the current challenges in clinical project management. By harnessing the power of AI, they are developing solutions like the HINT (hierarchical interaction network) model. It is a revolutionary deep learning model that predicts efficiency and success rates of clinical trials. For example, this model can predict the success of a trial based on the patient eligibility criteria and the type of drug. Based on the probability score, a low success score means potential failure of the trial and alerts trial managers to either redesign the trial design, use a different combination of the drug, or re-evaluate the development of the molecule.
AutoTrial, a large language model, plays a crucial role in matching patients to clinical studies. It achieves this by providing the user with trial descriptions and eligibility criteria. The accuracy of AutoTrial's predictions depends on SPOT (sequential predictive modeling of clinical trial outcome), which is used to sort for newer trials in its database, ensuring that patients are matched to the most suitable trials.
To help users better understand disease and eligibility criteria, systems like Trial Pathfinder and TrialGPT use a large language model to analyze previously completed trials and assign trial-level scores to determine patient needs and trial design.
The latest advance in the predictive model of in silco generative AI-assisted system, considers a large patient lab and clinical outcomes data set to create a digital twin, is used to predict and treat patients. The digital twin is a biosimulation that merges genetic and pharmacogenomic to support the patient's journey. It also follows patients using wearable devices that help address their needs in real-time alerts. Thus, using a digital twin helps reduce the need for control-arm patients up to 50%.
Thus, having usable data from large datasets that can bring meaningful insights into the patient journey requires broad consent. Research transparency is a third option for patients to share data for secondary use to protect human subjects’ data and subject matter experts who can address the ethical challenges of mining patient data to create data sets that help in finding new biological pathways or reproducible data that can augment drug development.
As patients progress through a conundrum of predictions and adjustments, it is a welcome relief to know that AI prediction models for drug accountability to monitor adverse drug reactions are moving away from drug toxicity (drug property, target property, and pharmacokinetics properties of absorption, distribution, metabolism, excretion, and toxicity) toward adopting the drug's efficacy in the patient as a criterion for dosage adjustments. Curate AI is one such system. Drug dosing now is a more personalized treatment away from one dosage that fits all.
Thus, the contracting world of clinical trials is reaching new heights. A prompt defines a patient journey under the watchful eyes of a clinical trial project manager tasked with learning new ways of running a trial by building capacities for negotiation, brainstorming, and collaboration with data scientists to apply project management tools like scheduling and data abstraction in real-time - a necessary skill that is on the rise for clinical trial project managers.
Interested in more information on this topic? References below:
