[vc_row][vc_column][vc_column_text css=”.vc_custom_1555864690658{margin-bottom: 0px !important;}”]The traditional method for evaluating clinical trial data in phase 2 and phase 3 studies requires sponsors to collect enough data to support an Independent Data Monitoring Committee (IDMC) review. Over half of these results are negative, causing sponsors to waste valuable time and millions in lengthy clinical trials.
The FDA has already issued guidance on adaptive trial design to begin to mitigate this risk. However, with the evolution of machine learning, the next step is here – machine learning on real-time patient data.
This concept, called “Dynamic Data Monitoring in Clinical Trials,” will dramatically improve the effectiveness of phase 2 and phase 3 studies by:
- Monitoring drug safety and signal detection in real-time
- Timely terminating “hopeless” trials
- Performing a formal futility analysis, or other adaptive procedures such as population enrichment, or sample size modification
- Intelligently estimating an optimal sample size for a trial and thus maximizing the probability of success of the trial
- Enabling a seamless, optimal phase 2/3 combination trial by identifying most potential doses for phase 3
- Intelligently identifying the subpopulation in which the drug is most effective
- Checking and verifying the assumptions set prior to initiation of the trial
- Optimizing on-going trials to maximize success
The CIMS Dynamic Data Monitoring engine utilizes machine learning and artificial intelligence (AI) technology in an integrated, closed system with eClinical and Interactive Web Response System (IWRS).
With the CIMS DDM platform, you can dynamically monitor and optimize your clinical trials — overcoming the drawbacks of classical adaptive group sequential designs to dynamically monitor trials without using an Independent Statistical Group (ISG) and/or Independent Data Monitoring Committee (IDMC).
As new data is cumulated, the system will automatically compute the score function for chosen endpoints, confidence intervals and conditional power; update stopping boundaries; and perform simulations to predict the trend of a clinical trial.
When Dynamic Data Monitoring becomes mainstream, it will reshape how clinical trials are designed and executed – improving patient safety and cutting sponsor costs buy up to 50%.
Contact us to see it live.
[/vc_column_text][/vc_column][/vc_row]
Leave a Reply