Protocol SmartDesign Glossary

Protocol SmartDesign ​leverages AI to recommend optimal primary endpoints and inclusion/exclusion criteria associated with the highest chance of operational success in clinical research, streamlining the protocol planning process, reducing complexities and saving time required for protocol design.​

The AI model analyzes Norstella proprietary historical data, real-world data and sources from the public domain to make protocol recommendations, increasing the likelihood of successful outcomes and reducing the risk of protocol design errors.​

“Successful trials” is defined as hitting a completion date within +/- 60 days of the initial anticipated completion date. These are the trials that feed the model and inform the recommendations generated by the application. The total number of successful trials for your study will appear in the banner of the main page.

The View study filters in Trialtrove link at the top of the page creates Trialtrove filters based on the user-selected criteria and opens them directly in Trialtrove. The filter is automatically limited to industry-sponsored trials, as only these trials are used by the product. This view allows users to see all trials available in this therapeutic space within our database.

The View study filters in Sitetrove link at the top of the page creates Sitetrove filters based on the user-selected criteria and opens them directly in Sitetrove. The filter is automatically limited to industry-sponsored trials, as only these trials are used by the product. This view allows users to see all investigators in this therapeutic space within our database.

The Completed trials number displayed in the banner represents the total number of completed trials that the model reviewed to identify successful trials that meet the product’s business rules. Clicking this number opens a Trialtrove link that applies the study criteria along with the following PSD business rules:

The Successful trials number represents the trials that met the operational success criteria and were used by the model to generate recommendations for primary endpoints and inclusion/exclusion criteria. Clicking this number opens a Trialtrove link that includes the study criteria, applicable business rules, and any model-applied relaxations. If the model relaxed any criteria, the Trialtrove filters will reflect those same relaxations.

This section provides a list of the successful trials that influenced the recommendations for the primary endpoint, inclusion, and exclusion criteria. You can view specifics about these trials in the provided table.

The title used to identify a trial, with preference given to the official or scientific title when available.

To dig into the trial source material, access Sitetrove by clicking the title of each trial to explore additional information.

This is the unique identifier (Sitetrove ID) assigned to the trial by Citeline, ensuring precise tracking and reference.

The trial phase is listed as reported publicly. If not reported, Citeline analysts evaluate the trial and assign a phase based on standard trial phase descriptions, reflecting the stage of the research.

This refers to the organizations that are responsible for conducting the study, including logistical and financial support. It denotes the entity that initiates and oversees the study, exercising authority and control over its execution.

Indicates the number of participants targeted for the trial, as documented in public sources.

Represents the total number of investigators associated with the trial, linking to the research efforts.

Calculated as the time between the First Patient First Visit (FPFV) to Last Patient Last Visit (LPLV) of the study. This value is derived from data available in the public domain and the Citeline database, providing insight into the trial’s overall timeframe.

The prevalence indicates how often the recommendation appeared in successful trials. The categories are defined as follows:

This feature highlights the source data and specific trials that the model utilized to generate each recommendation within the section. The information is organized by summary recommendations for each prevalence category, ensuring clarity in how different data sources influence the recommendations. Each trial ID listed will contain a link directly to the corresponding trial page, the allowing you to access detailed information about the trials that informed the recommendations, Trial Timing containing the Trial Start and end date and the Sponsor of the trial.

The model’s recommendations for primary endpoints are derived from an analysis of historically successful trials that match the selected study criteria. By examining these trials, the model identifies patterns and trends that are associated with higher success rates and fewer amendments. This analysis helps in determining which primary endpoints are most likely to contribute to the success of the study, based on empirical evidence from similar trials. The goal is to provide recommendations that optimize the likelihood of achieving positive outcomes and minimize the need for modifications during the trial.

The model’s recommendations for inclusion criteria are informed by analyzing a dataset of historically successful trials relevant to the selected study parameters. It examines these trials to identify criteria that are associated with higher success rates and minimal need for changes.

The model’s recommendations for exclusion criteria are based on an analysis of historical data from successful trials that align with the selected study parameters. By evaluating these trials, the model identifies exclusion criteria that are linked to improved success rates and reduced need for modifications. This ensures that the recommended criteria are both effective and practical, drawing from proven practices in similar studies.

The historical average number of sites used in similar trials conducted over the past five years is calculated based on the user-selected Therapeutic Area (TA), disease, patient segment, and trial phase. If no exact match is found for the selected criteria, values are derived using data from the selected disease only. Minimum, median, and maximum values are also provided to give users a clear view of the historical range within the selected indication.

The historical average number of participants reflects the mean count of participants in similar trials conducted over the past five years, based on the user-selected Therapeutic Area (TA), disease, patient segment, and trial phase. If no exact match is found for the selected criteria, values are derived using data from the selected disease only. Minimum, median, and maximum values are also provided to give users a clear view of the historical range within the selected indication.

The Completed Trials Enrollment Rate reflects the mean (average) enrollment rate observed in comparable clinical trials conducted globally over the past five years. Minimum, median, and maximum values are also provided to give users a clear view of historical trial performance within the selected indications.

This metric is derived from the user-selected Therapeutic Area (TA), disease, patient segment, and trial phase, and is calculated without applying country-specific constraints, ensuring a broad, international perspective. If the PSD model expands the patient segment or trial phase, this benchmark adjusts accordingly to maintain a fair and consistent comparison with the forecasted results.

It serves as a benchmark for evaluating forecasted enrollment rates against historical performance.

It represents the predicted enrollment rate for the study based on the entered criteria, utilizing data from similar trials to forecast trends. The forecasted rate will dynamically adjust in response to any changes made to the Primary Endpoints (PE) and Inclusion and Exclusion (I&E) recommendations, ensuring the prediction reflects the most accurate and current study parameters.

The model forecasts the completion time of the study based on your selected criteria. This prediction covers the timeline from First Patient First Visit (FPFV) to Last Patient Last Visit (LPLV). The forecasted duration may adjust as modifications are made to the study criteria and recommendations, ensuring the timeline remains accurate and reflective of the most recent changes.

The historical average number of sites used in the selected studies. Minimum, median, and maximum values are also provided to give users a clear view of the range within the selected studies.

The historical average number of participants used in the selected studies. Minimum, median, and maximum values are also provided to give users a clear view of the range within the selected studies.

The Completed Trials Enrollment Rate reflects the mean (average) enrollment rate observed in selected clinical. Minimum, median, and maximum values are also provided to give users a clear view of trials performance.

It represents the predicted enrollment rate for the study based on the entered criteria, utilizing data from similar trials to forecast trends. The forecasted rate will dynamically adjust in response to any changes made to the Primary Endpoints (PE) and Inclusion and Exclusion (I&E) recommendations, ensuring the prediction reflects the most accurate and current study parameters.

The model forecasts the completion time of the study based on your selected criteria. This prediction covers the timeline from First Patient First Visit (FPFV) to Last Patient Last Visit (LPLV). The forecasted duration may adjust as modifications are made to the study criteria and recommendations, ensuring the timeline remains accurate and reflective of the most recent changes.