Introduction
The Clinical Study Safety demo provides advanced visualizations and robust statistical analyses during clinical trials. It helps site managers, clinical statisticians, and trial overseers to spot trends, identify issues, and make informed decisions throughout the clinical trial process. This helps detect issues early and make necessary adjustments to improve the chances of FDA approval.
The dataset is from github.com/phuse-org/phuse-scripts and is centered on xanomeline. Xanomeline is used to treat neurological symptoms such as schizophrenia. The dataset tracks patients over the course of their treatment. It was used originally in a pilot study for developing CDISC ADaM standards, which are guidelines for submitting clinical trial safety data to the FDA. These standards help ensure reproducibility in clinical trials. We use Spotfire for analyzing various aspects of patient data, including demographics, adverse events, and lab results.
Plus, we detail how Spotfire Copilot bolsters Clinical Safety.
Dataset Composition
This demo contains several pages, each designed to capture different aspects of patient data:
Patient Demographics and Disposition: Details on the patients’ backgrounds split by treatment group
Lab Results: Information on lab test readings throughout the trial, changes from baseline, hepatotoxicity warnings
Adverse Events: Records of any negative symptoms and relative risk assessment
Patient Profiles: A consolidated view of individual patient journeys through the study
Demographics
The initial page focuses on the distribution of patients across different treatment groups: high dose, low dose, and placebo. The objective is to ensure an even distribution of demographic characteristics across these groups to maintain the integrity of the clinical trial.
For instance, the distribution of Caucasian patients is similar across different groups, but there can be fluctuations when examining minority groups. An example is the absence of male African participants in the low-dose group, which could sway the trial’s results.
Disposition
Disposition data tracks patient retention and categorizes their reasons for withdrawal from the study over time. This is crucial for maintaining the study's protocol and ensuring that as many patients as possible are retained. The disposition data is visualized to show the decline in participant numbers from preliminary screening to week 26 of the study. The most common reason for withdrawal is adverse events.
Adverse Events (AEs)
Adverse events are incidents occurring during the study that may or may not be related to the drug being tested. They are classified into various categories, such as cardiac disorders or nervous system disorders. The visualization tool enables users to slice and color-code AEs by the drug administered and the severity of the events.
Severe AEs are distinguished from serious AEs, with the latter posing significant risks to the patient's life or mobility. For instance, a severe headache is intense but not life-threatening, whereas a serious AE could be a condition that significantly impacts the patient’s health.
We can also study the relative risk of AEs by comparing the incidence rates between high-dose and placebo groups. The double dot plot on the left shows whether the drug increases or decreases the likelihood of adverse events compared to the placebo. The relative risk plot on the right shows the corresponding confidence intervals or error bars. High error bars indicate a low occurrence of events, implying those data points are less reliable for drawing conclusions.
Lab Results
The lab results section evaluates whether changes in lab test readings are significant compared to patients’ baseline. This is a collective view of all patients’ lab tests and the proportions that are normal, low, and high. HbA1c, measuring average recent blood sugar levels, remains high across patients.
A critical aspect is also identifying potential drug-induced liver injury (DILI). Lab test ALB (Albumin), for instance, is a protein in blood plasma that is used to measure liver and kidney functions. Low levels could indicate a disorder and high levels could indicate dehydration or Diarrhea. ALT (alanine aminotransferase) is an enzyme in the liver and high levels of ALT could indicate a liver disorder.
The scatterplot of Hy’s Law tracks liver function tests, highlighting patients who move into danger zones over the course of the study. This example shows a patient's progression through the study, showing how their liver function tests moved from normal ranges into danger zones and then back towards normal. This analysis helps in identifying trends that could indicate serious side effects of the drug, such as DILI.
The violin plot offers a different view of the distribution of these lab results over time. These plots are complemented by trend lines, making it easier to see overall changes in the population or in specific patient subgroups.
Patient Profile
The patient profile page consolidates data to present a comprehensive view of a patient’s journey through the study. This includes:
Lab Results: Highlighting normal and abnormal readings.
Dosing Events: Recording the administration of the drug.
Concomitant Medications: Other medications the patient is taking.
Adverse Events: Severity and type of AEs experienced.
Users can drill down into individual patient data to identify potential links between drug dosing and adverse events. This detailed view helps understand the specific experiences of each patient to facilitate analysis and intervention. For example, a patient profile might show a consistent pattern of high lab test results or a series of adverse events following drug administration. This information can be crucial for identifying potential risks associated with the drug and making necessary adjustments in the study protocol.
Recommended Comments
There are no comments to display.