An overview of features in JMP® Clinical to analyze adverse events

[#1]

The analysis of safety data, including investigations into the strengths and limitations of various data sources, has received a lot of attention in recent years. Much of this discussion focused specifically on the analysis of adverse events (AEs), defined in the International Council for Harmonisation Guideline for Good Clinical Practice (GCP, Revision 2) as “any untoward medical occurrence in a patient or clinical investigation subject administered a pharmaceutical product and which does not necessarily have a causal relationship with this treatment.” However, this definition implies that it is technically not possible for a patient on placebo or vehicle to experience an AE, so the definition from draft GCP Revision 3 may be preferred: “any unfavourable medical occurrence in a trial participant.”

In a clinical trial, investigators collect verbatim reports of patient experiences or reactions that are then coded by automated or manual means using a coding dictionary to a preferred set of AE terms for analysis and reporting purposes. For the development of new medical products, the Medical Dictionary for Regulatory Activities (MedDRA) is the dictionary of choice, and its use is required for submissions to the European Medicines Agency (EMA), the Pharmaceuticals and Medical Devices Agency (PMDA) of Japan, and to the Center for Drug Evaluation and Research (CDER) and Center for Biologics Evaluation and Research (CBER) within the United States Food and Drug Administration (US FDA).

Once coding is complete and the clinical trial is unmasked (I worked for years in ophthalmology where the terms masked/unmasked are preferred to blinded/unblinded), the analysis and summary of these coded AE terms provide the bulk of our understanding on the safety of the new treatment and its varying regimens and doses.

These summaries comprise a large portion of the clinical study report (CSR) used to describe the trial findings. Further, should the new treatment be efficacious enough to justify submission to one or more regulators, integrated analyses of safety across all clinical studies, with additional safety analyses produced within important subgroups, are produced as well. Upon approval, these analyses of AEs are reproduced within treatment labels so that patients and physicians understand the risk of using the new therapy.

Given how important AEs are to patient safety, it is no wonder why JMP Clinical contains so many reports dedicated to their analysis and summary (Figure 1).

Figure 1. Reports dedicated to adverse events

Given so many options for analyzing AEs, the user is often faced with the question “What report should I use?” That’s a great question! Not only does JMP Clinical analyze individual AEs, but it can also analyze various groupings of events as well, using Standardized MedDRA Queries (SMQs) and FDA Medical Queries (FMQs). So, without further ado, let’s see what JMP Clinical can do (Table 1).

Report	Description
Adverse Events Distribution	Provides a graphical and tabular summary of the frequency and percentage of patients with AEs by treatment or demographic groups as well as across all groups. Use this report when you want to describe patients experiencing certain events. These analyses tend to be most appropriate when the majority of patients complete the study treatment and study. Differential rates of treatment discontinuation or study discontinuation between treatment or demographic groups may be misleading in understanding the results. This report provides no information on whether an event occurs multiple times within a patient.
Adverse Events Multiple Occurrences Distribution	Provides a graphical and tabular summary of the frequency of AEs by treatment or demographic groups, as well as across all groups. Use this report when you want to understand the total frequency of each kind of event, especially when patients may experience multiple events of the same kind. When compared to Adverse Events Distribution, this report provides insight into the likelihood of events occurring multiple times within a patient.
Adverse Events Incidence Rates	Provides a graphical and tabular summary of the frequency of AEs by treatment or demographic groups, as well as across all groups normalized by the duration of patient exposure to therapy or the study duration. Use this report when you want to understand the total frequency of each kind of event adjusted for length of follow-up to obtain incidence rates, which summarize the number of events per unit of patient years (e.g., “ the number of events per 100 patient years”). This analysis includes all events that occur and adjusts for differential follow-up or dosing. This analysis is preferred when the treatment exposure or study duration varies significantly for patients within the study, or when analyzing multiple studies together with varying treatment or study durations.
Adverse Events Narrative	Generates written reports of selected events that can be used for clinical study reports to describe patient safety and recovery. Most often, use it to summarize serious adverse events (SAEs) or AEs that lead to treatment or study discontinuation.
Adverse Events Time to Event	Generates a Kaplan-Meier time-to-event analysis to the first instance of a particular AE. Patients who do not experience a particular AE are censored at the end of their follow-up. This analysis is useful when discussing patients experiencing certain events, as well as when the time at which the first event occurs from the start of treatment needs to be considered. It is useful for understanding if event occurrence may be delayed between groups. However, this analysis does not consider any events experienced beyond the first.
Adverse Events Risk Report	A more statistical graphical and tabular complement to Adverse Events Distribution, where the analysis focuses on comparing patients experiencing AEs. Pairwise treatment comparisons are performed using either the risk difference (difference in percentages or proportions), the relative risk, or the odds ratio with 95% confidence intervals.
Medical Query Risk Report	A more statistical graphical and tabular complement to Medical Query Distribution, where the analysis focuses on comparing patients experiencing FMQs or SMQs. Pairwise treatment comparisons are performed using either the risk difference (difference in percentages or proportions), the relative risk, or the odds ratio with 95% confidence intervals.
Algorithmic FDA Medical Query Risk Report	Similar to the Medical Query Risk Report but specific to the seven sex-specific FMQs (five for females, two for males) and the four FMQs (rhabdomyolysis, hypoglycemia, hyperglycemia, hypersensitivity) that include algorithms that use information from laboratory, concomitant medication, and medical history domains (LB, CM, and MH, respectively) while assessing the temporal relationships of these data.
Custom Medical Query Risk Report	Similar to the Adverse Events Risk Report and Medical Query Risk Report, in which the analysis focuses on comparing patients experiencing user-defined custom medical queries summarized in ADAE using CQxxNAM variables.
Treatment Emergent Adverse Event Summary	Provides a summary graphic and table of treatment emergent adverse events (TEAEs) commonly presented in CSRs. Frequency and percentages of patients experiencing any TEAE, treatment emergent SAEs, severe TEAEs, drug-related TEAE, TEAE leading to death, or TEAE leading to study discontinuation are included.
Medical Query Distribution (New in JMP Clinical 19)	Similar to Adverse Events Distribution, it summarizes the frequency and percentage of patients experiencing SMQs or FMQs at the narrow or broad scope. This analysis is useful to obtain a summary of medical concepts.
Recurrence Report (New in JMP Clinical 19)	This analysis considers multiple occurrences of an AE within patients and considers the explicit timing of when events occur to summarize the mean cumulative function (MCF), which is the average number of cumulative events that occur up to a particular point in time. This report is similar to Adverse Events Time to Event in that it considers the time at which events occur. Unlike Adverse Events Time to Event, this report considers all AEs that a patient may experience. It is the most comprehensive approach to considering all events and their timing, while accounting for differential follow-up.

Table 1. Summary of adverse event reports

And there are still more features in JMP Clinical that make AEs a major focus of the analysis!

Report	Description
Co-occurrence (New in JMP Clinical 19)	This report summarizes all pairs of occurrences that happen simultaneously within a patient overlapping in time; it also obtains estimates of this overlap as the number of days the paired occurrences happen together, with the ability to compare occurrence pair frequencies and duration by treatment. This report considers occurrences such as adverse events (AE), concomitant medications (CM), or deaths. Compared to Nearby Occurrences, this report is a more brute-force analysis across all occurrences,
DSUR/PSUR Report	The development/periodic safety update report (DSUR/PSUR) provides summaries of patient disposition, demographics, and exposure, as well as tables summarizing the frequency and percentage of patients experiencing AEs and SAEs. This report can be used to satisfy annual regulatory reporting requirements for investigational new drugs.
Nearby Occurrences (New in JMP Clinical 19)	This report allows users to define reference occurrences of interest and find nearby occurrences, either: According to time windows around start or stop dates. If there is any overlap in time between reference and nearby occurrences. This report considers occurrences such as adverse events (AE), concomitant medications (CM), or deaths. It is a more nuanced approach to understanding the temporal relationships of occurrences of interest compared to Co-occurrence.
Patient Profiles	Produces a patient-level view of all data that presents AEs alongside medications and study findings. Useful for a single snapshot of the entirety of a patient’s data.

All of this insight (and so much more) is available within JMP Clinical.