FDA’s Internal SOP on How To Do An NDA Safety Review Part 3
We are continuing and finishing the review of one of the FDA’s critical internal SOPs: MOPP 6010.3R: Attachment B: Clinical Safety Review of an NDA or BLA (Issued 12/14/2010; Posted 12/14/2010) available at:
We are now going to examine the FDA reviewer’s handling of AEs. As in the previous parts of this review, all sentences in italics represent Bart’s Corner comments and opinions unless otherwise specified.
Non-Fatal Serious AEs
Deaths, as noted earlier, are carefully scrutinized by the FDA. SAEs from all sources are also looked at very carefully. SAEs from all subjects in phase 2 & 3 are required in line listings and for each event a narrative should be provided by the sponsor. Rates for occurrences of SAEs and for subgroups and doses are provided.
The FDA reviewer then must assess which SAEs were likely explained by other factors and which by the drug. Similar non-serious AEs (e.g. syncope that did not meet serious criteria) should also be examined.
Dropouts & Discontinuations
First the reviewer should figure out how the company differentiated “dropouts” from “discontinuations” as companies do this differently. “Dropouts without adequate follow-up by the applicant may be caused by unreported AEs. Reviewers should consider those dropouts or discontinuations… to be related to possible AEs.” If there is no adequate explanation, the FDA reviewer should get back to the company for more information.
Red flags requiring heightened scrutiny by the reviewer will include:
- Dropouts classified as administrative, lost to follow-up, or a similar term
- Dropouts for which the applicant changed the investigator’s reason for the drop out.
- The reviewer should consider the possibility that what seemed to be another illness or the underlying illness was, in fact, an ADR.
- A substantial number of dropouts due to lost to follow-up and sites with disproportionately high dropout rates should be a sign of concern.
Analysis of dropouts due to AEs is critical:
- It identifies the type and frequency of AEs that patients were unable to tolerate in a clinical trial setting, where there is arguably more support than in a clinical practice setting.
- These adverse dropouts may provide a clue to unexpected, but important, ADRs that can easily be dismissed as intercurrent illness.
- The dose response and time dependency of the drop-outs and drug-demographic, drug-disease, and drug-drug interactions should be examined.
The FDA reviewer should also examine the sponsor’s tables to be sure that the dropouts and discontinuations were correctly categorized.
It is clear that the FDA reviewer is instructed to take a critical and skeptical eye to the deaths, SAEs, dropouts and discontinuations reported by the sponsor. The reviewer will look at the source document CRFs and even reconstruct the tables if necessary. Sponsors should be well aware of this in preparing their ISS and other safety data. It will not be accepted at face value.
These AEs (some serious some not) are defined in ICH E3 and include lab abnormalities that do not meet serious criteria but which may be medically important, events leading to dropout or additional concomitant therapy, other abnormalities judged important (e.g. a single seizure, syncopal episode). These events should be discussed by the sponsor. The level of suspicion for them should be lower than usual: “the causal relation of a drug to uncommon serious adverse events may be supported by less serious events that are more common.”
This is a concept found elsewhere in US and EU regulatory documents with various other names including “safety concerns”, “possible safety issues” and are now integral parts of risk management and minimization plans in the EU and US. They should be sought and noted by the sponsor.
Drug Induced Liver Disease (DILI)
Particular attention should also be paid to DILI. Cases meeting Hy’s Law of liver problems likely due to the drug should be sought (subjects with any elevated aminotransferase (AT) of >3x upper limit of normal (ULN), alkaline phosphotase (ALP) <2xULN, and associated with an increase in bilirubin ≥2xULN).
The FDA clearly states: “Finding one Hy’s Law case in the clinical trial is worrisome; finding two is considered highly predictive that the drug has the potential to cause severe DILI when given to a larger population.
So the sponsor should look for every case that meets the FDA’s “worrisome” criteria and make sure they are worked up and discussed:
An excess of AT elevations to >3xULN compared to a control group
- Marked elevations of AT to 5x-, 10x-, or 20xULN in modest numbers of subjects in the test drug group and not seen (or seen much less frequently) in the control group
- One or more cases of newly elevated total serum bilirubin to >2xULN in a setting of pure hepatocellular injury (no evidence of biliary obstruction, Gilbert syndrome, another drug cause, infection or some other explanation accompanied by an overall increased incidence of AT elevations >3xULN in the test drug group compared to placebo
Supportive Safety Results
The reviewer will also look at supportive or secondary safety issues (e.g. labs, ECGs etc.) unless these need to be reviewed in detail in conjunction with a major safety concern as noted above. Common SAEs/AEs will also be reviewed. The FDA recommends that the methodology for finding AEs in the clinical trials be customized to the study using open-ended questions or checklists.
Again the FDA stresses the lack of usefulness of sponsor/investigator causality:
“When assessing the critical question of whether an adverse event is caused by a drug, a greater interest to reviewers is whether the drug is capable of causing that adverse event in the population rather than whether the drug caused the event in each subject who reported the event…Where events are common and occur in multiple subjects in controlled trials, it is usually not necessary or helpful to consider each case individually. Rather, all reported cases can be considered potentially drug-related, and causality is assessed by comparing the rates of reports in subjects treated with the test drug and in control groups.”
This again, is an important concept for the sponsor to understand when preparing the NDA/BLA.
The FDA would like to see tables with rates of common AEs from phase 2 and phase 3 trials. The FDA also notes that the AE terms should not be divided in a way to mask the overall problem (“splitters vs lumpers”) giving the example that azotemia, BUN elevated and acute renal insufficiency, though somewhat different should be considered together in this exercise. The FDA gives detailed suggestions on how to handle pooling of data and table preparation. the sponsor should refer to this when preparing the NDA/BLA.
Identifying Common and Drug-Related AEs
“The most persuasive evidence for causality is a consistent difference from control across clinical trials, and evidence of dose response.”
Less Common AEs
“A large database is needed to evaluate less common AEs.” The overall database is usually heterogeneous and not useful for identifying these uncommon AEs. Rather the review should develop a condensed list of reactions by estimated incidence and include them in the Warnings & Precautions section of the labeling.
The FDA reviewer should provide an overview summary rather than doing a detailed comment about each lab in each trial. Particular attention should be paid to lab tests done at unscheduled visits. [FDA’s italics]. There are detailed instructions in this SOP on how to review, handle and present the lab data and the reader is referred to the SOP for the details. Particular attention should be paid to shift changes from normal to abnormal and to all outliers. “Decisions about what criteria to use to identify outliers should have been made at the pre-NDA meeting, if possible.” ECGs should be analyzed with particular attention to the ICH guidance E14 on QT/QTc prolongation.
If special studies were done these should be discussed.
Other Safety Explorations
Several should be done as a rule including dose-dependency for AEs, flexible dose titration, cumulative dose dependency, time of onset analyses, duration of AEs, drug-demographic, drug-drug and drug-disease interactions. The human carcinogenicity, pediatric, reproduction and pregnancy data should be examined. All overdoses should be reviewed and the drug abuse potential, withdrawal and rebound effects (if any) reviewed.
The FDA then has an extensive section of sample tables. These should be followed carefully; it is always painful when an NDA/BLA is delayed for format reasons rather than content and scientific reasons.
So in summary, this SOP should be fully reviewed and understood by the sponsor and the safety team in preparing an NDA/BLA. It points out that there are safety issues with drugs and that they should be discussed realistically with the sponsor understanding that they will appear in the approved labeling. The ISS should not attempt to “make them go away” abut rather should address them scientifically and honestly.
This SOP has useful tips and pointers as well as clear instructions on obligatory and nice to have sections, analyses, tables etc. This document should be the bible (or one of the bibles) used in preparing an FDA submission for a new product.