Adverse event / serious adverse event data entry when death event occurs

If the death event is an efficacy endpoint in the study, should the death event still be reported as AE/SAE? 

If the death (or mortality) is an efficacy endpoint, the death event will ordinarily not be reported as AE/SAE. Only in some special situation, for example, a death due to anaphylactic reaction, car accident,... that are not related to the underlying disease, will the death be reported as AE/SAE. 

FDA's guidance Safety Reporting Requirements for INDs and BA/BE Studies clearly stated this: 

Generally, study endpoints refer to outcomes that sponsors are measuring to evaluate efficacy. For trials designed to evaluate the effect of a drug on disease-related mortality or major morbidity, endpoint information should be collected, tracked, and monitored, usually by a Data Monitoring Committee (DMC), during the course of the study. The protocol would prespecify a monitoring plan for determining whether subjects receiving the drug treatment are at higher risk for the outcome (e.g., all-cause mortality), and such results would be reported according to the protocol. The study endpoints must be reported to FDA by the sponsor according to the protocol, and ordinarily would not be reported as IND safety reports, except when there is evidence suggesting a causal relationship between the drug and the event (21 CFR 312.32(c)(5)). For example, a death ordinarily would not be reported as an individual case in an expedited report from a trial designed to compare all-cause mortality in subjects receiving either drug treatment or a placebo. On the other hand, in the same trial with an all-cause mortality endpoint, if the death occurred as a result of an anaphylactic reaction that coincided with initial exposure to the drug, or as a result of fatal hepatic necrosis, the death must be reported as an individual case in an IND safety report because there would then be evidence suggesting a causal relationship between the drug and the event (21 CFR 312.32(c)(5)).  

In our clinical trial "Dinutuximab and Irinotecan Versus Irinotecan to Treat Subjects With Relapsed or Refractory Small Cell Lung Cancer", the overall survival (time to death) was the primary efficacy endpoint. The death event would not be reported as AE/SAE and not recorded on AE/SAE case report form. It would be odd and inappropriate to record the AEs like "Death due to disease progression', 'death due to small cell lung cancer', 'death due to the underlying disease'. With Death being an efficacy endpoint, the death information should be recorded on separate case report form - Death Details form according to CDASH. 

Should death be reported as an event or an outcome of an AE/SAE?

Usually, when death event occurred, there were the adverse events leading to the death. The adverse event leading to the death will need to be reported as series adverse event with fatal outcome. The death is the outcome of a SAE and is not entered as an adverse event on its own. 

There are situations that the death is instantaneous or within very short period (for example one hour) of onset of symptoms or an unobserved cessation of life that cannot be attributed to a specific AE term. The death in this situation will be reported as Sudden Death or Sudden Death NOS on the AE case report form. 

If there are multiple ongoing AEs at the time of trial participant's death, should all these ongoing AEs be considered as SAEs with fatal outcome? 

This issue was discussed in a previous post "Recording the outcome for AE/SAE when multiple events contribute to Death". It is preferred in my opinion that a single AE/SAE should be identified as the primary cause for the death. This AE/SAE will have the fatal outcome recorded on the AE case report form. Other ongoing AEs at the time of death will have the AE outcome recorded as 'not recovered/not resolved"

How to fill out other fields on AE forms when death event occurs? 

As discussed above, for AE outcome, 'Fatal' should be selected for the AE/SAE that directly or primarily contribute to the study participant's death. 

For other ongoing AEs at the time of death, the AE outcome 'Not recovered or Not Resolved' should be selected.  

'Action Taken with Study Treatment' is another field on AE form and it is difficult to decide which choice should be selected when a death event occurs. 

At the time of death, there are two appropriate choices for 'Action Taken with Study Treatment": Dose Not Changed or Drug Withdrawn. Which choice to select depends on the sequence of the events: the last dose date/time in relevance to the death date/time. Assuming a study treatment with QD dose frequency, if the last dose date is the same as the death date or if the last dose date is one day prior to the death date, it is reasonable to assume that there is no dose withdrawn. Therefore, it is appropriate to select 'Dose Not Changed' for 'Action Taken with Study Treatment' field. On the other hand, if the last dose date is two days or earlier than the death date, it is appropriate to select 'Dose Withdrawn' for 'Action Taken with Study Treatment' field. 

Sentinel Dosing (Sentinel Subject) and Staggering Enrollment in First-in-Human (FIH) Clinical Trials

First-in-human (FIH) study is a type of clinical trial in which a new drug, procedure, or treatment is tested in humans for the first time. FIH studies take place after the new treatment has been tested in laboratory and animal studies and are usually conducted as phase I clinical trials. 

FIH study can be conducted in healthy volunteers (usually the case) or in patients.(in some special situations). Even though the new drug, procedure, or treatment has been thoroughly tested in pre-clinical studies before initiating the FIH study, the conservative approaches may still needed to be taken to ensure the safety of the study participants when designing the FIH study . 

FIH study can also be designed as phase 0 study or exploratory IND study as discussed in a previous post. 

FIH study may be designed as a single ascending dose (SAD) study where the healthy volunteers are enrolled and dosed in cohorts in dose-escalation fashion, i.e., the next dose cohort will only be enrolled after the safety data from the previous cohorts has been reviewed. FIH study may also be designed as dose-escalation study to identify the maximum tolerable dose (MTD) - such as the "3+3 design". 

Even with the SAD or dose escalation study designs, if it is uncertain there are still potential risks to the participants, additional precautions may be taken: sentinel dosing (sentinel subject) and staggering enrollment. 

Sentinel Dosing (Sentinel Subject): 

For the FIH study in healthy volunteers, the subjects are recruited to the clinical research unit (CRU, also called Phase I clinic). A cohort of subjects will be confined in the CRU to be dosed, observed, and evaluated. All subjects in the same dose cohort will be dosed at the same time. The study starts with the lowest dose cohort and then moves to higher dose cohorts. 

While dosing by cohort approach is usually safe, unexpected incidences can still occur. If the unexpected adverse events cause the harm to the study participants, it affects all participants in the entire cohort. Below are two examples where the phase I trial participants died or severely injured after receiving the experiment treatment in FIH Phase I studies. 

• Cytokine Storm in a Phase 1 Trial of the Anti-CD28 Monoclonal Antibody TGN1412
• French study’s death recalls 2006 UK clinical trial disaster - Were mistakes repeated?

To prevent this from happening, a strategy called sentinel dosing is often practiced so that one person in the first cohort of participants is dosed in advance of the full study or in advance of any full cohort. The very first subject who receive the sentinel dose is called 'sentinel subject'. 

Sentinel dosing was mentioned in EMA guidance "Guideline on strategies to identify and mitigate risks for first-in-human and early clinical trials with investigational medicinal products":

It is considered appropriate to design the administration of the first dose in any cohort so that a single subject receives a single dose of the active IMP (often known as sentinel dosing). Flexibility in this approach is allowed but should be on a risk-proportionate basis with a clear scientific rationale for any proposals not to use this strategy.

When the study design includes the use of placebo it would be appropriate to allow for one subject on active and one on placebo to be dosed simultaneously prior to dosing the remaining subjects in the cohort. This approach is expected for all single and multiple dosing cohorts, in order to reduce the risks associated with exposing all subjects in a cohort simultaneously. This sentinel approach may continue or also start to be appropriate at later stages of study design, e.g. on the steep part of the dose response curve, when approaching target saturation levels or the maximum clinical exposure levels defined in the protocol (see sections 7.5 and 8.2.9), in case of non-linear PK, or in light of emerging clinical signs or adverse events that do not meet stopping criteria. There should be an adequate period of time between the administration of treatment to these first subjects in a cohort and the remaining subjects in the cohort to observe for any reactions and adverse events. The duration of the interval of observation will depend on the PK and PD characteristics and the level of uncertainty associated with the product (see section 4). At the end of the observation period, there should be a clearly defined review of all available data for the sentinel subjects before dosing of further subjects in the cohort, with dose stopping rules in place to prevent further dosing if any rule is met (see also section 8.2.10).

Staggering Enrollment

Majority of phase I studies are conducted in healthy volunteers where the same cohort of subjects are recruited and confined at the clinical research unit (a single center) for dosing and post-dose measures and observations. In some situations (such as oncology studies, gene therapy trials, studies using human-plasma derived products), phase I studies are conducted in patients and are not ethical to be conducted in healthy volunteers. The patients will usually be recruited from multiple sites - so called multi-center phase I clinical trials. 

In multi-center phase I clinical trials, before multiple sites can start to recruit patients, a 'staggering enrollment' approach may be employed to minimize the potential harms caused by the innovative therapies. With the 'staggering enrollment' approach, after the first patient is enrolled, the second patient will only be enrolled after the first patient has been followed-up for a period of time and thoroughly evaluated for the safety measures. The third patient or the parallel enrollment will only be started after the second patient has been followed-up and thoroughly evaluated. 

The 'staggering enrollment' approach was used in the first-in-human trials in CAR-T trial and in gene therapy trials.  

The first CAR-T approval was for Novartis’s Kymriah (tisagenlecleucel) for the treatment of Acute lymphocytic leukemia (ALL). The tisagenlecleucel was originally developed by UPENN and FIH study was conducted by the UPENN. In their FIH study for CAR-T, the enrollment was staggering: 

“Staggered enrollment on the CNS3 cohort: infusion of any subsequent patient on the CNS3 cohort will be delayed until 21 days after the prior CNS3 patient’s infusion to allow for toxicity monitoring.”

For Bluebird’s Beti-cel in treatment of β-thalassemia patients requiring regular red blood cell (RBC) transfusions, their FIH trial also employed a staggering enrollment strategy:

“Initially, subjects with β-thalassemia major of the βE/β0 genotype will be enrolled in this study, and treatment will be staggered. The second subject will begin myeloablative conditioning only after the first subject 1) engrafts (defined as an absolute neutrophil count [ANC] ≥0.5 × 109/L for 3 consecutive days); and 2) has no LentiGlobin® BB305 Drug Product treatment-related serious adverse event (SAE) unexpected to occur with autologous HSCT. After Subject 2 meets these same criteria, parallel enrollment will be opened to additional subjects with the βE/β0 genotype.” 

'Staggering enrollment' may also be employed for the logistic reason. For a specific investigational site, the investigator and the study coordinator may not have the resource to enroll multiple patients all at once. They just don't have the manpower to do that. 'Staggering enrollment' approach allows the site to enroll one patient at a time. 

Drug Development for Rare Diseases - Public Workshops

Drug development for rare diseases is challenging, complex, but absolutely necessary. FDA has special programs to manage and encourage the drug development for rare diseases. 

According to "Rare Diseases at FDA" website, all three divisions (CDER, CBER, and CDRH) have special programs to support the drug development in rare diseases areas:

• CDER's Rare Disease Program
• CBER's Rare Disease Program

At "Regulatory Education for Industry (REdl)" meeting this week, there was a session "Partnering Across FDA to Advance Therapies for Rare Diseases" featuring three prosentations by FDA officers:

• Dr Retzky presented "FDA Rare Disease Updates: Incentives and Resources"
• Dr Lee presented "CDER's Perspective on Working Together with Our Rare Disease Partners within CDER and across FDA"
• Ms Vaillancourt presented "CBER's Perspective on Working Together with Our Rare Disease Partners across FDA"
• Dr Sherafat presented "Design Considerations for Clinical Trials in Rare Diseases"

FDA/CDER has an ARC program (Accelerating Rare disease Cures) aiming to drive scientific and regulatory innovation and engagement to accelerate the availability of treatments for patients with rare diseases. In year 1 of the ARC program, two important public workshops had been organized (webcasts can be watched following the links below):

• FDA/NIH Regulatory Fitness in Rare Disease Clinical Trials conference, May 16-17, 2022

CDER’s Rare Diseases Team and National Center for Advancing Translational Sciences
Focus on academic investigators and those looking to learn how to bridge the gap between academic investigation and the regulatory aspects of drug development

• FDA and Duke Margolis Virtual Public Workshop: Translational Science in Drug Development: Surrogate Endpoints, Biomarkers, and More, May 24-25, 2022

Focus on translational science and the development of surrogate endpoints