Expansion Cohorts In Clinical Trials Of Oncology Drugs: FDA Issues Guidance

Expansion Cohorts In Clinical Trials Of Oncology Drugs: FDA Issues Guidance

FDA released the final guideline paper in March for first-in-human trial models that include both dosage escalation and various expansion cohorts to assess safety, anti-tumor efficacy, pharmacokinetics (PK), and other features of therapeutic cancer candidates all within a single protocol.

The principle is simple: after the optimal Phase 2 dosage has been established during the dose-escalation phase of the trial, expansion cohorts may be created to study other aspects of the cancer drug’s activity. Further dose or dose schedule investigation (e.g., once-daily vs. twice-daily dosing); comparison of safety, tolerability, and anti-tumor effect of monotherapy vs. combination therapy with another oncology drug; or characterization of the safety, tolerability, and anti-tumor effect in different indications, different lines of treatment, or different patient populations may be among the objectives for each cohort (including those with another tumor mutational status). Food impact, drug-drug interaction, and PK in individuals with organ failure may be evaluated using expansion cohorts.

The Guidance’s Main Points


The Most Important Consideration Is Safety.

The protocol should provide a scientific justification for including each group, demonstrating that the potential benefit to the patient exceeds the risks. Sponsors should have strategies to continuously monitor safety data in real time and quickly communicate developing information to investigators, institutional review boards, and the FDA during the trial to keep them aware of new adverse findings.

Safety monitoring and reporting strategies must be implemented by sponsors. The sponsor should include a communication procedure for when major safety problems develop, as well as a plan to notify clinical investigators and regulatory authorities so that protocol revisions may be implemented to address these issues. Cumulative safety data from a multiple expansion cohort trial should be provided more often than once a year.

An independent monitoring committee should be created for each cohort to conduct pre-specified and ad hoc safety and futility evaluations. The committee should also suggest protocol changes (such as adding new safety information or changing the eligibility criteria), updates to the informed consent document to address safety issues (including re-consenting patients if necessary), or adjusting the dosing schedule or safety monitoring plan as needed. For studies with a small sample size, a change from the draft to the final advice allows the committee to be composed of internal sponsor employees.

Additionally, because of the complexity of these trials and the higher safety concerns, sponsors must form an independent safety assessment committee (ISAC) and an independent data monitoring committee (IDMC). These two committees’ tasks “should include, but not be limited to, study of incoming accelerated safety reports, preparation of cumulative summaries of all adverse events, and providing protocol change suggestions to the IND sponsor.” These committees should also be entrusted with reviewing any major adverse events in real time and performing predetermined and ad hoc safety and futility evaluations for each cohort.

Finally, the advice states that all patients should get proper informed consent, and that informed consent papers should be updated on a regular basis based on new research data or protocol changes.

Content of Study Protocol

Because of a study’s complexity and dangers, the advice suggests that sponsors provide more information than is generally necessary for a clinical research protocol. A more thorough procedure will enable the FDA and others (investigators, IRBs) to assess whether the potential benefits exceed the risks to patients and that all potential concerns are handled. The guidelines also detail what should be included in an initial protocol and subsequent protocol modifications.


Considerations for Drug Products and Patients

Because of the higher risk of multiple expansion cohort studies, the FDA advises that they should only be used when the potential benefits outweigh the risk, and that “the patient group should be confined to individuals with severe conditions for whom there is no alternative curative treatment available.”

Drugs and biologics with “high inter- and intra-patient variability in pharmacokinetics suggestive of polymorphic enzyme mediated drug clearance for small molecules” are not appropriate for research in a multiple expansion cohort trial. “The investigational medication has the potential to achieve the criteria for breakthrough treatment designation to warrant continuation of the accelerated development program,” the FDA says of this study design.

FDA Collaboration

Given the complexities of this approach, sponsors should consider meeting with FDA before filing an IND to discuss trial design and the entire development plan for the cancer medicinal candidate. For example, suppose the sponsor plans to use the trial’s results as main support for a marketing application. In that case, they should prepare ahead to ensure they understand and fulfill the FDA’s expectations regarding acceptable data standards, independent evaluation of tumor-based endpoints, and so on.

This technique has been used to develop and subsequent expedite approval of many cancer treatments based on a single first-in-human Phase 1/2 clinical study since the FDA initially issued the draft guideline in 2018. 

Factors To Examine From A Statistical Standpoint

The sample size should be justified based on cohort goals in the statistical analysis strategy. It should also define analyses and stop rules in advance (for safety or futility).

The protocol should be written so that it will allow the planned analysis to proceed. For example, suppose the goal is to compare the anti-tumor impact of one cohort to another (e.g., a combination treatment cohort vs. a monotherapy cohort). In that case, patients should be randomized to those cohorts, and the sample size for each cohort should be large enough to detect any differences.

The stipulation that designs other than Simon’s two-stage may be utilized to test anti-tumor efficacy in a non-randomized cohort has been changed from the draft to the final advice. Adopting Bayesian or other alternative statistical designs may help decrease the number of patients exposed to ineffective drugs. 

The Risks and Benefits Of Expansion Cohort Trial Design

Risks involve difficulty to appropriately interpret safety signals when they develop because of the complexity of enrolling patients into numerous distinct cohorts simultaneously, each with its own goals and inclusion/exclusion criteria, dosages, dose schedules, or treatment regimens. Patients may be put in greater danger as a result of this. Similarly, estimating anti-tumor activity based on a limited number of patients in a cohort may overestimate efficacy, thereby exposing patients to an ineffective dosage. In addition, trial logistics may be complex, especially if specific cohorts need randomization while others do not.

Sponsors should avoid using this strategy when the investigational medicinal product has a restricted therapeutic window or substantial inter-patient PK variability to reduce the risk of increasing safety issues. The indications under investigation should also be confined to major oncologic disorders with an unmet demand for safe and effective treatments.

Expansion cohorts in first-in-human clinical trials may increase the efficiency of oncology medication development, and this method has previously been used to assist in the approval of several cancer medicines. Although these studies are complex and have many moving components, this strategy provides a viable and faster alternative to conventional Phase 1 and 2 trials with proper preparation and early FDA participation.

Benefits include the sponsor saving time and money by moving immediately from the dosage escalation phase to numerous expansion cohorts (generally conducted as a stand-alone clinical study). In addition, the capacity to evaluate various patient groups and hypotheses simultaneously may help speed up the approval process.

This expansion cohort technique also allows researchers to collect early data on patient groups who are often overlooked or ignored throughout the clinical development of cancer drugs, such as older individuals, those with concurrent malignancies or brain metastases, and pediatric patients.

Given the recently enacted Research to Accelerate Cures and Equity (RACE) for Children Act, which allows the FDA to mandate pediatric development of cancer therapies with molecular targets relevant to pediatric cancers, this strategy might be a viable alternative for generating evidence to support that requirement (though the final guidance document notes that, depending on the age of the pediatric population, a different formulation may be required for the cohort).


[1] https://www.cellandgene.com/doc/key-takeaways-from-fda-s-new-guidance-on-first-in-human-multiple-expansion-cohort-trials-0001

[2] https://www.clinicalleader.com/doc/fda-issues-guidance-on-expansion-cohorts-in-clinical-trials-of-oncology-drugs-0001

[3] https://www.fda.gov/media/115172/download



Navigating Genomic Therapy Development

Research on genomic medicine is growing rapidly, which raises challenges concerning the design of clinical trials, regulatory channels, manufacturing processes, and price structures. Sandy Macrae,

Read More