The Food and Drug Administration’s (FDA) guidance, “Cancer Clinical Trial Eligibility Criteria: Performance Status,” is one in a series of guidances addressing eligibility for clinical trials of oncology drugs. Specifically, this guidance includes recommendations regarding expanding eligibility criteria to include participants with a wider range of performance status (PS).
Performance status addresses how well a person is able to perform activities of daily living. Two common scales are Eastern Cooperative Oncology Group (ECOG) and Karnofsky Performance Scale (KPS). Persons with a low PS have typically been excluded from clinical trial participation in oncology since it has been reported that a low PS correlates with lower survival, and those with a low PS may not be well enough to receive investigational treatment or tolerate toxicities. The etiology of low PS may be important to consider. If the low PS is due to malignancy, the treatment may result in an improved PS.
PS determination is inherently subjective with possible variability between raters which invites bias. Studies have shown that older people are typically assigned a higher numeric ECOG PS score than a younger participant with no objective difference in physical activity. PS is less predictive of cancer-related outcomes for older adults and may be less relevant for more recently developed anticancer treatments that have different toxicity profiles than cytotoxic chemotherapy.
When considering inclusion of participants with low PS on clinical trials, sponsors should consider the following potential advantages and disadvantages:
Potential Advantages
- More rapid trial accrual.
- Improved external validity of trial results.
Potential Disadvantages
- Increased adverse events (AE).
- Potential impact on trial outcome data.
Recommendations
Patients with lower PS should be included in clinical trials in a way that contributes to a greater understanding of the efficacy and safety profile of the investigational drug while maintaining participant safety. Patients with ECOG PS2 (or KPS 60-70) should be included unless there is a scientific and/or clinical rationale for exclusion justified by established safety considerations. Given the potential for differences in AE rates, including PS2 patients could provide important safety data to facilitate decision-making for participants in the post-approval setting where the population may be more likely to be included in treatment with the drug intervention.
As clinical trial data is gathered, PS eligibility criteria should be modified as appropriate, with later stage trials generally mirroring the intended use population. If data precludes enrollment of ECOG PS2 or KPS 60-70 participants, the rationale for exclusion should be stated explicitly in the protocol.
Baseline performance status should be collected in all clinical trials. Where there is a large range of baseline PS participants, PS can be used as a stratification factor.
Alternative trial designs can be used to include those with low PS. Pre-specified cohorts meeting this criterion could be exempt from the primary analysis to encourage inclusion of these participants and collect safety data. These cohorts could be exploratory in nature and small. Consultation with the FDA when designing such cohorts may be appropriate.
Participant reported outcome assessment of Physical Function and Role Function are two core clinical outcomes that can provide baseline and longitudinal data that can complement clinician assessed PS. Wearable devices could be used to add objective activity data. Since existing PS scales are suboptimal for most participants with cancer over age 65, sponsors may consider using an available geriatric assessment tool to better characterize the functional status of older adults. This could include the comprehensive geriatric assessment which is more descriptive in evaluating geriatric overall health status than ECOG and better for predicting toxicity than KPS. Other available geriatric assessment tools may better characterize the functional status of older adults. A simple assessment tool evaluating single or multiple aspects of function with limited burden to the participant is preferred.
Please note a significant amount of information is present in the references. They are great resources and should be consulted.