Generalisability of Common Oncology Clinical Trial Eligibility Criteria in the Real World.

Affiliation

Department of Oncology, University of Calgary, Calgary, Alberta, Canada. Electronic address: [Email]

Abstract

OBJECTIVE : Strict oncology clinical trial eligibility criteria can contribute to low accrual and result in poorly generalisable study findings. Using common eligibility criteria, we sought to (i) determine how many patients would be eligible versus ineligible and (ii) describe real-world patterns of treatments and outcomes between those considered trial eligible and ineligible.
METHODS : The Alberta Cancer Registry was used to assemble a population-based cohort of patients diagnosed with 11 common malignancies between 2004 and 2015. We considered age >75 years, anaemia, comorbid conditions (heart disease, uncontrolled diabetes, kidney disease, liver disease) and history of a prior malignancy or immunosuppression to be exclusion criteria. Logistic regression was used to characterise the likelihood of receiving treatment. Cox regression models were constructed to determine cancer-specific and overall survival.
RESULTS : We identified 125 316 cancer patients, of whom 53% were men; the median age was 66 (interquartile range 48-84) years. Approximately 38% of patients were considered trial ineligible. The most common reasons for ineligibility were advanced age (24%) and heart disease (16%). In this ineligible group, 12, 47 and 19% still underwent chemotherapy, surgery and radiotherapy, respectively. Compared with ineligible patients, eligible patients were more likely to undergo chemotherapy (odds ratio 1.98, 95% confidence interval 1.89-2.07, P < 0.0001), surgery (odds ratio 1.39, 95% confidence interval 1.32-1.46, P < 0.0001) and radiotherapy (odds ratio 1.46, 95% confidence interval 1.4-1.52, P < 0.0001). Compared with ineligible patients who did not receive treatment, those considered ineligible but who still received treatment experienced improved cancer-specific survival (hazard ratio 0.75, 95% confidence interval 0.74-0.77, P < 0.0001) and overall survival (hazard ratio 0.89, 95% confidence interval 0.87-0.90, P < 0.0001).
CONCLUSIONS : A significant proportion of real-world patients are unable to participate in clinical trials due to stringent exclusion criteria, but many still receive treatment in routine practice. The eligibility criteria of oncology clinical trials should be broadened.

Keywords

Clinical trials,eligibility criteria,population-based study,real-world data,

OUR Recent Articles