Rates of Consent Withdrawal in Cancer Trials Low but Not ‘Trivial’

Rates of withdrawal of consent from cancer clinical trials were less than 10%, according to a multi-site observational cohort study.

Among nearly 12,000 patients from 58 clinical trials, 9% withdrew from their respective trials within 2 years, reported Aminah Jatoi, MD, of the Mayo Foundation for Medical Education and Research at the Mayo Clinic in Rochester, Minnesota, and colleagues.

From 2013 to 2019, 2-year rates of consent withdrawal were 5.7%, 7.6%, 8.5%, 7.8%, 8.4%, 9.5%, and 9.8%, they noted in JAMA Oncology.

Five percent of patients withdrew within 6 months, while 0.9% withdrew consent even before starting the trial intervention.

“The 9% of patients who withdrew consent from their respective clinical trials at 2 years is by no means trivial,” Jatoi and colleagues wrote. “As cancer clinical trials are designed and initiated, all of these factors should be taken into consideration with a planned expansion of sample size to offset anticipated withdrawal of consent.”

Multivariable analyses showed that several factors were associated with a higher likelihood of withdrawal of consent within 2 years from cancer trials, including:

• Hispanic ethnicity: OR 1.67, 95% CI 1.30-2.15, P<0.001
• Randomized design with placebo: OR 1.64, 95% CI 1.38-1.94, P<0.001
• Patient age 75 years and older: OR 1.39, 95% CI 1.12-1.72, P=0.003

On the other hand, use of radiation in cancer trials was associated with patient retention (OR 0.68, 95% CI 0.54-0.86, P=0.001).

The fact that two underserved groups — persons of Hispanic ethnicity and older patients — appeared to have a higher rate of withdrawal of consent was particularly concerning, Jatoi and colleagues said. They noted that while there have been increasing efforts to recruit patients from diverse backgrounds to clinical trials, “fewer such efforts have focused on retention once patients have been enrolled in a trial.”

“The current study suggests that successful recruitment of patients to a cancer clinical trial is only the beginning and that retention efforts are also required,” they added.

In an editorial accompanying the study, Narjust Florez, MD, of the Dana-Farber Cancer Institute in Boston, and colleagues noted that despite the fact that most of the patients in these trials were white and under the age of 65 — which limits generalizability — “it is an important study highlighting the issue of consent withdrawal within clinical trials and calls for research and interventions to mitigate these challenges at the start of clinical trial design.”

This study included therapeutic trials conducted within the Alliance for Clinical Trials in Oncology, a U.S. government-funded group that develops, undertakes, analyzes, and reports adult cancer trials as part of the National Cancer Institute’s National Clinical Trials Network.

The patients in these trials represented a wide variety of cancer types, treatment, and demographics from 2013 through 2019. The included trials intended to treat active cancer or cancer in an adjuvant/postoperative setting, and included either systemic therapy or radiation. Symptom control trials were also included.

Randomized and single-arm phase II trials, randomized phase III trials, randomized phase II/III trials (those that included both designs in one protocol), and placebo-controlled trials were included. Phase I studies were excluded.

Patient withdrawal was defined as a patient’s voluntary termination of consent to participate anytime during trial conduct.

The pooled study population included 11,993 patients. Median age was 62, 67% were women, 61% were younger than 65, and 81% were non-Hispanic white.

“Findings of this cohort study are robust, generalizable, and worthy of consideration as cancer clinical trials continue to be thoughtfully designed and executed with the goal of generating meaningful findings for the population at large,” the researchers concluded. “Factors associated with withdrawal of consent should be considered when designing trials and should be further studied to learn how they can be favorably modified.”