Do Clinical Trials Underreport Cancer Drug Toxicities?

Clinical trials may be underreporting patient toxicities from cancer drugs – by up to 75 percent in some cases – according to a new study that examined the discrepancies between physician-reported toxicities and reports from patients themselves. These study results, published recently in the Journal of Clinical Oncology, support greater incorporation of patient-reported outcomes into toxicity reporting in clinical trials.

“Information available to oncologists and their patients about symptomatic toxicities of anticancer treatments is not based on direct report by prior patients, but instead on reports made by clinician assessment in clinical trials,” the study authors, led by Massimo Di Maio, MD, explained. “The accurate description of occurrence and severity of toxicity of anticancer agents is crucial for an informed evaluation of their risk-benefit ratio.”

With the current study, Dr. Di Maio and colleagues examined the agreement (and disagreement) between physician- and patient-reported toxicities from three separate randomized trials of anticancer treatment:

  • the ELDA (Elderly Breast Cancer—Docetaxel Adjuvant) study of patients aged 65–79 years with early-stage breast cancer
  • the GECO (Gemcitabine–Coxib) trial of patients aged <70 years with non–small-cell lung cancer (NSCLC)
  • the TORCH (Tarceva or Chemotherapy) trial of patients with advanced NSCLC

Researchers assessed agreement between physicians and 1,090 patients for six symptomatic toxicities: anorexia, nausea, vomiting, constipation, diarrhea, and hair loss.

The toxicity was collected prospectively by investigators; then, at the end of each treatment cycle, patients were asked to complete the European Organisation for Research and Treatment of Cancer questionnaire, which asked patients whether they had experienced any of these toxicities during the previous week.

In all cases, percentages of toxicities reported by patients were higher than those reported by physicians. For the six toxicities, Cohen’s k (a measure of agreement) ranged between 0.15 and 0.45, “which can be interpreted as poor to moderate agreement,” the authors explained.

Underreporting was calculated as the rate of cases in which physicians reported no (grade 0) toxicities for a patient across all treatment cycles, but patients did report at least one toxicity in more than one cycle. Among all three trials, physician-reported rates of six symptomatic toxicities were always lower than those reported by the patients themselves. In some instances, the under-reporting by physicians was nearly 75 percent.

“For all six symptomatic toxicities, there was a low agreement between patient and physician reporting. Disagreement was almost completely in the direction of under-reporting by the doctors,” Francesco Perrone, MD, PhD, director of the Clinical Trials Unit at the National Cancer Institute of Naples and corresponding author in the study told ASH Clinical News.

The rates of underreporting were slightly lower – but still high – when researchers analyzed only patients who reported “very much” toxicity in any cycle, while physicians reported none: 50.0 percent for anorexia, 25.8 percent for nausea, 13.0 percent for vomiting, 44.2 percent for constipation, 24.1 percent for diarrhea, and 42.7 percent for hair loss.

Several factors may explain underreporting, the authors wrote:

  • On the physician side, there may be less attention paid to subjective toxicity – particularly for mild adverse effects that would not prompt treatment modification
  • Physicians could have failed to report an adverse event because they judged it unrelated to treatment
  • Clinicians could be less likely to report a toxicity that is largely expected with the drugs administered – but the opposite could also be true

“Effective communication between patients with cancer and their physicians about benefits and risks associated with treatments is a critical component of care, ensuring that patients’ preferences are taken into account in decision making,” Dr. Di Maio and colleague wrote.

The findings suggest that there may be an underestimation of subjective side effects in clinical trial reports that may bias information being given to patients. To have a clearer picture of a drug’s toxicity profile, researchers need to incorporate patients’ subjective toxicity reports into assessments.

“Our paper, along with other studies, shows that a full comprehension of side effects requires two points of view: the professional one (the doctor) and the subjective one (the patient),” Dr. Perrone added. “Each of them alone is not enough, but together they can be truly informative on the toxicity profile of new drugs.”


Reference
Di Maio M, Gallo C, Leighl NB, et al. Symptomatic toxicities experienced during anticancer treatment: agreement between patient and physician reporting in three randomized trials. J Clin Oncol. 2015 January 26. [Epub ahead of print]

SHARE