Effect of baseline QTc interval prolongation on oncology clinical trial enrollment.

Abstract

2612 Background: An increasing number of clinical trials in oncology require normal or near-normal intervals of electrocardiographic corrected QT (QTc) as an eligibility criterion. However the normal QTc reference range (conventionally defined as ≤450ms) has been developed in broad populations of patients, and it is unclear whether average QTc intervals are higher in adult cancer patients. Methods: All electrocardiograms (ECGs) performed for any indication at the Moffitt Cancer Center during a one-week period were reviewed. The QTc interval for each patient was calculated according to the Bazett and Fridericia formulas. Mean and median QTc intervals were calculated along with standard deviations. The percentage of QTc intervals above 450ms, 470ms, and 500ms were calculated and graded accordingly to the Common Terminology Criteria for Adverse Events v4.0 (CTCAE v4.0). The effects of gender on the QTc interval were also evaluated. Results: 257 patients underwent ECGs over a one-week period. A total of 90 patients (35%) had abnormally prolonged QTc. 50 patients (20%) had QTc prolongation >450 (≥ grade 1), 32 (12%) had QTc prolongation > 470 (≥grade 2) and 8 patients (3%) had QTc prolongation >500ms (≥grade 3) using the Bazett formula. The corresponding numbers according to the Fridericia formula were 23 (9%), 13 (5%) and 2 (1%) . The median QTc interval was 441 (Bazett) and 422 (Fridericia) with a standard deviation of 30 and 28 respectively. Women had a mean QTc of 438 (Bazett) versus 443 for men. Conclusions: A very large percentage of patients seen at a comprehensive cancer center have QTc intervals above the normal reference range on routine ECGs. The growth in number of trials requiring a normal QTc interval as an eligibility criteria suggests that an ever-larger number of patients are excluded from trial participation. Regulatory agencies may need to reconsider eligibility criteria that substantially restrict trial enrollment and that accrue patients who are not reflective of the general patient population.