A Prospective Study of Cognitive Complaints in Patients With Testicular Cancer

IntroductionCognitive complaints are frequent among patients with cancer. Oncologists and patients have become concerned that systemic chemotherapy may have cognitive side effects. After this concern became public there was a risk of expectation bias in studies of cognitive complaints. We prospectively explored cognitive complaints in patients with testicular cancer treated with chemotherapy or radiotherapy during the 1990s, before today's increased awareness of this possible side effect.Patients and MethodsThe European Organisation for Research and Treatment of Cancer (EORTC) quality of life (QoL) questionnaire (QLQ-C30) and a testicular cancer module (TC module) were completed before treatment (baseline), at 3 months, and at 12 months by 276 chemotherapy and 71 radiotherapy patients enrolled in 3 EORTC studies. Cognitive complaints were based on the cognitive function (CF) items (concentration and memory) of the QLQ-C30. Other QLQ-C30 functions and symptoms represented adjustment variables.ResultsThe chemotherapy group (CHEMO) showed a significant increase in prevalence of cognitive complaints from baseline to 3 months. At 12 months these rates were back at baseline levels. The radiotherapy group (RAD) showed no significant change in the prevalence of cognitive complaints over time. Significant differences between the 2 treatment groups were observed only at the 3-month follow-up. In multivariate analyses, treatment modality did not show significant association with cognitive complaints at any time point, whereas current fatigue showed significant associations at all time points.ConclusionIn patients with testicular cancer with no information or expectation bias, an increased rate of cognitive complaints was observed shortly after chemotherapy, with return to baseline levels at 12 months. Treatment modality (chemotherapy vs. radiotherapy) was not associated with cognitive complaints at any time point after adjustment for relevant QoL variables.