We agree that a larger study is warranted and state that in our summary (as we said in the title it is a pilot study, but we do not think that the study was fundamentally flawed and address your points:

Healthy volunteers were needed to produce the reliable change calculation, as stated clearly in the text. This gives an index of the variation to be expected with these tests, therefore allowing statistical analysis. If there were no healthy volunteers it would not be possible, with certainty, to calculate the expected performance on each cognitive task, taking into account practice effects (i.e. it is easier to do a task if you have seen it before). The real strength of this trial is that the patients were tested before, during and after treatments, and as such act as their own control.

These patients were not an overly anxious group; seven of the 32 had anxiety scores above the threshold on the GHQ12 at baseline and this decreased to five and two at the second and third test points, respectively. Also, those who scored above the threshold on the GHQ12 were no more likely to perform worse on the cognitive tasks than the other patients.

A significant difference in age or in a measure of intelligence is much more likely to show a difference in performance on cognitive tasks. We state in the text that in this pilot study, age and IQ were not perfectly matched between groups, which was one possible confounding factor. We acknowledge that the sample size is small and the differences found are subtle.

Contrary to the expectations of many healthcare professionals, a diagnosis of cancer itself does not impair cognition. We have conducted > 700 assessments of patients, many newly diagnosed with cancer, and found that their cognitive performance is not significantly impaired.

At the Sussex Cancer Centre, patients with localised prostate cancer scheduled for radical radiotherapy receive LHRH therapy as standard before radiotherapy, both to reduce the planning volume and improve outcome. This is not the practice at all centres, but finding a control group without radiotherapy would be difficult, as the patients may be a different group and the timings of treatment would also differ. Radiotherapy is not known to impair cognitive performance, unless it is directed at the brain. Nonetheless, we think several groups of patients need to be assessed, particularly those receiving different antiandrogen therapies.