The Brandes-Friesen case reports: how should we interpret the news?

In this issue (see pages 561 to 566 two case reports by Lorne J. Brandes and Linda A. Friesen illustrate a temporal relation between the discontinuation or initiation of therapy with certain non-antineoplastic agents (lithium, carbamazepine and an H1-antihistamine) and the clinical course of cancer. Only an accumulation of similar case reports and the results of epidemiologic and controlled clinical studies will show whether the temporal relation observed in these cases is also a causal one. In the absence of strong, let alone conclusive, evidence that antidepressants or antihistamines promote the growth of tumours in human beings, a recommendation to stop using these medications would be premature. The news about these case reports and about animal studies on the tumour-promoting potential of some antidepressants and antihistamines is likely to spread through the media. The realistic and responsible option for physicians and clinical investigators is to help patients and prospective participants in clinical trials to interpret this information in a balanced and reasonable fashion.