Has the Time Come for Routine Use of Amifostine in Clinical Radiotherapy Practice?

Abstract

In the study by Kruse et al. (1) published in this issue of Acta Oncologica , the heart is added to a growing and long list of preclinical normal tissues, in which damage induced by ionizing irradiation is ameliorated by amifostine. Such radiobiological observations nourish a long-standing dream concerning the possibility of improving outcome of clinical radiotherapy by use of radioprotectors (2). Ever since the Cold War this effort has waxed and waned, but has recently been boosted by the commercialization of amifostine, FDA approval and authoritative guidelines (3). But why has this research been so long underway and what are the prospects for an imminent introduction of amifostine into routine clinical radiotherapy. The primary hindrance is the subject matter itself, since there is no fail-safe modification of traditional radiotherapy (4). Numerous normal tissues are dose limiting and an evaluation of the overall therapeutic benefit of using a radioprotector is therefore very complex. Essentially, two scenarios are at hand: If the radioprotector can assure absolute normal tissue selectivity, fewer complications with unchanged rate of tumor control can be obtained for a given level of radiation dose. However, if selectivity is uncertain then an increase in radiation dose may be needed to maintain the same rate of tumor control when using the protector. However, this strategy requires that the protective effect on tumor tissues is predictable and exceeded by the protection offered to all relevant normal tissues. But even the perfect preclinical radioprotector must have an acceptable toxicity profile and must be easy to handle if generalized clinical use is to be expected with routine fractionated radiotherapy (5). Based on the available preclinical knowledge, including the present study by Kruse et al. (1), the variability of the protection observed in different normal tissue models is unfortunately heterogeneous. Some normal tissues such as the central nervous system, which is often the dose-limiting organ in radiotherapy, are not protected at all, since amifostine does not cross the blood /brain barrier. In other normal tissues such as the salivary glands and the hematopoietic system, amifostine affords significant radioprotection. To make things even more complicated, tumor protection has been impossible to rule out by preclinical experiments. Large single doses of irradiation have often been used and relevant comparisons between tumor effects have been absent or difficult to translate into a clinically meaningful context (5). On the clinical side, decades have passed with publication of phase I /II studies with limited numbers of patients and a few underpowered randomized studies. The problems involved in the interpretation of the results from such trials are exemplified by a recent correspondence (6). Despite the long list of preclinical normal tissues with proven effect of amifostine, disappointingly few have been confirmed in the clinical setting. Amelioration of acute radiation toxicity has been observed in studies that have often employed various types of treatment intensification such as concomitant chemotherapy or accelerated radiotherapy. However, definite confirmation regarding late morbidity such as fibrosis has not been obtained. Hope rose with the recent publication of the study by Brizel et al. (7) showing that amifostine significantly protected against radiation-induced xerostomia, with no apparent loss of tumor control in the head and neck. Unfortunately, the study was not powered as an equivalence trial and was therefore incapable of ruling out the possibility of amifostine-induced tumor protection. In addition, postoperative radiotherapy was used in two-thirds of the included patients, diluting potential protection in the onethird of patients treated with definitive radiotherapy. Finally, the study required that at least 75% of both INVITED EDITORIAL