The administration of paclitaxel without prophylaxis for the prevention of hypersensitivity reactions: is this a rationale and safe therapeutic strategy?

Abstract

Paclitaxel is currently one of the most widely employed antineoplastic agents in the oncologist's armamentarium. A unique toxicity of this agent is the development of hypersensitivity reactions, which occur in as many as 6 ± 10% of patient's receiving the drug (Weiss et al. 1990). During the initial clinical development of paclitaxel, concern for the incidence and severity of hypersensitivity reactions led to the decision that all patients receiving the agent would be treated with a three drug regimen designed to prevent these reactions (Weiss et al. 1990; McGuire et al. 1989). This program included the use of dexamethasone, taken the night before and morning of chemotherapy, along with histamine-1 and histamine-2 receptor antagonists delivered intravenously approximately 30 min prior to paclitaxel administration. All patients given paclitaxel were also required to receive the drug as a 24-h infusion, but a subsequently reported randomized trial (Eisenhauer et al. 1994), and the results of clinical experience, have revealed that shorter infusion schedules can be employed with an acceptable incidence of hypersensitivity reactions. This statement assumes patients continue to receive the three drug prophylactic program, regardless of the paclitaxel infusion schedule utilized. The need to employ prophylaxis for paclitaxel-associated hypersensitivity reactions if very long treatment infusion schedules (e. g., $96 h) are to be administered has recently been questioned (Wilson et al. 1994; Georgiadis et al. 1997; Younes et al. 1996; Seidman et al. 1996). The major rationale for this suggestion is data gathered from small phase 1 ±2 trials of longer paclitaxel infusion schedules ($96 h) where the incidence of hypersensitivity reactions appeared to be lower than that observed with shorter infusion regimens (#24 h) (Weiss et al. 1990; Spriggs and Tondini 1992). Is this a reasonable strategy based on knowledge of drug-induced hypersensitivity reactions, as well actual clinical data? Can the approach be considered to be safe? Several points can be made in addressing this important clinical issue. First, as the incidence of serious paclitaxel-induced hypersensitivity reactions is actually quite low (53% of treated patients) (Eisenhauer et al. 1994; McGuire et al. 1996), failure to observe one or more episodes in a phase 1 or 2 trial involving 15± 30 patients cannot be interpreted to indicate such reactions do not occur, or even to strongly suggest that the incidence is lower than that observed with shorter infusion schedules. Second, even from a theoretical perspective, it is difficult to understand why extending the duration of the paclitaxel infusion from 24 to 96 hours should significantly reduce the incidence and severity of hypersensitivity reactions. It is well-established that these reactions usually develop within the first few minutes of the initiation of a paclitaxel infusion, often with the delivery of only a few milligrams of the agent into the systemic circulation (Weiss et al. 1990; McGuire et al. 1989; Essayan et al. 1996; Peereboom et al. 1993). Therefore, why would a minimal slowing of the rate of entry of the drug into the circulation eliminate the risk of a serious reaction, such that prophylaxis for these reactions need not be administered? The signs, symptoms, and course of paclitaxel-induced hypersensitivity reactions are aclassico for those caused by release to histamine and other vasoactive substances (type 1) (Weiss et al. 1990). In cases of severe hypersensitivity reactions (e. g., penicillin allergy, iodinated radiographic contrast material, reaction to bee stings) a desensitization strategy can successfully reduce the incidence and severity of subsequent reactions. However, in these situaThe Journal Cancer Research and Clinical Oncology occasionally publishes Editorials and Guest editorials on current and controversial problems in experimental and clinical oncology. These papers reflect the personal opinions of the authors. Readers should send any comments directly to the authors