Expanded experience with an intradermal skin test to predict for the presence or absence of carboplatin hypersensitivity.

PURPOSE

Carboplatin-associated hypersensitivity is increasingly recognized as a potentially serious toxicity when this agent is administered for more than six total cycles.

PATIENTS AND METHODS

Our group has used a predictive skin test in women with gynecologic cancers who have previously received more than six cumulative cycles of platinum-based chemotherapy. Thirty minutes before all subsequent carboplatin courses, a 0.02-mL aliquot from the solution prepared for treatment is injected intradermally. A positive test is considered to be a > or = 5-mm wheal, with a surrounding flare.

RESULTS

From October 1998 through March 2003, 126 patients received a total of 717 carboplatin skin tests (median per patient, four tests; range, one to 54 tests). Of the 668 negative tests (93% of the total performed), 10 were associated with evidence of carboplatin hypersensitivity (1.5% false-negative rate; 95% CI, 0.6% to 2.4%), none of which were severe (eg, dyspnea, hypotension, cardiac/respiratory compromise). Of the 41 positive tests, the decision was made to not deliver the drug to 32 patients, although seven women ultimately underwent a future attempt at re-treatment with a platinum agent using a desensitization program. In seven episodes where patients received the carboplatin despite the finding of a positive test, six were associated with the development of symptoms of anaphylaxis (none severe).

CONCLUSION

A negative carboplatin skin test seems to predict with reasonable reliability for the absence of a severe hypersensitivity reaction with the subsequent drug infusion. The implications of a positive test remain less certain, but limited experience with continued treatment suggests this approach must be undertaken with considerable caution.