We thank Drs. Talpaz and Kantarjian for their comments and questions about our recent article on low-dose interferon- therapy for chronic myeloid leukemia [1]. The unpublished results they cite are better than their previously published data [2, 3] and those of other groups that used similar interferon- dosing schedules [1]. This includes the most recently published reports of randomized trials by the Italian Cooperative Study Group on Chronic Myeloid Leukemia [4] and by the German Chronic Myeloid Leukemia Group [5]. We are uncertain whether the data cited in their letter are from their original report or from an unexplained subset analysis. We also wonder why a discrepancy exists between their current and previous data and data of other groups that studied large numbers of patients and used similar dosing schedules. The answer to these questions awaits publication of their full-length article.

Talpaz and Kantarjian state that sufficient data suggest that "higher doses are more effective in inducing cytogenetic responses ... . " It is unclear, however, to which data they are referring, given that our study is the first to report the results of long-term, low-dose interferon- therapy in minimally pretreated patients with chronic myeloid leukemia. Although our sample comprised only 27 patients, our results were nearly identical to those reported by Talpaz and colleagues [2, 3] and the several other groups who have treated similar patient populations with higher interferon- doses. The important exception was the significantly decreased cost and toxicity found at the lower dose [1, 4, 5]. Previous studies suggesting that lower interferon- doses are inferior in treating chronic myeloid leukemia involved significantly pretreated patients, included patients treated for as little as 1 month, or continued low-dose therapy for only 4 to 8 weeks before patients were crossed over to the higher dose [1]. It has now been clearly substantiated that these factors, rather than the interferon- dose, probably accounted for their poorer response rates. The mechanism of action of interferon- in patients with chronic myeloid leukemia remains uncertain, and no evidence exists to suggest that "more is better." In fact, doses of interferon- as high as 12 x 10⁶ U/m² daily were initially used to treat hairy cell leukemia. However, the dose of 2 x 10⁶ U/m² three times weekly has subsequently been shown to be as effective as higher doses and is now the standard dose used to treat that disease. With an agent as costly and toxic as interferon-, the issue of dose is far from trivial.

Finally, we agree with Drs. Talpaz and Kantarjian that the next step is not to compare different doses in a randomized trial but rather to combine this most interesting agent in new and innovative ways with other therapies. We hope that investigative groups with large patient samples will produce scientifically thoughtful randomized trials that advance our knowledge.