Viana et al. [I] described their experience with etoposide (VP-16) as initial therapy in the management of patients with Langerhans cell histiocytosis (LCH). However, caution should be used in advocating the use of an epipodophyllotoxin as the "drug of choice" for the treatment of this disorder. Although these authors state that "the side effects (of etoposide) are minimal," there is an emerging body of evidence that both etoposide and teniposide (VM-26) are leukemogenic. Secondary cases of acute nonlymphocytic leukemia have been reported following the use of these agents in the treatment of many different types of neoplasms, including acute lymphocytic leukemia [2], non-small cell lung cancer [ 3 ] , germ cell tumors [4,5] and other neoplasms [summarized in 51. Co-existing risk factors, including the dose or schedule of epipodophyllotoxin, age of the patient, and the concurrent use of other chemotherapeutic agents such as cisplatin and cyclophosphamide have not yet been determined, but several patients have developed secondary leukemia after receiving less than the total dose of 1,350-2,700 mg/m2 of VP-16 used by Viana et al. [I]. Until risk factors for the development of epipodophyllotoxin-related secondary leukemia have been established, it would seem prudent to restrict the use of etoposide to situations where the benefits are clearly outweighed by the potential risks. Etoposide may be appropriate therapy for patients with recurrent or refractory LCH, but the risks of steroids and vinblastine are not sufficient in patients with low-risk LCH to warrant their replacement as initial therapy with an agent associated with the development of secondary leukemia. Much more study is needed before an epipodophyllotoxin can be recommended as initial therapy for what is frequently an indolent and non-life-threatening disorder, and for which other effective therapy is readily available.