Purpose: To determine the incidence of acute toxicity and treatment interruption associated with electron and photon craniospinal irradiation (CSI) in children treated with or without chemotherapy.
Materials and methods: A retrospective study involving a computerized search of the radiotherapy database at the University of Texas M. D. Anderson Cancer Center identified a total of 79 eligible patients </=18 years old who had received electron (n = 46) or photon (n = 33) CSI from October 1980 to March 2000. Acute toxicity was graded according to the 1998 National Cancer Institute Common Toxicity Criteria. Chemotherapy sequencing was categorized as before or after CSI or no chemotherapy. The incidences of weight loss and skin toxicity were recorded and differences in treatment interruption and hematologic values with respect to modality used (electron vs. photon), age (</=6 or >6 years), and sequencing of chemotherapy were compared using chi-square tests.
Results: The median age of the electron group was lower than that of the photon group (6.7 years and 11.7 years, respectively). The two groups were otherwise well matched in terms of median spinal dose (31.1 vs. 33.3 Gy), fraction size (1.57 vs. 1.63 Gy), and total treatment time (32.4 vs. 30.7 days). Only 2 patients in each group (photon and electron) had a treatment break (>3 days). The mean number of days interrupted was 0.94 (photon) and 1.1 (electron) (p = 0.72). The electron and photon groups were well balanced in terms of receiving pre-CSI chemotherapy (37% vs. 41%, p = 0.776). Chemotherapy given before radiotherapy vs. after or not at all was associated with an increased incidence of Grade 3-4 leukopenia (76% vs. 49%, p = 0.02), thrombocytopenia (90% vs. 10%, p = 0), and neutropenia (50% vs. 15%, p = 0.005). A younger age was associated with Grade 3-4 thrombocytopenia (29% vs. 8.7%, p = 0.034), and decreased hemoglobin (29% vs. 6.5%, p = 0.014). The incidence of leukocyte depression of Grade 3-4 toxicity was 62% in the electron group and 32% in the photon group (p = 0.018). The incidence of Grade 3-4 platelet toxicity was higher with electrons (21%) than with photons (4%), but the difference was of borderline significance (p = 0.053). The difference in Grade 1-2 weight loss was not statistically significant (86% electron vs. 55% photon groups, p = 0.53), and Grade 3-4 weight loss did not occur within the entire study group. Most patients experienced Grade 0-1 radiation dermatitis with either electrons or photons.
Conclusion: A younger patient age (</=6 years) and the sequence of chemotherapy before radiotherapy were associated with increased acute hematologic toxicity. Grade 3-4 weight loss, a surrogate for gastrointestinal toxicity, was not observed within the entire population. Despite the increased acute hematologic toxicities associated with CSI in very young children, the vast majority of patients were able to complete photon or electron CSI with minimal to no treatment interruptions. CSI with the electron technique can be used as an alternative to photon CSI in young children, and in our experience does not require the routine use of hematopoietic growth factors.