Response:Intrathecal methotrexate and central nervous system events

Response

We thank Brugières et al for their comments and the additional information coming from their study in children with anaplastic large cell lymphoma.¹  The authors raise important questions, which we would like to answer as follows:

Twenty of 58 patients with central nervous system (CNS) events (34.5%) in our study had achieved complete response or complete response, unconfirmed, and subsequently relapsed in the CNS; only 12 patients (20.6%) experienced isolated CNS relapses. These occurred in 9 of 47 (19.1%) patients who had not and in 3 of 11 (27.3%) patients who had received intrathecal methotrexate (MTX; P = .681). Thus, the percentage of patients with isolated CNS relapse was really small (0.99% of all study patients), with no significant differences between patients receiving or not receiving intrathecal MTX. All other patients either had progressed during or shortly after first-line therapy or had died from therapy or unknown causes.

In patients with isolated CNS relapse, no difference was observed between those treated with chemotherapy and rituximab (R) and those treated with chemotherapy alone. Seven of 36 (19.4%) patients treated with CHOP-14 and 5 of 22 (22.7%) patients treated with R-CHOP-14 had an isolated CNS relapse. One of the 7 patients treated with CHOP-14 and 2 of 5 treated with R-CHOP-14 had received intrathecal MTX.

With regard to the frequency and severity of toxicities occurring in patients who received or did not receive intrathecal MTX during the first 2 courses of therapy we performed a multivariate analysis (logistic regression) adjusting for International Prognostic Index factors except age and sex (for hematotoxicity only). We observed significantly more leukopenia of grades 3 or 4 with an odds ratio (OR) of 1.6 (95% confidence interval [CI] 1.7-3.2, P = .012). Thrombocytopenia of grades 3 or 4 was not significantly increased in patients receiving intrathecal MTX (OR 2.2, CI 0.8-5.9, P = .138). Mucositis of grades 3 or 4 occurred in significantly more patients if intrathecal MTX had been given (OR 2.8, CI 1.8-8.1, P = .001). No differences were seen for other toxicities including severe or life-threatening infections. Taken together, these additional analyses show that very few patients with aggressive lymphoma (< 1%), treated with modern therapy experience an isolated CNS relapse. Although the low number of such patients in our trial precluded a statistical analysis of whether at least these patients may have benefited from intrathecal MTX we agree with Brugières et al that intrathecal MTX was of limited value in the population at large and should no longer be administered, particularly if the side effects are taken into account. Alternative strategies like early systemic MTX (which should be able to prevent not only CSF but also parenchymal CNS involvement) and intrathecal liposomal cytarabine (Depocyte) are under evaluation in adults. It remains to be determined if these strategies administered to a “high-risk” population will be more successful than intrathecal MTX or whether patients with leptomeningeal and/or parenchymal disease detected with more sensitive methods (FACS analysis) should be treated on protocols for primary CNS lymphoma.