Peter[a]
Neuro-oncol 2002 Apr;4(2):109-114
A phase II trial of thymidine and carboplatin for recurrent malignant glioma:
A North American Brain Tumor Consortium Study.
Robins HI, Chang SM, Prados MD, Yung WK, Hess K, Schiff D, Greenberg H, Fink K, Nicolas K, Kuhn JG, Cloughesy T, Junck L, Mehta M.
Department of Medicine, University of Wisconsin, Madison, WI 53792 (H.I.R., M.M.); University of California at San Francisco, San Francisco, CA 94143 (S.M.C., M.D.P., K.N.); University of Texas MD Anderson Cancer Center, Houston, TX 77030 (W.K.A.Y., K.H.); University of Pittsburgh, Pittsburgh, PA 15213 (D.S.); Department of Neurology, University of Michigan, MI 48109 (L.J., H.G.); University of Texas Southwestern at Dallas, Dallas, TX 75390 (K.F.); University of Texas at San Antonio, San Antonio, TX 78284 (J.G.K.); University of California at Los Angeles, Los Angeles, CA 90095 (T.C.).
This phase II study in recurrent high-grade glioma evaluated the response rate, toxicities, and time to treatment failure of high-dose carboplatin modulated by a 24-h infusion of thymidine (75 g/m ( 2 ) ). The trial was based on preclinical data and a prior phase I study ( J. Clin. Oncol. 17, 2922-2931, 1999); a phase II recurrent high-grade glioma study was initiated in July of 1998. Thymidine was given over 24 h; carboplatin was given over 20 min at hour 20 of the thymidine infusion. The starting dose of carboplatin had a value of 7 for the area under the curve (AUC), with allowance for dose escalation of 1 AUC unit per cycle if grade 2 toxicity was observed. Treatment cycles were repeated every 4 weeks. Accrual as of September 1999 was 45 patients [4 were unevaluable]: 76% with glioblastoma multiforme (GBM), 20% with anaplastic oligodendroglioma, 2% with mixed type, and 2% with anaplastic astrocytoma. Most patients had prior chemotherapy (78%). As observed in the earlier phase I study (in which carboplatin pharmacokinetics were unaltered by thymidine or antiseizure medications), thymidine was myeloprotective, resulting in a minimal need for dose reduction for patients having a >2 grade toxicity (in only 4% of the courses of treatment). Of 101 total courses, the number of courses (at the AUCs) was 3 (5), 4 (6), 58 (7), 20 (8), 11 (9), and 5 (10). Grade 3 nonhematologic toxicities included headache (4%), altered consciousness (3%), fatigue (1%), and nausea (3%). Responses included 2 partial (1 oligodendroglioma, 1 GBM; 5%); 3 minor (1 anaplastic astrocytoma, 2 GBM; 7.3%); 6 stable disease (14.6%); and 30 progressive disease (73.2%). For GBM patients, median survival was 23 weeks (with a 95% confidence interval of 20 to 50 weeks), and progression-free survival was 8 weeks (with a 95% confidence interval of 7-16 weeks). These results in GBM were comparable to other phase II GBM trials and thus do not represent a therapeutic advance in the treatment of GBM. Taken collectively, however, results are consistent with continued investigation of thymidine in combination with chemotherapeutic agents for high-grade glioma and other malignant diseases. The significant myeloprotection afforded by thymidine may have particular relevance to polychemotherapeutic regimens.