Grem J L, Ellenberg S S, King S A, Shoemaker D D
Investigational Drug Branch, National Cancer Institute, Bethesda, MD 20892.
J Natl Cancer Inst. 1988 Oct 19;80(16):1313-8. doi: 10.1093/jnci/80.16.1313.
Trimetrexate, an investigational antifol, has been associated with marked variability in drug tolerance among patients. The agent is extensively protein bound, and hepatic biotransformation plays a major role in its elimination. In early phase II testing, nine of 15 patients who experienced life-threatening or fatal toxic effects from trimetrexate had albumin levels less than or equal to 3.5 g/dL prior to treatment. This prompted a review of the data base on 272 patients entered in phase I clinical trails. The incidence of severe or life-threatening anemia, leukopenia, neutropenia, thrombocytopenia, mucositis, and hepatic toxic effects during the first course of trimetrexate was analyzed according to dose, schedule, prior treatment, and baseline protein and albumin levels. The schedules using doses given by short infusions of 30-60 minutes daily for 5 days or weekly for 3 weeks were generally associated with higher incidence of toxic effects than the schedules using doses given every other week by short infusions or those using continuous infusion. The occurrence of leukopenia and mucositis was dose related. Patients with baseline albumin levels less than or equal to 3.5 g/dL had higher incidence of all types of severe or life-threatening toxic effects than those with albumin levels greater than or equal to 3.6 g/dL, and the differences were significant for the development of anemia, thrombocytopenia, and mucositis. Similar correlations were noted for pretreatment protein levels less than or equal to 6.0 g/dL. The small cohort of patients with leukemia experienced substantial toxic effects and tended to have low protein and albumin levels. Performance status and prior therapy did not emerge as strong predictors of severe toxic effects in the univariate analysis. Multivariate analysis confirmed that the type of cancer (leukemia vs. solid tumor), dose, schedule, and baseline albumin level were significant and independent predictors of severe and life-threatening toxic effects in the phase I patient population. Multivariate analysis including only patients with solid tumors indicated that albumin level, dose, and schedule remained significant predictors of toxic effects. Since normal liver function as reflected by bilirubin and transaminase values were a requirement for eligibility, the results suggest that albumin and protein levels may provide a more sensitive index of hepatic function. Patients with hypoalbuminemia and hypoproteinemia are at increased risk of experiencing severe or life-threatening toxic effects from trimetrexate and should be treated cautiously.
三甲曲沙是一种正在研究中的抗叶酸剂，患者对该药的耐受性存在显著差异。该药物与蛋白质广泛结合，肝脏生物转化在其消除过程中起主要作用。在II期早期试验中，15名因三甲曲沙出现危及生命或致命毒性反应的患者中，有9名在治疗前白蛋白水平低于或等于3.5g/dL。这促使对I期临床试验中纳入的272名患者的数据库进行回顾。根据剂量、给药方案、先前治疗以及基线蛋白质和白蛋白水平，分析了三甲曲沙首个疗程中严重或危及生命的贫血、白细胞减少、中性粒细胞减少、血小板减少、粘膜炎和肝毒性反应的发生率。与每隔一周短时间输注给药方案或持续输注给药方案相比，每天短时间输注30 - 60分钟、持续5天或每周输注3周的给药方案通常与更高的毒性反应发生率相关。白细胞减少和粘膜炎的发生与剂量相关。基线白蛋白水平低于或等于3.5g/dL的患者，所有类型的严重或危及生命的毒性反应发生率均高于白蛋白水平大于或等于3.6g/dL的患者，且在贫血、血小板减少和粘膜炎的发生方面差异显著。对于治疗前蛋白质水平低于或等于6.0g/dL的情况，也观察到了类似的相关性。一小群白血病患者经历了严重的毒性反应，且往往蛋白质和白蛋白水平较低。在单变量分析中，体能状态和先前治疗并未成为严重毒性反应的有力预测因素。多变量分析证实，癌症类型（白血病与实体瘤）、剂量、给药方案和基线白蛋白水平是I期患者群体中严重和危及生命的毒性反应的显著且独立的预测因素。仅对实体瘤患者进行的多变量分析表明，白蛋白水平、剂量和给药方案仍然是毒性反应的显著预测因素。由于胆红素和转氨酶值反映的正常肝功能是入选的一项要求，结果表明白蛋白和蛋白质水平可能提供一个更敏感的肝功能指标。低白蛋白血症和低蛋白血症患者发生三甲曲沙严重或危及生命的毒性反应的风险增加，应谨慎治疗。
