Abstract
In four phase II trials of trimetrexate given iv daily for 5 days, we noted marked variability among patients in the development of severe or life-threatening toxic effects. In an effort to define which of 15 patient characteristics were associated with toxic effects of this degree, we have carried out single-factor and multifactor analyses on toxic effects during the first cycle of therapy in 68 patients. The final logistic regression model identified both low pretreatment serum protein levels and metastatic liver disease as significant correlates of severe toxic effects (P = .02 and P = .0005, respectively). While drug dose was an important element of the best logistic model, its statistical significance was only borderline. Trimetrexate is extensively protein bound and is cleared primarily by hepatic metabolism, so it is not unreasonable to believe that alteration in protein binding of the drug or in metabolic capacity of the liver could produce enhanced toxic effects. Although the validity of these predictors should be confirmed, it seems prudent to recommend lower starting doses of trimetrexate for patients with metastatic liver disease and/or low protein levels and dose escalation if toxic effects allow it.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have