PHASE-I TRIAL OF CHLOROGUINOXALINE SULFONAMIDE, WITH CORRELATION OF ITS PHARMACOKINETICS AND PHARMACODYNAMICS

To define a maximum tolerable dose, chloroquinoxaline sulfonamide (CQS ) was given as a 1-h infusion every 28 days to cancer patients for who m no effective standard therapy was available. Doses were escalated in cohorts of at least three patients each. Plasma for characterization of the pharmacokinetics of free and total CQS was obtained during and after the initial infusion and, when possible, during and after subseq uent infusions of CQS if the dose had been reduce. A total of 101 cour ses of CQS in 55 patients were evaluated. Dose levels ranged from 18 t o 3,700 mg/m(2). The dose-limiting toxicity was hypoglycemia, first re cognized at the 3,700-mg/m(2) dose. When dose-limiting hypoglycemia wa s recognized, patients were entered at successively lower doses, with close monitoring of plasma glucose and insulin concentrations being do ne in 26 patients. Grade 1-3 hypoglycemia occurred within 4 h of the t ermination of CQS infusion and cleared by 24 h. Symptomatic hypoglycem ia was more frequent at doses of CQS above 1,000 mg/m(2). Concomitant administration of 5% glucose did not ameliorate the hypoglycemia assoc iated with CQS doses of > 1,000 mg/m(2). The total calorie intake, per centage of ideal body weight, or percentage of weight lost did not exp lain the incidence or severity of hypoglycemia in 12 patients in whom these data were obtained. Cardiac tachyarrhythmias occurred in 7 patie nts who received CQS at doses of greater than or equal to 1,000 mg/m(2 ), and tachyarrhythmia was associated with hypoglycemia in 3 patients. Other toxicities were sporadic, but the frequency of toxicity was hig her at CQS doses of greater than or equal to 1,000mg/m(2). These toxic ities included fever, rash, lightheadedness, leukopenia, thrombocytope nia, alopecia, diarrhea, nausea, and vomiting. All toxicities were rev ersible. Mean peak plasma [CQS] and AUC increased with dose, with a su ggestion that peak plasma [CQS] plateaued at higher doses. The decline in plasma [CQS] was fitted to a three-compartment, open linear model. The terminal half-life ranged from 28 to 206 h. Total body clearance ranged from 44 to 881 ml/h with no evidence of saturation. Urinary exc retion of the parent compound in 24 h averaged < 5%. CQS not bound to plasma protein (free CQS) comprised 1%-17% of total plasma CQS and was not related to dose. A relationship was defined between the magnitude of hypoglycemia and CQS and was not related to dose. A relationship w as defined between the magnitude of hypoglycemia and CQS pharmacokinet ic parameters. The percentage of decrease in plasma [glucose], i.e., ( predose [glucose]-nadir [glucose]/predose [glucose]) x 100, correlated with both free and total peak plasma [CQS] The relationship was descr ibed by the Hill equation: Effect = (Emax) (peak)(H)/(peak(50))(H) + ( penk)(H), where the maximal effect (Emax) equals the maximal possible percentage of decrease in plasma [glucose] equals 100%, peak(H), is th e peak total [CQS] at which E is half-maximal (326 mg/l), and H is the Hill constant, a measure of the sigmoidicity of the relationship (1.0 6). The relationship fit the data precisely with a mean absolute error (MAE) of 10.42 and was unbiased with a mean error (ME) of - 0.06. The recommended phase II dose of CQS is 1,000 mg/m(2). Because the magnit ude of hypoglycemia after CQS administration is related to peak plasma [CQS], repetitive CQS doses of less than or equal to 1,000 mg/m(2) wo uld probably be tolerated better than single large doses of equivalent intensity.