Phase I trial of the matrix metalloproteinase inhibitor BAY12-9566 in patients with advanced solid tumors.

PURPOSE: Matrix metalloproteinases (MMPs) are a family of proteolytic enzymes that are believed to be involved in primary and metastatic tumor growth by degrading the basement membrane and changing the extracellular matrix to facilitate invasion of malignant cells and angiogenesis. Overexpression of MMPs has been documented in various solid tumors. BAY12-9566 is a selective inhibitor of MMPs, in particular MMP-2, -3. and -9. The purpose of this trial was to define the maximum tolerated dose (MTD), dose-limiting toxicities (DLT), safety profile, pharmacokinetics and pharmacodynamics of orally administered BAY12-9566 in patients with incurable solid tumors. METHODS: The starting dose of BAY12-9566 for this single institution, outpatient phase I study was 100 mg/day orally. Patients were allowed to receive drug for up to 12 months. A total of 27 patients with various solid malignancies including colorectal, breast, lung, cervical and ovarian cancers were enrolled at doses from 100 to 1,600 mg/day. Patients were evaluated weekly while on treatment. Relevant radiologic examination was performed every 8 weeks to document and follow sites of measurable or evaluable disease. RESULTS: Toxicities from BAY12-9566 included liver injury test abnormalities, anemia, shoulder and back pain. thrombocytopenia, mild nausea and fatigue, diarrhea, rash and deep vein thrombosis. No toxicity greater than grade III was observed. As doses were increased from 100 to 400 to 1,600 mg/day, even in divided doses, less than proportional increases in AUC were observed. At the highest dose level of 1600 mg/day, the day 29 AUC (3778.00 mg x h/l) remained similar to the day 29 AUC (3312.60 mg x h/l) at the dose level of 1200 mg/day. No responses were seen, but 14 patients remained on study with stable disease for 4 to 26 months. CONCLUSIONS: BAY12-9566 was well tolerated at doses as high as 800 mg orally twice daily. Although mild alterations in liver injury tests, platelet count and hematocrit were noted, these were not dose-limiting. The drug was well absorbed. However, the absence of proportional increases in AUC with doses greater than 800 mg and the achievement of Css in the range associated with biologic activity in preclinical models led to the selection of 800 mg twice daily for further evaluation in phase III trials.

A basic compositional requirement of agents having heparin-like cell-modulating activities.

Heparin has been recognized as possessing a large variety of cell-modulating activities. Using compositionally simple, structurally rigid, and low molecular weight saccharide molecules (cyclodextrins), we demonstrated that these activities depend primarily on a single, gross compositional parameter: a minimum intramolecular density of neighboring anionic (sulfate) groups. This same critical parameter is shown to be involved in achieving cell-modulating behavior as diverse as angiogenesis, endothelial proliferation, inhibition of smooth muscle cell growth, and cell protection against virus invasion. Physical chemical evidence is presented that associates this property with multi-ionic complex formation between the clusters of anionic and cationic sites on the complexing partners. These observations revive early suggestions of the decisive role of electrostatic complexation capabilities of glycosaminoglycans like heparin; taken together with numerous observations on heparinoids and other agents reported in diverse specialized fields of cell biology and medicine, they provide evidence that molecular agents of critical anionic (sulfate) density (MACADs) represent a broad class of molecules that, in contrast to proteins, do not rely on structural detail for their cell biological activities, but function by ionic complexation with proteinic agents (e.g. growth factors), thereby modifying their structure-specific activities.

Impulsiveness and subjective effects of intravenous cocaine administration in the laboratory.

Cardiovascular and subjective responses to placebo and 40-mg intravenous (iv) cocaine injections were measured in 29 male iv cocaine users: most subjects received each of these injections on two separate occasions. Most of the subjects also completed various measures of psychopathology and personality. Although the small sample size made any conclusions tentative, an expected significant association between impulsivity and subjective euphoria following 40-mg cocaine administration was obtained, whereas associations of personality measures with cardiovascular responses to cocaine administration were inconsistent.

Protection of erythrocytes against hemolytic agents by cyclodextrin polysulfate.

Cyclodextrins generally exhibit hemolytic activity, some at concentrations as low as 1-10 mg/mL or lower. However, we found previously that a highly polysulfated cyclodextrin has no demonstrable hemolytic activity (Macarak et al., Biochem Pharmacol 42: 1502-1503, 1991). In the present study, we determined that, in fact, cyclodextrin polysulfate (CDS) actively protected erythrocytes against hemolysis induced by a wide spectrum of hemolytically active substances, ranging from pharmaceuticals, such as chlorpromazine, to solid suspensions of siliceous particles. The protective action was also effective against the hemolytic action of non-sulfated cyclodextrins. The similar kinetic responses of the erythrocytes to CDS protection against such chemically and structurally diverse hemolytic agents suggest a common mechanism involving the cell. Addition of the sulfated cyclodextrins to other cyclodextrin compounds used to solubilize poorly soluble pharmaceutical agents can extend the allowable maximum dosage without deleterious hemolytic action.

Test-retest stability of cardiovascular and subjective responses to intravenous cocaine in humans.

Sixteen male i.v. cocaine users were measured on their cardiovascular and subjective responses to placebo followed on a separate day by 40 mg i.v. cocaine injections. They were retested within 2 weeks, again receiving placebo and 40 mg i.v. cocaine injections in a random order on separate days. Significant increases in baseline (pre-injection) heart rates during the later sessions were interpreted as possibly reflecting conditioning effects. There were no significant differences in post-injection increases in cardiovascular or subjective responses between the initial and later 40 mg conditions, which might have been indicative of tolerance or sensitization development. Test-retest correlations, indicative of response stability, were moderate to high for any particular timepoint for blood pressure, heart rate, and subjective responses, but pre- vs. post-injection change scores were stable across testings on these measures only under placebo. With the possible exception of some subjective responses, there was little evidence of test-retest stability in responses (change scores) to the 40 mg cocaine injection.

A pharmacokinetic-pharmacodynamic model of heart rate during cocaine administration in humans.

We developed a pharmacokinetic-pharmacodynamic model of the heart rate response to cocaine consisting of the sum of the baseline heart rate, a chronotropic drug effect, and a conditioned heart rate response. The conditioned response was modeled as having a monoexponential decline. Cocaine's chronotropic effect was modeled as obeying a maximum (Emax) drug-effect relationship without tachyphylaxis. The model was tested by fitting the kinetic and heart rate data from an earlier study. The data from five subjects in that study were well fit by the model: mean Emax, 64 beats/min, and mean drug concentration producing 50% of Emax, 838 ng cocaine/ml. The data from the other three subjects in that study were fit well, employing a linear rather than an Emax drug effect relationship: mean effect coefficient, 0.035 beats/min/ng cocaine/ml. The findings emphasize the importance of conditioning in the responses to psychoactive drugs.

Effects of nifedipine pretreatment on subjective and cardiovascular responses to intravenous cocaine in humans.

The effects of oral nifedipine pretreatment on subjective and cardiovascular responses to intravenous cocaine infusions were studied in cocaine-using volunteers. Nifedipine, 10 mg or placebo, was administered 20-25 min before placebo, 20 mg, or 40 mg cocaine, using a repeated measures randomized double-blind design. The variables measured were self-reported subjective effects, general behavior rated by two observers, blood pressure and heart rate. Cocaine produced the expected dose-related effects on subjective and cardiovascular measures. Nifedipine pretreatment attenuated some subjective effects of cocaine. Nifedipine directly reduced blood pressure but did not antagonize the effects of cocaine on blood pressure. These findings suggest that dihydropyridine calcium channel modulators may be useful compounds in the clinical management of cocaine users.

Multidimensional scaling of subjective judgements of drug similarities among ketocyclazocine, morphine, cyclazocine, naloxone and placebo.

Profiles of the subjective and physiologic effects of opioid drugs in man cannot be assigned with precision to specific opioid drug-receptor interactions. We administered a set of training doses of ketocyclazocine, morphine, cyclazocine, naloxone and placebo to 10 drug-using volunteers and obtained similarity judgements between each of 2 test doses of the drugs and a training dose. These data were submitted to multidimensional scaling analysis (INDSCAL) using both neighboring cells estimates and root mean square estimates to estimate missing cells in the data matrices. The results of these analyses are convergent, appear valid and indicate that there are three drug dimensions expressed in this data set: morphine versus placebo and naloxone; cyclazocine and ketocyclazocine versus placebo and naloxone; and ketocyclazocine versus cyclazocine. We interpret this result as supporting evidence that in the set of five drugs studied, three subjective states are induced.

Intravenous cocaine infusions in humans: dose responsivity and correlations of cardiovascular vs. subjective effects.

Eight experienced IV cocaine users were intravenously administered 0, 10, 20, and 40 mg of cocaine hydrochloride on separate days in a pseudo-randomized ascending dose series, such that the 20 mg dose always preceded the 40 mg dose. They were subsequently administered 0, 20, and 40 mg of cocaine in a fully randomized presentation order. Cardiovascular effects of cocaine were significantly different from placebo for the 20 mg, but not the 10 mg dose, in contrast to subjective responses which differed from placebo for the 10 mg dose. Cardiovascular and subjective effects of cocaine did not differ between the 20 and 40 mg dose conditions for the pseudo-randomized trials, but did differ in the fully randomized trials. This lack of difference in responsivity between the 20 and 40 mg dose in the earlier trials may possibly have been due to contrast effects. Cardiovascular responses were not consistently correlated with subjective responses, either within a cocaine dose condition or across doses.

Effects of bromocriptine pretreatment on subjective and physiological responses to i.v. cocaine.

We studied the effect of pretreatment with single doses of bromocriptine on the pattern of subjective and physiologic responses to single doses of intravenous (IV) cocaine. Placebo, bromocriptine 2.5 mg and, in five subjects only, 5 mg were administered orally 120 minutes before a dose of placebo or cocaine 40 mg IV to 9 male cocaine-using volunteers. Bromocriptine pretreatment diminished blood pressure generally, including cocaine-induced blood pressure increases, and augmented the heart rate after cocaine. It caused virtually no change in either augmentation or diminution of subjective responses including "rush" and "good feeling" scores, and scores for the MBG of the Addiction Research Center Inventory (ARCI), all measures of euphoria. However, a trend for the scores for the item "Would a dose of drug (cocaine) make you feel better?," suggested that bromocriptine may decrease the urge to use cocaine that was evoked by cocaine itself. However, this decreased desire was associated with a trend toward an increase in dysphoria as measured on the LSD scale of ARCI. These data support the view that euphoria and some forms of craving may be pharmacologically separable. We found no potentially toxic interactions of bromocriptine with cocaine in these single dose experiments.

Subjective response during continuous infusion of cocaine.

The relationship between the subjective effects induced by IV cocaine injection(s) and cocaine plasma concentrations is complex and difficult to interpret. We designed a study in which bolus loading doses of cocaine followed by 4-hr placebo infusions were compared with the same bolus loading doses of cocaine followed by 4-hr infusions of cocaine calculated to maintain the peak plasma concentrations produced by the bolus. Seven cocaine-using volunteers were successfully studied using a randomized double-blind design, in which self- and observer-rating scales were used to measure drug effects. After the cocaine bolus loading doses, scores for most subjective measures remained elevated when the bolus was followed by a cocaine infusion. In contrast, the subjective responses returned to baseline when the bolus was followed by a placebo infusion. However, self-estimates of the intensity of the cocaine "rush" were not altered by the presence of active cocaine infusions and returned rapidly to baseline.

Effects of acute and chronic acyclovir on canine renal function.

Acyclovir-associated renal dysfunction is difficult to study in humans because it occurs in patient populations who are generally receiving multiple medicines concurrently and dysfunctions occur infrequently. We studied the effects of two regimens of acyclovir treatment, a short-term high dose (210 mg/kg/day for 43 hr) via constant infusion and a chronic lower dose by intermittent infusion (15 mg/kg, 3 times/day for 28 days), on selected renal functions in dogs. Urine concentrating capacity as compared to base-line and controls declined during both the short-term (40% decline) and chronic treatments (36% decline). The persistence of the concentrating deficit in the presence of stimulation with vasopressin suggested that the defect resided in the renal response to vasopressin. Glomerular filtration rates significantly decreased during the acute high-dose treatment (from 104 +/- 15 to 87 +/- 11 ml/min) but not during the chronic low-dose treatment. Both acyclovir treatments were associated with a small but significant decline (approximately 10% for both studies) in the plasma potassium concentrations although an increase in urine potassium clearance could not be demonstrated. We conclude that acyclovir can cause renal dysfunction in healthy animals at plasma concentrations higher but comparable to those achieved clinically. Although both treatment regimens decreased renal function, the shorter high-dose regimen which maintained acyclovir plasma concentrations was more detrimental than the longer exposure to a lower dose of the drug given intermittently. The data suggest that renal dysfunction is not related to a rare sensitivity, as all acyclovir-treated animals showed subtle decrements in renal function.

Cardiovascular responses to cocaine placebo in humans: a preliminary report.

Cardiovascular responses after placebo-cocaine injections were in the same direction as the effect of cocaine iv in 22 male volunteers. Subjects received iv placebo in a room where they had been given repeated doses of iv cocaine. The placebo response consisted of an increase from baseline values of systolic and diastolic blood pressure and pulse rate. The control group, 8 subjects, which was not exposed to a conditioning phase, showed a smaller increase in the pulse rate and systolic blood pressure after the placebo injection. The results, in accordance with animal literature, suggest the existence of cocaine-conditioned effects in humans.

Effects of intravenous cocaine are partially attenuated by haloperidol.

We examined the effect of the dopaminergic blocking agent, haloperidol, on the subjective and physiologic response to cocaine in cocaine-using volunteers. Five subjects received cocaine (40 mg) or placebo administered intravenously 20 min following pretreatment with haloperidol (8 mg) or placebo intramuscularly in a randomized double-blind study design. Haloperidol pretreatment attenuated cocaine-induced increases in systolic and diastolic blood pressure but not heart rate. Pretreatment with haloperidol reduced subject ratings of pleasant sensations but had no effect on drug "rush." Haloperidol (8 mg) has a small and limited effect on the subjective response to cocaine when given 20 min before cocaine.

Placebo responses to cocaine administration in humans: effects of prior administrations and verbal instructions.

Subjective and physiological responses of eight male cocaine-using research volunteers were studied after a double-blind saline infusion (placebo) was given when subjects were instructed that a cocaine infusion might be given. Cardiovascular and subjective responses to placebo were similar in pattern and direction, though of lesser magnitude, than after a 40 mg cocaine infusion. These placebo responses were compared to responses after an earlier saline infusion condition in which subjects were instructed prior to the infusion that they would receive saline (instructed placebo). The design was thus meant to test for the effects of instructions on placebo responses to cocaine. Heart rates at baseline (pre-infusion) were significantly higher in the placebo than in the instructed placebo condition. Similar trends were found for elevated baseline placebo responses on two subjective effects measures. A comparison with an initial placebo session prior to the placebo and instructed placebo conditions described above provided evidence for conditioning of placebo responses on diastolic blood pressure and heart rate. The present results suggest that verbal instructions, as well as conditioning in the laboratory, could contribute to the observed placebo responses to cocaine infusions.

Cocaine-induced cocaine craving.

In nine experienced users of cocaine, we examined the urge to use cocaine or other drugs following a 40 mg dose of intravenous (IV) cocaine with and without oral pretreatment with 2.5 mg bromocriptine. The urge to use cocaine was assessed with a questionnaire constructed to assess both "wanting" and "craving" for cocaine or other drugs. Fifteen minutes after the administration of cocaine (but not after placebo), subjects' ratings for both drug "wanting" and drug "craving" were significantly increased. Our results provide a laboratory demonstration of cocaine-induced increases in the urge to use drugs in humans. The findings, stressing the role of internal stimuli associated with drug administration, suggest the possibility of distinguishing among related, but perhaps distinct, components of the fluctuating levels of motivation to reuse drugs.