Phase I Study of Docetaxel in Combination with the P-Glycoprotein Inhibitor, Zosuquidar, in Resistant Malignancies Free

Development of resistance to antineoplastic agents presents an obstacle to the treatment of human malignancies. The phenomenon of multidrug resistance (MDR), whereby tumor cells develop cross-resistance to multiple unrelated agents, has been attributed, in part, to the expression of the MDR-1 gene, which encodes a M_r 170,000 membrane P-glycoprotein, and expressed in a wide variety of human malignancies and normal tissues (1). P-glycoprotein, a member of the ATP-binding cassette superfamily of transmembrane transporters, prevents the intracellular accumulation of many natural product-derived cytotoxic agents (2).

Because of the potential clinical importance of P-glycoprotein-mediated drug resistance, several agents have been administered in combination with cytotoxic agents to reverse this resistance (3, 4, 5, 6, 7, 8, 9). Eli Lilly and Company identified zosuquidar (LY335979), a difluorocyclopropyl quinoline, which binds with high affinity to P-glycoprotein (10). In vitro, zosuquidar inhibited P-glycoprotein activity with a Ki of ∼60 nmol/L with complete reversal of drug-resistance requiring concentrations at ∼100 nmol/L. In a Phase I study with doxorubicin, zosuquidar did not produce pharmacokinetics interactions with coadministered doxorubicin (11).

This study was designed as a single-arm, dose-escalation, Phase I clinical trial for the following reasons: (a) to determine the maximum tolerated dose and dose-limiting toxicity of docetaxel given in combination with zosuquidar; (b) to investigate the effect of zosuquidar on the pharmacokinetics of docetaxel; and (c) to evaluate the toxicities of docetaxel and zosuquidar combined. Treatment response was documented but was not a primary endpoint.

Patients at least 18 years of age were eligible for enrollment onto the study if they had locally advanced or metastatic cancer and no more than two prior regimens of chemotherapy, measurable disease, an Eastern Cooperative Oncology Group performance status of 0 to 2, and a life expectancy of at least 12 weeks. Laboratory criteria for eligibility included granulocytes ≥1.5 × 10⁹ cells/L; platelets ≥100 × 10⁹ cells/L; hemoglobin ≥9.0 g/dl; bilirubin ≤ upper limit of normal; alanine transaminase and aspartate transaminase ≤ 1.5 times upper limit of normal, alkaline phosphatase ≤2.5 times upper limit of normal, and a serum creatinine ≤1.5 mg/dl or a 24 hours clearance of >50 mL/minute. No chemotherapy, radiation therapy, or surgery was permitted within 3 weeks before study entry or 6 weeks if prior treatment was nitrosourea or mitomycin-C, and no investigational agent was permitted within 4 weeks before study entry. Patients with the following conditions were ineligible: hematologic malignancies, substantial comorbid conditions (e.g., congestive heart failure, angina pectoris, cardiac arrhythmias requiring medical therapy, or systemic infections requiring parenteral antibiotics), or a baseline neurotoxicity of National Cancer Institute Common Toxicity Criteria v.2 grade 2 or higher. Women who were either pregnant or breast-feeding were ineligible. Written informed consent was obtained from all of the patients before they were entered onto the study, in compliance with all institutional and federal regulations.

Each treatment cycle was 21 days. In cycle 1, patients received only docetaxel. In cycle 2 and all of the other cycles, patients received the combination of docetaxel and oral zosuquidar. The initial dose of zosuquidar was based on results of previous Phase 1 studies (11, 12).

Patients were eligible to continue treatment unless there was intolerable toxicity or evidence of progressive disease. Dose escalation was not allowed until the first 2 patients in the current dose level completed cycle 2 and recovered from any acute therapy-related toxicities. Before the start of any cycle, the patient was required to have an absolute neutrophil count ≥1.5 × 10⁹ cells/L, platelets ≥100 × 10⁹ cells/L, and hemoglobin ≥9 g/dl and evidence of recovery from any grade >2 nonhematopoietic toxicity other than alopecia. Treatment delays of 2 weeks were allowed for recovery of neutrophils or platelets. Patients were to be removed from study for delays >2 weeks. Docetaxel doses were to be adjusted for toxicity occurring in the previous cycle according to the following criteria: 25% dose reduction for grade 4 neutropenia lasting ≥7 days, febrile neutropenia, or grade 4 infection; 25% dose reduction or hold treatment for grade 3 nonhematologic toxicities excluding nausea and vomiting; 50% dose reduction or hold treatment for grade 4 nonhematologic toxicity; 25 to 50% dose reduction (at discretion of the investigator) for total bilirubin >upper limit of normal, alanine transaminase or aspartate transaminase >1.5 times upper limit of normal or alkaline phosphatase >2.5 times upper limit of normal. The prophylactic use of growth factors (granulocyte colony-stimulating factor or granulocyte macrophage colony-stimulating factor) was prohibited except for patients with an absolute neutrophil count <0.5 × 10⁹/L for at least 7 days, neutropenic fever, or documented neutropenic infections. Zosuquidar doses were reduced to the previous level for grade 3 toxicities, excluding hallucinations, and discontinued for any grade 4 nonhematopoietic toxicities. If there was a dose reduction, dose escalation to the original dose was allowed providing the patient tolerated the reduced dose. If a second dose reduction was necessary, the dose was not re-escalated for the remainder of the trial.

Dose-limiting toxicity was determined by the toxicity observed on cycle 2 when docetaxel was combined with zosuquidar. Dose-limiting toxicity was defined as grade 3 or higher nonhematopoietic toxicity (excluding alopecia, nausea, or vomiting), grade 3 hallucinations, and grade 4 hematopoietic toxicity lasting ≥7 days. If dose-limiting toxicity occurred in 1 or 2 of the 3 patients treated at a given dose level, up to 5 additional patients would be treated at that dose level. If 3 patients in this expanded dose level experienced a dose-limiting toxicity, enrollment would be terminated. If only 1 of 6 or 2 of 8 patients experienced a dose-limiting toxicity, dose escalation proceeded. The maximum tolerated dose of docetaxel and zosuquidar was defined as one dose level lower than that of which at least 3 of a possible 8 patients experienced a dose-limiting toxicity.

Antitumor response was evaluated by physical examination and/or by imaging after the second cycle of therapy and after every other subsequent cycle. Responses were defined according to the criteria defined by the World Health Association (WHO Handbook, 1979).

Plasma samples were obtained to determine the pharmacokinetics of docetaxel (cycles 1 and 2) and zosuquidar (cycle 2). For docetaxel (level 1 to 4), plasma samples were obtained before the infusion and during cycle 1 and cycle 2 at 0.5, 1 (just before the end of the infusion), 1.25, 1.75, 2.5, 4, 8, 12 (optional), 24, and 30 hours after the initiation of the infusion. In level 5 and subsequent dose levels, plasma sampling was carried out before docetaxel and 0.5, 1 (just before end of infusion), 1.25, 2, 3, 4, 6, 10, 11, 12, 13, 14, 16, 18, 22, and 28 hours after the initiation of the infusion. Plasma samples were analyzed for docetaxel with a validated high-performance liquid chromatography method with UV detection, over a concentration range of 10 ng/mL to 2,500 ng/mL (13).

For zosuquidar (dose levels 1 to 4), plasma sampling was carried out at predose, 1, 2, 4, 6, and 8 hours after the fourth and seventh doses. In dose levels 5 and above, zosuquidar plasma sampling was carried out before the first dose and 1, 2, 2.5, 3, 3.5, 4, 5, 6, 8, 12, 13, 14, 15, 16, 18, 20, 24, and 30 hours from the time of the first dose in the second cycles. Plasma samples were analyzed with a validated high-performance liquid chromatography/fluorescence detection method over the concentration range 20 ng/mL to 2000 ng/mL as published previously (11).

Pharmacokinetic analyses of docetaxel and zosuquidar were evaluated using noncompartmental methods (WinNonlin Professional Version 3.1). The areas under the plasma concentrations versus time curves (area under the curve; AUC) were obtained with a combination of the linear and logarithmic trapezoidal methods. The linear trapezoidal method was used up to t_max, and then log trapezoidal rule was used. Area under the curve was calculated according to the following:

where t_n is the last time point where the plasma concentration is above the limit of quantification, C(t_n)′ is the prediction for the concentration at the last quantifiable time point, and λ_z is the calculated terminal rate constant.

C_max and t_max were directly determined from the observed concentration-time profiles. Other pharmacokinetics parameters assessed were mean residence time (MRT), systemic clearance (CL), volume of distribution (Vss), and time above a threshold concentration of 0.2 μmol/L (T > 0.2 μmol/L) as described in the following equations:

Please note T is the infusion time.

The duration of time for which plasma docetaxel concentrations remained at or above 0.2 μmol/L were calculated according to the following:

Please note C2 is docetaxel last plasma concentration above 0.2 μmol/L measured at time T2; C3 is docetaxel first plasma concentration below 0.2 μmol/L measured at time T3; and T > 0.2 μmol/L is the time during which docetaxel plasma concentrations are above 0.2 μmol/L.

Because of the dose reduction of docetaxel between cycle 1 and cycle 2 in some subjects and variation in actual docetaxel doses between subjects, dose-normalized AUC_(0-tn), AUC_(0–12), AUC, and C_max were also calculated to facilitate treatment comparison.

After oral administration of zosuquidar, AUC_(0-tn), AUC_ι (AUC over the dosing interval), C_max and t_max were calculated after dose 7 in dose levels 1 to 4 and dose 1 in dose levels 5 and above. Dose-normalized AUC_(0-tn), AUC_ι, and C_max were also obtained.

Characteristics and prior therapy of the 41 eligible patients enrolled from March 1999 until April 2002 are summarized in Table 1. Patients from three centers, two in the United States and one in the Netherlands, received treatment. Most patients had prior treatment, although no patient had >2 prior regimens as required by protocol. Thirteen patients had received prior taxane therapy. The mean age was 56 years (range 26 to 79), 56% patients were female and 44% male.

Dose escalations were performed as listed in Table 2. Dose levels 1 to 4 of zosuquidar were based on body surface area and administered every 8 hours for a total of 7 doses beginning the day before the 2nd cycle of docetaxel. In dose levels 1, 2, and 3, zosuquidar at 100, 200, or 300 mg/m² every 8 hours was given for a total of 7 doses, respectively. In level 4, zosuquidar 300 mg/m² was given every 8 hours for a total of 7 doses with docetaxel escalated to 100 mg/m². In dose levels 1 to 4, docetaxel was administered 2 hours after the fourth dose of zosuquidar was given.

Pharmacokinetic data, from an ongoing study, failed to show any positive correlation between the clearance of zosuquidar and body surface area (11, 12). Thus, dosing by body surface area contributed to variability in plasma concentration and potentially to P-glycoprotein inhibition. To achieve more predictable drug levels and ensure optimal P-glycoprotein inhibition in all of the patients for a given dose, the regimen was altered to a flat dose. Additionally, prolonged administration did not result in the plasma accumulation of zosuquidar (11, 12). Consequently, the dose of zosuquidar was changed to a flat dose given twice 12 hours apart in patients enrolled in dose levels 5 to 8. Patients in dose levels 5 to 7 received zosuquidar 400 mg, 450 mg, and 500 mg, respectively, administered every 12 hours for a total of 2 doses. In dose levels 5 to 8, docetaxel was administered 2 hours after the first dose of zosuquidar and was given on day 1 at a dose of 75 mg/m². Patients in level 8 received zosuquidar 500 mg with docetaxel 100 mg/m².

Thirty-six of the 41 patients completed at least one cycle of the docetaxel and zosuquidar treatment and were evaluable for toxicity and response. One of five patients enrolled in level 4 did not receive the combination therapy, and one of five patients enrolled in level 5 chose to discontinue treatment after the first cycle. Only two were infused at level 6 because ongoing data concerning the pharmacokinetics of zosuquidar from previous cohorts suggested that the dose increment was insufficient to achieve a better pharmacodynamic effect than the dose at level 5 given the variability in the plasma pharmacokinetics. Two of 11 patients enrolled in level 7 were replaced after the first cycle, one for hypersensitivity reaction during the docetaxel infusion and the other for progressive disease. In level 8, docetaxel was escalated to 100 mg/m² with the same dose of zosuquidar as given in dose level 7. Ten patients were enrolled at this level, although one patient developed progressive disease after the first cycle and did not receive the combination therapy. A total of 152 cycles of docetaxel and zosuquidar were administered, with a median of 3 cycles per patient (range 1 to 39 cycles).

The majority of cycles of combination therapy (141 of 152 cycles) were administered on schedule. Dose adjustments including delays and reductions secondary to toxicity were made in 10 patients over 11 cycles. The majority of dose adjustments occurred in 5 patients from dose levels 4 and 8.

Neutropenia was the most frequent hematopoietic toxicity observed in this study (Table 3). Ten patients had febrile neutropenia, one patient each from dose levels 1, 2, 3, and 6 and 2 patients each from dose levels 4, 7, and 8. Seven of these patients had their first episode of febrile neutropenia occur during cycle 1, where they received docetaxel alone, with five receiving docetaxel 75 mg/m² and two receiving docetaxel 100 mg/m². The three remaining patients had febrile neutropenia during later cycles when they received the combination of docetaxel and zosuquidar. Each of the seven patients experiencing febrile neutropenia in cycle 1 received docetaxel at a 25% dose reduction in cycle 2. One receiving 75 mg/m² and one receiving 100 mg/m² went on to experience an additional 25% dose reduction of docetaxel because of febrile neutropenia in later cycles. The remaining five had no additional dose reductions. One patient experienced a delay in the start of cycle 6 because of neutropenia in the absence of fever. Grade 3 and 4 neutropenia was noted in 8 and 22 patients, respectively. Four patients developed grade 3 anemia. One patient each in level 8 and 1 evidenced grade 3 and 4 thrombocytopenia, respectively. As noted in the first two lines of Table 3, zosuquidar had little effect on myelosuppression. The absolute neutrophil count nadir and range of patients receiving docetaxel alone were not substantially different from the absolute neutrophil count nadir and range of the combination of docetaxel and zosuquidar.

The nonhematologic toxicities are summarized in Table 4. All of the other nonhematologic toxicities were ≤grade 2. Patients experienced fatigue, nausea, vomiting, and neurologic toxicity. All of the four patients in level 4 (docetaxel 100 mg/m² and zosuquidar 300 mg/m² for 7 doses) experienced grade 3 neurologic toxicities including ataxia, tremors, and motor neuropathy, consistent with the toxicity associated with high doses or prolonged administration of zosuquidar. These data, in combination with the pharmacokinetic data on zosuquidar, led to a flat dose of zosuquidar given twice 12 hours apart in dose levels 5 to 8. The occurrence of grade 2 or greater ataxia or tremor was substantially reduced with the shorter dosing schedule [5 of 14 patients in dose levels 1 through 4 versus 1 of 27 patients in dose levels 5 through 8 (data not shown)]. There were no occurrences of grade 3 ataxia or tremor in dose levels 5 through 8. Other grade 3 nonhematologic toxicities included hypocalcemia, hypophosphatemia, and elevated serum glutamic oxaloacetic transaminase, which did not occur during the first cycle of combination therapy and did not require dose adjustments or discontinuation from the study. There were no grade 4 nonhematologic toxicities.

Thirty-six patients with measurable disease were evaluated for response. A patient with ovarian carcinoma, previously treated with paclitaxel ∼1 year before study, had a partial response for 11 months. Fifteen patients had evidence of stable disease; three women with breast cancer had stable disease for 3, 6.5, and 37.5 months. Twenty patients had progressive disease.

Because of limited quantifiable concentrations of docetaxel in the terminal phase of the concentration profiles, reliable estimation of t_1/2, CL, and V_ss of docetaxel could not be obtained for more than 50% of the subjects. Therefore, treatment comparisons were performed with AUC_(0-tn), partial AUC (AUC _(0–12)), and C_max. These parameters were dose-normalized when included in the comparisons to eliminate the complication caused by dose reductions between cycles 1 and 2.

Table 5 shows that dose-normalized AUC_(0-tn), AUC _(0–12), and C_max of docetaxel are marginally increased with the presence of zosuquidar. Extended dosing (dose levels 1 to 4) did not show additional benefit over the reduced schedule of zosuquidar (dose levels 5 to 8). The presence of zosuquidar marginally increased T >0.2 μmol/L (an increase of 4.99%) in level 8. A zosuquidar concentration >100 nmol/L is considered adequate to reverse P-glycoprotein-mediated resistance in vitro (10). This concentration was obtained when administering zosuquidar 500 mg for 2 doses.

Fig. 1 shows the zosuquidar plasma concentrations versus time profiles after the fourth and seventh doses (dose levels 1 to 4, 8 hourly regimen) or the first dose (dose levels 5 to 8, 12 hourly regimen). With extended dosing (dose levels 1 to 4), the plasma concentrations of zosuquidar after the seventh dose tended to be lower than those after the fourth dose, especially because one would anticipate accumulation of zosuquidar plasma concentrations with multiple dosing given the terminal elimination of half life of zosuquidar. This may additionally suggest an autoinduction of zosuquidar metabolism as previously observed after multiple administrations (11). Hence, extended dosing in this manner could result in below zosuquidar plasma concentrations to achieve the desired pharmacodynamic effect. Reducing the duration of administration ameliorates this possible effect.

Table 6 shows that zosuquidar AUC_(0-tn), AUC_τ, and C_max were almost 20% lower in the presence of docetaxel 100 mg/m² than docetaxel 75 mg/m². To have a better understanding of the dose-dependent effect of docetaxel on the pharmacokinetics of zosuquidar, zosuquidar data were pooled from all of the dose levels within the extended and reduced dosing schedules. Fig. 2 represents a scatter plot of dose-normalized AUC_τ and C_max versus absolute docetaxel doses (dose levels 5 to 8). Because dexamethasone was given along with docetaxel treatment and is an inducer of CYP3A4 isozymes (14), the data points were also indicated by the dose of dexamethasone to illustrate the effect of dexamethasone. The dose-normalized AUC_τ and C_max of zosuquidar tended to decrease as the dose of docetaxel increased.

Because of the potential importance of P-glycoprotein in clinical drug resistance and the ability to reverse this resistance both in vitro and in vivo by modifying agents, Phase I/II clinical trials have been undertaken with agents such as verapamil, tamoxifen, toremifene, quinidine, trifluperazine, and cyclosporine A (15, 16). However, toxicities associated with the concentrations of these agents needed to modulate MDR have led to the development of agents with a more favorable therapeutic index. One such agent, zosuquidar, exhibited modulation of MDR in vitro and in vivo without substantial toxicity and no substantial interaction with the cytotoxic agent (10). As a result, zosuquidar appeared to be an excellent candidate modulator to be administered in combination with cytotoxic chemotherapy.

In this Phase I study, we determined that docetaxel at 75 mg/m² or 100 mg/m² can be administered safely in combination with zosuquidar 500 mg given for a total of 2 doses taken orally 2 hours before the docetaxel infusion and 12 hours later. The major toxicity of this study was substantial neutropenia. As predicted in a previous study with zosuquidar in combination with doxorubicin (11), equivalent neutropenia was noted when docetaxel dose was given alone or when it was administered in combination with zosuquidar. Unlike the enhanced myelosuppression noted with the combination of an antineoplastic agent and cyclosporine A or valspodar (4, 5, 6, 7, 8, 9), zosuquidar had little effect on myelosuppression. Nonhematologic toxicities were modest and not dose limiting. Neurologic toxicity including cerebellar ataxia, tremors, and dizziness were likely potential toxicities of zosuquidar. Sensory neuropathy, a known toxicity of docetaxel, may have been exacerbated by zosuquidar.

The pharmacokinetics parameters calculated in this study for docetaxel were consistent with that reported literature (17). A minor increase (∼10 to 15%) in the systemic exposure of docetaxel in the presence of zosuquidar was observed. Because the primary routes for systemic elimination of docetaxel is via hepatic metabolism by the cytochrome P450 (CYP) 3A4 system and biliary excretion, this increase may represent P-glycoprotein inhibition of biliary excretion (17). In addition, it has been well established that the neutropenia and increased fluid retention induced by docetaxel toxicity are related to the duration of time that plasma concentrations are at or above a threshold value of 0.2 μmol/L (17, 18). In the presence of zosuquidar, this time was increased by only 5% with docetaxel at a dose of 100 mg/m².

The zosuquidar concentrations were variable and lower than the range observed in previous studies (11, 12). The systemic exposure of zosuquidar was reduced in the presence of docetaxel in a dose-dependent manner. In this study, dexamethasone was given with docetaxel to reduce the incidence and severity of fluid retention and hypersensitivity reactions. Dexamethasone and docetaxel increase CYP3A activity (14). Reduced systemic exposure of zosuquidar after oral administration may be because of induced first-pass metabolism via CYP3A of zosuquidar by the combination of dexamethasone and docetaxel, but the data were insufficient to directly address this hypothesis. Data from an ongoing, randomized Phase 2 trial in combination with docetaxel in which pharmacokinetic samples are being collected over multiple cycles will better define the nature and degree of this interaction. In addition, the data will also allow for an assessment of the potential impact on the future development of this combination.

The plasma concentration of zosuquidar required to modulate clinical MDR is unknown. Binding of drug to plasma proteins and difficulties related to tumor drug delivery will likely require plasma concentrations to exceed 100 nmol/L, the concentration necessary to completely reverse P-glycoprotein drug resistance in vitro (10). Studies with isolated human liver microsomes have suggested that potentially substantial inhibition of the cytochrome P450 isoenzyme, CYP3A, will occur at concentrations of zosuquidar of 4000 nmol/L and higher. Thus, a rationale emerges for a target range of zosuquidar plasma concentrations when combined with a cytotoxic agent. The peak concentrations should optimally stay below 4000 nmol/L to avoid potential P450 interactions but achieve dose levels >100 nmol/L (in vitro concentration) needed to achieve full-modulating activity. In this study, zosuquidar C_max dose levels ranged from 46.7 μg/L to 414 μg/L for level 1 to 4 and 94.4 μg/L to 368 μg/L for dose levels 5 to 7. This corresponds approximately to 100 nmol/L to 700 nmol/L. The geometric mean C_max for dose levels 7 and 8 were 216 μg/L (∼400 nmol/L) and 175 μg/L.

In conclusion, this Phase I study showed that docetaxel with zosuquidar, a MDR reversing agent, can be administered with acceptable toxicity. The recommended Phase II doses are docetaxel 75 or 100 mg/m² as a 1 hour infusion, with oral zosuquidar 500 mg given 2 hours before the start of docetaxel followed by a second dose 12 hours later. The pharmacokinetics interaction of docetaxel and zosuquidar is minimal, allowing for the full dose administration of docetaxel. Given that taxane resistance has been attributed to several mechanisms including P-glycoprotein-mediated drug resistance (19), and the expression of P-glycoprotein is of prognostic importance in women with breast cancer (20), a Phase II study in women with advanced breast carcinoma is presently underway with this combination.

The authors thank nurses and data managers of the centers for caring for the patients on this trial, Dr. Teresa Vietti for her editorial assistance, and Maria Pinnell for her expert secretarial assistance.