A Phase I and Pharmacokinetic Study of Exisulind and Docetaxel in Patients with Advanced Solid Tumors Free

Currently, the induction of apoptosis in malignant cells is undergoing intense investigation as a therapeutic strategy for cancer patients. Exisulind (sulindac sulfone, FGN-1, Aptosyn), a proapoptotic drug, is a sulfone metabolite of the nonsteroidal anti-inflammatory drug (NSAID) sulindac. Exisulind promotes programmed cell death by inhibiting cyclic GMP (cGMP)-phosphodiesterases 2 and 5 that are often overexpressed in a variety of cancers (1). This inhibition results in sustained elevation of cGMP, which leads to the activation of c-GMP–dependent protein kinase G (PKG). Activation of PKG promotes the proteasomal degradation of β-catenin and activation of c-Jun NH₂-terminal kinase (JNK), leading to apoptosis (2). Interestingly, exisulind exhibits broad antitumor activity independent of cyclooxygenase-1, cyclooxygenase-2, p53, and bcl-2 inhibition (3). The proapoptotic effect of exisulind differs from the benzylamide analogs of sulindac, OSI-461 (formerly CP461) and OSIP486 821 (formerly CP248), by having no effect on microtubule polymerization (4, 5). In rodent models, exisulind demonstrated tumor growth inhibition of colon, bladder, breast, prostate, and lung cancers (6, 7, 8).

Based on the sulindac data in familial adenomatous polyposis (FAP), there was initial interest in developing exisulind as a chemopreventive agent. In one phase I trial, exisulind was administered twice daily for 6 months to 20 patients with FAP who had undergone subtotal colectomies (9). Dose-limiting toxicity (DLT) was observed at the 800 mg/d dose level and consisted of National Cancer Institute Common Toxicity Criteria (CTC) grade 3 reversible hepatic transaminitis. Side effects were common at all dose levels and included CTC grade 1 or 2 hepatic transaminase elevations, nausea, vomiting, dyspepsia, abdominal pain, diarrhea, and headache. The maximum tolerated dose (MTD) was determined to be 300 mg twice a day in patients with subtotal colectomies. There was no significant effect on polyp number or cellular proliferation, although there was a trend toward increased apoptosis in polyps removed from patients treated at the MTD. A randomized, double-blind, placebo-controlled 12-month trial in 73 patients with FAP demonstrated a 25% reduction in the formation of new polyps in the exisulind-treated group that was not statistically significant (10). However, during a 12-month extension, patients taking placebo who crossed over to exisulind and those who continued exisulind experienced statistically significant decreases in new polyp formation (11).

Exisulind was also studied alone as an anticancer agent. In one phase I study, exisulind was administered to 15 heavily pretreated patients with non–small-cell lung cancer [NSCLC (12)]. The MTD of exisulind was 250 mg twice daily (500 mg/d). Gastrointestinal discomfort and elevated liver function tests were dose-limiting. There were no objective responses; however, two patients maintained stable disease for 14 and 15 weeks, respectively. In a randomized, placebo-controlled trial in patients with prostate cancer, exisulind (500 mg/d) or placebo was administered to 96 men with increasing prostate-specific antigen (PSA) after radical prostatectomy (13). Exisulind significantly suppressed the rise in PSA overall and lengthened the interval to doubling of the PSA in high-risk patients. As with the prior studies, the most common toxicities included gastrointestinal symptoms (e.g., dyspepsia), increased transaminases, and nausea.

Based on the novel mechanism of action of exisulind, there was interest in conducting combination studies with cytotoxic agents that have complementary effects on tumor cells. Docetaxel (Taxotere; Aventis Pharmaceuticals Inc., Bridgewater, NJ) and paclitaxel, both semisynthetic taxanes, inhibit tumor cell growth through microtubule stabilization. Interfering with microtubules leads to activation of the JNK/stress-activated protein kinase signaling pathway, resulting in apoptosis (14). Thus, the hypothesis was that docetaxel, paclitaxel, and exisulind converge at JNK/stress-activated protein kinase activation and could combine to potentiate apoptosis. This was assessed in vitro when exisulind was combined with paclitaxel and demonstrated synergistic effects against six lung cancer cell lines regardless of drug resistance phenotype (15). These effects were then studied in vivo. In an orthotopic model of NSCLC (A549) in athymic nude rats, the efficacy of exisulind was studied alone and in combination with docetaxel (16). One week after tumor implantation, rats were given either (a) oral exisulind at 25 or 50 mg/kg/d, (b) docetaxel given intraperitoneally at 2.5 or 5 mg/kg weekly for 4 weeks, or (c) exisulind and docetaxel in four combinations of the doses listed above. After 80 days, the combination arm of exisulind (50 mg/kg/d) with docetaxel (5 mg/kg) resulted in a significant improvement of survival rate (60%) compared with control animals (P < 0.0004) or animals treated with either drug alone.

Based on these encouraging preclinical results, a phase I dose-escalation and pharmacokinetic study of exisulind and docetaxel was initiated in patients with advanced solid tumors using weekly intravenous docetaxel and daily oral exisulind. The main objectives of this study were as follows: (a) to characterize the toxicities of docetaxel and exisulind when administered concurrently over a 4-week cycle; (b) to determine the MTD and recommended dose for subsequent phase II trials; (c) to characterize the pharmacokinetic behavior of exisulind and docetaxel in combination in addition to describing any drug interactions; and (d) to seek preliminary evidence of antitumor activity of this combination in patients with advanced solid malignancies.

Patients with biopsy-confirmed solid malignancies refractory to standard therapy or for whom no effective therapy existed were eligible for this study. Other eligibility criteria included the following: (a) age ≥ 18 years; (b) Southwestern Oncology Group performance status of ≤1; (c) no prescription or over-the-counter NSAIDs for 2 weeks before enrollment (patients taking a cumulative monthly dose of aspirin of <3,250 mg for cardiovascular prevention were not excluded from the study); (d) life expectancy ≥ 3 months; (e) no chemotherapy or investigational agents within 4 weeks of study entry or within 6 weeks of study entry for nitrosoureas or mitomycin C; (f) adequate hematopoietic (absolute neutrophil count ≥ 1,500/μL, hemoglobin ≥ 9.0 g/dL, platelet count > 100,000/μL), hepatic [total bilirubin ≤ the institutional upper limit of normal (ULN); aspartate aminotransferase (AST) and alanine aminotransferase (ALT) ≤ 2.5× ULN and alkaline phosphatase ≤ ULN, or alkaline phosphatase ≤ 4× the ULN if transaminases were ≤ ULN, or transaminases ≤ 1.5× ULN and alkaline phosphatase ≤ 2.5× ULN)], and renal (serum creatinine concentration ≤ ULN) functions; (g) no palliative radiation therapy for the prior 2 weeks; and (h) resolution of all previous therapy-related toxicity. Patients with uncontrolled brain metastases (rapidly evolving neurologic symptoms or metastases that needed specific treatment before systemic chemotherapy) and those with significant medical conditions (e.g., uncontrolled hypertension, heart disease, or diabetes mellitus) were excluded. Female patients of child-bearing age were required to have a negative pregnancy test before study entry and were required to be on adequate birth control while on study. Informed consent was obtained according to federal and institutional guidelines.

Exisulind was started on day 3 of the first treatment cycle and continued without interruption throughout the remainder of the study period. The starting dose was 150 mg orally twice daily, with subsequent cohorts receiving 200 mg (cohort 2) and 250 mg twice daily (cohorts 3 and 4). The dose-escalation increments of exisulind were based on prior studies, with the MTD of 250 mg twice daily (BID) designated as the target dose for the combination, whereas 150 mg BID (60% of single-agent MTD) was thought to be a safe starting dose. Docetaxel was administered intravenously over 30 minutes on days 1, 8, and 15 of a 28-day treatment cycle. Patients received 8 mg of dexamethasone intravenously before docetaxel. The dose of docetaxel was 30 mg/m²/week, except for the fourth and final cohort, which received 36 mg/m²/week. Patients could receive docetaxel and exisulind concurrently for a maximum of six cycles. Patients who demonstrated disease stability in the absence of intolerable toxicity were allowed to remain on single-agent exisulind after six cycles of the combination. Toxicities were graded according to the National Cancer Institute CTC Version 2.0. DLT was defined as grade 4 neutropenia for >5 days, grade 4 neutropenia complicated by fever, grade 4 thrombocytopenia, or grade 3 nonhematologic toxicity (excluding alopecia or nausea and vomiting in the absence of optimal medical management) occurring in the first cycle. Patients were prohibited from taking NSAIDS (other than that described in Patient Selection) for the duration of the study.

The dose-escalation scheme was to proceed with three patients per cohort until one patient experienced DLT, in which case three additional patients were added to the cohort. If more than one of six patients in a given cohort experienced DLT, the MTD was exceeded, and the prior dose level was declared the MTD. A minimum of six patients were to be treated at the MTD.

If an elevated bilirubin or any grade 3 hepatotoxicity occurred, exisulind and docetaxel dosing was interrupted. When the patient had recovered to a grade 1 toxicity, exisulind could be restarted at the next lower dose level, and the docetaxel dose was reduced by 25%. The dose of docetaxel was also reduced by 25% for an elevated ALT to 1.6× the ULN, and by one dose level for any grade 3 or greater hematologic toxicity.

Exisulind was supplied by OSI Pharmaceuticals Inc., succesor-in-interest to Cell Pathways Inc. (Melville, NY) as 100- or 150-mg gelatin capsules. Docetaxel was purchased commercially from Aventis Pharmaceuticals Inc.

Complete medical histories, physical examinations, concurrent medication profiles, assessments of performance status, and routine laboratory studies were performed before study initiation and weekly during treatment. Routine laboratory studies included a complete blood count, differential white blood cell count, electrolytes, blood urea nitrogen, serum creatinine, glucose, total protein, albumin, calcium, phosphate, uric acid, lactate dehydrogenase, alkaline phosphatase, total bilirubin, ALT, and AST. A urinalysis and a serum pregnancy test (as appropriate) were performed before treatment. Pretreatment studies also included an electrocardiogram, relevant radiographic studies to evaluate all measurable and assessable sites of disease, and an assessment of all relevant tumor markers. Patient adherence was monitored by pill counts and dosing diaries. Radiographic evaluations for disease status assessment were repeated after every other course. A complete response was defined as the disappearance of all measurable and assessable disease for at least two measurements performed at least 4 weeks apart without worsening of disease-related symptomatology or declining performance status. A partial response required at least a 50% reduction in the sum of the product of the bidimensional measurements of all lesions documented by at least two measurements separated by a minimum of 4 weeks. Any increase in the size of a lesion by ≥25% or the appearance of a new lesion was considered disease progression. Patients continued on treatment in the absence of disease progression or intolerable toxicity.

To study the pharmacokinetics of docetaxel, whole blood samples were obtained from an indwelling venous catheter distant from the site of drug infusion. On days 1 and 8 of the first treatment course, samples were collected before the infusion; at the end of the infusion; at 5, 10, 20, and 30 minutes after the end of the infusion; and at 1, 2, 4, 8, 24, and 48 h after the end of the infusion. The 48 hour sample was drawn before the administration of exisulind on day 3. Exisulind concentrations were measured at baseline and then measured weekly, before dosing, on days 8, 15, and 22 of course 1 and on day 1 of course 2. The samples were collected in tubes containing EDTA, inverted several times, and immediately placed on ice. Within 15 minutes of blood collection, samples were centrifuged at room temperature to separate plasma and then frozen at −70°C until needed for analysis.

Plasma was assayed for docetaxel by high-performance liquid chromatography (HPLC) with ultraviolet detection and by liquid chromatography-tandem mass spectrometry (LC/MS/MS). The development of a LC/MS/MS assay for docetaxel was necessitated by the fact that later sampling points (8–24 hours) were often lower than the limit of quantitation for the HPLC/ultraviolet-based assay. The LC/MS/MS-based assay had a lower limit of quantitation of about 10-fold lower than any of the samples collected at the later time points, and thus full pharmacokinetic time courses could be evaluated. The initial docetaxel analysis was performed by HPLC with ultraviolet detection following solid-phase extraction from plasma as described previously (17). The LC/MS/MS assay developed and used in this study has been detailed in a separate report (18). In brief, quantitative analyses were performed with a PE Sciex API-3000 triple quadrupole mass spectrometer with a turbo ionspray source, interfaced with a PE Sciex 200 Autosampler and HPLC system (Applied Biosystems, Framingham, MA). The liquid chromatography system consisted of a Luna 5-cm C-18 column (2-mm internal diameter; Phenomenex, Torrance, CA), and the mobile phase was isocratic 90% acetonitrile in 10 mmol/L ammonium acetate. The flow rate was 200 μL/min, and the injection volume was 20 μL. The mass spectrometry settings used were as follows: turbo ion spray temperature of 300°C, needle voltage of 3,500 V, declustering potential of 25 V, focus plate of 175 V, collision energy of 20 V, needle position of 7, and collision gas (N₂) density set at 10. The instrument was operated in the selective reaction monitoring mode (positive ion), monitoring the ion transitions from m/z 808 to 226 for docetaxel and m/z 854 to 286 for paclitaxel (internal standard). The peaks coeluted at 1.6 minutes with a total analysis time of 4 minutes. Blank pooled plasma samples were spiked with docetaxel (LKT Laboratories, St. Paul, MN) and paclitaxel (internal standard; Sigma, St. Louis, MO) to yield final concentrations in plasma ranging from 0.25 to 1,000 nmol/L and extracted as described previously to generate a standard curve.

After the development of the LC/MS/MS assay, docetaxel was determined in samples using the method described above (18). The accuracy of the docetaxel assay was determined by preparation of standard plasma samples at 7.5, 25, and 50 nmol/L. The accuracy and precision % relative standard deviation (%RSD) observed were 94.4 ± 3.8% at 7.5 nmol/L, 97.3 ± 2.0% at 25 nmol/L, and 98.9 ± 2.0% at 50 nmol/L. Exisulind concentrations in plasma were determined by HPLC with detection at 329 nm using sulindac as an internal standard as described previously (19). The intra- and interday variability for the exisulind assay was 3.4% and 4.7%, respectively, as determined in spiked plasma samples.

Individual plasma concentration time-course measurements for docetaxel were analyzed by noncompartmental analysis using WinNonlin Version 3.0 (Pharsight Corp., Mountain View, CA). Area under the concentration-time curve (AUC) extrapolated to infinity (AUC_inf) was calculated using the linear trapezoidal method to the last desired point (24 hours) and then dividing the last measured concentration by the terminal rate constant (λ_z), which was determined from the slope of the linear, terminal phase of the curve. The terminal half-life (t_1/2λ) was calculated based on the terminal rate constant by 0.693/λ_z. The AUC_inf and t_1/2λ values are only reported for those patients for whom the docetaxel concentration was above the lower limit of quantitation for the assay at 24 hours, allowing for a full set of values for parameter calculations. The relationship between toxicity and exposure to exisulind was assessed using a χ² test, in which the sample size of 19 evaluable patients achieved 86% power to detect an effect size of 0.7 with 1 degree of freedom and a significance level of 0.05.

Twenty patients received 70 courses of treatment with exisulind and docetaxel. Patient characteristics are depicted in Table 1. All but one patient had received prior therapy, and the majority of patients had a performance status of 0 or 1. A wide variety of tumor types were represented among the 20 patients. Table 2 depicts the dose-escalation scheme. Ten courses of treatment did not include docetaxel because two patients had received six cycles of combination therapy. All patients were evaluable for toxicity, and there were no DLTs at any level. No patients required dose modification of exisulind secondary to toxicity, but two patients required dose reduction of docetaxel due to grade 3 neutropenia and grade 3 asthenia, respectively.

Seventeen of 70 courses (24%) were accompanied by mild to moderate AST elevations (Table 3). Mild to moderate elevations of ALT or alkaline phosphatase were evident in 15 of 70 courses (21%) each. Only one of the patients who developed elevation of alkaline phosphatase had metastatic disease to the bone. All but one of the patients with AST or ALT elevations was receiving 500 mg/d exisulind. Serum bilirubin was normal in all but four courses, which included three grade 1 elevations and one grade 2 elevation.

Other gastrointestinal toxicities included anorexia, nausea, vomiting, dyspepsia, diarrhea, and epigastic pain. The majority of patients experienced some degree of nausea, vomiting, and anorexia. Overall, 24% of courses were accompanied by mild to moderate nausea. Two patients experienced three courses of grade 3 nausea. Vomiting occurred during 15 of 70 courses (21%), with two patients experiencing two courses (3%) of grade 3 vomiting. No patients required hospitalization for symptom control, and most were adequately managed with standard antiemetic therapy. Anorexia was generally mild (grade 1–2), occurring in 14 of the 70 courses (20%), although one patient did experience grade 3 toxicity that improved without dose modification. Diarrhea occurred in 14 of 70 courses (20%), with grade 2 diarrhea occurring in four patients for a total of 4 courses. Dyspepsia was evident in 13 of 70 courses (19%), and epigastric pain was encountered in 10 of 70 courses (25%). Symptoms were generally relieved with proton pump inhibitor administration. One patient developed biliary pancreatitis during the second course of treatment and required hospitalization. This patient exhibited the highest AUC and lowest clearance of docetaxel and was taken off study for grade 3 toxicity. Two patients received exisulind for 10 and 12 cycles. One of the patients developed peptic ulcer disease after receiving 10 cycles, whereas the other patient did not develop any cumulative toxicity. None of the grade 3 gastrointestinal toxicities occurred in the first cycle and hence were not considered dose-limiting.

Grade 1 and 2 anemia was the most common hematologic toxicity that resulted from the combination. Fourty-four of 70 (63%) courses in the study were accompanied by mild to moderate anemia. The anemia appeared slightly more severe in the group treated with the highest dose of docetaxel (36 mg/m²). Three patients experienced four courses (6%) of neutropenia. One of the three patients required dose modification of the docetaxel secondary to grade 2 and 3 neutropenia. All of the patients with documented neutropenia initially received the highest dose of docetaxel (36 mg/m²/week). Only one patient experienced two courses of grade 1 thrombocytopenia, which did not require dose modification.

Fatigue was the other principal (occurring in >20% of courses) toxicity of exisulind and docetaxel in this study. Fatigue was reported in 34 of 70 courses (49%) and was not dose-related. Three patients experienced grade 3 fatigue, which generally corresponded to a decline in Eastern Cooperative Oncology Group performance status and disease progression. One patient experienced a grade 4 anaphylactic reaction several minutes after receiving the docetaxel infusion, which responded well to steroids and antihistamines. Although there was an apparent temporal correlation with the docetaxel infusion, it was noted that this patient also had the highest plasma exisulind concentration at the time of the docetaxel infusion.

No partial or complete responses were observed during the study period. Seven patients maintained stable disease for 3 to 10 cycles of therapy. One patient with prostate cancer and another with carcinoid remained on study with stable disease for 12 and 10 cycles, respectively.

Pharmacokinetic samples for docetaxel were obtained from 19 patients on day 1 and 18 patients on day 8. Docetaxel pharmacokinetic parameters were analyzed using noncompartmental methods (Table 4). Mean plasma concentration-time profiles on days 1 and 8 after administration of docetaxel doses of 30 and 36 mg/m² as a 30-minute intravenous infusion are shown in Fig. 1. The pharmacokinetic profile of docetaxel was compared on days 1 and 8 for patients receiving 30 and 36 mg/m² doses of docetaxel (Figs. 1 and 2). The AUC_inf and t_1/2λ values are displayed in Figs. 2 and 3, respectively, and the mean values were not statistically different between days 1 and 8 using Student’s t test.

The pharmacokinetic samples for exisulind were obtained from 18 patients on day 8, 17 patients on day 15, 15 patients on day 22, and 17 patients on day 1 of cycle 2. Mean steady-state concentrations of exisulind at all dose levels are illustrated in Fig. 4. There was no relationship between the plasma concentrations of exisulind and the administered dose. Inter- and intrapatient variability was substantial. Interpatient variability among the exisulind trough concentrations at the 150-, 200-, and 250-mg dose levels was 38.5%, 31.5%, and 63.8%, respectively, whereas the intrapatient variability was 36.7 ± 20.0% across all dose levels. There did not appear to be any impact of either dose of docetaxel on the plasma concentrations of exisulind (data not shown).

Interestingly, there was a significant correlation between the gastrointestinal toxicities observed and exisulind plasma concentrations (P = 0.0108; Tables 5 and 6). Independent of the exisulind dose, patients who demonstrated higher plasma concentrations of exisulind experienced more severe gastrointestinal toxicities such as epigastric pain, dyspepsia, nausea, vomiting, and an increase in alkaline phosphatase from baseline. Among the nine patients whose mean concentrations of exisulind exceeded 15 μmol/L, seven experienced grade 2 or greater toxicities related to the gastrointestinal tract, one had a grade 3 anaphylactic reaction, and one experienced grade 2 tinnitus. Of 10 patients with mean plasma concentrations of <15 μmol/L, 3 patients had grade 2 or greater gastrointestinal toxicities. There was no relationship between the increased plasma concentrations of exisulind and pretreatment renal function, pretreatment hepatic function, or administered dose of exisulind.

Exisulind is a selective proapoptotic agent that inhibits cGMP-phosphodiesterases 2 and 5, resulting in a sustained increase in cGMP and activation of PKG (1). Apoptosis by this agent is preceded by the proteasomal degradation of β-catenin and activation of JNK (2). Preclinical studies demonstrated synergistic effects when exisulind was combined with docetaxel or paclitaxel against human lung and prostate cancer cell lines in vitro (15). One hypothesis for the observed synergy was that both agents converge at the activation of JNK to promote apoptosis (2, 14). In an orthotopic model of NSCLC in athymic nude rats, an increase in apoptosis and survival was demonstrated when exisulind was combined with docetaxel, compared with either drug given alone (16). This phase I and pharmacological study was based on these promising in vivo results and designed to evaluate the feasibility and safety of the administration of twice-daily exisulind and weekly docetaxel in patients with advanced solid tumors.

The results of this study demonstrate that the combination of exisulind and docetaxel is tolerable, and no DLTs occurred during the first cycle. The principal toxicities noted were gastrointestinal in nature and included mild to moderate nausea, vomiting, abdominal pain, diarrhea, anorexia, dyspepsia, and elevated liver function enzymes. Neutropenia was encountered in four cycles in three patients treated with the highest dose of docetaxel (36 mg/m²), requiring dose reduction in one patient. Therefore, the cohort combining a lower dose of docetaxel (30 mg/m²) was also expanded and evaluated. At the third dose level (30 mg/m² docetaxel and 500 mg/d exisulind), no patient experienced neutropenia. Anemia was present in the majority of patients but was only mild to moderate in severity. Based on these results, the recommended phase II dose for the combination is 250 mg BID of exisulind and 36 mg/m² docetaxel weekly for 3 weeks every 4 weeks, with reduction of docetaxel to 30 mg/m²/week, if required for myelosuppression.

Gastrointestinal toxicities have been reported in previous studies with exisulind. In studies in patients with FAP, prostate cancer, or lung cancer, the predominant toxicities were elevation of liver function enzymes, nausea, and vomiting (9, 12, 13). In this trial, they were common at all dose levels but were mostly of mild to moderate severity. Grade 1 and 2 elevation of ALT, AST, and alkaline phosphatase accompanied the higher doses of exisulind, whereas nausea, vomiting, dyspepsia, and abdominal pain were present at all dose levels. Grade 3 gastrointestinal toxicities occurred in 5 of the 19 patients at some point during therapy. One patient developed pancreatitis during the second course of treatment, and an abdominal sonogram demonstrated borderline dilation of the extrahepatic bile ducts without evidence of cholelithiasis. Previous studies have reported a choleretic effect of sulindac and exisulind (20, 21), whereas only 2 of 45 patients with prostate cancer treated with exisulind developed a biliary event (13). Exisulind has been evaluated in combination with several chemotherapeutic agents, including capecitabine and docetaxel on an every 3-week schedule (22, 23). In combination with capecitabine, the 125-mg BID (250 mg/d) dose of exisulind was recommended for phase II testing because only 7 of 13 patients could tolerate the higher dose of 250 mg BID of exisulind. Interestingly, in this study, the DLTs were characteristic of capecitabine and did not appear to be more prominent with the combination (22). However, at the higher dose of exisulind, the rate of grade 2/3 nausea or vomiting was 21%, whereas in the present study, it was only 11% at the 250 mg BID dose levels of exisulind. These results imply that there may be somewhat synergistic effects between exisulind and other oral agents, such as capecitabine, that are associated with similar gastrointestinal toxicities.

The mean plasma concentrations of exisulind at the highest dose level ranged from 5 to 47 μmol/L and were not affected by docetaxel dosing. These concentrations are substantially lower than the range of IC₅₀ values that have been reported in previous in vitro studies of exisulind against human tumor cell lines, which ranged from 100 to 300 μmol/L (15, 24, 25). Exisulind plasma concentrations displayed substantial inter- and intrapatient variability. There was no relationship between the plasma concentration of exisulind and the administered dose or any other pretreatment condition including renal or hepatic function, weight, height, age, or gender. Interestingly, the majority of exisulind undergoes biliary excretion with no evidence of metabolism by the cytochrome P450 isoenzyme system, which can be a major source of interpatient variability (19, 26). Thus, the etiology of this variability may be related to hepatobiliary transporters that have also been implicated in the gastrointestinal toxic effects of a range of agents including NSAIDS, irinotecan, and methotrexate (27, 28). By contrast, the mean C_max concentrations of docetaxel achieved at the 30 and 36 mg/m² dose levels ranged from 67 to 1,278 nmol/L, which was significantly higher than the IC₅₀ (2.5 nmol/L) reported in the preclinical studies of docetaxel and exisulind against the A549 NSCLC cell line (16). A more sensitive LC/MS/MS method was developed to assess plasma concentrations of docetaxel on a weekly schedule, and no apparent impact of exisulind was noted between days 1 and 8 on the pharmacokinetics of docetaxel, when exisulind would have achieved steady-state concentrations.

Surprisingly, a significant relationship (P = 0.0016) was established between the mean plasma concentrations of exisulind and the occurrence of grade 2 or greater toxicities, which were predominantly gastrointestinal in nature. Although at first glance, this may appear to be a statistical anomaly, these effects, in conjunction with the interpatient variability observed, may relate to the biliary excretion of exisulind and the hepatobiliary transport system. Clearly, future studies of exisulind and structurally related analogs should incorporate studies to assess patients for single-nucleotide polymorphisms that may impact on the function of hepatobiliary transport proteins.

In this phase I study, no objective responses were observed with the combination of exisulind and docetaxel, although a third of patients maintained stable disease for 3 to 12 cycles and two patients (one with prostate cancer and one with carcinoid) remained on study for 10 and 12 cycles. Several factors could be responsible for this, including the inability to achieve exisulind plasma concentrations of 100 to 300 μmol/L, the extensive prior therapy received by the patients, and the variability of tumor types studied. However, prior studies of the combination of exisulind and docetaxel/carboplatin or capecitabine have yielded response rates of 45% and 16% in patients with untreated metastatic NSCLC and anthracycline- and taxane-refractory metastatic breast cancer, respectively (22, 29). These results suggest that combinations of exisulind and chemotherapeutic agents result in objective responses that are at least consistent with that expected with chemotherapy alone. Any additional benefit of exisulind may only be detected in randomized studies in these diseases. However, the fact that single-agent exisulind inhibited the increase in PSA overall and prolonged PSA doubling time in men after prostatectomy, compared with placebo, indicates that at least in prostate cancer, the agent has independent anticancer effects (13).

In summary, this phase I study demonstrated that exisulind and docetaxel may be safely administered in combination at full single-agent doses of each drug. Docetaxel (36 mg/m², weekly) and exisulind (500 mg/d) were reasonably well tolerated and are recommended for appropriate disease-directed studies. Future studies of exisulind and other analogs, such as OSI-461, should incorporate pharmacogenomic studies to assess the impact of genetic polymorphisms on the toxicity and pharmacology of these agents. Based on this phase I study and the preclinical data demonstrating the additive effects of exisulind and docetaxel on A549 NSCLC cells in vivo, a phase II study using this combination was initiated in patients with NSCLC.