|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Troxacitabine, an L-Stereoisomeric Nucleoside Analog, on a Five-Times-Daily Schedule: A Phase I and Pharmacokinetic Study in Patients With Advanced Solid MalignanciesByFrom the Institute for Drug Development, Cancer Therapy and Research Center, University of Texas Health Science Center at San Antonio; and Brooke Army Medical Center, San Antonio, TX; and Biochem Pharma Inc, Laval, Canada. Address reprint request to Eric K. Rowinsky, MD, Institute for Drug Development, Cancer Therapy and Research Center, 8122 Datapoint Dr, Ste 700, San Antonio, TX, 78229; email: erowinsk{at}saci.org
PURPOSE: To assess the feasibility of administering troxacitabine, a unique L-nucleoside that is not a substrate for deoxycytidine deaminase–mediated catabolism, as a 30-minute intravenous (IV) infusion daily for 5 days. PATIENTS AND METHODS: Patients with advanced solid malignancies were treated with escalating doses of troxacitabine as a 30-minute IV infusion daily for 5 days. Plasma and urine sampling was performed to characterize the pharmacokinetics and pharmacodynamics of troxacitabine.
RESULTS: Thirty-nine patients received 124 courses of troxacitabine at eight dose levels ranging from 0.12 to 1.8 mg/m2/d. Severe neutropenia that was protracted (> 5 days) and/or associated with fever, and skin rashes were consistently experienced by heavily (HP) and minimally pretreated (MP) patients at doses exceeding 1.2 and 1.5 mg/m2/d, respectively. At troxacitabine doses CONCLUSION: Recommended doses for phase II studies of troxacitabine as a 30-minute infusion daily for 5 days every 4 weeks are 1.5 and 1.2 mg/m2/d for MP and HP patients, respectively. Broad disease-directed evaluations of troxacitabine on this schedule and possibly less frequent schedules are warranted.
NATURALLY OCCURRING nucleosides and nucleoside analogs developed to date as anticancer therapeutics, such as cytarabine and gemcitabine, are in the beta-D stereochemical configuration.1 Until recently, their corresponding unnatural L-enantiomers have largely been considered to be unrecognizable by cellular enzymes and, therefore, biologically inactive.2 However, the discovery that the L-(-)-SDDC (2'-deoxy-3'-thiacytidine; 3TC) stereoisomer was more potent than its corresponding D-enantiomer against human immunodeficiency and hepatitis B viruses led to the identification of (-)OddC or troxacitabine ([-]-2'deoxy-3'-oxacytidine) (Troxatyl; BioChem Pharma, Inc, Laval, Canada) (Fig 1), a dioxolane that has unique mechanistic and pharmacologic properties and impressive anticancer activity in preclinical studies.1-5
Troxacitabine is phosphorylated intracellularly by deoxycytidine kinase to its monophosphate, diphosphate, and triphosphate forms, which are then incorporated into DNA but not RNA.1 Early work suggested that troxacitabine was transported into cells by both equilibrative-sensitive and equilibrative-insensitive nucleoside transport systems. However, recent detailed experiments have revealed that troxacitabine permeation is largely non–carrier-mediated.6 Troxacitabine does not induce apoptosis but, instead, induces degradation of DNA into large fragments, ultimately leading to cell death.7 Incorporation of troxacitabine triphosphate into DNA results in complete DNA chain termination caused by the absence of a hydroxyl group in the dioxolane ring of the L-nucleoside, which is required for chain elongation.7 Although the incorporation of troxacitabine into the terminal ends of DNA and the inhibition of chain elongation are the principal determinants of cytotoxicity, the potent inhibitory effects of troxacitabine triphosphate on DNA polymerases required for DNA replication and repair may also contribute to its antitumor activity.8 However, unlike cytarabine and gemcitabine, the L configuration confers on troxacitabine a resistance to degradation by deoxycytidine deaminase.1 In contrast to cytarabine and gemcitabine, the intracellular elimination of troxacitabine triphosphate is slow, with biphasic elimination and intracellular half-lives of 3.5 hours and more than 20 hours for each respective phase.7 The relatively high intracellular retention and low systemic clearance suggest that troxacitabine-induced cytotoxicity may be less dependent on dose and schedule than other nucleoside analogs.7 Troxacitabine has demonstrated broad activity against both solid and hematopoietic malignancies in vitro and in vivo.1,4-11 Of particular note, impressive activity has been observed against human cancer cell lines and xenografts of hepatocellular, prostate, and renal origin, as well as cancers with multidrug resistance conferred by overexpression of P-glycoprotein.1,5,9,11 Notable activity against the highly resistant human pancreatic cancer xenografts Panc-01 and MiaPaCa has also been observed, and troxacitabine is more efficacious than gemcitabine.10 In the human tumor colony-forming assay, troxacitabine inhibited the growth of a broad range of fresh human tumors, with treatment duration seemingly the most important determinant of cytotoxicity.4,5 In addition to troxacitabine’s activity against Panc-01 and MiaPaCa pancreatic cancers, prominent inhibitory activity against a variety of human tumor xenografts has also been observed, with tumor regression noted in several well established models.1,5,8-11 Notable activity has been observed against PC-3 and DU-145 prostate; CAKI-1, A498, RXF-393 and SN-12C renal; HT-29 colon; KBV head and neck; HepG2 liver; and NCI-H460 and NCI-H322M non–small-cell lung cancer xenografts.1,5,8-11 Troxacitabine has consistently demonstrated superior activity on frequent divided schedules (eg, daily x 5) compared with single and less frequent schedules (eg, weekly).4,5,8 Toxicology studies in rats and monkeys demonstrated that troxacitabine principally affects rapidly proliferating tissues, particularly the bone marrow, skin, and gonads.5 There were considerable interspecies differences in susceptibility to toxicity, with monkeys being the most sensitive. A frequent dosing schedule also resulted in greater toxicity than the single-dose schedule at any dose level.5 As a result of this interspecies variability, the comparative effects of troxacitabine on the growth of hematopoietic progenitor cells from humans and monkeys were evaluated. Human progenitor cells were five-fold more sensitive to treatment for 1 hour, but this difference was less pronounced for continuous exposure. Pharmacokinetics were not dose-dependent in either rats or monkeys, although clearance was lower in monkeys.5 In rats, renal clearance accounted for the majority of drug disposition, with 90% of the dose recovered from urine as unchanged troxacitabine within 24 hours after treatment.5 The unique mechanistic and pharmacologic properties, high potency, and broad activity of troxacitabine in preclinical studies, which seem to be related in part to its novel structural and stereochemical features, provided the impetus to develop this agent. Although the antitumor activity of troxacitabine has generally been superior on frequent or divided schedules, the relatively high intracellular retention of its active phosphorylated species, the fact that it is a poor substrate for deoxycytidine deaminase, and its low systemic clearance suggest that efficacy may occur with treatment regimens of infrequent high doses and daily repeated low doses. The principal objectives of this study were to (1) characterize the principal toxicities of troxacitabine administered as a 30-minute intravenous (IV) infusion daily for 5 days every 3 weeks in patients with advanced solid malignancies, (2) determine the maximum-tolerated dose (MTD) and recommend a safe starting dose on this schedule for phase II studies, (3) characterize the pharmacokinetic behavior of troxacitabine, and (4) seek preliminary evidence of antitumor activity.
Eligibility Patients with pathologically documented advanced solid tumors that were refractory to conventional therapy or for whom no effective therapy existed were candidates for this study. Eligibility criteria also included: age  18 years; Eastern Cooperative Oncology Group (ECOG) performance status  2 (ambulatory and capable of all self-care); a life expectancy of at least 12 weeks; no chemotherapy, wide-field radiotherapy, or other experimental therapy within 4 weeks of treatment (6 weeks for nitrosoureas and Mitomycin C); adequate hematopoietic (absolute neutrophil count [ANC] 1,500/µL, platelet count 100,000/µL, hemoglobin 9.0 g/dL), hepatic (total bilirubin level 1.5 times the upper normal limit; AST and ALT three times the upper normal limit), and renal (creatinine 1.5 mg/dL) functions; no coexisting medical condition of sufficient severity to limit full compliance with the study; no evidence of brain metastases; and no active serious infection. All patients gave written informed consent according to federal and institutional guidelines before the commencement of treatment.
Drug Administration and Dosage
Because the first dose level of 0.12 mg/m2/d was anticipated to be less than doses likely to produce relevant biologic activity, albeit selected because of interspecies differences in toxicology studies, a minimum of three patients were to be entered at each dose level, except for the starting dose level, in which a single patient was to be treated. Dose escalation proceeded in increments of 100% in cohorts of new patients until grade 1 dermatologic or mucosal (including diarrhea) or other grade
The MTD was to be defined separately for minimally pretreated (MP) and heavily pretreated (HP) patients if it seemed that HP patients were more susceptible to DLT. HP patients were defined a priori as those who had been previously treated with
Pretreatment and Follow-Up Studies
Plasma Sampling and Assay
Troxacitabine concentrations in plasma and urine were quantified by a validated assay using high-performance liquid chromatography in tandem with mass spectrometric detection. This system consisted of a 1090 series II liquid chromatograph (Hewlett Packard, Palo Alto, CA) coupled with an API 300 MS/MS Detector (PE Sciex, Foster City, CA). BCH-189, a nucleoside analog that is structurally similar to troxacitabine, was used as an internal standard. Plasma and urine samples were spiked with BCH-189 and troxacitabine extracted from each matrix by vortexing with 25 mmol/L ammonium formate buffer at pH 3.5. The samples were loaded on preconditioned PRS cartridges (200 mg/3 mL; Varian, Walnut Creek, CA) and centrifuged at 800 rpm at 20°C for 2 minutes. The cartridges were then washed with water and centrifuged at 800 rpm at 20°C for 2 minutes, after which the samples were eluted with 1% ammonium hydroxide in methanol and again centrifuged at 600 rpm for 5 minutes. The organic solvent was evaporated to dryness at 35°C, and the residue was reconstituted with acetonitrile/25 mmol/L ammonium acetate (v/v, 80/20). Five-µL aliquots of these reconstituted samples were then injected onto the liquid chromatography system which consisted of an Amino (YMC, Inc, Wilmington, NC) 5.0 cm x 4.0 cm, 3 µm analytic column, a mobile phase of acetonitrile/25 mmol/L ammonium acetate (v/v, 90/10), and a flow rate of 1 mL/min. Analysis was performed in the positive ion mode, and ionization was induced using a heated nebulizer inlet operating at 480°C, with nebulizer gas pressure of 80 psi. Multiple reaction monitoring of the transitions 214.0
Pharmacokinetic Analyses Pharmacokinetic parameters were depicted using descriptive statistics. Linear least-squares regression was used to assess the relationships between (1) troxacitabine dose and exposure (Cmax and AUC), (2) troxacitabine pharmacokinetic parameters (Cls and Vss) and body-surface area (BSA), and (3) troxacitabine Cls and creatinine clearance. Creatinine clearance was estimated according to the method of Cockcroft and Gault.13 Nonparametric methods (Wilcoxon Rank Sums and Kruskal-Wallis Tests) were used to examine the relationship between (1) troxacitabine dose level and pharmacokinetic parameters and (2) troxacitabine exposure and NCI-CTC–graded toxicity. In addition, the relationships between troxacitabine exposure and percentage of decrease in ANC were explored. The percentage of decrease in ANC was calculated as follows: 100% x (pretreatment ANC value-nadir ANC value)/pretreatment ANC value. These relationships were described using the maximum effect (Emax) model of drug effect, ie, percentage decrease in ANC = (Emax x AUC)/(AUC50 + AUC), in which Emax was fixed at 100% AUC and AUC50 was the AUC at which the effect was 50% of maximum. The Emax model was fitted to the data using nonlinear least-squares regression. The a priori level of significance was set at 0.05. Statistical analyses were performed using the JMP, version 3.1 statistical software program (SAS Institute, Cary, NC).
General Thirty-nine patients, the pertinent characteristics of whom are listed in Table 1, received 124 courses of troxacitabine through eight dose levels that ranged from 0.12 to 1.8 mg/m2/d. All courses were fully assessable for toxicity. The total numbers of new patients treated at each dose level, number of assessable courses, and dose escalation scheme are depicted in Table 2. The median number of courses administered per patient was two (range, one to 10 courses). Seven patients required dose reductions because of toxicity.
After no or negligible drug-related effects were noted in first courses, the dose of troxacitabine was successively doubled from 0.12 to 0.24 mg/m2/d and then to 0.48 mg/m2/d. Because one patient treated with the first course of troxacitabine at the 0.48 mg/m2/d dose level developed a skin rash that was generalized and tolerable, albeit classified as grade 3 according to the NCI-CTC criteria, the dose was escalated by 50% to 0.72 mg/m2/d. Although clinically-relevant toxicity was not noted in the first course of troxacitabine in patients treated at the 0.72 mg/m2/d dose level, severe hematologic effects were sporadically noted in two patients after successive dosing (grade 4 thrombocytopenia during course 3 and grade 3 neutropenia/grade 4 thrombocytopenia during course 4). Thereafter, dose escalation increments did not exceed 25%. There were several episodes of uncomplicated grade 3 and 4 neutropenia at doses of 0.96 and 1.2 mg/m2/d, but DLT was not observed. However, DLT was consistently noted at the 1.5-mg/m2/d dose level in HP patients only, leading to a divergence of the dose escalation process into two distinct schemes for MP and HP patients. At 1.5 mg/m2/d, three of seven HP patients experienced DLT in the first course. Five of the seven HP patients experienced grade 4 neutropenia, including one patient each with prolonged (> 5 days) grade 4 neutropenia and grade 4 neutropenia with sepsis. Treatment delay was also required in seven (33%) of 21 courses. In contrast, this dose level was well tolerated by MP patients, but treatment delay was also common. Thereafter, the treatment interval was modified from 3 to 4 weeks for both HP and MP patients, and dose escalation proceeded to 1.8 mg/m2/d every 4 weeks in MP patients, whereas additional HP patients were treated with troxacitabine 1.2 mg/m2/d every 4 weeks. The MTD for HP patients was determined to be 1.2 mg/m2/d because none of six HP subjects experienced DLT in the first course, and only one of 23 total courses was associated with dose-limiting events. At 1.8 mg/m2/d, MP patients consistently encountered dose-limiting hematologic and dermatologic events, with four of six patients experiencing DLT in the first course. In contrast, the 1.5-mg/m2/d dose level was considered the MTD for MP patients as DLT occurred in one of six patients in the first course and in two of 26 total courses.
Toxicity
Prolonged (> 5 days) grade 4 neutropenia was experienced by two patients in three courses (1.5 mg/m2/d and 1.8 mg/m2/d, one patient each, both in course 1). One HP and one MP patient each developed fever associated with grade 3 to 4 neutropenia after treatment with troxacitabine doses of 1.5 and 1.8 mg/m2/d dose levels, respectively. At the 1.5 mg/m2/d dose level, five of seven HP patients experienced grade 4 neutropenia, and three subjects experienced grade 3 neutropenia. Two of these patients experienced hematologic dose-limiting events (grade 4 neutropenia with fever/sepsis and grade 4 neutropenia lasting longer than 5 days) during the first course of troxacitabine. The next lower dose level, 1.2 mg/m2/d, was considered the MTD for HP patients because DLT did not occur in the first course in any of the six HP subjects, and only one of 23 total courses was associated with dose-limiting hematologic toxicity. At the 1.8 mg/m2/d dose level, one of six MP patients experienced a dose-limiting hematologic event (grade 4 neutropenia that lasted longer than 5 days) in the first course; three other MP patients experienced grade 3 skin rash (see Dermatologic Toxicity). Overall, the 1.5 mg/m2/d dose level was well tolerated by MP patients and considered the MTD; dose-limiting hematologic toxicity, consisting of brief grade 4 neutropenia with failure of neutrophil recovery to 1,500/µL until 8 weeks after treatment, occurred in one of six MP patients in the first course and two of 26 total courses was associated with dose-limiting myelosuppression. Severe thrombocytopenia and anemia were less common than severe neutropenia and generally noted concomitantly with severe neutropenia. These effects were also more severe in HP patients. Overall, seven (6%) of 124 courses were delayed because of thrombocytopenia of any grade that did not completely recede on the day of planned treatment, with severe (grade 4) thrombocytopenia noted in four courses (2%). Two MP patients, both treated with troxacitabine, 0.72 mg/m2/d every 3 weeks, developed grade 4 thrombocytopenia during courses 3 and 4. An MP patient in course 5 at the 1.8-mg/m2/d dose level experienced grade 4 thrombocytopenia. Despite dose reduction to 1.5 mg/m2/d, grade 4 thrombocytopenia recurred in course 7. This patient was subsequently treated with three additional courses at the 1.2-mg/m2/d dose level, which were well tolerated. Drug-related anemia was generally mild (grade 1) or moderate (grade 2). Severe (grade 3) anemia, possibly related to drug and requiring RBC transfusions, was noted in eight (7%) of 124 courses. Dermatologic. Troxacitabine treatment produced three types of cutaneous effects: skin rash, palmar-plantar erythrodysesthesia, and asymptomatic hyperpigmentation. The distributions of skin rash and hand-foot syndrome according to NCI-CTC toxicity grade and as a function of dose level are shown in Table 5. Skin rash was the most common cutaneous effect and nonhematologic toxicity of troxacitabine. Overall, 21 (54%) of 39 patients developed a rash during treatment, with 18 patients (46%) initially developing the rash during the first course of troxacitabine. Both the incidence and severity of the skin rash seemed dose-related. The rash was typified by an erythematous maculopapular eruption that was localized and patchy in the majority of patients (grade 1 to 2, 19 courses). However, four patients developed a generalized and symptomatic rash that was classified as grade 3 during four courses (0.48 mg/m2/d, one course; and 1.8 mg/m2/d, three courses). The rash had bullous features in one patient. The trunk, neck, and limbs were most commonly affected. Sun-exposed areas did not seem to be more prone to effects. The onset of the rash was usually on days 5 to 8. Manifestations were maximal on days 8 to 15, and the rash was self-limiting, generally resolving completely by day 22. Fourteen patients experienced pruritus, which was usually managed successfully with antihistamine medications. On several occasions, a brief course of anti-H1 histamine therapy and corticosteroids was administered. This generally resulted in rapid symptomatic improvement and was also efficacious in patients with grade 3 rashes, with rash resolution within 7 days of commencing antihistamine therapy.
Troxacitabine also induced cutaneous effects that involved the palms of the hands and plantar aspects of the feet, which were distinct from the aforementioned maculopapular rash. This second type of dermatologic toxicity was experienced by four patients, at troxacitabine doses at or above 1.2 mg/m2/d. The toxicity resembled palmar-plantar erythrodysesthesia (hand-foot syndrome), characterized by discomfort, erythema, dryness, desquamation, and pruritus of the palms and soles. In two MP patients treated at the 1.5-mg/m2/d and 1.8-mg/m2/d dose levels, the rash was painful, tender, and blistering; disrupted function; and scored as grade 3. In all four affected individuals, hand-foot syndrome developed after either two (two patients) or at least four (two patients) courses. These manifestations were usually maximal 8 to 15 days after treatment, with subsequent improvement, albeit incomplete, by the time of the next course. However, this toxicity progressively worsened with each successive course. Six additional patients treated with a wide range of troxacitabine doses developed hyperpigmentation that increased gradually with each successive course. Hyperpigmented regions were generally on the face and appendages; however, one subject developed progressive hyperpigmentation of the tongue and buccal mucosa with cumulative therapy. Miscellaneous. Fourteen patients developed nausea at some time during their treatment; eleven, one, and two patients developed grades 1, 2, and 3 nausea, respectively. Vomiting was reported in five individuals at some time during treatment (grade 1, four patients; grade 3, one patient). Nausea and vomiting were successfully managed with simple antiemetic regimens; routine treatment with antiemetic premedication was not undertaken. Eleven and two patients reported mild to moderate malaise (grade 1 to 2) and mild mucositis (grade 1), respectively. Malaise was noted across the entire troxacitabine dosing range, although definite temporal relationships could not be discerned, indicating that the underlying malignant process may have been a contributing factor.
Antitumor Activity
Pharmacokinetics and Pharmacodynamics
Representative plasma concentration–time profiles are shown in Fig 2, and mean troxacitabine pharmacokinetic parameters derived using noncompartmental methods are listed in Table 6. An inspection of the scatterplots of dose versus both Cmax and AUC0-
Troxacitabine Cls correlated significantly with BSA on day 1 (R2 = 0.4688, P < .0001) and day 5 (R2 = 0.2491, P = .0012). In addition, Cls values did not differ between male and female patients (day 1, P = .4652; day 5, P = .7808), although only eight (21%) of the 31 total patients were female. Troxacitabine Vss values were not related to BSA (R2 = 0.443, P = .1985). Renal excretion of unmetabolized drug was the principal mode of troxacitabine elimination. Urinary excretion of unchanged troxacitabine accounted, on average, for 50% and 61% of the administered troxacitabine dose during the first 24 hours after treatment on days 1 and 5, respectively (Fig 4). Urinary excretion was maximal (mean, 43.4%) in the first 12 hours after treatment. Only 16% of the troxacitabine dose administered on day 5, on average, was excreted in the urine as unchanged troxacitabine between 24 to 48 hours. As shown in Fig 5, troxacitabine Cls related moderately well to renal function, as assessed by creatinine clearance (day 1, R2 = 0.318, P = .0002; day 5, R2 = 0.3017, P = .0003).
Pharmacodynamics. The relationship between troxacitabine dose and effects on ANC counts was close in the dosing range evaluated. All pharmacokinetic parameters that reflected troxacitabine exposure seemed to be equally predictive of the percentage of decrements in the ANC counts. Scatterplots that depict the relationships between troxacitabine AUC values on days 1 and 5 and the percentage of decrements in ANC counts in course 1 are depicted in Fig 6. The relationships were well described by sigmoidal Emax models, as shown in Fig 6, particularly after the values of three apparent outliers (one each at the 0.72-mg/m2/d, 1.2-mg/m2/d, and 1.5-mg/m2/d dose levels) were excluded from the analyses; R2 values increased from 0.33 to 0.56.
Relationships between the other principal adverse effect of troxacitabine, dermatologic toxicity, and indices reflecting drug exposure (Cmax and AUC0-  ) were also sought. The distributions of NCIC-CTC grades of dermatologic toxicity experienced during course 1 at the three highest dose levels as functions of troxacitabine Cmax or AUC values on days 1 and 5 are shown in Fig 7. Although the propensity for dermatologic toxicity was directly related to troxacitabine dose across all dosing levels, relationships between the severity of skin toxicity and any parameter of drug exposure were not evident (Fig 8). This was irrespective of the comparisons performed (ie, grades 0 v grades 1 to 3; grades 0 to 1 v grades 2 to 3), however, the small numbers of patients with any categorical grade of toxicity limits the statistical power of such analyses.
The importance of stereochemistry in the action, metabolism, pharmacokinetics, and pharmacodynamics of new agents has been well recognized in many medical disciplines.14 However, relatively little attention has been focused on exploiting the potential of chiral specificity in the development of anticancer agents. The recent discovery that the L-enantiomers of several dideoxycytidine analogs (eg, 3TC) were potent antiviral agents led to the discovery and development of the L-(-) dioxolane-cytidine nucleoside troxacitabine.2,3 In preclinical studies, troxacitabine demonstrated impressive cytotoxic potency and activity against a broad spectrum of neoplasms, which led to its selection for clinical development.1,4-11 Although troxacitabine, cytarabine, and gemcitabine are all phosphorylated by deoxycytidine kinase, troxacitabine is a poor substrate for deoxycytidine deaminase, which rapidly inactivates the other nucleosides and has been linked to nucleoside resistance caused by enhanced deamination in some tumors in the clinic.15 Furthermore, the chiral specificity of troxacitabine may enhance the intracellular retention of phosphorylated metabolites, thereby conferring a biochemical advantage.7 In essence, chiral specificity seems to be an important determinant of substrate specificity of cellular enzymes and processes involved in nucleoside action and metabolism, and chiral specificity may be exploited to design more effective anticancer agents. Because the most important effects of nucleosides are on DNA synthesis, and because the duration of drug exposure is a principal determinant of drug action in vitro,16 the schedule of nucleoside administration may significantly influence the antitumor and toxic effects of these agents in the clinic.17 Schedule-dependence has not, however, been as prominent with troxacitabine as with other nucleosides in preclinical studies, possibly because of troxacitabine’s low clearance and the high intracellular retention of troxacitabine triphosphate.7,8,16 These features may obviate the need for protracted administration schedules to maximize drug exposure. Nevertheless, antitumor activity in preclinical studies has generally been superior with frequent divided dose-schedules (eg, daily x 5) compared with less frequent dose schedules.8 This, in part, provided the rationale for this phase I and pharmacokinetic study of troxacitabine administered as a 30-minute IV infusion daily for 5 days every 3 to 4 weeks. Hematologic toxicity, particularly neutropenia, was the most common type of adverse effect and the principal DLT of troxacitabine in the present study. MP and HP patients in the first course of treatment experienced a high frequency of unacceptable hematologic events with troxacitabine doses above 1.5 and 1.2 mg/m2/d, respectively. Because repetitive treatment of both HP and MP patients was feasible only after the treatment interval was increased to 4 weeks, troxacitabine doses of 1.5 and 1.2 mg/m2/d every 4 weeks are recommended for phase II studies in MP and HP patients, respectively. Although grade 3 to 4 neutropenia was common in MP and HP patients at these respective doses, the duration of severe neutropenia was generally brief (< 5 days) and uncomplicated. Despite this, the requirement, albeit infrequent, for treatment delay caused by prolonged recovery of blood cell counts to levels required for retreatment in MP and HP patients alike suggests that further study of risk factors is required, and that the use of troxacitabine at these doses should be restricted to patients similar to those in the present study. Troxacitabine resulted in several types of dermatologic effects, the most common of which was a maculopapular eruption that occurred in 23 (19%) of 124 courses. Rashes were generally mild to moderate and self limiting, and symptoms were usually managed successfully with simple pharmacologic measures, such as brief courses of antihistamines and/or corticosteroids. Although four patients in four courses experienced skin rashes of grade 3 severity, these more severe events resolved rapidly and did not preclude retreatment. Both the incidence and severity of the skin rashes seemed to be dose related, with an unacceptable frequency of severe dermatologic events in patients treated with 1.8 mg/m2/d of troxacitabine (three of six courses that involved three of six patients). Because the incidence of unacceptable hematologic toxicities also abruptly increased at troxacitabine doses above 1.5 mg/m2/d, the role of prophylactic measures to decrease the incidence and severity of skin toxicity and permit further dose escalation was not rigorously evaluated. Prophylactic administration of corticosteroids (dexamethasone 4 mg orally every 8 hours for 3 to 5 days) was effective in ameliorating or preventing skin rash in 11 subjects who were retreated with troxacitabine after experiencing dermatologic toxicity of grade 2 to 3 severity in a previous course. Troxacitabine also produced an erythematous desquamating rash that involved the hands and feet, which was readily distinguishable from the more common and aforementioned maculopapular rash. The eruption resembled the hand-foot syndrome associated with liposomal formulations of anthracyclines and protracted treatment with the fluoropyrimidines.18,19 Palmar-plantar erythrodysesthesia was experienced by four patients in four courses at troxacitabine doses of at least 1.2 mg/m2/d. Manifestations were always evident after treatment with at least two courses, seemed to progressively worsen with each successive course, and were not generally responsive to treatment with antihistamines and/or corticosteroids. Although both incidence and severity of hand-foot syndrome might be expected to increase with higher troxacitabine doses than those evaluated in the present study, hand-foot syndrome was not reported in a phase I study of troxacitabine on the schedule of 5 days daily every 3 to 4 weeks at doses below 8 mg/m2/d.20 Similarly, hand-foot syndrome was not evident at doses below 10 mg/m2 in a phase I study of troxacitabine as a single 30-minute infusion every 3 weeks.21 On the basis of this limited experience, it is interesting to speculate that pharmacokinetic parameters that reflect drug exposure, particularly Cmax, may relate to the development of this toxicity. Hand-foot syndrome has not been described with either gemcitabine or cytarabine use, which further indicates that drug exposure may be a critical determinant of this toxicity, perhaps enhancing the intracellular retention of troxacitabine triphosphate. In the present study, pharmacodynamic relationships that link the severity of dermatologic effects with pharmacokinetic parameters of drug exposure were not evident, irrespective of the type of comparison performed. However, higher AUC and Cmax values were generally noted in patients who experienced more severe dermatologic events. For example, three of six patients with day-5 AUC values in the upper ninetieth percentile experienced grade 3 dermatologic toxicity. Nevertheless, the small number of patients with any specific toxicity limit the statistical power of analyses aimed at determining the relative influence of potentially relevant pharmacokinetic parameters, dose, and schedule. In contrast to the pharmacokinetic behavior of nucleosides with a D configuration that undergo rapid systemic clearance caused by deamination mediated by deoxycytidine deaminase, troxacitabine exhibited a long plasma half-life (mean, 39.3 hours). Moreover, plasma concentrations capable of inhibiting the growth of a variety of human tumor cell lines (eg, 5 to 150 nmol/L, 1 to 30 ng/mL) were readily achieved and sustained for protracted periods. In contrast, half-lives of 7 to 20 minutes and 17 minutes have been reported for cytarabine and gemcitabine, respectively.22,23 The systemic clearance of troxacitabine (day 1 mean, 65.8 mL/min/m2) was also comparable to the glomerular filtration rate, and the renal excretion of unmetabolized troxacitabine in the 48-hour period after the fifth daily dose accounted for an average of 77% of the administered drug. Renal clearance may account for an even greater proportion of systemic clearance than estimated in the present study, in which urine sampling was limited to the 48-hour period immediately after the fifth dose of treatment, because troxacitabine was still detectable at days 15 and 21 in a proportion of patients both in this and the leukemia study.20 Troxacitabine clearance also related well to renal function, as assessed by creatinine clearance, which supports the importance of renal function in troxacitabine clearance. Furthermore, because decrements in neutrophil counts were related to drug exposure, it follows that renal function must be rigorously monitored in subsequent evaluations, and the dosing recommendation derived from this study should be applied only to patients with normal renal function. The dominance of renal excretion in the elimination of troxacitabine and the adequacy of a pharmacodynamic model to describe neutropenia indicate that, as with carboplatin, it may be feasible to devise individualized dosing algorithms largely on the basis of renal function to maximize troxacitabine’s therapeutic index and minimize interindividual variability.24-26 The preliminary antitumor activity observed in patients with various drug-refractory malignancies, including choroidal melanoma, renal cell carcinoma, carcinoma of unknown primary, and acute leukemia, in phase I studies is encouraging and provides further impetus for the development of troxacitabine, beginning with broad phase II evaluations.1,4-11 On the basis of the preclinical antitumor spectrum of troxacitabine and the preliminary clinical activity noted to date, troxacitabine may be active against neoplasms that constitutively overexpress P-glycoprotein, such as those derived from renal and gastrointestinal tissues, malignancies with acquired multidrug resistance, and cancers that are not typically responsive to nucleosides. Although the ultimate clinical activity of troxacitabine will be defined only in appropriate phase II/III trials, troxacitabine’s specific pattern of myelotoxicity, its readily manageable nonhematologic toxicities, and its activity against several types of neoplasms in early clinical evaluations warrant broad disease-directed evaluations of troxacitabine on this administration schedule and possibly less frequent schedules because of troxacitabine’s long plasma half-life, protracted intracellular retention of troxacitabine triphosphate, and antitumor activity noted in both preclinical and preliminary clinical evaluations.
Presented in part at the Thirty-Fifth Annual Meeting of the American Society of Clinical Oncology, Atlanta, GA, May 15-18, 1999.
1. Grove KL, Guo X, Liu SH, et al: Anticancer activity of beta-L-dioxolane-cytidine, a novel nucleoside analogue with the unnatural L configuration. Cancer Res 55: 3008-3011, 1995
2.
Chang CN, Doong SL, Zhou JH, et al: Deoxycytidine deaminase-resistant stereoisomer is the active form of (+/-)-2',3'-dideoxy-3'-thiacytidine in the inhibition of hepatitis B virus replication. J Biol Chem 267: 13938-13942, 1992
3.
Coates JA, Cammack N, Jenkinson HJ, et al: The separated enantiomers of 2'-deoxy-3'-thiacytidine (BCH-189) both inhibit human immunodeficiency virus replication in vitro. Antimicrob Agents Chemother 36: 202-205, 1992
4.
Siu LL, Attardo G, Izbicka E, et al: Activity of (-)-2'-deoxy-3'-oxacytidine (BCH-4556) against human tumor colony-forming units. Ann Oncol 9: 885-891, 1998 5. Biochem Pharma Inc: Troxacitabine (BCH-4556). Investigator’s Brochure. Laval, Quebec, Canada, 1999 6. Gourdeau H, Clarke ML, Ouellet F, et al: Mechanisms of uptake and resistance to Troxatyl, a novel deoxycytidine nucleoside analog, in solid-tumor and leukemic cell lines. Proc Am Assoc Cancer Res 42: 79, 2001 (abstr 423)
7.
Grove KL, Cheng YC: Uptake and metabolism of the new anticancer compound beta-L-(-)-dioxolane-cytidine in human prostate carcinoma DU-145 cells. Cancer Res 56: 4187-4191, 1996 8. Attardo G, Kadhim S, Angers E, et al: Dose scheduling effects on the anti-tumor activity of BCH-4556 against human tumor xenografts in nude mice. Proc Am Assoc Cancer Res 38: 124, 1997 (abstr 670)
9.
Kadhim SA, Bowlin TL, Waud WR, et al: Potent antitumor activity of a novel nucleoside analogue, BCH-4556 (beta-L-dioxolane-cytidine), in human renal cell carcinoma xenograft tumor models. Cancer Res 57: 4803-4810, 1997
10.
Weitman S, Marty J, Jolivet J, et al: The new dioxolane, (-)-2'-deoxy-3'-oxacytidine (BCH-4556, troxacitabine), has activity against pancreatic human tumor xenografts. Clin Cancer Res 6: 1574-1578, 2000
11.
Rabbani SA, Harakidas P, Bowlin T, et al: Effect of nucleoside analogue BCH-4556 on prostate cancer growth and metastases in vitro and in vivo. Cancer Res 58: 3461-3465, 1998 12. Gibaldi M, Perrier D: Pharmacokinetics (ed 2). New York, NY, Marcel Dekker, 1982, pp 409-417 13. Cockcroft DW, Gault MH: Prediction of creatinine clearance from serum creatinine. Nephron 16: 31-41, 1976[Medline] 14. Kroemer HK, Gross AS, Eichelbaum M: Enantioselectivity in drug action and metabolism: Influence on dynamics, in Cutler NR, Sramek JJ, Narang PK (eds): Pharmacodynamics and Drug Development: Perspectives in Clinical Pharmacology, New York, NY, Wiley, 1994, pp 103-114 15. Grant S: Ara-C: Cellular and molecular pharmacology. Adv Cancer Res 72: 197-233, 1998[Medline] 16. Huang P, Robertson LE, Wright S, et al: High molecular weight DNA fragmentation: A critical event in nucleoside analogue-induced apoptosis in leukemia cells. Clin Cancer Res 1: 1005-1013, 1995[Abstract] 17. Storniolo AM, Allerheiligen SR, Pearce HL: Preclinical, pharmacologic, and phase I studies of gemcitabine. Semin Oncol 24: S2-S7, 1997 (suppl 7) 18. Ranson MR, Carmichael J, O’Byrne K, et al: Treatment of advanced breast cancer with sterically stabilized liposomal doxorubicin: Results of a multicenter phase II trial. J Clin Oncol 15: 3185-3191, 1997[Abstract]
19.
Mackean M, Planting A, Twelves C, et al: Phase I and pharmacologic study of intermittent twice-daily oral therapy with capecitabine in patients with advanced and/or metastatic cancer. J Clin Oncol 16: 2977-2985, 1998
20.
Giles FJ, Cortes JF, Baker SD, et al: Troxacitabine, a novel dioxolane nucleoside analog, has activity in patients with advanced leukemia. J Clin Oncol 19: 762-771, 2001 21. Belanger K, Moore M, Siu L, et al: A phase I study of the L-nucleoside beta-L-dioxolane-cytidine (BCH-4556) given as a 30-minute infusion every 21 days. Proc Am Soc Clin Oncol 18: 197a, 1999 (abstr 760) 22. Capizzi RL, Yang JL, Cheng E, et al: Alteration of the pharmacokinetics of high-dose ara-C by its metabolite, high ara-U in patients with acute leukemia. J Clin Oncol 1: 763-771, 1983[Abstract] 23. Plunkett W, Huang P, Searcy CE, et al: Gemcitabine: Preclinical pharmacology and mechanisms of action. Semin Oncol 23: S3-S15, 1996 (suppl 10)
24.
Egorin MJ, Van Echo DA, Tipping SJ, et al: Pharmacokinetics and dosage reduction of cis-diamine(1,1-cyclobutanedicarboxylato)platinum in patients with impaired renal function. Cancer Res 44: 5432-5438, 1984 25. Calvert AH, Newell DR, Gumbrell LA, et al: Carboplatin dosage: Prospective evaluation of a simple formula based on renal function. J Clin Oncol 7: 1748-1756, 1989[Abstract]
26.
Chatelut E, Canal P, Brunner V, et al: Prediction of carboplatin clearance from standard morphological and biological patient characteristics. J Natl Cancer Inst 87: 573-580, 1995 Submitted March 26, 2001; accepted August 13, 2001. This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
112.0 m/z and 230.2
.
and 
, values of three apparent outliers, which were excluded from the model fits depicted here; —, fits of the Emax model to the data sets.
