Vinflunine (Javlor) is the first fluorinated microtubule inhibitor belonging to the Vinca alkaloids family. Vinflunine is obtained by semisynthesis using superacidic chemistry to selectively introduce two fluorine atoms at the 20′ position of the catharanthine moiety. This compound has been selected for clinical development on the basis of encouraging preclinical activity that warrants study in patients with a wide spectrum of solid tumors. Clinically significant activity has been seen in phase II studies, mainly in the treatment of transitional cell carcinoma of the urothelial tract, non–small cell lung cancer, and carcinoma of the breast. Vinflunine is currently in phase III trial assessment in patients with (second line) transitional cell carcinoma of the urothelium and first-line advanced breast cancer. The efficacy of vinflunine in patients with advanced non–small cell lung cancer previously treated with a platinum-containing regimen was confirmed by a large phase III trial.
There is a need for new, efficacious chemotherapeutic agents that are associated with increased response rates and with limited and manageable toxicity. Vinflunine (Javlor, Pierre Fabre Medicament Laboratories) is a novel microtubule inhibitor obtained by semisynthesis using superacidic chemistry to selectively modify the catharanthine moiety of the Vinca alkaloid molecule and optimize the therapeutic index of these derivatives. The optimal structural modification of this compound is the introduction of two fluorine atoms at the 20′ position, a part of the molecule previously inaccessible by classic chemistry (Fig. 1; ref. 1) that led to a different and broader spectrum of activity relative to other Vinca alkaloids (2). Interestingly, its nonfluorinated counterpart was deprived of activity against experimental models highlighting the essential contribution of the fluorine atoms to this identified antitumor activity (1). These features support the development of vinflunine to treat human cancer and improve the range of evaluable effective treatments.
Mechanism of Action
Vinflunine, at the lowest effective concentrations, interacts with the Vinca alkaloid binding site on tubulin, suppresses microtubule dynamics (switching at microtubule ends between phases of slow growth and rapid shortening) and microtubule treadmilling (growth at the plus end and shortening at the minus end of the microtubule; refs. 3–5), causes cell cycle arrest which appears on fluorescence-activated cell sorting analysis as a G2 + M phase arrest, and is associated with an accumulation of cells in mitosis (5–7) leading to cell death via apoptosis (8). It also prevents the assembly of microtubules without affecting their disassembly at higher concentrations comparable with to those of the other Vinca alkaloids tested (5), and induced concentration-dependent reduction of the microtubular network of interphase cells, accompanied by paracrystal formation (9, 10).
Although those features are common to the Vinca alkaloid class, differences in the effects of vinflunine have been shown relative to those of vinblastine and vinorelbine (4). Vinflunine inhibits treadmilling less powerfully than vinblastine and vinorelbine, and does not suppress the rate of microtubule shortening, whereas vinblastine does. These different actions might have varied effects during mitosis, which may lead to differential effects on the cell cycle, and therefore, on cell killing (4). Compared with other Vincas, vinflunine exhibits the weakest overall affinity for tubulin, which results in the formation of fewer and smaller spiral filaments, effects that may be associated with its reduced neurotoxicity (2, 11). In addition, certain effects of vinflunine are more readily reversible than the other Vinca alkaloids, as shown by the reversibility of drug-induced centrosome separation and the permanence of the mitotic block (4, 10).
Vinflunine has shown vascular-disrupting and antiangiogenic activities in extensive in vitro studies (12, 13), along with antimetastatic effects, as observed in two in vivo models (13, 14). The in vivo effects occur at considerably lower doses than the maximum tolerated dose for vinflunine and are apparent either only with vinflunine, or occur over a much wider dose range than vinorelbine or vinblastine, when tested concurrently. The clear demonstration of the antivascular, antiangiogenic, and antimetastatic activities of vinflunine is of major interest.
The significant in vivo antitumor activity of vinflunine, and its superiority over vinorelbine, was first identified against murine i.v.-grafted P388 leukemia and then confirmed in a series of murine and human solid tumor xenografts (15, 16). Compared with vinorelbine, vinflunine showed markedly superior tumor growth inhibition against a panel (7 of 11) of human tumors xenografted onto nude mice (ref. 16; Fig. 2). Clear dose dependency and some schedule dependency were noted. Vinflunine significantly prolonged survival in five murine tumors by factors ranging from 100% to 357%, and proved superior to vinorelbine. Using an orthotopically implanted murine bladder cancer model, Bonfil et al. clearly showed the higher activity of vinflunine over vinorelbine with a favorable toxicity profile (17). These results anticipate a further clinical development in this setting.
Vinflunine, like the other Vinca alkaloids, seems to be subject to P-glycoprotein–mediated drug resistance mechanisms, whereas atypical multidrug-resistant human tumor sublines retain full sensitivity to vinflunine (18). Bcl-2 down-regulation is associated with vinflunine resistance in an ovarian cancer model (19). More importantly, vinflunine induces drug resistance far less readily than vinorelbine in terms of the number of passages required to select for total resistance and the level of resistance ultimately obtained. In vitro full resistance was reached in 8 months versus 2 months after 2-fold IC50 exposures to vinflunine and vinorelbine, respectively. In another model, in vivo complete resistance to vinflunine was obtained after 36 weeks of exposure at subtherapeutic doses compared with only 11 weeks for vinorelbine (20). Although the mechanism of this lower propensity to induce resistance has not yet been explained, this may have important clinical implications.
Vinflunine exerts marked cytotoxicity against a selected panel of nine human solid tumor cell lines of different origins (colon, prostate, bladder, breast, and ovary) with IC50 values ranging from 27 nmol/L to 14 μmol/L (2). Vinflunine induces a typical dose-response curve with an initial exponential region tending to plateau with increasing concentrations.
Vinflunine shows more pronounced in vitro radiosensitization than vinorelbine (5). In vitro combinations identified a high level of synergy with vinflunine and various chemotherapeutic drugs with completely different modes of action including the DNA-damaging agents cisplatin and mitomycin C (21, 22).
Pharmacokinetics and Metabolism
Knowledge of vinflunine pharmacokinetics was obtained from several phase I studies. The first phase I study with vinflunine, administered once every 3 weeks, showed a dose-proportional increase between blood exposure to vinflunine (and its metabolites) and dose levels ranging from 30 up to 400 mg/m2 (ref. 23; see Fig. 3). In the same study, a close correlation was found between the neutrophil count at nadir and vinflunine blood exposure, whereas no relationship was observed with peak concentration (23). The mean pharmacokinetic variables calculated on a 168-hour time period from two other phase I studies (24, 25) are presented in Table 1.
Following i.v. administration to patients, vinflunine was eliminated according to a multiexponential decay with a rapid decrease of blood concentrations during the first hour. The terminal half-life is ∼40 h. The volume of distribution of the terminal elimination phase is large, 2,422 L (35.1 L/kg), suggesting important tissue distribution and uptake. Total clearance in blood is large (43.9 L/h, 0.640 L/h/kg). Vinflunine binding to blood cells is moderate, and that to platelets is negligible. The binding of vinflunine to human serum proteins is moderate (∼53%), and that to platelets is negligible (<5%). The binding of vinflunine to human serum proteins is also moderate (∼79%); there was no binding to α1-acid-glycoprotein. Several metabolites have been observed in the blood, the main one, 4-O-deacetylvinflunine, is the only active metabolite and is comparable with the parent compound in activity. The terminal half-life of 4-O-deacetylvinflunine (∼5 days) is longer than that of the unchanged compound. Excretion of vinflunine and its metabolites is higher in feces than in urine (2/3 and 1/3 of the recovered radioactivity, respectively; refs. 23, 26). A phase I study including 25 patients with either mild, moderate, or severe liver impairment indicated no modification of pharmacokinetic disposition from either vinflunine and 4-O-deacetylvinflunine compared with phase I patients with normal liver function (27).
Safety and Tolerability
Data drawn from 880 patients (3,335 cycles) evaluated at 320 mg/m2 every 3 weeks in phase II and III trials established that the dose-limiting toxicity of vinflunine was grade 3 to 4 neutropenia, which was observed in 49.2% of patients but was reversible and noncumulative. Febrile neutropenia was observed in 5.5% of patients (Table 2). Severe fatigue (grade 3-4) was experienced by 12.4% of patients. Constipation was frequently reported with vinflunine (54.1% of patients during treatment), with 10.4% at grades 3 to 4. This adverse event was manageable by prophylactic laxative treatment, however. Similarly, the incidence of grades 3 to 4 abdominal pain was at 5.9%.
Other symptoms occurred with the vinflunine use: grades 3 to 4 myalgia (4.2% of patients), injection site reactions (20.6% of patients), and neuropathy (all grades, 12%; grade 3, 0.5%) in patients previously exposed to platinum derivatives and taxanes. Vinflunine seems to be far less neurotoxic with respect to other Vincas and microtubule inhibitors such as taxanes and epothilones. Grade 3 peripheral neurotoxicity was observed primarily in patients previously treated with neurotoxic drugs and enrolled in trials allowing baseline neurologic impairment. Combining these observations with the low incidence of febrile neutropenia, vinflunine can be viewed as a new cytotoxic Vinca alkaloid with an excellent safety profile.
Clinical Efficacy: Promising Results in Transitional Cell Carcinoma of the Urothelium, Lung Carcinoma, and Breast Carcinoma
Phase I trials: Demonstration of a manageable profile of vinflunine. The clinical development of vinflunine began at the end of 1998. Three phase I trials (23–25) were conducted to determine the maximum tolerated dose and the recommended dose according to three different schedules of vinflunine administration: on day 1 every 3 weeks, weekly administration, and on days 1 and 8 every 3 weeks. The secondary objectives of these studies were to establish the toxicity profile, to determine duration and reversibility of toxicities, to establish the pharmacokinetic profile of each schedule, to establish the relationship between the pharmacokinetic results and toxicity observed, and eventually, to assess the antitumor activity. Results are described in the Table 3.
Vinflunine treatment every 3 weeks was considered adequate, based on pharmacokinetic and safety profiles from the three dose schedules evaluated in phase I trials. Vinflunine was administered at the beginning of the phase II trials at 350 mg/m2 in normal saline as a 10-min infusion according to the clinical results obtained in the phase I trial. A preliminary safety analysis performed on the first patients enrolled in the early phase II studies resulted in a dose reduction to 320 mg/m2 for all subsequent patients included in clinical trials.
Phase II and III trials with vinflunine. An international program of phase II studies with vinflunine as a single agent has been carried out in chemonaïve patients, and also as salvage therapy to determine the tumor response in a large spectrum of solid tumors: melanoma, renal carcinoma, ovarian cancer (after first line paclitaxel/platinum), mesothelioma (first line), colon carcinoma (after oxaliplatin/irinotecan-containing regimens), non–small cell lung cancer (NSCLC; second line after platinum-containing regimen), in transitional cell carcinoma of the urothelium (second line after platinum-containing regimen), and breast carcinoma (after anthracyclines/taxanes).
An open label phase II study of patients presenting with advanced transitional cell carcinoma of the urothelium who failed a previous platinum-containing regimen was designed to evaluate the activity and safety of i.v. vinflunine in this setting. Nine partial responses were seen in the 51 patients treated with vinflunine at 320 mg/m2 every 3 weeks [18% overall response rate; 95% confidence interval (CI), 8.4-30.9%]. Three out of nine responders had previously received treatment with a Vinca alkaloid-containing regimen and five had visceral involvement. The median duration of response, progression-free survival (PFS), and overall survival (OS) were established at 9.1 months (4.2-15.0), 3 months (2.4-3.8), and 6.6 (4.8-7.6) months (28). D. Petrylak and coworkers included 114 refractory patients previously treated with platinum-containing regimen and obtained an objective response rate of 15% with a median duration of response of 6.8 months (29). The overall benefit of cancer chemotherapy in the management of urothelial carcinoma patients whose disease has progressed after or during a prior platinum-containing regimen remains controversial and is a medically unmet need. These patients have a poor prognosis and a median survival of ∼4 months. This observation has raised the possibility of improving survival for patients with bladder cancer previously treated with platinum-containing regimens. This question can only be answered through a prospective randomized phase III trial in which vinflunine is compared with a non–chemotherapy approach because no standard is currently available. This trial is currently under analysis. Another trial has been initiated in the United States comparing vinflunine plus gemcitabine versus gemcitabine in patients who cannot receive a platinum-containing regimen as first line treatment. This is another unmet need as no standard exists.
In metastatic breast cancer, 60 patients previously treated with anthracyclines and taxanes were treated with i.v. vinflunine at 320 mg/m2 every 3 weeks. An independent external review reported an objective response rate of 30.0% according to WHO criteria. A total of 65.0% of disease control was achieved. Considering the time elapsed between the end of the treatment with taxane-containing regimens and disease progression, two groups were identified: (a) patients who relapsed <6 months, i.e., who are refractory or resistant (poor prognosis); and (b) patients relapsing during the six month or later after completion of a taxane regimen for advanced disease. A response rate of 34.4% was achieved in the first group. Among the 21 patients who relapsed during or less than 3 months after completion of the previous taxane regimen, 7 reached partial response (33.3%). The response rate observed in the second group was 29.2% with a disease control of 66.7%. Of note, 14 out of 53 patients (26.4%) who were enrolled with visceral involvement achieved a partial response. Among the patients enrolled with liver metastasis as the main target lesion, 11 (27.5%) responded to the treatment and 60% achieved disease control. The median duration of response, median PFS, and OS were 4.8 months (95% CI, 2.8-4.2. months), 3.7 months (95% CI, 2.8-4.4), and 14.3 months (9.2-19.6), respectively (30). Fumoleau and colleagues (31) assessed 58 patients treated with the abovementioned dose of vinflunine as third-line treatment after anthracycline/taxane failure. Thirteen percent of patients achieved partial response. The median OS was established at 11.4 months (95% CI, 7.4 to 14.2. months). These results supported a clinical phase III trial of vinflunine in a first line metastatic setting comparing vinflunine plus gemcitabine versus paclitaxel plus gemcitabine. This trial is ongoing.
In NSCLC, the initial assessment of vinflunine was also undertaken in patients who had failed a previous platinum-based regimen. The aim of this trial was to determine the response rate and safety of vinflunine as a single agent in patients with advanced NSCLC. Of the 62 evaluable patients, 5 achieved partial response (8%; 95% CI, 1-15%) and 38 (61%; 95% CI, 49-73%) achieved disease control. PFS was established at 2.6 months (95% CI, 1.4-3.8) and 7 months for OS (32). Based on these results, a large randomized phase III trial was performed comparing vinflunine with docetaxel in patients with advanced NSCLC previously treated with a platinum-based regimen. There were 547 patients with a median age of 62 years (range, 22-83), Eastern Cooperative Oncology Group performance status of 0 to 1 (89%), and stage IV (62%) and stage IIIB (23%) disease. All patients had one prior platinum-based regimen in combination with a Vinca alkaloid (22%), paclitaxel (21%), or gemcitabine (48%). The dose of vinflunine in NSCLC was 320 mg/m2 every 3 weeks and docetaxel at 75 mg/m2 every 3 weeks. Relevant grade 3 to 4 toxicities (vinflunine versus docetaxel) included neutropenia (32.8% versus 29.5%), febrile neutropenia (3% versus 5%), fatigue (11% versus 6%), abdominal pain (4% versus <1%), and constipation (7% versus <1%). Efficacy end points were achieved: median PFS [2.3 versus 2.3 months; hazard ratio, 1.004 (0.841-1.199)], response rate (4.4% versus 5.5%), stable disease (36% versus 39%), and median survival [6.7 versus 7.2 months; hazard ratio, 0.973 (0.805-1.176)]. The investigators concluded that both arms induced a low and manageable toxicity. Vinflunine therapy resulted in clinically similar efficacy outcomes, placing this agent as a new and useful alternative for second-line treatment of NSCLC (33).
Data are also now available on the activity of vinflunine in the treatment of malignant pleural mesothelioma. Nine out of 62 chemonaïve patients achieved partial response rates following external radiologic review (13.8%; 95% CI, 6.5-24.7). In these patients, median PFS and OS were 3.2 months (95% CI, 2.6-4.2) and 10.8 months (95% CI, 7-12.0), respectively. The 1-year survival rate was established at 37% (34).
There was no evidence from phase II trials that vinflunine had satisfactory activity in the treatment of metastatic malignant melanoma (35), renal cell carcinoma (36), or colorectal cancer (ref. 37; Table 4). Also, in ovarian carcinoma, results of the single agent phase II study after platinum-taxane failure in both sensitive or refractory patient populations remained unsatisfactory (37).
Given the evidence of activity against NSCLC in second-line therapy, it is appropriate that vinflunine should be evaluated in combination with other agents in the first-line treatment of this disease (Table 5). A phase I/II study has been undertaken combining vinflunine with cisplatin. The recommended dose has been established at 320 mg/m2 of vinflunine plus 80 mg/m2 of cisplatin on day 1 every 3 weeks; no pharmacokinetic interaction was observed. The results from phase II of the study, which was designed to determine the response rate and safety of the combination, showed that at the recommended dose there were 16 partial responses confirmed by an external radiologic review (33%; 95% CI, 20-48), with disease control achieved in 77% of the 49 evaluable patients (38).
Data from a study similarly designed to establish the recommended dose also confirmed significant activity for combinations of vinflunine with gemcitabine. Three dose levels of the combination were investigated in 12 patients. All patients were chemonaïve for advanced or metastatic disease. The recommended dose was established at 320 mg/m2 on day 1 and gemcitabine at 1,000 mg/m2 on days 1 and 8. To date, 44% of the patients treated at the recommended dose achieved partial response according to Response Evaluation Criteria in Solid Tumors standards with a good tolerance profile. Pharmacokinetic analysis did not detect any drug-drug interaction of vinflunine and gemcitabine when combined (39). Vinflunine was also tested in combination with carboplatin. The recommended dose of the combination was established at 320 mg/m2 in combination with carboplatin's area under the curve of 5; the response rate was 37% and the tolerability was acceptable (40).
Clinical trials with the vinflunine combination are ongoing in advanced/metastatic breast cancer: vinflunine with capecitabine (41), vinflunine plus trastuzumab (42), and vinflunine in combination with anthracyclines.
As discussed elsewhere in this issue of CCR Focus, other new chemotherapeutic agents are also under development, such as epothilones, macrolide compounds that stabilize microtubules, or satraplatin, an oral platinum analogue (43–45). The armamentarium of anticancer drugs will also be increased with new targeted therapies including proteasome inhibitors and aurora kinase inhibitors (46, 47). Vinflunine represents a well-founded development of the program by the Institut de Recherche Pierre Fabre to evaluate novel antimicrotubule compounds. There is a thorough preclinical evaluation that suggests that the compound will have advantages in terms of efficacy, tolerability, and range of activity over vinorelbine. The initial clinical evaluation has confirmed significant activity in heavily pretreated patients with breast cancer, bladder cancer, and NSCLCs, i.e., poor prognosis patient groups. The vinflunine toxicity profile is acceptable with neutropenia being the dose-limiting toxicity. Phase II studies have shown promising activity and tolerability for vinflunine given at 320 mg/m2 once every 3 weeks in the treatment of pretreated patients with breast cancer, bladder cancer, and NSCLC. This fully justified further evaluation in ongoing and completed phase III trials in pretreated patients with bladder and breast cancers. Results in lung cancer in the second line therapy setting are promising. Preclinical data support the possibility that vinflunine will be synergistic with other chemotherapeutic agents that have established activity. Combination trials are ongoing in Europe and the United States.
In conclusion, the new Vinca, vinflunine, has advantages in terms of efficacy, tolerability, and range of activity over vinorelbine. Given the three different well-established Vinca alkaloids in the clinical armamentarium, a development strategy attempting to show superiority may be difficult. On the other hand, given the improved safety profile, it can be suggested that vinflunine, in the future, may eventually supplant vinorelbine. Results in lung cancer treatment in a second-line setting support this hypothesis. The initial clinical evaluation in heavily pretreated patients with breast carcinoma is also promising and additional phase III trials are ongoing. The development of vinflunine has influenced the treatment of two less common solid tumors, transitional cell carcinoma and mesothelioma, in which preliminary results are encouraging. Whether or not vinflunine supplants vinorelbine in breast and lung carcinoma, or finds a niche in transitional cell carcinoma or mesothelioma, an important strategy for the future development of this agent will be to show its ability to combine with new targeted therapies or other chemotherapeutic agents to improve anticancer outcome.
- Received October 4, 2007.
- Revision received December 4, 2007.
- Accepted January 14, 2008.