A phase I study of the P-glycoprotein antagonist tariquidar in combination with vinorelbine

Abstract

Purpose: P-glycoprotein (Pgp) antagonists have had unpredictable pharmacokinetic interactions requiring reductions of chemotherapy. We report a phase I study using tariquidar (XR9576), a potent Pgp antagonist, in combination with vinorelbine.

Experimental design: Patients first received tariquidar alone to assess effects on the accumulation of (99m)Tc-sestamibi in tumor and normal organs and rhodamine efflux from CD56+ mononuclear cells. In the first cycle, vinorelbine pharmacokinetics was monitored after the day 1 and 8 doses without or with tariquidar. In subsequent cycles, vinorelbine was administered with tariquidar. Tariquidar pharmacokinetics was studied alone and with vinorelbine.

Results: Twenty-six patients were enrolled. Vinorelbine 20 mg/m(2) on day 1 and 8 was identified as the maximum tolerated dose (neutropenia). Nonhematologic grade 3/4 toxicities in 77 cycles included the following: abdominal pain (4 cycles), anorexia (2), constipation (2), fatigue (3), myalgia (2), pain (4) and dehydration, depression, diarrhea, ileus, nausea, and vomiting, (all once). A 150-mg dose of tariquidar: (1) reduced liver (99m)Tc-sestamibi clearance consistent with inhibition of liver Pgp; (2) increased (99m)Tc-sestamibi retention in a majority of tumor masses visible by (99m)Tc-sestamibi; and (3) blocked Pgp-mediated rhodamine efflux from CD56+ cells over the 48 hours examined. Tariquidar had no effects on vinorelbine pharmacokinetics. Vinorelbine had no effect on tariquidar pharmacokinetics. One patient with breast cancer had a minor response, and one with renal carcinoma had a partial remission.

Conclusions: Tariquidar is a potent Pgp antagonist, without significant side effects and much less pharmacokinetic interaction than previous Pgp antagonists. Tariquidar offers the potential to increase drug exposure in drug-resistant cancers.

Fig. 1.

A, vinorelbine pharmacokinetics were studied in the first treatment cycle after administration of the agent alone and in combination with tariquidar (days 1 and 8—alternating in patients). Frequent blood samples were obtained out to 120 h through a peripheral i.v. at a site distant from the infusion site. Left, the data in one patient. Right, data for all patients. * and accompanying line, the mean values for all patients. B, tariquidar concentration-time profiles from 2 patients treated with 150 mg of tariquidar, which resulted in >7-fold differences in drug exposure in these patients. The mean drug concentrations for all patients 24, 48, and 72 h after the dose of tariquidar without concomitant vinorelbine administration in cycle 1 were 121, 66, and 43 ng/mL, respectively, which are above the concentration of the drug required to restore chemosensitivity in Pgp-expressing tumor cell lines. The mean (±SD) clearance and terminal half-life of tariquidar were 287 ± 196 mL/min and 33.6 ± 13.5 h, respectively. The mean tariquidar plasma concentrations at 24 and 48 h were 119 and 79.4 ng/mL on cycle 1, and 121 and 69.5 ng/mL on cycle 2, respectively.

Fig. 2.

Rhodamine efflux from CD56+ cells shows the extent and duration of Pgp inhibition after a single 150-mg i.v. dose of tariquidar. The inhibition of rhodamine efflux by the tariquidar present in the patient’s blood is compared with the inhibition by exogenously added PSC 833 (valspodar), a potent Pgp inhibitor. Rhodamine efflux is completely blocked by the addition of PSC833, and was similarly blocked by the tariquidar administered as part of the study. A, individual fluorescence-activated cell sorting analyses obtained using CD56+ cells isolated from three study patients are shown. Rhodamine is rapidly extruded from CD56+ cells in the blood sample obtained before the administration of tariquidar (Pre), and very little remains inside the cells. In the Pre samples, there is a very large difference between the amount of rhodamine inside the cells in the portion of the sample that received no exogenous PSC833 (thick solid green line, middle), and the portion to which exogenous PSC833 was added (red dotted line, right). Black line, control endogenous fluorescence. By contrast, the difference in the 2 portions after the 8-, 24-, and 48-h administration of tariquidar i.v. is very small (thick solid green line and red dotted line are superimposed). In a majority of the blood samples obtained after the administration of tariquidar, the amount of rhodamine remaining inside CD56+ cells was similar to that in cells to which an excess of PSC833 was added exogenously. This indicates the i.v. administered tariquidar was as effective in blocking Pgp as an excess of PSC833 added exogenously. B, summary of the data for all patients. This panel shows plots of the difference (PSC Efflux − Efflux) between the amounts of rhodamine remaining inside the CD56+ cells after a period of efflux. The difference that is plotted is that between cells in the half of the sample that was treated with exogenous PSC833 before the efflux period (PSC Efflux) and that inside cells to which no exogenous PSC833 (Efflux) was added. Greater inhibition of rhodamine efflux by circulating tariquidar results in a smaller difference between the PSC Efflux and the Efflux values. Thus, increased inhibition of Pgp by tariquidar is reflected by the smaller values on the Y-axis. As can be seen, even 48 h after the administration of a single 150-mg i.v. dose of tariquidar, the differences in the large majority of samples is still very small indicating the inhibition of efflux even 48 h after the administration of tariquidar is similar to that achieved by freshly adding PSC833 exogenously.

Fig. 3.

Effect of tariquidar on the accumulation of ^99mTc-sestamibi. The in vivo efficacy of tariquidar was also shown by performing ^99mTc-sestamibi scans obtained in the patients enrolled on study. ^99mTc-sestamibi, a radionuclide used in cardiac imaging studies, is a Pgp substrate and has been used as a functional imaging agent for the multidrug resistance-1 phenotype. Baseline ^99mTc-sestamibi scans were obtained, followed 48 to 96 h later by a second scan 1 to 3 h after the administration of tariquidar. A, evidence that inhibition of ^99mTc-sestamibi efflux occurred only in organs expressing Pgp. The mean 99mTc-sestamibi AUC in the heart, right lung, left lung, and liver is shown for all study patients. Large differences were noted in Pgp-expressing liver but not in the heart nor in the lungs, organs with very little Pgp. B, the percent change in the ^99mTc-sestamibi AUC_0–180 after tariquidar in the two patients with measurable responses. The ^99mTc-sestamibi AUC_0–180 in the tumors of both patients was increased after the administration of tariquidar, albeit not to the extent seen in the normal liver that most likely had higher levels of Pgp. Left, the results are those of a patient with breast cancer who experienced a MR (8 prior regimens and 11 prior agents including 4 cycles of paclitaxel as a single agent 6 mo before enrollment, and 4 of Adriamycin as part of the CAF regimen 4 y before enrollment); right, the results in a patient with renal cell carcinoma who experienced a PR (see Supplementary Figs. S4 and S5).