Potential hazard of pharmacokinetic interactions between lopinavir–ritonavir protease inhibitors and irinotecan

Abstract

The use of concomitant chemotherapy and highly active antiretroviral therapy (HAART) has been demonstrated to be feasible and effective in reducing morbidity associated with opportunistic infections and to improve overall survival in patients with HIV-related malignances [1,2]. However, such combination therapy causes severe toxicity, including neutropenia [3,4]. It has been hypothesized that such unpredictable toxicity can be related to pharmacokinetic interactions. Here, we report on the effect of a lopinavir–ritonavir (LPV/r) protease inhibitor combination on the pharmacokinetics of irinotecan, in an HIV-infected, 42-year-old man with Kaposi's sarcoma. The patient was enrolled in an ongoing phase II clinical trial, approved by the Institutional Review Board and the Ethics Committee to evaluate the efficacy of HAART and irinotecan therapy in HIV patients with Kaposi's sarcoma. The HAART regime in this patient consisted of a twice daily oral administration of 400 mg lopinavir and 135 mg ritonavir protease inhibitor combination (kaletra) in combination with a twice daily 300 mg zidovudine and 150 mg lamivudine (combivir) nucleoside reverse transcriptase combination. Chemotherapy consisted of 150 mg/m2 irinotecan administered by 1.5 h infusion at days 1 and 10, repeated every 3 weeks. Standard ondasentron treatment 30 min before the irinotecan infusion and a 6-day treatment with 300 μg/day granulocyte colony stimulating factor was also administered. Before the chemotherapy course, the patient underwent a 2-month treatment with the above HAART regime. The pharmacokinetic profile of irinotecan administered alone was investigated on the first day of the first chemotherapy cycle, after a 2-day protease inhibitor wash out, whereas the pharmacokinetics of irinotecan in combination with LPV/r was performed on the first day of the second chemotherapy cycle. The patient on this occasion received 50% of the irinotecan dose (75 mg/m2) because of persistent grade G2 WHO neutropenia, in agreement with clinical protocols. The plasma concentration of irinotecan, its active metabolite SN38, the inactive SN38 glucoronide and the 7-ethyl-10[4-N-(5 aminopentanoic acid)-1-piperidine]-carbonyloxycamptothecin (APC) oxidized metabolite were determined over a 50-h time interval by using a validated high-pressure liquid chromatography method [5]. Irinotecan is a prodrug metabolized in vivo to form the active metabolite SN38, which is further eliminated as a glucoronide metabolite [6]. Furthermore, irinotecan undergoes significant oxidative metabolism by the same P450 isoenzyme CYP3A4 involved in the metabolic clearance of both lopinavir and ritonavir protease inhibitors, prevalently to form the inactive APC metabolite [6,7]. Therefore, interferences at the level of this common metabolic pathway could occur. In the case study reported here, we found that the co-administration of LPV/r protease inhibitors resulted in a marked inhibition of the APC production from CPT11. The area under the curve (AUC)APC: AUCIrinotecan metabolic ratio was 20-fold lower during the course of irinotecan plus LPV/r (Table 1). This was in agreement with the 10-fold reduction of APC AUC∞ normalized for dose (AUC∞/dose) during the co-administration of irinotecan with LPV/r compared with that observed when irinotecan was administered alone. The reduced production of APC metabolite was associated with a 34% reduction of the irinotecan total body clearance and a twofold increase of the AUC∞/dose of the active metabolite SN38. This suggests that the inhibition of CYP3A4 activity by LPV/r may drive the irinotecan metabolism to an activation pathway. The relative extent of irinotecan to SN38 conversion expressed as AUCSN38: AUCIrinotecan was 1.5-fold greater during LPV/r combination therapy compared with irinotecan alone. These results are in agreement with a previous study by Kehrer et al. [8], which reported that ketoconazole, used as model inhibitor of CYP3A4, determined an almost complete inhibition of APC production associated with a shunt of irinotecan metabolism to SN38. Although in our study the patient received 50% of the irinotecan dose, during the course of irinotecan plus LPV/r, he was subjected to approximately the same SN38 AUC reported when the 100% irinotecan dose was administered alone. A persistent grade G2 WHO neutropenia, still observed during the second cycle of irinotecan plus LPV/r, probably as a result of the relatively high SN38 plasma levels, forced the patient to abandon the clinical protocol.Table 1: Effect of lopinavir–ritonavir protease inhibitor combination on pharmacokinetics of irinotecan.These results suggest that to avoid hazard and potential severe toxicity, caution should be taken when an LPV/r protease inhibitor combination is co-administered with irinotecan. Clinicians who intend using irinotecan-based chemotherapy in combination with HAART regimes containing LPV/r should carefully monitor their patients for potential adverse effects, and irinotecan dose reduction should be considered to avoid severe toxicity. Further pharmacokinetic–pharmacodynamic investigations, involving a large number of homogeneously treated patients, will provide a definite indication on irinotecan dose adjustments in patients receiving irinotecan in combination with LPV/r protease inhibitors.