Assessment of Drug–Drug Interaction between Warfarin and Aprepitant and Its Effects on PT-INR of Patients Receiving Anticancer Chemotherapy

Abstract

Aprepitant is a known inducer of CYP2C9, the main warfarin-metabolizing enzyme. Consequently, co-administration of these two drugs may result in reduction of the anticoagulation activity of warfarin. However, the nature and degree of time-dependent changes in prothrombin time international normalized ratio (PT-INR) after aprepitant and warfarin co-treatment in patients receiving anticancer chemotherapy has not been elucidated. We retrospectively examined the changes in warfarin dose, PT-INR, and warfarin sensitivity index (WSI; average of PT-INR value/average of daily warfarin dose) during four weeks, i.e., one week before and three weeks after aprepitant administration. The mean and standard deviation values of WSI for one week before and one, two, and three weeks after the beginning of aprepitant administration were 0.51±0.22 (1.00, n=34), 0.74±0.30 (1.53±0.59, n=30), 0.38±0.15 (0.82±0.22, n=28), and 0.46±0.29 (0.87±0.23, n=24), respectively. Values in parentheses represent relative changes versus WSI of one week before and number of subjects. Although the mean value of WSI significantly increased one week after aprepitant administration compared to that at one week before the administration, it in turn significantly decreased two weeks after compared to one week before (paired t-test, p<0.05 after Bonferoni correction). In patients taking warfarin, PT-INR should be carefully monitored for at least two weeks after the beginning of aprepitant administration because it may fluctuate with both aprepitant and chemotherapy during this period.

Aprepitant is a potent, selective, brain-penetrant nonpeptide neurokinin-1 (NK₁)-receptor antagonist. It is indicated for the prevention of acute and delayed chemotherapy-induced nausea and vomiting (CINV) in combination with corticosteroids and a 5-hydroxy-tryptamine 3 (5-HT₃) receptor antagonists.¹⁾ Aprepitant has been reported to be a moderate inhibitor and inducer of CYP3A4, as well as an inducer of CYP2C9.^2,3) The administration of aprepitant (125 mg) is recommended one hour prior to initiating chemotherapy treatment, followed by doses of 80 mg in the morning of the second and third days.

Warfarin is the most widely prescribed anticoagulant drug, and has been used for the treatment and prevention of thromboembolic diseases.⁴⁾ Drug interactions with this oral anticoagulant are clinically relevant since an interaction leading to enhanced action may be associated with an increased risk of hemorrhage. Conversely, interactions that decrease warfarin plasma concentrations might decrease the anticoagulant effect of warfarin. Warfarin is clinically administered as a racemic mixture of R- and S-warfarin, which differ in the potency of their anticoagulation effect, with the potency of S-warfarin being three to five times higher than that of R-warfarin.⁵⁾ The R- and S-forms also have different metabolic pathways: while R-warfarin is metabolized by CYP3A4 and CYP1A2, S-warfarin is mainly metabolized by CYP2C9.⁶⁾

The manufacturer cautions that co-administration of aprepitant with warfarin may result in a clinically significant decrease in prothrombin time international normalized ratio (PT-INR) owing to the induction of CYP2C9 by aprepitant, which causes S-warfarin concentration to decrease. Compared to placebo-treated subjects, in healthy subjects stabilized on warfarin and administered a 3-d regimen of aprepitant, S-warfarin concentration and PT-INR were found to decrease, respectively, by 34 and 14% on day 8 after aprepitant administration.²⁾ It is clinically important to elucidate whether aprepitant causes a reduction of the anticoagulation activity of warfarin in patients receiving anticancer chemotherapy, as well as in healthy subjects. We previously reported two cases in which treatment with aprepitant persistently altered antithrombotic control in patients receiving warfarin.⁷⁾ However, the nature and degree of time-dependent changes in PT-INR after aprepitant and anticancer co-treatment remains unknown. In the present study, we aimed to clarify the effects of a drug–drug interaction between aprepitant and warfarin for a long-term monitoring on PT-INR in patients receiving anticancer chemotherapy.

MATERIALS AND METHODS

Study Design

In this retrospective study, 87 patients co-administered aprepitant and warfarin in the University of Tokyo Hospital from February 2010 to October 2014 were initially selected. From the selected patients, 29 patients were excluded because the aprepitant administration interval was less than four weeks or whose warfarin treatment initiation was less than 10 d before aprepitant dosing. In addition, 39 and 5 patients were excluded because the warfarin dose was unknown, and because they were concomitantly treated with other drugs known to strongly inhibit or induce CYP2C9⁸⁾ (Table 1), respectively. We examined the changes in warfarin dose and PT-INR over a four-week period (1 cycle, i.e., one week before and three weeks after aprepitant administration) in 14 patients who were administered warfarin for over 10 d before adding aprepitant at an interval longer than four weeks (Fig. 1A). The total number of aprepitant cycles administered to the 14 patients was 52. From these, we used data from 34 cycles for the analysis because 12, 5, and 1 cycle(s) were excluded, respectively, because there was no co-administration of aprepitant and warfarin, the daily warfarin dose was unknown, and PT-INR data of one week before aprepitant administration were unavailable (Fig. 1C).

Table 1. Drugs Known to Strongly Inhibit or Induce CYP2C9⁸⁾

Inhibitor of CYP2C9		Inducer of CYP2C9
TS-1	Miconazole	Rifampicin
Tegafur	Voriconazole	Phenobarbital
Fluorouracil	Fluconazole	Phenytoin
Doxifluridine	Sulfaphenazole	Carbamazepine
Capecitabine	Bucolome
UFT	Benzbromarone
Amiodarone

TS-1: Tegafur, Gimeracil, Oteracil Potassium. UFT: Tegafur, Uracil.

Fig. 1. Study Protocol (A), Warfarin Sensitivity Index (B) and Selection of Study Cohort (C)

Demographic and clinical data were collected from the hospital records over the four-week period including age, sex, cancer type, protocol of chemotherapy, reason for warfarin prescription, mean warfarin dosage, concomitant drugs, PT-INR, and serum biochemical markers (aspartate aminotransferase (AST), alanine aminotransferase (ALT), creatinine and albumin.) Furthermore, we evaluated the changes in Warfarin Sensitivity Index (WSI) during those four weeks, which was defined as the mean of PT-INR divided by the mean daily warfarin dose⁹⁾ (Fig. 1B). However, where a cycle included a week in which there was no PT-INR, the data were excluded.

Statistical Analysis

Statistical analysis was performed using PASW Statistics 17.0. Changes in warfarin dose, PT-INR, and WSI before and after administration of aprepitant were analyzed using a paired t-test. Values of p<0.05 after Bonferroni correction were considered statistically significant.

Ethical Considerations

This study was performed in accordance with the Declaration of Helsinki and its amendments, and conducted under the authority of the Graduate School of Medicine and the Faculty of Medicine at the University of Tokyo.

RESULTS

Study Population

Fourteen patients were ultimately included in the study, comprising six male and eight female subjects, with a mean age of 59±13 years (range: 33–78 years). Table 2 shows the patients’ characteristics at baseline. Table 3 shows the detailed list of drugs administered in combination with warfarin for each of the patients, and their suspected effect on warfarin activity. None of the patients involved in the present study presented adverse effects of grade 3 or above, according to the Common Terminology Criteria for Adverse Events (CTC AE) version 4.

Table 2. Patients’ Characteristics

Case	Sex	Age	Reason for anticoagulation	Cancer type	Chemotherapy regimen	Cycle	Monitor cycle (Total)
1	F	78	AF	Non-Hodgkin lymphoma	R-CHOP	5	5th (1)
2	F	70	DVT, PE	Ovarian tumors	Nogitecan	5	2nd–5th (4)
3	F	77	PE	Urothelial cancer	DIP	1	1st (1)
4	M	69	DVT	Lung cancer	NP	4	2nd, 3rd (2)
5	M	68	PE	Malignant melanoma	DAC-(tam)Feron	4	2nd–4th (3)
6	M	60	AF	Non-small cell lung cancer	CBDCA/ETP	3	1st–3rd (3)
7	F	33	DVT	Carcinoma of the uterine cervix	EP	4	4th (1)
8	F	47	DVT, PE	Carcinoma of the uterine corpus	IAP	4	1st, 3rd, 4th (3)
9	F	44	PE	Carcinoma of the uterine cervix	MEP	2	1st, 2nd (2)
10	M	73	DVT	Lung cancer	CBDCA/PTX	2	1st, 2nd (2)
11	M	53	DVT	Malignant femoral soft tissue tumor	MAID	2	2nd (1)
12	M	49	PE	Bladder tumor	GC	10	4–9th (6)
13	F	64	AF	Carcinoma of the uterine corpus	TC	4	2nd–4th (3)
14	F	49	DVT, PE	Ovarian tumors	CPT-P	2	1st, 2nd (2)

F: female; M: male; AF: atrial fibrillation; DVT: deep vein thrombosis; PE: pulmonary embolism. Details of supportive care drugs for the prevention of nausea and vomiting. Aprepitant dosage: 4 d (Case 7), 2 d (1st cycle of case 10), 5 d (Case 11), 3 d (The other cycle of all cases); Corticosteroid: Prednisolone (Case 1), Betamethasone (Case 10), Dexamethasone (The other cycle of all case); 5-HT₃ receptor antagonist: Palonosetron (Case 5, 6), Ramosetron (Case 9), Granisetron (The other cycle of all case). Details of chemotherapy regimen. R-CHOP: Rituximab day 1, Doxorubicin day 1, Cyclophosphamide day 1, Vincristine day 1, Prednisolone days 1–5; Nogitecan: Nogitecan days 1–5; DIP: Docetaxel day 1, Ifosfamide days 2–6, Cisplatin days 2–6; NP: Cisplatin day 1, Vinorelbine days 1, 8; DAC Feron: Interferon Beta days 1–4, Nimustine day 1, Cisplatin days 1–3, Dacarbazine days 1–3; CBDCA/ETP: Carboplatin day 1, Etoposide days 1–3; EP: Etoposide days 1–4, Cisplatin day 1; IAP: Doxorubicin day 1, Cisplatin day 1, Ifosfamide days 1–5; MEP: Cisplatin day 1, Mitomycin C day 1, Etoposide days 1, 3 and 5; CBDCA/PTX: Carboplatin day 1, Paclitaxel day 1, MAID: Pirarubicin days 1–3, Dacarbazine days 1–3, Ifosfamide days 1–3, Mesna days 1–3; GC: Cisplatin day 1, Gemcitabine days 1, 8, 15, TC: Paclitaxel day 1, Carboplatin day 1; CPT-P: Cisplatin day 1, Irinotecan days 1, 8, 15.

Table 3. List of Drugs Administered in Combination with Warfarin for Each of the Patients, and Their Suspected Effect on Warfarin Activity

Case	Drugs increasing the effect of warfarin			Drugs decreasing the effect of warfarin
1	Omeprazole	Gliclazide	Allopurinol	Prednisolone
	Aspirin	Acetaminophen
2	Acetaminophen	Levofloxacin		Dexamethasone
3	Thiamazole	Acetaminophen	Erythromycin	Dexamethasone
	Tramadol	Cisplatin
4	Prednisolone	Cisplatin		Dexamethasone
5	Bezafibrate	Cisplatin		Dexamethasone
6	Aspirin	Clopidogrel	Carboplatin	Dexamethasone
	Etoposide
7	Cisplatin	Etoposide		Dexamethasone
8	Levofloxacin	Sulfamethoxazole	Trimethoprim	Dexamethasone
	Cisplatin
9	Acetaminophen	Levofloxacin	Cisplatin	Dexamethasone
	Etoposide
10	Loxoprofen	Etoposide		Dexamethasone
11				Betamethasone
12	Loxoprofen	Acetaminophen	Cisplatin	Dexamethasone
13	Omeprazole	Acetaminophen	Naproxen	Dexamethasone
	Carboplatin	Paclitaxel
14	Acetaminophen	Cisplatin		Dexamethasone

Time-Dependent Changes in Mean Values of Warfarin Dose, PT-INR and WSI

As shown in Figs. 2A and B, the mean values of warfarin dose for one week before and one, two, and three weeks after aprepitant administration were 3.72±1.25 (mean±standard deviation (S.D.)), 3.37±0.89, 3.90±1.24, and 3.90±1.39, respectively. The mean values of warfarin dose the week after aprepitant treatment tended to be lower than those before, but the difference was not statistically significant (p>0.05 after Bonferroni correction).

Fig. 2. Time-Dependent Changes in Mean Values of Warfarin Dose, PT-INR and WSI

N.S.: Not significantly different (paired t-test. Bonferroni correction). ※: Erythromycin (1200 mg/d) was administered on days 11–13 after aprepitant administration and the PT-INR drastically increased three weeks after. (A, B) warfarin dose. (C, D) PT-INR. (E, F) WSI. (A, C, E) the mean and standard deviation values of warfarin dose, PT-INR or WSI. (B, D, F) warfarin dose, PT-INR or WSI values from individual patients.

The mean values of PT-INR for one week before and one, two, and three weeks after administration of aprepitant were 1.68±0.44, 2.30±0.67, 1.35±0.33, and 1.55±0.52, respectively. The mean values of PT-INR in 27 cycles out of 30 increased the week after aprepitant administration. In 25 cycles out of 28, PT-INR decreased two weeks after aprepitant treatment compared to the week before. As shown in Figs. 2C and D, the mean values of PT-INR significantly increased one week after administration (calculated p=0.0000149) and significantly decreased two weeks after (calculated p=0.00069), compared to the week before co-treatment.

The mean values of WSI one week before and one, two, and three weeks after administration of aprepitant were 0.51±0.22 (1.00, n=34), 0.74±0.30 (1.53±0.59, n=30), 0.38±0.15 (0.82±0.22, n=28), and 0.46±0.29 (0.87±0.23, n=24), respectively. Values in the parentheses represent relative changes versus WSI of one week before and number of subjects. The mean values of WSI in 26 cycles out of 30, increased the week after aprepitant administration. In 22 cycles out of 28 WSI decreased two weeks after aprepitant administration compared to the week before. As shown in Figs. 2E and F the mean values of WSI significantly increased one week after administration of aprepitant (calculated p=0.000139), and significantly decreased two weeks after (calculated p=0.00354), compared to the week before co-treatment.

DISCUSSION

We retrospectively examined the changes in warfarin dose, PT-INR, and WSI over a four-week period, i.e., one week before and three weeks after aprepitant administration. The mean and standard deviation values of WSI for one week before and one, two, and three weeks after the beginning of aprepitant administration were 0.51±0.22 (1.00, n=34), 0.74±0.30 (1.53±0.59, n=30), 0.38±0.15 (0.82±0.22, n=28), and 0.46±0.29 (0.87±0.23, n=24), respectively. The values in parentheses represent the relative changes with respect to the WSI from one week before treatment with its standard deviation, followed by the number of subjects in the cohort. Although the mean value of WSI significantly increased one week after aprepitant administration compared to that at one week before the administration, it in turn significantly decreased two weeks after compared to one week before. However, it was not a serious change for the patients compared to the known strong inhibitors or inducers of CYP2C9 (Table 1). For example, a small retrospective report found that in 32 patients who were stable whilst taking warfarin, and then administered miconazole oral gel, the INR increased from a mean of 2.44 (range of 1.92 to 3.18) to 8.8 (range of 4.9 to 16.9).¹⁰⁾ In the reports of the interaction between rifampicin and warfarin, the dose of warfarin required to maintain its clinical effect requires doubling.^11,12)

In a previous study in healthy subjects with a continued warfarin regimen and 3-d co-administration of aprepitant (125 mg on day 1, and 80 mg on days 2 and 3), S-warfarin concentration and PT-INR were, respectively, 34 and 14% lower on day 8 after aprepitant, compared to that in placebo-treated subjects.²⁾ Furthermore, the area under the curve (AUC) in healthy subjects after a single 500 mg dose of tolbutamide, a probe substrate of CYP2C9, and treatment with aprepitant for 3 d (125 mg on day 1, and 80 mg on days 2 and 3) decreased by 23, 27, and 15%, on days 4, 8, and 15, respectively, compared to a placebo.³⁾ In the present study, the mean values of PT-INR and WSI for two weeks after administration of aprepitant decreased significantly compared to the week before in patients receiving anticancer chemotherapy.

In contrast, the mean values of PT-INR and WSI one week after aprepitant administration increased significantly compared to those of the week before in patients receiving anticancer chemotherapy. Such increase was not observed in healthy subjects, but only in patients receiving anticancer chemotherapy. These results suggest the involvement of other factors apart from the interaction between aprepitant and warfarin. The type of cancer and cancer-associated bleeding, change of general condition and dietary intake after anticancer chemotherapy may affect the changes in the anticoagulation response to warfarin. During the cycles of our study, there was no observed case where dietary intake was dramatically decreased, or where PT-INR rose above 4 and caused severe bleeding episodes, either before or after the anticancer treatment.

Warfarin and aprepitant were co-administered with a number of other drugs as part of the present study, namely anticancer and antiemetic drugs. Interactions have been reported between warfarin and some anticancer drugs, such as carboplatin-etoposide,¹³⁾ paclitaxel,¹⁴⁾ and cisplatin,¹⁵⁾ which have been shown to increase the PT-INR. We assessed the anticancer drug cases individually and investigated the relationship between increased PT-INR one week after administration of aprepitant and the anticancer drugs used in the present study. We could not identify an anticancer drug that particularly prolonged the PT-INR. However, it has been shown in previous reports that the anticancer drugs play a major role in the increase of PT-INR and WSI one week after administration of aprepitant, and further studies are necessary to confirm this phenomenon.

Dexamethasone is used for the prevention of acute and delayed CINV and may be co-administered with aprepitant, as was the case in 32 out of 34 cycles carried out during the present study. In a previous report, when dexamethasone (20 mg on day 1, and 8 mg on days 2–5), and aprepitant (125 mg on day 1, and 80 mg on days 2 and 3) were co-administered in healthy subjects, the AUC of dexamethasone increased by 217 and 220% on days 1 and 5, respectively, compared to placebo.¹⁶⁾ At high doses, dexamethasone can induce enzyme expression. In addition, reports exist on the possible interaction between dexamethasone and warfarin. Sellam et al. reported that dexamethasone potentiated the effect of warfarin and substantially elevated INR.¹⁷⁾ In the present study, the mean dose of dexamethasone per cycle was approximately 30 mg, which is much lower than in the report mentioned above. Although we cannot rule out the possibility that dexamethasone played a role in the change in INR, the likelihood that it was involved is low.

In one of the cycles of our study, erythromycin (1200 mg/d) was administered on days 11–13 after aprepitant administration and the PT-INR drastically increased three weeks later (Fig. 1F). There was a case reported in the literature of a patient whose PT-INR increased after concurrent use of erythromycin and warfarin.¹⁸⁾ This suggests that the increase of PT-INR was caused by the interaction of erythromycin and warfarin. Apart from the anticancer drugs and erythromycin, no other drug caused any drastic change in PT-INR in this study (Table 3).

Our study has several limitations. First, the sample size was very small in the present study. Secondly, the patients suffered from various cancers and were administered different chemotherapeutic agents. Therefore, it is difficult to discuss whether the anticancer drugs play a role in the increase of WSI. It would be necessary to assess the interaction between warfarin and each anticancer drug individually to confirm this. Thirdly, there is no data about warfarin concentration in this study. Therefore, it is not possible to know the relationship between warfarin concentration and anticoagulation effect in patients receiving anticancer chemotherapy.

CONCLUSION

Our results indicate that, although the mean value of WSI significantly increased one week after aprepitant administration compared to one week before the administration, it in turn significantly decreased two weeks after compared to one week before.

In patients taking warfarin, PT-INR should be carefully monitored at least for two weeks after the beginning of aprepitant administration because it may fluctuate with both aprepitant and chemotherapy during this period.

It is necessary to further elucidate the interaction between aprepitant and warfarin.

Acknowledgments

This work was supported in part by a Grant-in-Aid for Young Scientists from the Ministry of Education, Culture, Sports, Science and Technology of Japan. We would like to thank Editage (www.editage.jp) for English language editing.

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Hesketh PJ, Grunberg SM, Gralla RJ, Warr DG, Roila F, de Wit R, Chawla SP, Carides AD, Ianus J, Elmer ME, Evans JK, Beck K, Reines S, Horgan KJ, Aprepitant Protocol 052 Study Group. The oral neurokinin-1 antagonist aprepitant for the prevention of chemotherapy-induced nausea and vomiting: a multinational, randomized, double-blind, placebo-controlled trial in patients receiving high-dose cisplatin—the Aprepitant Protocol 052 Study Group. J. Clin. Oncol., 21, 4112–4119 (2003).
• 2) Depré M, Van Hecken A, Oeyen M, De Lepeleire I, Laethem T, Rothenberg P, Petty KJ, Majumdar A, Crumley T, Panebianco D, Bergman A, de Hoon JN. Effect of aprepitant on the pharmacokinetics and pharmacodynamics of warfarin. Eur. J. Clin. Pharmacol., 61, 341–346 (2005).
• 3) Shadle CR, Lee Y, Majumdar AK, Petty KJ, Gargano C, Bradstreet TE, Evans JK, Blum RA. Evaluation of potential inductive effects of aprepitant on cytochrome P450 3A4 and 2C9 activity. J. Clin. Pharmacol., 44, 215–223 (2004).
• 4) Hirsh J. Oral anticoagulant drugs. N. Engl. J. Med., 324, 1865–1875 (1991).
• 5) O’Reilly RA. Studies on the optical enantiomorphs of warfarin in man. Clin. Pharmacol. Ther., 16, 348–354 (1974).
• 6) Kaminsky LS, Zhang ZY. Human P450 metabolism of warfarin. Pharmacol. Ther., 73, 67–74 (1997).
• 7) Ohno Y, Yamada M, Yamaguchi R, Hisaka A, Suzuki H. Persistent drug interaction between aprepitant and warfarin in patients receiving anticancer chemotherapy. Int. J. Clin. Pharm., 36, 1134–1137 (2014).
• 8) Hisaka A, Ohno Y, Yamamoto T, Suzuki H. Prediction of pharmacokinetic drug–drug interaction caused by changes in cytochrome P450 activity using in vivo information. Pharmacol. Ther., 125, 230–248 (2010).
• 9) Halkin H, Shapiro J, Kurnik D, Loebstein R, Shalev V, Kokia E. Increased warfarin doses and decreased international normalized ratio response after nationwide generic switching. Clin. Pharmacol. Ther., 74, 215–221 (2003).
• 10) Kovac M, Mitic G, Kovac Z. Miconazole and nystatin used as topical antifungal drugs interact equally strongly with warfarin. J. Clin. Pharm. Ther., 37, 45–48 (2012).
• 11) Romankiewicz JA, Ehrman M. Rifampin and warfarin: a drug interaction. Ann. Intern. Med., 82, 224–225 (1975).
• 12) Casner PR. Inability to attain oral anticoagulation: warfarin-rifampin interaction revisited. South. Med. J., 89, 1200–1203 (1996).
• 13) Le AT, Hasson NK, Lum BL. Enhancement of warfarin response in a patient receiving etoposide and carboplatin chemotherapy. Ann. Pharmacother., 31, 1006–1008 (1997).
• 14) Thompson ME, Highley MS. Interaction between paclitaxel and warfarin. Ann. Oncol., 14, 500 (2003).
• 15) Yano R, Kurokawa T, Tsuyoshi H, Shinagawa A, Sawamura Y, Matsunaga A, Nakamura T, Yoshida Y, Yoneda M, Kotsuji F, Masada M. Transient elevation of international normalized ratio during cisplatin-based chemotherapy in patients who are taking warfarin. Ann. Pharmacother., 45, e55 (2011).
• 16) McCrea JB, Majumdar AK, Goldberg MR, Iwamoto M, Gargano C, Panebianco DL, Hesney M, Lines CR, Petty KJ, Deutsch PJ, Murphy MG, Gottesdiener KM, Goldwater DR, Blum RA. Effects of the neurokinin1 receptor antagonist aprepitant on the pharmacokinetics of dexamethasone and methylprednisolone. Clin. Pharmacol. Ther., 74, 17–24 (2003).
• 17) Sellam J, Costedoat-Chalumeau N, Amoura Z, Aymard G, Choquet S, Trad S, Vignes BL, Hulot JS, Berenbaum F, Lechat P, Cacoub P, Ankri A, Mariette X, Leblond V, Piette JC. Potentiation of fluindione or warfarin by dexamethasone in multiple myeloma and AL amyloidosis. Joint Bone Spine, 74, 446–452 (2007).
• 18) Hassell D, Utt JK. Suspected interaction: warfarin and erythromycin. South. Med. J., 78, 1015–1016 (1985).