Capecitabine and Warfarin: Interaction Details

a) In a clinical study in four patients with cancer, chronic administration of capecitabine 1250 mg/m(2) twice daily with a single dose of warfarin 20 mg increased the mean AUC of S-warfarin by 57% and decreased its clearance by 37%. Baseline corrected AUC of INR in these 4 patients increased by 2.8-fold and the maximum observed mean INR value was increased by 91% ¹.

b) Altered coagulation parameters and/or bleeding, including death, have been reported in patients receiving capecitabine, tegafur, or fluorouracil concurrently with coumarin-derivative anticoagulants ².

c) A study of four patients with advanced/metastatic breast or colon cancer demonstrated a significant increase of S-warfarin exposure with the coadministration of capecitabine. Four patients received a single 20 mg warfarin dose on day 1, 8 days prior to the start of capecitabine treatment. From days 8 through 70 the subjects received three cycles of 1250 mg/m2 capecitabine orally twice daily for 14 days followed by a 7 day rest period. On day 61 (day 12 of the third cycle) the patients received a second single 20 mg warfarin dose. The pharmacokinetics (PK) and pharmacodynamics (PD) parameters of warfarin were assessed at 30 minutes predose then 1,2,4,8,12,24,32,48,72,96, and 144 hours after dose administration on both day 1 and day 61. Comparison of PK and PD parameters of warfarin alone to warfarin plus capecitabine revealed an increase in the mean AUC of S-warfarin by 57% (90% CI, 32% to 88%), a 51% (90% CI, 32% to 74%) prolongation of the elimination half-life, and a decrease in clearance of S-warfarin by 36% (90% CI, 24% to 47%). The baseline corrected AUC of INR in these subjects increased by 2.8-fold, and the maximum observed mean INR value increased by 91%. Exposure to R-warfarin was not significantly affected by the coadministration of capecitabine. There was some suggestion of increased factor VII activity, as well, but the difference did not reach statistical significance (P=0.1204) ³.

d) A 59-year-old male patient on chronic warfarin therapy for a mechanical mitral valve replacement had a greater than 85% daily dose reduction of warfarin over the course of three cycles of capecitabine/irinotecan therapy. This patient was diagnosed with stage IV colon cancer with hepatic metastasis. Previous therapy included fluorouracil/leucovorin and the FOLFOX 6 regimen. With a target INR of 2.5 - 3.5, the patient was taking 16% less warfarin at the end of the first cycle of capecitabine/irinotecan. Subsequent cycles required additional warfarin dose reduction of up to 58% with no other confounding factors such as additional drug interactions, vomiting or weight loss identified. Liver impairment could have been considered as a cause for decreased warfarin metabolism, however, when the capecitabine/irinotecan therapy was discontinued increased warfarin dosing was required to maintain the desired INR ⁴.

e) This report presents two case studies illustrating the significant coagulopathy and resulting bleeding episodes manifested with the coadministration of warfarin and capecitabine. The first case presents a 91-year-old woman diagnosed with adenocarcinoma of the rectum with metastases to the lungs and liver. She received capecitabine 2000 mg/m(2) per day in 2 divided doses for 2 weeks followed by one week of rest. After 4 cycles a CT scan revealed a transluminal thrombus in the left femoral vein. Warfarin 2.5 mg per day was initiated to maintain an INR of 2 - 2.5. After 2 additional cycles of capecitabine she was admitted to the hospital with rectal and vaginal bleeding. She had a PT of 72.9 seconds and an INR greater than 10. Both warfarin and capecitabine were discontinued immediately. Forty-eight hours after presentation her PT decreased to 15.6 seconds with an INR of 1.65. The patient did not receive any additional cancer therapy and her coagulation parameters remained in therapeutic range since discharge ⁵.

f) The second case presented in this report involves a 72-year-old female patient with recurrent metastatic breast cancer. Later she was diagnosed with pulmonary embolism for which she received warfarin 2.5 mg per day to maintain an INR of 2 - 3. After an 8 month progression-free interval, a bone scan revealed progressive metastatic disease. She began capecitabine 2000 mg/m2 per day in 2 divided doses for 14 days followed by 7 days rest. After 2 cycles she was admitted to the hospital with an H/H of 5.6 g/dL/16%, a PT greater than 100, and an INR greater than 10. Her warfarin, capecitabine and ibuprofen were discontinued immediately. She received 4 units fresh frozen plasma, 4 units packed red blood cells, and 10 mg vitamin K. Forty-eight hours after admission her vital signs were normal, the INR was 2.2 and her stool was guaiac negative. She did not resume chemotherapy. She did resume warfarin therapy at 2.5 mg per day and her coagulation parameters remained in normal range ⁵.

g) A 58-year-old male patient on warfarin 1 mg/day to prevent catheter-associated thrombosis experienced asymptomatic elevated PT after initiation of fluorouracil and vinBLAStine. His PT prior to chemotherapy was 11.6 sec. He received intravenous vinBLAStine 2 mg/day and intravenous fluorouracil 800 mg/day for 5 consecutive days at monthly intervals for the treatment of small cell lung cancer. PTs were not measured with each cycle; however, 16 days after the initiation of the last cycle, the PT was elevated at 62.3 sec. The following day the PT decreased to 48.8 sec. Warfarin was stopped and intravenous phytonadione 10 mg was administered on the next day. The PT continued to decrease to 19.6, 16.8 and 14.1 sec over a week. The patient remained well for several months but subsequently died from brain metastases ⁶.

h) After starting warfarin, a 70-year-old female patient treated with cycloPHOSphamide, methotrexate, and fluorouracil for breast cancer experienced elevated PT by day 15 on 4 chemotherapy cycles; furthermore, she experienced gross hematuria on the first occasion and epistaxis on each occasion. Her PT ranged from 8.2 to 21 sec with a stable warfarin dose. The PT increased to 44.2, 39, 31, and 29 sec, respectively, by day 15 of each subsequent cycle of chemotherapy. The PT returned to pretreatment levels within 48 to 72 hours. Epistaxis and hematuria responded to discontinuation and reduction of warfarin dosage ⁷.

i) After starting warfarin, a 57-year-old female patient treated with cycloPHOSphamide, methotrexate, and fluorouracil for breast cancer experienced elevated PT and microscopic hematuria by day 15 of each cycle. She also experienced epistaxis on 1 occasion. Her PT ranged from 14.8 to 17.2 sec over 3 weeks. The PT increased to a range of 24.6 to 26 sec by day 15 of each subsequent cycle of chemotherapy. The PT returned to pretreatment levels within 48 hours ⁷.

j) A case was reported where a 73-year-old male patient with prosthetic mitral and aortic valve replacement was stabilized on warfarin 22.5 mg weekly with a mean INR of 2.43. After colon resection for colon cancer, he received maintenance doses of levamisole (50 mg every 8 hours for 3 days every other week) and fluorouracil (450 mg/m(2)/week). Within 4 weeks of beginning antineoplastic therapy, his INR increased from 3.04 to 39.56; thereafter, his warfarin was reduced to 7.5 mg weekly. With a decline in neutrophil count, he discontinued the levamisole-fluorouracil for 5 weeks. During this period, his INR dropped below therapeutic levels. When he was restarted on levamisole-fluorouracil, he again experienced a significant rise in PT ⁸.

k) Five patients received warfarin and fluorouracil for a period of 3 years. Before fluorouracil, the weekly mean dose of warfarin was 40.66 mg and with fluorouracil the mean was 24 mg. As an average, all patients required a warfarin dose reduction of 44% ⁹.

l) Concomitant use of warfarin and fluorouracil resulted in fluctuating INR in a 60-year-old woman. The patient had colon cancer and received a chemotherapy regimen consisting of 6 cycles of fluorouracil and leucovorin, given for 5 consecutive days during a month. The patient was diagnosed with deep venous thrombosis in the left leg. Warfarin was initiated. Her baseline INR was 1.1. The INR rose above the targeted INR range of 2 to 3 three to four weeks after the last fluorouracil and leucovorin dose of each chemotherapy cycle. The INR peaked at 4.2 three weeks after the last cycle and the total weekly warfarin dose was reduced from 9 mg/week to 7 mg/week. The INR dropped to 1.58 four weeks after the completion of the final cycle of chemotherapy . In order to maintain an INR of 2 to 3, the weekly warfarin dose needed to be doubled from 7 mg/week to 14 mg/week. Fluorouracil may inhibit the synthesis of cytochrome P450 2C9 and thus inhibit the metabolism of warfarin. Close monitoring of the INR is recommended when warfarin is added to fluorouracil therapy. Warfarin dosage may need to be reduced by 20 to 70% during concomitant therapy with fluorouracil and then increased within 30 days after discontinuation of fluorouracil to maintain the INR within the desired range ¹⁰.

m) Continuous-infusion fluorouracil)-based chemotherapy and minidose warfarin (1 mg/day) has been shown to increase the incidence of INR abnormalities. Ninety-five cancer patients treated with continuous-infusion fluorouracil-based chemotherapy were evaluated. Four-hundred eighty eight INR determinations were performed, 50 of which (10%) showed an INR of more than 1.5. Thirty-one patients (33%) had a significant elevation in their INR levels. Of these, 6 patients had minimal INR elevations (between 1.18 and 1.5). Bleeding was observed in 8 patients (INR 1.34 to 8.8). Epistaxis with no gross blood loss was observed in 6 patients, and hematuria was observed in 2 patients. The proportion distribution of increased INR was the following: the INR became elevated in 12 (57%) of 21 patients treated with the FOLFOX regimen, 11 (27%) of 40 patients treated with the de Gramont regimen, and 5 (26%) of 19 patients treated with the FOLFIRI regimen. Analysis of 80 patients treated with these regimens showed a statistically significant association between INR elevation and FOLFOX treatment (p = 0.041). This study demonstrated that coadministration of warfarin and fluorouracil resulted in an INR elevation in 33% of patients and that of the 8 patients who developed bleeding problems, 7 had an elevated INR. No relationship was observed between liver metastasis, hepatic function, age, performance status, number of previous chemotherapy regimens administered or chemotherapy toxicity or INR alteration or bleeding. Clinicians should regularly monitor the PT in patients receiving warfarin and fluorouracil ¹¹.

n) Concomitant administration of multiple doses of capecitabine (1000 mg/m(2) twice daily for 14 days) increased celecoxib (CYP2D6 substrate) AUC by 28%, Cmax by 24% and Ctrough by 30% ¹.