Does Warfarin Prevent Catheter-related Thrombosis in Cancer Patients? Yes and No.

Lindsey R. Perez, Pharm.D. Candidate, The University of Texas at Austin
ClotCare Editorial Review by Henry I. Bussey, Pharm.D., FCCP, FAHA
May, 2009

Reference: Young AM, Billingham LJ, Begum G, et al. Warfarin thromboprophylaxis in cancer patients with central venous catheters (WARP): an open-label randomized trial. Lancet. 2009 Feb 14;373(9663):567-74.

Although widely touted as showing that warfarin does not protect against catheter-related thrombosis in cancer patients, the results indicate that adjusted-dose warfarin may be effective.

The study design was a little unusual in that patients were categorized by their physicians' views. Clinicians who believed warfarin to be effective assigned patients to two different doses of warfarin ("Dose-Evaluation Arm") while clinicians who were uncertain of the effect of warfarin randomized patients to no warfarin vs. warfarin ("Warfarin-Evaluation Arm"). Within the "Warfarin-Evaluation Arm", physicians willing to manage the warfarin dosing could randomize their patients to no warfarin, fixed-dose warfarin, or adjusted-dose warfarin. The target of adjusted-dose warfarin was an INR target of 1.5 to 2. The study enrolled 1590 cancer patients (16 years of age or older) who required a central venous catheter (CVC) in order to receive chemotherapy (Table 1).

Note: See the first paragraph for an explanation of how patients were randomized to the various groups.

The numbers in the results (Table 2) appear confusing at first but that is because of some randomization issues. Patients who received warfarin in the "Dose-Evaluation" arm were excluded from the warfarin vs no warfarin analysis because they were not eligible to be randomized to no warfarin. Similarly, those who received fixed-dose warfarin in the "Two-way" arm of the "Warfarin-evaluation" arm were excluded from the "fixed vs adjusted dose" warfarin analysis because they were not eligible to be randomized to adjusted dose warfarin. The results of this open-label randomized trial conducted in the United Kingdom showed that there was no difference in radiologically confirmed, symptomatic catheter-related thrombotic events in the warfarin vs no warfarin analysis. But dose-adjusted warfarin did reduce the rates compared to fixed-dose warfarin (3% vs 7%). Major bleeding events (intracranial, retroperitoneal, those requiring transfusion or hospitalization, or those leading to death) were not statistically significant in any comparison. Additionally, no survival benefit was found in any comparison.

Additional considerations:

• Late Start: While the study protocol included initiation of warfarin 3 days prior to CVC insertion in order to allow for warfarin accumulation, only 14% (n=155) of the warfarin treated patients adhered to the protocol.
  
  
• Early Stop: Of the treatment patients, 9% (n=106) discontinued warfarin early (more than 7 days before the catheter was removed).
  
  
• INR Control: Difficulties and variability in INR monitoring were reported as study complications. INR measurements analyzed from one study site showed a median INR of 1.69 in the 56 patients assigned to the dose-adjusted warfarin arm.

Previously conducted studies have shown 1 mg fixed-dose warfarin to be ineffective for thromboprophylaxis in this patient population. While the WARP trial results support these findings, the study also suggests that adjusted dose warfarin is effective in reducing cather-related thrombosis. Further studies are warranted to determine if dose-adjusted warfarin with a higher target INR and/or better INR may be more effective and/or safer in preventing catheter-related thrombosis.