Introduction:
The choice of anticoagulant for patients with non-valvular atrial fibrillation (AF) in the setting of active cancer has not been well studied given only a minority of patients with cancer included in clinical trials.
Hypothesis:
Cancer patients will have similar rates of stroke (CVA), gastrointestinal bleeding (GIB), and intracranial hemorrhage (ICH) when treated with direct oral anticoagulants (DOAC) compared to warfarin (VKA) for AF.
Methods:
A retrospective study was performed at MD Anderson Cancer Center on patients with cancer and AF being treated with VKA or DOAC. The outcomes assessed were overall mortality, CVA, GIB, or ICH and patients were censored at 5 years. 1:1 propensity score matching was conducted to select comparable patients receiving VKA versus DOAC. Multivariable Cox proportional hazards regression models were fitted for each outcome.
Results:
A total of 1133 patients were included in the dataset with mean age of 72 years, 42% women, and 74% received DOAC (57% received apixaban). At 5 year follow up, mortality occurred in 145, CVA in 59, GIB in 59, and ICH in 6 patients. The baseline risk scores were higher for those on VKA compared to DOAC (CHADSVASC 3.5±1.6 vs 3.3±1.7 p=0.04, HASBLED 2.0 ± 1.1 vs 1.8 ± 1.0 p<0.01). After propensity score matching 210 patients were included in each anticoagulant group. Survival analysis showed higher GIB free survival in the DOAC group however no difference was observed for composite GIB, CVA, or ICH (Figure). On multivariate analysis the risk of GIB was higher in the VKA group (HR 3.1, 95% CI 1.3-7.3); however, risk of mortality (HR 1.1, 95% CI 0.7-1.8), CVA (HR 0.9, 95% CI 0.4-1.8), ICH (HR 0.3, 95% CI 0.0-2.2), and composite of GIB, CVA, or ICH (HR 1.3, 95% CI 0.8-2.3) were not different comparing VKA to the DOAC propensity matched group.
Conclusion:
Patients with active cancer have higher risk of GIB but similar overall survival and risk for composite stroke and bleeding when treated with VKA compared to DOAC for AF.
