br See page for disclosure information
See page 786 for disclosure information.
NOACs remains unclear. We, therefore, investigated the relations between bleeding events and new-onset cancers in AF patients receiving NOACs in a prospective cohort. The aim was to determine whether bleeding would be a predic-tor of new-onset cancers after NOACs use.
Nonvalvular AF patients who received NOACs (dabigatran etexilate, rivaroxaban, apixaban, and edoxaban) treatment were enrolled at the cardiology and neurology clinics of Taipei Veterans General Hospital, Taipei, Taiwan from April 2013 to July 2017. All patients had approved indications for NOACs according to the guidelines.11,12 The clinical charac-teristics, concomitant medications, laboratory measurements, and history of cancer and treatment were recorded during enrollment. The time to initiate NOACs was recorded during the enrollment by the interview of the patient and medical record of drug prescription. The patients were prospectively followed every 3 to 6 months through their medical records, clinic visits, or phone interviews to determine the incidence of bleeding, stroke, and new-onset cancers. The bleeding sites and severity were evaluated by reviewing laboratory data or medical records.13,14 Ethical approval was granted by the Institutional Review Board of the Veterans General Hospital, Taipei, Taiwan. All subjects gave written informed consent.
New-onset cancer was defined as the diagnosis after the initiation of NOACs. Preexisting cancer was defined as
Arrhythmias & Conduction Disturbances/New-Onset Cancers and Nonvitamin K Anticoagulants 783
cancer that had been diagnosed before the initiation of NOACs. The clinical course after bleeding; the surveillance of cancer diagnosis, stage, and therapy (surgery, radiation, chemotherapy, hormone, target therapy); the interruption and reinitiation of NOACs; and the chronological order of bleeding and cancer diagnosis were explored in detail through medical records. The interval from the time point of first bleeding event after NOACs initiation to the diagno-sis of new-onset cancer was also explored.
We calculated the CHA2DS2-VASc and HAS-BLED scores at enrollment. The CHA2DS2-VASc risk score assigned 1 point for congestive HA-1077 failure, hypertension, diabetes, vascular disease, age 65 to 74 years, and female gender and 2 points for either age (75 years and older) or stroke.12 The HAS-BLED score is calculated by adding 1 point for hypertension, abnormal renal function, liver dis-ease, stroke, bleeding history, age (65 years and older), and either concomitant antiplatelet or nonsteroidal anti-inflam-matory drugs or excess alcohol consumption (>8 drinks per week). For previous warfarin users, a history of labile INRs (i.e., less than 60% of time in a therapeutic range) was scored as 1 point.12
Major bleeding episodes were defined as fatal bleeding, symptomatic bleeding in a critical area or organ, or bleed-ing causing a reduced hemoglobin level of 20 g/L according to the statement by the International Society on Thrombosis and Haemostasis.15
All continuous variables are expressed as the means § SDs. The 2-sample independent t test, 1-way ANOVA, and a chi-square test were used, as appropriate. Multivariate binary logistic regression analysis was used to compare the patients with and without new-onset cancer. Receiver
The baseline characteristics of the patients with and without cancer
operating characteristic (ROC) curve analysis was used to determine the cut-off point of the CHA2DS2-VASc score to predict new-onset cancers in patients with bleeding. Sta-tistical significance was established at p < 0.05. PSAW SPSS 22.0 was used for the statistical analysis.
We prospectively followed 395 patients who received NOACs for a mean duration of 2.8 § 1.5 years (0.5 to 4.7 years). There were 248 patients receiving dabigatran etexilate, 104 receiving rivaroxaban, 41 receiving apixaban, and 2 receiving edoxaban. A total of 18 (4.5%) patients were diagnosed with new-onset cancers after receiving NOACs (7 GI cancers, 6 urinary tract cancers, 4 lung can-cers, and 1 blood cancer). The mean duration from the initi-ation of NOACs to the diagnosis of new-onset cancer was 584 § 372 days (17 to 1314 days). There were 42 (10.6%) patients who had preexisting cancers before receiving NOACs. The patients with new-onset cancers tended to be older than those without new-onset cancers (Table 1). There were no differences in gender, body weight, co-morbidities, history of gastrointestinal bleeding, alcohol, or concomitant medications between the patients with and without cancers. Liver and cardiac function did not differ between the groups. The patients with cancers (either preexisting or new-onset) had higher levels of creatinine. The levels of PT and APTT did not differ between the 3 groups. The patients with new-onset cancers had higher HAS-BLED scores than those without new-onset cancers (Figure 1). There was a marginal trend of increased CHA2DS2-VASc scores in the patients with new-onset cancers (Figure 1).