Cancer Diagnoses Cluster in Bleeders on Anticoagulation Therapy

Cancer diagnoses tended to arise more in patients who bled on oral anticoagulants (OACs), a Canadian population-based cohort study found.

In Ontario, people with atrial fibrillation (Afib) on OAC therapy had a 21.8% incidence of bleeding within the subsequent 2 years, while 4.9% were diagnosed with a malignancy during that window. The incident malignancy was more likely in patients who had documented bleeding (HR 4.0, 95% CI 3.8-4.3), reported Husam Abdel-Qadir, MD, PhD, of Women’s College Hospital in Toronto, and colleagues in Circulation.

The strength of this finding varied depending on the site of malignancy and bleeding:

• Any cancer with gastrointestinal bleeding (adjusted HR 5.0, 95% CI 4.6-5.5)
• Cancer at the site of gastrointestinal bleeding (adjusted HR 15.3, 95% CI 13.2-17.7)
• Any cancer with genitourinary bleeding (adjusted HR 5.0, 95% CI 4.4-5.7)
• Cancer at the site of genitourinary bleeding (adjusted HR 11.8, 95% CI 10.1-13.9)
• Any cancer with respiratory bleeding (adjusted HR 4.0, 95% CI 3.5-4.6)
• Cancer at the site of respiratory bleeding (adjusted HR 10.1, 95% CI 8.1-12.5)

“These findings should promote education that timely investigations for malignancy should be considered for people with bleeding after anticoagulation, as they may offer an opportunity for cancer diagnosis at an earlier stage, which may increase the likelihood of cure,” wrote Abdel-Qadir and colleagues. “Appropriate investigations for a source of bleeding may also increase confidence in resuming anticoagulation if the underlying cause is addressed, or no serious pathology is uncovered.”

Weaker associations were made between cancer diagnosis and intracranial (adjusted HR 1.8, 95% CI 1.4-2.2) and nasopharyngeal bleeds (adjusted HR 1.5, 95% CI 1.2-2.0), and these relationships were not significant at the level of bleeding site-specific cancer. Breast cancer did not show any association with bleeding in OAC users.

OAC is the cornerstone therapy for stroke prevention in people with Afib; an increased risk of bleeding is considered an expected tradeoff for some people.

While previous reports have linked Afib to a higher incidence of cancer diagnosis, causation has not been established due to the possibility that antithrombotic therapy like OAC can unmask the bleeding from an occult malignancy.

“Nonetheless, patients and healthcare providers may attribute bleeding to the medication rather than intrinsic pathology, the differential diagnosis of which includes an underlying malignancy,” Abdel-Qadir’s team wrote.

The present report supports prior work associating OAC bleeding and new cancer diagnosis, the authors said, adding that it also expands the concept to the more contemporary direct oral anticoagulants (DOACs) used in Afib.

“This changes the message in a manner that encourages clinical action as it may allow for cancer diagnosis at a curable stage, rather than simply asking a clinician to become the bearer of bad news,” they wrote.

For their study, Abdel-Qadir and colleagues had relied on various datasets covering medical, prescription, and neighborhood data of Ontario residents. The study targeted people, 66 years or older, who newly initiated warfarin or DOACs after Afib diagnosis from 2008-2022.

The study cohort included 119,480 OAC users (median age 77 years, 52% men). The majority had first gotten a DOAC (69.3%) rather than a vitamin K antagonist. At 2 years, 73.0% still persisted with OAC therapy.

Notably, only 52.2% of those diagnosed with malignancy had cancer staging data. From a hypothesis-generating analysis of that limited information, it appeared that more cancers were diagnosed at an earlier stage after bleeding (27.6% stage 4 after bleeding vs 31.3% without bleeding, P=0.029).

Many patients hadn’t had a documented endoscopic evaluation after bleeding, however. “This is an actionable message not communicated in previous research. It is possible that physicians may have recommended bleeding investigations, but patients did not proceed with them. It is also possible that some of these bleeds were appropriately determined not to warrant investigation (e.g., traumatic bleeds or hemorrhoids) or occurred in patients with previously documented pathology,” Abdel-Qadir’s team wrote.

The observational design of the study was chief among its limitations. It was also possible that some patients were taking aspirin or dual antiplatelet therapy, in addition to OAC, that may have increased their risk of bleeding.

Finally, study authors noted that their study excluded patients with valvular disease, chronic dialysis, venous thromboembolism, previous cancer, or previously documented bleeding.

Please enable JavaScript to view the

comments powered by Disqus.