Risk Assessment Models for Thrombosis and Anticoagulant-Related Bleeding in Ambulatory Cancer Patients

Abstract

Cancer patients have a high risk of developing venous thromboembolism and arterial thrombosis, along with an increased risk of anticoagulant-related bleeding with primary and secondary prophylaxis of cancer-associated thrombosis. Decisions on initiation, dosing, and duration of anticoagulant therapy for prevention and treatment of cancer-associated thrombosis are challenging, as clinicians have to balance patients' individual risk of (recurrent) thrombosis against the risk of bleeding complications. For this purpose, several dedicated risk assessment models for venous thromboembolism in cancer patients have been suggested. However, most of these scores perform poorly and have received limited to no validation. For bleeding and arterial thrombosis, no risk scores have been developed specifically for cancer patients, and treatment decisions remain based on clinical gestalt and rough and unstructured estimation of the risks. The aims of this review are to summarize the characteristics and performance of risk assessment scores for (recurrent) venous thromboembolism and discuss available data on risk assessment for bleeding and arterial thrombosis in the cancer population. This summary can help clinicians in daily practice to make a balanced decision when considering the use of risk assessment models for cancer-associated venous thromboembolism. Future research attempts should aim at improving risk assessment for arterial thrombosis and anticoagulant-related bleeding in cancer patients.

Unanswered questions in cancer‐associated thrombosis

“…If recurrent VTE occurs despite therapeutic anticoagulation in patients who are fully adherent to treatment, a careful assessment of bleeding risk should be conducted before selection of therapy for the breakthrough event. Bleeding‐risk factors include prior bleeding, renal insufficiency and metastatic or intracranial disease (see subsection ‘Risk of bleeding in cancer‐associated VTE’) 66,75 . Given that no validated risk assessment models are currently available to accurately and reproducibly quantify bleeding risk in these patients, clinicians will have to rely on their own judgement and experience prior to changing or escalating therapy.…”

“…Similarly, available risk assessment models are not reliable due to modest discrimination and limited external validation. 66,67 Permanent discontinuation of anticoagulation might be preferred in patients with: (a) severe persistent thrombocytopenia, prior history of major bleeding, or ongoing bleeding without curative options; or (b) absence of risk factors for VTE recurrence, including patients in complete oncologic remission (e.g. after curative treatment) not receiving ongoing anticancer therapies and patients with very low-risk cancer types (e.g.…”

“…Bleedingrisk factors include prior bleeding, renal insufficiency and metastatic or intracranial disease (see subsection 'Risk of bleeding in cancer-associated VTE'). 66,75 Given that no validated risk assessment models are currently available to accurately and reproducibly quantify bleeding risk in these patients, clinicians will have to rely on their own judgement and experience prior to changing or escalating therapy. In patients at high risk of anticoagulant-related bleeding, the decision on the need for, type and dose of anticoagulant therapy should be individualized.…”

“…While biomarkers, such as D‐dimer, 65 might help to further risk‐stratify, none validated. Similarly, available risk assessment models are not reliable due to modest discrimination and limited external validation 66,67 . Permanent discontinuation of anticoagulation might be preferred in patients with: (a) severe persistent thrombocytopenia, prior history of major bleeding, or ongoing bleeding without curative options; or (b) absence of risk factors for VTE recurrence, including patients in complete oncologic remission (e.g.…”

Recurrent Thrombosis and Bleeding in Patients with Cancer-Associated Venous Thromboembolism Receiving Anticoagulation: Are These Modifiable Risk Factors for Mortality?

“…Notwithstanding the short-term consequences on mortality (fatal pulmonary embolism or fatal bleeding), both recurrent VTE and major bleeding could worsen prognosis through various indirect mechanisms, including the interruption of essential anticancer treatment. 9 10 11 So, a better understanding of the complex relations among VTE recurrence, bleeding, and mortality could improve risk stratification and spur the development of novel approaches that have the potential to improve outcomes.…”