Rethinking the approach to thrombosis in patients with cancer

The association between cancer and thrombosis has been known for at least a century. During the majority of this time, however, the clinical impact of this association has been limited to primarily an awareness of the risk of symptomatic deep vein thrombosis (DVT) or pulmonary embolism (PE). ¹ This approach started to change with a substantial paradigm shift occurring in the early 2000s when CLOT, a randomized trial of treatment of cancer-associated venous thromboembolism (VTE), was published evaluating dalteparin against the prior standard, vitamin K antagonists. ² This study was important for two reasons. First, from a practical clinical standpoint, the CLOT trial established that low-molecular-weight heparin (LMWH) monotherapy was the standard of care treatment of cancer-associated VTE, at least for the first 6 months after diagnosis. Second, and more importantly, CLOT established the concept that cancer-associated VTE was a disease that deserved study on its own – that cancer-associated VTE had a different clinical presentation and a greater risk of recurrent VTE as well as a greater risk of bleeding and now required a different treatment.

The publication of the CLOT study and the subsequent regulatory approval of dalteparin for the specific indication of cancer-associated VTE led to a renewed interest in this illness and a flurry of research activity encompassing multiple aspects: mechanisms, risk factors and epidemiology, impact on cancer outcomes, biomarker development, primary prevention, and new therapeutics. ³ Although several research advances were made as a result of these investigations, clinically another period of stagnation resulted with no substantial changes in bedside or outpatient practice. Clinicians utilized antithrombotic approaches in cancer for the following reasons: inpatient or postsurgical thromboprophylaxis and LMWH monotherapy or warfarin for treatment. The prevalence of cancer-associated VTE continued to rise, and it became clear that treatment pattern changes in cancer medicine – from chemotherapy to targeted agents and immunotherapy – would not reduce the burden of this complication. Indeed, several newer anticancer systemic therapeutic agents, including multitargeted tyrosine kinase inhibitors, antiepidermal growth factor antibodies, cyclin-dependent kinase inhibitors, and even immunotherapy agents, continued to be associated with high rates of cancer-associated thrombosis.^4,5

The 2000s have not been wasted years, however. Two streams of research – first, a better understanding of epidemiology cancer-associated thrombosis, and second, the advent of direct oral anticoagulants (DOACs) – have now brought us to a second paradigm shift in this field. Unlike the paradigm shift following the publication of CLOT, current research initiatives target not just treatment but also primary prevention efforts. The first wave of studies focused on expanding treatment options for cancer-associated VTE. The Hokusai VTE Cancer and the Select-D studies both randomized patients with active malignancy and confirmed diagnosis of VTE to a DOAC (edoxaban and rivaroxaban, respectively) or dalteparin.^6,7 In Hokusai, the primary outcome was a combination of recurrent VTE or major bleeding in 12 months, which was seen in 12.8% of patients randomized to the edoxaban arm and 13.5% in the dalteparin arm (hazard ratio (HR) 0.97; 95% CI 0.70–1.36; p = 0.006 for noninferiority). ⁶ Similarly, albeit on a smaller scale in Select-D, rivaroxaban resulted in a reduced 6-month VTE recurrence rate (4% vs 11%, 95% CI 7–16% vs 2–9%) at 6 months. ⁷ In both studies, increased bleeding was observed, primarily among those with gastrointestinal cancers. Although caution must be exercised in selecting appropriate patients, overall the news for patients with cancer who are suffering from VTE is a win–win: with proper selection (as outlined in various guidelines ⁸ ), over 90% of patients with cancer will not suffer from either recurrent VTE or major bleeding. This is an important clinical advance.

A second and even more recent wave of studies has the potential to impact a much larger number of patients with cancer. Primary prevention of thrombotic events in patients with cancer has been evaluated in studies for nearly two decades. Although some of the initial larger trials showed a statistically significant reduction in thrombosis, event rates were low and daily self-injection with LMWHs was considered a barrier to implementation. Thus, the results did not impact clinical practice. In 2019, two simultaneously published studies challenged this status quo. CASSINI and AVERT both targeted primarily patients with solid tumors and at intermediate-to-high risk of VTE (Khorana score ⩾ 2).^9,10 Both studies used oral agents (rivaroxaban and apixaban, respectively). Together, the studies demonstrated that this population was at high risk for thrombotic events (approximately 10% DVT or PE over 6 months, with additional events including arterial occurring in patients) and that this risk could be substantially diminished with DOAC prophylaxis.^9,10 This benefit was accompanied by a 1% increase in absolute risk of major bleeding (number needed to harm, approximately 100). These findings have led to a reassessment of the current standard of care and three updated societal guidelines (American Society of Clinical Oncology (ASCO), International Society on Thrombosis and Haemostasis (ISTH), International Initiative on Thrombosis and Cancer (ITAC)) now recommend primary prevention in at-risk patients, particularly highlighting patients with pancreas cancer and/or a Khorana score ⩾ 2. ¹¹

Given these rapidly emerging data, how are clinicians to approach cancer-associated thrombosis? First, awareness that the risk persists even with newer anticancer agents should remain high and clinicians should ensure patients with cancer are educated about the warning signs and symptoms. Second, risk assessment to identify high-risk patients is vital both for increased awareness and for selection for prophylaxis approaches. An interesting potential alternative is to consider screening high-risk patients for VTE: such an approach has been shown to identify a high number of patients with pre-existing DVT, possibly reducing hospitalizations and urgent visits.^9,12 Third, clinicians should be aware of the expanding number of treatment options for their patients with cancer: the ability to individualize treatment is a major advance that has occurred in these past few years.

Although much has been learned, much remains to be fully understood. Despite a century of knowing the association with cancer, the pathophysiology of thrombosis in this setting is still incompletely understood. Clinical studies have focused primarily on venous events in this setting; arterial events, including stroke and myocardial infarction, are also distressingly prevalent and deserving of further study. The implications of visceral vein thrombi (e.g. portal or mesenteric vein thrombosis) – generally incidentally discovered on imaging for cancer – are unknown, as is the need for treating these. Implementation science in this setting is still in its infancy, and system-based approaches are sorely needed.

Altogether, however, a great deal of progress has been made in the past few years, including work that has a direct clinical impact on current patients. We are finally at the cusp of reducing the public health burden of VTE in cancer with the advent of new primary and secondary prevention approaches. This indeed is cause for celebration.