Anticoagulation for the treatment of immune checkpoint inhibitor–associated venous thromboembolism

TO THE EDITOR:

Immune checkpoint inhibitors (ICIs) are associated with venous thromboembolic events (VTE) that carry significant morbidity and mortality.¹ Current guidelines recommend the use of direct oral anticoagulants (DOACs) or low molecular weight heparins (LMWHs) for the treatment of cancer-associated VTE, based on data in patients receiving conventional chemotherapy.² Evidence from preclinical studies suggest that DOACs may have a synergistic effect with ICIs.^3,4 For example, in mouse models of cancer, rivaroxaban has been shown to enhance the infiltration of dendritic cells and cytotoxic T cells in the tumor microenvironment.³ In contrast, heparins may increase antithrombin activities and enhance the response to ICI therapy.⁵ Given these observations, we performed a cohort study to compare the safety and efficacy of DOACs and LMWHs for the treatment of ICI-associated VTE.

We performed a propensity score–matched cohort study using the TriNetX Analytics Network database, a multicenter research network comprising deidentified data from electronic health records from over 70 participating health care institutions.⁶ This study has been deemed exempt by the National Cheng Kung University Hospital Institutional Review Board. We identified adult (age ≥18) cancer patients who received ICIs between March 2011 and March 2023 (supplemental Table 1). We defined ICI-associated VTE as pulmonary embolism (PE) or deep venous thrombosis (DVT) that occurred within 6 months of ICI initiation.⁷ The experimental group included patients who received a DOAC while the control group included patients who received a LMWH within 1 month after VTE. We excluded patients who received both DOACs and LMWHs or patients who received warfarin. We also excluded patients who had PE, DVT, or atrial fibrillation/flutter before ICI therapy. The index date was the start of anticoagulation after VTE occurrence. We defined the primary outcome as 12-month all-cause mortality and the secondary outcomes as intracranial bleeding, gastrointestinal (GI) bleeding, and recurrent VTE including PE and DVT within 1 year of anticoagulation (supplemental Table 1). We determined recurrent VTE as an event that occurred after 2 weeks of being started on anticoagulation for acute treatment of VTE.⁸ This definition was based on the International Society on Thrombosis and Hemostasis (ISTH) recommendations that VTE recurrence is defined as VTE occurring after a successful (clear clinical improvement of patient symptoms and signs) acute (first 2 weeks) treatment.⁸ The cohorts were matched in a 1:1 ratio by incorporating relevant variables such as age, sex, site of cancer, the presence of metastatic disease, history of bleeding, and underlying comorbidities used in the Charlson Comorbidity Index (supplemental Table 2) using a greedy nearest-neighbor matching with a caliper of 0.1 of the pooled standard deviations. For each outcome, the hazard ratio (HR) with 95% confidence intervals (CIs) was calculated using the Cox proportional hazards model to estimate the risk of DOAC on the outcomes. All P values were 2-tailed and considered statistically significant if <.05. All analyses were conducted on the TriNetX platform.

We identified 1100 patients who developed ICI-associated VTE, among which 727 and 373 patients received a DOAC and LMWH, respectively (supplemental Figure 1). The median time from ICI initiation to VTE occurrence was 117 days (interquartile range, 57-198). Patients on a DOAC tended to be older (mean age, 66.6 ± 10.8 vs 61.7 ± 12.9 years) than those on a LMWH (Table 1). After propensity score matching, all covariates were well balanced between the 2 groups (Table 1; supplemental Table 3). Apixaban (68.8%) was the most commonly used DOAC, whereas enoxaparin (93.1%) was the most commonly used LMWH. The most common type of cancer was lung cancer (45%-47%). More than 65% of the patients had metastatic disease. Over a median follow-up period of 365 days, a total of 154 and 214 deaths occurred among the DOAC and LMWH cohorts, respectively. In a Cox proportional hazards analysis, patients on a DOAC had significantly lower risk of mortality than patients on a LMWH (HR, 0.73; 95% CI, 0.59-0.90) (Table 2). There were no significant differences in the risk of intracranial (HR, 0.64; 95% CI, 0.25-1.63), GI bleeding (HR, 0.79; 95% CI, 0.40-1.57) or recurrent VTE (HR, 1.15; 95% CI, 0.58-2.28) between the 2 cohorts.

In this cohort study, patients on a DOAC had a similar risk of bleeding, recurrent thromboembolism, and a lower rate of mortality than patients on a LMWH. These results are supportive with the hypothesis from preclinical data showing that DOACs promote antitumor immunity within the tumor microenvironment.⁴ Limitations of this study include potential unmeasured or unmatched confounders and the heterogeneity in our study population. Because this was a database-driven study, we had to exclude patients with a history of VTE prior to the initiation of ICIs as we were not able to accurately identify new VTE events that occurred with the use of ICIs. We recognize that patients with a history of VTE likely have a higher risk of ICI-associated VTE and that excluding this high risk subset of patients may potentially introduce selection bias and affect the observations in this study.¹ As we utilized the International Classification of Diseases-10 codes to identify outcomes, we could not exclude the possibility of misclassification of outcomes. We were unable to utilize the ISTH criteria for major bleeding as we did not have data regarding the change in hemoglobin levels or the amount of blood transfused as part of the ISTH criteria.⁹ We did use serious bleeding events such as intracranial or GI bleeding as outcomes as these could be accurately identified using International Classification of Diseases-10 codes.¹⁰ Of note, we excluded patients who received both DOAC and LMWH to better isolate the effects of DOAC vs LMWH in the analysis. Nevertheless, some patients might have been treated initially with LMWH while in the hospital and transitioned to DOAC after discharge. These patients could not be evaluated in this analysis and this might be a source of bias in the risk estimates. Prospective studies are needed to validate these results and further explore whether these observations are related to antineoplastic effect with DOACs and ICI therapy.

Contribution: Cho-Han Chiang contributed to the conception or design of the work and the acquisition, analysis, and interpretation of data for the work, and drafted the manuscript; S.O. contributed to the interpretation of data for the work and drafted the manuscript; Y.-C.C. contributed to the acquisition and analysis of data for the work; K.-Y.C. contributed to the acquisition and analysis of data for the work; Cho-Hung Chiang contributed to the acquisition and analysis of data for the work; Y.C. contributed to the acquisition of data for the work; and R.P. contributed to the conception or design of the work and interpretation of data for the work and critically revised the manuscript.

Conflict-of-interest disclosure: R.P. reports consulting fees from Merck Research, outside the submitted work. R.P. is partially funded through a career development award from the Conquer Cancer Foundation. The remaining authors declare no competing financial interests.

Correspondence: Rushad Patell, Division of Hemostasis and Thrombosis, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Ave, Boston MA 02215; email: rpatell@bidmc.harvard.edu.