Finding a risk assessment model for thromboembolic events in hospitalized cancer patients: The Indicate Study.

Abstract

e24108 Background: Hospitalized cancer patients are at increased risk for Thromboembolic Events (TEs). Given the hemorrhagic risk associated with untailored thromboprophylaxis, the identification of patients at low TE risk who might not receive it during in-hospital stay would be clinically useful. Methods: The INDICATE study was a monocentric, observational study enrolling patients with active solid malignancy hospitalized for at least two nights for treatment administration, diagnostic procedures or acute medical illness (excluding TEs). The primary objective was to assess the Negative Predictive Value (NPV) of low-grade Khorana Score (e.g. KS = 0), evaluated at the time of patients’ in-hospital admission, for TEs prediction during and after (next 45 days) hospitalization. The primary analysis was conducted on patients with KS = 0. However, all-grade KS patients were enrolled for secondary outcomes analysis. Assuming a 7% of TEs as the unsatisfactory percentage and a 3% as the satisfactory percentage, expecting about 5% of collected data as incomplete, all consecutive patients who fulfil the inclusion criteria irrespective of the KS were enrolled, until a total of 149 patients with KS = 0 were included to detect the favorable NPV with one-sided alpha equal to 0.10 and power equal to 0.80. Results: Between November 2016 and May 2019, a total of 535 patients were enrolled. Among these, 153 (28.6%) had a KS = 0. The primary study endpoint was met: 29 (5.4%) patients received a diagnosis of TEs during or after hospitalization, with 7 (4.6%) cases in the KS = 0 group. However, patients with higher KS values did not show increasing TE incidence. Among the other evaluated risk assessment models, the ONKOTEV scoreshowed the best predictive potential, with significantly higher values in patients with TEs (p < 0.001).Of note, TEs were associated with poorer overall survival (median, 6.7 vs 24.8 months, log rank p < 0.001). Conclusions: The INDICATE study showed that hospitalized cancer patients with KS = 0 at admission have a low risk of TEs, and could thus be spared from routine thromboprophylaxis. Further studies are needed to better define a RAM in this population.