Association between Provoked and Unprovoked Venous Thromboembolism and Malignancy: A Cross-Sectional Retrospective Cohort Study

Introduction:

A strong association exists between incidences of deep venous thrombosis (DVT), pulmonary embolism (PE), or venous thromboembolism (VTE) in cancer patients. Evidence for subsequent incidence of new malignancy in patients with VTE is poor and controversial. There is even less evidence to demonstrate an association between malignancy and provoked and unprovoked VTE. Despite this, extensive invasive and non-invasive diagnostic investigations are often performed during hospitalizations for VTE to rule out malignancy in patients. Through our study, we sought to further investigate the association of incidences of malignancy between provoked and unprovoked VTE patients.

Methods:

A retrospective search was first conducted to identify adult hospitalizations for VTE at a single tertiary care center for 2019. Through manual chart review for inciting risk factors, these patients were categorized as provoked and unprovoked. The two cohorts were longitudinally followed over a subsequent 5-year period (2020-2024). for the incidence of new malignancies. Patients with known malignancy and those who were lost to follow-up over the 5 years were excluded from the final analysis pool. Final data was analyzed using StataBE18 software to perform the Chi-squared test and Risk Ratio.

Results:

We identified 221 VTE hospitalizations in the year 2019, of which 155 were categorized as provoked and 66 as unprovoked. After excluding patients lost to follow-up and patients with active malignancy (75 provoked and 12 unprovoked), the final analysis included 80 provoked and 54 unprovoked VTE cases. Of the 54 unprovoked VTE patients, 5 (9.26%) were diagnosed with new malignancy during the follow-up period. Of the 80 provoked patients, 7 (8.75%) patients were diagnosed with new malignancy. We performed a Pearson's Chi-squared test to evaluate statistically significant differences in incidences of new cancer between each of the cohorts and found that the incidences were similar (chi2=0.0103; p-value 0.919) with a risk ratio of 1.03 and a 95% Confidence Interval between 0.51-2.10.

Conclusions:

Our analysis indicates that there is no statistically significant relationship between unprovoked or provoked VTE and the diagnosis of new malignancy. The incidence rates were almost similar in both cohorts. Therefore, patients with either provoked or unprovoked VTE should not require additional aggressive testing for malignancy. They should be expected to undergo age-appropriate screening as otherwise indicated. Excessive testing can lead to anxiety, prolonged hospitalization, misallocation of resources, and increased healthcare spending. Our study limitations include a small sample size and single-center analysis. Future larger studies comparing outcomes at different centers can provide further clarity to streamline our findings, given historically conflicting reports.

Farber:Genmab/AbbVie: Membership on an entity's Board of Directors or advisory committees; BeiGene: Speakers Bureau; Genentech: Speakers Bureau; Genmab/AbbVie: Speakers Bureau; Gilead/Kite: Speakers Bureau; Incyte/MorphoSys: Speakers Bureau; Seagen: Speakers Bureau.