09/03/2026
For years, many of us have treated cancer-associated thrombosis as if it were a single disease.
But the data keep reminding us: it isn’t.
Colleagues analyzing patients in the RIETE Registry, we looked at outcomes in more than 18,000 patients with cancer-associated VTE, alongside over 88,000 patients without cancer.
What emerges is a story about heterogeneity.
Within the first 90 days after VTE, patients with cancer experienced:
• 4% recurrent VTE or VTE-related death
• 3.4% major bleeding
Compared with patients without cancer, the risks were markedly higher:
2.5× higher risk of recurrent VTE or fatal PE
~40% higher risk of major bleeding
But the most interesting signal appears when you stop thinking of “cancer” as one category.
Some cancers behave very differently. Lung and pancreatic cancers carry some of the highest thrombotic risk we see.
Meanwhile, gastrointestinal and genitourinary cancers disproportionately drive bleeding risk.
In practice, that tension is familiar to anyone managing these patients:
The same patient who is at very high risk of clotting may also be at meaningful risk of bleeding.
Large real-world registries like RIETE allow us to step back and see the patterns across thousands of patients. And the message is fairly clear:
Cancer-associated thrombosis is not one disease.
It is many biologically different thrombotic phenotypes.
For those of us interested in precision vascular medicine, this matters.
The next step is moving beyond generic anticoagulation strategies toward risk-adapted approaches that account for tumor biology, bleeding phenotype, and patient-level factors.
We’re getting closer to that reality.
