Our study presented at ASH this year was a real-world data study looking at the rates of both venous and arterial thrombosis among patients with a diagnosis of breast cancer who were on targeted therapies. We know that patients with breast cancer undergoing treatment in both the early stage and metastatic setting are at an increased risk of thrombotic events and we have previously seen in prospective clinical trials that certain medications such as CDK4/6 inhibitors, and particularly abemaciclib, increase one’s risk of venous thrombosis and maybe arterial thrombosis as well. So we wanted to study this in a real-world population because, as you know, certain side effects, certain toxicities, even certain outcomes, can differ within a real-world population compared to the very controlled patient population that are in prospective randomised multicentre clinical trials.
What was the study design?
The study design, we used a big real-world evidence database called TriNetX which is de-identified national data collected from numerous centres across the country. The database itself has over 129 million patients and we narrowed it down to patients with a diagnosis of breast cancer using ICD-10 codes and then limited that to patients who are on targeted therapies for breast cancer. What I mean by that is CDK4/6 inhibitors, PARP inhibitors, alpelisib, and mTOR inhibitor everolimus. We compared that population to a general population of patients with breast cancer for our multivariate analysis.
What are the key results?
The results showed, maybe not surprisingly, that the incidence of both venous and arterial thrombotic events were much higher in this real-world population than was previously seen in the prospective randomised trials for these drugs. Particularly among the CDK4/6 inhibitors, if you look at the original trials that got these drugs approved, abemaciclib looked like it probably had the highest thrombotic risk at about [??]% for venous thrombosis and arterial thrombotic events of around 1% or less. In this real-world population, among all CDK4/6 inhibitors the rate of thrombosis was about 13%, particularly palbociclib and abemaciclib seemed to have higher risks than ribociclib in our study.
Both CDK4/6 inhibitors and everolimus, the mTOR inhibitor, were associated with an increased risk of arterial thrombosis and then for venous thrombosis the risk was increased among patients on all CDK4/6 inhibitors, everolimus, alpelisib, but wasn’t really seen among patients taking olaparib, or the PARP inhibitor. So rates were higher than previously seen.
What is the clinical significance of these results?
These results may prompt clinicians to take a better look at your patient’s thrombosis risk. Let’s say you have a patient based on a clinical thrombosis risk score, such as the Khorana score, who may be higher risk or intermediate risk and you’re going to start one of these medications for their breast cancer treatment. I think that would prompt me to highly consider prophylactic anticoagulation among those patients. I would just make sure that you counsel these patients that there is a risk of thrombosis, both venous and arterial, and it may be attributed to these medications.
So this study adds to the existing body of data that these drugs are probably associated with an increased risk of thrombosis and preventing the morbidity and mortality that can come with thrombosis is probably the best thing we can do.
