Background: Cyclin dependent kinase (CDK) 4/6 inhibitors combined with endocrine therapy are an integral therapy for advanced hormone receptor positive breast cancer. Rates of venous thromboembolic events (VTE) were between 1-5% in clinical trials leading to CDK 4/6 inhibitor approval. Abemaciclib currently carries an FDA warning for VTE risk. However, rates of thrombosis in real world practice remain poorly described. We aim to better define the thrombotic risk associated with CDK 4/6 inhibitors, expanding to evaluate venous and arterial events. Additionally, factors that predispose to thrombosis while on CDK 4/6 inhibitors are unclear. We hypothesize the Khorana thrombosis risk score may predict which patients will experience thrombosis on CDK 4/6 therapy.

Methods: We conducted a retrospective analysis of breast cancer patients aged ≥18 years who were prescribed a CDK 4/6 inhibitor (palbociclib, ribociclib, abemiciclib) at the Knight Cancer Institute and affiliated clinics between February 2015 and March 2020. Venous and arterial thrombosis occurring during treatment or within 30 days of discontinuation were included. We collected data including pre-treatment lab values, age, BMI, prior thrombosis, and smoking status and calculated a Khorana risk score for each patient. Utilizing univariate and multivariate logistic regression we compared these variables in patients who developed thrombosis and those who did not. We calculated overall survival of the two groups. Results: There were 270 patients included in the analysis and 29 patients (10.7%) developed a thrombotic event. Of these, 66% were venous, 28% were arterial, and 10% experienced >1 clot. Ribociclib had the highest incidence of thrombosis; 17%, followed by palbociclib; 9% and abemaciclib; 5% (Table 1). Multivariate analysis evaluating risk factors for thrombosis did not find any statistically significant predictors of thrombosis (Table 2). Khorana scores did not predict which patients experienced thrombosis. Median overall survival did not significantly differ between those who developed thrombosis and those who did not (23 months vs 17.5 months respectively, p value = 0.37).

Discussion: Incidence of thrombosis in our institutional analysis is higher than reported in clinical trials. Interestingly, we found arterial thrombosis comprised over one-third of events. Patients on ribociclib experienced the highest incidence of thrombosis, suggesting the FDA thrombosis warning may need to be expanded to the drug class. Khorna scores were not predictive of thrombosis risk in our population, however only 1% of study patients had a high risk score over 2. Larger, real world studies are needed to define the risk of thrombosis with CDK 4/6 inhibitors. The role of prophylactic anticoagulation is yet to be defined in this patient population.

TABLE 1: Incidence of thrombosis

Cumulative Incidence:
Total CDK 4/6 inhibitor population n = 270 
Number of thrombotic events 29 (11%) 
Total arterial 8 (28%) 
Total venous 19 (66%) 
Total arterial + venous 2 (6.9%) 
Incidence by CDK4/6 inhibitor  
Abemaciclib (n = 20) 1 (5%) 
Palbociclib (n = 233) 22 (9.4%) 
Ribociclib (n = 17) 3 (17.6%) 
Site of event  
First thrombotic event n = 26 
CVA 2 (6.9%) 
CVA + DVT 1 (3.5%) 
CVA + MI 1 (3.5%) 
DVT 10 (34.5%) 
MI 1 (3.5%) 
PE 4 (13.8%) 
PE + DVT 2 (7%) 
Line associated DVT 1 (3.5%) 
Retinal vein thrombosis 1 (3.5%) 
TIA 3 (10.3%) 
Second thrombotic event n = 3 
DVT 1 (3.5%) 
PE + MI 1 (3.5%) 
TIA 1 (3.5%) 
Cumulative Incidence:
Total CDK 4/6 inhibitor population n = 270 
Number of thrombotic events 29 (11%) 
Total arterial 8 (28%) 
Total venous 19 (66%) 
Total arterial + venous 2 (6.9%) 
Incidence by CDK4/6 inhibitor  
Abemaciclib (n = 20) 1 (5%) 
Palbociclib (n = 233) 22 (9.4%) 
Ribociclib (n = 17) 3 (17.6%) 
Site of event  
First thrombotic event n = 26 
CVA 2 (6.9%) 
CVA + DVT 1 (3.5%) 
CVA + MI 1 (3.5%) 
DVT 10 (34.5%) 
MI 1 (3.5%) 
PE 4 (13.8%) 
PE + DVT 2 (7%) 
Line associated DVT 1 (3.5%) 
Retinal vein thrombosis 1 (3.5%) 
TIA 3 (10.3%) 
Second thrombotic event n = 3 
DVT 1 (3.5%) 
PE + MI 1 (3.5%) 
TIA 1 (3.5%) 

TABLE 2 Predictors of thrombosis

CharacteristicOR95% CIP-Value
Age 1.02 0.98 - 1.06 0.28 
BMI 1.04 0.98 - 1.1 0.24 
Hemoglobin 0.83 0.69 - 1.00 0.06 
History of thrombosis 1.91 0.72 - 5.09 0.20 
Platelet 1.0 0.99 - 1.01 0.86 
Smoking 2.61 0.75 - 9.11 0.13 
WBC 0.97 0.84 - 1.11 0.64 
CharacteristicOR95% CIP-Value
Age 1.02 0.98 - 1.06 0.28 
BMI 1.04 0.98 - 1.1 0.24 
Hemoglobin 0.83 0.69 - 1.00 0.06 
History of thrombosis 1.91 0.72 - 5.09 0.20 
Platelet 1.0 0.99 - 1.01 0.86 
Smoking 2.61 0.75 - 9.11 0.13 
WBC 0.97 0.84 - 1.11 0.64 

Citation Format: Malinda West, Claire Smith, Tia Kohs, Ramin Amirsoltani, Jessica Ribkoff, Joshua Choung, Alison Palumbo, Joseph Shatzel. CDK 4/6 inhibitors are associated with a high incidence of thrombotic events in real world practice [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS13-24.