Impact of Khorana Score, Tumor Histology, and Treatment Regimen on VTE Incidence in Lung Cancer

Introduction:

Venous thromboembolism (VTE) is common in cancer patients, with up to 15% developing VTE. (Agnelli, 2018) Lung cancer patients have further increased risk of VTE with a reported incidence as high as 30%. (Khorana, 2018) VTE occurrence is a prognosis modifying comorbidity, and active lung cancer is a high-risk predictor of VTE recurrence. (Heit, 2016)

The Khorana score (KRS) is a validated risk stratification tool developed to estimate risk of VTE in patients with cancer and prevent VTEs through recommendations for prophylactic anticoagulation. The KRS incorporates clinical and laboratory parameters to calculate individual risk. Following development of their model, Khorana et. al showed that the KRS had a high negative predictive value (98.5%) but a low positive predictive value (6.7%). (Khorana, 2008) Several studies have attempted to validate the score for higher risk populations, but limitations have been identified. Mansfield et.al demonstrated no difference in rates of VTE between patients with high and intermediate KRS. (Mansfield, 2016)

While the KRS might have applicability in populations at low risk, it would be of significant clinical utility to better understand those at higher risk of VTE occurrence and recurrence. It is unknown how the histology, presence of actionable genomic alterations (AGAs), and treatment type impact the performance and validity of the KRS in patients with lung cancer. Therefore, we performed a retrospective study to evaluate the KRS and these factors.

Methods:

We conducted a retrospective review of patients seen at our center for late-stage lung cancer, diagnosed between March 2015 and April 2023. This included both small cell (extensive stage) and non-small cell lung cancers (stages III & IV) treated with chemotherapy only (CTX), chemotherapy + immunotherapy (IO), immunotherapy only and tyrosine kinase inhibitor (TKI) monotherapy. Clinical information was obtained through review of the electronic medical record.

The primary endpoint was VTE incidence by histology type and initial systemic therapy. We examined VTE incidence by AGAs and calculated median time to VTE in each subgroup. Chi square analysis was used to assess the correlation between Khorana score and incidence of VTE. Secondary endpoints included VTE incidence in patients on antiplatelet or anticoagulation prior to lung cancer treatment.

Results:

The VTE incidence across 552 patients in our cohort was 23.2%. The VTE incidence in NSCLC was 24.8% and 11.9% in SCLC. Chi square analysis of VTE incidence in low (0-1) KRS patients is 23.4% and 21.6% in high KRS (score >2), with a Pearson correlation coefficient of 0.118.

We observed a significant difference in VTE incidence by age and race with younger patients and Black patients exhibiting higher incidence. BMI and sex did not correlate.

Presence of any AGA correlated with a higher risk of VTE incidence, with the highest incidence occurring in those who had mutations in MET and EGFR.

The highest VTE incidence was in those with TKI monotherapy (30.6%, p=0.169 ), then CTX (24.5%), then CTX + IO (21.5%) and lastly in IO only (16.8%). These variations, however, are not statistically significant.

Median time to VTE in patients that experienced VTE (VTE+) was 251 days, and there was no significant variation between histology groups, treatment groups or AGA. Overall survival was not impacted by VTE diagnosis.

VTE incidence among patients on anti-platelet prior to lung cancer treatment initiation was 20.3% versus 24.5% in those not on anti-platelet (p=0.276). However, VTE incidence among patients on anticoagulation prior to treatment was 7% versus 25.6% for those not on anticoagulation (p=0.001).

Conclusions:

The KRS did not predict development of VTE in our population of advanced lung cancer. NSCLC had a higher rate of VTE compared to SCLC. Prescence of AGA correlates with higher rates of VTE incidence. Our study did not show a difference in survival based on the occurrence of VTE. Antiplatelet use prior to lung cancer treatment did not significantly reduce VTE incidence, while anticoagulant use did.

VTEs are a common and significant complication of lung cancer. Our results show that the KRS inadequately predicts VTE incidence. The presence of AGAs impacted VTE incidence and should be considered in future risk assessment models. Better tools are needed to inform which patients with lung cancer should receive prophylactic anticoagulant therapy.

Hall:Jazz Pharmaceuticals: Consultancy; Takeda: Consultancy; Regeneron: Consultancy; Merck Sharp & Dohme: Research Funding; AstraZeneca/MedImmune: Research Funding; Mirati Therapeutics: Research Funding; Lilly: Research Funding; Daiichi Sankyo/Lilly: Research Funding; Genentech: Research Funding; Regeneron: Research Funding.