Hanny Al-Samkari1, Orly Leiva2, Ibiayi Dagogo-Jack3, Alice Shaw3, Jochen Lennerz4, Anthony J Iafrate4, Pavan K Bendapudi3, Jean M Connors5. 1. Division of Hematology Oncology, Massachusetts General Hospital, Boston, Massachusetts; Harvard Medical School, Boston, Massachusetts. Electronic address: hal-samkari@mgh.harvard.edu. 2. Harvard Medical School, Boston, Massachusetts; Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts. 3. Division of Hematology Oncology, Massachusetts General Hospital, Boston, Massachusetts; Harvard Medical School, Boston, Massachusetts. 4. Harvard Medical School, Boston, Massachusetts; Department of Pathology, Center for Integrated Diagnostics, Massachusetts General Hospital, Boston, Massachusetts. 5. Harvard Medical School, Boston, Massachusetts; Hematology Division, Brigham and Women's Hospital, Boston, Massachusetts.
Abstract
INTRODUCTION: Clinical venous thromboembolism (VTE) risk prediction scores, such as the Khorana Risk Score, perform poorly in NSCLC, possibly because the tumor molecular subtype is omitted. Previous studies suggest a possible increased VTE risk in ALK-rearranged NSCLC, but data are conflicting. METHODS: We performed a retrospective cohort study of patients with advanced-stage NSCLC diagnosed between 2009 and 2019. Multivariable, time-to-event analyses modeling the risk of first venous or arterial thrombosis in ALK and non-ALK NSCLC groups, controlling for covariates known to impact thrombosis risk (15 in VTE model and 17 in arterial thrombosis model), were performed using Cox proportional hazards regression and competing-risks regression. Multivariable negative binomial regression modeled the total VTE rate. RESULTS: A total of 422 patients with ALK-rearranged and 385 patients with non-ALK-rearranged NSCLC were included. Patients with an ALK rearrangement were younger, had better performance status, and had lower rates of most thrombotic risk factors but had significantly higher rates of initial VTE (42.7% versus 28.6%, p < 0.0001), recurrent VTE (13.5% versus 3.1%, p < 0.0001), and similar rates of arterial thrombosis (5.0% versus 4.4%, p = 0.71) compared with non-ALK NSCLC. VTE risk attributable to ALK was significant (Cox model: hazard ratio 3.70, [95% confidence interval [CI]: 2.51-5.44, p < 0.001], competing risks: subhazard ratio 3.91 [95% CI: 2.55-5.99, p < 0.001]). Negative binomial modeling revealed higher VTE rates in patients with an ALK rearrangement (incidence rate ratio 2.47 [95% CI: 1.72-3.55, p < 0.001]). The OR for recurrent VTE was 4.85 (95% CI: 2.60-9.52, p < 0.001). Arterial thrombosis risk attributable to ALK was significant (Cox model: hazard ratio 3.15 [95% CI: 1.18-8.37, p = 0.021], competing risks: subhazard ratio 2.80 [95% CI: 1.06-7.43, p = 0.038]). CONCLUSIONS: In time-to-event analyses controlling for thrombosis risk factors, the ALK rearrangement conferred a fourfold increase in VTE risk and a threefold increase in arterial thrombosis risk in NSCLC. These patients may benefit from pharmacologic thromboprophylaxis.
INTRODUCTION: Clinical venous thromboembolism (VTE) risk prediction scores, such as the Khorana Risk Score, perform poorly in NSCLC, possibly because the tumor molecular subtype is omitted. Previous studies suggest a possible increased VTE risk in ALK-rearranged NSCLC, but data are conflicting. METHODS: We performed a retrospective cohort study of patients with advanced-stage NSCLC diagnosed between 2009 and 2019. Multivariable, time-to-event analyses modeling the risk of first venous or arterial thrombosis in ALK and non-ALKNSCLC groups, controlling for covariates known to impact thrombosis risk (15 in VTE model and 17 in arterial thrombosis model), were performed using Cox proportional hazards regression and competing-risks regression. Multivariable negative binomial regression modeled the total VTE rate. RESULTS: A total of 422 patients with ALK-rearranged and 385 patients with non-ALK-rearranged NSCLC were included. Patients with an ALK rearrangement were younger, had better performance status, and had lower rates of most thrombotic risk factors but had significantly higher rates of initial VTE (42.7% versus 28.6%, p < 0.0001), recurrent VTE (13.5% versus 3.1%, p < 0.0001), and similar rates of arterial thrombosis (5.0% versus 4.4%, p = 0.71) compared with non-ALKNSCLC. VTE risk attributable to ALK was significant (Cox model: hazard ratio 3.70, [95% confidence interval [CI]: 2.51-5.44, p < 0.001], competing risks: subhazard ratio 3.91 [95% CI: 2.55-5.99, p < 0.001]). Negative binomial modeling revealed higher VTE rates in patients with an ALK rearrangement (incidence rate ratio 2.47 [95% CI: 1.72-3.55, p < 0.001]). The OR for recurrent VTE was 4.85 (95% CI: 2.60-9.52, p < 0.001). Arterial thrombosis risk attributable to ALK was significant (Cox model: hazard ratio 3.15 [95% CI: 1.18-8.37, p = 0.021], competing risks: subhazard ratio 2.80 [95% CI: 1.06-7.43, p = 0.038]). CONCLUSIONS: In time-to-event analyses controlling for thrombosis risk factors, the ALK rearrangement conferred a fourfold increase in VTE risk and a threefold increase in arterial thrombosis risk in NSCLC. These patients may benefit from pharmacologic thromboprophylaxis.
Authors: Malinda T West; Thomas Kartika; Ashley R Paquin; Erik Liederbauer; Tony J Zheng; Lucy Lane; Kyaw Thein; Joseph J Shatzel Journal: Curr Probl Cancer Date: 2022-01-10 Impact factor: 3.187
Authors: Alok A Khorana; Nigel Mackman; Anna Falanga; Ingrid Pabinger; Simon Noble; Walter Ageno; Florian Moik; Agnes Y Y Lee Journal: Nat Rev Dis Primers Date: 2022-02-17 Impact factor: 65.038
Authors: Malinda T West; Claire E Smith; Andy Kaempf; Tia C L Kohs; Ramin Amirsoltani; Jessica Ribkoff; Josh Lee Choung; Alison Palumbo; Zahi Mitri; Joseph J Shatzel Journal: Eur J Haematol Date: 2021-02-18 Impact factor: 2.997