Abstract Number: OC 12.1
Meeting: ISTH 2021 Congress
Theme: Venous Thromboembolism » Cancer Associated Thrombosis
Background: The Khorana Score (KS) predicts Venous Thromboembolism (VTE) rates in cancer patients. ISTH and other guidelines, recommend consideration of primary thromboprophylaxis in patients starting chemotherapy with KS ≥ 2.
Aims: Establish baseline VTE and death rates to assess effect of planned thromboprophylaxis intervention.
Methods: From 10/17/2017 to 11/29/2019 we identified all patients in MSKCC, with active cancer, a new chemotherapy order, and full KS data. This was IRB approved. VTE events were identified from contrast imaging studies, Doppler ultrasounds, V/Q scans, plus manual review. We also queried new prescriptions for anticoagulants and ICD10 codes. We tracked VTE events up to 6 months and all-cause mortality up to 12 months.
Results: 13,170 patients were identified. 850 experienced a VTE within 6 months (6.45%) and 3529 died within 12 months (26.8%). The KS strongly predicted VTE rates within 6 months, as well as death rates at 6 and 12 months. (See Figure/Table). The correlation of death rates and KS held after accounting for VTE events. By Cox regression with VTE (time-dependent) and KS as covariates, the hazard of death is 4.185 times higher for patients having VTE compared to no VTE (Table 1B). HGB, Platelets, and WBC predicted both VTE and mortality (p<0.0001). Cancers with higher VTE rates showed a trend of increased mortality.
Figure 1. Kaplan–Meier 12 months cumulative death probability.
| Six-Month Rate of VTE | Six-Month Death Rate by KS and VTE | Hazard Ratio of 6-Month Death by KS | Hazard Ratio of 12-Month Death by KS | ||
| Khorana Score |
Hazard ratio (95% CI) |
No VTE | With VTE | Hazard Ratio (95% CI) |
Hazard Ratio (95% CI) |
| 0 | – | 4.49% | 13.13% | – | – |
| 1 | 1.184 (0.9798, 1.431) |
7.08% | 14.89% | 1.536 (1.296, 1.820) * |
1.406 (1.250, 1.583) * |
| 2 | 1.989 (1.6474, 2.402) |
9.94% | 22.92% | 2.218 (1.866, 2.636) * |
2.276 (2.021, 2.564) * |
| 3 | 3.287 (2.6341, 4.102) |
16.50% | 35.38% | 3.622 (2.974, 4.412) * |
3.245 (2.808, 3.751) * |
| 4+ | 5.193 (3.7776, 7.139) |
22.60% | 46.81% | 5.201 (3.915, 6.909) * |
5.166 (4.191, 6.368) * |
| Univariate Cox cause-specific model analysis of VTE with death as the competing event. | Cox regression analysis of 6 months survival. VTE is a time-dependent covariate. | Univariate Cox regression analysis | |||
| * (p<0.0001) | |||||
VTE and Death Rates and Hazard Ratio as Function of Khorana Score and Occurrence of Venous Thromboembolism.
Conclusions: The KS predicts all-cause mortality as well as 6-month VTE rates. We interpret these findings as the KS is associated with a more aggressive cancer biology, and this aggressive phenotype predicts both VTE and mortality. For a given KS, the occurrence of VTE further predicts the risk of death, however it has yet to be determined if this is causative or an association. It also has not yet been demonstrated that reduction of VTE events may reduce mortality.
To cite this abstract in AMA style:
Soff G, Sathe C, Poon C, Yin S, Devlin S, Sarasohn D, Gilbert L, Khorana A, Mantha S. The Khorana Score Is a Marker of Poor Cancer Prognosis, Independent of Development of VTE [abstract]. Res Pract Thromb Haemost. 2021; 5 (Suppl 2). https://abstracts.isth.org/abstract/the-khorana-score-is-a-marker-of-poor-cancer-prognosis-independent-of-development-of-vte/. Accessed December 6, 2023.« Back to ISTH 2021 Congress
ISTH Congress Abstracts - https://abstracts.isth.org/abstract/the-khorana-score-is-a-marker-of-poor-cancer-prognosis-independent-of-development-of-vte/