PB0984 - Prediction of lung cancer-associated venous thromboembolism in middle and low income countries: The N & N score

Background: Venous thromboembolism (VTE) is a common cause of death among cancer patients. Incidentally, the risk of VTE in lung cancer patients can be as high as 4–20%. Several biomarkers have been identified, and several prediction models exist for assessing VTE among lung cancer patients. One major limitation in these models is the lack of systematic stratification of cancer-associated VTE risk factors, thus limiting their performance and cost-ineffectiveness.

Aims: To review the biomarkers of VTE amongst patients with lung cancer and apply meta-analysis to pool the risk estimate per biomarker.

Methods: We computed the stability index, which is a function of recurrence (f). Biomarkers were stratified in terms of the relative risk based on the risk weight (rw), which is a product of the risk estimate and recurrence (f). Furthermore, we classified biomarkers into level 1, level 2 and level 3 based on their risk weight. We fitted a critical point of onset, otherwise known as the cut-off score, within the upper quartile, given that preemptive interventions are targeted for the high-risk group. Finally, bearing in mind the cost of determination of the relevant biomarkers, especially in resource-scarce low and middle-income countries, we propose a predictive model for VTE.

Results: We identified different paths to predicting VTE amongst patients with lung cancer, including the D-dimer route. However, in view of cost-effectiveness, we fitted a cost-effective model comprising relevant biomarkers. Hence, the N & N score, which derives its name from the authors’ initials is a composite score derived from the risk weight of five relevant biomarkers namely albumin, leukocyte, and haemoglobin, age and body mass index (Table 1).

Conclusion(s): We identified and proposed a cost-effective predictive model for VTE in lung cancer. Further studies are required to validate the performance and suitability of the proposed N & N score towards detecting VTE in lung cancer