New Prediction Model Could Guide Shared Decision Making in NSCLC

A novel risk-prediction model can identify patients at an increased risk of mortality following surgical treatment for stage I and II non-small cell lung cancer (NSCLC), according to results presented at the International Association for the Study of Lung Cancer (IASLC) 17^th Annual World Conference on Lung Cancer in Austria.¹

Low-dose computed tomography (CT) scans can detect early-stage lung cancer cases and reduce lung cancer mortality. Screening benefits are reduced, however, if patients are poor candidates for surgery.

Researchers analyzed socio-demographic and medical history variables to develop an algorithm that estimates the risk of 30-day mortality following surgery for patients with early-stage NSCLC. Using Surveillance, Epidemiology, and End Results (SEER) and Medicare linked databases, they identified age, race, country of origin, urban/rural status, and comorbidities in the year prior to NSCLC diagnosis. All patients had at least 1 year of Medicare enrollment prior to diagnosis and received initial surgery within 6 months of diagnosis.

The researchers used an initial sample of 1571 surgical cases and conducted internal validation with an additional 4632 independent surgical cases. They identified 201 deaths that occurred within 30 days of surgery, and found an observed risk of 30-day mortality that was 9.3-fold greater in the highest 10% of predicted risk.

Pending additional development and validation, the risk-prediction model has potential clinical applications that can quickly and efficiently inform cancer screening and shared decision-making.

Reference

1. Roth JA, Ramsey SD. A lung cancer surgical mortality risk-prediction algorithm to inform lung cancer screening shared decision-making. Paper presented at: International Association for the Study of Lung Cancer 17^th World Conference on Lung Cancer; December 2016; Vienna, Austria.