Optimizing biomarker based strategies for lung cancer screening

Lung cancer remains the leading cause of cancer related mortality in the US and globally— responsible for ~1.8 million deaths annually. Randomized trials of low dose CT (LDCT) for lung cancer screening have demonstrated a clear mortality benefit and are recommended by the United States Preventive Services Task Force (USPSTF). However, most individuals currently eligible for screening do not have lung cancer and many at risk populations are not candidates for screening. Additionally, LDCT has a relatively high rate of false positives leading to potentially harmful tests and invasive procedures. Moreover, early experience suggests that LCS outcomes in clinical practice may differ from those observed in randomized trials. Given the concerns with the performance of LDCT, the use of blood and imaging biomarkers to complement the screening process has the potential to enhance its efficacy and effectiveness. Biomarkers could be employed at two different steps of the screening process – as the initial test for cancer screening, to better define the best candidates for screening, or following a positive LDCT scan, to inform lung nodule evaluation and management. While several screening and diagnostic biomarkers have been developed, they have not been formally evaluated in prospective comparative effectiveness trials, nor have they been subjected to cost-effectiveness analyses. More importantly, their impact on lung cancer burden at the population level is still unknown. Our proposed research will inform the following areas of LCS and biomarker discovery: 1) generate a computational framework to estimate and compare the population impact of emerging screening and diagnostic biomarkers, 2) identify the optimal strategy of integrating biomarker tests into LCS, and 3) highlight and identify the characteristics of biomarkers (sensitivity and specificity for histological subtypes, etc.) which are critical for cost-effective biomarker applications.