Graham Lidgard

PhD

Exact Sciences

CSO and SVP R&D

NLST (Learn more about this study)

NLST-190

Jan 20, 2016

Clinical and economic impact model of biomarkers used in conjunction with low dose CT in lung cancer screening of high risk individuals

The project is the design and development of a clinical and economic outcomes model to quantify the requirements for biomarkers to enhance the performance (overall sensitivity and specificity) of lung cancer screening by low dose CT. The model will be applicable to any biomarker development program. It will be a Markov-based decision analysis with microsimulation (using TreeAge Pro software) and will quantify clinical outcomes (e.g., life years, quality adjusted life years, or lung cancer deaths averted) and economic outcomes (will include all relevant costs).

We will use cost, health-state utility, and decision probability data from primary analysis of relevant databases, public sources and from the literature. We need the NLST dataset to inform probabilities of positive or negative follow-up CT results conditional on an initial negative CT result, where the definition of a “screen negative” LDCT result may be different than the one used in the NLST trial. Alternate definitions may include different nodule size thresholds, size determination methodology, nodule attenuation, or nodule growth criteria.

We would need the following information, which could be derived from the NLST database, to help design and power clinical trials for the development of biomarkers used in conjunction with LDCT lung cancer screening:

• Distribution of lung cancer stage at diagnosis by nodule size observed at last CT screen prior to diagnosis
o We would expect biomarker quantitation level to be proportional to nodule size
• Length of time from CT result to diagnosis of lung cancer.
o Absolute calendar dates are not needed—so long as time passage can be ascertained in a useful unit of measure. This information will be helpful in planning trial length.