Cancer stage at diagnosis of Non-Small Cell Lung Cancer (NSCLC) provides important information about the treatment options and prognosis. Early detection is critical for early stage NSCLC patients as treatment options with curative intent are available. We aim to investigate the spatiotemporal variations in NSCLC incidence rates in New York State (NYS). We used a scan-statistic approach to assess spatial clustering of incidence rates. We obtained county level age-adjusted incidence NSCLC rates and socioeconomic index in NYS between 1995 and 2015 from SEER*Stat maintained by the National Cancer Institute. We conducted weighted space-only normal model using the inverse of the age-adjusted rate standard error, and unweighted, space-time normal model, adjusting for the Yost Index. Models were ran separately for overall and each cancer stage. Significant (p< 0.01) high/low incidence clusters were detected, with variations seen in the cluster sizes and locations between space-only vs. space-time models. The results may be useful for identifying areas where further investigation and targeted NSCLC prevention and interventions, such as lung cancer screening programs, are needed.