Our data provide evidence that HCC may be a unique thyroid cancer distinct from papillary and follicular thyroid cancer.

Purpose: Low-dose CT screening can reduce lung cancer-related mortality. However, CT screening has an FDR of nearly 96%. We sought to assess whether urine samples can be a source for DNA methylation-based detection of non-small cell lung cancer (NSCLC).Experimental Design: This nested case-control study of subjects with suspicious nodules on CT imaging obtained plasma and urine samples preoperatively. Cases (n ¼ 74) had pathologic confirmation of NSCLC. Controls (n ¼ 27) had a noncancer diagnosis. We detected promoter methylation in plasma and urine samples using methylation on beads and quantitative methylation-specific realtime PCR for cancer-specific genes (CDO1, TAC1, HOXA7, HOXA9, SOX17, and ZFP42).Results: DNA methylation at cancer-specific loci was detected in both plasma and urine, and was more frequent in patients with cancer compared with controls for all six genes in plasma and in CDO1, TAC1, HOXA9, and SOX17 in urine. Univariate and multivariate logistic regression analysis showed that methylation detection in each one of six genes in plasma and CDO1, TAC1, HOXA9, and SOX17 in urine were significantly associated with the diagnosis of NSCLC, independent of age, race, and smoking pack-years. When methylation was detected for three or more genes in both plasma and urine, the sensitivity and specificity for lung cancer diagnosis were 73% and 92%, respectively.Conclusions: DNA methylation-based biomarkers in plasma and urine could be useful as an adjunct to CT screening to guide decision-making regarding further invasive procedures in patients with pulmonary nodules.

<p>Gene Methylation Detection, Sensitivity, Specificity Using Detectable vs. Non-detectable cutoff in NSCLC late stages</p>

<p>Figure S6</p>

<p>Univariate and multivariate logistic regression analysis for risk of having NSCLC in Early Stages (n=75)</p>

<p>Figure S2</p>

<p>Univariate and multivariate logistic regression analysis for risk of having NSCLC in Black (n=38)</p>

<p>Figure S7</p>