Skip to main content
. Author manuscript; available in PMC: 2009 Aug 1.
Published in final edited form as: Cancer Res. 2008 Aug 1;68(15):6208–6214. doi: 10.1158/0008-5472.CAN-07-6616

Fig. 3.

Fig. 3

Estimated probability of RUNX3 methylation (PMR ≥ 10%) is plotted as a function of age at diagnosis for current or former smokers (N=265, dashed grey line) and for never or irregular smokers (N=38, dashed black line). These fitted curves are superimposed over the raw data that has been smoothed using a LOESS procedure, with the smokers represented by the solid grey line and the nonsmokers by the solid black line. The logistic regression analysis was performed on tumors from 304 patients and yielded a statistically significant association between RUNX3 methylation and smoking history (p=0.009, likelihood ratio test), adjusted by age at diagnosis, tumor grade, and tumor stage.