Prognostic implication of microsatellite alteration profiles in early-stage non-small cell lung cancer

Development of non-small cell lung cancer (NSCLC) is a result of multiple accumulated genetic abnormalities. Profiles of genetic abnormalities may determine tumor behavior and impact on patient outcome. We used microsatellite markers at 3p14, 9p21, and 10q24 to analyze tumor samples from 91 patients with pathologically confirmed stage I NSCLC for microsatellite alterations. Loss of heterozygosity at any single locus was not significantly associated with length of survival. However, patients whose tumors had microsatellite instability (MI) at 10q24 had shortened disease-specific survival. Among 31 such patients, 32% (10 of 31 patients) had died of the disease within 5 years after surgery compared with 16% (9 of 58 patients) without MI at 10q24 (P = 0.07). Interestingly, in the adenocarcinoma subtype, 71% (5 of 7 patients) of the patients with MI at 10q24 succumbed to the disease as compared with only 12% (3 of 26) of the adenocarcinoma patients without such MI (P < 0.001), suggesting the presence of distinct mechanisms in tumorigenesis among different subtypes of lung cancer. It has been noticed that certain microsatellite alteration profiles provide additional values for risk assessment. Of 23 patients who had MI at 10q24 and an alteration at 3p14, 39% (9 of 23 patients) died of the disease within 5 years as compared with only 15% (10 of 66 patients) of the patients without such a profile (P = 0.02). Strikingly, among the 22 patients with no alteration at any loci tested or with loss of heterozygosity at 10q24 and retention of at least one of the other two loci, none died of lung cancer within 5 years after surgery, whereas 28% (19 of 67 patients) of the patients outside these profiles did so (P = 0.01). Our results support the hypothesis that microsatellite alterations can be used as biomarkers for the genetic classification of pathological stage I NSCLC, which may in turn influence treatment decisions dependent on an accurate forecast of patient survival time.