Purpose: An overlapping area of gain at 3q26 has been reported in lung squamous cell carcinoma (SCC), but whether this also occurs in preneoplastic/preinvasive squamous cell proliferations and early-stage invasive carcinomas of the lung is still unknown. Experimental Design: We evaluated the prevalence and the clinicopathologic implications of 3q26 amplification and polysomy of chromosome 3 in 31preneoplastic/preinvasive squamous cell lesions of the bronchial mucosa and in 139 early-stage invasive pulmonary SCC, both of limited growth within the bronchial wall [early hilar SCC (EHSCC)] and involving the pulmonary parenchyma [parenchyma-infiltrating SCC (PISCC)]. Moreover, mRNA expression of two candidate genes (h-TERC and SKI-like), both mapping to the minimal common amplification region, was also studied by quantitative real-time reverse transcription-PCR. Results: 3q26 amplification and polysomy of chromosome 3 were confined to malignant samples, with 37% of invasive SCC, and 27% of severe dysplasias/in situ carcinomas showing these chromosomal abnormalities. Amplification (with minimal common amplification region at 3q26.2), polysomy 3, concurrent amplification and polysomy 3, or other changes (monosomy) were found in 25 SCC and 1dysplasia, 24 and 2, 2 and 0, and 1and 0, respectively. Amplification was significantly associated with EHSCC, polysomy 3 with PISCC. 3q26 amplification correlated with increased tumor diameter and a history of smoking, whereas polysomy 3 correlated with tumor diameter, pT class, and p53, p21, and fascin immunoreactivity. No relationship of either 3q26 gain or polysomy was found with patients' survival. Overexpression of h-TERC or SKI-like mRNA was found in 3q26-amplified or polysomic SCC, with higher levels of h-TERC in the former and of SKI-like in the latter. Conclusions: 3q26 amplification and chromosome 3 polysomy may be related to the development of invasive SCC, with differential distribution in tumor subsets, despite substantial histologic uniformity. Both h-TERC and SKI-like may be involved in tumor progression.A large body of evidence supports the theory that pulmonary squamous cell carcinoma (SCC) is not caused by a single transforming event, but by a steady accrual of subsequent molecular genetic and epigenetic abnormalities, resulting in a spectrum of premalignant/preinvasive lesions, i.e., basal cell hyperplasia, squamous metaplasia and dysplasia, and in situ SCC (carcinoma in situ [CIS]), occurring over time, either preceding or accompanying invasive tumors (according to the so-called sequential theory of morphologic and molecular changes; refs. 1 -7). Whether basal cell hyperplasia, squamous metaplasia, and mild squamous dysplasia are common reactive changes or already true neoplastic changes is still unclear, whereas severe dysplasia and CIS are currently considered highgrade preinvasive lesions, with 40% or more individuals developing invasive cancer over time (8,9).In the lung, early-stage invasive SCC [corresponding to pT1-2N0M0 stage tumor patients accor...