Preinvasive lesions of the bronchus.

Preinvasive lesions are considered the precursors of squamous cell carcinoma of the bronchus. Treatment at the preinvasive stage, before the potential for metastasis, may improve survival from squamous cell carcinoma. An understanding of the natural history and outcome of preinvasive lesions is essential for the accurate interpretation studies of their treatment, and decisions regarding the management of individual lesions. The natural history of preinvasive lesions has only been reported in a small number of highly selected patients and uses different inclusion criteria, treatment criteria. and time-periods of follow-up, making it difficult to draw definitive conclusions. High-grade preinvasive lesions carry a risk of progression to carcinoma but most patients have multiple lesions and a significant probability of developing new lesions over time. Distinguishing lesions with malignant potential, the targets for therapy, from those that will regress or remain indolent is difficult. The American College of Chest Physicians guidelines recommend bronchoscopic follow-up of severe dysplasia and carcinoma-in situ. This review of the evidence regarding the natural history and outcome of preinvasive lesions supports this view, but also shows that further studies in individuals at risk for lung cancer are necessary before guidelines for the management of preinvasive lesions can be developed.

Surveillance for the detection of early lung cancer in patients with bronchial dysplasia.

BACKGROUND: The natural history of bronchial preinvasive lesions and the risk of developing lung cancer in patients with these lesions are not clear. Previous studies have treated severe dysplasia and carcinoma in situ (CIS) on the assumption that most will progress to invasive carcinoma. AIMS: To define the natural history of preinvasive lesions and assess lung cancer risk in patients with these lesions. HYPOTHESIS: Most preinvasive lesions will not progress to invasive carcinoma but patients with these lesions will be at high risk. METHODS: A cohort of patients with preinvasive lesions underwent fluorescence bronchoscopy every 4-12 months and computed tomography of the chest annually. The main end point was the development of invasive carcinoma. RESULTS: 22 patients with 53 lesions were followed up for 12-85 months. 11 cancers were diagnosed in 9 patients. Of the 36 high-grade lesions (severe dysplasia and CIS), 6 progressed to invasive cancers. 5 separate cancers developed at remote sites in patients with high-grade lesions. All cancers were N0M0 and curative treatment was given to 8 of the 9 patients. The cumulative risk of developing lung cancer in a patient with a high-grade lesion was 33% and 54% at 1 and 2 years, respectively. Of the 17 low-grade lesions, none progressed to invasive carcinoma. CONCLUSIONS: Although the risk of malignant progression of individual preinvasive lesions is relatively small, patients with high-grade lesions are at high risk of lung cancer. Surveillance facilitated early detection and treatment with curative intent in most patients.

Somatic genetic changes accompanying lung tumor development.

Carcinomas are believed to develop by incremental steps of increasingly abnormal morphology driven by accumulating somatic genetic changes. This process is often difficult to study, as the early stages are undetectable. We used fluorescence bronchoscopy, which enhances detection of preinvasive bronchial lesions, and have obtained sequential biopsies of carcinoma in situ (CIS) from a patient with no detectable tumor and from a squamous cell carcinoma that developed 19 months after presentation at the site of one of the previous CIS lesions. Biopsies of preinvasive CIS, which follow-up showed had different pathologic outcomes, and tumor were microdissected to obtain enriched cell populations and DNA prepared from them. Molecular characteristics of these biopsies were compared by loss of heterozygosity analysis, TP53 mutation analysis, and comparative genomic hybridization. Although all lesions examined had the same TP53 mutation and almost identical allelotypes, differences were observed. Loss in 5q21 and amplification of 3q25-26 were associated with the lesion that progressed and the subsequent carcinoma. Allele loss at 4p16 was detected in the tumor but not in any of the CIS lesions, suggesting it was a late event associated with tumor invasion. Amplification at 4q12 was specifically observed in the tumor and in the CIS at the site of eventual tumor formation. Although these findings may be unique to this one patient, the successful demonstration of sequential genetic changes raises the possibility that this approach, unencumbered by interpatient variability between lesions, will greatly facilitate the identification of molecular events driving the invasive process