Search

Type 1 congenital cystic adenomatoid malformation (CCAM), the most frequent malformation of the lung, is the only type to present intracystic mucinous cell clusters, which may form beyond the cysts extracystic mucinous proliferation resembling mucinous bronchioloalveolar carcinomas (BACs). As mucinous BACs are increasingly described in the literature in young patients with CCAM, we hypothesized that type 1 CCAM mucinous cells could represent BAC precursors. We reviewed 7 cases of type 1 CCAM including 6 with intracystic mucinous cell clusters, 3 with extracystic mucinous proliferations, and 4 with mucinous BAC or mixed adenocarcinoma with predominant BAC. K-ras mutations at codon 12 were detected in 3/3 intracystic mucinous cell clusters, in 2/3 extracystic mucinous proliferations, and in 3/4 BAC. Loss of heterozygosity (LOH) at p16(INK4) locus, with microsatellite alterations in 3 cases, was observed in 2/3 intracystic mucinous cell clusters, in 2/3 extracystic mucinous proliferations, and in all BAC. Two extracystic mucinous proliferations showed LOH at FHIT and Rb loci, respectively. P16(INK4) expression was lost in 2 intracystic mucinous cell clusters, 1 extracystic mucinous proliferation, and 1 BAC. Neither epidermal growth factor receptor mutation on exons 18, 19, and 21 nor P53 accumulation was observed. All lesions expressed MUC5AC, but were negative for MUC2, CDX2, and TTF-1. In conclusion, type 1 CCAM mucinous cells share the same differentiation profile with corresponding mucinous BAC, consistent with a common bronchial origin. Moreover, the high frequency of K-ras mutation and LOH and/or microsatellite alterations at p16(INK4) locus presented by these mucinous cells justifies their consideration as BAC precursors.

In lung cancer, an association was made between drastic clinical response to epidermal growth factor receptor (EGFR) inhibitors and the presence of somatic mutations within the tyrosine kinase domain of the EGFR. In some cases, patients with partial response or disease stabilization do not always have EGFR-mutated tumors. To go further in the characterization of the EGF pathway, we screened EGFR, ERBB2, ERBB3, KRAS, BRAF, and PIK3CA for mutations in 2 groups of White patients with nonsmall cell lung cancer (45 cancers from women and 46 cancers from men). Associations between TP53 mutations, clinicopathologic parameters, and EGF pathway molecular alterations were analyzed. All mutations were exclusive and essentially found in EGFR and KRAS. We demonstrated that EGFR mutations were linked to female sex, absence of smoking, late age at diagnosis, and adenocarcinoma (ADC) with bronchioloalveolar (BAC) features. Moreover, in invasive ADC with BAC component, microdissection assays showed that mutations were retrieved in both tumor subtypes suggesting that EGFR mutations appear early in lung carcinogenesis. On the contrary, KRAS mutations correlated with smoking, younger age at diagnosis, and ADC subtype regardless of BAC differentiation. These results suggest the existence of distinct carcinogenesis pathways both leading to disruption of EGF regulation and targeted either by tobacco carcinogens or by unidentified toxic. The identification of BAC features in ADC helps clustering patients that are more likely to fit the EGFR-mutated group.

Discrimination of invasive well-differentiated adenocarcinoma (IAD) from reactive bronchioloalveolar epithelium entrapped in pulmonary scars (PSE) may be difficult on routine histology, especially on small biopsies. Ancillary studies to help in this regard are desirable. Whereas IADs have been shown to harbor cumulative mutational damage of tumor suppressor genes, little is known about molecular changes in PSEs. In this study, we compared cumulative loss of heterozygosity (LOH) of tumor suppressor genes in PSEs (N = 12), bronchioloalveolar carcinomas (BACs, N = 15) and stage 1 IADs (N = 7). Unstained serial sections were microdissected to obtain lesional and normal tissue DNA. PCR was performed for up to 16 polymorphic markers. An allelic ratio of < 0.5 or >2.0 was designated as LOH. Fractional allelic loss (FAL) was calculated for each case as the number of markers with LOH divided by the total number of informative markers. Mean percentage of informative markers was 76.8%. PSEs showed significantly lower mean FAL compared with BACs and IADs (3.0% vs. 20.4% and 28.5%, respectively; P < 0.003). Only 1 case of PSE showed LOH of one marker in two different areas, whereas the majority of allelic losses in the neoplasms were present in two or more microdissected foci. Our study shows that PSEs harbor LOH of tumor suppressor genes at relatively low rates and in a random distribution compared with BACs and IADs, which show consistent allelic losses, and high FALs. These molecular differences may serve as an adjunct to histology in challenging glandular lesions of the lung.

We have previously reported that terminal-respiratory-unit (TRU) type adenocarcinoma is a distinct subset of lung adenocarcinoma in terms of molecular pathway for carcinogenesis and phenotypic profiles. This type of cancer shows TRU features, characterized by distinct cellular morphology and the expression of TTF-1 and surfactant proteins. Recently, two groups published novel mutations of the epidermal growth factor receptor (EGFR) that are closely associated with clinical response to gefitinib. The clinicopathologic features of gefitinib responders overlap with those of TRU-type adenocarcinoma, and the characteristics of TRU are likely to correspond to the bronchioloalveolar features reported as a predictor of gefitinib response. We therefore examined the characteristics of EGFR-mutated pulmonary adenocarcinomas with special reference to TRU-type adenocarcinoma. EGFR mutation was detected in 97 of 195 adenocarcinomas, 91 of 149 TRU-type adenocarcinomas and 6 of 46 tumors of other types. Conversely, 91 of 97 EGFR-mutated adenocarcinomas were categorized as TRU-type adenocarcinomas. This type-specific involvement was confirmed by logistic regression model. In addition, EGFR mutation was detected in some cases of atypical adenomatous hyperplasia, a preinvasive lesion of TRU-type adenocarcinoma. These findings further confirm that TRU-type-adenocarcinoma is a distinct adenocarcinoma subset in which a particular molecular pathway is involved.

To detect the possible genetic alterations characteristic of bronchioloalveolar carcinoma (BAC) and to study molecular genetic factors responsible for determining the biologic aggressiveness of pulmonary adenocarcinoma, comparative analysis of loss of heterozygosity (LOH) on 9 chromosomal regions was performed in 14 BACs and in 20 stage I adenocarcinomas (AD). The most frequently affected chromosome regions in BAC were 8q and 17p. In stage I AD, more than 60% of the cases showed LOH of 1p, 3p, 5q, 7q, 17p, and 18q loci, and LOH of 1p, 3p, 7q, and 18q was observed with greater frequency than in BAC (P < 0.05). Fractional allele loss (FAL) was significantly greater in stage I AD than in BAC (P < 0.001). In cases with microdissection of multiple sites, intratumoral heterogeneity of LOH status was observed in 73% of BAC and 94% of stage I AD, and homogeneous distribution of LOH of 9p was unique to BAC. The high FAL value was associated with a poor prognosis of BAC, but this trend did not reach statistical significance (P = 0.098). In stage I AD, no correlation was found between LOH of particular chromosomal region or FAL and clinical outcome. LOH of 1p, 3p, 7q, and 18q was associated with invasive properties of pulmonary AD and may be useful in identifying invasive adenocarcinoma when conventional histomorphological tools are not helpful.

In an attempt to understand the histogenesis and molecular pathogenesis of multifocal bronchioloalveolar lung carcinoma (BAC) we studied 28 cases of BAC using a topographic genotyping approach for the presence of K-ras exon 1 mutations and p53 loss of heterozygosity (LOH). This analytical approach demonstrated K-ras exon 1 mutations in 12.5% of solitary BACs, 40% of BACs with microscopic or macroscopic satellite lesions, and 60% of BACs with intrathoracic metastases. In all cases with K-ras mutations, the identical point mutation was present in the primary, satellite, and intrathoracic metastatic lesions. When p53 LOH was demonstrated in the primary lesion, it was also detected in the satellites and intrathoracic metastases. No significant association was noted between the presence of K-ras mutations and p53 LOH. The results strongly support a monoclonal origin of multifocal BACs. Furthermore, the findings support the theories explaining the origin of multifocal BAC by intraalveolar route of spread, intrapulmonary lymphatic spread, or aerosolization leading to implantation at different sites. A trend toward an increased frequency of K-ras mutations and p53 LOH in BACs with satellites or metastases compared to solitary BACs was noted.