The definition of bronchiolar adenoma was firstly proposed by Chang et al (1) in 2018, and CMPT is considered to belong to a subset of BA. BA is composed of bilayered cellular proliferation with a continuous basal cell layer, which was highlighted by basal cell markers p40 and/or CK5/6 (1). But in clinical practice, it was found that the tissue structure of some cases could not be completely conformed to the classical BA. Zhang et al (6). reported part of tumor only consist of a single layer of columnar or cubic epithelial cells in some cases of BA, without basal cell layer below, and this region is transitional to the tumor region with a double layer structure. Even some lesions were entirely composed of a single layer of columnar or cubic epithelial cells with glandular or papillary structures, which is smilar to the monolayer cell area of bronchial adenoma with monolayer cell changes. The authors defined such cases as atypical BA (6). And in our case, the pure GGN showed there were alveoli of different sizes, with single or double layers of columnar epithelial cells or flat cells, and was positive for TTF-1 and P63 (focal). The mixed GGN showed that there were predominantly papillary or glandular architecture covering single layer columnar epithelial cells with a few ciliums without basal cell layer, and was positive for TTF-1, Napsin-A (focal), CK7, P63 (partial) and CK5/6(a few). So, we consider these two lesions as atypical BA.
Considering the spectrum of airway epithelium from the proximal bronchus to the peripheral alveolar structures, BA is considered to be a benign tumor (1), like a bronchial adenoma. However, it is still controversial whether BA has malignant potential. More and more cases with malignant transformation have been reported in the literature recently (3–5, 7). Miyai et al(3) reported a case of CMPT with malignant transformation of basal tumor cell components, and Chen et al (7) descripted a case of mucinous adenocarcinoma caused by cancerization from CMPT and speculated CMPT is a precancerous lesion of mucinous adenocarcinoma. Han et al (5) reported a case of BA losing continuity of the basal cell layer in the junctional zone between BA and IMA, and BA harbors the same KRAS mutation with the adjacent invasive mucinous adenocarcinoma. Therefore, a common histological feature of these cases is the absence of basal cell layer or the cancerization of basal cell, which indicated BA may be carcinogenic, especially in atypical cases. For the following reasons, we consider malignant transformation of the mixed GGN: First, the lesion lost the typical structures of the bronchioles and alveoli, with disordered morphology. Second, the boundary of the lesion is not clear, and there are discontinuous, skipping microscopic lesions. Third,the cells of the tumor had relatively swollen nuclei and were proliferating in a papillary or micropapillary manner.
Usually, it is extremely difficult to distinguish BA from AIS, minimally invasive adenocarcinoma (MIA) or adenocarcinoma in frozen section, especially adenocarcinoma (1). Seven case of BA were misdiagnosed as adenocarcinomas in frozen sections. The reasons are as follows: first, some pathologists have insufficient understanding of the disease; second, some morphologic characteristics of BA are difficult to recognize in frozen sections; third, immunohistochemistry is required for differential diagnosis in some atypical cases. In this case, due to another case of typical BA has been identified 2 weeks ago in paraffin section, so the mixed GGN was diagnosed as BA quickly in frozen section. However, there are some findings to support malignant transformation as above. This significantly increases the difficulty of differential diagnosis in frozen section.
Some mutations of common driver genes for lung adenocarcinoma also have been identified in BAs, such as EGFR, KRAS and BRAF (1, 8, 9). Different from the high frequency of EGFR mutation in lung cancer, a high percentage (50%) of BRAF mutation was found in BA (8). Meanwhile, KRAS mutation has been found in 24% of BA (1). In this case, only CCNE1 gene mutation was identified in both lesions and her-2 mutation was identified in mixed GGN. The protein encoded by CCNE1 gene participates and plays an important role in G1/S cell-cycle regulation. Recent study has found that the CCNE1 gene could serve as an oncogene and a poor prognostic factor in ovarian cancer and lung cancer (10–12). However, CCNE1 gene mutation has not been reported in BA or CMPT so far. Further studies are needed to confirm whether CCNE1 promotes BA canceration.