Materials and Methods
Patient Material
Seven formalin-fixed and paraffin-embedded (FFPE) samples of surgically resected pulmonary adenofibroma diagnosed from 2009 to 2020 in the Pathology Department of the West China Hospital, Sichuan University were collected, on the basis of morphology and related IHC staining. Every patient underwent surgical resection, with wedge resection in 5 patients and lobectomy in 2 patients. These specimens were reviewed by two experienced pathologists. Relevant data of all cases was collected as follows: Sex, age, smoking status, tumor size, location and follow-up. After obtaining institutional authorization and the approval from Ethics Committee, inform consent can be exempted in this retrospective observational research.
Immunohistochemical Staining
Four-μm sections cut from representative available FFPE blocks of each case were evaluated for a panel of immunohistochemical markers, including Pan-CK (clone AE1/AE3, BIO), TTF-1 (clone 8G7G3/1, ZECA), EMA (clone GP1.4, BIO), ER (clone SP1, Roche), PR (clone 1E2, Roche), CD34 (clone EP88, BIO), STAT6 (clone, EP325, MXB), S-100 (clone 4C4.9, MXB), SMA (clone UMAB237, BIO), Desmin (clone MX046, MXB). The stain was performed on Leica Bond-Max or Roche Ventana system. The positive percentage and expression pattern (membrane, cytoplasm or nucleus) were evaluated. Staining results were regarded as positive when it expressed moderate or greater intensity in more than 10% of neoplastic cells. Reactivity less than 10% of neoplastic cells were considered as negative.
Tissue and Plasma DNA Isolation and Purification
Genomic DNA (gDNA) was extracted from the sufficient and qualified FFPE samples using the GeneRead DNA FFPE Kit (Qiagen, Germantown, MD, USA) and from peripheral blood mononuclear cells using the DNA Blood Midi/Mini kit (Qiagen) according to the manufacturer’s instructions. Plasma cell-free DNA (cfDNA) was isolated using the MagMAX Cell-Free DNA Isolation Kit (Thermo, Waltham, MA, USA) according to the manufacturer’s protocol. The quality of purified DNA was assayed by gel electrophoresis and quantified with a Qubit 4.0 Fluorometer (Life Technologies, Carlsbad, CA, USA).
Library Construction and ctDNA and Whole Exome Sequencing
Purified gDNA was first fragmented into DNA pieces of approximately 300 bp using an enzymatic method (5 × WGS Fragmentation Mix; Qiagen). After end-repairing and A tailing, T-adaptors were ligated on both ends, followed by PCR amplification to obtain a pre-library. The final sequencing libraries were prepared using the 96 rxn xGen Exome Research Panel v1.0 (Integrated DNA Technology, Coralville, IA, USA) according to the manufacturer’s protocol. For the targeted sequencing of cfDNA, the prelibraries were prepared according to the method previously described (6). In-house panels were designed to capture cfDNA fragments to generate sequencing libraries. The sequencing libraries were evaluated using the NovaSeq 6000 platform (Illumina, San Diego, CA, USA) in 150PE mode.
Bioinformatics Analysis of Whole Exome Sequencing Results
The raw sequencing reads were subjected to quality control by trimming adaptor sequences and removing poly-N sequences (>10%) and low-quality reads (<Q20) preprocessed using FASTP (7). The FASTQ files were aligned to the human reference genome (hg19/GRCh37) using Burrows–Wheeler Aligner (BWA, v0.7.15) (8). Picard (2.12.1) (http://picard.sourceforge.net/) was used to process PCR duplicates for mapped BAM files. GATK (the Genome Analysis Toolkit 4.0.11.0) (9) was used for local realignment and base quality recalibration was employed to compute sequencing coverage and depth. Single nucleotide variants (SNVs) and small insertions and deletions (indels) were identified using GATK MuTect2. Mutations in the ENCODE Data Analysis Consortium blacklist were removed.
Variants were annotated using ANNOVAR (10) based on multiple databases, including HGVS variant description and population frequency databases (1000G, ExAC, and dbSNP), disease or phenotype databases (OMIM, COSMIC, and ClinVar), and variant functional in silico prediction tools (PolyPhen-2 and SIFT). After annotation, SNVs annotated as genomicSuperDups with a variant allele frequency (VAF) < 0.2 or PopFreqMax > 0.05 were excluded and nonsynonymous SNVs with a VAF > 3% or with a VAF > 1% in cancer hotspots collected from patient databases were retained for further analyses.
Discussion
The understanding of the PAF progressed fairly slow on account of its rare incidence. First, PAF was described as a subtype belonging to the same tumor family as pulmonary hamartomas (1, 21). After that, the distinction between PAF and SFT was ambiguous due to morphological similarity. PAF was described as a special variant of SFT in some studies (3, 4), especially in one patient that PAF and SFT coincidentally occurred in the same lobe of the lung (17). In recent years, most scholars have supported that PAF belongs to an independent entity, while more cases and further studies are still needed to investigate the mechanism of tumorigenesis.
The detailed data of these 7 patients and cases in the literature are summarized in
Table 1
. The clinicopathological features of our cases were consistent with those of patients in the literature. Gender preference showed that PAF was more common in female than male patients. For predilection sites, the tumors were located slightly more in the left lung than in the right lung, with no significant difference between the upper and lower lobes. Most PAFs were asymptomatic and detected by physical examination, while some can presented with cough (14, 16, 20), chest pain (12, 13, 20) and breathlessness (13, 16, 20). PAF radiographically presented as a solitary, well-circumscribed solid nodule on the peripheral lung, consistent with the features of benign tumors. The maximum diameters mostly ranged from 0.5 to 4.5 cm (5, 14, 20). Some studies found that PAF grew very slowly, with no signs of malignant transformation during the 7 years of follow-up (12).
PAF is histologically heterogeneous. Typical cases are characterized by epithelial and stromal components: stromal spindle fibroblastic or myofibroblastic cells and glandular elements lined by slightly respiratory epithelium (some show a leaf-like epithelial pattern). Eosinophilic secretions can sometimes be found in the glandular lumina (14). Epithelial cells with a clear cytoplasm are frequently present (13), indicating that they may secrete glycogen or mucus. Calcification and myxoid changes could occasionally be seen, but no intratumoral hemorrhage or necrosis are reported (14, 20).
The neoplastic component of PAF remains controversial. Most scholars considered PAF as a biphasic tumor, composed of neoplastic epithelium and stromal spindle cells (14, 18–20). Other scholars argued that the glandular component of PAF is entrapped normal respiratory epithelium during tumor growth and only stroma is genuine neoplastic component (1, 15). Fusco et al. even thought that epithelium of PAF is bronchiolar metaplasia based on the expression of E-cardherin, TTF-1 and its ultrastructural features (5).
Regarding immunohistochemical studies, CK, EMA and TTF-1 were widely expressed in the epithelium, while the expression of mesenchymal components was different among different cases. Previous studies have described PAF as a peripheral lesion, and TTF-1 was invariably positive. In our series, however, two TTF-1-negative cases with the tumor located away from the pleura suggest that epithelium is derived from a larger bronchus. Only a few cases showed positive expression of Bcl-2 (11/26), CD34 (14/33), and S-100 (1/23) in the spindle cells of the stroma. Desmin (7/29) and SMA (10/29) were expressed in some PAFs, likely indicating an immunophenotype of myofibroblasts in those spindle cells.
In recent years, ER and PR were found to be occasionally positive in the stroma, possibly indicating that PAF is potentially hormone related (5, 20). Two female patients and one male patient in our series were positive for ER, PR, SMA and Desmin simultaneously, which does not support this point of view. No pelvis or breast mass was found in these 2 female patients, excluding the possibility of metastatic smooth muscle tumors of the uterus or malignant phyllodes tumors of the breast. What’s more, no statistically significant differences were found in the expression of hormone receptors between different sexes in all the cases.
Fusco et al. reported that 5 of 7 PAFs were nuclear positive for STAT6 protein. In their study, sequencing results also showed that 4 of 7 cases, similar to SFT cases, harbored the specific NAB2ex4-STAT6ex2 fusion, indicating that PAF has a similar genetic origin as SFT (5). Olson et al. even reported that a 60-year-old male patient developed PAF and SFT in the same lung. Next-generation sequencing was performed in these two lesions, but the NAB2-STAT6 fusion was only identified in SFT (17). Several large-scale studies on soft tissue tumors showed that nuclear positivity for the STAT6 antibody and STAT6 fusion gene were mostly characterized by SFT (22–25). Lindholm et al. noted that STAT6 protein was negative in their 13 PAFs, and the expression of CD34 and Bcl-2 was also significantly different between SFT and PAF (20). Given their findings, Lindholm et al. considered PAF as an independent entity that must be separated from SFT. All the cases in our series were negative for STAT6 protein. Combined with the negative sequencing results, we also agreed that using IHC staining or molecular detection to exclude SFT is essential to demonstrate diagnosis of PAF.
Currently, NGS has only been performed in one PAF in the literature (17). Hence, we performed NGS on 3 cases of PAF tissues together with one adenocarcinoma tissue coincidently occurring in the same lobe of one patient (Case 4) to further clarify its molecular characteristics. The results showed that PAF did not carry the NAB2-STAT6 gene fusion in SFT or the genetic changes in breast fibroepithelial tumors, such as MED2, RARA, and FLNA (26). The mutations detected in PAF were unrelated to the tumorigenesis of malignant tumors, and relevant studies were lacking. MUC17 mutated with SNVs or indels at three codon sites was detected in two PAF cases. The MUC17 gene is located on chromosome 7q22. RNA blotting and in situ hybridization proved that it could be expressed in intestinal absorption cells, colon cancer and pancreatic cancer cells (27), while corresponding research in the lung tissue was scant. Located on the surface of the cell membrane, the family of membrane mucins execute several crucial functions, including cytoprotection (28), migration promotion (29) and signal transduction (30, 31). Mutation of the MUC17 gene may also lead to abnormal function of goblet cells, which are widely arranged on the bronchial epithelium.
For the patient with PAF and adenocarcinoma in the same lobe, SNVs of HER2 were detected in adenocarcinoma, and SNVs of JAK1 and PTPN11 genes were found in PAF, suggesting that PAF is unlikely to be related to the occurrence of adenocarcinoma. As a collateral branch of the JAK-STAT pathway and upstream of the MAPK signaling pathway, JAK1 and PTPN11 are related to cell proliferation, differentiation and transformation (32).
In addition to SFT, PAF should be differentiated from other benign and malignant tumors. More than two mesenchymal components can be found in hamartoma, such as cartilage, smooth muscle, adipose tissue, loose mucus-like tissue or bone. In particular, when clear cartilage or obvious fat tissue is present in the stroma, the diagnosis of PAF should be excluded. Sclerosing pneumocytoma (SP) has similar gender preference, tumor size and even macroscopic appearance compared to PAF. SP typically demonstrates a more varied growth pattern consisting of papillary, sclerotic, and solid growth as well as comprising two cell types (2). The extensive existence of monomorphic round cells in stroma with immunoactivity of TTF-1 and epithelial membrane antigen is crucial to differentiate from PAF (33, 34). Papillary adenoma consists of papillary structures containing fibrovascular cores lined by a single layer of cuboidal epithelium, which is different from PAF (2). Of note, papillary adenoma is thought to have low malignant potential for its occasionally infiltrative growth (35–37). However, there are no reported cases of metastatic disease or recurrence in PAF, and the tumor is generally considered to be benign and is cured by resection. Alveolar adenomas usually form multiple microscopical spaces lined by type II pneumocytes, which is filled with finely granular proteinaceous material (2). Glandular structure accompany with fibroblasts proliferation can be seen in some alveolar adenomas (38). Compared to PAF, stressed lumen and prominent spindle cells in stroma are generally scant in alveolar adenoma. Mesothelioma can present with adenoid structures, but it is closely related to the pleura and expresses mesothelial markers (e.g. calretinin, D2-40, and WT-1) in the tumor components. The epithelial component of biphasic synovial sarcoma can be characterized as a slit-like or tube-like glandular cavity, while the epithelial component does not express TTF-1 or Napsin A. Additionally, the positivity of TLE-1 protein and translocation of the SS18 gene are helpful for the differential diagnosis. Sarcomatoid carcinoma, particularly pulmonary blastoma and carcinosarcoma with bidirectional differentiation, has apparent dysplasia and mitotic activity, which can be easily distinguished from the glandular epithelium with bland morphology in PAF. Metastatic breast phyllodes tumors can also show phyllodes structures, but they can be differentiated based on ER, PR, TTF-1 and Napsin A staining and clinical history. Given that various lung tumors can more or less mimic the morphological pattern of PAF, Erber et al. commended that the exceptionally rare genuine PAF should be considered an exclusive diagnosis (19).
In summary, as a rare benign pulmonary lesion, PAF has not been listed in the 2015 WHO Classification of Tumors of the Lung, Pleura, Thymus and Heart. Pathologists should be aware of its unique morphology, and a definite diagnosis should be confirmed by immunohistochemical staining and even genetic detection. Additionally, clinicians should be familiar with its indolent biological behavior and excellent prognosis to avoid overtreatment. Consequently, more cases and further researches are needed to further illustrate its clinicopathological characteristics.