Pathologists are required to categorize lung cancer into adenocarcinoma and SqCC due to the targetable driver genetic alterations identified in lung adenocarcinoma and inappropriate drugs for SqCCs due to side effects in patients with SqCC. Before the 2015 WHO classification, adenocarcinoma was defined as carcinoma with an acinar/tubular structure or mucin production, whereas SqCC was defined as carcinoma with keratinization or intercellular bridges. If poorly differentiated carcinoma lacking light microscopic evidence of glandular differentiation (Figure 1A) is proven by immunohistochemistry to express “adenocarcinoma markers,” such as TTF-1 (Figure 1B) and/or Napsin A (Figure 1C), it is diagnosed as a solid adenocarcinoma. If poorly differentiated carcinoma lacking light microscopic evidence of squamous differentiation (Figure 1D) is proven by immunohistochemistry to express “SqCC markers,” (28) such as p40 (Figure 1E), CK5/6 (Figure 1F), and p63, it is diagnosed as non-keratinizing SqCC. Because of this classification, the proportion of large cell carcinoma has been markedly reduced.

In the 2015 WHO classification, SqCCs are classified into keratinizing SqCC, non-keratinizing SqCC, and basaloid SqCC. Before this classification, basaloid SqCC was categorized as a variant of large cell carcinoma. However, basaloid SqCC immunohistochemically shows “SqCC markers” (e.g., p40, CK5/6, and p63) and is therefore categorized as SqCC.

In the 2015 WHO classification, a new category of “neuroendocrine tumors” was established. Invasive neuroendocrine tumors comprise three subtypes: SCLC, large cell neuroendocrine carcinoma (LCNEC), and carcinoid tumor (typical/atypical). Diffuse idiopathic pulmonary neuroendocrine cell hyperplasia is extremely rare and non-invasive; therefore, its clinical importance is low. On the other hand, the distinction between a high-grade neuroendocrine tumor (HGNET), comprising SCLC and LCNEC, and a carcinoid tumor is very important in both pathological and clinical practice. HGNET is one of the most aggressive subtypes and characterized by a history of heavy smoking in the patient, whereas carcinoid tumors usually carry a benign prognosis and frequently occur in patients with no history of smoking.

The comprehensive molecular profiling of 230 lung adenocarcinoma by TCGA was published in 2014 (3). The authors reported high rates of somatic mutations (mean: 8.9 mutations per megabase) and identified 18 statistically significant genetic mutations: TP53 (46%), KRAS (33%), KEAP1 (17%), STK11 (17%), EGFR (14%), NF1 (11%), BRAF (10%), SETD2 (9%), RBM10 (8%), MGA (8%), MET (7%), ARID1A (7%), PIK3CA (7%), SMARCA4 (6%), RB1 (4%), CDKN2A (4%), U2AF1 (3%), and RIT1 (2%).

Furthermore, approximately 75% of the lung adenocarcinomas examined harbored genetic alterations that promote the RTK/RAS/RAF signaling pathway. Of all the cases, 62% showed driver genetic alterations that promote the RTK/RAS/RAF pathway. Among them, mutations in KRAS, EGFR, and BRAF comprised 32, 11, and 7.0%, respectively. Other genetic alterations that promote the RTK/RAS/RAF pathway included MET exon 14 skipping (4.3%), ERBB2 (or HER2) mutation (1.7%), ROS1 fusion (1.7%), ALK fusion (1.3%), MAP2K1 mutation (0.9%), RET fusion (0.9%), NRAS mutation (0.4%), and HRAS mutation (0.4%). An examination of the DNA copy number of the remaining 38% cases without driver genetic alterations that promote the RTK/RAS/RAF pathway revealed amplification of oncogenes in the RTK/RAS/RAF pathway: ERBB2 amplification (0.9%) and MET amplification (2.2%). The authors also identified new genetic alterations in this pathway: mutations in NF1 and RIT1. NF1 is a tumor suppressor gene that regulates the RTK/RAS/RAF pathway, and the frequency of NF1 mutations was 8.3%. Similarly, RIT1 constitutes a part of the RTK/RAS/RAF pathway, and the frequency of RIT1 mutations was 2.2%. Consequently, 75% of lung adenocarcinomas have genetic alterations that promote the RTK/RAS/RAF pathway. This study on the comprehensive molecular characterization of lung adenocarcinoma widened the potential therapeutic targets of lung adenocarcinoma.

This study conducted mRNA profiling and provided new transcriptional subtypes: i.e., the terminal respiratory unit (TRU, formerly bronchioid), the proximal-inflammatory (PI, formerly squamoid), and the proximal-proliferative (PP, formerly magnoid) mRNA subtypes. The TRU subtype was enriched for EGFR mutation and kinase fusions. The PI subtype was characterized by solid morphology and co-mutation of NF1 and TP53. The PP subtype was enriched for KRAS mutation and STK11 inactivation.

DNA methylation profiling divided lung adenocarcinomas into three subtypes: i.e., CpG island methylator phenotype (CIMP)-high, CIMP-intermediate, and CIMP-low subtypes. CIMP-high tumors often showed DNA hypermethylation of CDKN2A, GATA2, GATA5, HIC1, HOXA9, HOXD13, RASSF1, SFRP1, SOX17, and WIF1. The CIMP-high subtype was enriched for MYC overexpression as well as for DNA hypermethylation of genes in WNT pathway.

Protein profiling divided lung adenocarcinomas into six subtypes. The top 50 differentially expressed proteins among the 6 subtypes included Cyclin D1, Smad4, p-mTOR, Rad50, beta-Catenin, and HER2. The six subtypes partially overlapped with the mRNA three subtypes.

The comprehensive molecular profiling of 178 cases of SqCC by TCGA was published in 2012 (4). As can be expected from the history of heavy smoking in SqCC patients, SqCCs are characterized by complex genomic alterations. The authors indeed detected a mean of 360 exonic mutations, 165 genomic rearrangements, and 323 segments of copy number alteration per tumor. They identified 11 statistically significant genetic mutations: TP53, CDKN2A, PTEN, PIK3CA, KEAP1, MLL2, HLA-A, NFE2L2, NOTCH1, RB1, and PDYN. The frequency of TP53 mutations was 90%. The authors identified novel loss-of-function mutations in the HLA-A class I major histocompatibility gene. They also conducted pathway analyses and identified pathways related to oxidative damage, including KEAP1 and NFE2L2 in 34% of cases, squamous cell differentiation pathway, including overexpression of SOX2 and TP63 in 44% of cases, PI3K/AKT pathway in 47% of cases, and inactivation of CDKN2A in 72% of SqCC cases. These results provided us multiple potential targets for the treatment of lung SqCC.

The mRNA profiling divided SqCCs into four subtypes: i.e., classical, basal, secretory, and primitive subtypes. The classical subtype was characterized by pronounced hypermethylation, chromosomal instability, and alterations in KEAP1, NFE2L2, and PTEN. The basal subtype showed NF1 alterations. The primitive subtype was enriched for RB1 and PTEN alterations.

MicroRNA profiling divided SqCCs into four subtypes. These four subtypes roughly overlapped with the mRNA four subtypes. DNA methylation profiling also divided SqCCs into four subtypes (methylation clusters 1–4). Methylation cluster 4 showed little DNA hypermethylation and included most of the primitive mRNA subtype. Cluster 3 was predominantly made up of the classical mRNA subtype and enriched for NFE2L2 mutations. Cluster 3 and Cluster 2 showed the highest levels of DNA hypermethylation. Cluster 1 showed intermediate DNA hypermethylation levels.

In 2012, comprehensive genomic analyses were reported by two groups. Rudin et al. identified SOX2 as a frequently amplified gene in SCLC (6). In vitro suppression of SOX2 blocked the proliferation of SOX2-amplified SCLC cell lines. Furthermore, they identified a recurrent RLF–MYCL1 fusion in SCLC with RNA sequencing. In vitro silencing of MYCL1 in SCLC cell lines with an RLF–MYCL1 fusion decreased cell proliferation. On the other hand, Peifer et al. reported recurrent mutations in genes that encode histone modifiers, including CREBBP, EP300, and MLL, suggesting histone modification as a major feature of SCLC (7).

In 2015, George et al. sequenced the genome of 110 SCLCs and found comprehensive genomic profiles (5). SCLC is characterized by highly complex genomic alterations, and C:G>A:T transversions were found in 28% of all mutations on average, which is a characteristic pattern of heavy smoking. Almost all examined SCLCs showed bi-allelic inactivation of TP53 and RB1. Genomic alterations of tumor suppressor gene TP73 was observed in 13% of SCLCs. Among TP73 genomic alterations, genomic rearrangement of TP73Δex2/3 was identified. Because TP73Δex2/3 promotes carcinogenesis, TP73Δex2/3-targeted strategy is a promising treatment for SCLC. The authors also observed inactivating mutations of NOTCH family genes (31), which suppressed neuroendocrine differentiation via the regulation of ASCL1 expression in 25% of SCLCs.

The mRNA profiling divided SCLCs into two groups (5). Most of SCLCs (83%) were categorized into group 2, which was characterized by higher expressions of CHGA, GRP, ASCL1, and DLK1. Group 1, comprising 17% of SCLCs, showed lower expressions of these four genes.

EGFR mutation is one of the most common driver mutations in lung adenocarcinoma, and EGFR-mutated adenocarcinoma is characterized by East-Asian ethnicity, female gender, and non/light-smoking history (32). Pathologically, EGFR-mutated lung adenocarcinomas typically show nuclear TTF-1 (NKX2-1) immunostaining and a hobnail cell type. Adenocarcinomas with a micropapillary pattern have a higher frequency of EGFR mutations than adenocarcinomas without this pattern (33, 34).

Fusion genes were recently identified as oncogenic drivers. In lung adenocarcinoma, rearranged genes, including ALK, ROS1, RET, NTRK1, and NRG1, have been reported. ALK-rearranged adenocarcinoma comprises 4–5% of adenocarcinomas (35). ROS1- and RET-rearranged adenocarcinoma each comprises approximately 1% (35, 36). These rearranged adenocarcinomas show a good clinical response to molecular-targeted drugs. ALK-rearranged adenocarcinoma is characterized by a TTF-1 cell lineage, an acinar structure with mucin/signet-ring cell pattern, non-/light-smoking history, and young onset (37–39). ROS1- and RET-rearranged adenocarcinomas have a similar histology to ALK-rearranged adenocarcinoma, such as mucinous cribriform pattern or solid signet-ring cell pattern (35, 40, 41).

NTRK1 fusion in lung adenocarcinoma was identified by Vaishnavi et al. (42) A NTRK1-rearranged adenocarcinoma identified by Shim et al. belonged to IMA subtype (43). NTRK1 is a TRKA kinase; thus, TRKA kinase inhibitors have the potential to be used in the treatments of NTRK1-rearranged adenocarcinomas. NRG1-rearranged adenocarcinoma is also characterized by IMA (44, 45). IMAs are frequently KRAS-mutated; therefore, Nakaoku et al. examined 34 IMA cases without KRAS mutations, partly by whole-transcriptome sequencing (45). They identified five oncogenic fusions: CD74–NRG1, SLC3A2–NRG1, EZR–ERBB4, TRIM24–BRAF, and KIAA1468–RET. These fusion genes were mutually exclusive from KRAS mutations. NRG1 fusions were present in 17.6% (6/34) of KRAS-wild-type IMAs. Because fusions of NRG1, ERBB4, BRAF, and RET are potential molecular targets, their clinical applications are promising.

Driver genetic alterations in lung adenocarcinomas differ between Caucasians and Asians, and between smokers and non-smokers. For example, KRAS mutations are frequently detected in lung adenocarcinomas in smokers, whereas genetic alterations in EGFR, ALK, ROS1, and RET are frequently detected in lung adenocarcinomas in non-smokers. The frequencies of driver genetic alterations in Caucasians were determined in TCGA study (3), whereas lung adenocarcinoma in Asians is characterized by a high frequency of EGFR mutations (approximately 50%) and low frequency of KRAS mutations (approximately 10%).

MicroRNAs are small single-stranded non-coding RNAs (19–22 nucleotides in length) that play important regulatory roles, including lung carcinogenesis (46–50). Nadal et al. performed microRNA profiling of adenocarcinomas subclassified by the 2015 WHO classification (51). They demonstrated that different histological subtypes of lung adenocarcinoma have distinct microRNA expression profiles. Unsupervised hierarchical clustering divided adenocarcinomas into three major clusters, which correlated with the histological subtypes of the 2015 WHO classification. Cluster 1 included fewer acinar and solid adenocarcinomas, and nearly all the tumors in cluster 1 were categorized as lepidic adenocarcinomas or IMAs. In contrast, clusters 2 and 3 included more acinar and solid adenocarcinomas and fewer lepidic adenocarcinomas and IMAs. Solid adenocarcinoma was characterized by the overexpression of miR-27a, miR-212, and miR-132 (51).

Enteric adenocarcinoma is one of the new variants of lung adenocarcinoma in the 2015 WHO classification. It is defined as an adenocarcinoma with a predominant component that shows enteric differentiation (30). Garajová et al. found that the microRNA signature of enteric adenocarcinomas shows similarities with NSCLCs and pancreatic adenocarcinomas, but not with colorectal adenocarcinomas. Enteric adenocarcinomas share oncogenic microRNAs (miR-31*, miR-126*, miR-506, miR-508-3p, and miR-514) with pancreatic adenocarcinomas (52).

Small cell lung carcinoma, which is categorized into neuroendocrine tumor, shows high expression of ASCL1, which is a transcription factor that promotes neuroendocrine differentiation. A study reported that miR-375 expression was promoted by ASCL1 in lung neuroendocrine carcinoma (53). This study suggested that miR-375 might reduce the YAP1-associated proliferative arrest by inhibiting YAP1. Another study examined microRNAs from 50 SCLC patients and 30 healthy individuals, and suggested that level of miR-92a-2 in plasma could be a potential non-invasive method for the SCLC diagnosis (54). Recent evidence suggests that miR-21 expression may be higher in HGNET (i.e., SCLC and LCNEC) than in typical/atypical carcinoid, and that high expression of miR-34a may be associated with atypical carcinoids (55).