Reference genome and annotation updates lead to contradictory prognostic predictions in gene expression signatures: a case study of resected stage I lung adenocarcinoma.

RNA-sequencing enables accurate and low-cost transcriptome-wide detection. However, expression estimates vary as reference genomes and gene annotations are updated, confounding existing expression-based prognostic signatures. Herein, prognostic 9-gene pair signature (GPS) was applied to 197 patients with stage I lung adenocarcinoma derived from previous and latest data from The Cancer Genome Atlas (TCGA) processed with different reference genomes and annotations. For 9-GPS, 6.6% of patients exhibited discordant risk classifications between the two TCGA versions. Similar results were observed for other prognostic signatures, including IRGPI, 15-gene and ORACLE. We found that conflicting annotations for gene length and overlap were the major cause of their discordant risk classification. Therefore, we constructed a prognostic 40-GPS based on stable genes across GENCODE v20-v30 and validated it using public data of 471 stage I samples (log-rank P < 0.0010). Risk classification was still stable in RNA-sequencing data processed with the newest GENCODE v32 versus GENCODE v20-v30. Specifically, 40-GPS could predict survival for 30 stage I samples with formalin-fixed paraffin-embedded tissues (log-rank P = 0.0177). In conclusion, this method overcomes the vulnerability of existing prognostic signatures due to reference genome and annotation updates. 40-GPS may offer individualized clinical applications due to its prognostic accuracy and classification stability.

RHBDF1 modulates cisplatin sensitivity of small cell lung cancer through YAP1/Smad2 signaling pathway.

Small cell lung cancer (SCLC) is a fatal tumor type that is prone to drug resistance. In our previous study, we showed that human rhomboid-5 homolog-1 (RHBDF1) was differentially expressed in 5 intrinsic cisplatin-resistant SCLC tissues compared with 5 intrinsic cisplatin-sensitive SCLC tissues by RNA sequencing, which intrigued us. We performed gain- and loss-of-function experiments to investigate RHBDF1 function, bioinformatics analysis, qRT-PCR, western blotting, and immunoprecipitation to elucidate the molecular mechanisms as well as detect RHBDF1 expression in SCLC by immunohistochemistry. We found that RHBDF1 knockdown promoted cell proliferation and cisplatin chemoresistance and inhibited apoptosis in vitro and in vivo. These effects could be reversed by overexpressing RHBDF1 in vitro. Mechanistically, RHBDF1 interacted with YAP1, which increased the phosphorylation of Smad2 and transported Smad2 to the nucleus. Among clinical specimens, the RHBDF1 was a low expression in SCLC and was associated with clinicopathological features and prognosis. We are the first to reveal that RHBDF1 inhibited cell proliferation and promoted cisplatin sensitivity in SCLC and elucidate a novel mechanism through RHBDF1/YAP1/Smad2 signaling pathway which played a crucial role in cisplatin chemosensitivity. Targeting this pathway can be a promising therapeutic strategy for chemotherapy resistance in SCLC.

Identification of mutational signature for lung adenocarcinoma prognosis and immunotherapy prediction.

There is no robust genomic signature to predict the prognosis of patients with early-stage lung adenocarcinoma (LUAD). It was known that clonal heterogeneity was closely associated to tumour progression and prognosis prediction. Herein, using stage I patients from The Cancer Genome Atlas, we identified the clonal/subclonal events of each gene and preselected a set of genes with prognosis-specific mutation patterns based on a robust published transcriptomic prognostic signature. Subsequently, we constructed a mutational prognostic signature (MPS), whose prognostic performance was independently validated in two datasets of stage I samples. The predicted high-risk patients had significantly higher immune cell infiltration, along with higher expression of cytotoxic and immune checkpoint genes, and an integrated dataset with 88 samples confirmed that high-risk patients could benefit from immunotherapy. The developed MPS can identify the high-risk patients with stage I LUAD and improve individualised treatment planning of high-risk patients who might benefit from immunotherapy. KEY MESSAGES: We creatively developed a prognostic signature (57-MPS) based on clonal diversity. The high-risk samples displayed an underlying immunosuppressive mechanism. 57-MPS improved the predictive performance of PD-L1 for immunotherapy.

Hierarchical identification of a transcriptional panel for the histological diagnosis of lung neuroendocrine tumors.

Background: Lung cancer is a complex disease composed of neuroendocrine (NE) and non-NE tumors. Accurate diagnosis of lung cancer is essential in guiding therapeutic management. Several transcriptional signatures have been reported to distinguish between adenocarcinoma (ADC) and squamous cell carcinoma (SCC) belonging to non-NE tumors. This study aims to identify a transcriptional panel that could distinguish the histological subtypes of NE tumors to complement the morphology-based classification of an individual. Methods: A public dataset with NE subtypes, including 21 small-cell lung cancer (SCLC), 56 large-cell NE carcinomas (LCNECs), and 24 carcinoids (CARCIs), and non-NE subtypes, including 85 ADC and 61 SCC, was used as a training set. In the training set, consensus clustering was first used to filter out the samples whose expression patterns disagreed with their histological subtypes. Then, a rank-based method was proposed to develop a panel of transcriptional signatures for determining the NE subtype for an individual, based on the within-sample relative gene expression orderings of gene pairs. Twenty-three public datasets with a total of 3,454 samples, which were derived from fresh-frozen, formalin-fixed paraffin-embedded, biopsies, and single cells, were used for validation. Clinical feasibility was tested in 10 SCLC biopsy specimens collected from cancer hospitals via bronchoscopy. Results: The NEsubtype-panel was composed of three signatures that could distinguish NE from non-NE, CARCI from non-CARCI, and SCLC from LCNEC step by step and ultimately determine the histological subtype for each NE sample. The three signatures achieved high average concordance rates with 97.31%, 98.11%, and 90.63%, respectively, in the 23 public validation datasets. It is worth noting that the 10 clinic-derived SCLC samples diagnosed via immunohistochemical staining were also accurately predicted by the NEsubtype-panel. Furthermore, the subtype-specific gene expression patterns and survival analyses provided evidence for the rationality of the reclassification by the NEsubtype-panel. Conclusion: The rank-based NEsubtype-panel could accurately distinguish lung NE from non-NE tumors and determine NE subtypes even in clinically challenging samples (such as biopsy). The panel together with our previously reported signature (KRT5-AGR2) for SCC and ADC would be an auxiliary test for the histological diagnosis of lung cancer.

Identifying metabolic reprogramming phenotypes with glycolysis-lipid metabolism discoordination and intercellular communication for lung adenocarcinoma metastasis.

Tumor metastasis imposes metabolic requirements for escaping from primary tissues, producing vulnerability in treatment. This study aimed to explore the metabolic reprogramming relevant to lung adenocarcinoma (LUAD) metastasis and decode the underlying intercellular alterations. Using the gene expression profiles of 394 LUAD samples derived from The Cancer Genome Atlas (TCGA), we identified 11 metastasis-related metabolic genes involved in glycolysis and lipid metabolism, and defined three metabolic reprogramming phenotypes (MP-I, -II, and -III) using unsupervised clustering. MP-III with the highest glycolytic and lowest lipid metabolic levels exhibited the highest metastatic potency and poorest survival in TCGA and six independent cohorts totaling 1,235 samples. Genomic analyses showed that mutations in TP53 and KEAP1, and deletions in SETD2 and PBRM1 might drive metabolic reprogramming in MP-III. Single-cell RNA-sequencing data from LUAD validated a metabolic evolutionary trajectory from normal to MP-II and MP-III, through MP-I. The further intercellular communications revealed that MP-III interacted uniquely with endothelial cells and fibroblasts in the ANGPTL pathway, and had stronger interactions with endothelial cells in the VEGF pathway. Herein, glycolysis-lipid dysregulation patterns suggested metabolic reprogramming phenotypes relevant to metastasis. Further insights into the oncogenic drivers and microenvironmental interactions would facilitate the treatment of LUAD metastasis in the future.

The value of AGR2 and KRT5 as an immunomarker combination in distinguishing lung squamous cell carcinoma from adenocarcinoma.

With the advancement of tumor subtype-specific treatments, precise histopathologic distinction between adenocarcinoma (ADC) and squamous cell carcinoma (SCC) is of significant clinical importance. Nevertheless, the current markers are insufficiently precise in poorly differentiated tissue. This study aimed to establish a histology-specific immunomarker combination to subclassify non-small cell lung cancer (NSCLC) specimens. Based on previous work, we assessed the differential expression of anterior gradient 2 (AGR2) and keratin 5 (KRT5) in ADC and SCC by analyzing public datasets and postoperative specimens. Subsequently, we established a train set (n = 188) and a validation set (n = 42) comprised of NSCLC surgical specimens for training and verifying the subtype-identification capabilities of the two biomarkers separately and in combination, and contrasted the diagnostic utility of AGR2-KRT5 with that of the classic immunomarker combination, TTF1-P40. Differential expression of the two genes was statistically significant in ADC and SCC samples, both at the mRNA and protein levels. The specificity and sensitivity of AGR2 to detect ADC in the training set were 97.0% and 94.4%, while the sensitivity and specificity of KRT5 to determine SCC were 93.9% and 98.9%, respectively. The accuracies of AGR2-KRT5 in ADC, SCC, and across all samples were 93.3%, 92.0% and 92.6% respectively. In the validation cohort, the predictive accuracy of AGR2-KRT5 was up to 100% for ADC and 86.7% for SCC. Compared with TTF1-P40 in ADC samples, AGR2-KRT5 had 8.4% higher accuracy. In summary, the AGR2-KRT5 immunomarker combination reliably distinguished SCC from ADC, and was more accurate than TTF1-P40 in ADC.