Next generation sequencing (NGS) is being increasingly applied for assisting cancer molecular diagnosis. However, it is still needed to validate NGS accuracy on detection of DNA alternations based on a large number of clinical samples, especially for DNA rearrangements and copy number variations (CNVs). This study is to set up basic parameters of targeted NGS for clinical diagnosis and to understand advantage of targeted NGS in comparison with the conventional methods of molecular diagnosis. Genomic DNA from 1000 Genomes Project and DNA from cancer cell lines have been used to establish the basic parameters for targeted NGS. The following confirmation was conducted by clinical samples. The multiple variants tested by amplification-refractory mutation system (ARMS), fluorescence in situ hybridization (FISH) and immunohistochemistry (IHC) were evaluated by targeted NGS to determine the sensitivity. Furthermore, the multiple variants detected by targeted NGS were confirmed by current conventional methods to elucidate the specificity. At sequencing depth of 500×, the maximal sensitivities on detecting single nucletic variances (SNVs) and small insertions/deletions (Indels) can reach 99% and 98.7% respectively, and in 20% of cancer cells, CNV detection can reach to the maximal level. The following confirmation of the sensitivity and specificity was conducted by a large cohort of clinical samples. For SNV and indel detection in clinical samples, targeted NGS can identify all hotspot mutations with 100% sensitivity and specificity. On ALK fusion detection, about 86% IHC-identified cases could be identified by targeted NGS and all ALK fusion detected by targeted NGS were confirmed by IHC. For HER2-amplification, 14 HER2-amplification cases identified by target NGS were all confirmed by FISH and about 93.3% of Her-2 IHC (3+) cases were identified by targeted NGS. Finally, the targeted NGS platform developed here has accurately detected EGFR hotspot mutations in 215 NSCLC patients. DNA from cancer cell lines is better than standard DNA as a reference to establish basic parameters for targeted NGS. Comparison of the conventional methods using a large cohort of patient samples confirmed the high preformance of targeted NGS on detecting DNA alterations.
Programmed cell death ligand 1 (PD-L1) is a critical biomarker for predicting the response to immunotherapy. However, traditional quantitative evaluation of PD-L1 expression using immunohistochemistry staining remains challenging for pathologists. Here we developed a deep learning (DL)-based artificial intelligence (AI) model to automatically analyze the immunohistochemical expression of PD-L1 in lung cancer patients. A total of 1,288 patients with lung cancer were included in the study. The diagnostic ability of three different AI models (M1, M2, and M3) was assessed in both PD-L1 (22C3) and PD-L1 (SP263) assays. M2 and M3 showed improved performance in the evaluation of PD-L1 expression in the PD-L1 (22C3) assay, especially at 1% cutoff. Highly accurate performance in the PD-L1 (SP263) was also achieved, with accuracy and specificity of 96.4 and 96.8% in both M2 and M3, respectively. Moreover, the diagnostic results of these three AI-assisted models were highly consistent with those from the pathologist. Similar performances of M1, M2, and M3 in the 22C3 dataset were also obtained in lung adenocarcinoma and lung squamous cell carcinoma in both sampling methods. In conclusion, these results suggest that AI-assisted diagnostic models in PD-L1 expression are a promising tool for improving the efficiency of clinical pathologists.
While methods for detecting SNVs and indels in circulating tumor DNA (ctDNA) with hybridization capture-based next-generation sequencing (NGS) have been available, copy number variations (CNVs) detection is more challenging. Here, we present a method enabling CNV detection from a 150-gene panel using a very low amount of ctDNA. First, a read depth-based CNV estimation method without a paired blood sample was developed and cfDNA sequencing data from healthy people were used to build a panel of normal (PoN) model. Then, in silico and in vitro simulations were performed to define the limit of detection (LOD) for EGFR, ERBB2, and MET. Compared to the WES results of the 48 samples, the concordance rate for EGFR, ERBB2, and MET CNVs was 78%, 89.6%, and 92.4%, respectively. In another cohort profiled with the 150-gene panel from 5980 lung cancer ctDNA samples, we detected the three genes’ amplification with comparable population frequency with other cohorts. One lung adenocarcinoma patient with MET amplification detected by our method reached partial response to crizotinib. These findings show that our ctDNA CNV detection pipeline can detect CNVs with high specificity and concordance, which enables CNV calling in a non-invasive way for cancer patients when tissues are not available.
Background: The identification of indeterminate pulmonary nodules (IPNs) following a low-dose computed tomography (LDCT) is a major challenge for early diagnosis of lung cancer. The inadequate assessment of IPNs’ malignancy risk results in a large number of unnecessary surgeries or an increased risk of cancer metastases. However, limited studies on non-invasive diagnosis of IPNs have been reported. Methods: In this study, we identified and evaluated the diagnostic value of circulating small extracellular vesicle (sEV) microRNAs (miRNAs) in patients with IPNs that had been newly detected using LDCT scanning and were scheduled for surgery. Out of 459 recruited patients, 109 eligible patients with IPNs were enrolled in the training cohort (n = 47) and the test cohort (n = 62). An external cohort (n = 99) was used for validation. MiRNAs were extracted from plasma sEVs, and assessed using Small RNA sequencing. 490 lung adenocarcinoma samples and follow-up data were used to investigate the role of miRNAs in overall survival. Results: A circulating sEV miRNA (CirsEV-miR) model was constructed from five differentially expressed miRNAs (DEMs), showing 0.920 AUC in the training cohort (n = 47), and further identified in the test cohort (n = 62) and in an external validation cohort (n = 99). Among five DEMs of the CirsEV-miR model, miR-101-3p and miR-150-5p were significantly associated with better overall survival (p = 0.0001 and p = 0.0069). The CirsEV-miR scores were calculated, which significantly correlated with IPNs diameters (p < 0.05), and were able to discriminate between benign and malignant PNs (diameter ≤ 1 cm). The expression patterns of sEV miRNAs in the benign, adenocarcinoma in situ/minimally invasive adenocarcinoma, and invasive adenocarcinoma subgroups were found to gradually change with the increase in aggressiveness for the first time. Among all DEMs of the three subgroups, five miRNAs (miR-30c-5p, miR-30e-5p, miR-500a-3p, miR-125a-5p, and miR-99a-5p) were also significantly associated with overall survival of lung adenocarcinoma patients. Conclusions: Our results indicate that the CirsEV-miR model could help distinguish between benign and malignant PNs, providing insights into the feasibility of circulating sEV miRNAs in diagnostic biomarker development. Trial registration: Chinese Clinical Trials: ChiCTR1800019877. Registered 05 December 2018, https://www.chictr.org.cn/showproj.aspx?proj=31346. Graphical
10 September 2021 in Clinical and Translational Medicine
Background: The differential diagnosis of pancreatic ductal adenocarcinoma (PDAC) from chronic pancreatitis (CP) is clinically challenging due to a lack of minimally invasive diagnosis methods. MicroRNAs (miRNAs) derived from small extracellular vesicles (EVs) in the blood have been reported as a promising diagnosis biomarker for various types of cancer. However, blood small EV miRNA signatures and their diagnostic value to differentiate between PDAC and CP remain to be determined.Methods: In this study, 107 patients with PDAC or CP were recruited, and 90 patients were finally enrolled for a training cohort (n = 48) and test cohort (n = 42). Small RNA sequencing was used to assess the expression of blood small EV miRNAs in these patients.Results: The linear model from the differentially expressed blood small EV miR‐95‐3p divided by miR‐26b‐5p showed an average sensitivity of 84.1% and an average specificity of 96.6% to identify PDAC from CP in the training cohort and the test cohort, respectively. When the model was combined with serum carbohydrate antigen 19‐9 (CA19‐9), the average sensitivity increased to 96.5%, and the average specificity remained at 96.4% of both cohorts, which demonstrated the best performance of all the published biomarkers for distinguishing between PDAC and CP. The causal analysis performed using the Bayesian network demonstrated that miR‐95‐3p was associated with a “consequence” of “cancer” and miR‐26b‐5p as a “cause” of “pancreatitis.” A subgroup analysis revealed that blood small EV miR‐335‐5p/miR‐340‐5p could predict metastases in both cohorts and was associated with an overall survival (p = 0.020).Conclusions: This study indicated that blood small EV miR‐95‐3p/miR‐26b‐5p and its combination with serum levels of CA19‐9 could separate PDAC from CP, and miR‐335‐5p/miR‐340‐5p was identified to associate with PDAC metastasis and poor prognosis. These results suggested the potentiality of blood small EV miRNAs as differential diagnosis and metastases biomarkers of PDAC.
Background: The characteristics of head and neck squamous cell carcinoma (HNSCC) across different anatomic sites in the Chinese population have not been studied. To determine the genomic abnormalities underlying HNSCC across different anatomic sites, the alterations of selected cancer-related genes were evaluated.Methods: Genomic DNA samples obtained from formalin-fixed, paraffin-embedded tissues were analyzed using targeted sequencing in a panel of 383 cancer-related genes to determine the genomic alterations.Results: A total of 317 formalin-fixed, paraffin-embedded HNSCC specimens were collected, and a total of 2,156 protein-coding mutations, including 1,864 single nucleotide variants and 292 insertions and deletions, were identified across more than six different anatomic sites. Mutation loads were distinct across the anatomic sites. Larynx carcinoma was found with the highest mutation loads, whereas nasopharynx carcinoma showed the lowest mutation loads. A total of 1,110 gains and 775 losses were identified in the 317 specimens. Patients who had at least one clinically actionable alteration (levels 1–4 in OncoKB) were identified. One patient had an actionable alteration with level 1 evidence in OncoKB, TEX10-NTRK2 fusion, who may benefit from larotrectinib or entrectinib treatment.Conclusion: The genomic profiling of HNSCC using targeted sequencing can identify rational therapeutic candidate genes suitable for the treatment of the HNSCCs.