US of thyroid nodules: can AI-assisted diagnostic system compete with fine needle aspiration?

OBJECTIVES: Artificial intelligence (AI) systems can diagnose thyroid nodules with similar or better performance than radiologists. Little is known about how this performance compares with that achieved through fine needle aspiration (FNA). This study aims to compare the diagnostic yields of FNA cytopathology alone and combined with BRAFV600E mutation analysis and an AI diagnostic system. METHODS: The ultrasound images of 637 thyroid nodules were collected in three hospitals. The diagnostic efficacies of an AI diagnostic system, FNA-based cytopathology, and BRAFV600E mutation analysis were evaluated in terms of sensitivity, specificity, accuracy, and the κ coefficient with respect to the gold standard, defined by postsurgical pathology and consistent benign outcomes from two combined FNA and mutation analysis examinations performed with a half-year interval. RESULTS: The malignancy threshold for the AI system was selected according to the Youden index from a retrospective cohort of 346 nodules and then applied to a prospective cohort of 291 nodules. The combination of FNA cytopathology according to the Bethesda criteria and BRAFV600E mutation analysis showed no significant difference from the AI system in terms of accuracy for either cohort in our multicenter study. In addition, for 45 included indeterminate Bethesda category III and IV nodules, the accuracy, sensitivity, and specificity of the AI system were 84.44%, 95.45%, and 73.91%, respectively. CONCLUSIONS: The AI diagnostic system showed similar diagnostic performance to FNA cytopathology combined with BRAFV600E mutation analysis. Given its advantages in terms of operability, time efficiency, non-invasiveness, and the wide availability of ultrasonography, it provides a new alternative for thyroid nodule diagnosis. CLINICAL RELEVANCE STATEMENT: Thyroid ultrasonic artificial intelligence shows statistically equivalent performance for thyroid nodule diagnosis to FNA cytopathology combined with BRAFV600E mutation analysis. It can be widely applied in hospitals and clinics to assist radiologists in thyroid nodule screening and is expected to reduce the need for relatively invasive FNA biopsies. KEY POINTS: • In a retrospective cohort of 346 nodules, the evaluated artificial intelligence (AI) system did not significantly differ from fine needle aspiration (FNA) cytopathology alone and combined with gene mutation analysis in accuracy. • In a prospective multicenter cohort of 291 nodules, the accuracy of the AI diagnostic system was not significantly different from that of FNA cytopathology either alone or combined with gene mutation analysis. • For 45 indeterminate Bethesda category III and IV nodules, the AI system did not perform significantly differently from BRAFV600E mutation analysis.

Inhibition of histamine receptor H3 suppresses the growth and metastasis of human non-small cell lung cancer cells via inhibiting PI3K/Akt/mTOR and MEK/ERK signaling pathways and blocking EMT.

Recent evidence shows that the expression levels of histamine receptor H3 (Hrh3) are upregulated in several types of cancer. However, the role of Hrh3 in non-small cell lung cancer (NSCLC) has not been elucidated. In the present study, we showed that the expression levels of Hrh3 were significantly increased in NSCLC samples, and high levels of Hrh3 were associated with poor overall survival (OS) in NSCLC patients. In five human NSCLC cell lines tested, Hrh3 was significantly upregulated. In NSCLC cell lines H1975, H460, and A549, Hrh3 antagonist ciproxifan (CPX, 10-80 µM) exerted moderate and concentration-dependent inhibition on the cell growth and induced apoptosis, whereas its agonist RAMH (80 µM) reversed these effects. Furthermore, inhibition of Hrh3 by CPX or siRNA retarded the migration and invasion of NSCLC cells through inhibiting epithelial-mesenchymal transition (EMT) progression via reducing the phosphorylation of PI3K/Akt/mTOR and MEK/ERK signaling pathways. In nude mice bearing H1975 cell xenograft or A549 cell xenograft, administration of CPX (3 mg/kg every other day, intraperitoneal) significantly inhibited the tumor growth with increased E-cadherin and ZO-1 expression and decreased Fibronectin expression in tumor tissue. In conclusion, this study reveals that Hrh3 plays an important role in the growth and metastasis of NSCLC; it might be a potential therapeutic target against the lung cancer.

Association between plasma somatic copy number variations and response to immunotherapy in patients with programmed death-ligand 1-negative non-small cell lung cancer.

OBJECTIVE: To determine how patients with non-small cell lung cancer (NSCLC) with programmed death-ligand 1 (PD-L1)-negative and/or a low tumor mutation burden status benefit from immune checkpoint inhibitors (ICI). METHODS: We determined the plasma cell-free DNA profiles of 25 patients with PD-L1-negative advanced NSCLC before ICI therapy using low-coverage whole-genome sequencing. RESULTS: Elevated cell-free copy number variations (CNVs) were associated with progressive disease, with a cutoff CNV score of 0.10 evaluated with an area under the curve of 0.790 in PD-L1-negative NSCLC. CNV changes were also correlated with poor survival. Progression-free survival and overall survival were both significantly shorter in CNVhigh compared with CNVlow patients. CONCLUSIONS: Cell-free CNV may be a useful peripheral blood biomarker for predicting the response to ICIs in patients with NSCLC.

Dysregulated Immune and Metabolic Microenvironment Is Associated with the Post-Operative Relapse in Stage I Non-Small Cell Lung Cancer.

The underlying mechanism of post-operative relapse of non-small cell lung cancer (NSCLC) remains poorly understood. We enrolled 57 stage I NSCLC patients with or without relapse and performed whole-exome sequencing (WES) and RNA sequencing (RNA-seq) on available primary and recurrent tumors, as well as on matched tumor-adjacent tissues (TATs). The WES analysis revealed that primary tumors from patients with relapse were enriched with USH2A mutation and 2q31.1 amplification. RNA-seq data showed that the relapse risk was associated with aberrant immune response and metabolism in the microenvironment of primary lesions. TATs from the patients with relapse showed an immunosuppression state. Moreover, recurrent lesions exhibited downregulated immune response compared with their paired primary tumors. Genomic and transcriptomic features were further subjected to build a prediction model classifying patients into groups with different relapse risks. We show that the recurrence risk of stage I NSCLC could be ascribed to the altered immune and metabolic microenvironment. TATs might be affected by cancer cells and facilitate the invasion of tumors. The immune microenvironment in the recurrent lesions is suppressed. Patients with a high risk of relapse need active post-operative intervention.

Correction to: Comparison of the Amplification Refractory Mutation System, Super Amplification Refractory Mutation System, and Droplet Digital PCR for T790 M Mutation Detection in Non-small Cell Lung Cancer after Failure of Tyrosine Kinase Inhibitor Treatment.

The affiliation of first author (Lucheng Zhu) should be Department of Oncology, Hangzhou Cancer Hospital and Department of Oncology, Hangzhou First People's Hospital, Nanjing Medical University.

Comparison of the Amplification Refractory Mutation System, Super Amplification Refractory Mutation System, and Droplet Digital PCR for T790 M Mutation Detection in Non-small Cell Lung Cancer after Failure of Tyrosine Kinase Inhibitor Treatment.

Plasma mutation detection has the advantages of non-invasiveness and accessibility. Here, we evaluated three methods, the amplification refractory mutation system (ARMS), second-generation ARMS (SuperARMS), and droplet digital PCR (ddPCR), to assess their concordance and feasibility for the detection of mutations in plasma samples. Non-small lung cancer patients with stage IIIB/IV that were resistant to epidermal growth factor receptor-tyrosine kinase inhibitor (EGFR-TKI) treatment were enrolled. Blood samples were collected within 14 days after TKI resistance. Each sample was simultaneously assessed by the three methods. In total, 169 patients were enrolled; 54.4% were female, 72.2% were diagnosed with stage IV disease; and 97.6% had adenocarcinoma. T790 M mutations were detected in 42 (24.8%) of the 169 samples using ARMS, one of which carried the T790 M alone, 22 that also encoded exon 19 deletions, and 19 with L858R mutations. For the SuperARMS assay, 59 (34.9%) samples exhibited the T790 M mutation, and 110 (65.1%) showed no detectable T790 M mutation. ddPCR showed that 61 (36.1%) samples contained the T790 M mutation, whereas 108 (63.9%) were not positive. T790 M abundance ranged from 0.04% to 38.2%. The median T790 M abundance was 0.15% for total samples and 2.98% for T790 M mutation samples. The overall concordance was 78.7% (133/169) among ARMS, SuperARMS, and ddPCR. Compared with patients with stage III disease, patients with stage IV disease exhibited a higher T790 M mutation detection rate (28.7% vs. 14.9% by ARMS; 37.7% vs. 27.7% by SuperARMS; and 41.8% vs. 21.3% by ddPCR). Liquid biopsy showed promise and has the advantages of non-invasiveness and accessibility. T790 M detection based on circulating tumor DNA showed high concordance. Compared with non-digital platforms, ddPCR showed higher sensitivity and provided both frequency and abundance information, which might be important for treatment decisions.