Discrimination of ground-glass nodular lung adenocarcinoma pathological subtypes via transfer learning: A multicenter study.

BACKGROUND: The surgical approach and prognosis for invasive adenocarcinoma (IAC) and minimally invasive adenocarcinoma (MIA) of the lung differ. However, they both manifest as identical ground-glass nodules (GGNs) in computed tomography images, and no effective method exists to discriminate them. METHODS: We developed and validated a three-dimensional (3D) deep transfer learning model to discriminate IAC from MIA based on CT images of GGNs. This model uses a 3D medical image pre-training model (MedicalNet) and a fusion model to build a classification network. Transfer learning was utilized for end-to-end predictive modeling of the cohort data of the first center, and the cohort data of the other two centers were used as independent external validation data. This study included 999 lung GGN images of 921 patients pathologically diagnosed with IAC or MIA at three cohort centers. RESULTS: The predictive performance of the model was assessed using the area under the receiver operating characteristic curve (AUC). The model had high diagnostic efficacy for the training and validation groups (accuracy: 89%, sensitivity: 95%, specificity: 84%, and AUC: 95% in the training group; accuracy: 88%, sensitivity: 84%, specificity: 93%, and AUC: 92% in the internal validation group; accuracy: 83%, sensitivity: 83%, specificity: 83%, and AUC: 89% in one external validation group; accuracy: 78%, sensitivity: 80%, specificity: 77%, and AUC: 82% in the other external validation group). CONCLUSIONS: Our 3D deep transfer learning model provides a noninvasive, low-cost, rapid, and reproducible method for preoperative prediction of IAC and MIA in lung cancer patients with GGNs. It can help clinicians to choose the optimal surgical strategy and improve the prognosis of patients.

Predictive Efficacy of a Radiomics Random Forest Model for Identifying Pathological Subtypes of Lung Adenocarcinoma Presenting as Ground-Glass Nodules.

Purpose: To establish and verify the ability of a radiomics prediction model to distinguish invasive adenocarcinoma (IAC) and minimal invasive adenocarcinoma (MIA) presenting as ground-glass nodules (GGNs). Methods: We retrospectively analyzed 118 lung GGN images and clinical data from 106 patients in our hospital from March 2016 to April 2019. All pathological classifications of lung GGN were confirmed as IAC or MIA by two pathologists. R language software (version 3.5.1) was used for the statistical analysis of the general clinical data. ITK-SNAP (version 3.6) and A.K. software (Analysis Kit, American GE Company) were used to manually outline the regions of interest of lung GGNs and collect three-dimensional radiomics features. Patients were randomly divided into training and verification groups (ratio, 7:3). Random forest combined with hyperparameter tuning was used for feature selection and prediction modeling. The receiver operating characteristic curve and the area under the curve (AUC) were used to evaluate model prediction efficacy. The calibration curve was used to evaluate the calibration effect. Results: There was no significant difference between IAC and MIA in terms of age, gender, smoking history, tumor history, and lung GGN location in both the training and verification groups (P>0.05). For each lung GGN, the collected data included 396 three-dimensional radiomics features in six categories. Based on the training cohort, nine optimal radiomics features in three categories were finally screened out, and a prediction model was established. We found that the training group had a high diagnostic efficacy [accuracy, sensitivity, specificity, and AUC of the training group were 0.89 (95%CI, 0.73 - 0.99), 0.98 (95%CI, 0.78 - 1.00), 0.81 (95%CI, 0.59 - 1.00), and 0.97 (95%CI, 0.92-1.00), respectively; those of the validation group were 0.80 (95%CI, 0.58 - 0.93), 0.82 (95%CI, 0.55 - 1.00), 0.78 (95%CI, 0.57 - 1.00), and 0.92 (95%CI, 0.83 - 1.00), respectively]. The model calibration curve showed good consistency between the predicted and actual probabilities. Conclusions: The radiomics prediction model established by combining random forest with hyperparameter tuning effectively distinguished IAC from MIA presenting as GGNs and represents a noninvasive, low-cost, rapid, and reproducible preoperative prediction method for clinical application.

Hypermethylation of mismatch repair gene hMSH2 associates with platinum-resistant disease in epithelial ovarian cancer.

PURPOSE: One major reason of the high mortality of epithelial ovarian cancer (EOC) is due to platinum-based chemotherapy resistance. Aberrant DNA methylation may be a potential mechanism underlying the development of platinum resistance in EOC. The purpose of this study is to discover potential aberrant DNA methylation that contributes to drug resistance. METHODS: By initially screening of 16 platinum-sensitive/resistant samples from EOC patients with reduced representation bisulfite sequencing (RRBS), the upstream region of the hMSH2 gene was discovered hypermethylated in the platinum-resistant group. The effect of hMSH2 methylation on the cellular response to cisplatin was explored by demethylation and knockdown assays in ovarian cancer cell line A2780. Matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry was employed to examine the methylation levels of hMSH2 upstream region in additional 40 EOC patient samples. RT-qPCR and IHC assay was used to detect the hMSH2 mRNA and protein expression in extended 150 patients. RESULTS: RRBS assay discovered an upstream region from - 1193 to - 1125 of hMSH2 was significant hypermethylated in resistant EOC patients (P = 1.06 × 10-14). In vitro analysis demonstrated that global demethylation increased cisplatin sensitivity along with a higher expression of the hMSH2 mRNA and protein. Knockdown hMSH2 reduced the cell sensitivity to cisplatin. MALDI-TOF mass spectrometry assay validated the strong association of hypermethylation of hMSH2 upstream region with platinum resistance. Spearman's correlation analysis revealed a significantly negative connection between methylation level of hMSH2 upstream region and its expression. The Kaplan-Meier analyses showed the high methylation of hMSH2 promoter region, and its low expressions are associated with worse survival. In multivariable models, hMSH2 low expression was an independent factor predicting poor outcome (P = 0.03, HR = 1.91, 95%CI = 1.85-2.31). CONCLUSION: The hypermethylation of hMSH2 upstream region is associated with platinum resistant in EOC, and low expression of hMSH2 may be an index for the poor prognosis.

Genome-wide association studies identify susceptibility loci for epithelial ovarian cancer in east Asian women.

OBJECTIVE: Genome-wide association studies (GWASs) for epithelial ovarian cancer (EOC) have focused largely on populations of European ancestry. We aimed to identify common germline variants associated with EOC risk in Asian women. METHODS: Genotyping was performed as part of the OncoArray project. Samples with >60% Asian ancestry were included in the analysis. Genotyping was performed on 533,631 SNPs in 3238 Asian subjects diagnosed with invasive or borderline EOC and 4083 unaffected controls. After imputation, genotypes were available for 11,595,112 SNPs to identify associations. RESULTS: At chromosome 6p25.2, SNP rs7748275 was associated with risk of serous EOC (odds ratio [OR] = 1.34, P = 8.7 × 10-9) and high-grade serous EOC (HGSOC) (OR = 1.34, P = 4.3 × 10-9). SNP rs6902488 at 6p25.2 (r2 = 0.97 with rs7748275) lies in an active enhancer and is predicted to impact binding of STAT3, P300 and ELF1. We identified additional risk loci with low Bayesian false discovery probability (BFDP) scores, indicating they are likely to be true risk associations (BFDP <10%). At chromosome 20q11.22, rs74272064 was associated with HGSOC risk (OR = 1.27, P = 9.0 × 10-8). Overall EOC risk was associated with rs10260419 at chromosome 7p21.3 (OR = 1.33, P = 1.2 × 10-7) and rs74917072 at chromosome 2q37.3 (OR = 1.25, P = 4.7 × 10-7). At 2q37.3, expression quantitative trait locus analysis in 404 HGSOC tissues identified ESPNL as a putative candidate susceptibility gene (P = 1.2 × 10-7). CONCLUSION: While some risk loci were shared between East Asian and European populations, others were population-specific, indicating that the landscape of EOC risk in Asian women has both shared and unique features compared to women of European ancestry.

The Associations of Genetic Variants in E-cadherin Gene With Clinical Outcome of Epithelial Ovarian Cancer.

OBJECTIVE: The E-cadherin protein plays major roles in tumor progression, invasion, and metastasis. Polymorphisms located in the E-cadherin gene (CDH1) may contribute to increased risks of specific cancers. In this study, we evaluated the associations between genetic variants in CDH1 and the clinical outcomes of patients with epithelial ovarian cancer (EOC). MATERIALS AND METHODS: We assessed the -160C/A and -347G/GA polymorphisms in the promoter region, as well as the 3'-UTR +54C/T polymorphism of E-cadherin, in 257 patients with EOC by ligase detection reaction-polymerase chain reaction. RESULTS: Multivariate analysis showed that patients with EOC with the CDH1 -347GA/GA genotype had shorter progression-free survival and overall survival (hazard ratio [HR], 2.16; 95% confidence interval [CI], 1.06-4.40 and HR, 2.06; 95% CI, 1.01-4.19, respectively) compared to those carrying the G/G genotype. Likewise, the patients with the CDH1 -160A/A genotype had a shorter progression-free survival than those with the C/C genotype (HR, 4.12; 95% CI, 1.43-111.88). No significant association was detected between the CDH1 3'-UTR +54C/T polymorphism and survival of the patients with EOC. CONCLUSIONS: The CDH1 -347GA/GA and -160A/A genotypes may be prognostic markers that can help to identify patients at increased risk of invasive/metastatic cancer in northern China.

Association between polymorphisms of ERCC1 and survival in epithelial ovarian cancer patients with chemotherapy.

AIM: We evaluated whether the ERCC1 polymorphisms had an effect on survival in epithelial ovarian cancer (EOC) patients with platinum-based chemotherapy. MATERIALS & METHODS: Clinical data of 209 EOC patients between 2002 and 2008 were reviewed. The genotypes of 19007T/C and 8092C/A polymorphisms were assessed in all patients using PCR-RFLP. RESULTS: The 19007T/C polymorphism was significantly associated with response to treatment. Compared with the patients carrying C/C genotype, the patients with the T/T genotype have a significantly decreased response to platinum-based chemotherapy (odds ratio: 32.26; 95% CI: 3.66-250.00). Cox's multivariate analysis suggested that EOC patients with the T/T genotype had an increased risk of disease progression (hazard ratio: 3.34; 95% CI: 1.77-6.29) and death (hazard ratio: 2.87; 95% CI: 1.38-5.96) compared with those carrying the C/C genotype. CONCLUSION: The 19007T/C polymorphism may be a useful prognostic marker in patients with EOC treated with platinum-based chemotherapy in Chinese women.

Association of three single nucleotide polymorphisms of the E-cadherin gene with endometriosis in a Chinese population.

Endometriosis, one of the most frequent diseases in gynecology, is a benign but invasive and metastatic disease. The altered expression of E-cadherin may play an important role in developing endometriosis. In this paper, we discuss the association of three single nucleotide polymorphisms (SNPs) on the E-cadherin gene and risk of endometriosis. We examined the genotype frequency of three polymorphisms in 152 endometriosis patients and 189 control women. There was a significant difference in the frequency of the E-cadherin 3'-UTR C --> T genotypes between endometriosis and controls (P = 0.01). The frequency of the C allele in patients (71.1%) was significantly higher than in the controls (63.8%; P = 0.04). When compared with the T/T + T/C genotypes, the C/C genotype had a significantly increased susceptibility to endometriosis, with an adjusted odds ratio of 1.79 (95% confidence interval = 1.17-2.76). No significant difference was found between endometriosis and control women on two polymorphisms (-160 C --> A, -347 G --> GA) at the gene promoter region of E-cadherin. The -160 C --> A and -347 G --> GA polymorphisms displayed linkage disequilibrium (D' = 0.999). The -160 A/-347 GA haplotype was only detected in endometriosis patients (2%). These data show a relation between the E-cadherin 3'-UTR C --> T polymorphism, the -160 A/-347 GA haplotype of two promoter polymorphisms and risk of endometriosis, suggesting a potential role in endometriosis development, at least in North Chinese women.

Polymorphisms in the promoter regions of the matrix metalloproteinases-7, -9 and the risk of endometriosis and adenomyosis in China.

Matrix metalloproteinases (MMPs) may contribute to the development of endometriosis. The aim of this study was to assess the effects of the polymorphisms in the promoters of MMP-7 (181A/G) and MMP-9 (1562C/T) on the risk of occurrence of endometriosis and adenomyosis. We genotyped 219 patients (143 women with endometriosis, 76 women with adenomyosis) and 160 control women in North China. There was a significant difference in frequency of the MMP-7 genotype between endometriosis and controls (P = 0.01) and also between adenomyosis and controls (P = 0.01). The frequency of the G allele in two groups of patients (7.3 and 7.9%) was significantly higher than in the controls (2.8%) (P = 0.01 and 0.01, respectively). Compared to the A/A genotype, the genotype with the -181G allele showed a significantly increased susceptibility to both diseases, with adjusted odds ratio of 2.62 [95% confidence interval (CI) = 1.17-5.87] for endometriosis and 3.14 (95% CI = 1.26-7.81) for adenomyosis. However, the overall genotype and allelotype distribution of the MMP-9 in the two case groups were not different from that of controls. We conclude that MMP-7-181A/G polymorphism has a potential to be a susceptibility factor for endometriosis and adenomyosis while MMP-9-1562C/T polymorphism may not provide a useful marker to predict susceptibility to endometriosis and adenomyosis, at least in women from North China.

The function of the SNP in the MMP1 and MMP3 promoter in susceptibility to endometriosis in China.

Matrix metalloproteinases (MMPs) may contribute to the development of endometriosis. Genetic variations in several MMP promoters may influence the transcription and expression of MMPs. The purpose of the present study was to assess how gene polymorphisms in the MMP1 and MMP3 promoters affect the risk of development of endometriosis. We genotyped 100 women with endometriosis and 150 control subjects in North China. There was a significant difference in frequency of the MMP1 genotype between cases and controls (P=0.03). The 2G homozygote in endometriosis and controls was significantly different (P=0.02). The frequency of the 2G allele among affected women (79%) was significantly higher than among the healthy controls (66.9%; P=0.003). However, the overall genotype and allelotype distribution of the MMP3 single nucleotide polymorphism (SNP) in patients was not different from that of controls (P> or =0.05). MMP1 and MMP3 polymorphisms were in linkage disequilibrium in cases and controls (D'=0.47; P=0.00). The haplotype frequency distribution derived from these two polymorphisms was significantly different between cases and controls (P=0.00). The haplotype analysis suggested an implication of both MMP1 and MMP3 polymorphisms in the susceptibility to endometriosis. We conclude that the MMP1 promoter SNP and MMP 2G/6A haplotype may modify susceptibility to endometriosis, but that the MMP3 promoter SNP is unlikely to be associated with endometriosis in the population of North China.

Enhanced expression of telomerase activity in thymoma and thymic carcinoma tissues: a clinicopathologic study.

BACKGROUND: Telomerase is a nucleoprotein complex that caps the physical termini of all eukaryotic chromosomes. Because most malignant cells and reproductive cells have telomerase activity, which elongates telomeric DNA, telomerase may play important roles in unlimited cell division acquisition of the malignant phenotype. The current study examined the relation of telomerase activity in thymoma and thymic carcinoma with the clinicopathologic features of these lesions. METHODS: Tissue specimens were surgically resected from patients with thymoma and thymic carcinoma. Telomerase activity was evaluated according to a modified telomeric repeat amplification protocol assay. Paraffin sections of tumor were immunostained by MIC2 antibody, a marker of immature T cells. RESULTS: Telomerase activity was detected in all thymic epithelial tumors. The activity (mean +/- SD; unit per microg protein) in thymoma (n = 17) was significantly higher than that in thymic carcinoma (n = 7) (431.8 +/- 400.1 vs. 68.8 +/- 39.8; P < 0.01). Telomerase activities in thymoma and thymic carcinoma were significantly higher than that in primary lung adenocarcinoma (33.5 +/- 39.2, n = 47), studied as a control (P < 0.01). In patients with thymoma, telomerase activity did not correlate with tumor stage according to Masaoka classification (P = 0.776). In patients with thymic carcinoma, however, telomerase activity positively correlated with tumor stage (P = 0.02). In thymoma, telomerase activity positively correlated with the ratio of induced lymphocytes according to Rosai's classification (P = 0.045). MIC2-positive lymphocytes were identified in all cases of thymoma (n = 12). In contrast, lymphocytes infiltrating thymic carcinoma did not react with MIC2. CONCLUSIONS: In thymoma, telomerase activity reflects the presence of immature T-cell lymphocytes in tumor tissue rather than tumor stage or malignant phenotype. In thymic carcinoma, telomerase activity derived directly from cancer cells may relate to tumor stage.