Combination of DCE-MRI and NME-DWI via Deep Neural Network for Predicting Breast Cancer Molecular Subtypes.

BACKGROUND: To explore whether the combination of dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) and nonmono-exponential (NME) model-based diffusion-weighted imaging (DWI) via deep neural network (DNN) can improve the prediction of breast cancer molecular subtypes compared to either imaging technique used alone. PATIENTS AND METHODS: This prospective study examined 480 breast cancers in 475 patients undergoing DCE-MRI and NME-DWI at 3.0 T. Breast cancers were classified as follows: human epidermal growth factor receptor 2 enriched (HER2-enriched), luminal A, luminal B (HER2-), luminal B (HER2+), and triple-negative subtypes. A total of 20% cases were withheld as an independent test dataset, and the remaining cases were used to train DNN with an 80% to 20% training-validation split and 5-fold cross-validation. The diagnostic accuracies of DNN in 5-way subtype classification between the DCE-MRI, NME-DWI, and their combined multiparametric-MRI datasets were compared using analysis of variance with least significant difference posthoc test. Areas under the receiver-operating characteristic curves were calculated to assess the performances of DNN in binary subtype classification between the 3 datasets. RESULTS: The 5-way classification accuracies of DNN on both DCE-MRI (0.71) and NME-DWI (0.64) were significantly lower (P < .05) than on multiparametric-MRI (0.76), while on DCE-MRI was significantly higher (P < .05) than on NME-DWI. The comparative results of binary classification between the 3 datasets were consistent with the 5-way classification. CONCLUSION: The combination of DCE-MRI and NME-DWI via DNN achieved a significant improvement in breast cancer molecular subtype prediction compared to either imaging technique used alone. Additionally, DCE-MRI outperformed NME-DWI in differentiating subtypes.

SSTU: Swin-Spectral Transformer U-Net for hyperspectral whole slide image reconstruction.

Whole Slide Imaging and Hyperspectral Microscopic Imaging provide great quality data with high spatial and spectral resolution for histopathology. Existing Hyperspectral Whole Slide Imaging systems combine the advantages of the techniques above, thus providing rich information for pathological diagnosis. However, it cannot avoid the problems of slow acquisition speed and mass data storage demand. Inspired by the spectral reconstruction task in computer vision and remote sensing, the Swin-Spectral Transformer U-Net (SSTU) has been developed to reconstruct Hyperspectral Whole Slide images (HWSis) from multiple Hyperspectral Microscopic images (HMis) of small Field of View and Whole Slide images (WSis). The Swin-Spectral Transformer (SST) module in SSTU takes full advantage of Transformer in extracting global attention. Firstly, Swin Transformer is exploited in space domain, which overcomes the high computation cost in Vision Transformer structures, while it maintains the spatial features extracted from WSis. Furthermore, Spectral Transformer is exploited to collect the long-range spectral features in HMis. Combined with the multi-scale encoder-bottleneck-decoder structure of U-Net, SSTU network is formed by sequential and symmetric residual connections of SSTs, which reconstructs a selected area of HWSi from coarse to fine. Qualitative and quantitative experiments prove the performance of SSTU in HWSi reconstruction task superior to other state-of-the-art spectral reconstruction methods.

Risk model-guided identification of MTDH expression as a marker for ferroptosis induction therapy in head and neck squamous cell carcinoma.

Head and neck squamous cell carcinoma (HNSCC) are a prevalent malignancy with high mortality and morbidity rates. Therefore, in this study, we aimed to develop a novel risk score model by using a DNA methylation signature associated with ferroptosis to enhance the prognosis prediction of HNSCC. The transcriptome, methylome, and clinical data of HNSCC patients were collected from The Cancer Genome Atlas (TCGA) database. Additionally, data from a methylation dataset in the Gene Expression Omnibus (GEO) database were used for validation. The ferroptosis score (FS) in each patient was calculated using the transcriptome data, and the single-sample gene set enrichment analysis (ssGSEA) was performed to assess ferroptosis activity. Furthermore, a series of biochemical experiments including CCK8, colony formation, wound healing, and ROS detection were carried out to evaluate the influence of MTDH on the malignancy of HNSCC. Our results revealed that the FS was associated with patient prognosis, as the patients with high FS had a poor prognosis. The receiver operating characteristic (ROC) curve established based on the ferroptosis-associated DNA methylation signature, demonstrated the excellent predictive power of FS for the 1-, 3-, and 5-year survival of HNSCC. Importantly, this predictive model was successfully validated in the GEO dataset. The nomogram also demonstrated excellent accuracy and reliability, as determined by the calibration curves and the decision curve analysis (DCA) plot. Interestingly, the risk score model was found to be correlated with immune cell infiltration and immunotherapy-related biomarkers, suggesting its potential in predicting the immunotherapy response in HNSCC treatment. Moreover, we found that the expression of two risk score model component genes, SETD1B and MTDH, was significantly different between tumor and the adjacent tissues in patients with LSCC, which was also significantly correlated with patient prognosis. Further experimental validation showed that the upregulated expression of MTDH significantly inhibited ferroptosis through regulating GPX4 expression and enhanced the cytotoxicity of ferroptosis inducers in HNSCC cells. In conclusion, we have developed a risk score model by using a ferroptosis-related DNA methylation signature, which can be used as an alternative tool to predict the prognosis of patients with HNSCC. SETD1B and MTDH were identified as the pivotal genes in this model and might play important role in the progression of HNSCC.

The significance of spread through air spaces in the prognostic assessment model of stage I lung adenocarcinoma and the exploration of its invasion mechanism.

PURPOSE: Spread through air spaces (STAS) is a crucial invasive mode of lung cancer and has been shown to be associated with early recurrence and metastasis. We aimed to develop a prognostic risk assessment model for stage I lung adenocarcinoma based on STAS and other pathological features and to explore the potential relationship between CXCL-8, Smad2, Snail, and STAS. METHODS: 312 patients who underwent surgery at Harbin Medical University Cancer Hospital with pathologically diagnosed stage I lung adenocarcinoma were reviewed in the study. STAS and other pathological features were identified by H&E staining, and a prognostic risk assessment model was established. The expression levels of CXCL8, Smad2, and Snail were determined by immunohistochemistry. RESULTS: The nomogram was established based on age, smoking history, STAS, tumor lymphocyte infiltration, tissue subtype, nuclear grade, and tumor size. The C-index for DFS was (training set 0.84 vs validation set 0.77) and for OS was (training set 0.83 vs validation set 0.78). Decision curve analysis showed that the model constructed has a better net benefit than traditional reporting. The prognostic risk score validated the risk stratification value for stage I lung adenocarcinoma. STAS was an important prognostic factor associated with stronger invasiveness and higher expression of CXCL8, Smad2, and Snail. CXCL8 was associated with poorer DFS and OS. CONCLUSIONS: We developed and validated a survival risk assessment model and the prognostic risk score formula for stage I lung adenocarcinoma. Additionally, we found that CXCL8 could be used as a potential biomarker for STAS and poor prognosis, and its mechanism may be related to EMT.

Mining of immunological and prognostic-related biomarker for cervical cancer based on immune cell signatures.

Background: Immunotherapy has changed the therapeutic landscape of cervical cancer (CC), but has durable anti-tumor activity only in a subset of patients. This study aims to comprehensively analyze the tumor immune microenvironment (TIME) of CC and to mine biomarkers related to immunotherapy and prognosis. Methods: The Cancer Genome Atlas (TCGA) data was utilized to identify heterogeneous immune subtypes based on survival-related immune cell signatures (ICSs). ICSs prognostic model was constructed by Cox regression analyses, and immunohistochemistry was conducted to verify the gene with the largest weight coefficient in the model. Meanwhile, the tumor immune infiltration landscape was comprehensively characterized by ESTIMATE, CIBERSORT and MCPcounter algorithms. In addition, we also analyzed the differences in immunotherapy-related biomarkers between high and low-risk groups. IMvigor210 and two gynecologic tumor cohorts were used to validate the reliability and scalability of the Risk score. Results: A total of 291 TCGA-CC samples were divided into two ICSs clusters with significant differences in immune infiltration landscape and prognosis. ICSs prognostic model was constructed based on eight immune-related genes (IRGs), which showed higher overall survival (OS) rate in the low-risk group (P< 0.001). In the total population, time-dependent receiver operating characteristic (ROC) curves displayed area under the curve (AUC) of 0.870, 0.785 and 0.774 at 1-, 3- and 5-years. Immunohistochemical results showed that the expression of the oncogene (FKBP10) was negatively correlated with the degree of differentiation and positively correlated with tumor stage, while the expression of tumor suppressor genes (S1PR4) was the opposite. In addition, the low-risk group had more favorable immune activation phenotype and higher enrichment of immunotherapy-related biomarkers. The Imvigor210 and two gynecologic tumor cohorts validated a better survival advantage and immune efficacy in the low-risk group. Conclusion: This study comprehensively assessed the TIME of CC and constructed an ICSs prognostic model, which provides an effective tool for predicting patient's prognosis and accurate immunotherapy.

Expression and Prognostic Value of Glucose Transporter 3 in Diffuse Large B Cell Lymphoma.

BACKGROUND: Several reports have suggested that glucose transporter 3 (GLUT-3) promotes tumor metastasis. The aim of this study was to examine the relationship between the expression level of GLUT-3 and the prognosis of patients with diffuse large B cell lymphoma (DLBCL). METHODS: The GLUT-3 expression levels in 91 DLBCL patients were evaluated by immunohistochemistry. The relationships between GLUT-3 expression level and clinicopathological characteristics and progression-free survival (PFS) of DLBCL patients were analyzed. The use of validation cohorts confirmed the predictive value of GLUT-3 expression. The correlation between GLUT-3 and immune cell infiltration was investigated using the Cell-type Identification By Estimating Relative Subsets Of RNA Transcripts system and the analysis of the infiltrating score was obtained by single sample Gene Set Enrichment Analysis. RESULTS: Expression of GLUT-3, which is highly expressed in DLBCL patients, was significantly associated with elevated serum LDH level, recurrence and Ki-67 status. Kaplan-Meier analysis showed that high GLUT-3 expression levels in DLBCL were related to poor PFS. Univariate and multivariate analyses results showed that low GLUT-3 expression level was significantly but independently associated with favorable PFS in DLBCL patients. GLUT-3 expression was also correlated with immune cell infiltration and the analysis of the infiltrating score. CONCLUSION: Our results indicate that GLUT-3 may act as a potential independent prognostic factor in DLBCL patients. The difference of the immune microenvironment in DLBCL patients may be predicted by the expression level of GLUT-3.

Focus on the Predictive Value of Subclassification of Extratumoral Structural Abnormalities for Malignant Nonspiculate and Noncalcified Masses on Digital Mammography.

Purpose: To determine the independent risk factors associated with malignant nonspiculate and noncalcified masses (NSNCMs) and evaluate the predictive values of extratumoral structural abnormalities on digital mammography. Methods: A total of 435 patients were included between January and May 2018. Tumor signs included shape, density, and margin, which were evaluated. Extratumoral signs were classified into extratumoral structural abnormalities (parenchymal and trabecular) and halo; subclassification included contraction, distortion, pushing and atrophy sign of parenchyma, parallel, vertical, and reticular trabecula sign, and narrow and wide halo. Univariate and multivariate analysis was performed. The positive predictive value (PPV) of the independent predictor was calculated, and diagnostic performance was evaluated using the receiver operating characteristic curve. Results: Of all cases, 243 (55.8%) were benign and 192 (44.2%) were malignant. Extratumoral contraction sign of parenchyma was the strongest independent predictor of malignancy (odds ratio [OR] 36.2, p < 0.001; PPV = 96.6%), followed by parenchymal distortion sign (OR 10.2, p < 0.001; PPV = 92%), parallel trabecula sign (OR 7.2, p < 0.001; PPV = 85.6%), and indistinct margin of tumor (OR 4.3, p < 0.001; PPV =70.9%), and also parenchymal atrophy sign, wide halo, vertical trabecula, age ≥ 47.5 years, irregular shape, and size ≥ 22.5 mm of tumor (OR range, 1.3-4.0; PPV range, 56.6-83.6%). The diagnostic performance of most of the extratumoral signs was between that of indistinct margin and irregular shape of tumor. Conclusion: The subclassification of extratumoral structural abnormalities has important predictive value for mammographic malignant NSNCM, which should be given more attention.

Clinical and Therapeutic Factors Vary by Prognosis in Patients with Pathological Complete Response After Neoadjuvant Therapy for Breast Cancer.

PURPOSE: To determine clinical predictors of recurrence and metastasis in patients with pathological complete response (pCR) after neoadjuvant chemotherapy (NCT). METHODS: Patients treated with NCT who achieved pCR (n=285) were classified into three groups according to pre-NCT clinical stage (cStage): group I (IIa-IIb), group II (IIIa), and group III (IIIb-IIIc). Survival was analysed using the Kaplan-Meier method. The relationships between clinicopathological factors and recurrence were determined using Cox regression analysis. RESULTS: The median follow-up was 31 months. The 3-year disease-free survival and overall survival rates in groups I, II, and III were 92.7%, 87.8%, and 66.7% (P < 0.01) and 98.6%, 98.3%, and 90.6% (P=0.370), respectively. Lymph node status and tumour size were independent risk factors for recurrence and metastasis after NCT. In the human epidermal growth factor receptor 2-positive subgroup, advanced cStage and lymph node metastasis were associated with recurrence (P < 0.01). In the hormone receptor-positive subgroup, disease-free survival rates differed for cStages I-II compared to cStage III (P=0.049) and clinical node status 0-2 compared to clinical node status 3 (P=0.037). CONCLUSION: Pre-NCT cStage predicted the prognosis of pCR for different breast cancer subtypes. Patients with advanced cStage, lymph node metastasis, and large tumour sizes had a higher risk of recurrence or metastasis.

High expression of CIP2A protein is associated with tumor aggressiveness in stage I-III NSCLC and correlates with poor prognosis.

The aim of this work was to examine the expression of cancerous inhibitor of protein phosphatase 2A (CIP2A) in non-small cell lung cancer (NSCLC) and analyze its correlation with clinical outcomes. CIP2A protein levels were detected by immunohistochemistry (IHC). One hundred and eighty-four of 209 (88.3%) primary stage I-III NSCLC specimens and 4 of 38 (10.5%) adjacent normal lung tissue specimens expressed CIP2A protein. High expression of CIP2A was detected in 38.8% (81/209) of the NSCLC specimens. Patients diagnosed histologically with late-stage NSCLC (p<0.001) and malignant nodes (p=0.001) exhibited high CIP2A expression. Univariate analysis using the log-rank test identified CIP2A expression as a prognostic predictor for overall survival (p=0.005). In multivariate analyses using the Cox regression test, CIP2A expression, T stage, N stage, histological type, and chemotherapy were identified as independent prognostic factors (p=0.007, 0.001, 0.003, <0.001, and <0.001, respectively). Furthermore, Kaplan-Meier survival curves demonstrated that high CIP2A expression indicated poor prognosis in the subgroup of patients with squamous cell carcinoma (p=0.008). Similar results were noted in the subgroup of patients with adenocarcinoma, but the results did not reach statistical significance (p=0.084). We also used univariate analysis and multivariate analysis to assess the prognostic factors for overall survival in the subgroup of patients who received postoperative chemotherapy. CIP2A expression was also an independent prognostic factor in NSCLC patients who received postoperative chemotherapy (p=0.009), along with histological type (p=0.001) and N stage (p=0.034). In conclusion, adding to the accumulating evidence, our research suggested that the CIP2A expression is associated with aggressiveness and correlates with poor prognosis in NSCLC. Our findings also indicated that CIP2A might be a potential therapeutic target against NSCLC.

Increased BTLA and HVEM in gastric cancer are associated with progression and poor prognosis.

PURPOSE: Deregulation of immune checkpoint molecules by tumor cells is related to immune escape. This study was conducted to investigate the relationship between the appearance of B- and T-lymphocyte attenuator (BTLA) and its ligand herpesvirus entry mediator (HVEM) with the prognosis in gastric cancer patients. PATIENTS AND METHODS: A total of 136 patients with curative gastrectomy were included. The expression of BTLA and HVEM was detected by immunohistochemistry, and its correlation with the clinical significance of gastric cancer was further analyzed. RESULTS: The positivity of BTLA and HVEM was detected in 74.3% (101/136) and 89.0% (121/136) of the gastric cancer specimens, respectively. A high expression of BTLA and HVEM was detected, respectively, in 28.7% (39/136) and 44.9% (61/136) of the specimens. Characteristics analysis showed that the high expression of BTLA was significantly associated with lymph node metastasis (P=0.030). Similarly, the high expression of HVEM was also significantly correlated with lymph node metastasis (P=0.007) and depth of invasion (P=0.011). In addition, there was a positive correlation between the expression of BTLA and HVEM in gastric cancer specimens (r=0.245, P=0.004). Univariate analysis revealed that the high expression of BTLA and HVEM was associated with overall survival of patients along with tumor size, Borrmann type, depth of invasion, lymph node metastasis, and histological grade (P<0.05). Multivariate analysis established that the high expression of HVEM (P=0.010), depth of invasion (P=0.001), lymph node metastasis (P<0.001), and histological grade (P=0.027) were independent prognostic factors associated with overall survival in patients with gastric cancer. CONCLUSION: The increased BTLA and HVEM levels correlate with the development and poor prognosis of gastric cancer. HVEM is an important prognostic indicator, and BTLA/HVEM pathway is considered to be a promising candidate for immunotherapy of gastric cancer.

Clinicopathological Features and Prognosis of Papillary Thyroid Microcarcinoma for Surgery and Relationships with the BRAFV600E Mutational Status and Expression of Angiogenic Factors.

OBJECTIVE: To investigate the clinicopathological characteristics of papillary thyroid microcarcinoma (PTMC) for surgery by comparing the difference between PTMC and larger papillary thyroid carcinoma (LPTC). METHODS: We analyzed the differences in the clinicopathological characteristics, prognosis, B-type RAF kinase (BRAF)V600E mutational status and expression of angiogenic factors, including pigment epithelium-derived factor (PEDF), Vascular Endothelial Growth Factor (VEGF), and hypoxia-inducible factor alpha subunit (HIF-1α), between PTMC and LPTC by retrospectively reviewing the records of 251 patients with papillary thyroid carcinoma, 169 with PTMC, and 82 with LPTC (diameter >1 cm). RESULTS: There were no significant differences in the gender, age, multifocality, Hashimoto's thyroiditis, TNM stage, PEDF protein expression, rate of recurrence, or mean follow-up duration between patients with PTMC or LPTC. The prevalence of extrathyroidal invasion (EI), lymph node metastasis (LNM), and BRAF mutation in patients with PTMC was significantly lower than in patients with LPTC. In addition, in PTMC patients with EI and/or LNM and/or positive BRAF (high-risk PTMC patients), the prevalence of extrathyroidal invasion, Hashimoto's disease, lymph node metastasis, tumor TNM stage, PEDF positive protein expression, the rate of recurrent disease, and the mRNA expression of anti-angiogenic factors was almost as high as in patients with larger PTC, but with no significant difference. CONCLUSIONS: Extrathyroid invasion, lymph node metastases, and BRAFV600E mutation were the high risk factors of PTMC. PTMC should be considered for the same treatment strategy as LPTC when any of these factors is found. Particularly, PTMC with BRAFV600E gene mutations needed earlier surgical treatment. In addition, the high cell subtype of PTMC with BRAFV600E gene mutation is recommended for total thyroidectomy in primary surgery to reduce the risk of recurrence.

The overexpression of MDM4: an effective and novel predictor of gastric adenocarcinoma lymph node metastasis.

BACKGROUND: MDM4 is the important negative regulator of the tumor suppressor protein p53, which is overexpressed in various human cancers. This study evaluates the MDM4 expression in patients with gastric adenocarcinoma (GTAC) at the mRNA and protein levels and examines relationships among MDM4 expression, clinicopathological features, and prognosis. RESULTS: The qRT-PCR and the Western blot analysis showed that the MDM4 expression level was high in GTACN+ but not in GTACN-. The high expression level of MDM4 was significantly associated with age (P = 0.047), lymph node metastasis (LNM) (P < 0.001), pathological stage (P < 0.001), differentiation status (P = 0.001), and preoperative serum CA19-9 level (P < 0.001). Moreover, the survival analysis showed that Borrmann type, depth of invasion, LNM, and preoperative serum CA19-9 level were independent prognostic factors. The univariate analysis revealed that MDM4 expression influenced GTAC prognosis. Furthermore, the influence of overall prognosis relies on whether or not the high MDM4 expression level could lead to LNM. MATERIALS AND METHODS: We investigated MDM4 expression in primary GTAC and paired normal gastric tissues (30 pairs) through qRT-PCR and Western blot analyses. We also performed immunohistochemistry analysis on 336 paraffin-embedded GTAC specimens and 33 matched normal specimens. CONCLUSIONS: MDM4 expression may result in LMN of GTAC. High MDM4 expression levels are associated with LMN of GTAC and influence the prognosis of patients with GTAC.

The prognostic and predictive value of the lymphocyte to monocyte ratio in luminal-type breast cancer patients treated with CEF chemotherapy.

Several reports have suggested that peripheral blood-based parameters are associated with host immunity response, which is an essential component of the pathogenesis and progression of cancer. The purpose of the present study was to identify the prognostic significance of various peripheral blood-based biomarkers and to determine the optimal cut-off value suitable for luminal breast cancer patients. We found that lymphocyte-to-monocyte ratio (LMR) was significant prognostic predictors. And the patients with a CEF regimen and LMR ratio ≥ 5.2 gained a good prognosis. This study suggested that the LMR could be regarded as an independent prognostic factor in luminal breast cancer patients. The elevated LMR level also had enhanced 5-fluorouracil sensitivity in luminal breast cancer patients.

The Clinicopathologic and Prognostic Value of Altered Chromosome 17 Centromere Copy Number in HER2 Fish Equivocal Breast Carcinomas.

Chromosome 17 centromere (CEP17) gain is frequently observed in breast cancer by fluorescence in situ hybridization (FISH). To address the biologic characteristics and clinical significance of CEP17 gain in a large population of breast cancer patients, we performed FISH on a series of 770 breast cancer tissues to evaluate the status of human epidermal growth factor receptor 2 (HER2) gene and CEP17 by immunohistochemistry (IHC) and FISH. Among the 770 specimens, 184 cases showed CEP17 gain (23.9%). Histological grade, nodal status, HER2 by IHC, Ki 67 index, and p53 expression were significantly different between CEP17 gain tumors and HER2-positive tumors. In contrast with HER2-negative tumors, CEP17 gain tumors showed higher histological grade, higher HER2 score by IHC, and higher Ki 67 index. The patients with CEP17 gain tumors had an intermediate survival between HER2-negative and HER2-positive patients. By comparison to HER2-negative and HER2-positive patients, survival in luminal B patients with CEP17 gain tumors also fell in between. In conclusion, CEP17 gain tumors show specific differences compared with HER2-negative and HER2-positive tumors in clinical parameters and prognosis.

Tumor-associated macrophages are correlated with tamoxifen resistance in the postmenopausal breast cancer patients.

Tumor-associated macrophages (TAMs) have been correlated with increased angiogenesis and poor prognosis in breast cancer. However, the precise role of TAMs in tamoxifen resistance remains unclear. We used immunohistochemical method to examine the expression of epidermal growth factor receptor (EGFR) and CD163+ macrophages in 100 breast cancer tissues. The clinical and biological features of 100 patients were estrogen receptor (ER)-positive and human epidermal growth factor receptor 2(Her-2)-negative tumors. The tamoxifen resistant tissues (n = 48) were the surgical excision samples from patients who developed recurrence or metastasis at the time of adjuvant tamoxifen treatment. The tamoxifen resistant tissues were contrast to tamoxifen sensitive tissues (n = 52). Positive staining for EGFR and CD163+ macrophages were observed in 21 samples (43.8 %) and in 26 samples (54.2 %) respectively in tamoxifen resistance group, which were higher than that of tamoxifen sensitive group (P = 0.001 and P = 0.000279 respectively). Significant positive correlations were found between the expression of EGFR and CD163+ macrophages (r = 0.567, P < 0.01). CD163+ macrophages were positively correlated with tumor size, lymph node metastasis and obesity. Obesity was also related to tamoxifen resistance (P < 0.05). The patients with higher density of CD163+ macrophages infiltration suffered from shorter time to develop recurrence or metastasis (P < 0.05). TAMs may be associated with tamoxifen resistance. Further studies are needed to investigate the potential mechanism between TAMs and tamoxifen resistance.

Low ING4 protein expression detected by paraffin-section immunohistochemistry is associated with poor prognosis in untreated patients with gastrointestinal stromal tumors.

BACKGROUND: Inhibitor of growth 4 (ING4) has deserved attention as a tumor suppressor gene in many malignant tumors. In our study, we investigated ING4 immunoexpression in gastrointestinal stromal tumors (GISTs) and its prognostic value. METHOD: The expression of ING4 and Ki67 was investigated in 41 samples of various risk gastrointestinal stromal tumors by immunohistochemical technique. The associations of ING4 expression and clinicopathological parameters, and prognosis of the patients, were analyzed by multivariate Cox regression analysis. RESULTS: ING4 expression showed a decreased trend from lower-risk to high-risk gastrointestinal stromal tumors, and an opposite trend for Ki67 expression. In lower-risk tumors, it was found the expression level of ING4 was 78.95 % ± 27.90 % and that of Ki67 was 4.42 % ± 3.75 %. However, in high-risk tumors, the expression level of ING4 was 9.23 % ± 7.66 % and that of Ki67 was 18.50 % ± 9.09 %. There was a strongly negative correlation between the expression levels of ING4 and Ki67. A significant difference was observed in the expression of ING4 between invasion and non-invasion (p < 0.001). The expression of ING4 was markedly correlated with tumor size (p < 0.001), mitotic index (p < 0.001), tumor necrosis (p = 0.021), invasion (p < 0.001), recurrence and metastasis (p = 0.021), and mortality (p < 0.001). CONCLUSION: The low expression level of ING4 protein was correlated with high-risk GISTs. ING4 might be involved in the progression of GISTs and inhibit its invasion. ING4 might be one of the prognostic indicators in GISTs.