Dataset Information

Hypothesis-free deep survival learning applied to the tumour microenvironment in gastric cancer.

ABSTRACT: The biological complexity reflected in histology images requires advanced approaches for unbiased prognostication. Machine learning and particularly deep learning methods are increasingly applied in the field of digital pathology. In this study, we propose new ways to predict risk for cancer-specific death from digital images of immunohistochemically (IHC) stained tissue microarrays (TMAs). Specifically, we evaluated a cohort of 248 gastric cancer patients using convolutional neural networks (CNNs) in an end-to-end weakly supervised scheme independent of subjective pathologist input. To account for the time-to-event characteristic of the outcome data, we developed new survival models to guide the network training. In addition to the standard H&E staining, we investigated the prognostic value of a panel of immune cell markers (CD8, CD20, CD68) and a proliferation marker (Ki67). Our CNN-derived risk scores provided additional prognostic value when compared to the gold standard prognostic tool TNM stage. The CNN-derived risk scores were also shown to be superior when systematically compared to cell density measurements or a CNN score derived from binary 5-year survival classification, which ignores time-to-event. To better understand the underlying biological mechanisms, we qualitatively investigated risk heat maps for each marker which visualised the network output. We identified patterns of biological interest that were related to low risk of cancer-specific death such as the presence of B-cell predominated clusters and Ki67 positive sub-regions and showed that the corresponding risk scores had prognostic value in multivariate Cox regression analyses (Ki67&CD20 risks: hazard ratio (HR) = 1.47, 95% confidence interval (CI) = 1.15-1.89, p = 0.002; CD20&CD68 risks: HR = 1.33, 95% CI = 1.07-1.67, p = 0.009). Our study demonstrates the potential additional value that deep learning in combination with a panel of IHC markers can bring to the field of precision oncology.

SUBMITTER: Meier A

PROVIDER: S-EPMC7578283 | biostudies-literature | 2020 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Hypothesis-free deep survival learning applied to the tumour microenvironment in gastric cancer.

Meier Armin A Nekolla Katharina K Hewitt Lindsay C LC Earle Sophie S Yoshikawa Takaki T Oshima Takashi T Miyagi Yohei Y Huss Ralf R Schmidt Günter G Grabsch Heike I HI

The journal of pathology. Clinical research 20200627 4

The biological complexity reflected in histology images requires advanced approaches for unbiased prognostication. Machine learning and particularly deep learning methods are increasingly applied in the field of digital pathology. In this study, we propose new ways to predict risk for cancer-specific death from digital images of immunohistochemically (IHC) stained tissue microarrays (TMAs). Specifically, we evaluated a cohort of 248 gastric cancer patients using convolutional neural networks (CN ...[more]

PMID: 32592447

Similar Datasets

Project description:BackgroundPrognostic prediction for surgical treatment of gastric cancer remains valuable in clinical practice. This study aimed to develop survival models for postoperative gastric cancer patients.MethodsEleven thousand seventy-five patients from the Surveillance, Epidemiology, and End Results (SEER) database were included, and 122 patients from the Chinese database were used for external validation. The training cohort was created to create three separate models, including Cox regression, RSF, and DeepSurv, using data from the SEER database split into training and test cohorts with a 7:3 ratio. Test cohort was used to evaluate model performance using c-index, Brier scores, calibration, and the area under the curve (AUC). The new risk stratification based on the best model will be compared with the AJCC stage on the test and Chinese cohorts using decision curve analysis (DCA), the net reclassification index (NRI), and integrated discrimination improvement (IDI).ResultsIt was discovered that the DeepSurv model predicted postoperative gastric cancer patients' overall survival (OS) with a c-index of 0.787; the area under the curve reached 0.781, 0.798, 0.868 at 1-, 3- and 5- years, respectively; the Brier score was below 0.25 at different time points; showing an advantage over the Cox and RSF models. The results are also validated in the China cohort. The calibration plots demonstrated good agreement between the DeepSurv model's forecast and actual results. The NRI values (test cohort: 0.399, 0.288, 0.267 for 1-, 3- and 5-year OS prediction; China cohort:0.399, 0.288 for 1- and 3-year OS prediction) and IDI (test cohort: 0.188, 0.169, 0.157 for 1-, 3- and 5-year OS prediction; China cohort: 0.189, 0.169 for 1- and 3-year OS prediction) indicated that the risk score stratification performed significantly better than the AJCC staging alone (P < 0.05). DCA showed that the risk score stratification was clinically useful and had better discriminative ability than the AJCC staging. Finally, an interactive native web-based prediction tool was constructed for the survival prediction of patients with postoperative gastric cancer.ConclusionIn this study, a high-performance prediction model for the postoperative prognosis of gastric cancer was developed using DeepSurv, which offers essential benefits for risk stratification and prognosis prediction for each patient.

Project description:BackgroundGastric cancer is one of the leading causes of cancer death in the world. Improving gastric cancer survival prediction can enhance patient prognostication and treatment planning.MethodsIn this study, we performed gastric cancer survival prediction using machine learning and multi-modal data of 1061 patients, including 743 for model learning and 318 independent patients for evaluation. A Cox proportional-hazard model was trained to integrate clinical variables and CT imaging features (extracted by radiomics and deep learning) for overall and progression-free survival prediction. We further analyzed the prediction effects of clinical, radiomics, and deep learning features. Concordance index (c-index) was used as the model performance metric, and the predictive effects of multi-modal features were measured by hazard ratios (HRs) at pre- and post-operative settings.ResultsAmong 318 patients in the independent testing group, the hazard predicted by Cox from multi-modal features is associated with their survival. The highest c-index was 0.783 (95% CI, 0.782-0.783) and 0.770 (95% CI, 0.769-0.771) for overall and progression-free survival prediction, respectively. The post-operative variables are significantly (p<0.001) more predictive than the pre-operative variables. Pathological tumor stage (HR=1.336 [overall survival]/1.768 [progression-free survival], p<0.005), pathological lymph node stage (HR=1.665/1.433, p<0.005), carcinoembryonic antigen (CEA) (HR=1.632/1.522, p=0.02), chemotherapy treatment (HR=0.254/0.287, p<0.005), radiomics signature [HR=1.540/1.310, p<0.005], and deep learning signature [HR=1.950/1.420, p<0.005]) are significant survival predictors.ConclusionOur study showed that CT radiomics and deep learning imaging features are significant pre-operative predictors, providing additional prognostic information to the pathological staging markers. Lower CEA levels and chemotherapy treatments also increase survival chances. These findings can enhance gastric cancer patient prognostication and inform treatment planning.

Project description:BackgroundPersonalized therapy planning remains a significant challenge in advanced colorectal cancer care, despite extensive research on prognostic and predictive markers. A strong correlation of sarcopenia or overall body composition and survival has been described. Here, we explore whether automated assessment of body composition and liver metastases from standard of care CT images can add to clinical parameters in personalized survival risk prognostication.MethodsWe retrospectively analysed clinical imaging data from 85 patients (50.6% female, mean age 58.9 SD 12.2 years) with colorectal cancer and synchronous liver metastases. Pretrained deep learning models were used to assess body composition and liver metastasis geometry from abdominal CT images before the initiation of systemic treatment. Abdominal muscle-to-bone ratio (MBR) was calculated by dividing abdominal muscle volume by abdominal bone volume. MBR was compared with body mass index (BMI), abdominal muscle volume, and abdominal muscle volume divided by height squared. Differences in overall survival based on body composition and liver metastasis parameters were compared using Kaplan-Meier survival curves. Results were correlated with clinical and biomarker data to develop a machine learning model for survival risk prognostication.ResultsThe MBR, unlike abdominal muscle volume or BMI, was significantly associated with overall survival (HR 0.39, 95% CI: 0.19-0.80, P = 0.009). The MBR (P = 0.022), liver metastasis surface area (P = 0.01) and primary tumour sidedness (P = 0.007) were independently associated with overall survival in multivariate analysis. Body composition parameters did not correlate with KRAS mutational status or primary tumour sidedness. A prediction model based on MBR, liver metastasis surface area and primary tumour sidedness achieved a concordance index of 0.69.ConclusionsAutomated segmentation enables to extract prognostic parameters from routine imaging data for personalized survival modelling in advanced colorectal cancer patients.

Project description:BackgroundTumour budding has been reported to reflect invasiveness, metastasis and unfavourable prognosis in colorectal cancer. The aim of the study was to examine the relationship between tumour budding and clinicopathological characteristics, tumour microenvironment and survival in patients with primary operable colorectal cancer.MethodsA total of 303 patients from a prospective data set of patients with primary operable colorectal cancer were included in the study. The presence of budding was determined through assessment of all tumour-containing H&E slides and the number of tumour buds was counted using a 10 high-powered field method. Routine pathologic sections were used to assess: tumour necrosis, the tumour inflammatory cell infiltrate using Klintrup-Makinen (KM) grade and tumour stroma percentage (TSP) combined as the Glasgow Microenvironment Score (GMS).ResultsHigh-grade tumour budding was present in 39% of all tumours and in 28% of node-negative tumours respectively. High-grade budding was significantly associated with T stage (P<0.001), N stage (P<0.001), TNM stage (P<0.001), serosal involvement (P<0.001), venous invasion (P<0.005), KM grade (P=0.022), high tumour stroma (P<0.001) and GMS (P<0.001). Tumour budding was associated with reduced cancer-specific survival (CSS) (HR=4.03; 95% confidence interval (CI), 2.50-6.52; P<0.001), independent of age (HR=1.47; 95% CI, 1.13-1.90; P=0.004), TNM stage (HR=1.52; 95% CI, 1.02-2.25; P=0.040), venous invasion (HR=1.73; 95% CI, 1.13-2.64; P=0.012) and GMS (HR=1.54; 95% CI, 1.15-2.07; P=0.004).ConclusionsThe presence of tumour budding was associated with elements of the tumour microenvironment and was an independent adverse prognostic factor in patients with primary operable colorectal cancer. Specifically high tumour budding stratifies effectively the prognostic value of tumour stage, venous invasion and GMS. Taken together, tumour budding should be assessed routinely in patients with primary operable colorectal cancer.

Project description:BackgroundInflammation-related predisposition to cancer plays an essential role in cancer progression and is associated with poor prognosis. A hypoxic microenvironment and neutrophil infiltration are commonly present in solid tumours, including gastric cancer (GC). Neutrophil extracellular traps (NETs) have also been demonstrated in the tumour immune microenvironment (TIME), but how NETs affect GC progression remains unknown. Here, we investigated the role of NET formation in the TIME and further explored the underlying mechanism of NETs in GC tumour growth.MethodsHypoxia-induced factor-1α (HIF-1α), citrulline histone 3 (citH3) and CD66b expression in tumour and adjacent nontumor tissue samples was evaluated by western blotting, immunofluorescence and immunohistochemical staining. The expression of neutrophil-attracting chemokines in GC cells and their hypoxic-CM was measured by qRT‒PCR and ELISA. Neutrophil migration under hypoxic conditions was evaluated by a Transwell assay. Pathway activation in neutrophils in a hypoxic microenvironment were analysed by western blotting. NET formation was measured in vitro by immunofluorescence staining. The protumour effect of NETs on GC cells was identified by Transwell, wound healing and cell proliferation assays. In vivo, an lipopolysaccharide (LPS)-induced NET model and subcutaneous tumour model were established in BALB/c nude mice to explore the mechanism of NETs in tumour growth.ResultsGC generates a hypoxic microenvironment that recruits neutrophils and induces NET formation. High mobility group box 1 (HMGB1) was translocated to the cytoplasm from the nucleus of GC cells in the hypoxic microenvironment and mediated the formation of NETs via the toll-like receptor 4 (TLR4)/p38 MAPK signalling pathway in neutrophils. HMGB1/TLR4/p38 MAPK pathway inhibition abrogated hypoxia-induced neutrophil activation and NET formation. NETs directly induced GC cell invasion and migration but not proliferation and accelerated the augmentation of GC growth by increasing angiogenesis. This rapid tumour growth was abolished by treatment with the NET inhibitor deoxyribonuclease I (DNase I) or a p38 MAPK signalling pathway inhibitor.ConclusionsHypoxia triggers an inflammatory response and NET formation in the GC TIME to augment tumour growth. Targeting NETs with DNase I or HMGB1/TLR4/p38 MAPK pathway inhibitors is a potential therapeutic strategy to inhibit GC progression. Video Abstract.

Dataset Information

Hypothesis-free deep survival learning applied to the tumour microenvironment in gastric cancer.

Publications

Hypothesis-free deep survival learning applied to the tumour microenvironment in gastric cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets