Development of a semi-automated method for tumour budding assessment in colorectal cancer and comparison with manual methods

Development of a digital algorithm for assessing tumor-stroma ratio, tumor budding and tumor infiltrating lymphocytes in vulvar squamous cell carcinomas

Tumor-stroma ratio (TSR), tumor budding (TB), and tumor-infiltrating lymphocytes (TILs) are prognostic markers in some cancers but with unknown significance in vulvar squamous cell carcinoma (VSCC). This pilot study primarily aimed to develop a digital method for evaluating TSR, TB and TILs in VSCC and secondarily to investigate variation in these factors by p16 status. An independent training set stained with CD3/cytokeratin and CD8/cytokeratin was used to develop a deep learning-based Application Protocol Package (APP) segmenting tissue into background, epithelium, or stroma. TSR was defined as percentage of tumor epithelium relative to total tumor area, and tumor buds were defined as clusters of 1-4 tumor cells. A second APP quantified CD3+ and CD8+ lymphocytes in the intraepithelial and stromal compartments, respectively. The digital algorithms were applied to the study cohort of 41 VSCC cases, achieving satisfactory performance without manual corrections. TSR ranged between 33 and 91% with median of 64%, and median number of buds was 4 (range: 0-48) buds/mm2. Median density and range of CD3+ lymphocytes were 222 (13-2320) cells/mm2 in the intraepithelial and 1978 (397-6683) cells/mm2 in the stromal compartment, respectively. CD8+ lymphocyte counts were lower. There was a tendency towards lower TSR and higher number of buds in p16-negative compared with p16-positive VSCC. Finally, automated measures were compared with manual evaluations showing high concordance. The developed automated method provided precise and objective measurements of TSR, TB and TILs. The algorithms should be validated in a larger cohort and correlated with clinicopathological characteristics and prognosis to determine their clinical relevance.

Tumour Budding Is a Useful Predictor to Identify High-Risk Stage II Colon Cancer Patients After Curative Surgery

Aim. Although it is now accepted in the literature that tumour budding (TB) is a useful survival indicator in colon cancer (CC), there are still uncertainties about daily use. Here we methodologically examined the role of TB on survival in CC. Methods. In our study, we examined colon cancer patients who had surgery up to 15 years before presentation. TB was calculated separately using different comprehensive methodological methods. Results. We first investigated an optimal evaluation method. Relationship with prognostic factors was better (Venous invasion [p = .001], advanced pT [p = .003], perineural invasion [p = .040], MSS [p = .016], advanced size [p = .001], tumour obstruction [p = .005], margin involvement [p = .043], and nodal involvement [p = .028]) in Method-1. Similarly, with the same method, the success of the cut-off value, the correlation of TB data (r = .724), and the repeatability of the method (Κappa = .53-.75) were quite good (ROC = .816 [.707-.925]). Then, survival analysis was performed using the best three methods, including this method. In univariate analysis using Method-1, survival analyses were worse in high TB patients (RFS: 81%, p < .001; OS: 84%, p < .001). Multivariate analyses using the same method confirmed that high TB for RFS and OS was an independent poor prognostic parameter for survival (p = .002, Hazard ratio [HR]: 1.42 [1.13-1.80]) and OS (p = .014, HR: 1.38 [1.07-1.79]). Conclusions. With our study, we showed that tumour budding calculated by the standard method is a very valuable prognostic parameter in stage II CC and can contribute to the detection of patients with poor prognosis in stage II CC.

Artificial intelligence (AI) for tumor microenvironment (TME) and tumor budding (TB) identification in colorectal cancer (CRC) patients: A systematic review

Evaluation of the parameters such as tumor microenvironment (TME) and tumor budding (TB) is one of the most important steps in colorectal cancer (CRC) diagnosis and cancer development prognosis. In recent years, artificial intelligence (AI) has been successfully used to solve such problems. In this paper, we summarize the latest data on the use of artificial intelligence to predict tumor microenvironment and tumor budding in histological scans of patients with colorectal cancer. We performed a systematic literature search using 2 databases (Medline and Scopus) with the following search terms: ("tumor microenvironment" OR "tumor budding") AND ("colorectal cancer" OR CRC) AND ("artificial intelligence" OR "machine learning " OR "deep learning"). During the analysis, we gathered from the articles performance scores such as sensitivity, specificity, and accuracy of identifying TME and TB using artificial intelligence. The systematic review showed that machine learning and deep learning successfully cope with the prediction of these parameters. The highest accuracy values in TB and TME prediction were 97.7% and 97.3%, respectively. This review led us to the conclusion that AI platforms can already be used as diagnostic aids, which will greatly facilitate the work of pathologists in detection and estimation of TB and TME as instruments and second-opinion services. A key limitation in writing this systematic review was the heterogeneous use of performance metrics for machine learning models by different authors, as well as relatively small datasets used in some studies.

Immunohistochemical analysis of tumor budding in stage II colon cancer: exploring zero budding as a prognostic marker

Tumor budding, a biomarker traditionally evaluated using hematoxylin and eosin (H&E) staining, has gained recognition as a prognostic biomarker for stage II colon cancer. Nevertheless, while H&E staining offers valuable insights, its limitations prompt the utilization of pan-cytokeratin immunohistochemistry (IHC). Consequently, this study seeks to evaluate the prognostic significance of tumor budding using IHC in a contemporary cohort of stage II colon cancer patients, aiming to deepen our understanding of this critical facet in cancer prognosis. We conducted a retrospective, population-based cohort study including 493 patients with stage II colon cancer and evaluated tumor budding using IHC, following the H&E-based guidelines proposed by the International Tumor Budding Consensus Conference Group. Correlation between H&E-based and IHC-based tumor budding was assessed using a four-tiered scoring system that included a zero budding (Bd0) category. Survival analyses explored the prognostic significance of tumor budding assessed by IHC and H&E. As expected, IHC-based tumor budding evaluation yielded significantly higher bud counts compared to H&E (p < 0.01). Interestingly, 21 patients were identified with no tumor budding using IHC. This was associated with significantly improved recurrence-free survival (HR = 5.19, p = 0.02) and overall survival (HR = 4.47, p = 0.04) in a multivariate analysis when compared to tumors with budding. The Bd0 category demonstrated a 100% predictive value for the absence of recurrence. In conclusion, IHC-based tumor budding evaluation in stage II colon cancer provides additional prognostic information. The absence of tumor budding is associated with a favorable prognosis and may serve as a potential marker for identifying patients with no risk of recurrence.

AI-Powered Segmentation of Invasive Carcinoma Regions in Breast Cancer Immunohistochemical Whole-Slide Images

AIMS

The automation of quantitative evaluation for breast immunohistochemistry (IHC) plays a crucial role in reducing the workload of pathologists and enhancing the objectivity of diagnoses. However, current methods face challenges in achieving fully automated immunohistochemistry quantification due to the complexity of segmenting the tumor area into distinct ductal carcinoma in situ (DCIS) and invasive carcinoma (IC) regions. Moreover, the quantitative analysis of immunohistochemistry requires a specific focus on invasive carcinoma regions.

METHODS AND RESULTS

In this study, we propose an innovative approach to automatically identify invasive carcinoma regions in breast cancer immunohistochemistry whole-slide images (WSIs). Our method leverages a neural network that combines multi-scale morphological features with boundary features, enabling precise segmentation of invasive carcinoma regions without the need for additional H&E and P63 staining slides. In addition, we introduced an advanced semi-supervised learning algorithm, allowing efficient training of the model using unlabeled data. To evaluate the effectiveness of our approach, we constructed a dataset consisting of 618 IHC-stained WSIs from 170 cases, including four types of staining (ER, PR, HER2, and Ki-67). Notably, the model demonstrated an impressive intersection over union (IoU) score exceeding 80% on the test set. Furthermore, to ascertain the practical utility of our model in IHC quantitative evaluation, we constructed a fully automated Ki-67 scoring system based on the model's predictions. Comparative experiments convincingly demonstrated that our system exhibited high consistency with the scores given by experienced pathologists.

CONCLUSIONS

Our developed model excels in accurately distinguishing between DCIS and invasive carcinoma regions in breast cancer immunohistochemistry WSIs. This method paves the way for a clinically available, fully automated immunohistochemistry quantitative scoring system.

Molecular Insight into Gastric Cancer Invasion-Current Status and Future Directions

Gastric cancer (GC) is one of the most common malignancies worldwide. There has been no efficient therapy for stage IV GC patients due to this disease's heterogeneity and dissemination ability. Despite the rapid advancement of molecular targeted therapies, such as HER2 and immune checkpoint inhibitors, survival of GC patients is still unsatisfactory because the understanding of the mechanism of GC progression is still incomplete. Invasion is the most important feature of GC metastasis, which causes poor mortality in patients. Recently, genomic research has critically deepened our knowledge of which gene products are dysregulated in invasive GC. Furthermore, the study of the interaction of GC cells with the tumor microenvironment has emerged as a principal subject in driving invasion and metastasis. These results are expected to provide a profound knowledge of how biological molecules are implicated in GC development. This review summarizes the advances in our current understanding of the molecular mechanism of GC invasion. We also highlight the future directions of the invasion therapeutics of GC. Compared to conventional therapy using protease or molecular inhibitors alone, multi-therapy targeting invasion plasticity may seem to be an assuring direction for the progression of novel strategies.

Fully Automated Tumor Bud Assessment in Hematoxylin and Eosin-Stained Whole Slide Images of Colorectal Cancer

Tumor budding (TB), the presence of single cells or small clusters of up to 4 tumor cells at the invasive front of colorectal cancer (CRC), is a proven risk factor for adverse outcomes. International definitions are necessary to reduce interobserver variability. According to the current international guidelines, hotspots at the invasive front should be counted in hematoxylin and eosin (H&E)-stained slides. This is time-consuming and prone to interobserver variability; therefore, there is a need for computer-aided diagnosis solutions. In this study, we report an artificial intelligence-based method for detecting TB in H&E-stained whole slide images. We propose a fully automated pipeline to identify the tumor border, detect tumor buds, characterize them based on the number of tumor cells, and produce a TB density map to identify the TB hotspot. The method outputs the TB count in the hotspot as a computational biomarker. We show that the proposed automated TB detection workflow performs on par with a panel of 5 pathologists at detecting tumor buds and that the hotspot-based TB count is an independent prognosticator in both the univariate and the multivariate analysis, validated on a cohort of n = 981 patients with CRC. Computer-aided detection of tumor buds based on deep learning can perform on par with expert pathologists for the detection and quantification of tumor buds in H&E-stained CRC histopathology slides, strongly facilitating the introduction of budding as an independent prognosticator in clinical routine and clinical trials.

Semi-Supervised Learning to Automate Tumor Bud Detection in Cytokeratin-Stained Whole-Slide Images of Colorectal Cancer

Tumor budding is a histopathological biomarker associated with metastases and adverse survival outcomes in colorectal carcinoma (CRC) patients. It is characterized by the presence of single tumor cells or small clusters of cells within the tumor or at the tumor-invasion front. In order to obtain a tumor budding score for a patient, the region with the highest tumor bud density must first be visually identified by a pathologist, after which buds will be counted in the chosen hotspot field. The automation of this process will expectedly increase efficiency and reproducibility. Here, we present a deep learning convolutional neural network model that automates the above procedure. For model training, we used a semi-supervised learning method, to maximize the detection performance despite the limited amount of labeled training data. The model was tested on an independent dataset in which human- and machine-selected hotspots were mapped in relation to each other and manual and machine detected tumor bud numbers in the manually selected fields were compared. We report the results of the proposed method in comparison with visual assessment by pathologists. We show that the automated tumor bud count achieves a prognostic value comparable with visual estimation, while based on an objective and reproducible quantification. We also explore novel metrics to quantify buds such as density and dispersion and report their prognostic value. We have made the model available for research use on the grand-challenge platform.

The relationship between primary colorectal cancer histology and the histopathological growth patterns of corresponding liver metastases

Background

The histopathological growth patterns (HGPs) are a prognostic and predictive biomarker in colorectal cancer liver metastasis (CRLM). This study evaluates the relationship between the HGP and primary colorectal cancer (CRC) histopathology.

Methods

A total of 183 treatment-naive patients with resected CRC and CRLM were included. Thirteen CRC histopathology markers were determined and compared between the desmoplastic and non-desmoplastic HGP; tumour sidedness, pT&pN stage, tumour grade, tumour deposits, perineural- (lympho-)vascular- and extramural venous invasion, peritumoural budding, stroma type, CRC growth pattern, Crohn’s-like lymphoid reaction, and tumour-infiltrating lymphocyte (TIL) density. Logistic regression analysis was performed using both CRC and CRLM characteristics.

Results

Unfavourable CRC histopathology was more frequent in non-desmoplastic CRLM for all markers evaluated, and significantly so for a lower TIL density, absent Crohn’s-like lymphoid reaction, and a “non-mature” stroma (all p < 0.03). The cumulative prevalence of unfavourable CRC histopathology was significantly higher in patients with non-desmoplastic compared to desmoplastic CRLM, with a median (IQR) of 4 (3–6) vs 2 (1–3.5) unfavourable characteristics observed, respectively (p < 0.001). Multivariable regression with 9 CRC histopathology markers and 2 CRLM characteristics achieved good discriminatory performance (AUC = 0.83).

Conclusions

The results of this study associates primary CRC histopathology with the HGP of corresponding liver metastases.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-022-09994-3.

Tumour invasion and dissemination

Activating invasion and metastasis are one of the primary hallmarks of cancer, the latter representing the leading cause of death in cancer patients. Whilst many advances in this area have been made in recent years, the process of cancer dissemination and the underlying mechanisms governing invasion are still poorly understood. Cancer cells exhibit multiple invasion strategies, including switching between modes of invasion and plasticity in response to therapies, surgical interventions and environmental stimuli. The ability of cancer cells to switch migratory modes and their inherent plasticity highlights the critical challenge preventing the successful design of cancer and anti-metastatic therapies. This mini-review presents current knowledge on the critical models of tumour invasion and dissemination. We also discuss the current issues surrounding current treatments and arising therapeutic opportunities. We propose that the establishment of novel approaches to study the key biological mechanisms underlying the metastatic cascade is critical in finding novel targets that could ultimately lead to complete inhibition of cancer cell invasion and dissemination.

Molecular mechanisms of tumour budding and its association with microenvironment in colorectal cancer

Colorectal cancer (CRC) is the third most common cancer worldwide. Poor survival of CRC associated with the development of tumour metastasis led to the investigation of the potential biomarkers to predict outcomes in CRC patients. Tumour budding (TB) is a well-known independent prognostic marker for poor survival and disease metastasis. Therefore, it has been suggested that TB status is included in routine clinicopathological factors for risk assessment in CRC. In contrast with a vast majority of studies regarding the prognostic power of TB, there is no clear evidence pertaining to the underlying molecular mechanism driving this phenotype, or an understanding of TB relationship with the tumour microenvironment (TME). The aim of the present study is to present a comprehensive review of TB and tumour cell signalling pathways together with the cross-talk of immune cells that could drive TB formation in CRC.