High reproducibility is attainable in assessing histoprognostic parameters of pT1 colorectal cancer using routine histopathology slides and immunohistochemistry analyses

Prediction of Lymph Node Metastasis in T1 Colorectal Cancer Using Artificial Intelligence with Hematoxylin and Eosin-Stained Whole-Slide-Images of Endoscopic and Surgical Resection Specimens

According to the current guidelines, additional surgery is performed for endoscopically resected specimens of early colorectal cancer (CRC) with a high risk of lymph node metastasis (LNM). However, the rate of LNM is 2.1-25.0% in cases treated endoscopically followed by surgery, indicating a high rate of unnecessary surgeries. Therefore, this study aimed to develop an artificial intelligence (AI) model using H&E-stained whole slide images (WSIs) without handcrafted features employing surgically and endoscopically resected specimens to predict LNM in T1 CRC. To validate with an independent cohort, we developed a model with four versions comprising various combinations of training and test sets using H&E-stained WSIs from endoscopically (400 patients) and surgically resected specimens (881 patients): Version 1, Train and Test: surgical specimens; Version 2, Train and Test: endoscopic and surgically resected specimens; Version 3, Train: endoscopic and surgical specimens and Test: surgical specimens; Version 4, Train: endoscopic and surgical specimens and Test: endoscopic specimens. The area under the curve (AUC) of the receiver operating characteristic curve was used to determine the accuracy of the AI model for predicting LNM with a 5-fold cross-validation in the training set. Our AI model with H&E-stained WSIs and without annotations showed good performance power with the validation of an independent cohort in a single center. The AUC of our model was 0.758-0.830 in the training set and 0.781-0.824 in the test set, higher than that of previous AI studies with only WSI. Moreover, the AI model with Version 4, which showed the highest sensitivity (92.9%), reduced unnecessary additional surgery by 14.2% more than using the current guidelines (68.3% vs. 82.5%). This revealed the feasibility of using an AI model with only H&E-stained WSIs to predict LNM in T1 CRC.

Additional staining for lymphovascular invasion is associated with increased estimation of lymph node metastasis in patients with T1 colorectal cancer: Systematic review and meta-analysis

OBJECTIVES

Lymphovascular invasion (LVI) is a critical risk factor for lymph node metastasis (LNM), which requires additional surgery after endoscopic resection of T1 colorectal cancer (CRC). However, the impact of additional staining on estimating LNM is unclear. This systematic review aimed to evaluate the impact of additional staining on determining LNM in T1 CRC.

METHODS

We searched five electronic databases. Outcomes were diagnostic odds ratio (DOR), assessed using hierarchical summary receiver operating characteristic curves, and interobserver agreement among pathologists for positive LVI, assessed using Kappa coefficients (κ). We performed a subgroup analysis of studies that simultaneously included a multivariable analysis for other risk factors (deep submucosal invasion, poor differentiation, and tumor budding).

RESULTS

Among the 64 studies (18,097 patients) identified, hematoxylin-eosin (HE) and additional staining for LVI had pooled sensitivities of 0.45 (95% confidence interval [CI] 0.32-0.58) and 0.68 (95% CI 0.44-0.86), specificities of 0.88 (95% CI 0.78-0.94) and 0.76 (95% CI 0.62-0.86), and DORs of 6.26 (95% CI 3.73-10.53) and 6.47 (95% CI 3.40-12.32) for determining LNM, respectively. In multivariable analysis, the DOR of additional staining for LNM (DOR 5.95; 95% CI 2.87-12.33) was higher than that of HE staining (DOR 1.89; 95% CI 1.13-3.16) (P = 0.01). Pooled κ values were 0.37 (95% CI 0.22-0.52) and 0.62 (95% CI 0.04-0.99) for HE and additional staining for LVI, respectively.

CONCLUSION

Additional staining for LVI may increase the DOR for LNM and interobserver agreement for positive LVI among pathologists.

Risk assessment in pT1 colorectal cancer

Colorectal cancer (CRC) is a common malignancy worldwide and tumour stage is closely related to clinical outcome. A small but significant proportion of submucosal-invasive (ie, pT1) CRC are associated with regional lymph node metastases (LNM) and a worse prognosis. The likelihood of LNM in pT1 CRC needs to be balanced against the operative risk and costs of surgical resection when determining the best patient management. A wide range of histopathological and clinical factors may affect LNM risk in this setting. This script provides a comprehensive overview of the tumour and patient-associated features that have been linked to LNM risk in pT1 CRC. Some of the features are well established within the literature and are included in published guidelines, while others are novel and emerging in nature. Odds ratios for LNM that are associated with key predictive features are provided where appropriate, and published models developed as an aid to the calculation of LNM risk are discussed.

Use of artificial intelligence in the management of T1 colorectal cancer: a new tool in the arsenal or is deep learning out of its depth?

The field of artificial intelligence is rapidly evolving, and there has been an interest in its use to predict the risk of lymph node metastasis in T1 colorectal cancer. Accurately predicting lymph node invasion may result in fewer patients undergoing unnecessary surgeries; conversely, inadequate assessments will result in suboptimal oncological outcomes. This narrative review aims to summarize the current literature on deep learning for predicting the probability of lymph node metastasis in T1 colorectal cancer, highlighting areas of potential application and barriers that may limit its generalizability and clinical utility.

Whole slide image-based prediction of lymph node metastasis in T1 colorectal cancer using unsupervised artificial intelligence

OBJECTIVES

Lymph node metastasis (LNM) prediction for T1 colorectal cancer (CRC) is critical for determining the need for surgery after endoscopic resection because LNM occurs in 10%. We aimed to develop a novel artificial intelligence (AI) system using whole slide images (WSIs) to predict LNM.

METHODS

We conducted a retrospective single center study. To train and test the AI model, we included LNM status-confirmed T1 and T2 CRC between April 2001 and October 2021. These lesions were divided into two cohorts: training (T1 and T2) and testing (T1). WSIs were cropped into small patches and clustered by unsupervised K-means. The percentage of patches belonging to each cluster was calculated from each WSI. Each cluster's percentage, sex, and tumor location were extracted and learned using the random forest algorithm. We calculated the areas under the receiver operating characteristic curves (AUCs) to identify the LNM and the rate of over-surgery of the AI model and the guidelines.

RESULTS

The training cohort contained 217 T1 and 268 T2 CRCs, while 100 T1 cases (LNM-positivity 15%) were the test cohort. The AUC of the AI system for the test cohort was 0.74 (95% confidence interval [CI] 0.58-0.86), and 0.52 (95% CI 0.50-0.55) using the guidelines criteria (P = 0.0028). This AI model could reduce the 21% of over-surgery compared to the guidelines.

CONCLUSION

We developed a pathologist-independent predictive model for LNM in T1 CRC using WSI for determination of the need for surgery after endoscopic resection.

TRIAL REGISTRATION

UMIN Clinical Trials Registry (UMIN000046992, https://center6.umin.ac.jp/cgi-open-bin/ctr/ctr_view.cgi?recptno=R000053590).

Poorly differentiated cluster grade-a vital predictor for lymph node metastasis and oncological outcomes in patients with T1 colorectal cancer: a retrospective study

Background

T1 colorectal cancers have a low lymph node metastasis rate and good prognosis. Thus, endoscopic resection is an attractive choice. This study aimed to describe the value of poorly differentiated cluster grade in identifying endoscopically curable T1 colorectal cancers.

Methods

We included 183 T1 colorectal cancer patients who underwent curative resection. Univariate and multivariate logistic regressions were used to identify lymph node metastasis predictors. The Akaike information criterion was used to determine whether poorly differentiated cluster grade was the best predictor. Backward regression was used to screen the variables. Survival analyses were conducted to determine the prognostic predictive power of poorly differentiated cluster grade. Correlations among predictors and concordance between our pathologists were also investigated.

Results

Poorly differentiated cluster grade was an independent predictor for lymph node metastasis (adjusted odds ratio [OR]_{G 3} = 0.001; 95% confidence interval [95% CI]_{G 3} =  < 0.001, 0.139) in T1 colorectal cancer patients; moreover, it had the best predictive value (AIC = 61.626) among all indicators. It was also screened for inclusion in the predictive model. Accordingly, a high poorly differentiated cluster grade independently indicated shorter overall survival (hazard ratio [HR]_{G 2} = 4.315; 95% CI_{G 2} = 1.506, 12.568; HR_{G 3} = 5.049; 95% CI_{G 3} = 1.326, 19.222) and disease-free survival (HR_{G 3} = 6.621; 95% CI_{G 3} = 1.472, 29.786).

Conclusions

Poorly differentiated cluster grade is a vital reference to manage T1 colorectal cancer. It could serve as an indicator to screen endoscopically curable T1 colorectal cancers.

Utility of artificial intelligence with deep learning of hematoxylin and eosin-stained whole slide images to predict lymph node metastasis in T1 colorectal cancer using endoscopically resected specimens; prediction of lymph node metastasis in T1 colorectal cancer

BACKGROUND

When endoscopically resected specimens of early colorectal cancer (CRC) show high-risk features, surgery should be performed based on current guidelines because of the high-risk of lymph node metastasis (LNM). The aim of this study was to determine the utility of an artificial intelligence (AI) with deep learning (DL) of hematoxylin and eosin (H&E)-stained endoscopic resection specimens without manual-pixel-level annotation for predicting LNM in T1 CRC. In addition, we assessed AI performance for patients with only submucosal (SM) invasion depth of 1000 to 2000 μm known to be difficult to predict LNM in clinical practice.

METHODS

H&E-stained whole slide images (WSIs) were scanned for endoscopic resection specimens of 400 patients who underwent endoscopic treatment for newly diagnosed T1 CRC with additional surgery. The area under the curve (AUC) of the receiver operating characteristic curve was used to determine the accuracy of AI for predicting LNM with a fivefold cross-validation in the training set and in a held-out test set.

RESULTS

We developed an AI model using a two-step attention-based DL approach without clinical features (AUC, 0.764). Incorporating clinical features into the model did not improve its prediction accuracy for LNM. Our model reduced unnecessary additional surgery by 15.1% more than using the current guidelines (67.4% vs. 82.5%). In patients with SM invasion depth of 1000 to 2000 μm, the AI avoided 16.1% of unnecessary additional surgery than using the JSCCR guidelines.

CONCLUSIONS

Our study is the first to show that AI trained with DL of H&E-stained WSIs has the potential to predict LNM in T1 CRC using only endoscopically resected specimens with conventional histologic risk factors.

Current problems and perspectives of pathological risk factors for lymph node metastasis in T1 colorectal cancer: Systematic review

With the prevalence of endoscopic submucosal dissection and endoscopic full thickness resection, which enable complete resection of T1 colorectal cancer with a negative margin, the treatment strategy following endoscopic resection has become more important. The necessity of secondary surgical resection is determined on the basis of the risk of lymph node metastasis according to the histopathological findings of resected specimens because ~10% of T1 colorectal cancer cases have lymph node metastasis. The current Japanese treatment guidelines state four risk factors for lymph node metastasis: lymphovascular invasion, histological differentiation, depth of submucosal invasion, and tumor budding. These guidelines have succeeded in stratifying the low-risk group for lymph node metastasis, in which endoscopic resection alone is acceptable for cure. On the other hand, there are some problems: there is variation in diagnosis methods and low interobserver agreement for each pathological factor and 90% of surgical resections are unnecessary, with lymph node metastasis negativity. To ensure patients with T1 colorectal cancer receive more appropriate treatment, these problems should be addressed. In this systematic review, we gave some suggestions to these practical issues of four pathological factors as predictors.

Establishment of a Dynamic Nomogram for Predicting the Risk of Lymph Node Metastasis in T1 Stage Colorectal Cancer

Background

Accurate prediction of the risk of lymph node metastasis in patients with stage T1 colorectal cancer is crucial for the formulation of treatment plans for additional surgery and lymph node dissection after endoscopic resection. The purpose of this study was to establish a predictive model for evaluating the risk of LNM in patients with stage T1 colorectal cancer.

Methods

The clinicopathological and imaging data of 179 patients with T1 stage colorectal cancer who underwent radical resection of colorectal cancer were collected. LASSO regression and a random forest algorithm were used to screen the important risk factors for LNM, and a multivariate logistic regression equation and dynamic nomogram were constructed. The C index, Calibration curve, and area under the ROC curve were used to evaluate the discriminant and prediction ability of the nomogram. The net reclassification index (NRI), comprehensive discriminant improvement index (IDI), and clinical decision curve (DCA) were compared with traditional ESMO criteria to evaluate the accuracy, net benefit, and clinical practicability of the model.

Results

The probability of lymph node metastasis in patients with T1 colorectal cancer was 11.17% (20/179). Multivariate analysis showed that the independent risk factors for LNM in T1 colorectal cancer were submucosal invasion depth, histological grade, CEA, lymphovascular invasion, and imaging results. The dynamic nomogram model constructed with independent risk factors has good discrimination and prediction capabilities. The C index was 0.914, the corrected C index was 0.890, the area under the ROC curve was 0.914, and the accuracy, sensitivity, and specificity were 93.3, 80.0, and 91.8%, respectively. The NRI, IDI, and DCA show that this model is superior to the ESMO standard.

Conclusion

This study establishes a dynamic nomogram that can effectively predict the risk of lymph node metastasis in patients with stage T1 colorectal cancer, which will provide certain help for the formulation of subsequent treatment plans for patients with stage T1 CRC after endoscopic resection.

The Importance of Being “That” Colorectal pT1: A Combined Clinico-Pathological Predictive Score to Improve Nodal Risk Stratification

The management of endoscopically resected pT1 colorectal cancer (CRC) relies on nodal metastasis risk estimation based on the assessment of specific histopathological features. Avoiding the overtreatment of metastasis-free patients represents a crucial unmet clinical need. By analyzing a consecutive series of 207 pT1 CRCs treated with colectomy and lymphadenectomy, this study aimed to develop a novel clinicopathological score to improve pT1 CRC metastasis prediction. First, we established the clinicopathological profile of metastatic cases: lymphovascular invasion (OR: 23.8; CI: 5.12–110.9) and high-grade tumor budding (OR: 5.21; CI: 1.60–16.8) correlated with an increased risk of nodal metastasis, while age at diagnosis >65 years (OR: 0.26; CI: 0.09–0.71) and high tumor-infiltrating lymphocytes (OR: 0.19; CI: 0.06–0.59) showed a protective effect. Combining these features, we built a five-tier risk score that, applied to our series, identified cases with a higher risk (score ≥ 2) of nodal metastasis (OR: 7.7; CI: 2.4–24.4). Notably, a score of 0 was only assigned to cases with no metastases (13/13 cases) and all the score 4 samples (2/2 cases) showed nodal metastases. In conclusion, we developed an effectively combined score to assess pT1 CRC nodal metastasis risk. We believe that its adoption within a multidisciplinary pT1 unit could improve patients' clinical management and limit surgical overtreatment.

Ostomy prevalence and survival in elderly patients with stage III and IV rectal cancer

AIM

Oncological strategies in the elderly population are often debated. The objective of this study was to investigate the survival rates and prevalence of ostomy in elderly patients operated on for stage III and IV rectal cancers.

METHODS

This retrospective multicentric population-based study included 151 patients aged ≥75 years with stage III and IV rectal adenocarcinoma who underwent surgery between 2007 and 2014. Multivariable logistic regression was used to assess the impact of different prognostic factors.

RESULTS

The median age of the patients was 81 years (range: 75-97 years) with 40 patients >85 years of age. Age was significantly correlated with overall survival (OS) in both stage III and IV cancers (P < 0.001). For patients ≥80 years the presence of comorbid conditions was associated with a lower chance of survival (P = 0.02). A digestive stoma was created in 67 (76.1%) patients with stage III cancer and 26 (29.54%) had a stoma reversal. A palliative derivative stoma was performed in half of patients with stage IV cancer. Adjuvant chemotherapy was independently associated with improved 5-year OS (P < 0.001).

CONCLUSIONS

Age, comorbidities and adjuvant chemotherapy were independent predictors for OS. Resection of rectal tumors in fit elderly patients should be promoted; however, patients should be aware of the high risk of stoma. Geriatr Gerontol Int 2021; 21: 670-675.

Artificial Intelligence System to Determine Risk of T1 Colorectal Cancer Metastasis to Lymph Node

BACKGROUND & AIMS

In accordance with guidelines, most patients with T1 colorectal cancers (CRC) undergo surgical resection with lymph node dissection, despite the low incidence (∼10%) of metastasis to lymph nodes. To reduce unnecessary surgical resections, we used artificial intelligence to build a model to identify T1 colorectal tumors at risk for metastasis to lymph node and validated the model in a separate set of patients.

METHODS

We collected data from 3134 patients with T1 CRC treated at 6 hospitals in Japan from April 1997 through September 2017 (training cohort). We developed a machine-learning artificial neural network (ANN) using data on patients' age and sex, as well as tumor size, location, morphology, lymphatic and vascular invasion, and histologic grade. We then conducted the external validation on the ANN model using independent 939 patients at another hospital during the same period (validation cohort). We calculated areas under the receiver operator characteristics curves (AUCs) for the ability of the model and US guidelines to identify patients with lymph node metastases.

RESULTS

Lymph node metastases were found in 319 (10.2%) of 3134 patients in the training cohort and 79 (8.4%) of /939 patients in the validation cohort. In the validation cohort, the ANN model identified patients with lymph node metastases with an AUC of 0.83, whereas the guidelines identified patients with lymph node metastases with an AUC of 0.73 (P < .001). When the analysis was limited to patients with initial endoscopic resection (n = 517), the ANN model identified patients with lymph node metastases with an AUC of 0.84 and the guidelines identified these patients with an AUC of 0.77 (P = .005).

CONCLUSIONS

The ANN model outperformed guidelines in identifying patients with T1 CRCs who had lymph node metastases. This model might be used to determine which patients require additional surgery after endoscopic resection of T1 CRCs. UMIN Clinical Trials Registry no: UMIN000038609.

Evaluation of KRAS, NRAS and BRAF mutational status and microsatellite instability in early colorectal carcinomas invading the submucosa (pT1): towards an in-house molecular prognostication for pathologists?

AIM

We aimed to study the prognostic value of KRAS, NRAS, BRAF mutations and microsatellite stable (MSS)/instable (MSI) in the field of colorectal cancer invading the submucosa (ie, pT1 colorectal cancer (CRC)).

METHODS

We led a case-control study in tumour samples from 60 patients with pT1 CRC with (20 cases) and without (40 cases) metastatic evolution (5 years of follow-up) which were analysed for KRAS, NRAS, BRAF mutations (Idylla testing and next generation sequencing, NGS) and MSS/MSI status (Idylla testing and expression of mismatch repair (MMR) proteins using immunohistochemistry).

RESULTS

KRAS mutations were encountered in 11/20 (55%) cases and 21/40 (52.5%) controls (OR=1.11 (0.38 to 3.25), p=0.8548), NRAS mutations in 1/20 (5%) cases and 3/40 (7.5%) controls (OR=3.08 (0.62 to 15.39), p=0.1698) and BRAF mutations in 3/20 (15%) cases and 6/40 (15%) controls (OR=1.00 (0.22 to 4.5), p=1.00). A MSI status was diagnosed in 3/20 (15%) cases and 5/40 (12.5%) controls (OR=1.2353 (0.26 to 5.79), p=0.7885). Beyond the absence of significant association between the metastatic evolution and any of the studied molecular parameters, we observed a very good agreement between methods analysing KRAS, NRAS and BRAF mutations (Kappa value of 0.849 (0.748 to 0.95) between Idylla and NGS) and MSS/MSI (Idylla)-proficient MMR/deficient MMR (immunohistochemistry) status (Kappa value of 1.00).

CONCLUSION

Although being feasible using the fully automated Idylla method as well as NGS, the molecular testing of KRAS, NRAS, BRAF and MSS/MSI status does not seem useful for prognostic purpose in the field of pT1 CRC.