BACKGROUND:The low prevalence of the BRAF V600E mutation in colorectal cancers (CRCs) in Chinese populations has stimulated concern about the efficacy of BRAF mutation analysis for Lynch syndrome (LS) screening.METHODS:In total, 169 of 4104 consecutive CRC patients with absent MLH1 staining were analyzed to compare the utility of the BRAF V600E mutation testing with MLH1 promoter methylation analysis in the Chinese population. Germline genetic testing was performed in patients with wild-type BRAF/methylated MLH1.RESULTS:Compared with BRAF genotyping, the use of MLH1 methylation testing alone to evaluate patients with MLH1 deficiency reduced referral rates for germline testing by 1.8-fold (82.8% vs. 47.1%). However, 6 patients harboring MLH1 promoter methylation were verified to have LS through germline genetic testing. It is notable that all 6 patients had a family history of CRC in at least 1 first-degree relative (FDR) or second-degree relative (SDR). The combination of MLH1 promoter methylation analysis and a family history of CRC could preclude significantly more patients from germline genetic testing than from BRAF mutation testing alone (45.5% vs. 17.2%, p<0.001) and decrease the number of misdiagnosed LS patients with MLH1 promoter methylation.CONCLUSION:The combination of a family history of CRC with MLH1 promoter methylation analysis showed better performance than BRAF mutation testing in the selection of patients in the Chinese population for germline genetic testing.

Lynch syndrome (LS) is the most common hereditary colorectal cancer (CRC) predisposition syndrome. We performed a large‐scale study to assess a screening strategy for identifying LS in Chinese CRC patients in routine clinical testing. A total of 4,195 eligible CRCs were universally screened. Then, 8.7% of CRCs were detected with dMMR. The incidence of LS was 2.7% (115 of 4,195) in this cohort; among patients over 70 years of age, only 0.3% (2 of 678) were diagnosed as LS. Then, 17.4% of LS cases showed large genomic deletions/duplications. LS probands developed CRCs predominantly at proximal colon location. The frequency of BRAF V600E mutation among Chinese CRCs was significantly lower than that among Western populations, and MLH1 promoter methylation significantly improved the efficiency of genetic screening for LS among MLH1‐deficient patients. A comprehensive molecular testing strategy that includes detection of large genomic rearrangements is imperative for the diagnosis of LS. Among CRC patients aged 70 years or younger, a selective strategy for LS screening might be considered for routine clinical testing.

Objectives The microscopic evaluation of slides has been gradually moving towards all digital in recent years, leading to the possibility for computer-aided diagnosis. It is worthwhile to know the similarities between deep learning models and pathologists before we put them into practical scenarios. The simple criteria of colorectal adenoma diagnosis make it to be a perfect testbed for this study. Design The deep learning model was trained by 177 accurately labelled training slides (156 with adenoma). The detailed labelling was performed on a self-developed annotation system based on iPad. We built the model based on DeepLab v2 with ResNet-34. The model performance was tested on 194 test slides and compared with five pathologists. Furthermore, the generalisation ability of the learning model was tested by extra 168 slides (111 with adenoma) collected from two other hospitals. Results The deep learning model achieved an area under the curve of 0.92 and obtained a slide-level accuracy of over 90% on slides from two other hospitals. The performance was on par with the performance of experienced pathologists, exceeding the average pathologist. By investigating the feature maps and cases misdiagnosed by the model, we found the concordance of thinking process in diagnosis between the deep learning model and pathologists. Conclusions The deep learning model for colorectal adenoma diagnosis is quite similar to pathologists. It is on-par with pathologists’ performance, makes similar mistakes and learns rational reasoning logics. Meanwhile, it obtains high accuracy on slides collected from different hospitals with significant staining configuration variations.

AbstractEsophageal squamous cell carcinoma (ESCC) is more prevalent than esophageal adenocarcinoma in Asia, especially in China, where more than half of ESCC cases occur worldwide. Many studies have reported that the automatic detection of lymph node metastasis using semantic segmentation shows good performance in breast cancer and other adenocarcinomas. However, the detection of squamous cell carcinoma metastasis in hematoxylin‐eosin (H&E)‐stained slides has never been reported. We collected a training set of 110 esophageal lymph node slides with metastasis and 132 lymph node slides without metastasis. An iPad‐based annotation system was used to draw the contours of the cancer metastasis region. A DeepLab v3 model was trained to achieve the best fit with the training data. The learned model could estimate the probability of metastasis. To evaluate the effectiveness of the detection model of learned metastasis, we used another large cohort of clinical H&E‐stained esophageal lymph node slides containing 795 esophageal lymph nodes from 154 esophageal cancer patients. The basic authenticity label for each slide was confirmed by experienced pathologists. After filtering isolated noise in the prediction, we obtained an accuracy of 94%. Furthermore, we applied the learned model to throat and lung lymph node squamous cell carcinoma metastases and achieved the following promising results: an accuracy of 96.7% in throat cancer and an accuracy of 90% in lung cancer. In this work, we organized an annotated dataset of H&E‐stained esophageal lymph node and trained a deep neural network to detect lymph node metastasis in H&E‐stained slides of squamous cell carcinoma automatically. Moreover, it is possible to use this model to detect lymph nodes metastasis in squamous cell carcinoma from other organs. This study directly demonstrates the potential for determining the localization of squamous cell carcinoma metastases in lymph node and assisting in pathological diagnosis.

493 Background: Within MLH1-deficient colorectal cancer (CRC), BRAF mutation and MLH1 promotor hypermethylation strongly suggest a sporadic origin, providing exclusion criteria for Lynch syndrome (LS). However, the prevalence of BRAF mutation in Chinese CRCs are lower than that reported in Western countries, leading to concerns on the specificity of BRAF mutation for LS screening. The aim of this study was to assess the clinical usefulness of BRAF V600E mutation tesing and MLH1 promoter hypermethylation analyses to improve the yield of the diagnostic algorithm for discriminating LS in Chinese population. Methods: 4104 consecutive patients who had undergone surgery for CRCs between December 2011 and December 2014 in Chinese National Cancer Center were reviewed. 169 patients with MLH1 deficiency were analyzed. The MLH1 methylation status was evaluated by methylation-specific PCR; the BRAF mutation status was assessed by real-time PCR. Germline mutation testing was performed on 28 patients with BRAF-negative staining/MLH1 hypermethylation and family cancer history. Results: 52.9% (89/169) of patients displayed MLH1 hypermethylation. However, the BRAF mutation was found in 17.2% (29/169) of patients, and only 29.2% (26/89) of CRCs displaying MLH1 promoter hypermethylation harbored BRAF mutation, much lower from that reported by the western countries. Somatic methylation was identified in five LS patients. A CRC history in at least one first-degree relative had an outstanding ability to select patients with MLH1 hypermethylation for germline mutation testing, with 100% sensitivity and 82.6% specificity. Conclusions: The prevalence of BRAF mutation in Chinese CRCs are lower than that reported in Western countries leading to a decreased efficacy on triage patients for germline mutation. On the countary, MLH1 methylation analysis can improve the diagostic yield of LS, therefore play an indispensable role in LS screening among Chinese population.

Purpose To better understand the gene mutational status and heterogeneity between primary and metastatic CRC (mCRC) using a sensitive sequencing method. Methods The mutational status of EGFR, KRAS, NRAS, PIK3CA, ERBB2, BRAF, KIT , and PDGFRA was analyzed in 65 patients, with 147 samples of primary and paired live or lung metastatic CRC, using next-generation sequencing (NGS), quantitative RT-PCR (qPCR), and Sanger sequencing. Results Fifteen cases (15/22, 68.2%) of lung mCRC and thirteen cases (13/20, 65%) of liver mCRC harboured the same mutation profiles of KRAS, NRAS , or BRAF in the primary lesions. To all detected genes, 11 cases (11/22, 50%) of lung mCRC and 11 cases (11/20, 55%) of liver mCRC showed different mutational genes in the primary tumours. KRAS and BRAF mutations were more frequent in lung metastatic lesions ( p = 0.004 and 0.003, respectively). The gene mutations in KRAS, NRAS, BRAF , and PIK3CA in the lung metastatic sites were more frequent than those in the liver metastatic sites (86.7 vs. 44%, respectively, p = 0.000). Some new mutations were not covered in the qPCR ranges but were detected by NGS. Conclusion The study demonstrated that the discordance of gene mutational status between paired primary and metastatic tumours is rather high when detected by NGS. Evaluating the mutational status of both the primary and metastatic tumours should be considered in clinical mutation testing.

我国大肠癌(包含结肠癌及直肠癌)发病率和病死率在恶性肿瘤中均居前5位,并呈逐年快速上升态势.2016年ASCO年会上,基于CALGB/SWOG80405研究的回顾性亚组分析显示,原发左半结肠癌患者的总生存期(OS)显著优于右半结肠癌(33.3个月 vs 19.4个月).其中,在接受化疗联合贝伐珠单抗治疗组中,KRAS 野生型左、右半结肠癌患者的 OS 分别为31.4个月与24.2个月.而接受化疗联合西妥昔单抗治疗组中,KRAS野生型左、右半结肠癌患者的OS分别为36.0个月与16.7个月,提示2种分子靶向药物治疗左、右半结肠癌的效果存在明显差异.左半结肠癌患者从表皮生长因子受体(EGFR)单抗治疗中有更明显的获益,而右半结肠癌患者使用贝伐珠单抗较西妥昔单抗明显改善生存.基于上述研究,2017年NC-CN 指南进行了重要更新,仅对左半结肠癌的 KRAS/NRAS野生型患者推荐一线行化疗联合 EGFR 单抗治疗.

Lynch syndrome is the most common form of hereditary CRC, accounting for approximately 2~3% of population-based CRC. Individuals with Lynch syndrome have an increased risk for extracolonic cancer and second primary colorectal cancer. It has been dedicated that Lynch syndrome is an autosomal dominant cancer predisposition syndrome caused by germline mutations in the DNA mismatch repair genes (MMR)—MLH1, MSH2, MSH6 and PMS2, and EPCAM mutation is also found associated with MSH2 deficiency recently. Even though clinical diagnostic criteria (Amsterdan I/II, revised Bethesda) have been released decades ago, molecular testing of MMR genes is still the golden standard of Lynch syndrome diagnosis. Colorectal cancer patients less than 70y, are recommended to get into the procedure of Lynch syndrome screening, starting with microsatellite instability and/or immunohistochemical analysis on the tumor specimen followed by germline genetic testing and possibly further studies, such as MLPA. Notably, Testing on BRAF mutation is necessary, when the expression of MLH1 is lost, in order to exclude the sporadic colorectal cancer. Individuals with identified germline mutations need further screening for their immediate relatives.