Artificial neural network for prediction of distant metastasis in colorectal cancer

Background and Objectives: Artificial neural networks (ANNs) are flexible and nonlinear models which can be used by clinical oncologists in medical research as decision making tools. This study aimed to predict distant metastasis (DM) of colorectal cancer (CRC) patients using an ANN model. Methods: The data of this study were gathered from 1219 registered CRC patients at the Research Center for Gastroenterology and Liver Disease of Shahid Beheshti University of Medical Sciences, Tehran, Iran (January 2002 and October 2007). For prediction of DM in CRC patients, neural network (NN) and logistic regression (LR) models were used. Then, the concordance index (C index) and the area under receiver operating characteristic curve (AUROC) were used for comparison of neural network and logistic regression models. Data analysis was performed with R 2.14.1 software. Results: The C indices of ANN and LR models for colon cancer data were calculated to be 0.812 and 0.779, respectively. Based on testing dataset, the AUROC for ANN and LR models were 0.82 and 0.77, respectively. This means that the accuracy of ANN prediction was better than for LR prediction. Conclusion: The ANN model is a suitable method for predicting DM and in that case is suggested as a good classifier that usefulness to treatment goals.     

Application of artificial intelligence to the diagnosis and therapy of colorectal cancer

Artificial intelligence (AI) is a relatively new branch of computer science involving many disciplines and technologies, including robotics, speech recognition, natural language and image recognition or processing, and machine learning. Recently, AI has been widely applied in the medical field. The effective combination of AI and big data can provide convenient and efficient medical services for patients. Colorectal cancer (CRC) is a common type of gastrointestinal cancer. The early diagnosis and treatment of CRC are key factors affecting its prognosis. This review summarizes the research progress and clinical application value of AI in the investigation, early diagnosis, treatment, and prognosis of CRC, to provide a comprehensive theoretical basis for AI as a promising diagnostic and treatment tool for CRC.     

Prospective evaluation of prognostic factors in patients with colorectal cancer undergoing curative resection

BACKGROUND AND OBJECTIVES: Streak type, defined by the presence of white streaks at the advancing margin of tumor invasion; focal dedifferentiation, seen as undifferentiated cancer cells with an infiltrative pattern at the invasive front; and perineural invasion have been demonstrated to be prognostic factors in patients with colorectal cancer. We performed a prospective study to examine the usefulness of these features as prognostic factors. METHODS: We reviewed a total of 391 patients who underwent curative surgery for colorectal cancer between May 1997 and June 1999. Of these, 50 patients with multiple cancers were excluded, and a total of 341 patients were finally entered into the study. RESULTS: Of the prognostic factors investigated, depth of invasion, lymph node status, lymphatic invasion, venous invasion, growth type, streak type, focal dedifferentiation, and perineural invasion were significant prognostic factors in univariate analysis. In multivariate analysis, lymph node status, depth of invasion, and perineural invasion were significant prognostic factors. The survival of patients with perineural invasion was significantly poorer than that of patients without perineural invasion in both stage II and III cancer. CONCLUSIONS: Perineural invasion status can be used to facilitate the selection of colorectal cancer patients for adjuvant chemotherapy and should be described in routine pathology reports.     

Use of microsatellite analysis in young patients with colorectal cancer to identify those with hereditary nonpolyposis colorectal cancer

BACKGROUND AND OBJECTIVES: The frequency of microsatellite instability (MSI) in young patients with colorectal cancer was evaluated, including reexamination of the medical and family history of each patient, and interviews with the patients to determine any possible new occurrence of hereditary nonpolyposis colorectal cancer (HNPCC) in the patients themselves or their family members. METHODS: Fifty-three young patients (younger than 40 years of age) with colorectal cancer were selected and investigated. DNA was extracted from paraffin sections and microsatellite analysis was performed. RESULTS: The frequency of MSI among the young patients with colorectal cancer was 50.9%, which was significantly higher than the rate of 12-21% noted in older patients with colorectal cancer (P < 0.001). For the 24 young patients with colorectal cancer who did not have MSI, only one case of HNPCC kindred and two cases with a family history of cancer were identified. In contrast, among the 20 young patients with colorectal cancer who had MSI, five cases of HNPCC kindred, two cases with metachronous patients with colorectal cancer, and three cases with a family history of cancer were identified. CONCLUSION: Our results suggest that a defect in the DNA mismatch repair system may play some role in carcinogenesis in young patients with colorectal cancer. Microsatellite analysis and subsequent interviews regarding medical and family history are useful tools for efficiently identifying possible cases of HNPCC among young patients with colorectal cancer.     

直肠癌脑转移患者预后的影响因素分析

目的：分析影响直肠癌脑转移患者预后的相关因素。方法：搜集2008年6月－2010年6月期间来我院治疗的100例直肠癌脑转移患者的临床资料,根据治疗方法分为实验组和对照组,实验组为行手术治疗的46例患者,对照组为行放化疗的54例患者。采用Logistic逐步回归分析,分析影响脑卒中瘫痪经手术治疗患者预后的因素。结果：实验组患者的平均生存时间、1年生存率、3年生存率以及5年生存率均优于对照组患者,具有明显差异（P＜0．05）。Logistic回归分析示,性别、原发肿瘤组织学分级等因素不是影响直肠癌脑转移患者预后的因素;影响直肠癌脑转移患者预后的相关因素是原发肿瘤直径、脑转移灶数目、血清CEA水平、治疗方法、脑转移灶治疗与否以及年龄。经Logistic回归逐步分析可知,血清CEA水平对患者预后的影响最大（OR＝3．743）,其次是脑转移灶数目（OR＝3．153）、治疗方法（OR＝2．946）,年龄（＞60岁）影响最小（OR＝1．900）。结论：影响直肠癌脑转移患者预后的相关因素是原发肿瘤直径、脑转移灶数目、血清CEA水平、治疗方法、脑转移灶治疗与否以及年龄等。及时掌握这些预后影响因素能够显著提高直肠癌脑转移患者生存状况,在临床上具有重要的临床意义。     

结直肠癌预后的多因素回归分析

目的 探讨影响结直肠癌预后的因素在预测结直肠癌术后生存中的价值。 方法 应用多因素回归的分析方法,回顾性分析有完整临床病理资料和随访资料的941例结直肠癌患者的临床特点、病理特征及其对预后的影响。 结果 结直肠癌患者总的3,5年生存率分别为63.2％和60.8％,中位生存时间为1841d。单因素分析显示,其预后与肿瘤的大体分型、侵袭程度、转移情况、分化等级、病理分期以及癌性肠梗阻均有相关性。应用Cox比例危险回归模型分析,则显示肿瘤的大体分型、分化程度、肠壁的侵袭深度和病理分期是影响结直肠癌患者术后生存的独立因素。 结论 病理分期是影响结直肠癌预后最重要的一个指标(P<0.0005),对于指导手术治疗、术后辅助治疗和判断预后方面具有重要作用。     

Neutrophil-lymphocyte ratio as a prognostic factor in colorectal cancer

BACKGROUND AND OBJECTIVES: Simple methods to identify colorectal cancer patients at risk of recurrence are needed. This study aimed to determine if neutrophil-to-lymphocyte ratio (NLR) predicts survival in colorectal cancer patients. METHODS: Two-hundred thirty patients diagnosed with colorectal cancer over a two-year period were identified from a prospectively maintained colorectal cancer database. NLR was calculated from pre-operative full blood counts. In the case of patients who did not undergo surgery, the full blood count from their out-patient visit was used. Known prognostic factors were recorded. Overall and cancer-specific survival were calculated. RESULTS: Pre-operative NLR greater than 5 correlated with overall and cancer-specific survival in univariate analyses. NLR was not independent of Dukes stage. CONCLUSIONS: Pre-operative NLR may represent a simple method of identifying colorectal cancer patients with a poor prognosis pre-operatively.     

A cohort study of the prognostic and treatment predictive value of SATB2 expression in colorectal cancer

Background:

Special AT-rich sequence-binding protein 2 (SATB2) is a novel diagnostic marker of colorectal cancer (CRC), and loss of SATB2 has been linked to poor survival from the disease. In this study, we validated the prognostic ability of SATB2 expression in a large, prospective CRC cohort.

Methods:

Immunohistochemical SATB2 expression was assessed in 527 incident CRC cases from the Malmö Diet and Cancer Study. Kaplan–Meier analysis and Cox proportional hazards modelling were used to explore the impact of SATB2 expression on cancer-specific survival (CSS) and overall survival (OS).

Results:

High SATB2 expression was associated with a prolonged CSS in the full cohort (hazard ratio (HR)=0.61; 95% CI 0.41–0.92) and in colon cancer (HR=0.39; 95% CI 0.20–0.75), remaining significant in multivariable analysis of colon cancer (HR=0.49; 95% CI 0.25–0.96), with similar findings for OS. In curatively resected stage III-IV patients, a significant benefit from adjuvant and/or neoadjuvant therapy was observed for SATB2 high tumours (P_interaction=0.037 for OS) and high SATB2 expression in rectal cancer correlated with an enhanced effect of neoadjuvant therapy (P_interaction=0.033 for OS).

Conclusion:

High SATB2 expression is an independent marker of good prognosis in colon cancer and may modulate sensitivity to chemotherapy and radiation.     

The prognostic effect of histological tumor grade in node-negative breast cancer patients

Summary The prognostic effect of histological tumor grade was evaluated in 1036 patients with early breast cancer (pT1 pN0 M0) entered into a trial comparing mastectomy and breast preserving treatment. All analyses were adjusted for the factors treatment, patients' age, and tumor size. Tumor grade was defined according to Bloom and Richardson based on the sum of scores assigned to each of three histological features: 1) degree of differentiation, 2) pleomorphism, and 3) mitotic index. The relative importance of these factors with regard to disease-free survival was evaluated. In univariate as well as in multivariate analyses the pleomorphism was the only factor showing a significant effect (univariate: p=0.0024, multivariate: p=0.015). It was investigated how the factors should be combined to define a histological grading score which yields the best possible classification of the patients with respect to prognosis. A new grading system was defined splitting the patients into three groups: 1) pleomorphism 1; 2) pleomorphism 2 or pleomorphism 3 and mitotic index 1; 3) pleomorphism 3 and mitotic index 2 or 3. This yields a good classification of the patients with respect to prognosis (p=0.0004). The prognostic effect of this score was compared with the effects of the grading systems proposed in the literature. According to Bloom and Richardson and in the modified version by Schauer and Weiss, grading is based on the sum of scores of the various histological factors. Therefore, the strong effect of the pleomorphism was diluted in these grading definitions (Bloom and Richardson: p=0.03, Schauer and Weiss: p=0.028). The grading system proposed by Le Doussalet al. consists only of the scores of pleomorphism and mitotic index (p=0.014). In summary, the factor pleomorphism showed a stronger effect on disease-free survival by itself than the grading systems proposed in the literature.     

Surgical resection of pulmonary metastases from colorectal cancer: Four favourable prognostic factors

BACKGROUND: Surgical resection has been the first choice for treatment of pulmonary metastases from colorectal cancer; however, indications for surgery have yet to be adequately clarified. In considering strategies for the treatment of pulmonary metastases from colorectal carcinoma, determination of disease status as either systemic or pre-systemic is of primary importance. The aim of this study is to define the characteristics of those patients who are most likely to benefit from surgical resection. METHODS: Fifty-eight patients who underwent pulmonary resection for colorectal metastases were retrospectively reviewed and examined for clinicopathological factors. RESULTS: Overall, 5-year survival rates were 29%, with a median survival time (MST) of 27 months. Multivariate analysis identified four factors that indicate independent and favourable prognostic impact: three or less tumours, metachronous metastasis, negative hilar and/or mediastinal lymph node metastasis and normal prethoracotomy serum carcinoembryonic antigen level. The 5-year survival rate for 16 patients who satisfied all of these favourable characteristics was 62% (MST = 86 months), which was significantly better than those patients lacking these characteristics. The 5-year survival rate for 13 patients who underwent repeated metastasectomy was 37% (MST = 32 months). CONCLUSIONS: The four factors selected in our multivariate analysis appear to be favourable factors for the practical identification of those patients who are most likely to benefit from surgical resection. Repeated pulmonary resection for lung-only recurrence may benefit carefully selected patients.     

Identifying patients with undetected colorectal cancer: an independent validation of QCancer (Colorectal)

Background:

Early identification of colorectal cancer is an unresolved challenge and the predictive value of single symptoms is limited. We evaluated the performance of QCancer (Colorectal) prediction model for predicting the absolute risk of colorectal cancer in an independent UK cohort of patients from general practice records.

Methods:

A total of 2.1 million patients registered with a general practice surgery between 01 January 2000 and 30 June 2008, aged 30-84 years (3.7 million person-years) with 3712 colorectal cancer cases were included in the analysis. Colorectal cancer was defined as incident diagnosis of colorectal cancer during the 2 years after study entry.

Results:

The results from this independent and external validation of QCancer (Colorectal) prediction model demonstrated good performance data on a large cohort of general practice patients. QCancer (Colorectal) had very good discrimination with an area under the ROC curve of 0.92 (women) and 0.91 (men), and explained 68% (women) and 66% (men) of the variation. QCancer (Colorectal) was well calibrated across all tenths of risk and over all age ranges with predicted risks closely matching observed risks.

Conclusion:

QCancer (Colorectal) appears to be a useful tool for identifying undetected cases of undiagnosed colorectal cancer in primary care in the United Kingdom.     

Prediction of KRAS,NRAS and BRAF status in colorectal cancer patients with liver metastasis using a deep artificial neural network based on radiomics and semantic features

There is a critical need for development of improved methods capable of accurately predicting the RAS (KRAS and NRAS) and BRAF gene mutation status in patients with advanced colorectal cancer (CRC). The purpose of this study was to investigate whether radiomics and/or semantic features could improve the detection accuracy of RAS/BRAF gene mutation status in patients with colorectal liver metastasis (CRLM). In this retrospective study, 159 patients who had been diagnosed with CRLM in two hospitals were enrolled. All patients received lung and abdominal contrast-enhanced CT (CECT) scans prior to radiation therapy and chemotherapy. Semantic features were independently assessed by two radiologists. Radiomics features were extracted from the portal venous phase (PVP) of the CT scan for each patient. Seven machine learning algorithms were used to establish three scores based on the semantic, radiomics and the combination of both features. Two semantic and 851 radiomics features were used to predict the mutation status of RAS and BRAF using an artificial neural network method (ANN). This approach performed best out of the seven tested algorithms. We constructed three scores which were based on radiomics, semantic features and the combined scores. The combined score could distinguish between wild-type and mutant patients with an AUC of 0.95 in the primary cohort and 0.79 in the validation cohort. This study proved that the application of radiomics together with semantic features can improve non-invasive assessment of the gene mutation status of RAS (KRAS and NRAS) and BRAF in CRLM.     

Microsatellite instability in sporadic colorectal cancer is not an independent prognostic factor.

Hereditary non-polyposis colorectal cancer (HNPCC) is linked to an inherited defect in the DNA mismatch repair system. DNA from HNPCC tumours shows microsatellite instability (MSI). It has been reported that HNPCC patients have a better prognosis than patients with sporadic colorectal cancer. We examined whether the presence of MSI in a series of unselected colorectal tumours carries prognostic information. In a series of 181 unselected colorectal tumours, 22 tumours (12%) showed MSI. Survival analysis at 5-10 years follow-up showed no statistically significant difference in prognosis between MSI-positive and -negative tumours. Our results suggest that the MSI phenotype as such is not an independent prognostic factor.     

神经/脉管浸润与结直肠癌临床病理特征及预后的相关分析

目的 神经/脉管浸润对于结直肠癌患者的预后具有一定的影响,但尚未达成共识.本研究探讨神经/脉管浸润与结直肠癌临床病理特征的相关性及其对患者预后的影响.方法 收集2010-01-01-2011-12-31,在新疆医科大学附属肿瘤医院行根治性手术切除,并经病理确诊的226例结直肠癌患者的临床病理资料.根据是否有神经/脉管浸润分为神经/脉管浸润组(PVG) 42例和非神经/脉管浸润组(nPVG) 184例.统计分析2组患者的临床病理因素特点及预后.结果 单因素分析显示,肿瘤大小、浸润深度、淋巴结转移和TNM分期与结直肠癌神经/脉管浸润具有相关性,均P＜0.05;多因素Logistic回归分析显示,肿瘤大小和浸润深度与神经/脉管浸润密切相关,均P＜0.05;Kaplan-Meier法单因素分析显示,年龄、浸润深度、淋巴结转移、TNM分期及神经/脉管浸润与患者生存率具有相关性,均P＜0.01.PVG与nPVG组患者的3年总生存率分别为54.5％和84.7％,差异有统计学意义,x2=12.806,P＜0.001;3年无病生存率分别为42.3％和73.7％,差异有统计学意义,x2=13.629,P＜0.001.多因素Cox回归分析显示,浸润深度和神经/脉管浸润与结直肠癌患者的生存率密切相关,均P＜0.05.结论 肿瘤大小和浸润深度是影响结直肠癌神经/脉管浸润的独立危险因素;浸润深度和神经/脉管浸润是结直肠癌患者预后的独立影响因素.     

Assessment of prognostic factors in long-term survival of male and female patients with colorectal cancer using non-mixture cure model based on the Weibull distribution

ObjectiveColorectal cancer (CRC) is known as one of the malignant form of cells growing in the inner lining of colon and rectum which could seriously affect the cure rate of patients. We aimed to evaluate the effect of prognostic factors on cure fraction of CRC patients.MethodsA total of 1043 CRC patients were included to the study from December 2001 to January 2007 at the Research Center of Gastroenterology and Liver Disease in Shahid Beheshti University of Medical Sciences, Tehran, Iran. Patients’ information was extracted from their medical records, then they were followed to identify their death status via phone-call. Weibull non-mixture cure model was used to evaluate the effect of the risk factors on cure fraction of CRC patients.ResultsThe five-years survival rate was 0.66 (males: 0.64 and female: 0.69). The median survival time for non-cured CRC patients were 3.45 years (males: 3.46; females = 3.45 years). In the single Weibull model, BMI≥30 (OR = 4.61, p-value = 0.033), poorly differentiated tumor grade (OR = 0.36, p-value = 0.036), tumor size≥25 mm (OR = 0.22, p-value = 0.046), and N1-stage (OR = 0.42, p-value = 0.005) had significant effect on females’ cure fraction. Also, cure fraction of male CRC patients significantly affected by BMI (levels:25.0–29.9-OR = 12.13-p-value<0.001; ≥30-OR = 7.00-p-value = 0.017), T1-stage (OR = 0.52, p-value = 0.021), M1-stage (OR = 0.45, p-value = 0.007), IV-staging (OR = 0.36, p-value = 0.041) and IBD (OR = 0.26, p-value = 0.017). In multiple Weibull model, females were associated with tumor size≥25 mm (OR = 0.20, p-value = 0.044) and N1-stage (OR = 0.45, p-value = 0.013) and males were affected by M1-stage (OR = 0.41, p-value = 0.011) and IBD (OR = 0.20, p-value = 0.022).The cure fraction of males and females CRC patients was 64% and 69%, respectively.ConclusionsThe prognostic factors for cure fraction of patients with CRC may be different among males and females. Further multicenter studies are required to assess the effect of common prognostic factors between males and females.     

Prognostic value of preoperative serum CEA level compared to clinical staging: III. An approach to scoring of prognostic factors in colorectal cancer

In a clinical study of observed postoperative survival of colorectal cancer patients, we investigated the application of a risk score based on tumor-related prognostic parameters. Six hundred seventy-four patients have been registered for primary surgery of colorectal cancer since 1974 who did not receive further postoperative treatments. The prognostic parameters included operability, tumor extension, and preoperative serum carcinoembryonic antigen (CEA) level. The scoring system was based on the average death-rate ratios of subgroups of patients and their age and sex-matched reference groups derived from the general life table of the population of the Federal Republic of Germany. The individual score sums of the patients exhibited score sum ranges which characterized groups of patients with entirely different observed survival. The prediction of individual survival after primary operation was only partly possible. In the plot of individual survivals vs individual score sums, a marginal risk zone was obtained which evidently represents the zone of maximum expected survival of patients who do not receive further postoperative treatment.     

乳腺癌肝转移患者预后的多因素分析

目的:分析影响乳腺癌肝转移患者的预后因素。方法:1996年1月~2003年12月,本院收治114例乳腺癌肝转移住院患者。利用Cox模型分析患者的预后因素。结果:乳腺癌肝转移患者的一线、二线、三线和四线化疗的疗效依此下降,分别为31.9%(22/69)、27.8%(10/36)、16.7%(3/18)和0%(0/10)。单因素分析提示无肝功能损害和手术后无疾病复发间歇期长者预后好。Cox多因素分析模型分析结果提示有无肝功能损害是影响生存率的独立预后因素。结论:晚期乳腺癌化疗疗效随着使用方案次数的增多而降低。有无肝功能损害是影响乳腺癌肝转移患者的独立预后因素。     

肝外胆管癌预后因素分析

目的分析肝外胆管癌患者的预后因素.方法单因素按不同水平分层的人时死亡率计算方法和多因素Cox模型分析预后因素.结果:手术方式和术后生存期的关系最为密切,行切除术的患者术后生存期明显延长,其次是病理类型、术后并发症和术后生存期关系亦非常密切,建立了肝外胆管癌术后生存危险状态函数.在此基础上比较了不同术式的1、3、5年生存率.探查术的1、3、5年生存率分别为11.48%、0.06%、0.003%;外引流术的1、3、5年生存率30%、16.12%、0.31%,切除术的1、3、5年生存率分别为69.23%、28.10%、17.04%.结论: 手术切除是治疗肝外胆管癌患者的最佳方案.