Correlation of pretreatment hemoglobin and platelet counts with clinicopathological features in colorectal cancer in Saudi population

Prediction of Survival and Recurrence Patterns by Machine Learning in Gastric Cancer Cases Undergoing Radiation Therapy and Chemotherapy

Purpose

Radical surgery is the most important treatment modality in gastric cancer. Preoperative or postoperative radiation therapy (RT) and perioperative chemotherapy are the treatment options that should be added to surgery. This study aimed to evaluate the overall survival (OS) and recurrence patterns by machine learning in gastric cancer cases undergoing RT.

Methods and Materials

Between 2012 and 2019, the OS and recurrence patterns of 75 gastric cancer cases receiving RT ± chemotherapy at the Department of Radiation Oncology were evaluated by machine learning. Logistic regression, multilayer perceptron, XGBoost, support vector classification, random forest, and Gaussian Naive Bayes (GNB) algorithms were used to predict OS, hematogenous distant metastases, and peritoneal metastases. After the correlation analysis, the backward feature selection was performed as the variable selection method, and the variables with P values less than .005 were selected.

Results

Over the median 23-month follow-up, recurrence was seen in 33 cases, and 36 patients died. The median OS was 23 (min: 7; max: 82) months, and the disease-free survival was 18 (min: 5, max: 80) months. The most common recurrence pattern was hematogenous distant metastasis, followed by peritoneal metastasis. In this study, the most successful algorithms in the prediction of OS, distant metastases, and peritoneal metastases were found to be GNB with an accuracy of 81% (95% confidence interval [CI], 0.65-0.97, area under the curve [AUC]: 0.89), XGBoost with 86% accuracy (95% CI, 0.74-0.97, AUC: 0.86), and random forest with 97% accuracy (95% CI, 0.92-1.00, AUC: 0.97), respectively.

Conclusions

In gastric cancer, GNB, XGBoost, and random forest algorithms were determined to be the most successful algorithms for predicting OS, distant metastases, and peritoneal metastases, respectively. To determine the most accurate algorithm and perhaps make personalized treatments applicable, more precise machine learning studies are needed with an increased number of cases in the coming years.

Extricating the Association Between the Prognostic Factors of Colorectal Cancer

PURPOSE

Colorectal cancer (CRC) is one of the recurring and lethal gastrointestinal tract disease rankings as the primary cause of worldwide morbidity and mortality. In general, the tumour node metastasis (TNM) and Dukes classification assist in diagnosis, prognosis and treatments of CRC along with haematological examinations and tumour demographic characterisations in patients.

METHODS

The present investigation is carried out on clinically acknowledged sixty-five CRC patients based on haematological findings and are sorted into stages using TNM and Dukes. The present study is to find the association between haematological findings, demographic characters, differentiation position, lymph node invasion and tumour node metastasis in CRC patients in accordance with their age.

RESULTS

We observed significant (p < 0.05) nexus between lymph node metastasis and tumour node metastasis on the basis of tumour's differentiation demographic positioning and age of the individuals.

CONCLUSION

Earlier location tracing and medicinal treatment or surgery lessen the chance of CRC morbidity and mortality along with prolonging survival rate via prognostic factors and disease position determination.

Prediction of Survival and Recurrence Patterns by Machine Learning in Gastric Cancer Cases Undergoing Radiation Therapy and Chemotherapy

Purpose

Radical surgery is the most important treatment modality in gastric cancer. Preoperative or postoperative radiation therapy (RT) and perioperative chemotherapy are the treatment options that should be added to surgery. This study aimed to evaluate the overall survival (OS) and recurrence patterns by machine learning in gastric cancer cases undergoing RT.

Methods and Materials

Between 2012 and 2019, the OS and recurrence patterns of 75 gastric cancer cases receiving RT ± chemotherapy at the Department of Radiation Oncology were evaluated by machine learning. Logistic regression, multilayer perceptron, XGBoost, support vector classification, random forest, and Gaussian Naive Bayes (GNB) algorithms were used to predict OS, hematogenous distant metastases, and peritoneal metastases. After the correlation analysis, the backward feature selection was performed as the variable selection method, and the variables with P values less than .005 were selected.

Results

Over the median 23-month follow-up, recurrence was seen in 33 cases, and 36 patients died. The median OS was 23 (min: 7; max: 82) months, and the disease-free survival was 18 (min: 5, max: 80) months. The most common recurrence pattern was hematogenous distant metastasis, followed by peritoneal metastasis. In this study, the most successful algorithms in the prediction of OS, distant metastases, and peritoneal metastases were found to be GNB with an accuracy of 81% (95% confidence interval [CI], 0.65-0.97, area under the curve [AUC]: 0.89), XGBoost with 86% accuracy (95% CI, 0.74-0.97, AUC: 0.86), and random forest with 97% accuracy (95% CI, 0.92-1.00, AUC: 0.97), respectively.

Conclusions

In gastric cancer, GNB, XGBoost, and random forest algorithms were determined to be the most successful algorithms for predicting OS, distant metastases, and peritoneal metastases, respectively. To determine the most accurate algorithm and perhaps make personalized treatments applicable, more precise machine learning studies are needed with an increased number of cases in the coming years.

The pretreatment thrombocytosis as one of prognostic factors for gastric cancer: A systematic review and meta-analysis

BACKGROUND & AIMS

At present, increasing reports have shown that pretreatment platelet count was associated with the prognosis of many types of cancer. We performed rounded analysis to comprehensively analyze and evaluate the prognostic significance of pretreatment thrombocytosis for patients with gastric cancer.

METHODS

We identified relevant studies by searching database including PubMed, Embase, Cochrane Library and Web of Science. The relative risk (RR) with its 95% confidence interval (CI) was used to assess the correlation between thrombocytosis and overall survival (OS) of gastric cancer patients. We also conducted subgroup analysis and sensitivity analysis for the prognostic effect of thrombocytosis on OS. The analysis was performed and assessed using Review Manager 5.2.

RESULTS

A total of nine studies including 7158 participants were included in this systematic review. Analysis results showed that pretreatment thrombocytosis had a close relationship with 1, 3 and 5 years survival of gastric cancer, with the pooled RRs being 0.80 (95% CI 0.71-0.90; P = 0.0004), 0.65 (95% CI 0.45-0.92; P = 0.02) and 0.64 (95% CI 0.47-0.87; P = 0.004) respectively.

CONCLUSIONS

The present rounded analysis suggests that pretreatment thrombocytosis may have significant association with poor survival of patients with gastric cancer.

Elevated red blood cell distribution width contributes to poor prognosis in patients undergoing resection for nonmetastatic rectal cancer

Several studies have reported that elevated red blood cell distribution width (RDW) was associated with the poor prognosis of different kinds of cancers. The aim of this study was to investigate the prognostic role of RDW in patients undergoing resection for nonmetastatic rectal cancer.We retrospectively reviewed a database of 625 consecutive patients who underwent curative resection for nonmetastatic rectal cancer at our institution from January 2009 to December 2014. The cutoff value of RDW was calculated by receiver-operating characteristic curve.The results demonstrated that patients in high RDW-cv group had a lower overall survival (OS) (P = .018) and disease-free survival (P = .004). We also observed that patients in high RDW-sd group were associated with significantly lower OS (P = .033), whereas the disease-free survival (DFS) was not significantly different (P = .179).In multivariate analysis, we found elevated RDW-cv was associated poor DFS (hazard ratio [HR] = 1.56, P = .010) and RDW-sd can predict a worse OS (HR = 1.70, P = .009).We confirmed that elevated RDW can be an independently prognostic factor in patients undergoing resection for nonmetastatic rectal cancer. So more intervention or surveillance might be paid to the patients with nonmetastatic rectal cancer and elevated RDW values in the future.

Is whole-colon investigation by colonoscopy, computerised tomography colonography or barium enema necessary for all patients with colorectal cancer symptoms, and for which patients would flexible sigmoidoscopy suffice? A retrospective cohort study

BACKGROUND

For patients referred to hospital with suspected colorectal cancer (CRC), it is current standard clinical practice to conduct an examination of the whole colon and rectum. However, studies have shown that an examination of the distal colorectum using flexible sigmoidoscopy (FS) can be a safe and clinically effective investigation for some patients. These findings require validation in a multicentre study.

OBJECTIVES

To investigate the links between patient symptoms at presentation and CRC risk by subsite, and to provide evidence of whether or not FS is an effective alternative to whole-colon investigation (WCI) in patients whose symptoms do not suggest proximal or obstructive disease.

DESIGN

A multicentre retrospective study using data collected prospectively from two randomised controlled trials. Additional data were collected from trial diagnostic procedure reports and hospital records. CRC diagnoses within 3 years of referral were sourced from hospital records and national cancer registries via the Health and Social Care Information Centre.

SETTING

Participants were recruited to the two randomised controlled trials from 21 NHS hospitals in England between 2004 and 2007.

PARTICIPANTS

Men and women aged ≥ 55 years referred to secondary care for the investigation of symptoms suggestive of CRC.

MAIN OUTCOME MEASURE

Diagnostic yield of CRC at distal (to the splenic flexure) and proximal subsites by symptoms/clinical signs at presentation.

RESULTS

The data set for analysis comprised 7380 patients, of whom 59% were women (median age 69 years, interquartile range 62-76 years). Change in bowel habit (CIBH) was the most frequently presenting symptom (73%), followed by rectal bleeding (38%) and abdominal pain (29%); 26% of patients had anaemia. CRC was diagnosed in 551 patients (7.5%): 424 (77%) patients with distal CRC, 122 (22%) patients with cancer proximal to the descending colon and five patients with both proximal and distal CRC. Proximal cancer was diagnosed in 96 out of 2021 (4.8%) patients with anaemia and/or an abdominal mass. The yield of proximal cancer in patients without anaemia or an abdominal mass who presented with rectal bleeding with or without a CIBH or with a CIBH to looser and/or more frequent stools as a single symptom was low (0.5%). These low-risk groups for proximal cancer accounted for 41% (3032/7380) of the cohort; only three proximal cancers were diagnosed in 814 low-risk patients examined by FS (diagnostic yield 0.4%).

LIMITATIONS

A limitation to this study is that changes to practice since the trial ended, such as new referral guidelines and improvements in endoscopy quality, potentially weaken the generalisability of our findings.

CONCLUSIONS

Symptom profiles can be used to determine whether or not WCI is necessary. Most proximal cancers were diagnosed in patients who presented with anaemia and/or an abdominal mass. In patients without anaemia or an abdominal mass, proximal cancer diagnoses were rare in those with rectal bleeding with or without a CIBH or with a CIBH to looser and/or more frequent stools as a single symptom. FS alone should be a safe and clinically effective investigation in these patients. A cost-effectiveness analysis of symptom-based tailoring of diagnostic investigations for CRC is recommended.

TRIAL REGISTRATION

Current Controlled Trials ISRCTN95152621.

FUNDING

This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 21, No. 66. See the NIHR Journals Library website for further project information.

The elevated NLR, PLR and PLT may predict the prognosis of patients with colorectal cancer: a systematic review and meta-analysis

Recently, several studies have reported that inflammatory response and elevated platelet counts may be associated with the poor prognosis of colorectal cancer. This meta-analysis was designed to analyze and evaluate the prognostic role of elevated preoperative or pretreatment neutrophils-to-lymphocytes ratio, platelet-to-lymphocytes ratio or platelet counts in patients with colorectal cancer. We searched PubMed, EMBASE, Cochrane Library and Web of Science to April, 2016. A total of 23 studies (N = 11762 participants) were included for this meta-analysis. Elevated neutrophils-to-lymphocytes ratio have a close relationship with the poor Overall Survival of colorectal cancer with the pooled HR being 1.92 [95% CI 1.57–2.34; P < 0.00001]. This meta-analysis indicated that elevated neutrophils-to-lymphocytes ratio, platelet-to-lymphocytes ratio or platelet counts may be a cost-effective and noninvasive serum biomarker for poor prognosis for patients with colorectal cancer.

Prognostic relevance of hematological profile before resection for colorectal liver metastases

BACKGROUND

Although alterations of hematological profile and especially elevated platelet counts were reported to influence survival in primary colorectal cancer, its prognostic relevance before the surgical treatment of colorectal liver metastases (CLM) is mainly unclear. Therefore, the aim of this study was to analyze the impact of these factors on overall survival following liver resection of CLM.

MATERIALS AND METHODS

The surgical treatment of primary CLM between 1994 and 2012 in 983 patients was retrospectively analyzed using univariable and multivariable Cox regression models.

RESULTS

In the multivariable analyses, a preoperative anemia was independently associated with inferior overall outcome (P = 0.005, hazard ratio: 1.355). However, with only 2.7% of all cases, an elevation of preoperative platelets was not a frequent finding and no independent impact on survival (P = 0.834). Furthermore, abnormal hemoglobin and platelet values had no impact on rate of surgical revisions due to bleeding complications (P = 0.962 and P = 0.671, respectively), but a potential interaction between abnormal hemoglobin and platelet values and the amount of transfused packed red blood cells (P = 0.004 and P < 0.001, respectively) was observed.

CONCLUSIONS

Preoperative anemia is statistically significantly associated with inferior overall survival following resection of CLM and might define a new prognostic marker. Preoperative elevated platelets were not a frequent finding and showed no influence on overall survival.

Five-year outcomes of preoperative chemoradiation for rectal carcinoma in Saudi population: single-institutional experience

OBJECTIVES

Preoperative chemoradiation (CRT) followed by surgery is the standard treatment for locally advanced rectal cancer (LARC). The outcomes of preoperative CRT in Saudi patients with LARC have not been widely studied. The study reports long-term outcomes after preoperative CRT followed by curative surgery in Saudi patients with LARC.

DESIGN AND SETTINGS

A retrospective, single-institutional study performed in the tertiary care oncology center in Saudi Arabia.

MATERIALS AND METHODS

A total of 154 out of 204 patients with LARC were treated with preoperative CRT and followed by surgery at the oncology center between September 2005 and November 2012. Data regarding the response rates, toxicity profile, locoregional control (LRC), distant metastasis control (DMC), overall survival (OS), and disease-free survival (DFS) rates were analyzed.

RESULTS

The median age of the study population was 56.6 years (range: 26-89). Predominant clinical stages were IIA (70 patients; 45.4%) and IIIB (49 patients; 31.8%). Majority of patients (79.8%) underwent a complete total mesorectal excision (TME). Complete pathological response (ypT0N0) was seen in 26 patients (16.8%). At 5 years, locoregional recurrence (LR) was reported in 12 patients (7.8%), and distant metastases were noted in 33 patients (21.4%). The 5-year cumulative LRC, DMC, OS, and DFS rates were 91%, 71.3%, 78%, and 64.8%, respectively. Stage, nodal status, circumferential margins, ypT0N0, and adjuvant chemotherapy were found to be important prognostic factors for DFS.

CONCLUSION

The results of preoperative CRT followed by surgery and adjuvant chemotherapy in Saudi population are comparable with international data.