Preoperative Predictors of Lymph Node Metastasis in Colon Cancer

Association of clinicopathological factor with lymph node metastasis in rectal cancer patients: a retrospective cohort study

INTRODUCTION

Systemic inflammatory response (SIR) indicators serve as predictive factors for lymph node metastasis (LNM) in various cancers. This study aimed to investigate the association of platelet-to-lymphocyte ratio (PLR) and neutrophil-to-lymphocyte ratio (NLR) with LNM in rectal cancer and to identify clinicopathological factors linked to LNM.

METHODS

We retrospectively analyzed 181 rectal cancer patients who underwent surgical resection. Preoperative NLR and PLR were calculated from blood samples, with optimal cutoff values determined by receiver operating characteristic (ROC) analysis. Associations between NLR/PLR and clinicopathological features were evaluated, risk factors for LNM were analyzed via univariate and multivariate logistic regression.

RESULTS

No significant differences were observed between the high NLR (H-NLR) and low NLR (L-NLR) groups in terms of clinicopathological characteristics, including TNM stage, perineural invasion (PNI), lymphovascular invasion (LVI), or serum levels of CEA and CA19-9 respectively (p > 0.05).In contrast, the high PLR (H-PLR) group showed significantly higher prevalence of several adverse pathological features: The H-PLR group had a higher positive PNI (54.2% vs.25.0%,p = 0.04), greater positive LVI(51.6% vs.28.6%,p = 0.025),and more positive TDs (14.4% vs.0,p = 0.028), increased lymph node metastasis (52.9% vs.17.9%,p < 0.001), more elevated CEA (43.1% vs.14.3%,p = 0.005) and more advanced tumor stage (stage II + stage III,81% vs.67.9%,p = 0.003).Univariate analysis identified several factors significantly associated with LNM: T stage (OR = 3.156, 95%CI:1.580-6.303),positive PNI (OR = 6.182,95%CI:3.242-11.787),positive LVI (OR = 10.271,95%CI:5.177-20.375),H-PLR(OR = 5.175,95%CI:1.870-14.321),positive TDs (OR = 3.390,95%CI:1.261-9.117),TLN(OR = 1.053,95%CI:1.005-1.103),elevated CEA(OR = 3.313,95%CI:1.655-5.920) and elevated CA199 (OR = 2.248,95%CI:1.012-4.992) were correlated with LNM using univariate analysis, but only positive LVI(adjusted OR = 6.203,95%CI:2.892-13.303,p < 0.001) and positive PNI (adjusted OR = 3.086,95%CI:1.341-7.102,p = 0.008) were the independent risk factors for LNM using multivariate analysis.

CONCLUSION

H-PLR but not H-NLR may be associated with LNM, positive LVI and PNI were independent risk factors for LNM in RC.

Predicting the risk of lymph node metastasis in colon cancer: development and validation of an online dynamic nomogram based on multiple preoperative data

BACKGROUND

Predicting lymph node metastasis (LNM) in colon cancer (CC) is crucial to treatment decision-making and prognosis. This study aimed to develop and validate a nomogram that estimates the risk of LNM in patients with CC using multiple clinical data from patients before surgery.

METHODS

Clinicopathological data were collected from 412 CC patients who underwent Radical resection of CC. The training cohort consisted of 300 cases, and the external validation cohort consisted of 112 cases. The LASSO and multivariate logistic regression were used to select the predictors and construct the nomogram. The discrimination and calibration of the nomogram were evaluated by the ROC curve and calibration curve, respectively. The clinical application of the nomogram was assessed by decision curve analysis(DCA) and clinical impact curves(CIC).

RESULTS

Eight independent factors associated with LNM were identified by multivariate logistic analysis: LN status on CT, tumor diameter on CT, differentiation, ulcer, intestinal obstruction, anemia, blood type, and neutrophil percentage. The online dynamic nomogram model constructed by independent factors has good discrimination and consistency. The AUC of 0.834(95% CI: 0.755-0.855) in the training cohort, 0.872(95%CI: 0.807-0.937) in the external validation cohort, and Internal validation showed that the corrected C statistic was 0.810. The calibration curve of both the training set and the external validation set indicated that the predicted outcome of the nomogram was highly consistent with the actual outcome. The DCA and CIC indicate that the model has clinical practical value.

CONCLUSION

Based on various simple parameters collected preoperatively, the online dynamic nomogram can accurately predict LNM risk in CC patients. The high discriminative ability and significant improvement of NRI and IDI indicate that the model has potential clinical application value.

Prediction of lymph node metastasis in stage I-III colon cancer patients younger than 40 years

AIMS

Developing a clinical model to predict the individual risk of lymph node metastasis (LNM) in young colon cancer (CC) patients may address an unmet clinical need.

METHODS

A total of 2,360 CC patients under 40 years old were extracted from the SEER database and randomly divided into development and validation cohorts. Risk factors for LNM were identified by using a logistic regression model. A weighted scoring system was built according to beta coefficients (β) calculated by a logistic regression model. Model discrimination was evaluated by C-statistics, model calibration was evaluated by H-L test and calibration plot.

RESULTS

Risk factors were identified as T stage, tumor site, grade and histology. The area under the receiver operating characteristic curve (AUC-ROC) was 0.66 in both cohorts, indicating acceptable discriminatory power. The H-L test showed good calibration in the development cohort (χ2=2.869, P=0.837) and validation cohort (χ2=10.103, P=0.120) which also had been proved by calibration plot. Patients with total risk score of 0-1, 2-3 and 4-6 were considered as low, medium and high risk group.

CONCLUSION

This clinical risk prediction model is accurate enough to identify young CC patients with high risk of LNM and can further provide individualized clinical reference.

F-18 FDG PET/CT based Preoperative Machine Learning Prediction Models for Evaluating Regional Lymph Node Metastasis Status of Patients with Colon Cancer

OBJECTIVE

This study aimed to develop a simple machine-learning model incorporating lymph node metastasis status with F-18 Fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) and clinical information for predicting regional lymph node metastasis in patients with colon cancer.

METHODS

This retrospective study included 193 patients diagnosed with colon cancer between January 2014 and December 2017. All patients underwent F-18 FDG PET/CT and blood test before surgery. One categorical variable (lymph node FDG uptake [LNFDG]) and six continuous variables (age, neutrophil-to-lymphocyte ratio [NLR], carcinoembryonic antigen [CEA], carbohydrate antigen 19-9 [CA19-9], C-reactive protein, and maximal standardized uptake value (SUVmax) of the primary tumor) were used as input variables. Four supervised machine learning methods were used to build predictive models: logistic regression (LR), random forest (RF), gradient boosting machine (GBM), and support vector machine (SVM). Area under the receiver operating characteristic curve (AUC) of the validation set were used for evaluating and comparing model performance.

RESULTS

The number of patients with lymph node metastasis were 63 (33%). The mean number of harvested lymph nodes was 28.8 ± 11.4. The mean CEA, CA19-9, and CRP levels were 4.8 ± 9.3 ng/ml, 15.6 ± 42.8 U/ml, and 1.0 ± 3.0 mg/dl, respectively. The mean NLR was 2.2 ± 1.2. The mean SUVmax levels of the primary tumor were 15.2 ± 7.9. Fifty-one (26%) patients showed FDG uptake in the pericolic lymph nodes.  The mean AUC of the LR, RF, GBM, and SVM methods for the LNFDG model was 0.7046, 0.7047, 0.7040, and 0.7058, respectively. The mean AUC of the LR, RF, GBM, and SVM methods for the LNFDG plus clinical information model was 0.7046, 0.7302, 0.7444, and 0.7097, respectively.

CONCLUSION

Machine learning methods using LNFDG and clinical information could predict the lymph node metastasis status in patients with colon cancer with higher accuracy than a model using only FDG uptake of the lymph nodes.

Predicting lymph node metastasis in colorectal cancer patients: development and validation of a column chart model

Lymph node metastasis (LNM) is one of the crucial factors in determining the optimal treatment approach for colorectal cancer. The objective of this study was to establish and validate a column chart for predicting LNM in colon cancer patients. We extracted a total of 83,430 cases of colon cancer from the Surveillance, Epidemiology, and End Results (SEER) database, spanning the years 2010-2017. These cases were divided into a training group and a testing group in a 7:3 ratio. An additional 8545 patients from the years 2018-2019 were used for external validation. Univariate and multivariate logistic regression models were employed in the training set to identify predictive factors. Models were developed using logistic regression, LASSO regression, ridge regression, and elastic net regression algorithms. Model performance was quantified by calculating the area under the ROC curve (AUC) and its corresponding 95% confidence interval. The results demonstrated that tumor location, grade, age, tumor size, T stage, race, and CEA were independent predictors of LNM in CRC patients. The logistic regression model yielded an AUC of 0.708 (0.7038-0.7122), outperforming ridge regression and achieving similar AUC values as LASSO regression and elastic net regression. Based on the logistic regression algorithm, we constructed a column chart for predicting LNM in CRC patients. Further subgroup analysis based on gender, age, and grade indicated that the logistic prediction model exhibited good adaptability across all subgroups. Our column chart displayed excellent predictive capability and serves as a useful tool for clinicians in predicting LNM in colorectal cancer patients.

Histogram analysis based on intravoxel incoherent motion diffusion-weighted imaging for determining the perineural invasion status of rectal cancer

Background

Unfortunately, the morphologic magnetic resonance imaging (MRI) is unable to determine perineural invasion (PNI) status. This study applied histogram analysis of intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) in the assessment of PNI status of rectal cancer (RC).

Methods

The retrospective analysis enrolled 175 patients with RC confirmed by postoperative pathology in The First Affiliated Hospital of Shandong First Medical University from January 2019 to December 2021. All patients underwent preoperative rectal MRI. Whole-tumor volume histogram features from IVIM-DWI were extracted using open-source software. Univariate analysis and multivariate logistic regression analysis were used to compare the differences in histogram parameters and clinical features between the PNI-positive group and PNI-negative group. Receiver operating characteristic curve analysis was used to evaluate the diagnostic performance, while the Delong test was used to compare the area under the curve of the models.

Results

The interobserver agreement of the histogram features derived from DWI, including apparent diffusion coefficient (ADC), true diffusion coefficient (D), pseudo-diffusion coefficient (D*), perfusion fraction (f), water molecular diffusion heterogeneity index (α), and distributed diffusion coefficient (DDC) were good to excellent. A total of eight histogram features including DWI_maximum, DWI_skewness, D_kurtosis, D_minimum, D_skewness, D*_energy, D*_skewness, and f_minimum were significantly different between the PNI-positive and PNI-negative groups in the univariate analysis (P<0.05); among the clinicoradiologic factors, percentage of rectal wall circumference invasion (PCI) was significantly different between the two groups (P<0.05). Multivariate analysis demonstrated that the values of D*_energy, D*_skewness, and f_minimum differed significantly between the PNI-positive patients and PNI-negative patients (P<0.05), with the independent risk factors being D*_skewness [odds ratio (OR) =1.157; 95% confidence interval (CI): 1.050-1.276; P=0.003] and PCI (OR =11.108, 95% CI: 1.767-69.838; P=0.0002). The area under the curve of the model combining the three histogram features and PCI to assess PNI status in RC was 0.807 (95% CI: 0.741-0.863). The results of the Delong test showed that the combined model was significantly different from each single-parameter model (P<0.05).

Conclusions

The combined model constructed on the basis of IVIM-DWI histogram features may help to assess the status of RC PNI.

A prediction nomogram for perineural invasion in colorectal cancer patients: a retrospective study

BACKGROUND

Perineural invasion (PNI), as the fifth recognized pathway for the spread and metastasis of colorectal cancer (CRC), has increasingly garnered widespread attention. The preoperative identification of whether colorectal cancer (CRC) patients exhibit PNI can assist clinical practitioners in enhancing preoperative decision-making, including determining the necessity of neoadjuvant therapy and the appropriateness of surgical resection. The primary objective of this study is to construct and validate a preoperative predictive model for assessing the risk of perineural invasion (PNI) in patients diagnosed with colorectal cancer (CRC).

MATERIALS AND METHODS

A total of 335 patients diagnosed with colorectal cancer (CRC) at a single medical center were subject to random allocation, with 221 individuals assigned to a training dataset and 114 to a validation dataset, maintaining a ratio of 2:1. Comprehensive preoperative clinical and pathological data were meticulously gathered for analysis. Initial exploration involved conducting univariate logistic regression analysis, with subsequent inclusion of variables demonstrating a significance level of p < 0.05 into the multivariate logistic regression analysis, aiming to ascertain independent predictive factors, all while maintaining a p-value threshold of less than 0.05. From the culmination of these factors, a nomogram was meticulously devised. Rigorous evaluation of this nomogram's precision and reliability encompassed Receiver Operating Characteristic (ROC) curve analysis, calibration curve assessment, and Decision Curve Analysis (DCA). The robustness and accuracy were further fortified through application of the bootstrap method, which entailed 1000 independent dataset samplings to perform discrimination and calibration procedures.

RESULTS

The results of multivariate logistic regression analysis unveiled independent risk factors for perineural invasion (PNI) in patients diagnosed with colorectal cancer (CRC). These factors included tumor histological differentiation (grade) (OR = 0.15, 95% CI = 0.03-0.74, p = 0.02), primary tumor location (OR = 2.49, 95% CI = 1.21-5.12, p = 0.013), gross tumor type (OR = 0.42, 95% CI = 0.22-0.81, p = 0.01), N staging in CT (OR = 3.44, 95% CI = 1.74-6.80, p < 0.001), carcinoembryonic antigen (CEA) level (OR = 3.13, 95% CI = 1.60-6.13, p = 0.001), and platelet-to-lymphocyte ratio (PLR) (OR = 2.07, 95% CI = 1.08-3.96, p = 0.028).These findings formed the basis for constructing a predictive nomogram, which exhibited an impressive area under the receiver operating characteristic (ROC) curve (AUC) of 0.772 (95% CI, 0.712-0.833). The Hosmer-Lemeshow test confirmed the model's excellent fit (p = 0.47), and the calibration curve demonstrated consistent performance. Furthermore, decision curve analysis (DCA) underscored a substantial net benefit across the risk range of 13% to 85%, reaffirming the nomogram's reliability through rigorous internal validation.

CONCLUSION

We have formulated a highly reliable nomogram that provides valuable assistance to clinical practitioners in preoperatively assessing the likelihood of perineural invasion (PNI) among colorectal cancer (CRC) patients. This tool holds significant potential in offering guidance for treatment strategy formulation.

D-dimer to lymphocyte ratio can serve as a potential predictive and prognostic value in colorectal cancer patients with liver metastases

BACKGROUND

The intent of this research was to generate and investigate the D-dimer to lymphocyte ratio (DLR) capacity to forecast the risk and prognosis of colorectal cancer liver metastases (CRCLM).

METHODS

From January 2010 to December 2019, 177 clinicopathologically confirmed colorectal cancer (CRC) patients (89 in the control group and 88 in the experimental group) were identified at the Affiliated Cancer Hospital of Guangxi Medical University. Multivariate Cox regression analysis was used to screen independent predictive diagnostic and prognostic factors of liver metastasis in CRC, and receiver operating characteristic (ROC) curves and Kaplan‒Meier (K‒M) curves were established to analyze the diagnostic and predictive prognostic efficacy of the DLR in the development of CRCLM.

RESULTS

Patients with CRCLM had higher DLR levels and D-dimer levels in their blood, with statistically significant differences (p < 0.001). DLR might be employed as a predictor for the development of CRCLM, according to ROC curve research (sensitivity 0.670, specificity 0.775, area under the curve 0.765). D-dimer, lymphocyte count CEA, CA125, and CA199 were not linked to prognosis in patients with CRCLM in Cox regression analysis of dichotomous variables. In contrast, DLR level was a possible risk factor for the prognosis of patients with CRCLM (HR = 2.108, p = 0.047), and age, T stage, and DLR level (DLR < 0.4) were connected with the prognosis of patients with CRCLM (p < 0.05).

CONCLUSION

DLR serves as a risk indicator for the development of CRCLM.

A novel potential inflammation-nutrition biomarker for predicting lymph node metastasis in clinically node-negative colon cancer

Background

The purpose of this study is to investigate the predictive significance of (platelet × albumin)/lymphocyte ratio (PALR) for lymph node metastasis (LNM) in patients with clinically node-negative colon cancer (cN0 CC).

Methods

Data from 800 patients with primary CC who underwent radical surgery between March 2016 and June 2021 were reviewed. The non-linear relationship between PALR and the risk of LNM was explored using a restricted cubic spline (RCS) function while a receiver operating characteristic (ROC) curve was developed to determine the predictive value of PALR. Patients were categorized into high- and low-PALR cohorts according to the optimum cut-off values derived from Youden's index. Univariate and multivariate logistic regression analyses were used to identify the independent indicators of LNM. Sensitivity analysis was performed to repeat the main analyses with the quartile of PALR.

Results

A total of eligible 269 patients with primary cN0 CC were retrospectively selected. The value of the area under the ROC curve for PALR for predicting LNM was 0.607. RCS visualized the uptrend linear relationship between PALR and the risk of LNM (p-value for non-linearity > 0.05). PALR (odds ratio = 2.118, 95% confidence interval, 1.182-3.786, p = 0.011) was identified as an independent predictor of LNM in patients with cN0 CC. A nomogram incorporating PALR and other independent predictors was constructed with an internally validated concordance index of 0.637. The results of calibration plots and decision curve analysis supported a good performance ability and the sensitivity analysis further confirmed the robustness of our findings.

Conclusion

PALR has promising clinical applications for predicting LNM in patients with cN0 CC.