A Novel Nomogram Based on Machine Learning-Pathomics Signature and Neutrophil to Lymphocyte Ratio for Survival Prediction of Bladder Cancer Patients

Artificial Intelligence in Digital Pathology for Bladder Cancer: Hype or Hope? A Systematic Review

Bladder cancer (BC) diagnosis and prediction of prognosis are hindered by subjective pathological evaluation, which may cause misdiagnosis and under-/over-treatment. Computational pathology (CPATH) can identify clinical outcome predictors, offering an objective approach to improve prognosis. However, a systematic review of CPATH in BC literature is lacking. Therefore, we present a comprehensive overview of studies that used CPATH in BC, analyzing 33 out of 2285 identified studies. Most studies analyzed regions of interest to distinguish normal versus tumor tissue and identify tumor grade/stage and tissue types (e.g., urothelium, stroma, and muscle). The cell's nuclear area, shape irregularity, and roundness were the most promising markers to predict recurrence and survival based on selected regions of interest, with >80% accuracy. CPATH identified molecular subtypes by detecting features, e.g., papillary structures, hyperchromatic, and pleomorphic nuclei. Combining clinicopathological and image-derived features improved recurrence and survival prediction. However, due to the lack of outcome interpretability and independent test datasets, robustness and clinical applicability could not be ensured. The current literature demonstrates that CPATH holds the potential to improve BC diagnosis and prediction of prognosis. However, more robust, interpretable, accurate models and larger datasets-representative of clinical scenarios-are needed to address artificial intelligence's reliability, robustness, and black box challenge.

Prognostic factors and nomogram for the overall survival of bladder cancer bone metastasis: A SEER-based study

Bone metastasis has a poor prognosis in patients with bladder cancer (BC). This study aimed to construct a prognostic nomogram for predicting the overall survival of patients with bone-metastatic BC (BMBC). The Surveillance, Epidemiology, and End Results database was used to recruit patients with BMBC between 2010 and 2018. Univariate and multivariate analyses were performed to screen for prognostic factors and construct a nomogram. Harrell concordance index, receiver operating characteristic curve, and calibration curve were used to verify the prognostic nomograms. All statistical analyses and chart formation were performed using SPSS 23.0 and R software 4.1.2. A total of 1361 patients diagnosed with BMBC were identified in the Surveillance, Epidemiology, and End Results database. Six independent prognostic factors, including marital status, histological type, T stage, other metastases, surgery, and chemotherapy, were identified and included in the nomogram construction. Among them, chemotherapy contributed the most to the prognosis in the nomogram. The concordance index of the nomogram was 0.745 and 0.753 in the training and validation groups, respectively, and all values of the area under the curve were >0.77. The calibration curves showed perfect consistency between the observed and predicted survival rates. The prognostic nomogram developed in this study is expected to become an accurate and individualized tool for predicting overall survival in patients with BMBC and providing guidance for appropriate treatment or care.

A Predictive Model for the 10-year Overall Survival Status of Patients With Distant Metastases From Differentiated Thyroid Cancer Using XGBoost Algorithm-A Population-Based Analysis

Purpose: To explore clinical and non-clinical characteristics affecting the prognosis of patients with differentiated thyroid cancer with distant metastasis (DTCDM) and establish an accurate overall survival (OS) prognostic model.

Patients and methods: Study subjects and related information were obtained from the National Cancer Institute’s surveillance, epidemiology, and results database (SEER). Kaplan‐Meier analysis, log-rank test, and univariate and multivariate Cox analysis were used to screen for factors influencing the OS of patients with DTCDM. Nine variables were introduced to build a machine learning (ML) model, receiver operating characteristic (ROC) was used to evaluate the recognition ability of the model, calibration plots were used to obtain prediction accuracy, and decision curve analysis (DCA) was used to estimate clinical benefit.

Results: After applying the inclusion and exclusion criteria, a total of 3,060 patients with DTCDM were included in the survival analysis from 2004 to 2017. A machine learning prediction model was developed with nine variables: age at diagnosis, gender, race, tumor size, histology, regional lymph node metastasis, primary site surgery, radiotherapy, and chemotherapy. After excluding patients who survived <120 months, variables were sub-coded and machine learning was used to model OS prognosis in patients with DTCDM. Patients 6–50 years of age had the highest scores in the model. Other variables with high scores included small tumor size, male sex, and age 51–76. The AUC and calibration curves confirm that the XGBoost model has good performance. DCA shows that our model can be used to support clinical decision-making in a 10-years overall survival model.

Conclusion: An artificial intelligence model was constructed using the XGBoost algorithms to predict the 10-years overall survival rate of patients with DTCDM. After model validation and evaluation, the model had good discriminative ability and high clinical value. This model could serve as a clinical tool to help inform treatment decisions for patients with DTCDM.

The feasibility of proteomics sequencing based immune-related prognostic signature for predicting clinical outcomes of bladder cancer patients

Background

Bladder cancer (BCa) shows its potential immunogenity in current immune-checkpoint inhibitor related immunotherapies. However, its therapeutic effects are improvable and could be affected by tumor immune microenvironment. Hence it is interesting to find some more prognostic indicators for BCa patients concerning immunotherapies.

Methods

In the present study, we retrospect 129 muscle-invasive BCa (MIBC) patients with radical cystectomy in Shanghai General Hospital during 2007 to 2018. Based on the results of proteomics sequencing from 9 pairs of MIBC tissue from Shanghai General Hospital, we focused on 13 immune-related differential expression proteins and their related genes. An immune-related prognostic signature (IRPS) was constructed according to Cancer Genome Atlas (TCGA) dataset. The IRPS was verified in ArrayExpress (E-MTAB-4321) cohort and Shanghai General Hospital (General) cohort, separately. A total of 1010 BCa patients were involved in the study, including 405 BCa patients in TCGA cohort, 476 BCa patients in E-MTAB-4321 cohort and 129 MIBC patients in General cohort.

Result

It can be indicated that high IRPS score was related to poor 5-year overall survival and disease-free survival. The IRPS score was also evaluated its immune infiltration. We found that the IRPS score was adversely associated with GZMB, IFN-γ, PD-1, PD-L1. Additionally, higher IRPS score was significantly associated with more M2 macrophage and resting mast cell infiltration.

Conclusion

The study revealed a novel BCa prognostic signature based on IRPS score, which may be useful for BCa immunotherapies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-022-09783-y.

Development of Nomograms for Predicting Prognosis of Pancreatic Cancer after Pancreatectomy: A Multicenter Study

Surgical resection is the only curative treatment for pancreatic ductal adenocarcinoma (PDAC). Currently, the TNM classification system is considered the standard for predicting prognosis after surgery. However, the prognostic accuracy of the system remains limited. This study aimed to develop new predictive nomograms for resected PDAC. The clinicopathological data of patients who underwent surgery for PDAC between 2006 and 2015 at five major institutions were retrospectively reviewed; 885 patients were included in the analysis. Cox regression analysis was performed to investigate prognostic factors for recurrence and survival, and statistically significant factors were used for creating nomograms. The nomogram for predicting recurrence-free survival included nine factors: sarcopenic obesity, elevated carbohydrate antigen 19–9, platelet-to-lymphocyte ratio, preoperatively-identified arterial abutment, estimated blood loss (EBL), tumor differentiation, size, lymph node ratio, and tumor necrosis. The nomogram for predicting overall survival included 10 variables: age, underlying liver disease, chronic kidney disease, preoperatively found portal vein invasion, portal vein resection, EBL, tumor differentiation, size, lymph node metastasis, and tumor necrosis. The time-dependent area under the receiver operating characteristic curve for both nomograms exceeded 0.70. Nomograms were developed for predicting survival after resection of PDAC, and the platforms showed fair predictive performance. These new comprehensive nomograms provide information on disease status and are useful for determining further treatment for PDAC patients.

A Novel Machine Learning Algorithm Combined With Multivariate Analysis for the Prognosis of Renal Collecting Duct Carcinoma

Objectives

To investigate the clinical and non-clinical characteristics that may affect the prognosis of patients with renal collecting duct carcinoma (CDC) and to develop an accurate prognostic model for this disease.

Methods

The characteristics of 215 CDC patients were obtained from the U.S. National Cancer Institute’s surveillance, epidemiology and end results database from 2004 to 2016. Univariate Cox proportional hazard model and Kaplan-Meier analysis were used to compare the impact of different factors on overall survival (OS). 10 variables were included to establish a machine learning (ML) model. Model performance was evaluated by the receiver operating characteristic curves (ROC) and calibration plots for predictive accuracy and decision curve analysis (DCA) were obtained to estimate its clinical benefits.

Results

The median follow-up and survival time was 16 months during which 164 (76.3%) patients died. 4.2, 32.1, 50.7 and 13.0% of patients were histological grade I, II, III, and IV, respectively. At diagnosis up to 61.9% of patients presented with a pT3 stage or higher tumor, and 36.7% of CDC patients had metastatic disease. 10 most clinical and non-clinical factors including M stage, tumor size, T stage, histological grade, N stage, radiotherapy, chemotherapy, age at diagnosis, surgery and the geographical region where the care delivered was either purchased or referred and these were allocated 95, 82, 78, 72, 49, 38, 36, 35, 28 and 21 points, respectively. The points were calculated by the XGBoost according to their importance. The XGBoost models showed the best predictive performance compared with other algorithms. DCA showed our models could be used to support clinical decisions in 1-3-year OS models.

Conclusions

Our ML models had the highest predictive accuracy and net benefits, which may potentially help clinicians to make clinical decisions and follow-up strategies for patients with CDC. Larger studies are needed to better understand this aggressive tumor.

Can Routine Blood and Urine Parameters Reveal Clues to Detect Bladder Cancer? A Case–Control Study

Objective

Limited attention has been paid to abnormal blood and urine test results for patients with bladder cancer. The present study aimed to identify whether blood and urine parameters are associated with bladder cancer.

Methods

We used a case–control design and matched each patient with bladder cancer with three healthy controls of the same age and sex. Univariate conditional logistic regression was used to calculate the crude and adjusted odds ratio (OR) and its 95% CI. Multivariate conditional logistic regression was performed for confounders adjustment, and Spearman’s correlation coefficient was used to assess the correlation between tumor T stages and urine parameters.

Results

Patients with bladder cancer (n = 360) and controls (n = 1050) were recruited. In the univariate conditional logistic analysis, higher urine pH was associated with a decreased risk of bladder cancer (OR = 0.67, 95% CI = 0.57–0.78), while higher values of urine protein (OR = 4.55, 95% CI = 3.36–6.15), urine glucose (OR = 1.56, 95% CI = 1.18–2.05), and urine occult blood (OR = 4.27, 95% CI = 3.44–5.29) were associated with an increased risk of bladder cancer. After adjustment for body mass index, fasting blood glucose, hypertension, red blood cells, white blood cells, lymphocytes, neutrophils, and platelets, significance still remained for urine pH (OR = 0.68, 95% CI = 0.53–0.88), urine protein (OR = 1.97, 95% CI = 1.21–3.19), urine glucose (OR = 2.61, 95% CI = 1.39–4.89), and urine occult blood (OR = 3.54, 95% CI = 2.73–4.58).

Conclusion

This study indicated that lower urine pH and higher values of urine protein, urine glucose, and urine occult blood might be risk factors for bladder cancer.