Sensitivity of monoclonal antibodies to carcinoembryonic antigen, tissue polypeptide antigen, alpha-fetoprotein, carbohydrate antigen 50, and carbohydrate antigen 19-9 in the diagnosis of colorectal adenocarcinoma

BH-index: A predictive system based on serum biomarkers and ensemble learning for early colorectal cancer diagnosis in mass screening

BACKGROUND AND OBJECTIVE

Colorectal cancer is one of the most common malignancies among the general population. Artificial Intelligence methodologies based on serum parameters are in continuous development to obtain less expensive tools for highly sensitive diagnoses. This study proposes a predictive system based on serum biomarkers and ensemble learning to predict colorectal cancer presence and the related TNM stage in patients.

METHODS

We have selected 17 significant plasmatic proteins, i.e., Carcinoembryonic Antigen, CA 19-9, CA 125, CA 50, CA 72-4, Tissue Polypeptide Antigen, C-Reactive Protein, Ceruloplasmin, Haptoglobin, Transferrin, Ferritin, α-1-Antitrypsin, α-2-Macroglobulin, α-1 Acid Glycoprotein, Complement C4, Complement C3, and Retinol Binding Protein, regarding 345 patients (248 affected by the neoplastic disease). The proposed system consists of two predictors, i.e., binary and staging; the former predicts the presence/absence of cancer, while the latter identifies the related TNM stage (I, II, III, or IV). The experiments were conducted by deploying and comparing Random Forest, XGBoost, Support Vector Machine, and Multilayer Perceptron with feature selection based on Gini Importance and with dimensionality reduction via PCA.

RESULTS

The results show that the system composed of XGBoost as binary and staging predictor reaches 91.30% accuracy, 90% sensitivity, and 93.33% specificity for the absence/presence outcome, while 66.66% accuracy for the staging response. With the expansion of the training set in favor of positive patients and majority voting, the system composed of the combination of Support Vector Machine, XGBoost, and Multilayer Perceptron as the binary predictor reaches 98.03% accuracy, 100% sensitivity, and 92.30% specificity, while the combination of Random Forest, XGBoost, and Multilayer Perceptron as staging predictor achieves 60% accuracy. The final system reaches, in terms of accuracy, 98.03%, and 66.66% for the binary and staging predictors, respectively. It was also found that the biomarkers which contribute most to the binary decision are Ceruloplasmin and α-2-Macroglobulin, while the least significant dimensions are CA 50 and α-1-Antitrypsin; instead, Carcinoembryonic Antigen and α-1 Acid Glycoprotein are the most significant to the staging decision.

CONCLUSIONS

The present study proves the effectiveness of deploying serum biomarkers as feature dimensions for early colorectal cancer diagnosis and of using majority voting for noise reduction in the prediction.

Predictive performance of TPA testing for recurrent disease during follow-up after curative intent surgery for colorectal carcinoma

BACKGROUND

The aim of the present study was to investigate the predictive performance of serial tissue polypeptide antigen (TPA) testing after curative intent resection for detection of recurrence of colorectal malignancy.

METHODS

Serum samples were obtained in 572 patients from three different hospitals during follow-up after surgery. Test characteristics of serial TPA testing were assessed using a cut-off value of 75 U/L. The relation with American Joint Committee on Cancer stage and the potential additive value of tissue polypeptide antigen testing upon standard carcinoembryonic antigen (CEA) testing were investigated.

RESULTS

The area under the receiver operating characteristic curve of TPA for recurrent disease was 0.70, indicating marginal usefulness as a predictive test. Forty percent of cases that were detected by CEA testing would have been missed by TPA testing alone, whilst most cases missed by CEA were also not detected by TPA testing. In the subpopulation of patients with stage III disease predictive performance was good (area under the curve 0.92 within 30 days of diagnosing recurrent disease). In this group of patients, 86% of cases that were detected by CEA were also detected by TPA.

CONCLUSIONS

Overall, TPA is a relatively poor predictor for recurrent disease during follow-up. When looking at the specific subpopulation of patients with stage III disease predictive performance of TPA was good. However, TPA testing was not found to be superior to CEA testing in this specific subpopulation.

Transient human anti-mouse antibody generated with immune enhancement in a carbohydrate antigen 19-9 immunoassay after surgical resection of recurrent cancer

We report a case of transient human anti-mouse antibody from a 64-year-old man in a carbohydrate antigen 19-9 immunoassay using an AIA 1800 analyser that generated immune enhancement after surgical resection of recurrent cancer.

METHODS

The carbohydrate antigen 19-9 concentration was measured using an AIA 1800 analyser and a UniCel Dxl 800. Size-exclusion high-performance liquid chromatography was carried out on a Superose 12 column to estimate the carbohydrate antigen 19-9 elution profile using an AIA 1800 analyser. To determine whether IgM in the patient contributed to the carbohydrate antigen 19-9 immunoassay, immunoprecipitation was performed. Furthermore, mouse immunoglobulins were added to the patient's serum to verify that the patient's IgM reacted with it.

RESULTS

The carbohydrate antigen 19-9 concentration was >400 and 9.5 kU/L using an AIA 1800 analyser and using a UniCel Dxl 800, respectively. In the single carbohydrate antigen 19-9 peak, the molecular weight corresponded to IgM by size-exclusion high-performance liquid chromatography on a Superose 12 column. In the immunoprecipitation reaction and addition of mouse immunoglobulins, there was interference for anti-human IgM and mouse immunoglobulins whose recoveries were 3.2 and 14.2%, respectively. These results indicated that IgM in the patient's serum interfered with the carbohydrate antigen 19-9 immunoassay using an AIA 1800 analyser.

CONCLUSION

A novel transient human anti-mouse antibody generated with immune activation in a carbohydrate antigen 19-9 immunoassay using an AIA 1800 analyser was identified in a patient with rectal cancer after surgical resection. These findings demonstrate the importance of monitoring tumour markers in patients after treatment with mouse monoclonal antibody.

Rapid detection of S. mutans surface antigen I/II using a sensitive monoclonal anti-Ag I/II antibody by ELISA

The cell-surface protein antigen I/II (Ag I/II) is expressed in oral streptococci, which are known as the causative agent of a number of diseases including dental caries, endocarditis, gingivitis, and periodontal disease. Consequently, monoclonal antibodies (MAb) capable of recognizing the streptococcal Ag I/II protein could be a useful tool for the diagnosis and cure of these diseases. In this study, a previously generated monoclonal anti-Ag I/II antibody, ckAg I/II, was used to detect a small amount of Streptococcus mutans (S. mutans) surface antigen Ag I/II. The ckAg I/II was proved to be very sensitive and able to detect as little as 1 ng of recombinant Ag I/II protein within 5 min and Ag I/II in saliva within 10 min, as well as native Ag I/II in 20 μL of culture supernatant by ELISA. These results suggest that ckAg I/II can be used as a fast and efficient diagnostic tool to detect Ag I/II.

Utility of slot-blot-ELISA as a new, fast, and sensitive immunoassay for detection of carcinoembryonic antigen in the urine samples of patients with various gastrointestinal malignancies

Carcinoembryonic antigen (CEA) is the most widely used clinical tumor marker. CEA immunoassay has found acceptance as a diagnostic adjunct in clinical diagnosis of gastrointestinal tumors (GIT). Several immunoassays have been established for detection of CEA in plasma, serum, tissue, feces, and urine of cancer patients using polyclonal or monoclonal antibodies raised against CEA. Some of these assays display both high sensitivity and specificity for the detection of CEA. However, these assays require special and highly expensive equipment and the procedures require long periods for their completion. In the present study, we established a Slot-Blot Enzyme Linked Immunosorbent Assay (SB-ELISA), based on anti-CEA monoclonal antibody (CEA-mAb), as a new, simple, fast, cheap, and non-invasive immunodiagnostic technique for detection of CEA in the urine of GIT patients. Urine and serum samples were collected from 248 GIT patients (58 with pancreatic cancer, 20 with hepatoma, 23 with ampullary carcinoma, 15 with hilar cholangiocarcinoma, 28 with gastric cancer, 14 with esophageal cancer, and 90 with colorectal cancer). Moreover, urine and serum samples were collected from 50 healthy individuals to serve as negative controls. The traditional ELISA technique was used for determination of CEA in the sera of GIT patients using anti-CEA monoclonal antibody. A comparison between the results of both techniques (ELISA and SB-ELISA) was carried out. The traditional ELISA detected CEA in the sera of 154 out of 248 GIT patients with a sensitivity of 59.8%, 51.7% positive predictive value (PPV) and 75.37% negative predictive value (NPV). In addition, it identified 15 false positive cases out of 50 healthy individuals with a specificity of 70%. The urinary CEA was identified by a Western blotting technique and CEA-mAb at a molecular mass of 180 Kda. The developed SB-ELISA showed higher sensitivity, specificity, PPV, and NPV (70.1%, 78%, 62.4%, and 82.13%, respectively) for detection of CEA in the urine of GIT patients. The semi-quantitative SB-ELISA showed a higher overall efficiency of 72.8% versus 63.4% in the case of the quantitative ELISA, for detection of CEA. In conclusion, SB-ELISA is more efficient for detection of CEA in gastrointestinal tumors. It is a simple, rapid, non-invasive, and sensitive assay. Moreover, all steps of the SB-ELISA are performed at room temperature, without the use of expensive equipment; this may enhance the application of this assay in field studies and mass screening programs.

Plasma sialic acid as a marker of the effect of the treatment on metastatic colorectal cancer

The concentration of total sialic acid (TSA) is increased in the plasma of patients with many types of cancer. The purpose of this study was to assess the usefulness of the TSA marker in predicting the efficacy of the treatment, and to compare TSA with two common markers, carcinoembryonic antigen (CEA) and the carbohydrate antigen 19-9 (CA 19-9). The study was performed on 44 patients treated for advanced colorectal carcinoma by a weekly 8 h continuous infusion of 5-fluorouracil (1300 mg/m2) plus bolus injection of L-folinic acid (100 mg/m2). TSA, CEA and CA 19-9 levels were measured before and after 3 months of treatment and their variations analysed as a function of the response to the treatment. TSA levels of patients with metastatic colorectal carcinoma before treatment (959 +/- 265 mg/l) were significantly higher than those of 32 healthy people (584 +/- 99 mg/l). The percentage of patients with TSA concentration above the cut-off level (782 mg/l) was 73% before treatment and 23% after. All patients who experienced an objective response to the treatment (complete, partial or minor response) (n = 29) had a significant decrease of TSA levels (t = 5.96; P < 0.001). When the disease was considered as stabilised (n = 10), TSA changed slightly, but it increased with progressive disease (4 out of 5 patients). Changes in CEA and CA 19-9 did not correlate as well as TSA to the treatment efficacy. Initial levels of TSA did not permit prediction of the efficacy of the treatment since they were not significantly different between the five response groups. TSA seems to be more likely involved in tumour changes than in tumour volume. Its determination could provide useful information about the spreading and metastatic properties of the tumour. TSA normalisation is an indicator of probable tumour growth arrest and its elevation could be a marker of relapse.