Serum carcinoembryonic antigen and acute phase reactant proteins in the pre-operative detection of fixation of colorectal tumours

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.

Association of N-Linked Glycoprotein Acetyls and Colorectal Cancer Incidence and Mortality

Background

Acute phase proteins highlight the dynamic interaction between inflammation and oncogenesis. GlycA, a novel nuclear magnetic resonance (NMR) inflammatory marker that identifies primarily circulating N-acetyl glycan groups attached to acute phase proteins, may be a future CRC risk biomarker.

Methods

We examined the association between GlycA and incident CRC and mortality in two prospective cohorts (N = 34,320); Discovery cohort: 27,495 participants from Women's Health Study (WHS); Replication cohort: 6,784 participants from Multi-Ethnic Study of Atherosclerosis (MESA). Multivariable Cox models were adjusted for clinical risk factors and compared GlycA to acute phase proteins (high-sensitivity C-reactive protein [hsCRP], fibrinogen, and soluble intercellular adhesion molecule-1 [sICAM-1]).

Results

In WHS (median follow-up 19 years, 337 cases, 103 deaths), adjusted HRs (95% CIs) per SD increment of GlycA for CRC incidence and mortality were 1.19 (1.06–1.35; p = 0.004) and 1.24 (1.00–1.55; p = 0.05), respectively. We replicated findings in MESA (median follow-up 11 years, 70 cases, 23 deaths); HRs (95% CIs) per SD of GlycA for CRC incidence and mortality were 1.32 (1.06–1.65; p = 0.01) and 1.54 (1.06–2.23; p = 0.02), respectively, adjusting for age, sex, and race. Pooled analysis, adjusted HR (95% CI) per SD of GlycA for CRC incidence and mortality was 1.26 (1.15–1.39; p = 1 x 10⁻⁶). Other acute phase proteins (hsCRP, fibrinogen, and sICAM-1) had weaker or no association with CRC incidence, while only fibrinogen and GlycA were associated with CRC mortality.

Conclusions

The clinical utility of GlycA to personalize CRC therapies or prevention warrants further study.

Trial Registration

ClinicalTrials.gov: WHS NCT00000479, MESA NCT00005487

Clinical and prognostic significance of pathological and inflammatory markers in the surgical treatment of locally advanced colorectal cancer

Background

Locally advanced colorectal cancer (CRC) may vary in its clinical and pathological appearance. It is now accepted that progression of disease in patients with locally advanced CRC is determined not only by local tumor characteristics but also by the immune system and inflammatory response in the body.

Methods

We investigated patients with confirmed CRC who were treated in the surgical clinic at the University Hospital Alexandrovska over a 10-year period and retrospectively evaluated the histological features of the preoperative biopsies and operative specimens removed during radical multivisceral resections. We also collected prospective data for serum C-reactive protein levels and Jass-Klintrup score, Petersen Index score, and Glasgow Prognostic Score in patients with locally advanced CRC.

Results

Of 1,105 patients with CRC, 327 (29.6%) were diagnosed with locally advanced disease. In total, 108 combined multivisceral resections (79 for primary tumors and 29 for recurrent tumors) were performed. Overall survival was 34 months for pR0 cases and 12 months for pR1 cases (P<0.05). Our data confirmed that C-reactive protein is a prognostic marker of overall survival. Data for 48 patients with histologically confirmed locally advanced tumors showed significantly increased survival with a higher Jass-Klintrup score (P=0.037). In patients with node-negative disease, 5-year survival was 49%. However, where there were high-risk pathological characteristics according to the Petersen Index, survival was similar to that for node-positive disease (P=0.702). Our data also showed a significant difference in survival between groups divided according to whether they had a modified Glasgow Prognostic Score of 1 or 2 (P=0.031).

Conclusion

In order to maintain a reasonable balance between an aggressive approach and so-called meaningless “surgical exorbitance”, we should focus on certain histopathological and inflammatory markers that can be identified as additional factors for planning the type and volume of surgical treatment.

Preoperative elevation of serum C-reactive protein as an independent prognostic indicator of colorectal carcinoma

PURPOSE

The preoperative elevation of serum C-reactive protein (CRP) is thought to be a prognosticator of carcinomas of the digestive tract. We conducted this study to investigate the clinical importance of the preoperative elevation of serum CRP in patients with colorectal carcinoma (CRC).

METHODS

We investigated the correlation between an elevated preoperative serum CRP level and the clinicopathologic factors, including prognosis, of 116 patients who underwent resection of CRC.

RESULTS

Forty-seven (40.5%) patients had an elevated serum CRP value preoperatively (group H) and 69 (59.5%) did not (group L). There were significant differences in the tumor size, proportion of poorly differentiated tumors, depth of invasion, lymph node metastasis, lymphatic invasion, and tumor stage between the two groups. Survival was significantly lower in group H than in group L (P < 0.0001). Multivariate analysis showed that the preoperative elevation of serum CRP (P = 0.0007), as well as poor differentiation (P = 0.027) and advanced tumor stage (P = 0.007) were independent prognostic factors in patients with CRC.

CONCLUSION

We found the preoperative elevation of serum CRP to be an independent prognostic indicator of CRC.

Serum proteins profile as an indicator of malignancy: multivariate logistic regression and ROC analyses

The electrophoretic pattern of serum proteins of 85 patients carrying different types of neoplasia and 85 matched healthy adults were comparatively studied by agarose gel electrophoresis, to find out if there is a specific protein pattern common to different types of cancer. Each protein fraction was analyzed quantitatively by densitometry. Multivariate logistic regression and receiver operating characteristic (ROC) analyses were performed using SPSS software. When total protein and albumin were measured by colorimetric methods, cancer patients, compared to controls, had a decreased concentration of total protein (66.0+/-11.5 g/L vs. 76.4+/-6.8 g/L, p< or =0.0001) and of albumin (39.0+/-8.1 g/L vs. 46.0+/-4.3 g/L, p< or =0.0001). The electrophoretic data of serum proteins showed that the ratio of albumin to globulin (0.92+/-0.30 vs. 1.21+/-0.16, p< or =0.0001), percent of the fractions albumin (46.7+/-8.5% vs. 54.4+/-3.5%, p< or =0.0001) and beta-globulin (11.6+/-4.4% vs. 13.0+/-1.9%, p< or =0.001) were decreased and alpha(1)- (5.3+/-2.5% vs. 2.9+/-0.8%, p< or =0.0001), alpha(2)- (13.5+/-4.8% vs. 11.3+/-2.1%, p< or =0.0001) and gamma-globulins (23.0+/-7.7% vs. 18.3+/-3.1%, p< or =0.0001) were significantly increased in cancer patients relative to controls. Cancer patients also had higher counts of leukocytes (7.98+/-3.11, x10(9) cells/L vs. 6.33+/-1.68 x10(9) cells/L, p< or =0.0001) and erythrocyte sedimentation rate (35.9+/-23.5 mm/h vs. 14.1+/-9.5 mm/h,p< / =0.0001). On the basis of univariate analysis, a protein profile out of the normal ranges was more prevalent in cancer patients than in controls. Analysis of the data using multiple logistic regression indicated that the prevalence of cancer was strongly associated with the serum proteins' profile, and alpha(1)-globulin, erythrocyte sedimentation rate, total protein and the ratio of albumin to globulin were the best parameters to discriminate between malignant and healthy states. The area under the ROC curves were the same for most components of the serum proteins' profile at about 0.75+/-0.09, p< or =0.001. We conclude that the profile of serum proteins indicates high diagnostic values for discriminating between cancer patients and healthy individuals and may be useful as an adjunct diagnosis for detection of malignancy.

Serum amyloid A protein (SAA) in colorectal carcinoma

The changes in serum levels of serum amyloid A protein were studied in 67 patients suffering from colorectal carcinoma and compared to three other major acute phase proteins: C-reactive protein, alpha1-antichymotrypsin and alpha1-acid glycoprotein. Although the presence of colorectal carcinoma caused an increase in serum levels of all the acute phase reactants studied, serum amyloid A protein showed the most powerful reaction in pre-operative disease stage, with the mean value of 330 mg/l (range 7-2506 mg/l) as compared to the normal values of <1.2 mg/l obtained in 30 healthy adults. The mean serum amyloid A protein concentration increased to 487 mg/l after surgery, declining during the post-operative clinical course until the sixth chemotherapy cycle (from 167 mg/l to 64 mg/l), but never returned to the normal range. In the later chemotherapy cycles, mean serum amyloid A protein increased to 163 mg/l, probably as a result of the disease relapse. According to the statistical relations among exact confidence intervals for proportions, serum amyloid A protein showed the best specificity for colorectal carcinoma of all the acute phase proteins studied (83-100%) and also a sensitivity of 100%. We concluded that serum amyloid A protein seems to be a reliable parameter, which could be recommended for clinical routine as a non-specific tumour marker for colorectal carcinoma.

Preoperative plasma plasminogen activator inhibitor type-1 and serum C-reactive protein levels in patients with colorectal cancer. The RANX05 Colorectal Cancer Study Group

BACKGROUND

Preoperative plasma plasminogen activator inhibitor-1 (PAI-1) is a prognostic variable in patients with colorectal cancer. It has been suggested, however, that plasma PAI-1 is a nonspecific prognostic parameter similar to the acute-phase reactant C-reactive protein (CRP). In the present study we analyzed the association between plasma PAI-1 and serum CRP in patients scheduled for elective resection of colorectal cancer. In addition, the prognostic value of PAI-1 and CRP was studied in this patient cohort.

METHODS

PAI-1 and CRP were analyzed in citrated plasma and serum, respectively, obtained preoperatively from 594 patients. Patients who required preoperative blood transfusion received SAGM blood, in which soluble PAI-1 is not present. None of the patients received pre- or postoperative adjuvant chemotherapy, and all were followed in the outpatient clinic for at least 5 years or until death. The association of PAI-1 and CRP, respectively, with survival was tested using the median value of PAI-1 and the upper normal limit for CRP. Analyses were performed by inclusion of all patients, and in the subgroup of patients, who underwent curative resection.

RESULTS

The median follow-up period was 6.8 (5.4-7.9) years. The median value of plasma PAI-1 was 35.8 ng/ml, and values greater than 94 nmol/L identified patients with increased CRP levels. Comparison of the molecules showed that PAI-1 was weakly correlated with CRP (r = .26; P <.0001). Both molecules showed a Dukes independent distribution. In univariate survival analyses high levels of PAI-1 were found associated with poor prognosis and low levels with good prognosis (P = .02, HR: 1.3). Similarly, high levels of CRP were found associated with poor prognosis and low levels with good prognosis (P <.0001, HR: 1.9). In a multivariate statistical analysis including Dukes classification, gender, age, tumor location, perioperative blood transfusion, PAI-1 and CRP, plasma PAI-1 was a dependent prognostic variable, while serum CRP (P <.0001; HR: 1.4; 95% CI: 1.3-1.5) was found to be a Dukes independent prognostic variable. Similar analyses, excluding patients with Dukes' D disease showed serum CRP to be an independent prognostic variable (P <.0001; HR: 1.3: 95% CI: 1.2-1.5).

CONCLUSIONS

This study did not show a strong correlation between plasma PAI-1 and serum CRP in patients with colorectal cancer. Serum CRP was found to be a Dukes independent prognostic variable in this patient cohort, and was found to identify a subgroup of curatively resected patients at risk for short survival.

Pre-operative plasma levels of C-reactive protein, albumin and various plasma protease inhibitors for the pre-operative assessment of operability and recurrence in cancer surgery

Pre-operative levels of the acute phase protein C-reactive protein (CRP), albumin (assessing nutritional status), the tumour marker CEA and three plasma protease inhibitors, i.e. C1-esterase inhibitor, alpha-2-macroglobulin and antithrombin III, were prospectively studied in 183 patients with various solid cancers. First, the predictive value of abnormal levels for operability at the primary operation was studied. Secondly, the predictive value of abnormal levels for cancer recurrence and metastases was evaluated during 2 years of follow-up. The results show that malignancy induces increased CRP and C1-esterase inhibitor levels and decreased albumin levels in serum. These changes, as well as raised alkaline phosphatase and lowered haemoglobin levels, also correlate to the 'overall' tumour burden. The most important conclusion is, that increased pre-operative CRP levels (CRP > or = 10 mg/l; sensitivity, 79%; specificity, 71%) and/or low albumin levels (albumin <37 g/l; sensitivity, 94%; specificity, 54%) are seen in inoperable cancer patients compared with patients having operable cancers. The second main important conclusion is, that high pre-operative C1-esterase inhibitor levels (C1-esterase inhibitor >152%; sensitivity, 45%; specificity, 90%), and in some patients a high alkaline phosphatase level, are seen in patients exhibiting early cancer recurrence (within 2 years post-operatively).

Staging and surgery for colorectal cancer

Staging of colorectal cancer has become increasingly important to select groups of patients for limited or more extensive surgery, and for adjuvant radiotherapy and chemotherapy. The main treatment is still surgery, but subgroups may benefit from adjuvant therapy, even accepting additional side effects. Accurate staging is necessary to define different treatment groups. A critical review is given of the present methods of clinicopathological staging.

Prognostic factors in colorectal cancer: current status and new trends

The clinical, laboratory, and pathologic data of 361 patients who had curative resections for colorectal cancers were collected and analyzed in a multiple stepwise regression model. In univariate analysis, among clinical factors, bowel obstruction and emergency surgery showed the most significant prognostic value (P = 0.002, P = 0.004, respectively). Vegetating growth, Astler-Coller stage of tumor, intramural spread, lymph node involvement, and synchronous liver metastases resulted in the pathologic variable significantly affecting the prognosis (P = 0.006, P less than 0.001, P = 0.036, P less than 0.001, P less than 0.001, respectively). In the multivariate analysis, stage was the predictive factor with the highest hazard ratio in conjunction with bowel obstruction (P less than 0.0001 in both cases). Processing data excluding stage ("multiparametric factor" itself), hepatic metastases, lymph node involvement, bowel obstruction, and intramural spread appeared as independent predictors of survival (P less than 0.0001, P less than 0.0001, P = 0.0004, P = 0.0316, respectively). Other variables, as biologic and molecular factors, should be more widely tested in order to assess their impact on prognosis.

The Dijon clinical staging system for early rectal carcinomas amenable to intracavitary treatment techniques

The Dijon clinical and endoscopic staging system for intracavitary radiotherapy of rectal cancer takes into account the size and the depth of penetration of the rectal wall. Its prognostic value was evaluated in a series of 72 patients with rectal adenocarcinoma treated at the Centre de Lutte Contre le Cancer G. F. Leclerc in Dijon: 30 presented with a clinical stage (CS) T1A (purely exophytic tumors of less than 3 cm). The 5-year local relapse-free actuarial survival (LRFS) was 97%. Fourteen patients with CS T1B (infiltrative component and less than 3 cm diameter) had a LRFS of 77%. Nine patients with CS T2A tumors (with larger exophytic tumors) has a LRFS of 65%. Nineteen CS T2B cases (larger than 3 cm with an infiltrative component) presented a LRFS of 60%. The size of the tumor and the clinical estimate of the infiltration of the rectal wall both have a significant prognostic value: adenocarcinoma of less than 3 cm (n = 44) had a LRFS of 93% versus 59% in larger ones (n = 39; p = less than 0.01). Free mobile lesions (n = 39) did better (n = 33; LRFS = 86%) than infiltrated tumors (n = 33; LRFS = 66%; p = 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)

Colorectal cancer. The bases for a comprehensive follow-up

The purpose of this article was to review the effectiveness of follow-up in patients with colorectal cancer submitted to curative treatment. A comprehensive follow-up involves rational initial management of the primary tumor, knowledge of prognostic factors, selection of the patient to be followed, determination of the time for follow-up, use of the most appropriate tests for early diagnosis of recurrence, and eventual curative treatment. The updated answers to all these questions are given through an extensive review of the world literature and confronted with the authors' experience of eight years of follow-up in a series of 170 colorectal cancer patients treated for cure. Although the future might be more promising, past world experience suggests only a few patients could be saved. It is concluded that there is no place for incomplete and disperse screening tests, and only comprehensive, intensive, and very well-coordinated follow-up programs should be undertaken if better results are hoped to be achieved.

Pre-operative staging of rectal neoplasm and its impact on clinical management

A sophisticated staging technique (extended staging, ES) employing modern technology has been compared prospectively with conventional clinical assessment (initial staging, IS) in 45 patients with low rectal carcinoma (less than 12 cm from the anal verge) to determine its potential merit and its impact on clinical management. ES consisted of computerized tomography of liver and pelvis, ultrasound scan of liver, measurement of serum concentrations of carcinoembryonic antigen (CEA) and acute phase reactant proteins and multiple superficial and deep biopsies to determine not only the histological grade of the tumour but also its DNA cellular content as measured by flow cytometry. ES proved statistically superior to IS in the assessment of local spread and dissemination. Although histological grade proved to be more accurate on ES than IS this was only true when one specialist pathologist interpreted the slides. When several pathologists were involved interobserver variation made interpretation unreliable. Assessment of DNA content using flow cytometry, being quantitative, was more accurate and perhaps should be used in the future as a prognostic indicator. The improved accuracy of ES would have altered both pre- and intra-operative clinical decisions. It would have prevented some patients receiving inappropriate adjuvant therapy as well as selecting patients more accurately for the correct treatment.

The role of peptidases in cancer of the rectum and sigmoid colon

Increased levels of peptidases are found in some human carcinomas and may be related to invasive potential. We therefore measured the activity of four peptidases in 50 specimens of tumour and normal colonic wall from patients with a rectal or sigmoid carcinoma, and correlated this with the stage, differentiation, fixity of the tumour and presence of venous invasion, determined histologically. Since acute phase reactant proteins (APRP) may inhibit these proteolytic enzymes we have also measured serum levels of two relevant APRPs, alpha 1 acid glycoprotein (AGP) and C-reactive protein (CRP) pre-operatively. Activity of cathepsin B, cathepsin H and collagenase-like peptidase (CLP) was determined fluorimetrically and collagenase photometrically. Significantly elevated activity of cathepsin B, CLP and collagenase was found in tumour compared with normal colonic wall (median values: (nmol (mg protein)-1 min-1) Cat B 0.71 and 0.42 (P less than 0.001), CLP 25.24 and 12.25 (P less than 0.0001) and collagenase 0.49 and 0.31 (P less than 0.001). There was no correlation between the activity of these enzymes expressed as a ratio of tumour/colonic wall, and differentiation or Dukes' stage of the tumour. However, there was significant elevation of activity of cathepsin B in tumours with local spread (n = 13) compared with those with no spread (n = 37) (median values 2.76 and 1.36 respectively (P less than 0.001] and also in tumour with venous invasion (n = 24) compared with tumours without (n = 26) (median values 1.82 and 1.18 respectively (P less than 0.01]. Pre-operative serum levels of CRP were inversely correlated with the activity of CLP and cathepsin H and collagenase in the tumours (rs = 0.332, 0.359 (P less than 0.05) and 0.302 (P = 0.05) respectively). Thus certain peptidases are raised in rectal and sigmoid tumours. Activity of cathepsin B appears related to local tumour invasion. APRP may have a role in inhibiting the activity of these enzymes. These findings may have therapeutic implications.