Tumor Markers are parameters released by tumor cells, which enter the bloodstream or other biological fluids and are useful for the diagnosis, prognosis and treatment monitoring.
Most Tumor Markers are not specific to any type of cancer and the differences between benign and malignant diseases are quantitative (for example, patients with epithelial tumors tend to have significantly higher levels of these Tumor Markers than patients without malignancy).
There are now more than 20 well known parameters that are widely regarded as Tumor Markers such as PSA ―related to Prostate Cancer―, CA 15.3 ―related to Breast Cancer―, CA 125 and HE4 ―both related to Ovarian Cancer―, CEA and CA 19.9 ―both related to different gastrointestinal cancers (Colorectal, Gastric and Pancreatic Cancer)―, or NSE and ProGRP ―both related to in Lung Cancer―.
However, there are a variety of factors that can affect the accuracy of Tumor Markers by increasing its levels without malignancy presence. The main reason are benign diseases, among others, such as technical interferences.
In this sense, the Spanish Society of Clinical Biochemistry and Molecular Pathology, Cancer Biomarker Commission established the Barcelona Criteria, 4 criteria that help to correctly distinguish and value Tumor Markers results and reduce False Positives (FP):
- Tumor Markers Serum concentrations.
- Discard benign pathology by the exclusion of main source False Positive results.
- Follow-up if Tumor Markers moderate results (Grey Zone/Undetermined).
- Technical interference.
Statistical measurements in diagnostic tests
Unfortunately, the use of Tumor Markers in routine presents also other problems such as low Sensitivity in early stages, or nonexistence of any specific Tumor Marker for each malignant tumor. However, the combination of 2 or more Tumor Markers has a better outcome, especially in advanced stages.
In this regard, the combination of several Tumor Markers ―as well as the inclusion of patient history information in the equations―, using complex algorithms with multiple variables, results in higher Sensitivity and Specificity: that is what we have christened Multi-Biomarker Disease Activity Algorithm (MBDAA).
The Sensitivity of a diagnostic test is the percentage of actual positives that are correctly identified, and Specificity is the proportion of true negatives that are correctly classified. Both variables are closely linked together and give an idea of the accuracy of a test.
A test that correctly identifies all true positive as positive, but has many false negatives would have a Sensitivity of 100%, but low Specificity. For example, Sensitivity measures the number of cancerous tumors that are correctly identified as cancerous, whereas Specificity measures how many benign tumors are correctly identified as benign. A high Sensitivity means fewer cancers diagnosed as benign and high Specificity means fewer benign tumors diagnosed as cancerous.
Besides, the positive predictive value (PPV) is the number of true positives correctly identified on total real positive. A test with many false positives will have a low VPP. Moreover, the negative predictive value (NPV) is the number of true negatives correctly identified on the total actual negative. A high NPV value means that very few true positives were incorrectly identified as negative.
All these different values can be plotted together in a graphic that it is known as Receiving Operator Curve (ROC), where better results are displayed with curves that tends to come near to the upper left corner of the image (where 100% Sensitivity and 100% Specificity are reached).