The association between the acute-phase reactant proteins (APRPs) and cancer has long been established. There have been numerous reports correlating altered levels of various APRPs with different types of cancers. However, researchers are often quick to dismiss the use of these APRPs as potential biomarkers for the diagnosis and monitoring of cancer because alterations in APRP concentrations are observed in a wide range of diseases. Recent progress in proteomics studies which profiled the serum proteins of cancer patients and those of normal individuals indicated that the altered APRP expressions were different for distinct types, subtypes, and even stages of cancer. Interestingly, these data are in agreement with those observed earlier using immunochemical and biochemical assays. In view of this compelling association of different patterns of APRPs with various types of cancers and in an apparent shift of paradigm, we present in this review some indications that APRP fingerprinting may be used as complementary cancer biomarkers.
The objective of this study was to compare CA 72-4 with CEA and CA 19-9 in gastrointestinal malignancies. CA 72-4 was assayed by radioimmunoassay and CEA and CA 19-9 with the Abbott IMx analyser. The study included 52 patients with gastrointestinal cancer and 20 controls with benign gastrointestinal diseases. The 52 cases showed marker sensitivities of 39%, 49% and 35% for CA 72-4, CEA and CA 19-9, respectively, and 64% when the markers were combined. Marker expression in serum was highest in colorectal carcinoma followed by gastric and esophageal carcinoma. The sensitivities of the individual markers in colorectal, gastric and esophageal carcinomas, respectively, were: CA 72-4, 56%, 32% and 18%; CEA, 83%, 33% and 18%; CA 19-9, 53%, 25% and 18%. The sensitivity of the three markers in combination was 89%, 50% and 46% in colorectal, gastric and esophageal cancer, respectively. The specificity of CA 72-4, CEA and CA 19-9 was 100%, 72% and 86%, respectively. However, CA 72-4 is not a useful a marker for gastrointestinal cancers because of its poor sensitivity. CEA, which had the best overall sensitivity and a reasonable specificity, was the most useful single marker, especially for colorectal cancer. Whereas the single markers were not useful in gastric and esophageal cancer, the combination of the three may be.
This study was undertaken to investigate whether serum CA 125 could complement alpha-fetoprotein (AFP) to improve the diagnosis of hepatocellular carcinoma (HCC). CA 125 showed a sensitivity of 92% for HCC against the 58.8% sensitivity of AFP at the cutoff value of 200 ng/ml. However, the former was less specific (48.5% versus 97.4%) in relation to benign liver diseases (BLD). CA 125 had a higher negative predictive value (NPV) of 84.6% compared to 69.2% for AFP; when both markers were combined, however, the NPV rose to 91.7%. Overall, AFP was more efficient than CA 125 for the diagnosis of HCC. While a positive AFP result was highly indicative of HCC, a negative result did not rule out the disease; however, negative AFP and CA 125 meant that the likelihood of the disease was low. In situations of low HCC prevalence, CA 125 could serve as a first-line screening test followed by confirmation of positives by AFP.
Increased concentrations of insulin-like growth factor I (IGF-I) and decreased insulin-like growth factor binding protein 3 (IGFBP-3) in serum have been proposed as markers of prostate cancer (CaP). The evidence for this, however, is contradictory. We assayed serum for IGF-I, IGFBP-3 and prostate-specific antigen (PSA) in patients with CaP and benign prostatic hyperplasia (BPH) and in healthy controls (HC). The mean +/- SD concentration of IGF-I in CaP (98.3 +/- 39.3 ng/mL; n = 15) was lower than in BPH (119 +/- 31.1 ng/mL; n=24) and HC (119 +/- 36.1 ng/mL; n=46), but the differences between the three groups were not statistically significant (p > 0.05). The mean IGFBP-3 concentrations in CaP (2691 +/- 1105 ng/mL; n = 16; p = 0.029) and BPH (2618 +/- 816 ng/mL; n = 26; p = 0.006) patients were significantly lower than that of the HC (3119 +/- 618 ng/mL; n=59), but the difference between the two groups of patients was not significant (p > 0.05). PSA concentrations in CaP (median = 80.8 ng/mL; n = 25) were significantly higher than those in BPH (median = 8.6 ng/mL; n = 39) (p < 0.001). Ninety-six percent of CaP and 72% of BPH patients had PSA concentrations >4.0 ng/mL; the proportions of patients with concentrations exceeding 20 ng/mL were 76% and 10%, respectively. We conclude that IGF-I and IGFBP-3 are inferior to PSA for CaP detection.