Serum Ischemia Modified Albumin as a promising tumor marker in cancer patients

Background

Ischemia Modified Albumin (IMA) is the first blood test for cardiac ischemia (innstable angina pectoris and early myocardial infarction) approved by US FDA. IMA is based on the discovery that albumin, an abundant protein in blood is altered when exposed to ischemic tissue. It rises within minutes of the onset of ischemia and remains elevated for several hours after cessation of ischemia. IMA rises in the presence of ischemia and not as a result of necrosis. IMA, which appears to be an indicator of oxidative stress, may not be specific for cardiac, brain or lower extremities ischemia.

Anecdotal evidence suggests that IMA may increase as a result of oxidative stress in serum of cancer patients. Oxidative stress (OS) in cancer patients is caused by many reasons. Above all, it is predominance of glycolysis also in the presence of oxygen, which is energetically disadvantageous. The another factors supporting the IMA production are cancer anemia, hypoxia, acidosis, cancer cachexia (catabolism), vomiting, production of pro-inflammatory cytokins, and mainly radiotherapy and/or antineoplastic chemotherapy; they relevantly increase free radicals production. There is suppression of T-cells during OS. Moreover, OS is a potent factor in vascular cell proliferation. At the same time, there is depletion of plasma and tissue antioxidants (redox systems). All these factors contribute to progression of neoplastic diseases. Moreover, myocardium has less activity of catalase, superoxid dismutase and peroxidase; consequently it is more sensitive to oxidative stress and to development od cardiomyopathy. We intended to try use IMA as a relatively universal tumor marker and as a test of cadiotoxicity of antineoplastic drugs (anthracyclines, paclitaxel, trastuzumab; 5-fluorouracil which is responsible for coronary spasms mediated probably by endothelin production).

Patients and Methods

From 2nd January till end October 2007 the IMA was examined in serum of 429 cancer patients; reference value (cut-off point) was determined by examination of IMA serum level in 74 healthy adults.

IMA is detected in sérum using the Albumin Cobalt Binding test(ACBR test, ISCHEMIA Technologies, Inc., Denver, CO USA). A cobalt solution is first added to the serum, followed by the addition of dithiothreitol (DDT). The unbound cobalt reacts with the colour forming agent and is measured spectrophometrically. The higher the amount of unbound cobalt, the higher the ACB test value, and therefore the greater the concentration of IMA. Maximum time from vein to result is 2 ^m hours. It was found that increases in endogenous lactate inhibited the test. The results of ACB test was interpreted cautiously in cases with serum albumin <20 g/L or >55 g/L. Therefore, serum levels of lactate and albumin were determined. Data about IMA concentrations in noncardiac ischemia are limited but cancer patients with renal and/or liver failure, uncontrolled diabetes and sepsis were excluded. Results were evaluated by methods of decision analysis (the area under the Receiver Operator Characteristic Curve - ROC). The area under the ROC was calculated for each modeFs ability to discriminate cancer patients from healthy individuals. Ap value< 0,05 was considered as significant.

Results

Arithmetic mean of healthy adults and of cancer patients was 83.6 TJ/mL and 91,5 U/mL, respectively (p-value < 2.2 e-16). ROC curve including sensitivity (true positivity), specificity (true negativity) and value of AUC is in Fig.l. Changes of IMA sérum level according to age (< 55 years, 55 - 75 years, > 75 years) were statistically non significant (NS). We found signficant increasing tendency of IMA levels dependent on the rise of clinical stage. Statistically significant were differences between lst and 3rd and 4th and between 2nd and 4th clinical stage. The statistically significant increase of IMA levels were observed in majority of examined kinds of malignant tumors in descendent order (gastric cancer, lung cancer, rectosigmoidal cancer, colon cancer, renal cell carcinoma, and others). Interestingly, the increase of serum IMA levels in rectal cancer and breast carcinoma was not significant. There was close correlation of IMAand albumin levels (p-value < 2.2e-16; 95% CI). Serum IMA levels were not influenced by lactate levels (p-value=0,59 (95% CI).

Statistically significant increase of IMA levels were found in patients treated with different regimens of chemotherapy and/or radiation therapy. The IMA levels in descendent order were as follows: cisplatin +gemcitabine (100.98 U/mL), cisplatin + etoposide (91,88 U/mL), 5-fluorouracil + folinic acid (90,79 U/mL) and other (90,58 U/mL); significant increase was observed also during radiotherapy (90,26 U/mL). Very interesting was the observation in patients with non evidence of disease (NED) less than one month after chemotherapy and more than one year after chemotherapy; the mean values were 95,4 U/mL and 86,9 U/mL, respectivelly (p-value = 1.076e-08).

The ROC curve for IMA in all patients was convex against axis y; the area under the curve (AUC) = 0,718. The high AUCs (0,653 - 0,857) in increasing order were observed in the group of other cancers, renal cancer, colon cancer, rectosigmoidal cancer, gastric cancer and lung cancer. The best course of ROC curve was found out in patients with lung cancer. These results, i.e. sensitivity 69,9%, specificity 93,0%, and AUC 0,857 are in Fig. 2. The very high AUC was found in patients with IIIrd and IVth clinical stage (0,806 and 0,739, respectively).

The highest AUCs were recorded in patients treated by combination cisplatin + gemcitabine and cisplatin + etoposide (0,916 and 0,779 respectively). The ROC curve including extremely high sensitivity (85,7%), specificity (94,4%) and AUC value 0,916 is in Fig. 3. At the optimum determined cut-off point, there was in general recorded the high specificity of serum DVIA examination at the expence of lower sensitivity. The range of specificity (true negativity) in all groups of patients was 74,6 - 97,2% (mean = 95,8%).

Discussion

Low levels of reactive oxygen species (ROS) are indispensable as mediators in many of cell processes, including differentiation, cell cycle progression or the growth arrest, apoptosis and imunity. In contrast, high levels and/or inadequate removal of ROS result in oxidative stress, which may cause severe metabolic malfunctions and damage of biological macromolecules. Arecent study indicated an association between activities of antioxidant enzymes and increased levels of DNA lesions in lung cancer tissue, suggesting that free radical reactions may be increased in malignant cells in vivo. Lately, persistent OS in cancers has been suggested to explain partly activation of protooncogenes, genome instability, chemotherapy resistance and metastasis.

Prime targets of ROS are the polyunsaturated fatty acids in the membrane lipids. This attack causes lipid peroxidation. Further, the decomposition of peroxidized lipids yields a wide variety of end-products, including malondialdehyde (MDA) that is widely used in practice as an indicator of free radical damages.

Conclusions

On the base of our preliminary experience, the serum IMA as a result of oxidative stress, could be a promising tumor marker for stratification of cancer patients and monitoring of antineoplastic therapy.

References:

1. Banerjee, D., Madhusoodanan, U. K., Nayak, S., Jacob, J. Urinary hydrogen peroxide: a probable marker of oxidative stress in malignancy. Clinica Chimica Acta 2003; 334: 205 - 209.
   
2. Hrizostov, D., Gadjeva, V., Vlaykova, T, Dimitrov, G. Evaluation of oxidative stress in patients with cancer. Physiology and Biochemistry 2001; 109: 331-336.
   
3. Ilhan, N., Ilhan, N. Vascular endothelial growth factor and oxidative damage in cancer. Clinical Biochemistry 2003; 36: 225-228.
   
4. Kim, Y. T., Kim, W. T., Choi, J. S., Kim, S. H., Choi, E. K., Choi, N. H. Relation between deranged antioxidant system and cervical neoplasia. Int J Gynecol Cancer 2004; 14: 889-895.
   
5. Lases, E. C, Duurkens, V. A. M, Gerritsen W. B. M, Haas, F. J. L. M. Oxidative stress after lung resection therapy: A Pilot study. Chest2000; 117: 999-1003.
   
6. Peacock, F. et al. Metaanalysis of ischemia-modified albumin to rule out acute coronary syndromes in the emergency department. American Heart Journal 2006; 152: 253-262.
   
7. Wu, A. H. B. Markers for early detection of cardiac diseases. Scand J Clin Lab Invest 2005; 65 (Suppl 240): 112-121.

Supported by Grant ofSt. Elisabeth Univ. of Health Care and Social Work Bratislava.