David K. Walburger, Department of Microbiology
Abstract
The ability to follow a cancer’s aggressiveness and response to treatment throughout the disease’s course is critical in treatment programs. The tumor markers used as indicators have not been able to consistently do this. Thymidine Kinase, an enzyme linked with cellular proliferation, has been shown to be persistently elevated in almost all cancer serums, and corresponds well with disease stage in breast cancer patients. It has also been shown that TK levels in serum, as well as in tumors, decrease drastically after treatment; subsequent rises are indicative of disease recurrence. Currently, TK levels are measured by radionucleoside phosphorylation assays. There are considerable obstacles to the feasibility of these types of measurements in a clinical setting: radioactive waste, narrow activity range, and specialized equipment. This paper presents evidence supporting the clinical use of thymidine kinase as a prognostic marker using an immunoassay from a panel of monoclonal antibodies. Using 218 serum samples from breast cancer patients; a comparative study was conducted measuring TK with both the immunoassay and 3H-radioassay. Our findings show that the immunoassay is as effective in measuring TK levels as the radioassay (P<O.OO!).
Introduction
In most industrialized and many developing nations, cancer claims the lives of millions of individuals and is the leading cause of mortality among women. It is estimated that I in 5 people will die of cancer and over 1.5 mlllion women in the United States will acquire breast cancer in the 1990’s. Of these, approximately 30% will die. Even with improved diagnostic and treatment methods, mortality rates of breast cancer in the United States continue to lead all other cancers (second only recently to lung cancer). While breast cancer screening has reduced mortality by 30%, metastasis are often present in even the smallest tumors detectable by mammography.
Thymidine kinase (TK) is a pyrimidine nucleotide salvage pathway enzyme which catalyzes the phosphorylation of thymidine to thymidine monophosphate (TMP) (I). It has been shown that TK levels are elevated in the serum of breast cancer patients and corresponds with the disease stage (2). In a study of patients treated with primary hormonal therapy, TK levels decreased in patients responding to treatment and increased in patients with cancer progression (3). Other studies have shown similar results (4,5). Elevated TK levels In serum and tumors of post-treatment patients have also been shown to predict the recurrence of disease (6). Furthermore, the prognostic value of TK is not limited to breast cancer. In patients with acute and chronic leukemias and Hodgkin’s and non-Hodgkin’s lymphomas, elevated serum TK levels corresponding with disease stage have been observed (7 -I 7). Other studies have made similar findings in patients with brain, colon, bladder cancers, and other solid tumors (15; 18-25).
A comparative study of prognostic indicators in breast cancer confirmed that TKI is useful as a prognostic indicator. Cox multivariate analysis has been used to show a relationship between TK and both overall survival and disease free survival (26). Although the data supporting the prognostic value of TK has been established, this information has not been utilized clinically. Perhaps this is due to the fidelity and feasibility of the current assays for TK levels. We have developed a panel of anti-TK monoclonal antibodies which can be applied in an immunoassay. Using standard ELISA techniques and the monoclonal antibody (Mab), we compared the efficacy of an immunoassay of two Mabs with the current radioassay.
Materials and Methods
Serum Samples. A total of 218 serum samples from postoperative breast ancer patients at the Utah Valley Regional Medical Center Provo, Utah, were collected over a 36 month period. The samples were subsequently frozen at -20 C until assayed for TK levels.
TK Radioassay. The samples were assayed for TK activity using the radiolabeled 3H thymidine assay (26). The assay mix contains 0.02 M TrisHCI (pH 7.9), 2 x 10-6M 3H thymidine (85 Ci I mmol; Amersham Corporation, Arlington Heights, IL), 0.002 M MgCJ,,0.2 M KCI, 0.1 M NH,CI, 0.005 M mercaptoethanol, and 0.004 M ATP prepared fresh before each assay. Tubes containing I 00 111 of assay mix and 100 111 serum were incubated together for 60 minutes at 37′ C. A total of 4 diethylaminoethyl cellulose (DE-81) paper discs (Whatman, Inc., Clifton, NJ) were spotted with 25 111 of the sample and allowed to dry. The discs were then washed three times in 0.00 I M ammonium formate, distilled water, and then fixed in I 00% ethanol. Disks were placed in scintillation vials with 4 ml of Cyto-Scint (ICN Biomedicals, Inc., Irivine, CA), and radioactivity was measured in a scintillation counter. A linearization assay was preformed on Hela cell extract obtain through 3 consecutive diluted 1:10 at 20, 40 and 60 minutes check the potency of the assay compounds (Figure 1).
TK Immunoassay. Anti-TK monoclonal antibodies (KORG-1 & TKCM-1) were generated in our laboratory and employed in ELISAs as follows: Each sample was measured for protein content using the Bradford assay (27) and diluted in PBS to a concentration of 1.0 11g/ ml. 96 well plates (Corning, Corning, NY) were coated with 50 111 of serum and allowed to dry overnight. The wells were then rehydrated with 200 111 of PBS containing 0.0 1% gelatin and 0.05% Tween-20 to block any regions not coated with protein. Next, 100 111 of KORG-1 or TKCM-1 antibodies were applied to each well and incubated at 37′ C for 60 minutes. The plates were then washed three times with PBS Tween-20, blotted dry, and 100 111 of goat anti- mouse IgG (H+L) conjugated to horseradish peroxidase (Bio-Rad, Hercules, CA) diluted I :4000 in PBS, were added and incubated for 60 minutes at 37′ C. The plates were washed as previously described, and 200 111 of 3, 3′, 5′, 5′-tetramethylbenzidine substrate was added to each well and allowed to develop for 60 minutes. The reaction was stopped by the addition of 50 111 2M H2S04, and the plates were read at 450 nm on a Bio-Rad 450 microplate reader.
Results and Discussion
Comparison of the radioassay and the ELISA absorbance data of total TK level measurements (Figure I) using the Pearson correlation indicates that there is a strong correlation between the two methods. According to our analysis, the P value is less than 0.001, which indicates a strong correlation between the two tests. The Pearson correlation shawl (Figure 2) indicates a factor of 0. 794 when comparing absorption (Korg I) with the standard ATP radioassay. This correlation demonstrates that an acceptable margin of error exists between the radioassay and the immunoassay. Deviations from the radioassay may be due to inaccuracies in the radioassay as a crude measurement of total TK levels as the radioassay measures activity of the dimerized enzyme (29) or minute deviations in ELISA protein concentrations. The relatedness of the two Mab’s epitopes to each other for minute differences was compared as TK is a sensitive enzymes easily degraded form Its active 3-D configuration {Figure 3).
Even though substantial evidences point to the prognostic value of TK, it has not been applied in the management of breast cancer. This may be due to the concerns associated with the TK radioassay. Currently TK assays are based upon the phosphorylation of radioactive (3H or 1251) nucleosides by active enzymes (27, 29). Due to the nature of the assays, many factors discourage their practical application. First, the results of the radioassay only represent the active enzyme level and not necessarily the total amount of TK present. Second, since enzyme activity is highly sensitive to pH and temperature changes, slight deviations can produce inaccurate and irreproducible results. Furthermore, proper storage, handling, and disposal of radioactive materials, along with the availability of scintillation counters, are potential concerns for the practical application of these assays in a clinical setting. In contrast, an immunoassay measurement of TK would be simpler and more practical and efficient. Since ELISA is a standard clinical laboratory technique, the practicality and accuracy of an immunoassay may facilitate the use of TK as a prognostic marker for the management of cancer.
As tumor screening has reduced the fatality of breast neoplasms by 30%, pre-cancer TK levels could be taken through regular blood samples as a monitoring tool. As the amount needed in an immunoassay is small (<< I ml), levels be could determined from blood taken for other reasons and eventually using only a drop of blood from a pricked finger. As well, the monoclonal antibody used in the immunoassay may be used to stain tumor samples; which results could prove useful in predicting recurrence as previously shown in serums (6). Preliminary studies using the monoclonal antibody to stain tumor samples have commenced in our laboratory, and further emphasize the significance of TK levels in cancer.
References
- Kit S., and Dubbs D. R. “Acquisition of thymidine kinase activity by herpes simplex infected mouse fibroblast cells.” Biochem. Biophys. Res. Comm. 11:55-59,1963.
- McKenna R. G., O’Neill K. L., Abram W. P., and Hannigan B. M. “Thymidine kinase activities in mononuclear Leukocytes an serum from breast cancer patients.” Br. J. Cancer 57:619-622, 1988.
- Robertson,]. F. R., O’Neill, K. L., McKenna, P. G. and Blarney. R. W. “Serum thymidine kinase-a tumor marker in advanced Breast cancer.” !k.__L Cancer 60:487,1989.
- Galloux, H., javre, ]. L .. Guerin, D., Samperez, S .. and Jouan, P. “Prognostic value of fetal thymidine kinase measurements in breast cancer.” C. R. Acad. Sci. III 306:89-92, 1988.
- Borzenko, B. G. “Use of enzyme test in chemotherapy of patients with cancer of the breast.” Kiln. Med. Mask. 68:66-69, 1990.
- O’Neill, K. L., Hoper, M., and Odling-Smee, G. “Can thymidine kinase levels In breast tumors predict disease recurrence?” I. Natl Cancer Inst. 84:1825- 1828, 1992.
- Ellims, P. H., and Vander Weyden, M. B. “Thymidine kinase isoenzymes in human malignant lymphoma.” Cancer Res. 41:691-695, 1981.
- Ellims, P. H., Gan, T. E., and Vander Weyden M. B. “Thymidine kinase isozymes in chronic lymphocytic leukemia.” Br. I. Haem. 49:479-481, 1981.
- Ellims, P. H., Cant T. E., Medley, G., and Van der Weyden, M. B. “Prognostic relevance of thymidine kinase isozymes in adult non-Hodgkin’s lymphoma.” Blood 58:926-930, 1981.
- Gronowitz,]. S., Hagberg, H., Kallander,C. F. R., and Simonsson, B. “The use
of serum deoxythymidine kinase as a prognostic marker and in the
monitoring of patients with non-Hodgkin’s lymphoma.” lk,__,L_Cancer
47:487495,1983. - Hagberg, H., Gronowltz, J. S., Killander, A., eta!. “Serum thymidine kinase
in acute leukemia.” Br. I. Cancer 49:537-540, 1984. - Kallander, C. F. R., Simonsson, B. Hagberg, H., and Gronowitz, J. S. “Serum
Deoxythymidine kinase gives prognostic information in chronic lymphocytic
leukemia.” Cancer 54:2450-2455, 1984. - Vertongen, F., Fondu, P., Vanden Heule, B., Cauchie, C., and Mandelbaum, L.
M. “Thymidine kinase and thymidine phosporylase activities in various
types of leukemia and lymphoma.” Tumor Bioi. 5:303-311, 1984. - Eriksson, B., Hagberg, H., Glimelius, B., Sundstrom, C., Gronowitz]. S., and
Kallander, C. “Serum thymidine kinase as a prognostic marker In
Hodgkin’s disease.” Acta Rad. Oneal. 24:167-171, 1985. - O’Neill, K. L., Abram, W. P., and McKenna, P. G. “Serum thymidine kinase
levels in cancer patients.” Jr. I. Med. Science 155:272-274, 1986. - Russo, S. A., Harris, M. B., and Greengard, 0. “Lymphocyte thymidine kinase
and treatment response in acute lymphocytic leukemia.” Leuk. _Res.
11:149-154,1987. - Hagberg, H., Glimelius, B., Gronowitz,]. S., Killander, A., Kallander, C. F. R.,
and Schroder, T. “Biochemical markers in non-Hodgkin’s lymphoma
stages 3 and 4 and prognosis-A multivariate analysis.” Scan. I. Haemat.
33:5967,1984. - Sakamoto, S., Sagara, T., lwama, T., Kawasaki, T., and Okamoto, R. “Increased
activities of thymidine kinase isozyme in human colon polyp and carcinoma.”
Carcinogenesis 6:917-919, 1985. - Mevin, G. B., O’Neill, K. L., and McKenna, P. G. “Thymidine kinase activities
in pleural effusions.” Br. J. Cancer 58: 252, 1988. - Stewart, L. H., O’Neill, K. L., McKelvey, V. ]., Gillespie, E., Johnston, S. R., and
McKenna, P. G. “Why do most primary bladder neoplasms occur around
the ureteric orifices?” Br. I. Urology 71:34-37, 1991. - Stewart, L. H., O’Neill, K. L., McKelvey, V. J., Gillespie, E., Johnston, S. R.,
McKenna, P. G. “Thymidine kinase activities in bladder cancer.” I. Blamed.
Sci. 3(1):13-17,1992. - O’Neill, K. L., Abram, P., and McKenna, P. G. “Deoxythymidine kinase activities
in sera from cancer and non-cancer patients.” Tumor Bioi. 7:236,
1986. - O’Neill, K. L., McKenna, P. G., Hannigan, B. M., and Abram, W. P. “Isozymes
ofleukocyte thymidine kinase in malignancy.” Bioi. chem. 367:238, 1986. - O’Neill, K. L., Abram, P., Hannigan, B., and McKenna, P. G. “Elevated serum
and mononuclear leukocyte thymidine kinase activities in patients with
cancer.” Jr. Med. I. 80:264-265, 1987. - Weissman, S. M., Smellie, R. M. S, and Paul,]. “Studies on the biosynthesis of
deoxyribonucleic acid by extracts of mammalian cell: The phosphorylation
of thymidine.” Biochem. Biophys. Acta 45:101-110, 1960. - Romain S, Javre JL, Sam perez S, et a!. “Prognostic value of thymidine kinase
in cancer of the breast.” Bull Cancer (Paris) 1990:77:973-83. - Bradford, M. “A rapid and sensitive method for the quantitation of microgram
quantities of protein, utilizing the principle of protein binding.”
Ann. Biochem. 72:248, 1976. - Gronowitz,]. S., and Kallander, C. F. R. “Optimized assay for thymidine
kinase and its application for the detection of antibodies against herpes
simplex type 1- and 2-tnduced thymidine kinase.” Infect. Immun. 29:425-
434, 1980. - O’Neill, K. L., Grigsby, R. V., and Fairbairn, D. W. “A review of thymidine kinase: the future in breast cancer prognosis.” The Breast. 4:2:79·83,
1995.