TUMOR ASSOCIATED TRYPSIN INHIBITOR AS A PROGNOSTIC FACTOR IN RENAL CELL CARCINOMA

0022-5347/01/1653-0959/0 THE JOURNAL OF UROLOGY® Copyright © 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.®

Vol. 165, 959 –962, March 2001 Printed in U.S.A.

TUMOR ASSOCIATED TRYPSIN INHIBITOR AS A PROGNOSTIC FACTOR IN RENAL CELL CARCINOMA ANNUKKA PAJU, JAN JACOBSEN, TORGNY RASMUSON, ULF-HÅKAN STENMAN ¨ AND BORJE LJUNGBERG From the Department of Clinical Chemistry, Helsinki University Central Hospital, Helsinki, Finland, and Departments of Urology and Andrology and Oncology, Umeå University, Umeå, Sweden

ABSTRACT

Purpose: We analyzed the prognostic significance of pretreatment serum tumor associated trypsin inhibitor in renal cell carcinoma. Materials and Methods: Serum samples were obtained before surgery from 188 patients who underwent radical nephrectomy for renal cell carcinoma. Median followup of living patients was 8.5 years. Serum tumor associated trypsin inhibitor was measured by a time resolved immunofluorometric assay. Statistical analysis was performed using the Kaplan-Meier method, log rank and stratified log rank tests. Results: Preoperatively serum tumor associated trypsin inhibitor was elevated with a cutoff 16 ␮g./l. in 48% of the patients with normal serum creatinine. The concentration in patients with cancer was significantly higher than in controls (p ⬍0.0001). The serum level correlated with clinical stage and nuclear grade. Patients with an elevated level had significantly shorter survival time than those with a normal level (p ⫽ 0.005). Stratified log rank test demonstrated that tumor associated trypsin inhibitor was a prognostic factor independent of stage and grade in all patients as well as in those with nonmetastatic disease. Conclusions: Increased preoperative serum tumor associated trypsin inhibitor was associated with poor survival in renal cell carcinoma. The serum level was an independent prognostic variable. Preoperative serum tumor associated trypsin inhibitor appears to be a useful predictive factor that may be used to identify patients at increased risk of aggressive disease. KEY WORDS: kidney; carcinoma, renal cell; trypsin inhibitors; prognosis

Renal cell carcinoma is the most common malignancy of the kidney. Recent advances in molecular genetic analysis have led to the recognition of 5 distinct types of renal cell carcinoma, including conventional or nonpapillary, chromophobe, collecting duct, papillary and unclassified.1 The prognosis depends on tumor type with the best prognosis in chromophobe renal cell carcinoma and the worst in collecting duct carcinoma.2 Whatever the histological type the extent of tumor spread or stage and nuclear grade are considered the main prognostic factors.3, 4 However, in subsets of patients with a given disease stage there are often large variations in clinical course and survival time.3, 5 Thus, there is a need for prognostic factors for identifying patients with high risk disease who may benefit from additional treatment modalities. Others have focused on the evaluation of new markers. The prognostic value of cell proliferation markers and p53 mutation was recently investigated but they were not independent prognostic factors.6 Intratumoral microvessel density provided independent prognostic information in one study7 but not in another.6 To our knowledge no specific serum markers are available but interleukin-10, neuron specific enolase and tumor necrosis factor-␣ reportedly have prognostic value.8 –10 Tumor associated trypsin inhibitor is a 6 kDa. peptide that is identical to pancreatic secretory trypsin inhibitor.11 It is expressed by several tumors and cancer cell lines.12 As a serum marker, tumor associated trypsin inhibitor is especially useful in mucinous ovarian cancer13 and it is also a prognostic indicator in stage III epithelial ovarian cancer.14

It has been suggested that tumor associated trypsin inhibitor may also be used as a serum marker for renal cell carcinoma.15 We have recently reported that it is expressed by normal and malignant renal tissue as well as by several renal cancer cell lines.16 We evaluated the prognostic information provided by tumor associated trypsin inhibitor in patients with renal cell carcinoma in regard to established prognostic factors. MATERIALS AND METHODS

Data in this retrospective analysis were collected from 188 patients with renal cell carcinoma who had undergone radical nephrectomy at University Hospital of Umeå, Sweden between 1983 and 1995 and for whom serum samples were available. The study included 115 men and 73 women with a mean age of 65 years (range 25 to 85). After informed consent was obtained serum samples were collected before surgery and stored at ⫺80C until analysis. Staging procedures included physical examination, chest radiography, ultrasonography and computer tomography. Patients with skeletal symptoms or elevated serum alkaline phosphatase were assessed by bone scintigraphy. After nephrectomy all cases were followed with clinical and radiological examinations. At the latest followup 55 of 188 patients were alive at a median followup of 102 months (range 34 to 176). A total of 133 patients had died, including 101 of renal cell carcinoma and 32 of other causes. We excluded 17 of the 188 patients from statistical analyses because serum creatinine data were not available. Table 1 shows the characteristics of patients with normal serum creatinine. We obtained serum from 14 male and 70 female apparently healthy volunteers 20 to 59 years old (mean age 40), who served as a control group. Tumor

Accepted for publication September 5, 2000. Supported by Lions Cancer Research Foundation of Umeå University, Finnish Cancer Foundation, Finnish Academy of Sciences, Sigrid Juselius Foundation, Helsinki University Central Hospital and University of Helsinki. 959

960

TUMOR ASSOCIATED TRYPSIN INHIBITOR IN RENAL CELL CARCINOMA

TABLE 1. Characteristics of 158 patients with renal cell carcinoma and normal serum creatinine Mean age (range) No. sex: M F No. clinical outcome: Alive Dead of Ca Dead of other cause No. tumor stage: I II III IV No. tumor grade: 1 2 3 4 No. tumor type: Conventional Papillary Chromophobe Not available No. DNA ploidy: Diploid Aneuploid Not available No. venous invasion: Absent Present

65 (25–85)

were censored at the last clinical followup and others were censored if death resulted from unrelated disease. All tests were 2-sided with statistical significance considered at p ⫽ 0.05.

93 65 50 83 25 48 22 41 47 3 37 80 38 124 25 6 3 41 107 10 113 45

stage, nuclear grade and DNA ploidy were determined according to the 1997 TNM stage classification,17 and Skinner18 and Ljungberg19 et al, respectively. Tumor associated trypsin inhibitor in serum was determined by a modification of an previously described time resolved immunofluorometric assay using a combination20 of mAb 6E8 as the catcher antibody on the solid phase and an Eu labeled mAb 11B3 as the tracer. Fluorescence was measured in a research fluorometer. With a sample volume of 25 ␮l. in a total assay volume of 225 ␮l. the assay detection limit was 0.15 ␮g./l. For serum tumor associated trypsin inhibitor the upper reference limit, defined as the 97.5th percentile, was 16 ␮g./l. Intra-assay and interassay coefficients of variation in the concentration range of 1.5 to 450 ␮g./l. were 3.0% to 8.6% and 13% to 14%, respectively.20 Serum creatinine was measured by a routine method at the clinical chemistry laboratory at University Hospital. The Mann-Whitney U test was performed to compare differences in tumor associated trypsin inhibitor in patients with various disease stages and grades, and in those with serum creatinine less and greater than 125 ␮mol./l. Survival curves were constructed by the Kaplan-Meier method and we compared survival times with the log rank test. The independence of serum tumor associated trypsin inhibitor as a predictor of survival was analyzed by the stratified log rank test adjusted for stage and grade with the statistical end point of overall survival, as measured from the date of nephrectomy to the date of death or the latest followup. Surviving patients

RESULTS

Preoperatively the serum creatinine value was available for 171 patients. Median serum creatinine was 91 ␮mol./l. (range 45 to 353). Serum creatinine was elevated with a cutoff of 125 ␮mol./l. in 13 patients (8%), who had significantly higher serum tumor associated trypsin inhibitor than those with serum creatinine less than 125 ␮mol./l. (p ⬍0.001). Because impaired renal function increases serum tumor associated trypsin inhibitor,21 patients with a serum creatinine of greater than 125 ␮mol./l. were excluded from further analysis. Serum tumor associated trypsin inhibitor was elevated with a cutoff of 16 ␮g./l. in 76 patients with renal cell carcinoma (48%) and in 4 controls (5%). The median concentration in renal cell carcinoma was 15 ␮g./l. (range 6.1 to 1,140), which was significantly higher than in controls (11 ␮g./l, range 6.6 to 38, p ⬍0.0001). We noted no significant difference in tumor associated trypsin inhibitor in males and females, in the various cancer and control age groups, or in the various renal cell carcinoma tumor types. However, no conclusion regarding chromophobe tumors may be drawn because only 6 patients were evaluated. Tumor associated trypsin inhibitor was significantly greater in patients with aneuploid than diploid tumors and those with versus without thrombus invasion into the vena cava or vena renalis (p ⫽ 0.04 and 0.004, respectively). The serum level correlated with stage and grade (table 2). The concentration was significantly less in stages I and II than in stage IV disease but the difference in stage I or II versus stage III disease was not significant (p ⫽ 0.01 and 0.5, respectively). Grades 1 and 2 disease were associated with a significantly lower concentration than grade 4 disease (p ⫽ 0.003, table 2), while the difference in grade 1 or 2 versus grade 3 tumors approached statistical significance (p ⫽ 0.06). Univariate analysis revealed that survival was strongly associated with clinical stage and nuclear grade (p ⬍0.0001, table 3). Patients with elevated serum tumor associated trypsin inhibitor with a cutoff of 16 ␮g./l. also had significantly shorter survival time than those with a lower level (p ⫽ 0.005, A of figure and table 3). A significant survival benefit was observed in nonmetastatic stages I to III disease (p ⫽ 0.02, B of figure). The stratified log rank test adjusted for stage and grade showed that the preoperative serum level was a prognostic factor independent of stage and grade in all patients (p ⫽ 0.03 and 0.02, respectively) and in those with nonmetastatic disease (p ⴝ 0.03 and p ⴝ 0.04, respectively). When 30 to 50 ␮g./l. were used as the cutoff, we observed a difference in survival time in patients with metastatic stage IV tumor (p ⬍0.004, C of figure and table 3). However, at these higher cutoffs tumor associated trypsin inhibitor was

TABLE 2. Serum tumor associated trypsin inhibitor versus tumor grade and stage No. Pts.

No. Elevated (%)

Stage: I, II 70 27 (39) III 41 22 (54) IV 47 27 (57) Grade: 1, 2 40 15 (38) 3 80 37 (46) 4 38 24 (63) * Mann-Whitney U test p ⫽ 0.05 and 0.01 versus III and IV, respectively. † Mann-Whitney U test p ⫽ 0.06 and 0.003 versus III and IV, respectively.

␮g./l. Tumor Associated Trypsin Inhibitor Range

Median

Interquartile Range

6.2–493 8.2–155 6.1–1,140

13.8* 17.0 22.4

11.1–21.0 10.2–39.5 12.2–38.1

8.5–493 8.5–1,440 6.2–1,140

13.1† 14.1 23.2

9.9–20.4 11.1–24.5 11.7–67.2

TUMOR ASSOCIATED TRYPSIN INHIBITOR IN RENAL CELL CARCINOMA

961

TABLE 3. Univariate analysis of variables predicting survival by the log rank test Tumor associated trypsin inhibitor cutoff (␮g./l.): 16 30–50 Grade Stage

All Stages

Stages I-III

Stage IV

0.005 ⬍0.0005 ⬍0.0001 ⬍0.0001

0.02 ⬍0.006

0.18 ⬍0.004

not an independent prognostic factor on the stratified log rank test. DISCUSSION

Renal cell carcinoma has a varying histological appearance and clinical behavior. Pathological stage and nuclear grade are indisputably valuable for determining the prognosis in renal cell carcinoma.3, 4 In our study they were also significant prognostic variables. However, disease stage and grade are usually available only after surgery. Therefore, markers that accurately predict the outcome of renal cell carcinoma before therapy would be valuable. Several serum markers have been evaluated but only a few have prognostic value in renal cell carcinoma.8 –10 In our study elevated preoperative serum tumor associated trypsin inhibitor was associated with poor survival. Patients with an elevated serum concentration had significantly shorter survival time than those with a normal level (p ⫽ 0.005). The stratified log rank test revealed that tumor associated trypsin inhibitor independently predicted the outcome in all patients as well as in those without metastatic disease. These results suggest that determining tumor associated trypsin inhibitor in serum may aid in the selection of candidates for nephron sparing surgery. When 30 to 50 ␮g./l. were used as the cutoff, the outcome of metastatic disease was also predictable. Tumor associated trypsin inhibitor is expressed by several types of tumor as well as by several tumor cell lines derived from various cancer tissues.12 We have recently shown that it is expressed in histologically normal kidneys, renal cell carcinoma and several renal cancer cell lines.16 Thus, tumor production of tumor associated trypsin inhibitor may explain the elevated serum level in renal cell carcinoma. As a marker, tumor associated trypsin inhibitor has been most useful for monitoring mucinous ovarian cancer, in which it is often elevated in early disease.13 It is also a prognostic factor in stage III ovarian cancer of various types.14 The association of high serum concentrations of tumor associated trypsin inhibitor with poor prognosis is seemingly paradoxical. However, in the pancreas tumor associated trypsin inhibitor-pancreatic secretory trypsin inhibitor is thought to protect the gland against the premature intracellular activation of trypsinogen,12 which may be its role also in tumors. In ovarian cancer elevated levels are associated with tumor trypsinogen expression.12 Trypsinogen may contribute to the aggressiveness of cancer by facilitating the invasion of cancer cells directly by degrading the extracellular matrix22 or by activating proteinases associated with tumor invasion.23 Because trypsinogen is also expressed in kidneys,24 it is tempting to speculate that elevated serum tumor associated trypsin inhibitor in patients with renal cell carcinoma reflects trypsinogen expression. An elevated level may also occur in nonmalignant diseases, for example severe inflammatory disease.12 Thus, inflammation associated with extensive disease may also contribute to the elevation in some cases. Tumor associated trypsin inhibitor is cleared from the circulation by renal excretion and it is partially reabsorbed by the renal tubules. Thus, it is also increased in patients with markedly impaired renal function.21

Kaplan-Meier cancer specific survival according to preoperative serum tumor associated trypsin inhibitor (TATI) concentration in patients with renal cell carcinoma. A, all stages, cutoff 16 ␮ g./l. B, stages I to III, cutoff 16 ␮g./l. C, stage IV, cutoff 30 ␮g./l.

Meria et al previously reported that tumor associated trypsin inhibitor is a more sensitive marker of renal cell carcinoma than carcinoembryonic antigen, cancer antigens 15-3, 125 and 19-9, and ferritin.15 We also evaluated the sensitivity of tumor associated trypsin inhibitor for detecting renal cell carcinoma. Overall its sensitivity was 48%, increasing from 39% to 57% with increasing disease stage. This finding suggests that about half of the tumors expressed tumor associated trypsin inhibitor. Based on the correlation of the

962

TUMOR ASSOCIATED TRYPSIN INHIBITOR IN RENAL CELL CARCINOMA

level and disease grade, expression seems to be associated with aggressive disease and an adverse outcome. Sensitivity in the study of Meria et al was 69%, which is higher than in our study. This difference may be the result of excluding patients with elevated serum creatinine from our study. Furthermore, Meria et al suggested that tumor associated trypsin inhibitor may be used for following patients treated with radical nephrectomy. However, its use for monitoring of renal cell carcinoma remains to be evaluated in a larger study. CONCLUSIONS

Our results indicate that preoperative serum tumor associated trypsin inhibitor is a useful predictive factor in patients with renal cell carcinoma and in the subgroup without metastatic disease. Tumor associated trypsin inhibitor may be used as an aid for selecting treatment strategies and for stratification in randomized studies. REFERENCES

1. Kovacs, G., Akhtar, M., Beckwith, B. J. et al: The Heidelberg classification of renal cell tumours. J Pathol, 183: 131, 1997 2. Gelb, A. B.: Renal cell carcinoma. Current prognostic factors. Cancer, 80: 981, 1997 3. Fuhrman, S. A., Lasky, L. C. and Limas, C.: Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol, 6: 655, 1982 4. Medeiros, L. J., Gelb, A. B. and Weiss, L. M.: Renal cell carcinoma. Prognostic significance of morphologic parameters in 121 cases. Cancer, 61: 1639, 1988 5. Reis, M. and Faria, V.: Renal carcinoma. Reevaluation of prognostic factors. Cancer, 61: 1192, 1988 6. Gelb, A. B., Sudilovsky, D., Wu, C. D. et al: Appraisal of intratumoral microvessel density, MIB-1 score, DNA content, and p53 protein expression as prognostic indicators in patients with locally confined renal cell carcinoma. Cancer, 80: 1768, 1997 7. Nativ, O., Sabo, E., Reiss, A. et al: Clinical significance of tumor angiogenesis in patients with localized renal cell carcinoma. Urology, 51: 693, 1998 8. Wittke, F., Hoffmann, R., Buer, J. et al: Interleukin 10 (IL-10): an immunosuppressive factor and independent predictor in patients with metastatic renal cell carcinoma. Br J Cancer, 79: 1182, 1999 9. Rasmuson, T., Grankvist, K., Roos, G. et al: Neuroendocrine differentiation in renal cell carcinoma. Evaluation of chromogranin A and neuron-specific enolase. Acta Oncol, 38: 623, 1999

10. Dosquet, C., Coudert, M.-C., Lepage, E. et al: Are angiogenic factors, cytokines, and soluble adhesion molecules prognostic factors in patients with renal cell carcinoma? Clin Cancer Res, 3: 2451, 1997 11. Huhtala, M. L., Pesonen, K., Kalkkinen, N. et al: Purification and characterization of a tumor-associated trypsin inhibitor from the urine of a patient with ovarian cancer. J Biol Chem, 257: 13713, 1982 12. Stenman, U.-H., Koivunen, E. and Itkonen, O.: Biology and function of tumor-associated trypsin inhibitor. Scand J Clin Lab Invest, suppl., 51: 5, 1991 13. Halila, H., Lehtovirta, P. and Stenman, U.-H.: Tumourassociated trypsin inhibitor (TATI) in ovarian cancer. Br J Cancer, 57: 304, 1988 14. Venesmaa, P., Stenman, U.-H., Forss, M. et al: Pre-operative serum level of tumour-associated trypsin inhibitor and residual tumour size as prognostic indicators in stage III epithelial ovarian cancer. Br J Obstet Gynaecol, 105: 508, 1998 15. Meria, P., Toubert, M. E., Cussenot, O. et al: Tumour-associated trypsin inhibitor and renal cell carcinoma. Eur Urol, 27: 223, 1995 16. Lukkonen, A., Lintula, S., von Boguslawski, K. et al: Tumorassociated trypsin inhibitor in normal and malignant renal tissue and in serum of renal-cell carcinoma patients. Int J Cancer, 83: 486, 1999 17. Sobin, L. H. and Wittekind, C. H.: TNM Classification of Malignant Tumors, 5th ed. International Union Against Cancer. New York, Wiley-Liss, p. 180, 1997 18. Skinner, D. G., Colvin, R. B., Vermillion, C. D. et al: Diagnosis and management of renal cell carcinoma. A clinical and pathological study of 309 cases. Cancer, 28: 1165, 1971 19. Ljungberg, B., Mehle, C., Stenling, R. et al: Heterogeneity in renal cell carcinoma and its impact on prognosis: a flow cystometric study. Br J Cancer, 74: 123, 1996 20. Osman, S., Turpeinen, U., Itkonen, O. et al: Optimization of a time-resolved immunofluorometric assay for tumor-associated trypsin inhibitor (TATI) using the streptavidin-biotin system. J Immunol Methods, 161: 97, 1993 21. Tramonti, G., Donadio, C., Ferdeghini, M. et al: Serum tumourassociated trypsin inhibitor (TATI) and renal function. Scand J Clin Lab Invest, 56: 653, 1996 22. Koivunen, E., Ristima¨ki, A., Itkonen, O. et al: Tumor-associated trypsin participates in cancer cell-mediated degradation of extracellular matrix. Cancer Res, 51: 2107, 1991 23. Sorsa, T., Salo, T., Koivunen, E. et al: Activation of type IV procollagenases by human tumor-associated trypsin-2. J Biol Chem, 272: 21067, 1997 24. Koshikawa, N., Hasegawa, S., Nagashima, Y. et al: Expression of trypsin by epithelial cells of various tissues, leukocytes, and neurons in human and mouse. Am J Pathol, 153: 937, 1998