ANTICANCER RESEARCH 25: 1765-1772 (2005)
CYFRA 21.1 in Patients with Cervical Cancer: Comparison with SCC and CEA RAFAEL MOLINA1, XAVIER FILELLA1, JOSE M. AUGÉ1, ELVIRA BOSCH1, AURELI TORNE2, JAUME PAHISA2, JOSE A. LEJARCEGUI2, ANGELS ROVIROSA3, BEGOÑA MELLADO4, JAUME ORDI5 and ALBERT BIETE3 1Unit
of Cancer Research (Laboratory of Clinical Biochemistry), 2Institute of Gynecology and Obstetrics, 3Unit of Radiotherapy, 4Unit of Clinical Oncology and 5Laboratory of Pathology, School of Medicine, Hospital Clinic, Barcelona, Spain
Abstract. The serum levels of CYFRA 21.1, CEA and SCC were prospectively determined in 156 patients diagnosed with carcinoma of the uterine cervix from 1995 to 2003. Histology revealed squamous cancer in 119 patients, adenocarcinoma in 25 patients and adenosquamous carcinoma in the remaining 12 patients. We considered 3.3 ng/ml, 5 ng/ml and 2 ng/ml as the upper limits of normality for CYFRA 21.1, CEA and SCC, respectively. The sensitivity of CYFRA 21.1, CEA and SCC was 26%, 25% and 43%, respectively, at diagnosis. SCC was clearly related to tumor histology, with significantly higher levels in squamous tumors than in other histological types (p83%) of parametrial invasion in squamous tumors. Likewise, pretreatment SCC and CYFRA 21.1 serum levels were of prognostic value, with a shorter DFS and OS in patients with abnormal levels. Multivariate analysis indicated that stage, histological grade and parametrial invasion were independent prognostic factors, but not tumor markers. In conclusion, SCC is the tumor marker of choice in squamous tumors and the addition of CEA or CYFRA 21.1 does not significantly increase the sensitivity obtained by using SCC alone.
Correspondence to: Dr. R. Molina, Unit of Cancer Research, Laboratory of Clinical Biochemistry, Hospital Clinic, Barcelona 08036, Spain. e-mail:
[email protected] Key Words: SCC, CEA, cervical cancer, tumor markers, prognosis.
0250-7005/2005 $2.00+.40
In industrialized countries cervical cancer is the second cause of death by cancer in women (1). Cervical cancer has different histological types, with squamous cancer being the most frequent histology at about 90%. The other types are mostly adenocarcinomas or adenosquamous tumors. Squamous cell carcinoma antigen, SCC, is the marker of choice in squamous tumors. Its sensitivity ranges from less than 30% for Stage I disease to over 90% for Stage IV disease (2-4). As a diagnostic tool it is, therefore, not useful in view of these sensitivities, particularly in the early stages of the disease. However, SCC is reportedly useful in the prognosis, early diagnosis of recurrence and monitoring of patients with cervical cancer (2, 5-12). To increase the sensitivity of SCC in patients with squamous cell carcinoma, other markers have been evaluated with contradictory results (11, 13, 14). The usefulness of tumor markers in other histological types is unclear, although several markers have been suggested, such as carcinoembryonic antigen (CEA), CA 125, CA 19.9, hCG beta core fragment and cytokeratins (TPA and CYFRA 21.1) (15-18). CEA and CA 125 are the most frequently studied tumor markers, with sensitivities ranging from 25-60% (19). Based on the available evidence, the use of these markers for cervical cancer cannot be recommended at present. CYFRA 21.1 is a tumor marker that measures the cytokeratin 19 fragment and was first suggested as a tumor marker for lung cancer, with no clear relationship to the histological type, although it is accepted that its highest values are found in squamous tumors (20). Moreover, abnormal levels of this marker have been reported in some benign conditions (renal failure, liver cirrhosis, benign lung diseases) and in most epithelial tumors (21). The possible CYFRA 21.1 utility in patients with cervical cancer has recently been suggested (22-24). However, the advantages of its use alone or in combination with other tumor markers have not been well established.
1765
ANTICANCER RESEARCH 25: 1765-1772 (2005) Table I. Tumor marker serum levels in patients with cervical cancer subdivided according to tumor histology. Tumor marker
Histology
Sensitivity Mean±SD Median Range ng/ml ng/ml ng/ml
SCC
Squamous 61/119 51% 6.5±11.2 Adenocarcinoma 3/25 12% 0.9±0.8 Adenosquamous 3/12 25% 1.3±0.7
2.1 0.5 1.45
0.1-58 0.2-4.1 0.2-2.4
CEA
Squamous 36/119 30% 5.4±7.9 Adenocarcinoma 3/25 12% 20±57 Adenosquamous 0/12 0% 2±1.1
2.5 2.4 1.7
0.2-50 0.7-252 0.6-4.1
CYFRA Squamous 31/119 26% 4.4±9.1 21.1 Adenocarcinoma 7/25 28% 2.6±2.4 Adenosquamous 3/12 25% 2.6±2
2 1.6 1.9
0.2-73 0.3-11.1 0.9-6.7
The objectives of this study were: i) to evaluate the sensitivity of three tumor markers (CEA, CYFRA 21.1 and SCC) in patients with cervical carcinoma, alone or in combination; ii) to study the relationships among the tumor markers and the most important prognostic factors in this malignancy: stage, tumor size, myometrial invasion, histology, histological grade and nodal involvement; iii) to evaluate the prognostic value of these three markers.
Materials and Methods Serum concentrations of CYFRA 21.1, CEA and SCC were prospectively evaluated in 156 untreated patients with cervical carcinoma, from January 1995 to October 2003. Histology was squamous in 119 patients, adenocarcinoma in 25 patients and adenosquamous in the remaining 12 patients. Clinical staging was established in accordance with FIGO and was stage Ia in 2 patients, stage Ib in 45 patients, stage IIa in 10 patients, stage IIb in 55 patients, stage IIIb in 39 patients and stage IV in 5 patients (25). Treatment consisted of surgical resection in patients with stages I or IIa disease (not bulky), radical radiotherapy (external and brachytherapy) with intention to cure (>70Gy) in patients with stages IIb and IIIa ineligible for surgery, and palliative or postoperative irradiation (2 ng/ml/total
Mean±SD ng/ml
CEA > 5 ng/ml/total
Mean±SD ng/ml
4/45 9% 27/74 37% 6/58 10% 13/36 36% 6/32 19% 25/87 29% 2/22 9% 16/56 29% 13/41 32%
1.9±2.3 5.0±7.8 1.9±1.5 5.6±9.4 2.3±1.9 4.4±7.4 3.0±7.3 4.4±7.7 3.6±4.1
10/45 22% 51/74 69% 18/58 31% 23/36 64% 13/32 41% 48/87 55% 5/22 27% 36/56 64% 20/41 49%
1.5±1.6 9.6±13.0 2.2±2.8 11.0±16.0 1.4±0.5 7.4±12.1 4.6±11.9 7.0±9.4 7.0±13
5/45 11% 31/74 42% 11/58 19% 12/36 33% 7/32 22% 29/87 33% 6/22 27% 20/56 36% 10/41 24%
2.4±2.1 7.2±9.4 3.3±3.4 8.0±12.0 4.8±8.7 5.6±7.6 4.0±4.3 5.7±7.7 6.0±9.5
(p=0.013) and SCC (p=0.09), but not to CEA. It is interesting to point out that abnormal levels of tumor markers suggested a high probability of developing parametrial invasion: SCC 83.6%, CEA 86.1% and CYFRA 21.1 87.1%. Likewise, patients without parametrial invasion and with abnormal levels had a higher proportion of relapse than those with normal levels: SCC 4/10 (40%) versus 4/35 (11.4%), CEA 2/5 (40%) vs 8/32 (25%) and CYFRA 21.1 2/4 (50%) vs 7/41 (17.1%). No relationship was found among these prognostic factors and tumor marker serum levels in adenocarcinomas (data not shown). Figures 1 to 4 show the disease-free survival (DFS) and overall survival (OS) according to pretreatment CYFRA 21.1 and SCC concentrations in squamous tumors. Both tumor markers were prognostic factors in DFS, but only CYFRA 21.1 was so in OS. By contrast, pretreatment CEA serum levels were not of prognostic interest, in either DFS or OS (data not shown). Table IV indicates the univariate and multivariate analysis of the most important prognostic factors in squamous cervical cancer. Parametrial invasion and histological grade were independent prognostic factors in both DFS and OS, and age, diagnosis and stage only in OS.
Discussion Our results indicate that SCC is a useful tumor marker in cervical cancer, with a clear relationship to histological type (mainly in squamous tumors) tumor stage, tumor size, nodal involvement and parametrial invasion. These results are similar to those reported previously by other authors, indicating that SCC is the tumor marker of choice in squamous tumors (3, 5, 6, 9, 13, 14). The sensitivity of CEA and CYFRA 21.1 was lower than that of SCC, but was also related to the most important prognostic factors in this malignancy such as parametrial invasion, nodal involvement and tumor size. Other authors have also reported that the sensitivity of CYFRA 21.1 or CEA was lower than that of SCC, as well as the absence of a relationship of CYFRA 21.1 with histological type (13-15, 18, 23, 24, 26). Gaaranstron et al. (15) reported a similar relationship with parametrial invasion but not with nodal invasion. It is interesting to point out that abnormal pretreatment tumor marker serum levels indicated a high probability of parametrial invasion, which is mainly evaluated clinically and is of great importance in determining tumor stage and treatment. Surgery is the most
1767
ANTICANCER RESEARCH 25: 1765-1772 (2005)
Figure 1. Disease-free survival in patients with squamous cervical cancer subdivided according to pretreatment CYFRA 21-1 serum levels.
Figure 2. Overall survival in patients with squamous cervical cancer subdivided according to pretreatment CYFRA 21-1 serum levels.
commonly used treatment in patients without parametrial invasion and radiotherapy is performed in those with parametrial invasion. Seventy-nine percent of the 73 patients with parametrial invasion had abnormal levels of one tumor
1768
marker or another. Likewise, patients with abnormal levels of these tumor markers had more than an 83% probability of parametrial invasion. It is interesting to note that the patients with abnormal pretreatment tumor marker levels
Molina et al: CYFRA 21.1 in Cervical Cancer
Figure 3. Disease-free survival in patients with squamous cervical cancer subdivided according to pretreatment SCC serum levels.
Figure 4. Overall survival in patients with squamous cervical cancer subdivided according to pretreatment SCC serum levels.
without parametrial invasion had a higher proportion of recurrence than those with normal tumor marker concentrations. In summary, these results clearly indicate that pretreatment tumor marker levels are useful tools in the staging of patients with squamous cervical cancer. CEA and CYFRA 21.1 are useful tumor markers in cervical cancer, but have a lower sensitivity than SCC. The question,
however, is to determine the advantages of their use in combination with SCC in squamous tumors. Our results clearly indicate that the use of CEA and CYFRA 21.1 scarcely increased the sensitivity obtained with SCC alone. Moreover, the only possible clinical application of CEA and CYFRA 21.1 may be in pretreatment evaluation to determine the patients with a higher probability of parametrial invasion (14). Eight of
1769
ANTICANCER RESEARCH 25: 1765-1772 (2005) Table IV. Univariate and multivariate analysis of DFS and OS in 119 patients with squamous cervical cancer. Disease-free survival Parameter Stage Tumor size Histological grade Parametrial invasion Age SCC CEA CYFRA 21.1
Overall survival
Univariate Multivariate Univariate Multivariate 0.004 0.021 0.001 0.001 n.s.s. 0.02 n.s.s. 0.004
n.s.s. n.s.s. 0.003 0.001 n.s.s. n.s.s. n.s.s. n.s.s.
0.03 n.s.s. 0.012 0.0001 n.s.s. n.s.s. n.s.s. 0.03
0.08 n.s.s. 0.004 0.0001 0.032 n.s.s n.s.s. n.s.s.
the 23 patients with normal SCC with parametrial invasion had abnormal CEA and/or CYFRA 21.1. By contrast, SCC was not useful in adenocarcinomas and the combination of CYFRA 21.1 and CEA significantly increased the sensitivity obtained with one of these tumor markers used alone. The relationship among tumor marker serum levels and the most important prognostic factors suggests their possible utility in establishing prognosis in patients with cervical cancer. Several studies have reported that high pretreatment levels of SCC have been found to predict poor outcome in patients with squamous cell cancer (2, 12, 14, 27). In some of these studies, SCC was reported to be an independent prognostic marker (2, 12, 14, 27), i.e. independent of histology and FIGO stage and grade, but in other reports it was not (15). In our experience, SCC and CYFRA 21.1 are prognostic factors in univariate analysis but not in multivariate analysis. Gaaranstron et al. (15) reported that CYFRA 21.1 is not an independent prognostic factor. Discrepancies may be related to the number of patients studied, the length of follow-up, the tumor stage and treatment. Moreover, the relationship of these tumor markers with parametrial invasion, their prognostic value in univariate analysis, as well as the fact that they have the most important characteristics found in prognostic factors, e.g. easy to perform, cheap, reproducible, makes them useful in prognosis. In summary, our results confirm that SCC is the marker of choice in squamous cell carcinoma of the cervix, with a higher sensitivity than CYFRA 21.1 and CEA. The addition of CYFRA 21.1 or CEA does not significantly increase the sensitivity of this marker, but may be useful in the pretreatment evaluation of patients with cervical cancer.
Acknowledgements Supported by Grant: Red Temática del Cáncer, Instituto de Salud Carlos III, nÆ C03/10, Spain.
1770
References 1 Boring CC, Squires TS, Tong T and Montgomery S: Cancer statistics 1994. CA Cancer J Clin 4: 7-26, 1994. 2 Duk JM, De Bruijn H, Groenier KH, Hollema H, ten Hoor KA, Krans M et al: Cancer of the uterine cervix: sensitivity and specificity of serum squamous cell carcinoma antigen determinations. Gynecol Oncol 39: 186-194, 1990. 3 Borras G, Molina R, Xercavins J, Ballesta A and Iglesias X: Squamous cell carcinoma antigen in cervical cancer. Eur J Gynaecol Oncol 13: 414-418, 1992. 4 Kato H, Morioka H, Tsutsui H, Aramaki S and Torigoe T: Value of tumor antigen (TA-4) of squamous cell carcinoma in predicting the extent of cervical cancer. Cancer 50: 1294-1296, 1982. 5 Leminen A, Alftan H, Stenman UH and Lehtovirta P: Chemotherapy as initial treatment for cervical carcinoma: clinical and tumor marker response. Acta Obstet Gynecol Scand 71: 293-297, 1992. 6 Ngan HY, Cheng GT, Yeung WS, Wong LC and Ma HK: The prognostic value of TPA and SCC in squamous cell carcinoma of the cervix. Gynecol Oncol 52: 63-68, 1994. 7 Ngan HYS, Chan SYM, Wong LC, Choy DTK and Ma HK: Serum squamous cell carcinoma antigen assays in the monitoring of radiotherapy response in carcinoma of the cervix. Gynecol Oncol 37: 260-263, 1990. 8 Rose PG, Baker S, Fournier L, Nelson BE and Huner RE: Serum squamous cell carcinoma antigen levels in invasive cervical cancer: prediction of response and recurrence. Am J Obstet Gynecol 10: 942-946, 1993. 9 Avåll-Lundqvist EH, Sjovall K, Nilsson BR and Eneroth PH: Prognostic significance of pretreatment serum levels of squamous cell carcinoma antigen and CA 125 in cervical carcinoma. Eur J Cancer 28A: 1695-1702, 1992. 10 Chou CY, Wang ST, Kuo HC, Tzeng CC and Yao BL: Serum level of squamous cell carcinoma antigen and tumor size are useful to identify preoperatively patients at high risk of cervical cancer. Cancer 74: 2497-2501, 1994. 11 De Bruijn HW, Duk JM, van der Zee AG, Pras E, Willemse PH, Boonstra H et al: The clinical value of squamous cell carcinoma antigen in cancer of the uterine cervix. Tumor Biol 19: 505-516, 1998. 12 Scambia G, Benedette Panici P, Foti E, Amoroso M, Salerno G et al: Squamous cell carcinoma antigen: significance and role in the monitoring of neo-adjuvant chemotherapy response in cervical cancer. J Clin Oncol 12: 2309-2316, 1994. 13 Kornafel J and Wawrzkiewicz M: Evaluation of diagnostic usefulness of CEA, hGC and SCC antigens in cervical cancer patients. Eur J Gynaecol Oncol 39: 358-363 1989. 14 Molina R, Filella X, Lejarcegui JA, Pahisa J, Torné A, Rovirosa A et al: Prospective evaluation of squamous cell carcinoma and carcinoembryonic antigen as prognostic factors in patients with cervical cancer. Tumor Biol 24: 109-164, 2003. 15 Gaarenstroom KN, Bonfrer JMG, Kenter GG, Korse CM, Hart AAM et al: Clinical value of pre-treatment serum CYFRA 21-1, tissue polypeptide antigen and squamous cell carcinoma antigen levels in patients with cervical cancer. Cancer 76: 807-813, 1995. 16 Duk MJ, Aalders JG, Fleuren GJ, Krans M and de Bruijn HWA: Tumor markers CA 125, squamous cell carcinoma antigen, and carcinoembryonic antigen in patients with adenocarcinoma of the uterine cervix. Obstet Gynec 73: 661-667, 1989.
Molina et al: CYFRA 21.1 in Cervical Cancer
17 Borras G, Molina R, Xercavins J, Ballesta AM and Iglesias J: Tumor antigens CA 19.9, CA 125 and CEA in carcinoma of the uterine cervix. Gynecol Oncol 57: 205-211, 1995. 18 Dodd JK, Henry RJW, Tyler JPP and Houghton CRS: Cervical carcinoma: a comparison of four potential biochemical tumor markers. Gynecol Oncol 32: 248-252, 1989. 19 Duk JM, de Bruijn H, Klaas KH, Fleuren GJ and Aalders JG: Adenocarcinoma of the uterine cervix: prognostic significance of pre-treatment serum CA125, squamous cell carcinoma antigen and carcinoembryonic antigen levels in relation to clinical and histopathologic tumor characteristics. Cancer 65: 1830-1837, 1990. 20 Molina R, Filella X, Auge JM, Rifa J, Moreno V, Canals E et al: Tumour markers (CEA, CA125, CYFRA 21-1, SCC and NSE) in patients with non-small cell lung cancer as aid in histological diagnosis and prognosis: comparison with the main clinical and pathological prognostic factors. Tumor Biol 24: 209218, 2003. 21 Molina R, Agustí C, Filella X, Jo J, Joseph J, Giménez N and Ballesta AM: Study of a new tumor marker, CYFRA 21-1 in malignant and nonmalignant diseases. Tumor Biol 15: 318321, 1994. 22 Ferdeghini M, Gadducci A, Annicchiarico C, Prontera C, Melagnino G, Castellani C et al: Serum CYFRA 21-1 assay in squamous cell carcinoma of the cervix. Anticancer Res 13: 1841-1844, 1993.
23 Pras E, Willemse PH, Canrinus AA, de Bruijn HWA, Sluiter WJ, ten Hoor KA et al: Serum squamous cell carcinoma antigen and CYFRA 21-1 in cervical cancer treatment. Int J Rad Oncol Biol Phys 52: 23-32, 2002. 24 Yazigi R, Castillo R, Aliste G, Garrido J, Opazo A, Prado S et al: CYFRA 21-1 marker in carcinoma of the cervix. Int J Gynecol Cancer 10: 203-206, 2000. 25 Creaseman WT: New gynecologic cancer staging. Gynecol Oncol 38: 157-158, 1995. 26 Suzuki Y, Nakano T, Ohno T, Abe A, Morita S and Tsujii H: Serum CYFRA 21-1 in cervical cancer patients treated with radiation therapy. J Cancer Res Clin Oncol 126: 332-336, 2000. 27 Duk JM, Klaas H, de Bruijn H, Hollema H, ten Hoor KA, van der Zee A et al: Pre-treatment serum squamous cell carcinoma antigen: a newly identified prognostic factor in early stage cervical cancer. J Clin Oncol 14: 111-118, 1996.
Received February 9, 2004 Accepted February 8, 2005
1771
