GYNECOLOGIC
ONCOLOGY
39, 358-363 (1990)
Tumor Markers CA 125, Carcinoembryonic Antigen and TumorAssociated Trypsin Inhibitor in Patients with Cervical Adenocarcinoma ARTO LEMINEN, Departments
I and
II of Obstetrics
and
Gynecology,
Helsinki
University
M.D. Central
Hospital,
Haartmaninkatu
2, ~~-00290
Helsinki,
Finland
Received May 24. 1990
Tumor markersCA 125,carcinoembryonicantigen(CEA) and tumor-associated trypsin inhibitor (TATI) were studiedin 42 patients with cervical adenocarcinoma.Pretreatmentlevels of CA 125 were elevated in 73% of 33 patients, CEA in 48% of 27 patients, serumTATI in 23% of 22 patients, and urine TAT1 in 38%of 26 patients. Elevated CA 125levelswere associatedwith histologicalgrade(P = 0.002), and elevatedCEA levelswith the presenceof lymph nodemetastases (P = 0.008), respectively.No associationswere found betweenelevated tumor marker levels and stage,or tumor size. SerumCA 125levelsincreasedin 71% of the patientswith progressivedisease,CEA levelsin 36%,serum TAT1 levelsin 46%, and urine TATI levelsin 20%of the patients. In all patients with regressivediseasethe tumor marker levels decreasedor stayed unchanged.Regressionof the diseasewas significantly correlated(P < 0.05) with stage,histologicalgrade, tumor size, and nodal status. The resultssuggestthat CA 125 and, to a lesserextent, CEA and TATI are useful in the followup of patients with cervical adenocarcinoma. 0 WJO Academic Press, Inc.
INTRODUCTION
The determination of serum tumor markers has been of great value for monitoring the clinical course of gynecologic cancers. Tumor progression and recurrence have been detected by serial samples during follow-up of cervical squamous cell carcinoma [l] and cervical adenocarcinoma [2]. The incidence of carcinoma of the uterine cervix has been declining in most Western countries. This, however, is true only for cervical squamous cell carcinoma. In Finland the incidence of cervical adenocarcinoma has been stable since 1960 [3]. The clinical diagnosis of cervical adenocarcinoma is difficult and a reliable noninvasive method for post-treatment monitoring has not been available. Measurements of serum squamous cell carcinoma antigen (XC) have been used to monitor response to therapy in patients
with cervical squamous cell carcinoma [4,5]. Tumor markers CA 125 and CEA appear to be useful in monitoring cervical adenocarcinoma [6-91. The usefulness of another well-characterized tumor marker, tumor-associated trypsin inhibitor (TATI) [lo], in cervical adenocarcinoma has not been studied before. In the present study, the levels of three tumor markers were studied in a representative series of patients with cervical adenocarcinoma. The levels were correlated with stage, histological type and grade, tumor size, presence of lymph node metastases, and clinical course of the disease. MATERIAL
Patients. The patient material consisted of 42 patients with cervical adenocarcinoma treated between 1986 and 1988 at the Departments of Obstetrics and Gynecology, Helsinki University Central Hospital. The histological diagnosis and staging was performed according to the FIG0 classification [ 111. Serum and urine samples were available from 37 patients before the first treatment. Serial samples were collected during the follow-up period of 1 to 48 months (mean 18 months) and were available from 29 patients. Treatment. The principles of treatment consisted of pre- and postoperative radiation, surgery (mainly radical hysterectomy with pelvic lymphadenectomy), cytotoxic chemotherapy, or hormonal therapy, and have been described in detail elsewhere [12]. Evaluation of the effect of treatment was based on serial clinical examinations, laboratory and radiologic examinations, and findings at surgery performed for recurrent disease or for other reasons. The response to treatment was recorded according to the recommendations of the American Cancer Society [ 131. Progression of disease was defined as the appearance of any new lesions not previously identified, or an
358 oo!30-8258/90$1.50 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
AND METHODS
TUMOR
MARKERS:
CERVICAL
359
ADENOCARCINOMA
TABLE 1 Clinical and Histological Characteristics of 42 Cases of Cervical Adenocarcinoma: Relation to Elevated Tumor Marker Levels CA 125 >35 U ml-’
Stage I-II III-IV Histological type Adeno Adenosquamous Histological grade 1 2-3 Not known Tumor size 4 cm Not known Lymph node metastases No Yes Not known
CEA >3 pg liter-’
N
(%I
No./total
31 11
(74)
16129 8/11
(55) (73)
8/25 5110
(32)
(26)
(50)
39 3
(93) (7)
22137 213
(6’3
12123 113
10 16 6
(24)
(62)
319 11/15 216
(33)
319 16124 114
14126 9112 112
(54) (75) (50)
6126 7112 O/l
(26) (58)
12124 9110 316
(50) (W (50)
4122 619 315
(I@*
(14)
28 12 2
6’) (28)
25 11 6
(60)
(5)
(26) (14)
@Jo)
(67) (33)*
VW*
No./total
S-TAT1 >20 pg liter-’ (%)
No./total
(%)
3125 2110
(12) cw
6125 418
(23)
(36)
5133
(15)
9133
(33)
o/2
(‘3
O/I
(27) (0)
(33) (67)
l/19 4124 o/4
(II) (17) (0)
218 6124
(23 (25)
216
(33)
2118 3111 017
(27)
5123 519 012
(22) (56)
(%)
No./total
U-TAT1 >50 pg g-’ treat.
(25)
(0)
(11) (‘3
3121 l/8 115
(67)* (fm
(14) (13)
3121 318 415
cm
(50)
(0) (14)
(38) (75)
Note. S, serum; U, urine. * P < 0.05.
estimated increase in tumor size of 25% or more in existing lesions. For disease regression, a decrease in tumor size of 50% or more for at least 4 weeks (partial response) or disappearance of all clinical signs of malignancy (complete response) was required. The clinical course was correlated to serum and urine tumor marker levels. The time interval between the first and last sample ranged from 1 to 43 months (mean 12 months). A 100% increase, or a 50% decrease in marker levels was considered significant [8]. Statistics. Statistical analyses were performed by using x2 test and Fischer’s Exact test.
instructions (Centocor, Malvern, PA). Cutoff levels of 35 U ml-’ were used [8,14]. CEA levels were measured with an immunoradiometric kit using monoclonal antibodies (Abbot Laboratories, North Chicago, IL). Normal serum levels are below 3 pg liter-‘. TAT1 was measured by radioimmunoassay as earlier described [ 15,161. Cutoff levels of 20 pg liter- ’ in serum [ 151 and 50 pg g- ’ creatinine in urine [16] were used.
CA 125 was measured by imassay according to the manufacturer’s
and tumor marker jindClinical, pathological, ings. Clinical and histological characteristics of the
Radioimmunoassays.
munoradiometric 1000
q q
s
C E
loo? i
10:
6
a q
I q
2
=
r
m
q
$100; q
q q
z q
0
D
Y
10:
q
1-r
1000)
q
q
q
:
w
5
RESULTS
1
0
u
8
I
II
III
IV
STAGE
10.
q
I ;. I
0
0 II
q
I II
: Ill
3 IV
STAGE
FIG. 1. Pretreatment CA 125 and CEA levels in patients with cervical adenocarcinoma by stage (FIGO). The cutoff levels for normal values are indicated by a horizontal line.
360
ART0 200
-
LEMINEN 400
q
;
-350
\
6
q
c
3
150-
-300
B
100
.250 -200
F
-150
5
s
s
0
ch 50 -
q
8
i
o-
.?. I
0
q
I’
I’
I
II
III
IV
‘-
I
2
Concordance of Elevated Pretreatment Tumor Marker Levels in 19 Patients with Cervical Adenocarcinoma CA 125 ~ No. (%)
CEA ___ No. (%)
S-TAT1 No. (%)
U-TAT1 ~ No. (%)
6 (75) 4 (100) 5 (71)
6 (40) 2 (50) 2 (29)
4 (27) 2 (25) 2 (29)
5 (30)
Nole. S, serum; U, urine.
II
Ill
IV
STAGE
study population are shown in Table 1. The mean age of the patients was 55.3 years (range 25 to 87). Of all cases, 83% were primary cancers, the remaining seven (17%) representing residual or recurrent disease. Recurrent disease was found in 26% (11 of 42) during the follow-up. Only three (7%) patients had adenosquamous carcinoma of the cervix, and therefore comparison between different histologic tumor types could not be done. Early stages (1 and II) represented 74% (31 of 42) of the cases and lymph node metastases were found in 3 1% (11 of 36). The overall survival rate was 71% (30 of 42). Pretreatment serum CA 125 was elevated (range 35 to 736 U ml-‘) in 73% (24 of 33), serum CEA in 48% (range 3 to 433 pug liter-‘) (13 of 27) (Fig. l), serum TAT1 in 23% (range 20 to 194 pg liter-‘) (5 of 22), and urine TAT1 in 38% (range 50 to 151 pg g-’ creatinine) (10 of 26) (Fig. 2) of the patients. Elevated tumor marker levels were not associated with different clinical or histological characteristics (Table l), except that elevated CA 125 levels correlated with histological grade (P = 0.002), and elevated CEA levels correlated with nodal status (P = 0.008). The concordance of elevated pretreatment tumor marker levels is shown in Table 2. The other markers were elevated in 27-40% of the cases when CA 125 was elevated, in 25-75% of the cases when CEA was ele-
CA I25 (15) CEA (8) Serum TAT1 (4) Urine TAT1 (7)
50
q
10
FIG. 2. Pretreatment serum and urine TAT1 levels in patients with cervical adenocarcinoma by a horizontal line.
Tumor marker (No. of cases)
CI
I
STAGE
TABLE
-100
0
B
2
2 (25) 2 (50) -
by stage (FIGO). The cutoff levels are indicated
vated, in 50-100% of the cases when serum TAT1 was elevated, and in 29-71% of the cases when urine TAT1 was elevated. The situation where one tumor marker was negative in the pretreatment phase and the other tumor markers were positive is shown in Table 3. While CA 125 was negative the others were positive in O-75% of the cases, in 18-91% with CEA, in 27-73% with serum TATI, and in 17-83% with urine TATI. Tumor marker levels and clinical course. During the follow-up period progression of the disease was found in 49% (14 of 29). Forty-three percent (6 of 14) of them represented advanced stages (III-IV) and the rest (8/14) early stages. Regression of the disease was found in 34% (10 of 29) of the patients, and they all represented early stages, and stable disease in 5 (17%) patients of which 80% (4/5) represented early stages. There was no correlation between stable disease and stage, histological grade, tumor size, or lymph node metastases (Table 4). Progression of the disease correlated only with histological grade (P < 0.05). Regression of the disease correlated (P < 0.05) with all of these parameters. Eighty percent of patients with stable disease represented early stages (I and II) and had no lymph node metastases (Table 4). TABLE 3 Correlation of Negative Pretreatment Tumor Marker with the Positive Pretreatment Tumor Markers in 19 Patients with Cervical Adenocarcinoma Positive tumor marker Negative tumor marker (No. of cases) CA I25 (4) CEA (11) Serum TAT1 (15) Urine TAT1 (12)
CA 125
CEA
No. (%)
No. (%)
10 (91) 11 (73) 10 (83)
3 (75) 6 (40)
Note. S, serum; U, urine.
6 (50)
S-TAT1 ~ No. (%) 0
2 (18) 2 (17)
U-TAT1 ~ No. (%) 2 (50) 4 (36) 4 (27) -
361
TUMOR MARKERS: CERVICAL ADENOCARCINOMA
considering tumor markers in patients with cervical adenocarcinoma, and most of them refer only to CEA
TABLE 4 Stage, Histological Grade, Tumor Size, and Lymph Node Metastases in Progressive, Stable, and Regressive Disease in Patients with Cervical Admocarcinoma Progressive disease
Stable disease
Regressive disease
N14 (%)
NS (%)
NlO (%)
8 (58) 6 (42)
4 (80) 1 (20)
10 (loo)*
4 (29)* 10 (71)*
3 (60) 2 (40)
2 (20)* 8 (80)*
7 (50) 7 (50)
4 (80) 1 (20)
9 m* 1 (lo)*
6 (43) 5 (36) 3 (21)
4 (80)
8 (80)* 2 (20)* 0
Stage I-II
III-IV Histological grade 1 2-3 Tumor size 4 cm Lymph node metastases No Yes Not known
1 (20) 0
w71. According to this study CA 125 was the most useful tumor marker for cervical adenocarcinoma. In 73% of the cases the pretreatment serum levels were elevated, and serum levels increased in 72% of patients with progressive disease. In patients with regressive disease, the CA 125 levels decreased or were unchanged. These results are in accordance with earlier reports [6,8,9], although Duk et al. used a lower cutoff level (16 U ml-‘) for CA 125. Pretreatment serum CA 125 levels did not correlate with clinical stage, tumor size, or nodal status. On the contrary Duk et al. in their study found correlation between elevated serum CA 125 levels and stage. CEA and TAT1 were not as sensitive tumor markers as CA 125 in the diagnosis and follow-up of patients with cervical adenocarcinoma. Thus, these results are comparable with those by Duk et al. who found elevated serum CEA levels in 19% of the cases (a different cutoff level; 5 ng ml-‘). CA 125 showed better correlation with clinical course of the disease than CEA or TATI. According to this study and recent reports [9], serum CA 125 is the most useful marker for cervical adenocarcinoma. However, in cases with negative pretreatment CA 125 levels, CEA and TAT1 measurements may also be of value, as shown clearly in this study. When CA 125 was negative, up to 75% of the cases were CEApositive, and 50% of the cases were urine TATI-positive. When CA 125 levels were elevated before treatment, CEA, serum TATI, and urine TAT1 were negative in 73-91% of the patients, showing that majority of the cervical adenocarcinomas are CA 125-positive only. The same predominance of CA 125 is seen in the concordance of the elevated tumor marker levels in the pretreatment phase. If CEA, serum TATI, or urine TAT1 levels were elevated, CA 125 was also elevated in 71-100% of the cases. Urine TAT1 was elevated more often than serum TATI. It is apparent that assay of urine TAT1 is more
0*
* P < 0.05.
In progressive disease, there was a significant increase in serum CA 125 levels in 71% (10 of 14) of the patients. The corresponding rate for serum CEA levels was 36% (5 of 14), 46% (6 of 13) for serum TAT1 levels, and 20% (2 of 10) for urine TAT1 levels (Table 5). In all cases with regression the tumor marker levels were unchanged or decreased (Table 5). Three examples of changes in the tumor marker levels in relation to clinical course of the disease are shown in Fig. 3. DISCUSSION
The determination of tumor markers has proved useful in the pretreatment evaluation, in the evaluation of treatment effect, and in the follow-up of patients with gynecologic cancers. There are a limited number of studies
TABLE 5 Tumor Marker Levelsand Disease Progression (Pr, N 14) or Regression (Re, N 10) in Patients with Cervical Adenocarcinoma Tumor marker levels Increased No./total (%) FY CA 125 CEA S-TAT1 U-TAT1
10/14 5/14 6/13 2/10
Note. S, serum; U, urine.
(71) (36) (46) (20)
Unchanged No./total (%) Re 0 0 0 0
Pr 4/14 6/14 7/13 8/10
(29) (43) (54) (80)
Decreased No./total (%) Re 4 (40) 8 (80) 9 ww
7/8 (88)
pr
Re
0 3/14 (21) 0
6 W-3 2 cm
0
118(12)
0
362
ART0 LEMINEN
0
2
4
0 CA125 +cEA
Q CA125 +cEA
* *
* *
S-TAT1 U-TAT1
6 8 10 12 14 16 TIME (months)
0
5
10 15 TIME (months)
0 * * *
0
2
4
S-TAT1 U-TAT1
20
CA 125 CEA S-TAT1 U-TAT1
6 6 10 12 14 16 TIME (months)
FIG. 3. Tumor marker levels in patients with cervical adenocarcinoma. (A) Progressive disease; 7%year-old woman with stage III carcinoma treated with radiation and cytotoxic chemotherapy. (B) Stable disease; 62-year-old woman with stage I carcinoma treated with radiation, surgery, and cytotoxic chemotherapy. (C) Regressive disease: 2%year-old woman with stage II carcinoma treated with radiation, surgery, and cytotoxic chemotherapy.
useful than the serum assay. This conclusion has also reported by Halila et al. [18] in ovarian cancer. It might be most useful to determine the pretreatment levels of all these tumor markers and subsequently monitor them all, even if they were initially negative, because during follow-up the tumors may become marker-positive. With this method, earlier detection of recurrent disease would be possible. ACKNOWLEDGMENTS This study was supported by grants from Paulo Foundation, the Finnish Cancer Foundation, and the Medical Research Council Academy of Finland. I am grateful to Jorma Paavonen, M.D., for critical reading of this manuscript.
REFERENCES Te Velde, E. R., Persijn, J. P., Ballieux, R. E., and Faber, J. Carcinoembryonic antigen serum levels in patients with squamous cell carcinoma of the uterine cervix. Clinical significance, Cancer 49, 1866-1873 (1982). Kjorstad, K. E., and ijrjasater, H. The prognostic significance of carcinoembryonic antigen determinations in patients with adenocarcinoma of the cervix, Gynecol. Oncol. 19, 284-289 (1984). Leminen, A., Paavonen, J., Forss, M., Wahlstrom, T., and Vesterinen, E. Adenocarcinoma of the uterine cervix, Cancer 65, 5359 (1990).
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. Senekjian, E. K., Young, J. M., Weiser, P. A., Spencer, C. E., Magic, S. E., and Herbst, A. L. An evaluation of squamous cell carcinoma antigen in patients with cervical squamous cell carcinoma, Amer. J. Obstet. Gynecol. 157, 433-439 (1987). Niloff, J. M., Klug, T. L., Schltzl, E., Zurawski, V. R., Knapp, 6. R. C., and Bast, R. C. Elevation of serum CA 125 in carcinomas of the fallopian tube, endometrium, and endocervix, Amer. J. Obstet. Gynecol. 48, 1057-1058 (1984). 7. Kabawat, S. E., Bast, R. C., Bhan, A. K., Welch, W. R., Knapp, R. C., and Colvin, R. B. Tissue distribution of a coelomic-epithelium-related antigen recognized by the monoclonal antibody OC 125, In?. J. Gynecol. Pathol. 2, 275-285 (1983). 8. Halila, H., Stenman, U-H., and Sepplll, M. Ovarian cancer antigen CA I25 levels in pelvic inflammatory disease and pregnancy, Cancer 57, 1327-1329 (1986). 9. Duk, J. M., Aalders, J. Cl., Jan Fleuren, Cl., Krans, M., and de Bruijn, H. W. A. Tumor markers CA 125, squamous cell carcinoma antigen, and carcinoembryonic antigen in patients with adenocarcinoma of the uterine cervix, Obstet. Gynecol. 73, 661-668 (1989). 10. Huhtala, M-L., Pesonen, K., Kalkkinen, N., and Stenman, U-H. Purification and characterization of a tumor-associated trypsin inhibitor from the urine of a patient with ovarian cancer, J. Biol. Chem. 257, 13,713-13,716 (1982). 11. International Federation of Gynecology and Obstetrics. Classifi-
TUMOR
MARKERS:
CERVICAL
cation and staging of malignant turnouts in the female pelvis, Acta Obstet. Gynecol. Stand. 50, 1-7 (1971). 12. Vesterinen, E., Forss, M., and Nieminen, U. Increase of cervical adenocarcinoma: a report of 520 cases of cervical carcinoma including 112 tumors with glandular elements, Gynecol. Oncol. 33, 49-53 (1989). 13. Miller, A. B., Hoogstraten, B., Staquet, M., and Winkler, A. Reporting results of cancer treatment, Cancer 47, 207-214 (1981). 14. Bast, R. C., Klug, T. L., St. John, E., Jenison, E., Niloff, J. M., Lazarus, H., Berkowitz, R. S., Leavitt, T., Griffiths, T. C., Parker, L., Zurawski, V. R., and Knapp, R. C. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer, N. Engl. J. Med. 309, 883-887 (1983).
ADENOCARCINOMA
363
15. Stenman, U-H., Huhtala, M-L., Koistinen, R., and Seppala, M. Immunochemical demonstration of an ovarian cancer-associated urinary peptide, ht. J. Cancer 30, 53-57 (1982). 16. Huhtala, M-L., Kahanpaii, K., Seppalill, M., Halila, H., and Stenman, U-H. Excretion of a tumor-associated trypsin inhibitor (TATI) in urine of patients with gynecological malignancy, Znt. J. Cancer 31, 711-714 (1983). 17. Van Nagell, J. R., Donaldson, E. S., Gay, E. C., Rayburn, P., Powell, D. F., and Goldenberg, D. M. Carcinoembryonic antigen in carcinoma of the uterine cervix, Cancer 42, 2428-2434 (1978). 18. Halila, H., Lehtovirta, P., and Stenman, U-H. Tumour-associated trypsin inhibitor (TATI) in ovarian cancer, Brit. J. Cancer 57,304307 (1988).
ONCOLOGY
39, 358-363 (1990)
Tumor Markers CA 125, Carcinoembryonic Antigen and TumorAssociated Trypsin Inhibitor in Patients with Cervical Adenocarcinoma ARTO LEMINEN, Departments
I and
II of Obstetrics
and
Gynecology,
Helsinki
University
M.D. Central
Hospital,
Haartmaninkatu
2, ~~-00290
Helsinki,
Finland
Received May 24. 1990
Tumor markersCA 125,carcinoembryonicantigen(CEA) and tumor-associated trypsin inhibitor (TATI) were studiedin 42 patients with cervical adenocarcinoma.Pretreatmentlevels of CA 125 were elevated in 73% of 33 patients, CEA in 48% of 27 patients, serumTATI in 23% of 22 patients, and urine TAT1 in 38%of 26 patients. Elevated CA 125levelswere associatedwith histologicalgrade(P = 0.002), and elevatedCEA levelswith the presenceof lymph nodemetastases (P = 0.008), respectively.No associationswere found betweenelevated tumor marker levels and stage,or tumor size. SerumCA 125levelsincreasedin 71% of the patientswith progressivedisease,CEA levelsin 36%,serum TAT1 levelsin 46%, and urine TATI levelsin 20%of the patients. In all patients with regressivediseasethe tumor marker levels decreasedor stayed unchanged.Regressionof the diseasewas significantly correlated(P < 0.05) with stage,histologicalgrade, tumor size, and nodal status. The resultssuggestthat CA 125 and, to a lesserextent, CEA and TATI are useful in the followup of patients with cervical adenocarcinoma. 0 WJO Academic Press, Inc.
INTRODUCTION
The determination of serum tumor markers has been of great value for monitoring the clinical course of gynecologic cancers. Tumor progression and recurrence have been detected by serial samples during follow-up of cervical squamous cell carcinoma [l] and cervical adenocarcinoma [2]. The incidence of carcinoma of the uterine cervix has been declining in most Western countries. This, however, is true only for cervical squamous cell carcinoma. In Finland the incidence of cervical adenocarcinoma has been stable since 1960 [3]. The clinical diagnosis of cervical adenocarcinoma is difficult and a reliable noninvasive method for post-treatment monitoring has not been available. Measurements of serum squamous cell carcinoma antigen (XC) have been used to monitor response to therapy in patients
with cervical squamous cell carcinoma [4,5]. Tumor markers CA 125 and CEA appear to be useful in monitoring cervical adenocarcinoma [6-91. The usefulness of another well-characterized tumor marker, tumor-associated trypsin inhibitor (TATI) [lo], in cervical adenocarcinoma has not been studied before. In the present study, the levels of three tumor markers were studied in a representative series of patients with cervical adenocarcinoma. The levels were correlated with stage, histological type and grade, tumor size, presence of lymph node metastases, and clinical course of the disease. MATERIAL
Patients. The patient material consisted of 42 patients with cervical adenocarcinoma treated between 1986 and 1988 at the Departments of Obstetrics and Gynecology, Helsinki University Central Hospital. The histological diagnosis and staging was performed according to the FIG0 classification [ 111. Serum and urine samples were available from 37 patients before the first treatment. Serial samples were collected during the follow-up period of 1 to 48 months (mean 18 months) and were available from 29 patients. Treatment. The principles of treatment consisted of pre- and postoperative radiation, surgery (mainly radical hysterectomy with pelvic lymphadenectomy), cytotoxic chemotherapy, or hormonal therapy, and have been described in detail elsewhere [12]. Evaluation of the effect of treatment was based on serial clinical examinations, laboratory and radiologic examinations, and findings at surgery performed for recurrent disease or for other reasons. The response to treatment was recorded according to the recommendations of the American Cancer Society [ 131. Progression of disease was defined as the appearance of any new lesions not previously identified, or an
358 oo!30-8258/90$1.50 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
AND METHODS
TUMOR
MARKERS:
CERVICAL
359
ADENOCARCINOMA
TABLE 1 Clinical and Histological Characteristics of 42 Cases of Cervical Adenocarcinoma: Relation to Elevated Tumor Marker Levels CA 125 >35 U ml-’
Stage I-II III-IV Histological type Adeno Adenosquamous Histological grade 1 2-3 Not known Tumor size 4 cm Not known Lymph node metastases No Yes Not known
CEA >3 pg liter-’
N
(%I
No./total
31 11
(74)
16129 8/11
(55) (73)
8/25 5110
(32)
(26)
(50)
39 3
(93) (7)
22137 213
(6’3
12123 113
10 16 6
(24)
(62)
319 11/15 216
(33)
319 16124 114
14126 9112 112
(54) (75) (50)
6126 7112 O/l
(26) (58)
12124 9110 316
(50) (W (50)
4122 619 315
(I@*
(14)
28 12 2
6’) (28)
25 11 6
(60)
(5)
(26) (14)
@Jo)
(67) (33)*
VW*
No./total
S-TAT1 >20 pg liter-’ (%)
No./total
(%)
3125 2110
(12) cw
6125 418
(23)
(36)
5133
(15)
9133
(33)
o/2
(‘3
O/I
(27) (0)
(33) (67)
l/19 4124 o/4
(II) (17) (0)
218 6124
(23 (25)
216
(33)
2118 3111 017
(27)
5123 519 012
(22) (56)
(%)
No./total
U-TAT1 >50 pg g-’ treat.
(25)
(0)
(11) (‘3
3121 l/8 115
(67)* (fm
(14) (13)
3121 318 415
cm
(50)
(0) (14)
(38) (75)
Note. S, serum; U, urine. * P < 0.05.
estimated increase in tumor size of 25% or more in existing lesions. For disease regression, a decrease in tumor size of 50% or more for at least 4 weeks (partial response) or disappearance of all clinical signs of malignancy (complete response) was required. The clinical course was correlated to serum and urine tumor marker levels. The time interval between the first and last sample ranged from 1 to 43 months (mean 12 months). A 100% increase, or a 50% decrease in marker levels was considered significant [8]. Statistics. Statistical analyses were performed by using x2 test and Fischer’s Exact test.
instructions (Centocor, Malvern, PA). Cutoff levels of 35 U ml-’ were used [8,14]. CEA levels were measured with an immunoradiometric kit using monoclonal antibodies (Abbot Laboratories, North Chicago, IL). Normal serum levels are below 3 pg liter-‘. TAT1 was measured by radioimmunoassay as earlier described [ 15,161. Cutoff levels of 20 pg liter- ’ in serum [ 151 and 50 pg g- ’ creatinine in urine [16] were used.
CA 125 was measured by imassay according to the manufacturer’s
and tumor marker jindClinical, pathological, ings. Clinical and histological characteristics of the
Radioimmunoassays.
munoradiometric 1000
q q
s
C E
loo? i
10:
6
a q
I q
2
=
r
m
q
$100; q
q q
z q
0
D
Y
10:
q
1-r
1000)
q
q
q
:
w
5
RESULTS
1
0
u
8
I
II
III
IV
STAGE
10.
q
I ;. I
0
0 II
q
I II
: Ill
3 IV
STAGE
FIG. 1. Pretreatment CA 125 and CEA levels in patients with cervical adenocarcinoma by stage (FIGO). The cutoff levels for normal values are indicated by a horizontal line.
360
ART0 200
-
LEMINEN 400
q
;
-350
\
6
q
c
3
150-
-300
B
100
.250 -200
F
-150
5
s
s
0
ch 50 -
q
8
i
o-
.?. I
0
q
I’
I’
I
II
III
IV
‘-
I
2
Concordance of Elevated Pretreatment Tumor Marker Levels in 19 Patients with Cervical Adenocarcinoma CA 125 ~ No. (%)
CEA ___ No. (%)
S-TAT1 No. (%)
U-TAT1 ~ No. (%)
6 (75) 4 (100) 5 (71)
6 (40) 2 (50) 2 (29)
4 (27) 2 (25) 2 (29)
5 (30)
Nole. S, serum; U, urine.
II
Ill
IV
STAGE
study population are shown in Table 1. The mean age of the patients was 55.3 years (range 25 to 87). Of all cases, 83% were primary cancers, the remaining seven (17%) representing residual or recurrent disease. Recurrent disease was found in 26% (11 of 42) during the follow-up. Only three (7%) patients had adenosquamous carcinoma of the cervix, and therefore comparison between different histologic tumor types could not be done. Early stages (1 and II) represented 74% (31 of 42) of the cases and lymph node metastases were found in 3 1% (11 of 36). The overall survival rate was 71% (30 of 42). Pretreatment serum CA 125 was elevated (range 35 to 736 U ml-‘) in 73% (24 of 33), serum CEA in 48% (range 3 to 433 pug liter-‘) (13 of 27) (Fig. l), serum TAT1 in 23% (range 20 to 194 pg liter-‘) (5 of 22), and urine TAT1 in 38% (range 50 to 151 pg g-’ creatinine) (10 of 26) (Fig. 2) of the patients. Elevated tumor marker levels were not associated with different clinical or histological characteristics (Table l), except that elevated CA 125 levels correlated with histological grade (P = 0.002), and elevated CEA levels correlated with nodal status (P = 0.008). The concordance of elevated pretreatment tumor marker levels is shown in Table 2. The other markers were elevated in 27-40% of the cases when CA 125 was elevated, in 25-75% of the cases when CEA was ele-
CA I25 (15) CEA (8) Serum TAT1 (4) Urine TAT1 (7)
50
q
10
FIG. 2. Pretreatment serum and urine TAT1 levels in patients with cervical adenocarcinoma by a horizontal line.
Tumor marker (No. of cases)
CI
I
STAGE
TABLE
-100
0
B
2
2 (25) 2 (50) -
by stage (FIGO). The cutoff levels are indicated
vated, in 50-100% of the cases when serum TAT1 was elevated, and in 29-71% of the cases when urine TAT1 was elevated. The situation where one tumor marker was negative in the pretreatment phase and the other tumor markers were positive is shown in Table 3. While CA 125 was negative the others were positive in O-75% of the cases, in 18-91% with CEA, in 27-73% with serum TATI, and in 17-83% with urine TATI. Tumor marker levels and clinical course. During the follow-up period progression of the disease was found in 49% (14 of 29). Forty-three percent (6 of 14) of them represented advanced stages (III-IV) and the rest (8/14) early stages. Regression of the disease was found in 34% (10 of 29) of the patients, and they all represented early stages, and stable disease in 5 (17%) patients of which 80% (4/5) represented early stages. There was no correlation between stable disease and stage, histological grade, tumor size, or lymph node metastases (Table 4). Progression of the disease correlated only with histological grade (P < 0.05). Regression of the disease correlated (P < 0.05) with all of these parameters. Eighty percent of patients with stable disease represented early stages (I and II) and had no lymph node metastases (Table 4). TABLE 3 Correlation of Negative Pretreatment Tumor Marker with the Positive Pretreatment Tumor Markers in 19 Patients with Cervical Adenocarcinoma Positive tumor marker Negative tumor marker (No. of cases) CA I25 (4) CEA (11) Serum TAT1 (15) Urine TAT1 (12)
CA 125
CEA
No. (%)
No. (%)
10 (91) 11 (73) 10 (83)
3 (75) 6 (40)
Note. S, serum; U, urine.
6 (50)
S-TAT1 ~ No. (%) 0
2 (18) 2 (17)
U-TAT1 ~ No. (%) 2 (50) 4 (36) 4 (27) -
361
TUMOR MARKERS: CERVICAL ADENOCARCINOMA
considering tumor markers in patients with cervical adenocarcinoma, and most of them refer only to CEA
TABLE 4 Stage, Histological Grade, Tumor Size, and Lymph Node Metastases in Progressive, Stable, and Regressive Disease in Patients with Cervical Admocarcinoma Progressive disease
Stable disease
Regressive disease
N14 (%)
NS (%)
NlO (%)
8 (58) 6 (42)
4 (80) 1 (20)
10 (loo)*
4 (29)* 10 (71)*
3 (60) 2 (40)
2 (20)* 8 (80)*
7 (50) 7 (50)
4 (80) 1 (20)
9 m* 1 (lo)*
6 (43) 5 (36) 3 (21)
4 (80)
8 (80)* 2 (20)* 0
Stage I-II
III-IV Histological grade 1 2-3 Tumor size 4 cm Lymph node metastases No Yes Not known
1 (20) 0
w71. According to this study CA 125 was the most useful tumor marker for cervical adenocarcinoma. In 73% of the cases the pretreatment serum levels were elevated, and serum levels increased in 72% of patients with progressive disease. In patients with regressive disease, the CA 125 levels decreased or were unchanged. These results are in accordance with earlier reports [6,8,9], although Duk et al. used a lower cutoff level (16 U ml-‘) for CA 125. Pretreatment serum CA 125 levels did not correlate with clinical stage, tumor size, or nodal status. On the contrary Duk et al. in their study found correlation between elevated serum CA 125 levels and stage. CEA and TAT1 were not as sensitive tumor markers as CA 125 in the diagnosis and follow-up of patients with cervical adenocarcinoma. Thus, these results are comparable with those by Duk et al. who found elevated serum CEA levels in 19% of the cases (a different cutoff level; 5 ng ml-‘). CA 125 showed better correlation with clinical course of the disease than CEA or TATI. According to this study and recent reports [9], serum CA 125 is the most useful marker for cervical adenocarcinoma. However, in cases with negative pretreatment CA 125 levels, CEA and TAT1 measurements may also be of value, as shown clearly in this study. When CA 125 was negative, up to 75% of the cases were CEApositive, and 50% of the cases were urine TATI-positive. When CA 125 levels were elevated before treatment, CEA, serum TATI, and urine TAT1 were negative in 73-91% of the patients, showing that majority of the cervical adenocarcinomas are CA 125-positive only. The same predominance of CA 125 is seen in the concordance of the elevated tumor marker levels in the pretreatment phase. If CEA, serum TATI, or urine TAT1 levels were elevated, CA 125 was also elevated in 71-100% of the cases. Urine TAT1 was elevated more often than serum TATI. It is apparent that assay of urine TAT1 is more
0*
* P < 0.05.
In progressive disease, there was a significant increase in serum CA 125 levels in 71% (10 of 14) of the patients. The corresponding rate for serum CEA levels was 36% (5 of 14), 46% (6 of 13) for serum TAT1 levels, and 20% (2 of 10) for urine TAT1 levels (Table 5). In all cases with regression the tumor marker levels were unchanged or decreased (Table 5). Three examples of changes in the tumor marker levels in relation to clinical course of the disease are shown in Fig. 3. DISCUSSION
The determination of tumor markers has proved useful in the pretreatment evaluation, in the evaluation of treatment effect, and in the follow-up of patients with gynecologic cancers. There are a limited number of studies
TABLE 5 Tumor Marker Levelsand Disease Progression (Pr, N 14) or Regression (Re, N 10) in Patients with Cervical Adenocarcinoma Tumor marker levels Increased No./total (%) FY CA 125 CEA S-TAT1 U-TAT1
10/14 5/14 6/13 2/10
Note. S, serum; U, urine.
(71) (36) (46) (20)
Unchanged No./total (%) Re 0 0 0 0
Pr 4/14 6/14 7/13 8/10
(29) (43) (54) (80)
Decreased No./total (%) Re 4 (40) 8 (80) 9 ww
7/8 (88)
pr
Re
0 3/14 (21) 0
6 W-3 2 cm
0
118(12)
0
362
ART0 LEMINEN
0
2
4
0 CA125 +cEA
Q CA125 +cEA
* *
* *
S-TAT1 U-TAT1
6 8 10 12 14 16 TIME (months)
0
5
10 15 TIME (months)
0 * * *
0
2
4
S-TAT1 U-TAT1
20
CA 125 CEA S-TAT1 U-TAT1
6 6 10 12 14 16 TIME (months)
FIG. 3. Tumor marker levels in patients with cervical adenocarcinoma. (A) Progressive disease; 7%year-old woman with stage III carcinoma treated with radiation and cytotoxic chemotherapy. (B) Stable disease; 62-year-old woman with stage I carcinoma treated with radiation, surgery, and cytotoxic chemotherapy. (C) Regressive disease: 2%year-old woman with stage II carcinoma treated with radiation, surgery, and cytotoxic chemotherapy.
useful than the serum assay. This conclusion has also reported by Halila et al. [18] in ovarian cancer. It might be most useful to determine the pretreatment levels of all these tumor markers and subsequently monitor them all, even if they were initially negative, because during follow-up the tumors may become marker-positive. With this method, earlier detection of recurrent disease would be possible. ACKNOWLEDGMENTS This study was supported by grants from Paulo Foundation, the Finnish Cancer Foundation, and the Medical Research Council Academy of Finland. I am grateful to Jorma Paavonen, M.D., for critical reading of this manuscript.
REFERENCES Te Velde, E. R., Persijn, J. P., Ballieux, R. E., and Faber, J. Carcinoembryonic antigen serum levels in patients with squamous cell carcinoma of the uterine cervix. Clinical significance, Cancer 49, 1866-1873 (1982). Kjorstad, K. E., and ijrjasater, H. The prognostic significance of carcinoembryonic antigen determinations in patients with adenocarcinoma of the cervix, Gynecol. Oncol. 19, 284-289 (1984). Leminen, A., Paavonen, J., Forss, M., Wahlstrom, T., and Vesterinen, E. Adenocarcinoma of the uterine cervix, Cancer 65, 5359 (1990).
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. Senekjian, E. K., Young, J. M., Weiser, P. A., Spencer, C. E., Magic, S. E., and Herbst, A. L. An evaluation of squamous cell carcinoma antigen in patients with cervical squamous cell carcinoma, Amer. J. Obstet. Gynecol. 157, 433-439 (1987). Niloff, J. M., Klug, T. L., Schltzl, E., Zurawski, V. R., Knapp, 6. R. C., and Bast, R. C. Elevation of serum CA 125 in carcinomas of the fallopian tube, endometrium, and endocervix, Amer. J. Obstet. Gynecol. 48, 1057-1058 (1984). 7. Kabawat, S. E., Bast, R. C., Bhan, A. K., Welch, W. R., Knapp, R. C., and Colvin, R. B. Tissue distribution of a coelomic-epithelium-related antigen recognized by the monoclonal antibody OC 125, In?. J. Gynecol. Pathol. 2, 275-285 (1983). 8. Halila, H., Stenman, U-H., and Sepplll, M. Ovarian cancer antigen CA I25 levels in pelvic inflammatory disease and pregnancy, Cancer 57, 1327-1329 (1986). 9. Duk, J. M., Aalders, J. Cl., Jan Fleuren, Cl., Krans, M., and de Bruijn, H. W. A. Tumor markers CA 125, squamous cell carcinoma antigen, and carcinoembryonic antigen in patients with adenocarcinoma of the uterine cervix, Obstet. Gynecol. 73, 661-668 (1989). 10. Huhtala, M-L., Pesonen, K., Kalkkinen, N., and Stenman, U-H. Purification and characterization of a tumor-associated trypsin inhibitor from the urine of a patient with ovarian cancer, J. Biol. Chem. 257, 13,713-13,716 (1982). 11. International Federation of Gynecology and Obstetrics. Classifi-
TUMOR
MARKERS:
CERVICAL
cation and staging of malignant turnouts in the female pelvis, Acta Obstet. Gynecol. Stand. 50, 1-7 (1971). 12. Vesterinen, E., Forss, M., and Nieminen, U. Increase of cervical adenocarcinoma: a report of 520 cases of cervical carcinoma including 112 tumors with glandular elements, Gynecol. Oncol. 33, 49-53 (1989). 13. Miller, A. B., Hoogstraten, B., Staquet, M., and Winkler, A. Reporting results of cancer treatment, Cancer 47, 207-214 (1981). 14. Bast, R. C., Klug, T. L., St. John, E., Jenison, E., Niloff, J. M., Lazarus, H., Berkowitz, R. S., Leavitt, T., Griffiths, T. C., Parker, L., Zurawski, V. R., and Knapp, R. C. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer, N. Engl. J. Med. 309, 883-887 (1983).
ADENOCARCINOMA
363
15. Stenman, U-H., Huhtala, M-L., Koistinen, R., and Seppala, M. Immunochemical demonstration of an ovarian cancer-associated urinary peptide, ht. J. Cancer 30, 53-57 (1982). 16. Huhtala, M-L., Kahanpaii, K., Seppalill, M., Halila, H., and Stenman, U-H. Excretion of a tumor-associated trypsin inhibitor (TATI) in urine of patients with gynecological malignancy, Znt. J. Cancer 31, 711-714 (1983). 17. Van Nagell, J. R., Donaldson, E. S., Gay, E. C., Rayburn, P., Powell, D. F., and Goldenberg, D. M. Carcinoembryonic antigen in carcinoma of the uterine cervix, Cancer 42, 2428-2434 (1978). 18. Halila, H., Lehtovirta, P., and Stenman, U-H. Tumour-associated trypsin inhibitor (TATI) in ovarian cancer, Brit. J. Cancer 57,304307 (1988).
