GYNECOLOGIC ONCOLOGY
38, 6-11 (1990)
Squamous Cell Carcinoma Antigen and Carcinoembryonic Antigen Levels as Prognostic Factors for the Response of Cervical Carcinoma to Chemotherapy W.
MEIER,
W.
P. STIEBER,* A.
EIERMANN,
FATEH-MOGHADAM,*
A. SCHNEIDER, AND H. HEPP
Departments of Obstetrics and Gynecology and *Clinical Chemistry, Klinikum Grosshadern, Ludwig-Maximilians-University, Marchioninistrasse 15, 8000 Munich 70, West Germany Received September 5, 1989; revised November 2, 1989
With the introduction of tumor markers in gynecological oncology, it has become possible to identify tumor recurrence early in the follow-up of some types of tumor and thus to start the necessary treatment early [9-121. In the case of cervical carcinoma, squamous cell carcinoma antigen (SCC) and carcinoembryonic antigen (CEA) are two markers which are available as diagnostic tools-above all in combination [ 13- 181. Our study aimed to answer the question whether SCC and CEA levels can be used as a prognostic factor for tumor response to chemotherapy, and thus whether a developing resistance to cytostatic drugs can be determined early. In these cases, the patients could be saved from the continuation of an ineffective treatment with very severe side effects,
Between January 1986 and December 1988, 36 patients with primary advanced or recurrent cervical carcinoma were treated with cytostatic drugs in our department. Treatment at first was a combination of cisplatin and etoposide. After August 1987, a combination of carboplatin and ifosfamide was used. In all patients showing primary response to therapy, the squamous cell carcinoma antigen @CC) and carcinoembryonic antigen (CEA) levels fell rapidly to normal after one or two cycles. In contrast, clinical remission was not obtained in those patients with levels which remained high or rose again following an initial decrease. Chemotherapy is often the only available therapy for advanced cervical carcinoma or recurrent disease, although the results of treatment, especially in squamous cell carcinoma, remain poor. The course of the SCC or CEA levels can help to decide whether the patient would profit from a continuation of the therapy. With the tumor markers, treatment can be individualized so that, above all, cases of therapy failure or further tumor progression can be detected early and the patient can be spared the severeside effects of the treatment. 0 1990 Academic Press, Inc.
PATIENTS AND METHODS Between January 1986 and December 1988,36 patients with cervical carcinoma were treated with cytostatic drugs in our department. In 11 patients with stage IVB and one patient with stage IVA disease, primary chemotherapy was given following histological diagnosis at staging laparotomy or by scalenus biopsy. Three patients were treated with adjuvant chemotherapy following Wertheim-Meigs radical hysterectomy after the extent of the disease became known histologically (e.g., blood vessel involvement, lymphangiosis carcinomatosa). Twenty-one patients received chemotherapy because of recurrent disease: In 8 cases, the recurrence was local and in 13 cases there were distant metastases. Both vaginal stump and pelvic wall disease were classified as local recurrences. The primary therapy was in 13 cases a Wertheim-Meigs radical hysterectomy, with postop-
INTRODUCTION Almost 35% of patients with cervical carcinoma develop a recurrence usually within 2 years of surgery or radiotherapy [l-3]. Often, the only possible treatment available for these patients and those who present with metastatic disease is chemotherapy. Since the introduction of cisplatin, remission rates of more than 50% have been described, including patients with complete remission. The length of remission remains, however, relatively short [l-5]. Above all, because of the severe side effects of the treatment, the question often arises, whether in the individual case chemotherapy is justified, especially as there are no reliable methods of chemosensitivity testing in cervical carcinoma to date [6-81.
6 owe-8258/90 $1.50 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
SQUAMOUS CELL CARCINOMA
erative radiotherapy in 11 cases. Eight patients were treated primarily with combined radiotherapy. Eight patients had stage IB, one stage IIA, seven stage IIB, four stage IIIB, and one stage IVA disease. There was no relation between the localization of recurrence and the stage of disease or the type of primary therapy. Chemotherapy consisted of a combination of cisplatin 80 mg/qm Day 1 and etoposide 150 mg/qm Days 3 to 5 up to July 1987. From August 1987, carboplatin 300 mg/qm and ifosfamide 5 g/qm both on Day 1 were used [19]. The chemotherapy cycles were applied every 4 weeks dependent on blood count. The success of the treatment was judged according to the WHO recommendations [20]. Tumor marker levels were taken every 1 to 2 months. The serum levels of SCC were determined using a radioimmunoassay (Abbott). CEA levels were determined with an enzyme immunoassay (Abbott). The upper limit of normal was taken as 2 rig/ml for SCC and 3 rig/ml for CEA. RESULTS Of the 12 patients with primary chemotherapy, 9 had a raised SCC level and 8 a raised CEA level before beginning therapy (Table 1). In all 12 patients, either SCC, CEA, or both markers were raised. In patients with recurrent disease, SCC and CEA were each elevated in 13 cases, and in a total of 19 patients, at least one of the marker levels was raised. There were no percentage differences in raised tumor marker levels between patients with local recurrence and patients with distant metastases. In three patients who received adjuvant chemotherapy because of the histological extent of disease following radical hysterectomy, the initial SCC levels and in one patient the CEA level were also raised. Postoperative levels were within the normal range, so that in these three cases, the tumor markers could not be used to monitor the success of chemotherapy. Overall, 31 of 33 patients with either primary chemotherapy or chemotherapy for recurrent disease showed raised SCC or CEA levels. In 12 patients, both markers were raised. Only two patients did not have raised levels TABLE 1 Total Number of Elevated XC and CEA Levels before Beginning Chemotherapy see Primary therapy Recurrence Adjuvant therapy Total
N
(>2 rig/ml)
CEA (>3 rig/ml)
Elevated SCC and/or CEA
12 21 3 36
9 13 0 22
8 13 0 21
12 19 0 31
ANTIGEN
7
AND CHEMOTHERAPY
TABLE 2 SCC and CEA Levels in Relation to Histology
see Squamous cell carcinoma Adenocarcinoma Adenosquamous carcinoma
N
(>2 rig/ml)
CEA (>3 rig/ml)
Elevated SCC and/or CEA
26
19
14
24
5 2
2 I
5 2
5 2
before beginning chemotherapy. This made it possible to evaluate the results of chemotherapy in the majority of cases, using the tumor marker levels. Histologically, 26 patients had a squamous cell carcinoma, 5 patients an adenocarcinoma, and 2 patients an adenosquamous carcinoma (Table 2). There were no significant differences with respect to the number of raised SCC and CEA levels and the difference in patients’ distribution in the individual groups; however, all patients with adenocarcinoma showed raised CEA levels. In cases of squamous cell carcinoma, SCC levels were raised in 19 cases and CEA in 14 cases. Combined, either SCC or CEA were raised in 24 of 26 patients. Both patients with adenosquamous carcinoma demonstrated raised marker levels. The three patients, who received adjuvant chemotherapy all had a squamous cell carcinoma, with raised SCC levels in all three cases and a raised CEA in one patient. To judge the success of treatment, only those patients were included in whom at least three cycles of chemotherapy had been completed. In two patients, the chemotherapy was discontinued after one cycle; in one case because of poor tolerance of side effects, in the other because of rapid tumor progression. The outcome of therapy could be assessed in the remaining 31 patients. There was complete remission in 3 cases (10%). In 11 cases (35%) a partial remission and in 17 cases a primary progression were seen. Of these 17 women with progressive disease, in 5 the disease remained stable initially (“no change”). However, in each of these 5 cases, progression occurred within 3 months and they were therefore classified as having progressive disease. If cases of complete and partial remission are grouped together, 45% fall into this category. There were no differences in response with respect to the localization of the recurrence or the form of primary treatment. Table 3 shows the primary response to chemotherapy in relation to tumor marker levels. In two cases, neither SCC nor CEA levels were raised at first, and two further patients received only one treatment cycle, so that 29 of the 33 patients with therapeutical chemotherapy remained for assessment. Thirteen patients showed an in-
8
MEIER ET AL.
The semilogarithmic scale shows that in no case did the SCC level fall below 2 rig/ml. The levels begin to climb again steeply as early as 1 or 2 months after beginning treatment. Figure 2 shows six of these patterns. Decreasing Increasing Tumor markers can also be helpful in the often difficult SCCfCEA SCC/CEA values values assessment of tumors with bone metastases. Figure 3 shows the course of SCC in a 54-year-old patient who 0 Remission 13 developed bone metastases 14 months after primary ra16 0 Progressive disease diotherapy of a cervical carcinoma stage IIIB. After induction of chemotherapy, the SCC levels fell rapidly in normal and provided evidence of remission. The radiolitial response to the therapy with either a complete or ogical controls and bone scintigrams confirmed the parpartial remission. In all these cases, the tumor marker tial remission. Nine months later, the tumor marker levvalues dropped quickly to normal levels. None of these els began to rise again despite chemotherapy. Two patients demonstrated clinical progression at the same months later, the relapse was detectable radiologically, time as having normal tumor marker values. On the other and chemotherapy was stopped. hand, none of the 16 patients with increasing tumor marker levels showed clinical remission at this time. DISCUSSION Figure 1 shows tumor marker levels in five patients Despite the introduction of new cytostatic drugs, the with primary response to chemotherapy. In all cases, the XC levels were raised before beginning treatment. Fol- results of treatment in patients with recurrent disease or lowing one or two therapy cycles, the tumor marker primary metastatic disease in cervical carcinoma have levels can be seen to drop off to normal levels. The only been fractionally improved [l-3,8]. With the use of semilogarithmic scale clearly shows the rapid decrease platinum-containing compounds in single or combined in SCC levels to less than 2 rig/ml. Following remission schedules, up to 50% partial or even complete remissions periods of in some cases more than 6 to 8 months, the have been achieved. The period of remission has relevels begin to increase again. Shortly afterward, the mained relatively short [l-51. In our study, a remission progression was also confirmed clinically. Similar tumor rate of 45% was obtained, including three patients with complete remission. Patients with initial response to marker levels can be seen with CEA. In patients with primary progression during chemo- chemotherapy have significantly better survival rates therapy, only a short-term reduction in SCC levels can than patients with primary progression [ 1,3,8]. The combe seen, which, however, do not return to normal levels. plications due to tumor such as anus praeter, renal fisTABLE 3 Correlation of Tumor Marker Levels and Primary Response or Primary Progression during Chemotherapy
SCC(ng/ml) 1ooE
I
I
1
I
I
I
I
I
I
I
I
I
I
I
8
9
10
11
12
13
14
al
0
1
2
3
4
5
7 6 Months
FIG. 1. SCC levels of five patients with primary response to chemotherapy (semilogarithmic scale).
SQUAMOUS
100
CELL
CARCINOMA
ANTIGEN
9
AND CHEMOTHERAPY
SCC(ng/ml)
1’ 0
I 1
I 2
I 3
I 4
1 5
I 6 Months
I 7
0 8
1 9
I 10
I 11
1 12
FIG. 2. SCC levels of six patients with primary progression during chemotherapy (semilogarithmic scale).
tula, and intractable pain from deep pelvic recurrence occur significantly later in the course of the disease [2,3,5,8]. The price of this improvement in the quality of life due to reduction in tumor complications is paid for with the severe side effects of the chemotherapy, the incidence of which has not been much reduced despite dosage modification, change of interval, and use of highdose antiemetics and sedatives. In view of the short remission times and the known side effects of the drugs,
the question often arises as to whether or not the use of chemotherapy can be justified. There has been no reliable parameter available up until now to help monitor the therapy results, and demonstrate a response or signal a developing resistance to chemotherapy. A reliable method of in vitro chemosensitivity testing for cervical carcinoma has not yet been developed [6-81. The methods described in the literature are unsatisfactory in cases of cervical carcinoma [6,7]. The
SCC(ng/ml) 40 1 35 Begin Chemo
End Chemo
/
30 25 20 -
RI.,54 Y. CC IIIB
0 0
1 I
1 2
I 3
I 4
1 6
1 8
I I 7 8 Month8
I 9
1 10
I 11
1 12
1 13
I 14
FIG. 3. SCC levels during chemotherapy of a patient with bone metastases.
I 16
10
MEIER ET AL.
introduction of tumor markers, especially SCC, has meant that a parameter has become available that can indicate at a relatively early point whether the patient is responding to therapy [15-171. In 73 and 54% of cases of squamous cell carcinoma of the cervix, there are elevated SCC and CEA levels, respectively. In contrast, CEA can be seen to be superior to SCC in cases of adenocarcinoma [ 13,181. Prerequisite for the use of tumor markers for monitoring effects of therapy are the initial raised serum levels. In our patient collection, all 12 patients with primary chemotherapy showed a raised SCC or CEA level: SCC was raised in 75%, CEA in 66%. In recurrent disease 62% each of SCC and CEA levels were pathological, while 19 of 21 patients showed a raised SCC or CEA level. There was no significant difference between cases with local recurrent disease and those with distant metastases. These results agree with earlier studies and the results of other groups [ 13-181. As a result, monitoring the results of treatment or the response to chemotherapy using tumor markers becomes possible. In our patients, the positive predictive value and the negative predictive value for response to therapy was 100%. In those patients who are responding to therapy, the SCC levels fall to normal after two treatment cycles at the latest. There is a good correlation between the tumor marker levels and the course of the disease. After stopping treatment because of complete or partial remission, the values begin to climb again after 8-12 months. In our opinion, at this time chemotherapy should be stopped or, following a treatment break after complete remission, restarted to try to induce remission again. The optimal length of a time a chemotherapy should be continued after complete remission is not known [1,2,8]. Piver describes a chemotherapy schedule with cisplatin which was given over the period of 5 years to patients with cervical carcinoma and was tolerated without excessive side effects [3]. The patients received a total of 1325-3520 mg cisplatin during the treatment. A moot point arises when tumor marker levels remain within normal limits for several months and clinically the patient is in complete remission: Is maintenance chemotherapy justified? It would be more tolerable for the patient to begin again with the chemotherapy when the tumor marker levels begin to increase. Further studies are necessary to evaluate the percentage of patients which will have a second response to chemotherapy in these cases. In cases of primary progression, the SCC and CEA levels decrease for a short time initially, but never approach normal values. In all cases, the levels increase steeply after 2-3 months. In all these cases, chemotherapy should be discontinued, even if the clinical and
radiological examinations have not yet demonstrated the progression. Supportive measures should be introduced. Above all, in cases which are clinically difficult to assess, additional treatment monitoring using tumor marker levels would appear to be useful. Included in this category are bone metastases and histologically proven pelvic recurrence following surgery and radiotherapy [8]. The course of the tumor marker levels can help with an early decision as to whether a patient can be helped by continuing treatment. In advanced cervical carcinoma or in the case of recurrent disease, the cytostatic treatment often remains the only possible treatment. On the other hand, the success rate of chemotherapy with respect to the length of remission is not outstanding, especially in cases of squamous cell carcinoma. The course of the SCC and CEA levels can help to decide early whether patients will profit from continuation of treatment. If the levels are not within the normal range after, at the latest, three cycles of treatment, the chemotherapy should be discontinued. A rapid decrease in tumor marker levels to normal indicates primary response to therapy and treatment can be continued while monitoring tumor marker levels further. In our opinion, on complete remission, chemotherapy should be stopped if the tumor marker levels are normal. It should be restarted only when the tumor marker levels begin to increase again. Thus, with the help of tumor markers, individualized treatment can be given, whereby nonresponse or relapse can be diagnosed early and the patient saved from the severe side effects of the therapy. REFERENCES 1. Lele, S. B., and Piver, M. S. Weekly cisplatin induction chemotherapy in the treatment of recurrent cervical carcinoma, Gynecol. Oncol. 33, 6-8 (1989). 2. Piver, M. S., Barlow, J. J., Lele, S. B., and Maniccia, M. Weekly
cis-diamminedichloroplatinum II as induction chemotherapy in recurrent carcinoma of the cervix, Gynecol. Oncol. 18, 313-319 (1984). 3. Piver, M. S., Lele, S. B., Patsner, B., and McPhee, M. E. 1.5-5
years of uninterrupted cis-diamminedichloroplatinum II chemotherapy for metastatic cervical cancer, Gynecol. 0~01. 27, 24-33 (1987). 4. Rotmensch, J., Senekjian, E. K., Javaheri, G., and Herbst, A. L. Evaluation of bolus cisplatinum and continuous 5-fluorouracil infusion for metastatic and recurrent squamous cell carcinoma of the cervix, Gynecol. Oncol. 29, 76-81 (1988). 5. Thigpen, J. T., Blessing, J. A., Disaia, P. J., Fowler, W. C., and Hatch, K. D. A randomized comparison of a rapid versus prolonged (24 hr) infusion of cisplatin in therapy of squamous cell carcinoma of the uterine cervix: A Gynecologic Oncology Group study, Gynecol. Oncol. 32, 198-202 (1989). 6. Rockwell, S. Effects of clumps and clusters on survival measurements with clonogenic assays, Cancer Res. 45, 1601-1607 (1985).
SQUAMOUS CELL CARCINOMA 7. Umbach, G. E., Poethen, J., v. Matthiessen, H., Bender, H. G.. and Koldovsky, U. Experiences with the human tumor colonyforming assay in gynecologic malignancies. J. Cuncev Res. Clin. Oncol. 110, 234-237 (1985).
8. Umbach, G. E., v. Matthiessen, H., and Bender, H. G. Die Chemotherapie des fortgeschrittenen Zervixkarzinoms. Ein Uberblick, Tumor Diugn. 7, 89-98 (1986). 9. Crombach, G., Zippel, H. H., and Wtirz. H. Erfahrungen mit CA 125, einem Tumormarker fur maligne epitheliale Ovarialtumoren, Gehurtsh. Frauenheilk. 45, 205-212 (1985). 10. Kreienberg, R. Moglichkeiten und Grenzen der Tumormarkeruntersuchungen in der Nachsorge bei Ovarialkarzinom-Patientinnen, Gyniikologie 19, 178-185 (1986). 1I. Meier, W., Stieber. P.. Fateh-Moghadam, A., Eiermann, W., and Hepp, H. CA 125 Serumwerte und histologischer Befund zum Zeitpunkt der Second-look-Laparotomie beim Ovarialkarzinom, Geburtsh. Frauenheilk. 48, 331-333 (1988). 12. Niloff. J. M., Knapp, R. C., Lavin, P. T.. Malkasian, G. D., Berek, J. S., Mortel, R., Whitney, C.. Zurawski, V. R., and Bast, R. C. The CA I25 assay as a predictor of clinical recurrence in epithelial ovarian cancer. Amer. /. Ohstet. Gynecol. 155, 56-60 (1986). 13. Crombach, G., Wurz, H., and Bolte, A. Bestimmungen des SCCAntigens im Serum von Patientinnen mit Zervixkarzinom, Gehurtsh. Fruuenheilk. 47, 439-445 (1987). 14. Geyer, H., Schworer, D., and Pfleiderer. A. SCC (Squamous Cell Carcinoma)-Antigen als Tumormarker bei Zervixkarzinomen, Gehurtsh. Frouenheilk. 49, 266-271 (1989).
ANTIGEN
AND CHEMOTHERAPY
11
15. Kato, H., Tamai, K., Nagaya, T., Nagai, M., and Torigoe, T. The use of tumor antigen TA-4 for the management of squamous cell carcinoma, Cancer Detect. Prevent. 8, 155-157 (1985). 16. Maruo, T., Shibata, K., Kimura, A., Hoshina. M., and Mochizuki. M. Tumor-associated antigen, TA-4, in the monitoring of the effects of therapy for squamous cell carcinoma of the uterine cervix, Cancer 56, 302-308 (1985). 17. Meier, W., Stieber, P., Fateh-Moghadam, A., Eiermann, W., and Hepp, H. Erfahrungen mit dem Tumormarker SCC, insbesondere zur Therapieerfolgskontrolle beim Zervixkarzinom (OP und Radiatio), Gynukol. Rdsch. 27, 50-51 (1987). 18. Meier, W., Stieber, P., Fateh-Moghadam, A., Eiermann, W., and Hepp, H. Erfahrungen mit SCC-Antigen, einem neuen Tumormarker fur Karzinome der Cervix uteri, Geburtsh. Fruuenheilk. 49, 625-629 (1989). 19. Eiermann, W., Kuhnle, H., Meerpohl, H. G., Achterrath, W., Lenaz, L., and Preusser, P. Phase II trial with carboplatin/ ifosfamid in previously untreated patients with advanced squamous cell carcinoma of the cervix, Proc. Amer. Sot. Clin. Oncol. 8, 159 (1989). 20. WHO Handbook for Reporting Results of Cuncer Treatment, World Health Organisation, Geneva, Offset Publication 48, p. I (1979). 21. Bonomi, P., Blessing, J. A., Stehman, F. B., Disaia, P. J., Walton, L.. and Major, F. J. Randomized trial of three cisplatin dose schedules in squamous cell carcinoma of the cervix: A Gynecologic Oncology Group study, J. C/in. On&. 3, 1079-1085 (1985).
38, 6-11 (1990)
Squamous Cell Carcinoma Antigen and Carcinoembryonic Antigen Levels as Prognostic Factors for the Response of Cervical Carcinoma to Chemotherapy W.
MEIER,
W.
P. STIEBER,* A.
EIERMANN,
FATEH-MOGHADAM,*
A. SCHNEIDER, AND H. HEPP
Departments of Obstetrics and Gynecology and *Clinical Chemistry, Klinikum Grosshadern, Ludwig-Maximilians-University, Marchioninistrasse 15, 8000 Munich 70, West Germany Received September 5, 1989; revised November 2, 1989
With the introduction of tumor markers in gynecological oncology, it has become possible to identify tumor recurrence early in the follow-up of some types of tumor and thus to start the necessary treatment early [9-121. In the case of cervical carcinoma, squamous cell carcinoma antigen (SCC) and carcinoembryonic antigen (CEA) are two markers which are available as diagnostic tools-above all in combination [ 13- 181. Our study aimed to answer the question whether SCC and CEA levels can be used as a prognostic factor for tumor response to chemotherapy, and thus whether a developing resistance to cytostatic drugs can be determined early. In these cases, the patients could be saved from the continuation of an ineffective treatment with very severe side effects,
Between January 1986 and December 1988, 36 patients with primary advanced or recurrent cervical carcinoma were treated with cytostatic drugs in our department. Treatment at first was a combination of cisplatin and etoposide. After August 1987, a combination of carboplatin and ifosfamide was used. In all patients showing primary response to therapy, the squamous cell carcinoma antigen @CC) and carcinoembryonic antigen (CEA) levels fell rapidly to normal after one or two cycles. In contrast, clinical remission was not obtained in those patients with levels which remained high or rose again following an initial decrease. Chemotherapy is often the only available therapy for advanced cervical carcinoma or recurrent disease, although the results of treatment, especially in squamous cell carcinoma, remain poor. The course of the SCC or CEA levels can help to decide whether the patient would profit from a continuation of the therapy. With the tumor markers, treatment can be individualized so that, above all, cases of therapy failure or further tumor progression can be detected early and the patient can be spared the severeside effects of the treatment. 0 1990 Academic Press, Inc.
PATIENTS AND METHODS Between January 1986 and December 1988,36 patients with cervical carcinoma were treated with cytostatic drugs in our department. In 11 patients with stage IVB and one patient with stage IVA disease, primary chemotherapy was given following histological diagnosis at staging laparotomy or by scalenus biopsy. Three patients were treated with adjuvant chemotherapy following Wertheim-Meigs radical hysterectomy after the extent of the disease became known histologically (e.g., blood vessel involvement, lymphangiosis carcinomatosa). Twenty-one patients received chemotherapy because of recurrent disease: In 8 cases, the recurrence was local and in 13 cases there were distant metastases. Both vaginal stump and pelvic wall disease were classified as local recurrences. The primary therapy was in 13 cases a Wertheim-Meigs radical hysterectomy, with postop-
INTRODUCTION Almost 35% of patients with cervical carcinoma develop a recurrence usually within 2 years of surgery or radiotherapy [l-3]. Often, the only possible treatment available for these patients and those who present with metastatic disease is chemotherapy. Since the introduction of cisplatin, remission rates of more than 50% have been described, including patients with complete remission. The length of remission remains, however, relatively short [l-5]. Above all, because of the severe side effects of the treatment, the question often arises, whether in the individual case chemotherapy is justified, especially as there are no reliable methods of chemosensitivity testing in cervical carcinoma to date [6-81.
6 owe-8258/90 $1.50 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
SQUAMOUS CELL CARCINOMA
erative radiotherapy in 11 cases. Eight patients were treated primarily with combined radiotherapy. Eight patients had stage IB, one stage IIA, seven stage IIB, four stage IIIB, and one stage IVA disease. There was no relation between the localization of recurrence and the stage of disease or the type of primary therapy. Chemotherapy consisted of a combination of cisplatin 80 mg/qm Day 1 and etoposide 150 mg/qm Days 3 to 5 up to July 1987. From August 1987, carboplatin 300 mg/qm and ifosfamide 5 g/qm both on Day 1 were used [19]. The chemotherapy cycles were applied every 4 weeks dependent on blood count. The success of the treatment was judged according to the WHO recommendations [20]. Tumor marker levels were taken every 1 to 2 months. The serum levels of SCC were determined using a radioimmunoassay (Abbott). CEA levels were determined with an enzyme immunoassay (Abbott). The upper limit of normal was taken as 2 rig/ml for SCC and 3 rig/ml for CEA. RESULTS Of the 12 patients with primary chemotherapy, 9 had a raised SCC level and 8 a raised CEA level before beginning therapy (Table 1). In all 12 patients, either SCC, CEA, or both markers were raised. In patients with recurrent disease, SCC and CEA were each elevated in 13 cases, and in a total of 19 patients, at least one of the marker levels was raised. There were no percentage differences in raised tumor marker levels between patients with local recurrence and patients with distant metastases. In three patients who received adjuvant chemotherapy because of the histological extent of disease following radical hysterectomy, the initial SCC levels and in one patient the CEA level were also raised. Postoperative levels were within the normal range, so that in these three cases, the tumor markers could not be used to monitor the success of chemotherapy. Overall, 31 of 33 patients with either primary chemotherapy or chemotherapy for recurrent disease showed raised SCC or CEA levels. In 12 patients, both markers were raised. Only two patients did not have raised levels TABLE 1 Total Number of Elevated XC and CEA Levels before Beginning Chemotherapy see Primary therapy Recurrence Adjuvant therapy Total
N
(>2 rig/ml)
CEA (>3 rig/ml)
Elevated SCC and/or CEA
12 21 3 36
9 13 0 22
8 13 0 21
12 19 0 31
ANTIGEN
7
AND CHEMOTHERAPY
TABLE 2 SCC and CEA Levels in Relation to Histology
see Squamous cell carcinoma Adenocarcinoma Adenosquamous carcinoma
N
(>2 rig/ml)
CEA (>3 rig/ml)
Elevated SCC and/or CEA
26
19
14
24
5 2
2 I
5 2
5 2
before beginning chemotherapy. This made it possible to evaluate the results of chemotherapy in the majority of cases, using the tumor marker levels. Histologically, 26 patients had a squamous cell carcinoma, 5 patients an adenocarcinoma, and 2 patients an adenosquamous carcinoma (Table 2). There were no significant differences with respect to the number of raised SCC and CEA levels and the difference in patients’ distribution in the individual groups; however, all patients with adenocarcinoma showed raised CEA levels. In cases of squamous cell carcinoma, SCC levels were raised in 19 cases and CEA in 14 cases. Combined, either SCC or CEA were raised in 24 of 26 patients. Both patients with adenosquamous carcinoma demonstrated raised marker levels. The three patients, who received adjuvant chemotherapy all had a squamous cell carcinoma, with raised SCC levels in all three cases and a raised CEA in one patient. To judge the success of treatment, only those patients were included in whom at least three cycles of chemotherapy had been completed. In two patients, the chemotherapy was discontinued after one cycle; in one case because of poor tolerance of side effects, in the other because of rapid tumor progression. The outcome of therapy could be assessed in the remaining 31 patients. There was complete remission in 3 cases (10%). In 11 cases (35%) a partial remission and in 17 cases a primary progression were seen. Of these 17 women with progressive disease, in 5 the disease remained stable initially (“no change”). However, in each of these 5 cases, progression occurred within 3 months and they were therefore classified as having progressive disease. If cases of complete and partial remission are grouped together, 45% fall into this category. There were no differences in response with respect to the localization of the recurrence or the form of primary treatment. Table 3 shows the primary response to chemotherapy in relation to tumor marker levels. In two cases, neither SCC nor CEA levels were raised at first, and two further patients received only one treatment cycle, so that 29 of the 33 patients with therapeutical chemotherapy remained for assessment. Thirteen patients showed an in-
8
MEIER ET AL.
The semilogarithmic scale shows that in no case did the SCC level fall below 2 rig/ml. The levels begin to climb again steeply as early as 1 or 2 months after beginning treatment. Figure 2 shows six of these patterns. Decreasing Increasing Tumor markers can also be helpful in the often difficult SCCfCEA SCC/CEA values values assessment of tumors with bone metastases. Figure 3 shows the course of SCC in a 54-year-old patient who 0 Remission 13 developed bone metastases 14 months after primary ra16 0 Progressive disease diotherapy of a cervical carcinoma stage IIIB. After induction of chemotherapy, the SCC levels fell rapidly in normal and provided evidence of remission. The radiolitial response to the therapy with either a complete or ogical controls and bone scintigrams confirmed the parpartial remission. In all these cases, the tumor marker tial remission. Nine months later, the tumor marker levvalues dropped quickly to normal levels. None of these els began to rise again despite chemotherapy. Two patients demonstrated clinical progression at the same months later, the relapse was detectable radiologically, time as having normal tumor marker values. On the other and chemotherapy was stopped. hand, none of the 16 patients with increasing tumor marker levels showed clinical remission at this time. DISCUSSION Figure 1 shows tumor marker levels in five patients Despite the introduction of new cytostatic drugs, the with primary response to chemotherapy. In all cases, the XC levels were raised before beginning treatment. Fol- results of treatment in patients with recurrent disease or lowing one or two therapy cycles, the tumor marker primary metastatic disease in cervical carcinoma have levels can be seen to drop off to normal levels. The only been fractionally improved [l-3,8]. With the use of semilogarithmic scale clearly shows the rapid decrease platinum-containing compounds in single or combined in SCC levels to less than 2 rig/ml. Following remission schedules, up to 50% partial or even complete remissions periods of in some cases more than 6 to 8 months, the have been achieved. The period of remission has relevels begin to increase again. Shortly afterward, the mained relatively short [l-51. In our study, a remission progression was also confirmed clinically. Similar tumor rate of 45% was obtained, including three patients with complete remission. Patients with initial response to marker levels can be seen with CEA. In patients with primary progression during chemo- chemotherapy have significantly better survival rates therapy, only a short-term reduction in SCC levels can than patients with primary progression [ 1,3,8]. The combe seen, which, however, do not return to normal levels. plications due to tumor such as anus praeter, renal fisTABLE 3 Correlation of Tumor Marker Levels and Primary Response or Primary Progression during Chemotherapy
SCC(ng/ml) 1ooE
I
I
1
I
I
I
I
I
I
I
I
I
I
I
8
9
10
11
12
13
14
al
0
1
2
3
4
5
7 6 Months
FIG. 1. SCC levels of five patients with primary response to chemotherapy (semilogarithmic scale).
SQUAMOUS
100
CELL
CARCINOMA
ANTIGEN
9
AND CHEMOTHERAPY
SCC(ng/ml)
1’ 0
I 1
I 2
I 3
I 4
1 5
I 6 Months
I 7
0 8
1 9
I 10
I 11
1 12
FIG. 2. SCC levels of six patients with primary progression during chemotherapy (semilogarithmic scale).
tula, and intractable pain from deep pelvic recurrence occur significantly later in the course of the disease [2,3,5,8]. The price of this improvement in the quality of life due to reduction in tumor complications is paid for with the severe side effects of the chemotherapy, the incidence of which has not been much reduced despite dosage modification, change of interval, and use of highdose antiemetics and sedatives. In view of the short remission times and the known side effects of the drugs,
the question often arises as to whether or not the use of chemotherapy can be justified. There has been no reliable parameter available up until now to help monitor the therapy results, and demonstrate a response or signal a developing resistance to chemotherapy. A reliable method of in vitro chemosensitivity testing for cervical carcinoma has not yet been developed [6-81. The methods described in the literature are unsatisfactory in cases of cervical carcinoma [6,7]. The
SCC(ng/ml) 40 1 35 Begin Chemo
End Chemo
/
30 25 20 -
RI.,54 Y. CC IIIB
0 0
1 I
1 2
I 3
I 4
1 6
1 8
I I 7 8 Month8
I 9
1 10
I 11
1 12
1 13
I 14
FIG. 3. SCC levels during chemotherapy of a patient with bone metastases.
I 16
10
MEIER ET AL.
introduction of tumor markers, especially SCC, has meant that a parameter has become available that can indicate at a relatively early point whether the patient is responding to therapy [15-171. In 73 and 54% of cases of squamous cell carcinoma of the cervix, there are elevated SCC and CEA levels, respectively. In contrast, CEA can be seen to be superior to SCC in cases of adenocarcinoma [ 13,181. Prerequisite for the use of tumor markers for monitoring effects of therapy are the initial raised serum levels. In our patient collection, all 12 patients with primary chemotherapy showed a raised SCC or CEA level: SCC was raised in 75%, CEA in 66%. In recurrent disease 62% each of SCC and CEA levels were pathological, while 19 of 21 patients showed a raised SCC or CEA level. There was no significant difference between cases with local recurrent disease and those with distant metastases. These results agree with earlier studies and the results of other groups [ 13-181. As a result, monitoring the results of treatment or the response to chemotherapy using tumor markers becomes possible. In our patients, the positive predictive value and the negative predictive value for response to therapy was 100%. In those patients who are responding to therapy, the SCC levels fall to normal after two treatment cycles at the latest. There is a good correlation between the tumor marker levels and the course of the disease. After stopping treatment because of complete or partial remission, the values begin to climb again after 8-12 months. In our opinion, at this time chemotherapy should be stopped or, following a treatment break after complete remission, restarted to try to induce remission again. The optimal length of a time a chemotherapy should be continued after complete remission is not known [1,2,8]. Piver describes a chemotherapy schedule with cisplatin which was given over the period of 5 years to patients with cervical carcinoma and was tolerated without excessive side effects [3]. The patients received a total of 1325-3520 mg cisplatin during the treatment. A moot point arises when tumor marker levels remain within normal limits for several months and clinically the patient is in complete remission: Is maintenance chemotherapy justified? It would be more tolerable for the patient to begin again with the chemotherapy when the tumor marker levels begin to increase. Further studies are necessary to evaluate the percentage of patients which will have a second response to chemotherapy in these cases. In cases of primary progression, the SCC and CEA levels decrease for a short time initially, but never approach normal values. In all cases, the levels increase steeply after 2-3 months. In all these cases, chemotherapy should be discontinued, even if the clinical and
radiological examinations have not yet demonstrated the progression. Supportive measures should be introduced. Above all, in cases which are clinically difficult to assess, additional treatment monitoring using tumor marker levels would appear to be useful. Included in this category are bone metastases and histologically proven pelvic recurrence following surgery and radiotherapy [8]. The course of the tumor marker levels can help with an early decision as to whether a patient can be helped by continuing treatment. In advanced cervical carcinoma or in the case of recurrent disease, the cytostatic treatment often remains the only possible treatment. On the other hand, the success rate of chemotherapy with respect to the length of remission is not outstanding, especially in cases of squamous cell carcinoma. The course of the SCC and CEA levels can help to decide early whether patients will profit from continuation of treatment. If the levels are not within the normal range after, at the latest, three cycles of treatment, the chemotherapy should be discontinued. A rapid decrease in tumor marker levels to normal indicates primary response to therapy and treatment can be continued while monitoring tumor marker levels further. In our opinion, on complete remission, chemotherapy should be stopped if the tumor marker levels are normal. It should be restarted only when the tumor marker levels begin to increase again. Thus, with the help of tumor markers, individualized treatment can be given, whereby nonresponse or relapse can be diagnosed early and the patient saved from the severe side effects of the therapy. REFERENCES 1. Lele, S. B., and Piver, M. S. Weekly cisplatin induction chemotherapy in the treatment of recurrent cervical carcinoma, Gynecol. Oncol. 33, 6-8 (1989). 2. Piver, M. S., Barlow, J. J., Lele, S. B., and Maniccia, M. Weekly
cis-diamminedichloroplatinum II as induction chemotherapy in recurrent carcinoma of the cervix, Gynecol. Oncol. 18, 313-319 (1984). 3. Piver, M. S., Lele, S. B., Patsner, B., and McPhee, M. E. 1.5-5
years of uninterrupted cis-diamminedichloroplatinum II chemotherapy for metastatic cervical cancer, Gynecol. 0~01. 27, 24-33 (1987). 4. Rotmensch, J., Senekjian, E. K., Javaheri, G., and Herbst, A. L. Evaluation of bolus cisplatinum and continuous 5-fluorouracil infusion for metastatic and recurrent squamous cell carcinoma of the cervix, Gynecol. Oncol. 29, 76-81 (1988). 5. Thigpen, J. T., Blessing, J. A., Disaia, P. J., Fowler, W. C., and Hatch, K. D. A randomized comparison of a rapid versus prolonged (24 hr) infusion of cisplatin in therapy of squamous cell carcinoma of the uterine cervix: A Gynecologic Oncology Group study, Gynecol. Oncol. 32, 198-202 (1989). 6. Rockwell, S. Effects of clumps and clusters on survival measurements with clonogenic assays, Cancer Res. 45, 1601-1607 (1985).
SQUAMOUS CELL CARCINOMA 7. Umbach, G. E., Poethen, J., v. Matthiessen, H., Bender, H. G.. and Koldovsky, U. Experiences with the human tumor colonyforming assay in gynecologic malignancies. J. Cuncev Res. Clin. Oncol. 110, 234-237 (1985).
8. Umbach, G. E., v. Matthiessen, H., and Bender, H. G. Die Chemotherapie des fortgeschrittenen Zervixkarzinoms. Ein Uberblick, Tumor Diugn. 7, 89-98 (1986). 9. Crombach, G., Zippel, H. H., and Wtirz. H. Erfahrungen mit CA 125, einem Tumormarker fur maligne epitheliale Ovarialtumoren, Gehurtsh. Frauenheilk. 45, 205-212 (1985). 10. Kreienberg, R. Moglichkeiten und Grenzen der Tumormarkeruntersuchungen in der Nachsorge bei Ovarialkarzinom-Patientinnen, Gyniikologie 19, 178-185 (1986). 1I. Meier, W., Stieber. P.. Fateh-Moghadam, A., Eiermann, W., and Hepp, H. CA 125 Serumwerte und histologischer Befund zum Zeitpunkt der Second-look-Laparotomie beim Ovarialkarzinom, Geburtsh. Frauenheilk. 48, 331-333 (1988). 12. Niloff. J. M., Knapp, R. C., Lavin, P. T.. Malkasian, G. D., Berek, J. S., Mortel, R., Whitney, C.. Zurawski, V. R., and Bast, R. C. The CA I25 assay as a predictor of clinical recurrence in epithelial ovarian cancer. Amer. /. Ohstet. Gynecol. 155, 56-60 (1986). 13. Crombach, G., Wurz, H., and Bolte, A. Bestimmungen des SCCAntigens im Serum von Patientinnen mit Zervixkarzinom, Gehurtsh. Fruuenheilk. 47, 439-445 (1987). 14. Geyer, H., Schworer, D., and Pfleiderer. A. SCC (Squamous Cell Carcinoma)-Antigen als Tumormarker bei Zervixkarzinomen, Gehurtsh. Frouenheilk. 49, 266-271 (1989).
ANTIGEN
AND CHEMOTHERAPY
11
15. Kato, H., Tamai, K., Nagaya, T., Nagai, M., and Torigoe, T. The use of tumor antigen TA-4 for the management of squamous cell carcinoma, Cancer Detect. Prevent. 8, 155-157 (1985). 16. Maruo, T., Shibata, K., Kimura, A., Hoshina. M., and Mochizuki. M. Tumor-associated antigen, TA-4, in the monitoring of the effects of therapy for squamous cell carcinoma of the uterine cervix, Cancer 56, 302-308 (1985). 17. Meier, W., Stieber, P., Fateh-Moghadam, A., Eiermann, W., and Hepp, H. Erfahrungen mit dem Tumormarker SCC, insbesondere zur Therapieerfolgskontrolle beim Zervixkarzinom (OP und Radiatio), Gynukol. Rdsch. 27, 50-51 (1987). 18. Meier, W., Stieber, P., Fateh-Moghadam, A., Eiermann, W., and Hepp, H. Erfahrungen mit SCC-Antigen, einem neuen Tumormarker fur Karzinome der Cervix uteri, Geburtsh. Fruuenheilk. 49, 625-629 (1989). 19. Eiermann, W., Kuhnle, H., Meerpohl, H. G., Achterrath, W., Lenaz, L., and Preusser, P. Phase II trial with carboplatin/ ifosfamid in previously untreated patients with advanced squamous cell carcinoma of the cervix, Proc. Amer. Sot. Clin. Oncol. 8, 159 (1989). 20. WHO Handbook for Reporting Results of Cuncer Treatment, World Health Organisation, Geneva, Offset Publication 48, p. I (1979). 21. Bonomi, P., Blessing, J. A., Stehman, F. B., Disaia, P. J., Walton, L.. and Major, F. J. Randomized trial of three cisplatin dose schedules in squamous cell carcinoma of the cervix: A Gynecologic Oncology Group study, J. C/in. On&. 3, 1079-1085 (1985).
