Adjuvant chemotherapy after radical hysterectomy for cervical carcinoma

GYNECOLOGIC

ONCOLOGY

35, 193-198 (1989)

Adjuvant Chemotherapy after Radical Hysterectomy for Cervical Carcinoma CHYONG-HUEY

LAI,

M.D.,” TSONG-SHAN LIN, M.D.,* YUNG-KUEI HSIN-FU CHEN, M.D., PH.D.+

SOONG, M.D..

M. PHIL.,“.’

AND

Received July 14. 1988

than squamous histology, and extension to uterine corpus and vagina are frequently associated with increased risk of recurrence [I-II]. Among the poor prognostic factors, however, lymph node metastasis is probably the most important and consistent finding linked with grave outcomes [l-5,12-14]. Adjuvant radiation therapy was recommended in treating high-risk patients with variable results. In general, local control was improved, but long-term survival remained unchanged due to distant metastases [4,12,13]. With regard to the high response rates of recurrent and advanced cervical carcinoma to cisplatin-based combination chemotherapy [15-191, and our previous disappointing experience with adjuvant radiotherapy [20], we tried to use chemotherapy alone as adjuvant therapy for patients at high risk of recurrence. This report presents the results of our pilot study.

Three hundred and sixty-eight cases of invasive cervical cancer (stage IB through early stage IIB) were treated with radical abdominal hysterectomy and bilateral pelvic lymphadenectomy at Chang Gung Memorial Hospital. Of these patients, 172 were classified postoperatively as a high-risk group after surgical-pathological assessment of tumor extent. Among these high-risk patients, 40 received adjuvant chemotherapy with cisplatin, vinblastine, and bleomycin (PVB), 38 received adjuvant radiotherapy, and 79 refused adjuvant treatment. The 3-year cumulative disease-free survival rate was 91.6% for the low-risk group and 59.7% for those at high risk. Among patients in the high-risk group, the 3-year survival rate was 75.0% for patients treated with adjuvant chemotherapy and 46.8% for those not treated with adjuvant therapy (P < 0.05). The preliminary results of this pilot study showed a significant activity of adjuvant chemotherapy, which warrants further investigation of its role in the treatment of cervical cancer. ‘C: 1989 Academic Press. Inc. INTRODUCTION

MATERIALS

a well-established mode of therapy for early cervical cancer. The results have been comparable to those for radiation therapy [ 11. The surgical complication rate is reasonably low. Advantages of the primary surgical approach include (1) precise definition of the extent of disease, which provides additional information in determining further therapy; (2) preservation of ovarian function for young females; and (3) avoidance of long-term radiation injuries to normal tissue. Many factors influence the prognosis. Several investigators have suggested that lymph node metastasis, parametrial invasion, lymphovascular space tumor emboli, bulky tumor, deep cervical stromal penetration other Radical

hysterectomy

is

AND METHODS

From June 1983 to December 1985, 368 patients were treated at Chang Gung Memorial Hospital with radical hysterectomy and bilateral pelvic lymphadenectomy for primary cervical cancer, Stage IB through early Stage IIB. Surgical procedure and radicality were comparable to those for the Meigs operation. Stages were assigned according to the FIG0 clinical staging system (192 in stage IB, 15 in stage IIA, 161 in early stage IIB). Of these patients, 327 (88.9%) had squamous carcinoma, 39 (10.6%) adenocarcinoma or adenosquamous carcinoma, and 2 small cell undifferentiated carcinoma. One hundred and seventy-two patients were categorized as the high-risk group using the criteria listed in Table 1. The remaining 196 cases were classified as the low-risk group. When the final surgical pathology was carefully evaluated, the increased risk over that of the

’ To whom reprint requests should be addressed. 193

All

0090-8258189 $1.50 Copyright 0 1989by AcademicPress.Inc. rights

of reproduction

in any

form

reserved.

194

LA1 ET AL.

TABLE 1 Indications of Adjuvant Therapy 1. Positive pelvic lymph nodes 2. Parametrial extension 3. Vaginal or corpus extension + cervical stromal penetration > twothirds thickness + lymphatic permeation 4. Adenocarcinoma, adenosquamous carcinoma, or small cell undifferentiated carcinoma + cervical stromal penetration > one-half thickness

low-risk group, the need for adjuvant therapy (although not yet proven to be efficacious), and the complications of adjuvant chemotherapy and radiotherapy were explained to each patient. The patient herself decided whether to receive adjuvant therapy and which mode of treatment. Seventy-nine patients refused any adjuvant therapy. Forty-four patients accepted chemotherapy, but four of them dropped out of treatment after fewer than three courses of treatment. In the beginning, 41 accepted radiotherapy but 3 of them did not receive the full schedule. Five patients had a positive surgical margin and were administered postoperative radiotherapy. One had a ureterovaginal fistula; hence adjuvant therapy was not given within 3 months postoperation. Two received preoperative radiotherapy. In total, 15 cases were excluded from these three groups for the ineligibility mentioned above, but they are included in analysis of the whole series. Therefore, 40 cases receiving at least three courses of chemotherapy served as the study group and 79 cases without adjuvant therapy as the control group. Evaluation of adjuvant radiotherapy was not the purpose of this study; moreover, those who received radiotherapy actually were at greater risk than the other two groups. Hence, a detailed analysis is not provided for them. Patient characteristics in this study were listed in Table 2. TABLE 2 Patient Characteristics of Study Groups

Age Range Mean Stage IB II Lymph node positive Parametrium invaded Vagina invaded Cervical penetration Xwo-thirds thickness Lymphatic permeation Special histology

Chemotherapy (N = 40)

No adjuvant therapy (N = 79)

Total (N = 119)

24-65 45.1

30-73 53.0

24-73 50.4

I1 29 34 6 IO

22 57 24 I8 37

33 86 68 24 47

26 28 4

60 39 I8

86 67 22

Our protocol for adjuvant chemotherapy is as follows: cisplatin 60 mg/m2 IV drip on Day 1 with vigorous hydration and diuresis, vinblastine 4 mg/m’ IV bolus on Days 1 and 2, bleomycin 15 mg IV drip over 24 hr on Day 2. Treatment interval was 3 weeks. Eight courses of treatment were planned. The pretreatment workup included chest X-ray, ECG, complete blood count, platelet count, SGOT, BUN, creatinine, sodium, potassium, and chloride. Pulmonary function tests and creatinine clearance were not routinely done, but were arranged when indicated clinically. The prerequisites for administering chemotherapy were a white count above 3000/mm3, a platelet count greater than 100,000/mm3, and serum creatinine less than 2 mg/dl. If these values were not reached, chemotherapy was delayed 1 week. If the white count was still below 3000/mm3, the vinblastine dose was cut 50%, but the cisplatin and bleomycin doses remained unchanged. No therapy was given when the white count dropped below 2000/mm3 and the platelet count below 50,000/mm3. A history was taken and physical and pelvic examinations performed on each patient on admission. Vaginal Pap smears were taken every 2 months and an intravenous pyelogram was performed 6 months postoperation. After completion of their chemotherapy, the patients were followed on an outpatient basis every 2 months in the first 2 years, every 4 months the next 2 years, and every 6 months after 4 years. The disease-free survival rate was calculated with the Kaplan-Meier life-table method, and differences between individual groups were analyzed with the log-rank test. Differences in group means were calculated with Student’s t test and differences in ratios by the x2 test or Fisher’s exact test. RESULTS The follow-up period extended from the primary surgical treatment to the date of confirmed recurrence or loss to follow-up. Patients recurred as early as 3 months; the minimal follow-up for the disease-free patients was 6 months, the longest follow-up 4 years 3 months, and the group mean 22 months. Clinical staging deviated from surgical-pathological staging as only 56.5% were compatible. All others were either under- or overstaged, especially in early stage IIB, in which only 36 in 161 had actually parametrial invasion proved by surgical pathology (Table 3). The cumulative disease-free survival rate for the whole series of 368 patients is 77.4% at 3 years, 91.6% for the low-risk group and 59.7% for the high-risk group (Fig. 1). The difference is statistically significant (P < 0.05). Within the high-risk group, the 3-year disease-free survival rate was 75.0% for those receiving adjuvant che-

ADJUVANT

CHEMOTHERAPY

FOR CERVICAL

195

CARCINOMA

TABLE 3 Compatibility of Clinical Stage with Surgical-Pathological Stage Surgical-pathological

stage

Clinical stage (FIGO)

IB

HA

IIB

Total

IB IIA IIB

160 2 84

20 12 41

12 1 36

192 15 161

Total

246

13

47

368

motherapy and 46.8% for those not (Fig. 2). The time to recurrence was longer (15.6 months) for the chemotherapy group than for those receiving no adjuvant therapy (10.5 months; 0.1 > P > 0.05). The relationship between rate of recurrence and use of chemotherapy is shown in Table 4. Only 19 patients completed eight courses, 9 patients received five to seven courses, and 12 patients received three or four courses. The crude recurrence rates are similar no matter which two are compared. When treatment failures are analyzed with respect to the site of recurrence (Table 5), it is apparent that the adjuvant chemotherapy group and the control group experienced similarly high rates of relapse in the pelvis (100% versus 92%). Two of five in the chemotherapy group had both pelvic and distant failures, but none of them had isolated distant metastases. When

loo- -Q--l.-.. ‘. I* ---.__ 90‘. ‘. --._---_ n. ‘.

cm-

-.

----$.-.

‘.

‘.

‘.

---_

-.._

-. =..

TO-

-.

-.

-.

‘.

60-

50-

40-

30.

-1.. --.*

‘. .‘-m

.---.

LOW

RISK

t---t

ALL

CASES

.--.

HIGH

1 YEARS

PRIMARY

2 AFTER

PRIMARY

3 TREATMENT

FIG. 2. Cumulative new survival rates of high-risk patients. with and without adjuvant chemotherapy.

the study groups are subdivided with regard to risk factors, use of adjuvant chemotherapy, and outcome (Table 6), it is obvious that the failed cases in the chemotherapy group all had multiple risk factors. Four of five had positive lymph nodes, parametrial spread, and vaginal involvement. All had deep cervical stromal invasion and lymphatic permeation. In contrast to the chemotherapy group, of those receiving no adjuvant therapy for their increased risk, 12 in 39 relapsed, although they had only one or two risk factors. In the chemotherapy group, 34 had pelvic lymph node metastases but only 4 of them recurred in 3 years, whereas 8 of 24 receiving no adjuvant therapy recurred. The toxicity of adjuvant chemotherapy was overall acceptable (Table 7). Although nausea and vomiting were distressing and frequently encountered in the earlier TABLE 4 Recurrence Rates and Performance of Adjuvant Chemotherapy” Chemotherapy course No.

Case No.

Recurrence No. (%)

8 5-7 3-4

19 9 12

3 (15.1) I (11.1) 1 ( 8.3)

40

5 (12.5)

3

2 AFTER

0.05

I

RISK

1 YEARS

P c

TREATMENT

FIG. 1. Cumulative NED survival (with all cases included).

Total ” All P > 0.05.

196

LA1 ET AL.

TABLE 5 Sites of First Recurrence Site of recurrence

TABLE 7 Toxicity of Adjuvant Chemotherapy (N = 40)

No adjuvant therapy (N = 79)

Chemotherapy (N = 40)

Central Pelvis only Pelvis + distant Distant only

2 (40) 1 (20) 2 (40) 0

12 (48) 9 (36)

Total

5 (1W

25 (100)

Toxicity Total (N = 119) 14 (47) IO (33) 4 (13)

2 ( 8) 2 ( 8)

2 ( 6) 30 (100)

phase of this study, these became less problematic after the introduction of high-dose metoclopramide combined with dexamethasone and diphenhydramine. Alopecia was generally moderate, and recovery was quick after discontinuation of chemotherapy. Clinical pulmonary toxicity was not noted in this series, perhaps because bleomycin was given only on Day 2, so the accumulated dose was low. Renal function impairment was rare with adequate hydration and diuresis. Some patients developed severe myelosuppression. Six patients had a white count below 1000/mm3; 12 had a hemoglobin less than 8 g/dl; and 7 had a platelet count less than 50,000/mm3 and 1 less than 20,000/mm3. Seven patients developed infections, but no treatment-related deaths occurred. DISCUSSION Cervical cancer is the most frequent female malignancy and the leading cause of female cancer deaths in Taiwan. For many years, radical hysterectomy and pelvic lymphadenectomy constituted the preferred mode of treatment for localized invasive cervical cancer (clinical stage IB through early stage IIB) in this area. So, we

Nausea, vomiting Alopecia Infection Renal function impairment Hematology WBC Hemoglobin Platelet count

Grade

Case No.

Mild-moderate Severe Mild-moderate Severe

25 10 30 8 I

Mild

1


Death

6 14 12 21 6 1 0

have a unique opportunity to explore the role of surgical treatment of cervical cancer with extensive experience. Early cervical cancer can be cured at around 80% with either primary surgery or radiotherapy [l]. Nevertheless, not much improvement has been noted in the treatment of this disease worldwide in the past 20 years. Pelvic lymph node metastasis caused a 40 to 50% reduction in 5-year survival [ 1,4-6]. Parametrial involvement, lymphatic permeation, bulky tumor, and other conditions are also frequently reported as risk factors for recurrence [3,4-71. Usually, adjuvant radiotherapy is given with the hope of improving the possibility of cure. Nevertheless, many investigators have concluded that although local control is improved, distant metastases make the longterm survival the same and adjuvant chemotherapy might be considered in the treatment of microscopic systemic disease [4-6,7,11,13].

TABLE 6 Distribution of Risk Factors by Treatment Modalities and Outcomes No adjuvant therapy

Chemotherapy High-risk factor Positive nodes Parametrial spread Special histology Vagina involved Deep cervical invasion (atwo-thirds thickness) Lymphatic permeation Number of risk factors 1 2 3 34 L?R. recurrence.

R + ‘I (N = 5)

5 5

R(N = 35)

Total (N = 40)

R+ (N = 25)

R(N = 54)

Total (N = 79)

30 2 3 6

34 6 4 10

8 7 5 10

16 11 13 27

24 18 18 31

21 23

26 28

19 14

41 25

60 39

5

5 9 16 10

2 10 8 5

8 19 20 6

10 29 28 11

9 16 5

ADJUVANT

CHEMOTHERAPY

FOR CERVICAL

CARCINOMA

197

The failed cases in the so-called low-risk group were As cervical cancer was thought to be chemoresistant in the past, chemotherapy was employed only for re- also reviewed. All but one patient were at some, although current and advanced cases. Several agents such as cis- not very high, risk as defined by the selection criteria of this study. Three had deep cervical penetration of platin, bleomycin, mitomycin, adriamycin, and vincristine are reported to produce responses ranging from 10 tumor with lymphatic permeation. Two had deep cervical invasion only. Two had lymphatic permeation only. Two to 50% when used as single agents. Many combinations, especially cisplatin-based regimens, have resulted in a had large exophytic tumor with superficial invasion; one substantial number of complete remissions. Friedlander of them also had lymphatic permeation. In one patient, rt al. attained a 66% overall response rate and an 18% both corpus and vagina were invaded without lymphatic complete remission rate with the PVB regimen [l7]. It or deep cervical stromal invasion. One had corpus exis logical to have trials of combination chemotherapy on tension and lymphatic permeation. In only one patient were no risk factors identified. This subgroup should an adjuvant basis. better be defined as a median-risk group. These patients In reviewing the current literature on administration may need a treatment strategy different from those for of adjuvant chemotherapy after radical hysterectomy, only the Wertheim et al. study is available [ 191. They the low- and high-risk patients. Since this study was retrospective, background risk combined radiotherapy with a chemotherapy regimen of bleomycin and cisplatin. The continuous disease-free factors in either group were not stratified (Table 2). survival rate for the 32 evaluable patients was 84% at a Moreover, the choice of therapy was not randomized. mean and median follow-up time of 28 months. No con- These factors may bias the results. Our present report trol group was included in this study, so we do not know has its limitations, and the conclusion is tentative. Nevertheless, with the initial encouraging results, we are now whether this excellent result is due to chemotherapy, radiotherapy, or both. In Chang Gung Memorial Hos- conducting a prospective study with stratification and pital, we started to use chemotherapy alone as adjuvant randomization to explore the role of adjuvant chemotherapy for patients undergoing radical hysterectomy and therapy and radiotherapy in cervical cancer treated pripelvic lymphadenectomy who were identified to be at marily surgically. high risk of recurrence. ACKNOWLEDGMENTS The 3-year disease-free survival rate for patients who received adjuvant chemotherapy was significantly imThe authors thank J. T. Shen. M.D. (Columbia-Presbyterian Hosproved compared with patients not given the adjuvant pital, New York), and P. E. Schwartz. M.D. and J. M. Morris. M.D. chemotherapy (75% versus 46.8%. P < 0.05), and the (Yale University, New Haven, Connecticut), for kindly reviewing the time to recurrence was prolonged (15.6 months versus manuscript and making informative recommendations. 10.5 months) though not statistically significant. REFERENCES The chemotherapeutic regimen in this study was adopted from the Friedlander et trl. study, which showed I. Delgado. G. Stage IB squamous cancer of the cervix: The choice substantial activity [ 171; however, which combination is of treatment. Ohstrr. C;yrfe&. &,rr. 33, 174-183 (1978). optimal remains to be determined, as do the dose and 2. Noguchi, H., Shiozawa, I., Sakai, Y.. Yamazsaki. T., and Fukuta. schedule. We cannot tell from our study the number of T. Pelvic lymph node metastasis of uterine cervical cancer, G.vnrcol. Ond. 27, 150-15X (1987). courses of chemotherapy that is appropriate, because the _3 Van Nagell, J. R., Rayburn. W., Donaldson, E. S.. Hanson, M.. crude recurrence rates for different numbers of courses Gay. E. C.. Yoneda, J., Marayuma. Y.. and Powell. D. F. Therare similar (Table 4), and background risk factors were apeutic implications of patterns of recurrence in cancer of the somewhat more severe for those given more courses. uterine cervix. Crrncer 44, 2354-2361 (1979). Analysis of the failed cases revealed that pelvic re4. Jobson, V. W.. Girtanner. R. E.. and Averette, E. Therapy and currence was the main problem (Table 5). All recurrent survival of early invasive carcinoma of the cervix uteri with mecases had multiple risk factors (Table 6). Adjuvant chetastases to the pelvic nodes. Ohsr~r. G.vneco/. 151, 27-29 ( 1980). motherapy had poor local control for those with major 5. Figge. D. C., and Tamimi, H. K. Patterns of recurrence of carcinoma following radical hysterectomy. Amer. J. Ohster. Gynecol. local disease. No patients had distant metastases without 140, 213-220 (1981). concomitant pelvic recurrence. This may be due to the 6. Barber, H. R. K., Sommers, F. S. C., Rotterdam, H., and Kwon, efficacy of adjuvant chemotherapy in treatment of miT. Vascular invasion as a prognostic factor in stage IB cancer of croscopic systemic disease. Those who have lymph node the cervix, Ohsret. G.vned. 52, 343-348 (1978). metastases with tumor confined to the cervix probably 7. Piver, M. S.. and Chung, W. S. Prognostic significance of cervical could be helped by chemotherapy most, but those who lesion size and pelvic node metastases in cervical carcinoma, Ohhave lymph node metastases as well as extension to sret. Gynwol. 46, 507-510 (1975). parametrium, vagina, and corpus may need a combinax. Ballon, S. C., Berman, M. L., Lagasse, L. D., Petrilli, E. S., and Castaldo. T. W. Survival after extraperitoneal pelvic and paraaortion of chemotherapy and radiotherapy.

198

LA1 ET AL.

tic lymphadenectomy and radiation therapy in cervical carcinoma, Obsfet. Gynecol. 57, 90-95 (1981). 9. Krebs, H. B., Helmkamp, B. F., Sevin, B. U., Poliakoff, T. R., Nadji, M., and Averette, E. E. Recurrent cancer of the cervix following radical hysterectomy and pelvic node dissection, Obster. Gynecol. 59, 422-427 (1982). 10. Prempree, T., Patanaphan, V., Viravathana, T., Sewchand, W., Cho, Y. K., and Scott, R. M. Radiation treatment of carcinoma of the cervix with extension into the endometrium. A reappraisal of its significance, Cancer 49, 2015-2020 (1982). 11. Van Nagell, J. R., Donaldson, E. S., Wood, E. G., and Parker, J. C. The significance of vascular invasion and lymphocytic infiltration in invasive cervical cancer, Cancer 41, 228-234 (1978). 12. Russell, A. H., Tong, D. Y., Figge, D. C., Tamimi, H. K., Greer, B. E., and Elder, S. J. Adjuvant postoperative pelvic radiation for carcinoma of the uterine cervix: Pattern of cancer recurrence in patients undergoing elective radiation following radical hysterectomy and pelvic lymphadenectomy, Znt. J. Radiaf. Uncol. Biol. Phys. 10, 211-214 (1984). 13. Hogan, W. M., Littman, P., Griner, L., Miller, C. L., and Mikuta, J. J. Results of radiation therapy given after radical hysterectomy, Cancer 49, 1278-1285 (1982).

14. Himmelmann, A., Holmberg, E., Jansson, I., Oden, A., and Skogsberg, K. The effect of postoperative external radiotherapy on cervital carcinoma stage IB and HA, Gynecol. Oncol. 22, 73-84 (1985).

IS. Thigpen, T., Shingleton, H., Homesley, H., LaGasse, L., and Blessing, J. cis-Dichlorodiammineplatinum(I1) in the treatment of gynecologic malignancies: Phase II trials by the Gynecologic Oncology Group, Cancer Treaf Rep. 63, 1549-1555 (1979). 16. Daghestani, A. N., Hakes, T. B., Lynch, G., and Lewis, J. L., Jr. Cervix carcinoma: Treatment with combination cisplatin and bleomycin, Gynecol. Oncol. 16, 334-339 (1983). 17. Friedlander, M., Kaye, S. B., Sullivan, A., Atkinson, K., Elliott, P., Coppleson, M., Houghton, R., Solomon, J., Green, D., Russell, P., Hudson, C. N., Langlands, A. O., and Tattersall, M. H. N. Cervical carcinoma: A drug-responsive tumor-experience with combined cisplatin, vinblastine, and bleomycin therapy, Gynecol. Oncol. 16, 275-281 (1983). 18. Vogl, S. E.. Seltzer, V., Camacho, F., and Davis, T. E. Chemotherapy for recurrent and metastatic cervical cancer with standard-dose methotrexate-leucovorin, bleomycin and cis-platinum, J. Reprod. Med. 31, 327-329 (1986). 19. Wertheim, M. S., Hakes, T. B., Daghestani, A. N., Nori, D., Smith, D. H., and Lewis, J. L. A pilot study of adjuvant therapy in patients with cervical cancer at high risk of recurrence after radical hysterectomy and pelvic lymphadenectomy, J. C/in. Oncol. 3, 912-916 (1985). 20. Lai, C. H., Chen, H. F., and Lin, F. Z. Adjuvant radiotherapy after radical hysterectomy for cervical carcinoma-6 years’ experience, J. O&e?. Gynecol. R.O.C. 24, 24 (1985) (suppl. abstr.).