In: L Radmtmn Onco/o$~ Blol f’hn Vol. 20. pp. 927-932 Printed I” the U.S.A. All nghts reserved.
Copyright
0360.3016/91 $3.00 + .OO 8 1991 Pergamon Press plc
??Original Contribution
HIGH DOSE RATE INTRACAVITARY CERVICAL CANCER OF THE VAGINAL
BRACHYTHERAPY FOR RECURRENT STUMP FOLLOWING HYSTERECTOMY
HISAO ITO, M.D., HIDEO KUMAGAYA, M.D., NAOYUKI SHIGEMATSU, IKU NISHIGUCHI, M.D., TOSHITAKE NAKAYAMA, M.D. AND SHOZO HASHIMOTO, M.D. Department
of Radiology, Keio University, School of Medicine, 35 Shinanomachi,
Shinjukuku,
M.D.,
Tokyo 160 Japan
Forty-eight patients with recurrent cervical cancer of the vaginal stump following hysterectomy for cervical cancer, were treated with high dose rate intracavitary brachytherapy with or without external irradiation. The intervals between primary surgery and vaginal recurrences varied widely from 3 months to 36 years. Patients were classified into two groups, either with or without palpable tumor of the vaginal stump. Tumor size was determined by bimanual rectovaginal examination at the time of recurrence. Survival rates were 8% for the group with palpable tumors and 80% for those without, respectively. The survival rate of patients who did not have palpable masses and were treated with brachytherapy alone was not improved by combination with external irradiation. The time interval from the primary hysterectomy to the recurrence did not influence survival. These results suggest that the only significant prognostic factor for recurrent cervical cancer after hysterectomy is tumor size. The relationship between recurrent cervical cancer of the vagina1 stump and second primary vagina1 cancer is also discussed. High dose rate brachytherapy, Recurrence at the vaginal stump, Recurrent cervical cancer, Vaginal cancer, Intracavitary brachytherapy, Vaginal stump cancer.
etc. For example, central recurrences are salvaged easier than peripheral pelvic wall recurrences (5) and patients with postsurgical recurrences have more treatment options and a better prognosis than patients with recurrences developed after radiotherapy (6). In our institution, we treated with radiotherapy 48 patients who developed recurrent cervical cancer following simple or radical hysterectomy. Here, we report our analysis of the clinical prognostic factors in patients with recurrent cervical cancer of the vaginal stump.
INTRODUCTION
In Japan, periodic check-ups for uterine cancer have become a common practice for women over the age 30 years. As a result, the incidence of patients with early stages of cervical cancer has increased. The treatment modality for early cervical cancer has been a source of controversy among surgeons and radiotherapists (23). In Japan, pa-
tients present first at gynecologic clinics, and it is gynecologists in most institutions who determine the treatment modality without additional inputs of radiotherapists (10). Most gynecologists consider that surgical treatment is superior to radiotherapy, and, as a result, the majority of patients of stages 0-IIb become subjected to radical hys-
METHODS
terectomy and lymphadectomy. However, more recently there has been a trend to replace radical surgery with less radical procedures such as modified radical hysterectomy and simple hysterectomy to reduce the operative complications. A less radical surgery may be complicated with higher incidence of tumor recurrences at the vaginal stump. In general, recurrent tumors are difficult to manage and overall prognosis of the patients is poor ( 14, 15). The prognosis, however, depends on a number of factors, such as the site of recurrence, previous treatment modality,
AND
MATERIALS
Between January 1975 and August 1987, 48 patients were referred to the Department of Radiology at Keio University Hospital with cancer of the vaginal stump following hysterectomy for cervical cancer. All patients included in this study had squamous cell carcinoma both at the time of the original diagnosis and at the recurrence. The breakdown by stage at the time of the original diagnosis was as follows: Stage 0, 3 patients (6%); Stage I, 16 patients (33%); Stage II, 18 patients (37%); Stage III, 3 patients (6%); unknown, 9 patients (18%). Figure 1 illus-
Reprint requested to: Dr. Hisao Ito.
Accepted
927
for publication
9 November
1990.
928
I. J. Radiation Oncology 0 Biology 0 Physics
12
May 1, 1991, Volume 20, Number 5
withoutmas ECi patient 0
patientwith mass
:_ !
10 fn E at 3
_._
8
dose e&luati,tq point
:
(1 cm) b
6
z +
4
z’
,. 2 0 mder 30 31- 40 41- 50 51- 60 61_ 70 716 over
Age by Decades Fig. 1. Age by decades is shown for 48 patients cell carcinoma of the vaginal stump.
with squamous
the age distribution of the patients. Most of the patients in this series were in their 5th-6th decades. Patients were classified into two groups, those with palpable tumor masses and those without, according to the tumor size at the vaginal stump. Tumor size was determined by bimanual rectovaginal examination when recurrences were found. Tumor evaluation of the vaginal stump by CT scan was performed in about half of the patients; however, the findings were not consistent with the bimanual examination. In this study, information obtained from CT scan was not included in the tumor evaluation. When bimanual rectovaginal examination of the vaginal stump gave a questionable result, the patient was classified as not having a palpable mass. There was not difference in the age distributions between the patients with tumor masses at the vaginal stump and those without (Fig. 1). All recurring patients were treated with brachytherapy or a combination of brachytherapy and external irradiation. External irradiation, using a 6 MV photon beam, was given to all patients with palpable masses and to 10 of 28 patients without palpable masses. Whole pelvis irradiation (50 Gy) was delivered using conventional anterior-posterior parallel opposed fields. Following 30 Gy, a central shield (4 cm width) was inserted to cover the small pelvic cavity, to avoid overdosage of the bladder and the rectum (10). All patients in this series received high dose rate intracavitary brachytherapy using a Cesium137 source (7.4 X 10” Bq) for each colpostat. The irradiation dose to the vaginal stump was evaluated at a point that was 1 cm from the surface of the tumor (Fig. 2). The surface of the tumor was indicated on a roentgen film by metal chips inserted at the tumor margins. Patients received 5 Gy intracavitary irradiation twice a week to a total dose of 30 Gy in 6 fractions. As an alternative fractionation schedule, some patients received 6 Gy irradiation once a week to a total dose of 24 Gy.
trates
Fig. 2. The schema of hig dose rate brachytherapy. The metal chins were inserted into both sides of the tumor surface at the vaginal stump. The irradiated dose was evaluated at a depth of I cm from the surface of the vagina.
Survival analysis and survival using the Kaplan-Meier method.
curves were performed
RESULTS 1. Interval between primary surgery and tumor recurrence Figure 3 illustrates the cumulative frequency of cervical cancer recurrences from the time of radical or simple hysterectomy. The median time to recurrence was 3.3 years. Sixty-two percent of the recurrences occured within 5 years of the primary treatment. Fifteen of 19 patients originally diagnosed with Stage O-I disease failed within 5 years, and 14 of 2 1 patients with Stage II-III developed recurrences within the same period. Clinical stage at the original diagnosis did not influence the interval between
8 5 z z
40
g ‘E Q 1
20
5
0
+ 0 -A-
0
1 0
I
5
10
allpatients
” : 40 without mass n : 26 ll:20 with mass
1
15
20
L
25
30
35
Years after Radiotherapy Fig. 3. Cumulative incidence of recurrence time length after primary surgery.
in relation
to the
Brachytherapy for recurrent cervical cancer 0 H.
primary surgery and the recurrence. When the time intervals from primary surgery to recurrence were compared between the groups with and without palpable tumor masses, there was no difference. The cumulative incidence of the recurrences has a steep slope within first 5 years after surgery but thereafter the slope becomes shallower. 2. Survival rates Figure 4 shows actuarial survival of the patients, calculated by the Kaplan-Meier method. The 5- and IO-year survival rates for all patients were 56% and 47%, respectively. However, the survival was greatly influenced by the presence of palpable tumors at the vaginal stump. Patients without palpable tumors had 5- and IO-year survival rates of 95% and 80%, respectively, whereas patients with palpable tumors had 5- and IO-year survival of only 8%. The actuarial survival rates for patients without palpable tumors were further analyzed with respect to radiotherapy management (Fig. 5). Ten of 28 patients were treated with a combination of external and intracavitary irradiation, whereas the remaining 18 patients received brachytherapy alone. The survival rates were no different between patients that received combination therapy and those who received brachytherapy alone. Figure 6 shows that the length of time between the primary treatment and appearance of recurrences has an impact on the survival probability. The intervals between primary surgery and recurrence were divided into two groups at 4 points, at 2 years, 3 years, 6 years, and 10 years after the primary treatment. The IO-year actuarial survival rate of patients that developed recurrences within 2 years from the primary treatment was 51% compared with a survival rate of 38% for patients whose tumor recurred after 2 years from the primary treatment. When the survival rates were
= .,* 2 5
40
IT0
929
et a/.
L
80 -
T
.\
(n : 28)
:!I
(n : 10)
I
60-
o)
% a d
patlents
brachyther. alone ext. ired I
0 0
2
I
4
& bmchyttte~. I
I
6
Years after
I
I
8
10
Treatment
Fig. 5. Actuarial survival rates of patients with no palpable mass at the vaginal stump. Patients were treated with brachytherapy alone (-a -. -), or brachytherapy combined with external irradiation (------).
compared at 3, 6, and 10 years, the results were similar to those at the 2-year interval point. The length of time between primary treatment and recurrence did not influence the survival rate. 3. Late complications of radiotherapy Forty-four of 48 patients who were alive more than 1 year after radiotherapy for recurrence were included in the study of late complications. In this series, late complications of radiotherapy were mostly intestinal and the patients complained of anal bleeding and/or abdominal pain. Late intestinal complications were determined only by subjective symptoms. Table 1 shows the incidence of late intestinal complications. Eighteen patients were given
-
200 0
2
4
aipatiits - patiit withoutmass patientwith mass (n : 20) 1 6
8
10
40 In : 18, 20
I (C)
o-
O
2
C” 12) I
0)
I4
6
6
10
0
2
4
6
6
10
Years after Treatment
Years after Fig. 4. Actuarial
Treatment
survival rates of patients with carcinoma of the vaginal stump. All patients (), patients with no palpable mass at the vaginal stump (- . -a -), and patients with a palpable mass (------).
Fig. 6. Survival curves were compared between patients who had a recurrence within a certain period and those with a later recurrence. Four time periods were arbitrarily selected for the comparisons: (A) 2 years, (B) 3 years, (C) 6 years, and (D) 10 years.
1. J. Radiation Oncology ??Biology 0 Physics
930 Table 1. Incidence
of late intestinal Number
complications of patients
Total
With complication
Need treatment
B.T.* alone B.T. and E.I.+
18 26
3 (17%) 18 (69%)
1 (6%) 8 (31%)
Total
44
21 (47%)
9 (20%)
Treatment
* Brachytherapy. ’ Brachytherapy
and external
irradiation.
brachytherapy alone and 26 patients received a combination of external irradiation and brachytherapy. Three of 18 patients given brachytherapy alone had intestinal symptoms but only one patient required medication, On the other hand, 18 of 26 patients given external irradiation and brachytherapy complained of intestinal symptoms and 8 patients received medication. No patients needed surgical treatment for late complications in this study.
DISCUSSION In this study, all patients who developed cancer of the vaginal stump after hysterectomy for cervical cancer were diagnosed as having recurrences of the primary cancer. This was based on two factors, a) pathological examination could not determine a difference between the original (primary) cancer and second tumor (recurrence), and b) the second cancer arose at the same site as the primary one. The length of time between initial treatment and recurrence varied widely, from 3 months to 36 years. Our study raises an important question whether the carcinoma of the vaginal stump is a recurrence or a second primary cancer. Some surgical studies have reported no recurrent carcinoma of the cervix later than 5 years after the initial surgery (2, 16). The development of a second primary tumor has been described in a number of malignancies, most notably Hodgkin’s disease, in patients following successful treatment of their primary tumor (21). The concept of multifocal origins of neoplasms of the lower genital tract has been credited to Marcus (15). It has been suggested that many cancers developing in the upper vagina and cervix 5 or more years after radiation therapy for cervical cancer should be considered as new primary tumors of the vagina (17, 19). Upper vaginal carcinoma in sitzf, as well as invasive squamous cell carcinoma, may occur many years following radiation therapy (14, 19). Furthermore, it cannot be excluded that upper vaginal cancer might be a radiation related cancer (3, 8, 17). However, Koch et al. investigated the “time to second primary” in gynecologic cancer patients and suggested that there were no statistically significant differences in the 25 years following diagnosis of the first primary (13).
May I, 199I, Volume 20, Number 5
Another report has also suggested that the second primary tumor is not associated with radiation therapy (20). Note that all patients in our study were treated with surgery alone and no radiotherapy was applied at the first treatment. Overall our study supports the theory that patients with cervical cancer have a tendency to develop another cancer in the upper vagina. The estimated risk of developing a second tumor in patients with head and neck cancer was 10% within 3 years of radiotherapy, 15% within 5 years, and 23% within 8 years (4). However, the frequency in the lower genital tract was very low compared with head and neck tumors (1.2-4.4%) (20). The development of a second epithelial tumor in aerodigestive tract regions of the head and neck, lung, and esophagus may be due to diffuse mucosal initiation and promotion by exogenous carcinogenic factors such as tobacco and alcohol (4). We cannot make assumptions as to which exogenous factors may be associated with cancer of the genital tract. It is possible that the frequency of developing a second primary cancer in the upper vagina is related to intrinsic factors in patients with cancer. Examination of the cumulative incidence of recurrence in this series shows a biphasic curve, that is, an initial steep slope within 5 years of the first treatment and a shallow slope after 5 years. We expect that the initial slope reflects the recurrence of cervical cancer and few second primaries, and that the second slope shows mostly second primary cancers. Murad and associates have suggested that many vagina1 neoplasms occuring 5 or more years after irradiation for invasive squamous carcinoma of the cervix had pathologic features of primary vaginal neoplasms ( 17). In this study, we could find no definitive histologic difference between primary and recurrent tumors and could make no distinction between recurrence or development of a second primary at the vaginal stump in patients who had cervical cancer treated with surgery. However, we still expect that some or many patients in this series developed a second primary vaginal cancer. The treatment of recurrent cancer is difficult and the prognosis of the patients is poor. Evans reported that all patients in an untreated group with unresectable recurrent carcinoma of the cervix died within 15 months. On the other hand, 10% of patients lived for 15 months or longer and the 5-year survival rate was 6.2% in the treated group (7). Other authors describe 3.1% to 6% 5-year survival rates for patients with recurrent cervical cancers (12, 14, 15). Creasman suggested that 2-year survival seemed to be a reasonable compromise for patients with recurrent disease (5). In the present series, the overall actuarial 5year survival rate was 5 l%, which was extremely good compared to other treatment results. When treatment for pelvic recurrence is considered, the location is very important (5). Cervical cancer recurrence can be divided into three groups according to site: peripheral pelvic wall, central vaginal stump, and vaginal stump with extension
Brachytherapy for recurrent cervical cancer 0 H.
to the pelvic cavity. Deutsch et al. suggested that the best results were obtained in cases where recurrent cervical cancer after surgery appeared to be confined to the vagina (6). Another factor in determining the prognosis of patients with recurrent cancer is the tumor size. In this series, all patients had recurrence at a favorable site, that is, vaginal stump, and more than half of the patients had small tumors. We think these are reasons why our treatment results were excellent. This is supported by the fact that the 5-year survival rates of the groups with and without palpable tumor masses were completely different, that is, 8% for the former group but 80% for the latter. There are several possible treatment modalities for recurrent cervical cancer. The most common are radiotherapy and surgery. When recurrence develops after radiotherapy and is resectable, surgery might be preferable. If radiotherapy is to be applied to patients with recurrent cervical cancer, a combination of external irradiation and interstitial irradiation is recommended (18, 22). Deutsch et al. suggested that whole pelvic external irradiation in addition to intracavitary irradiation was necessary in cases where tumor that appeared after surgery were confined to the vagina (6). We treated patients without a palpable tumor by intracavitary irradiation alone. In this group, there was no difference in survival rates between patients with brachytherapy alone and those who received a combination of external irradiation and brachytherapy. On the other hand, a combination of external irradiation and brachytherapy increased the incidence of late radiation damage. We think that brachytherapy alone may be adequate treatment for patients without palpable tumor
IT0
et al.
931
masses of the vaginal stump. The prognosis of patients with palpable tumor masses was very poor. We delivered only 30 Gy of whole pelvic irradiation followed by brachytherapy to these patients, as we feared serious intestinal complications. Fletcher et al. recommend that external irradiation is important in treating patients with recurrences after simple hysterectomy, and 60 Gy should be given to the entire pelvis with a field reduction after 50 Gy (9). It is possible that higher doses of external irradiation may result in a better prognosis for these patients and the difference in survival rates between patients with and without palpable tumor masses will be decreased. Some authors suggest that the length of time between primary treatment and recurrence influences the prognosis of the patient. Bedwinek et al. reported that breast cancer patients with a chest wall recurrence later than 24 months after surgery had a 56% 5-year survival rate as compared with only 17% survival for those with a chest wall recurrence less than 2 years after surgery (1). We did not find any relationship between the intervals from the first treatment to recurrence and prognosis, but only a small number of patients was evaluated. In this study, the only significant prognostic factor was tumor size. Therefore, we conclude that early detection of the tumor is extremely important. Hernandez-Linares et al. also emphasize the importance of follow-ups with frequent vaginal Papanicolaou smears for patients after treatment for preinvasive or invasive malignancies (11). Cytologic surveillance should be continued over an extended period of time, as vaginal CIS may appear 20 years or more after the original treatment ( 11).
REFERENCES 1. Bedwinek, J. M.; Lee, J.; Fineberg, B.; Ocwieza, M. Prognostic indicators in patients with isolated local-regional recurrences of breast cancer. Cancer 47:2232-2235; 198 I. 2. Brunschwig, A. Surgical treatment of carcinoma of the cervix, recurrent after irradiation or combination of irradiation and surgery. Am. J. Roentgenol. 99:365-370; 1967. 3. Castro, E. B.; Rosen, P. P.; Quan, S. H. Q. Carcinoma of large intestine for carcinoma of the cervix and uterus. Cancer 3 1:45-52; 1973. 4. Cooper, J. S.; Pajak, T. F.; Rubin, P.; Tupchong, L.; Brady, L. W.; Liebel, S. A.; Laramore, G. E.; Martial, V. A.; Davis, L. W.; Cox, J. D. Second malignancies in patients who have head and neck cancer. Incidence, effect on survival and implication based on the RTOG experience. Int. J. Radiat. Oncol. Biol. Phys. 17:449-456; 1989. 5. Creasman, W. T.; Rutledge, F. Preoperative evaluation of patients with recurrent carcinoma of the cervix. Gynecol. Oncol. 1:ll l-1 18; 1972. 6. Deutsch, M.; Parsons, J. A. Radiotherapy for carcinoma of the cervix recurrent after surgery. Cancer 34:2051-2055; 1974. 7. Evans, S. D. Jr.; Hilaris, B. S.; Barber, H. R. K. External vs Interstitial irradiation in unresectable recurrent cancer of the cervix. Cancer 28:1284-1288; 1971. 8. Fehr, E. P.; Prem, K. A. Malignancy of the uterine corpus
9.
10.
11.
12.
13.
14.
15.
following irradiation therapy for squamous cell carcinoma of the cervix. Am. J. Obstet. Gynecol. 119:685-692; 1974. Fletcher, G. H.; Hamberger, A. D. Female pelvis: Squamous cell carcinoma of the uterine cervix. In: Fletcher, G. H., ed. Textbook of radiotherapy, 3rd edition. Philadelphia, PA: Lea & Febiger; 1980:732-772. Hashimoto, S.; Ito, H.; Dokiya, T. High dose rate brachytherapy of uterine cervix cancer: Comparison of two fraction regimens. In: Pailwal, B. R., ed. Optimization of cancer radiotherapy. New York: American Institute of Physics; 1985: 93-102. Hernandez-Linares, W.; Puthawala, A.; Nolan, J. F.; JernStrom, P. H.; Morrow, C. P. Carcinoma in situ of the vagina: past and present management. Obstet. Gynecol. 56:356360; 1980. Keettel, W. G.; Voorhis, L. W. V.; Latourette, H. B. Management of recurrent carcinoma of the cervix. Am. J. Obstet. Gynecol. 102:67 l-677; 1968. Koch, M.; Hauson, J.; Hill, G. B.; Starrevelt, A. A.; Gaedke, H. Analysis of “time to second primary” in patients with gynecological cancers. J. Chron. Dis. 38:443-447; 1985. Koss, L. G.; Melaned, M. R.; Daniel, W. W. In situ epidermoid carcinoma of the cervix and vagina following radiotherapy for cervical cancer. Cancer 14:353-360; 196 1. Marcus, S. L. Multiple squamous cell carcinoma involving
932
16.
17.
18.
19.
1. J. Radiation Oncology 0 Biology 0 Physics
the cervix, vagina and vulva. Am. J. Obstet. Gynecol. 80: 802-212; 1960. Munnell, E. W.; Bonney. W. A. Jr. Critical points of failure in the therapy of cancer of the cervix: a study of 250 recurrences. Am. J. Obstet. Gynecol. 8 1:52 l-534; 196 I. Murad, T. M.: Durant, J. R.; Maddox, W. A.; Galakatos, A. The pathologic behavior of primary vaginal carcinoma and its relationship to cervical cancer. Cancer 35:787-794; 1975. Nori, D.; Hilaris. B. S.; Kim, H. S.; Clark, D. G.: Kim, W. S.; Jones, W. B.; Lewis, J. L. Jr. Interstitial irradiation in recurrent gynecological cancer. Int. J. Radiat. Oncol. Biol. Phys. 7:1513-1517; 1981. Perez, C. A.; Arneson, A. V.; Dehner. L. P.; Dowling, E. A. Radiation therapy in carcinoma of the vagina. Obstet. Gynecol. 44:862-872; 1974.
May I. 199 I. Volume 20, Number
5
20. Pride, G. L.; Buchler, D. A. Carcinoma of vagina or cervix ten or more years following pelvic irradiation. Am. J. Obstet. Gynecol. 127:513-517; 1977. 21. Rubin. P.; Zagars, G.; Chuang, C.; Thomas, E. M. Hodgkin’s disease: Is there a price for successful treatment? A 25 year experience. Int. J. Radiat. Oncol. Biol. Phys. 12: I53- 166; 1985. 22. Puthawala, A.; Syed, A. M. N.; Nalick, D. R.; McNamara, C.: DiSaia, P. J. Integrated external and interstitial radiation therapy for primary carcinoma of the vagina. Obstet. Gynecol. 62:367-372; 1983. 23. Volterrani, F.; Feltre, L.; Sigurta, D.; Giuseppe, M. D.; Luciani, L. Radiotherapy versus surgery in the treatment of cervix stage Ib cancer. Int. J. Radiat. Oncol. Biol. Phys. 9: 1781-1784; 1983.
Copyright
0360.3016/91 $3.00 + .OO 8 1991 Pergamon Press plc
??Original Contribution
HIGH DOSE RATE INTRACAVITARY CERVICAL CANCER OF THE VAGINAL
BRACHYTHERAPY FOR RECURRENT STUMP FOLLOWING HYSTERECTOMY
HISAO ITO, M.D., HIDEO KUMAGAYA, M.D., NAOYUKI SHIGEMATSU, IKU NISHIGUCHI, M.D., TOSHITAKE NAKAYAMA, M.D. AND SHOZO HASHIMOTO, M.D. Department
of Radiology, Keio University, School of Medicine, 35 Shinanomachi,
Shinjukuku,
M.D.,
Tokyo 160 Japan
Forty-eight patients with recurrent cervical cancer of the vaginal stump following hysterectomy for cervical cancer, were treated with high dose rate intracavitary brachytherapy with or without external irradiation. The intervals between primary surgery and vaginal recurrences varied widely from 3 months to 36 years. Patients were classified into two groups, either with or without palpable tumor of the vaginal stump. Tumor size was determined by bimanual rectovaginal examination at the time of recurrence. Survival rates were 8% for the group with palpable tumors and 80% for those without, respectively. The survival rate of patients who did not have palpable masses and were treated with brachytherapy alone was not improved by combination with external irradiation. The time interval from the primary hysterectomy to the recurrence did not influence survival. These results suggest that the only significant prognostic factor for recurrent cervical cancer after hysterectomy is tumor size. The relationship between recurrent cervical cancer of the vagina1 stump and second primary vagina1 cancer is also discussed. High dose rate brachytherapy, Recurrence at the vaginal stump, Recurrent cervical cancer, Vaginal cancer, Intracavitary brachytherapy, Vaginal stump cancer.
etc. For example, central recurrences are salvaged easier than peripheral pelvic wall recurrences (5) and patients with postsurgical recurrences have more treatment options and a better prognosis than patients with recurrences developed after radiotherapy (6). In our institution, we treated with radiotherapy 48 patients who developed recurrent cervical cancer following simple or radical hysterectomy. Here, we report our analysis of the clinical prognostic factors in patients with recurrent cervical cancer of the vaginal stump.
INTRODUCTION
In Japan, periodic check-ups for uterine cancer have become a common practice for women over the age 30 years. As a result, the incidence of patients with early stages of cervical cancer has increased. The treatment modality for early cervical cancer has been a source of controversy among surgeons and radiotherapists (23). In Japan, pa-
tients present first at gynecologic clinics, and it is gynecologists in most institutions who determine the treatment modality without additional inputs of radiotherapists (10). Most gynecologists consider that surgical treatment is superior to radiotherapy, and, as a result, the majority of patients of stages 0-IIb become subjected to radical hys-
METHODS
terectomy and lymphadectomy. However, more recently there has been a trend to replace radical surgery with less radical procedures such as modified radical hysterectomy and simple hysterectomy to reduce the operative complications. A less radical surgery may be complicated with higher incidence of tumor recurrences at the vaginal stump. In general, recurrent tumors are difficult to manage and overall prognosis of the patients is poor ( 14, 15). The prognosis, however, depends on a number of factors, such as the site of recurrence, previous treatment modality,
AND
MATERIALS
Between January 1975 and August 1987, 48 patients were referred to the Department of Radiology at Keio University Hospital with cancer of the vaginal stump following hysterectomy for cervical cancer. All patients included in this study had squamous cell carcinoma both at the time of the original diagnosis and at the recurrence. The breakdown by stage at the time of the original diagnosis was as follows: Stage 0, 3 patients (6%); Stage I, 16 patients (33%); Stage II, 18 patients (37%); Stage III, 3 patients (6%); unknown, 9 patients (18%). Figure 1 illus-
Reprint requested to: Dr. Hisao Ito.
Accepted
927
for publication
9 November
1990.
928
I. J. Radiation Oncology 0 Biology 0 Physics
12
May 1, 1991, Volume 20, Number 5
withoutmas ECi patient 0
patientwith mass
:_ !
10 fn E at 3
_._
8
dose e&luati,tq point
:
(1 cm) b
6
z +
4
z’
,. 2 0 mder 30 31- 40 41- 50 51- 60 61_ 70 716 over
Age by Decades Fig. 1. Age by decades is shown for 48 patients cell carcinoma of the vaginal stump.
with squamous
the age distribution of the patients. Most of the patients in this series were in their 5th-6th decades. Patients were classified into two groups, those with palpable tumor masses and those without, according to the tumor size at the vaginal stump. Tumor size was determined by bimanual rectovaginal examination when recurrences were found. Tumor evaluation of the vaginal stump by CT scan was performed in about half of the patients; however, the findings were not consistent with the bimanual examination. In this study, information obtained from CT scan was not included in the tumor evaluation. When bimanual rectovaginal examination of the vaginal stump gave a questionable result, the patient was classified as not having a palpable mass. There was not difference in the age distributions between the patients with tumor masses at the vaginal stump and those without (Fig. 1). All recurring patients were treated with brachytherapy or a combination of brachytherapy and external irradiation. External irradiation, using a 6 MV photon beam, was given to all patients with palpable masses and to 10 of 28 patients without palpable masses. Whole pelvis irradiation (50 Gy) was delivered using conventional anterior-posterior parallel opposed fields. Following 30 Gy, a central shield (4 cm width) was inserted to cover the small pelvic cavity, to avoid overdosage of the bladder and the rectum (10). All patients in this series received high dose rate intracavitary brachytherapy using a Cesium137 source (7.4 X 10” Bq) for each colpostat. The irradiation dose to the vaginal stump was evaluated at a point that was 1 cm from the surface of the tumor (Fig. 2). The surface of the tumor was indicated on a roentgen film by metal chips inserted at the tumor margins. Patients received 5 Gy intracavitary irradiation twice a week to a total dose of 30 Gy in 6 fractions. As an alternative fractionation schedule, some patients received 6 Gy irradiation once a week to a total dose of 24 Gy.
trates
Fig. 2. The schema of hig dose rate brachytherapy. The metal chins were inserted into both sides of the tumor surface at the vaginal stump. The irradiated dose was evaluated at a depth of I cm from the surface of the vagina.
Survival analysis and survival using the Kaplan-Meier method.
curves were performed
RESULTS 1. Interval between primary surgery and tumor recurrence Figure 3 illustrates the cumulative frequency of cervical cancer recurrences from the time of radical or simple hysterectomy. The median time to recurrence was 3.3 years. Sixty-two percent of the recurrences occured within 5 years of the primary treatment. Fifteen of 19 patients originally diagnosed with Stage O-I disease failed within 5 years, and 14 of 2 1 patients with Stage II-III developed recurrences within the same period. Clinical stage at the original diagnosis did not influence the interval between
8 5 z z
40
g ‘E Q 1
20
5
0
+ 0 -A-
0
1 0
I
5
10
allpatients
” : 40 without mass n : 26 ll:20 with mass
1
15
20
L
25
30
35
Years after Radiotherapy Fig. 3. Cumulative incidence of recurrence time length after primary surgery.
in relation
to the
Brachytherapy for recurrent cervical cancer 0 H.
primary surgery and the recurrence. When the time intervals from primary surgery to recurrence were compared between the groups with and without palpable tumor masses, there was no difference. The cumulative incidence of the recurrences has a steep slope within first 5 years after surgery but thereafter the slope becomes shallower. 2. Survival rates Figure 4 shows actuarial survival of the patients, calculated by the Kaplan-Meier method. The 5- and IO-year survival rates for all patients were 56% and 47%, respectively. However, the survival was greatly influenced by the presence of palpable tumors at the vaginal stump. Patients without palpable tumors had 5- and IO-year survival rates of 95% and 80%, respectively, whereas patients with palpable tumors had 5- and IO-year survival of only 8%. The actuarial survival rates for patients without palpable tumors were further analyzed with respect to radiotherapy management (Fig. 5). Ten of 28 patients were treated with a combination of external and intracavitary irradiation, whereas the remaining 18 patients received brachytherapy alone. The survival rates were no different between patients that received combination therapy and those who received brachytherapy alone. Figure 6 shows that the length of time between the primary treatment and appearance of recurrences has an impact on the survival probability. The intervals between primary surgery and recurrence were divided into two groups at 4 points, at 2 years, 3 years, 6 years, and 10 years after the primary treatment. The IO-year actuarial survival rate of patients that developed recurrences within 2 years from the primary treatment was 51% compared with a survival rate of 38% for patients whose tumor recurred after 2 years from the primary treatment. When the survival rates were
= .,* 2 5
40
IT0
929
et a/.
L
80 -
T
.\
(n : 28)
:!I
(n : 10)
I
60-
o)
% a d
patlents
brachyther. alone ext. ired I
0 0
2
I
4
& bmchyttte~. I
I
6
Years after
I
I
8
10
Treatment
Fig. 5. Actuarial survival rates of patients with no palpable mass at the vaginal stump. Patients were treated with brachytherapy alone (-a -. -), or brachytherapy combined with external irradiation (------).
compared at 3, 6, and 10 years, the results were similar to those at the 2-year interval point. The length of time between primary treatment and recurrence did not influence the survival rate. 3. Late complications of radiotherapy Forty-four of 48 patients who were alive more than 1 year after radiotherapy for recurrence were included in the study of late complications. In this series, late complications of radiotherapy were mostly intestinal and the patients complained of anal bleeding and/or abdominal pain. Late intestinal complications were determined only by subjective symptoms. Table 1 shows the incidence of late intestinal complications. Eighteen patients were given
-
200 0
2
4
aipatiits - patiit withoutmass patientwith mass (n : 20) 1 6
8
10
40 In : 18, 20
I (C)
o-
O
2
C” 12) I
0)
I4
6
6
10
0
2
4
6
6
10
Years after Treatment
Years after Fig. 4. Actuarial
Treatment
survival rates of patients with carcinoma of the vaginal stump. All patients (), patients with no palpable mass at the vaginal stump (- . -a -), and patients with a palpable mass (------).
Fig. 6. Survival curves were compared between patients who had a recurrence within a certain period and those with a later recurrence. Four time periods were arbitrarily selected for the comparisons: (A) 2 years, (B) 3 years, (C) 6 years, and (D) 10 years.
1. J. Radiation Oncology ??Biology 0 Physics
930 Table 1. Incidence
of late intestinal Number
complications of patients
Total
With complication
Need treatment
B.T.* alone B.T. and E.I.+
18 26
3 (17%) 18 (69%)
1 (6%) 8 (31%)
Total
44
21 (47%)
9 (20%)
Treatment
* Brachytherapy. ’ Brachytherapy
and external
irradiation.
brachytherapy alone and 26 patients received a combination of external irradiation and brachytherapy. Three of 18 patients given brachytherapy alone had intestinal symptoms but only one patient required medication, On the other hand, 18 of 26 patients given external irradiation and brachytherapy complained of intestinal symptoms and 8 patients received medication. No patients needed surgical treatment for late complications in this study.
DISCUSSION In this study, all patients who developed cancer of the vaginal stump after hysterectomy for cervical cancer were diagnosed as having recurrences of the primary cancer. This was based on two factors, a) pathological examination could not determine a difference between the original (primary) cancer and second tumor (recurrence), and b) the second cancer arose at the same site as the primary one. The length of time between initial treatment and recurrence varied widely, from 3 months to 36 years. Our study raises an important question whether the carcinoma of the vaginal stump is a recurrence or a second primary cancer. Some surgical studies have reported no recurrent carcinoma of the cervix later than 5 years after the initial surgery (2, 16). The development of a second primary tumor has been described in a number of malignancies, most notably Hodgkin’s disease, in patients following successful treatment of their primary tumor (21). The concept of multifocal origins of neoplasms of the lower genital tract has been credited to Marcus (15). It has been suggested that many cancers developing in the upper vagina and cervix 5 or more years after radiation therapy for cervical cancer should be considered as new primary tumors of the vagina (17, 19). Upper vaginal carcinoma in sitzf, as well as invasive squamous cell carcinoma, may occur many years following radiation therapy (14, 19). Furthermore, it cannot be excluded that upper vaginal cancer might be a radiation related cancer (3, 8, 17). However, Koch et al. investigated the “time to second primary” in gynecologic cancer patients and suggested that there were no statistically significant differences in the 25 years following diagnosis of the first primary (13).
May I, 199I, Volume 20, Number 5
Another report has also suggested that the second primary tumor is not associated with radiation therapy (20). Note that all patients in our study were treated with surgery alone and no radiotherapy was applied at the first treatment. Overall our study supports the theory that patients with cervical cancer have a tendency to develop another cancer in the upper vagina. The estimated risk of developing a second tumor in patients with head and neck cancer was 10% within 3 years of radiotherapy, 15% within 5 years, and 23% within 8 years (4). However, the frequency in the lower genital tract was very low compared with head and neck tumors (1.2-4.4%) (20). The development of a second epithelial tumor in aerodigestive tract regions of the head and neck, lung, and esophagus may be due to diffuse mucosal initiation and promotion by exogenous carcinogenic factors such as tobacco and alcohol (4). We cannot make assumptions as to which exogenous factors may be associated with cancer of the genital tract. It is possible that the frequency of developing a second primary cancer in the upper vagina is related to intrinsic factors in patients with cancer. Examination of the cumulative incidence of recurrence in this series shows a biphasic curve, that is, an initial steep slope within 5 years of the first treatment and a shallow slope after 5 years. We expect that the initial slope reflects the recurrence of cervical cancer and few second primaries, and that the second slope shows mostly second primary cancers. Murad and associates have suggested that many vagina1 neoplasms occuring 5 or more years after irradiation for invasive squamous carcinoma of the cervix had pathologic features of primary vaginal neoplasms ( 17). In this study, we could find no definitive histologic difference between primary and recurrent tumors and could make no distinction between recurrence or development of a second primary at the vaginal stump in patients who had cervical cancer treated with surgery. However, we still expect that some or many patients in this series developed a second primary vaginal cancer. The treatment of recurrent cancer is difficult and the prognosis of the patients is poor. Evans reported that all patients in an untreated group with unresectable recurrent carcinoma of the cervix died within 15 months. On the other hand, 10% of patients lived for 15 months or longer and the 5-year survival rate was 6.2% in the treated group (7). Other authors describe 3.1% to 6% 5-year survival rates for patients with recurrent cervical cancers (12, 14, 15). Creasman suggested that 2-year survival seemed to be a reasonable compromise for patients with recurrent disease (5). In the present series, the overall actuarial 5year survival rate was 5 l%, which was extremely good compared to other treatment results. When treatment for pelvic recurrence is considered, the location is very important (5). Cervical cancer recurrence can be divided into three groups according to site: peripheral pelvic wall, central vaginal stump, and vaginal stump with extension
Brachytherapy for recurrent cervical cancer 0 H.
to the pelvic cavity. Deutsch et al. suggested that the best results were obtained in cases where recurrent cervical cancer after surgery appeared to be confined to the vagina (6). Another factor in determining the prognosis of patients with recurrent cancer is the tumor size. In this series, all patients had recurrence at a favorable site, that is, vaginal stump, and more than half of the patients had small tumors. We think these are reasons why our treatment results were excellent. This is supported by the fact that the 5-year survival rates of the groups with and without palpable tumor masses were completely different, that is, 8% for the former group but 80% for the latter. There are several possible treatment modalities for recurrent cervical cancer. The most common are radiotherapy and surgery. When recurrence develops after radiotherapy and is resectable, surgery might be preferable. If radiotherapy is to be applied to patients with recurrent cervical cancer, a combination of external irradiation and interstitial irradiation is recommended (18, 22). Deutsch et al. suggested that whole pelvic external irradiation in addition to intracavitary irradiation was necessary in cases where tumor that appeared after surgery were confined to the vagina (6). We treated patients without a palpable tumor by intracavitary irradiation alone. In this group, there was no difference in survival rates between patients with brachytherapy alone and those who received a combination of external irradiation and brachytherapy. On the other hand, a combination of external irradiation and brachytherapy increased the incidence of late radiation damage. We think that brachytherapy alone may be adequate treatment for patients without palpable tumor
IT0
et al.
931
masses of the vaginal stump. The prognosis of patients with palpable tumor masses was very poor. We delivered only 30 Gy of whole pelvic irradiation followed by brachytherapy to these patients, as we feared serious intestinal complications. Fletcher et al. recommend that external irradiation is important in treating patients with recurrences after simple hysterectomy, and 60 Gy should be given to the entire pelvis with a field reduction after 50 Gy (9). It is possible that higher doses of external irradiation may result in a better prognosis for these patients and the difference in survival rates between patients with and without palpable tumor masses will be decreased. Some authors suggest that the length of time between primary treatment and recurrence influences the prognosis of the patient. Bedwinek et al. reported that breast cancer patients with a chest wall recurrence later than 24 months after surgery had a 56% 5-year survival rate as compared with only 17% survival for those with a chest wall recurrence less than 2 years after surgery (1). We did not find any relationship between the intervals from the first treatment to recurrence and prognosis, but only a small number of patients was evaluated. In this study, the only significant prognostic factor was tumor size. Therefore, we conclude that early detection of the tumor is extremely important. Hernandez-Linares et al. also emphasize the importance of follow-ups with frequent vaginal Papanicolaou smears for patients after treatment for preinvasive or invasive malignancies (11). Cytologic surveillance should be continued over an extended period of time, as vaginal CIS may appear 20 years or more after the original treatment ( 11).
REFERENCES 1. Bedwinek, J. M.; Lee, J.; Fineberg, B.; Ocwieza, M. Prognostic indicators in patients with isolated local-regional recurrences of breast cancer. Cancer 47:2232-2235; 198 I. 2. Brunschwig, A. Surgical treatment of carcinoma of the cervix, recurrent after irradiation or combination of irradiation and surgery. Am. J. Roentgenol. 99:365-370; 1967. 3. Castro, E. B.; Rosen, P. P.; Quan, S. H. Q. Carcinoma of large intestine for carcinoma of the cervix and uterus. Cancer 3 1:45-52; 1973. 4. Cooper, J. S.; Pajak, T. F.; Rubin, P.; Tupchong, L.; Brady, L. W.; Liebel, S. A.; Laramore, G. E.; Martial, V. A.; Davis, L. W.; Cox, J. D. Second malignancies in patients who have head and neck cancer. Incidence, effect on survival and implication based on the RTOG experience. Int. J. Radiat. Oncol. Biol. Phys. 17:449-456; 1989. 5. Creasman, W. T.; Rutledge, F. Preoperative evaluation of patients with recurrent carcinoma of the cervix. Gynecol. Oncol. 1:ll l-1 18; 1972. 6. Deutsch, M.; Parsons, J. A. Radiotherapy for carcinoma of the cervix recurrent after surgery. Cancer 34:2051-2055; 1974. 7. Evans, S. D. Jr.; Hilaris, B. S.; Barber, H. R. K. External vs Interstitial irradiation in unresectable recurrent cancer of the cervix. Cancer 28:1284-1288; 1971. 8. Fehr, E. P.; Prem, K. A. Malignancy of the uterine corpus
9.
10.
11.
12.
13.
14.
15.
following irradiation therapy for squamous cell carcinoma of the cervix. Am. J. Obstet. Gynecol. 119:685-692; 1974. Fletcher, G. H.; Hamberger, A. D. Female pelvis: Squamous cell carcinoma of the uterine cervix. In: Fletcher, G. H., ed. Textbook of radiotherapy, 3rd edition. Philadelphia, PA: Lea & Febiger; 1980:732-772. Hashimoto, S.; Ito, H.; Dokiya, T. High dose rate brachytherapy of uterine cervix cancer: Comparison of two fraction regimens. In: Pailwal, B. R., ed. Optimization of cancer radiotherapy. New York: American Institute of Physics; 1985: 93-102. Hernandez-Linares, W.; Puthawala, A.; Nolan, J. F.; JernStrom, P. H.; Morrow, C. P. Carcinoma in situ of the vagina: past and present management. Obstet. Gynecol. 56:356360; 1980. Keettel, W. G.; Voorhis, L. W. V.; Latourette, H. B. Management of recurrent carcinoma of the cervix. Am. J. Obstet. Gynecol. 102:67 l-677; 1968. Koch, M.; Hauson, J.; Hill, G. B.; Starrevelt, A. A.; Gaedke, H. Analysis of “time to second primary” in patients with gynecological cancers. J. Chron. Dis. 38:443-447; 1985. Koss, L. G.; Melaned, M. R.; Daniel, W. W. In situ epidermoid carcinoma of the cervix and vagina following radiotherapy for cervical cancer. Cancer 14:353-360; 196 1. Marcus, S. L. Multiple squamous cell carcinoma involving
932
16.
17.
18.
19.
1. J. Radiation Oncology 0 Biology 0 Physics
the cervix, vagina and vulva. Am. J. Obstet. Gynecol. 80: 802-212; 1960. Munnell, E. W.; Bonney. W. A. Jr. Critical points of failure in the therapy of cancer of the cervix: a study of 250 recurrences. Am. J. Obstet. Gynecol. 8 1:52 l-534; 196 I. Murad, T. M.: Durant, J. R.; Maddox, W. A.; Galakatos, A. The pathologic behavior of primary vaginal carcinoma and its relationship to cervical cancer. Cancer 35:787-794; 1975. Nori, D.; Hilaris. B. S.; Kim, H. S.; Clark, D. G.: Kim, W. S.; Jones, W. B.; Lewis, J. L. Jr. Interstitial irradiation in recurrent gynecological cancer. Int. J. Radiat. Oncol. Biol. Phys. 7:1513-1517; 1981. Perez, C. A.; Arneson, A. V.; Dehner. L. P.; Dowling, E. A. Radiation therapy in carcinoma of the vagina. Obstet. Gynecol. 44:862-872; 1974.
May I. 199 I. Volume 20, Number
5
20. Pride, G. L.; Buchler, D. A. Carcinoma of vagina or cervix ten or more years following pelvic irradiation. Am. J. Obstet. Gynecol. 127:513-517; 1977. 21. Rubin. P.; Zagars, G.; Chuang, C.; Thomas, E. M. Hodgkin’s disease: Is there a price for successful treatment? A 25 year experience. Int. J. Radiat. Oncol. Biol. Phys. 12: I53- 166; 1985. 22. Puthawala, A.; Syed, A. M. N.; Nalick, D. R.; McNamara, C.: DiSaia, P. J. Integrated external and interstitial radiation therapy for primary carcinoma of the vagina. Obstet. Gynecol. 62:367-372; 1983. 23. Volterrani, F.; Feltre, L.; Sigurta, D.; Giuseppe, M. D.; Luciani, L. Radiotherapy versus surgery in the treatment of cervix stage Ib cancer. Int. J. Radiat. Oncol. Biol. Phys. 9: 1781-1784; 1983.
