- Email: [email protected]
Analysis of preoperative intracavitary cesium application versus postoperative external beam radiation in stage I endometrial carcinoma
fnr.J. Radimon Printed
Oncdogy Bid Phys Vol. 18. pp. 1011-1017 in the U.S.A. All rights reserved.
Copyright
0360-3016/90 $3.M) + .oO 0 1990 Pergamon Pres plc
0 Original Contribution ANALYSIS OF PREOPERATIVE INTRACAVITARY CESIUM APPLICATION VERSUS POSTOPERATIVE EXTERNAL BEAM RADIATION IN STAGE 1 ENDOMETRIAL CARCINOMA WILLIAM T. SAUSE,M.D., DONALD B. FULLER, M.D., W. GARY SMITH, M.D., GARY H. JOHNSON, M.D., HENRY P. PLENK, M.D. AND RONALD B. MENLOVE, PH.D. LDS Hospital,
Salt Lake City, Utah
Two groups of patients with surgical Stage 1 endometrial carcinoma treated at the LDS Hospita1 in Salt Lake City are analyzed. Group 1 comprises 112 patients treated from 1974 through 1976, during which time prewrative intracavitary cesium was routinely used in al1 patients. Group 2 comprises 117 patients treated 1981 through 1983
under the treatment policy of hysterectomy without preoperative cesium. High risk patients from each group (grade 3 and/or deep myometrial invasion) generally received similar postoperative external beam pelvic radiotherapy (4500-5000 cGy). While S-year actuarial disease-free survival rates were similar in each group (94% Group 1 VS 91% Group 2), multivariate analysis by the Cox Regression Method revealed that inclusion within treatment Group 2 carried independent adverse prognostic significante (p = 0.018). Other independent predictors of adverse 5-year disease-free survival included deep myometrial invasion and increasing histologie grade. Group 1 patients with grade 3 lesions had a superior 5-year actuarial disease-free survival(76% VS53%) compared to those from Group 2. Group 1 patients with deep myometrial invasion also had a superior 5-year disease-free survival(84% VS 69%). The remaining low risk patients (grade 1 or 2,
It has generally been acknowledged that combined therapy is associated with a significantly lower vagina1 and pelvic failure rate (1, 3, 4, 8, 9, 10, 12, 15, 16, 17). This benefit is principally limited to patients with adverse prognostic factors such as high grade histology or deep myometrial invasion (1, 3, 8). Even authors who favor the combined approach have expressed many varying opinions regarding optimal radiotherapeutic technique and timing (1, 2, 3, 4, 6, 8, 12, 15. 16, 17, 19). We report a series of 229 patients with FIGO Stage 1 endometrial carcinoma from the LDS Hospital in Salt Lake City treated during two different periods utilizing two different standard treatment polities. The first group of 112 patients was treated during the years 1974 through 1976 with a policy of routine preoperative intracavitary cesium application for al1 patients. The second group of 117 patients was treated during the years 198 1 through 1983 with the policy of no preoperative treatment.
INTRODUCTION Adenocarcinoma of the endometrium is now the most common gynecologic cancer and the fourth most common cancer overall among women. The majority of pa-
tients are postmenopausal and present with vagina1 bleeding early in their disease. Seventy percent or more have FIGO Stage 1 lesions at diagnosis (7). The optimal treatment method for Stage 1 endometrial carcinoma has never been precisely defined. Surgery is the mainstay of treatment and total abdominal hysterectomy with bilateral salpingo-oopherectomy (TAH-BSO) cures for most patients (6,8, 15, 16, 18,2 1). Many centers have utilized a combination to improve treatment results, domized trial documenting viva1 with this approach over reported.
of surgery and radiotherapy although a prospective ransignificantly improved sursurgery alone has never been
Acknowledgements-The authors wish to acknowledge the excellent secretarial work of Janet Lynn, and to thank Janet Gillette of the LDS Hospita1 Tumor Registry for providing the data base for this analysis. Accepted for publication 16 November 1989.
Presented at the October 1988 ASTRO Meeting in New Orleans, LA. Reprint requests to: William T. Sause, M.D., LDS Hospita1 Radiation Therapy Center, 8th Avenue & C Street, Salt Lake City, Utah 84143. Funded by a grant from the LDS Hospital Deseret Foundation. 1011
1012
1. J. Radiation Oncology 0 Biology0 Physics
METHODS Patients with surgical treated surgically during
AND MATERIALS
May 1990. Volume 18, Number 5 Table 1. Patient characteristics and distribution of prognostic factors according to treatment group
Stage 1 endometrial carcinoma the time petiods 1974 through
1976 and 1981 through 1983 were identified through the LDS Hospita1 Tumor Registry. These time periods were selected to provide a distinct separation between patients treated with two different standard treatment polities. The first group consisted of 118 patients treated during 1974 through 1976 of whom 6 were inevaluable because of sarcomatous histology. The second group consisted of 125 patients treated during 1981 through 1983 with 8 inevaluables due to sarcoma. The remaining 112 patients in group 1 and 117 patients in group 2 had surgically treated endometrial adenocarcinoma and formed the basis of this analysis. Our pathologists have used a standard grading criteria based on architecture throughout both periods of this analysis (wel1 differentiated = grade 1, moderately differentiated = grade 2 and poorly differentiated/anaplastic = grade 3). A minority of lesions (27 of 229, 12%) were ungraded. The treatment policy in effect for group 1 consisted of preoperative intrauterine and intravaginal cesium for virtually al1 patients. The intracavitary cesium was administered by a Fletcher-Suit after loading tandem and ovoids applicator and delivered a dose of 4500 to 5500 mg hr radium equivalent in one application. Extrafascial total abdominal hysterectomy and bilateral salpingo-oopherectomy were then performed within 24 to 48 hr following removal of the cesium application. Patients with grade 3 histology or deep (greater than i) myometrial invasion usually received supplemental postoperative pelvic externa1 beam radiation from an 8 MeV or 10 MeV linear accelerator to a parallel opposed or four field pelvic volume. The external beam dose was generally 4000 to 4500 cGy. Changes in the attending staff during the late 1970’s led to a new treatment policy for patients in group 2. These patients generally received no preoperative treatment. Those with grade 3 histology or deep myometrial invasion, usually received postoperative pelvic external beam radiotherapy to a dose of 4500 to 5000 cGy. Postoperative intravaginal cesium was rarely used and patients with low grade, superficially invasive lesions typically received no postoperative treatment. Some exceptions to the genera1 treatment polities occurred during both time periods but al1 patients are included in this analysis to avoid a potential source of selection bias. Table 1 summarizes prognostic factors and treatment given to patients in both treatment groups. Statistical methods Overall and disease-free survival curves were generated using the method of Kaplan and Meier. (11) Differences between survival curves were analyzed by the Mantel-
Group 1: 112 pts.
Mean age (yrs.) 64 Grade 1 2 3 Unk. Myometrial invasion <) >i Unk. depth Stage IA IB 1. Unk. Preopperative intracavity Cesium Proportion of high risk pts. receiving postoperative extemal pelvic irradiation
Group 2: 117 pts.
No.
(%)
No.
(%)
51 36 13 12
(46) (32) (12) (10)
83 21 9 4
(71) (18) (8) (3)
81 26 5
(72) (24) (4)
84 21 12
(72) (18) (10)
60 37 15 100
(54) (33) (13) (89)
68 44 5 5
(58) (38) (4) (4)
21130
(70)
19127
(70)
* High risk equals Grade 3 and/or deep myometrial invasion (>!).
Cox Method. (13) Multivariate analysis was performed by the Cox Proportional Hazards Method. (5) The median duration of follow-up was 130 months for group 1 and 55 months for group 2. Al1 patients followed longer than 60 months were censored at 60 months to allow for more accurate statistical comparison between the two groups. We fee1 this is valid because no tumor recurrences were seen later than 59 months after treatment and 8 1% were seen less than 36 months after treatment. To generate disease free survival curves, patients dying without evidente of cancer were treated as censored observations at the time of their death. RESULTS The 5-year overall and disease-free survival rates for al1 patients measured 84% and 92%, respectively (Fig. 1). The corresponding rates were 88% and 94%, respectively, for group 1 and 80% and 91% for group 2 (p = N.S., Fig. 2). Increasing histologie grade and deep (> 4) myometrial invasion were each associated with an adverse effect on disease-free and overall survival. Five-year disease-free survival rates measured 98% for grade 1, 9 1% for grade 2 and 67% for grade 3 lesions (p < 0.0000 1, Fig. 3). Diseasefree 5-year survival patients with superficial myometrial invasion measured 96%, compared to 69% for those with
Cesium vs. external beam radiation 0 W. T.
100.
E
E 5
s
a
SAUSE et al.
1013
_
60-
6040-
20-
-1
I
0
.
12
BO -omuw
Fig. 1. Overall and disease-free
deep invasion (p < 0.00001, Fig. 4). There was no difference in disease-free survival or overall survival in Stage IA when compared to Stage IB. Table 2 reveals a correlation between increasing histologic grade and probability of deep myometrial invasion. It is apparent that the risk of having deep myometrial invasion increased with increasing histologie grade (p < 0.000 1, chi square). To characterize risk factors more precisely, multivariate analysis was performed using the Cox Proportional Hazards Method. Variables entered included age, Stage (IA VSIB), histologie grade (1 VS2 VS3), depth of myometrial invasion (5: VS>$) and treatment group (1 VS2). Deep myometrial invasion (p < 0.0001) and increasing histologie grade (p = 0.007) each carried independent adverse prognostic significante. Inclusion within treatment group 1 carried independently favorable prognostic significante compared to group 2 for both 5-year disease-free (p = 0.0 18) and overall (p = 0.001) survival. Patient age and FIGO tumor Stage (IA VSIB) carried no independent prognostic significante. Further analysis of high risk patients revealed that those
ovonll
survival: al1 patients.
with grade 3 histology appeared to have superior diseasefree and overall survival during the period of routine preoperative intracavitary cesium application ( 1974 through 1976, group 1) compared to those treated under the more recent policy of no preoperative therapy (198 1 through 1983, group 2). Disease-free 5-year survival rates in these two groups measured 76% VS53%, respectively (Fig. 5). Because of relatively smal1 patient numbers in each subgroup, this finding did not achieve statistical significante. A similar trend towards improved disease-free and overall survival for patients with deep myometrial invasion was also noted in patient group 1. This group had an 84% 5year disease-free survival compared to 69% for corresponding patients treated under the group 2 policy (Fig. 6). Figure 7 reveals the benefit to routine preoperative cesium use (group 1) appeared to be restricted to “high risk” patients, that is, those with deep myometrial invasion and/ or grade 3 lesions. The remaining majority of “10~ risk” patients (grade 1 or 11,
§urvlvd
FNO Sunlval
Dbnn
-80 60
J
0 OAupl . uao2
;
w 40
: 20
0
12
24
36
m0(hmth8)
48
80 0
I
12
-
.
24
-
I 36
-W-W
Fig. 2. Overall and disease-free survival; Group 1 vs. Group 2.
-
I
48
-
+o
60
f I
1. J. Radiation Oncology 0 Biology 0 Physics
1014
Disease Free Survival
F
80 -
5 '5 3 cn
60 -
E 0
40 -
Pài
20 -
i
0 0
0 A
Gradel GradeII ?? Grade 111
1. 12
,* 24
ps 0.0001
I. 36
1. 46
I 60
Time (Months) Fig. 3. Disease-free
survival for al1 patients
according
to Grade.
Patterns offailure Local control was excellent in both patient groups. There were only five locoregional failures in the entire series of 229 patients (2%) and this did not vary between group 1 (2 of 112 patients, 1.7%) and group 2 (3 of 117 patients, 2.5%). A single, isolated vagina1 failure was observed in each group and in both patients this was noted in the distal vagina. There were no infield pelvic failures noted in the 40 high risk patients treated postoperatively with external beam radiotherapy, with or without a preoperative cesium application. Three of these 40 patients did have regional failure but two were in the distal vagina and one was in the labia, al1 outside the treated volume. Abdominal failure was the most common relapse pattem in our series, accounting for 12 of 2 1 total failures. This usually took the form of peritoneal carcinomatosis. The risk of abdominal failure for high risk patients was 12% (7 of 57 patients). Only 5 of the remaining 172 patients (2.9%) failed in the abdomen. DISCUSSION The 5-year actuarial disease-free and overall survival rates of 92% and 84% for our patients with surgically treated Stage 1 endometrial carcinoma are similar to most other reported series. (1, 2, 3, 4, 6, 8, 9, 10, 14, 15, 16, 18, 19) Our findings of adverse prognostic significante associated with increasing histologie grade (p < 0.0000 1) or deep myometrial invasion (p < 0.0001) have also been reported in multiple previous series. (1, 3, 7, 14, 15, 18, 19, 21) The unexpected finding in our analysis was that patients treated under a routine policy of an immediate properative tandem and ovoid intracavitary cesium application (group 1) fared better than the more recently treated group 2 patients (no preoperative therapy) according to the Cox Proportional Hazards Method. Superiority for 5-year disease-free survival (p = 0.018) and overall survival (p = 0.001) was noted in group 1 patients. This occurred even with high risk patients from both groups receiving
May 1990, Volume 18, Number 5
a similar volume, full dose postoperative extemal beam pelvic radiotherapy to 4500-5000 cGy. Inspection of Table 1 reveals that group 1 contained a smaller proportion of patients with grade 1 lesions than group 2 (46% VS 7 1%). Conversely, grade 2 or 3 lesions were more prevalent in group 1 compared to group 2 (44% VS 26%). Additionally, deep myometrial invasion was slightly more prevalent in group 1 (24% VS 18%). Slightly superior 5-year actuarial disease-free and overall survival rates were obtained in group 1 (Fig. 2) although this group contained a higher proportion of patients with unfavorable histologie features. Further analysis revealed that the improved prognosis in group 1 appeared to be restricted to high risk (grade 3 and/or > i myometrial invasion) patients (Fig. 5-8). Since the only major differente in treatment policy for group 1 was the routine administration of preoperative cesium and since similar proportions of high risk patients in each group received postoperative extemal beam pelvic radiotherapy, our data suggest that the preoperative cesium application may have been responsible for the decreased recurrence and death rate seen in group 1. Pelvic failure was extremely uncommon in either treatment group ( 1.7% group 1 VS2.5% group 2) and the predominant failure pattern was distant. Since the overall risk of recurrence was significantly lower in group 1, this implies that the preoperative cesium application may have exerted its favorable prognostic effect by devitalizing tumor cells and diminishing their ability to metastasize at the time of surgical manipulation. The advantage to performing the intracavitary cesium application immediately before surgery was that histopathologie findings in the resected uterus were not significantly altered and high risk patients could stil1 be selected for additional postoperative treatment. A review of literature on Stage 1 endometrial carcinoma reveals a wide diversity of opinion regarding optimal treatment for these patients. Approximately 80% of al1
Disease Free Survlval
Depltls1n cmpth>l)3
0 8
01
0
12
24 Tlme
36
ps o.oca
46
60
(Monthr)
Fig. 4. Disease-free survival for al1 patients of myometrial invasion.
according
to depth
Cesium vs. external beam radiation 0 W. T. SAUSEet al. Table 2. Correlation
between
1015
grade and depth of myometrial
invasion
Grade 1 Depth
3
Unk.
Total
No.
(W)
No.
((70)
No.
(%ó)
No.
@)
(85)
35
(61)
5
(23)
21
(78)
166
(72)
13 5
(11) (4)
193
(33) (6)
12 5
(22) (55)
3
(11)
47 16
(21) (7)
122
(100)
57
(100)
22
(100)
27
(100)
229
(100)
No.
@)
104 %k. q Total
2
Stage 1 endometrial carcinoma patients are cured by TAHBSO and some authors question the ability of combined modality therapy to improve these results. ( 19,2 1) Many other authors advocate the use of combined radiation and surgery in al1 patients or in selected patients with adverse prognostic features such as increased histologie grade or deep myometrial invasion. ( 1. 2, 3, 4, 6, 7. 8, 9, 10, 12. 14, 17, 18, 19) The principle benefit of combined modality therapy in most reported series has been a reduction in the vagina1 and pelvic failure rate. ( 1, 3, 4, 8, 9, 10, 15, 17) Brady et al. (4) reported that the rate of vagina1 persistente after hysterectomy alone for Stage 1 adenocarcinoma of the endometrium is 11% compared to 2% for combined preoperative radiation therapy and hysterectomy. Aalders et al. (1) showed in a large prospective randomized Norwegian trial that postoperative extemal pelvic radiotherapy significantly reduced the vagina1 and pelvic failure risk in Stage 1 endometrial adenocarcinoma when added to postoperative intravaginal radium application, compared to postoperative intravaginal radium alone. This benefit appeared to be most striking in patients with grade 3 histology and deep myometrial invasion in whom the addition of postoperative extemal beam pelvic radiotherapy reduced the risk of vagina1 and pelvic recurrence from 19.6% to 4.5%. Overall survival was not improved
in any patient subgroup, however, due to an increased incidence of distant metastases in the more aggressively treated group. Another prospective randomized trial in Stage 1 endometrial carcinoma reported by Graham (8) and updated by Piver et al. ( 15, 16) showed a slight but statistically insignificant survival improvement for either preoperative intrauterine or postoperative intravaginal radium application compared to hysterectomy alone at 5 and 10 years. Again, vagina1 failure rates were significantly reduced in the combined modality arms. Weigensburg (20) compared preoperative external beam radiation to preoperative intracavitary radiation in clinical Stage 1 carcinoma of the endometrium patients. He concluded that preoperative intracavity treatment was superior to extemal beam therapy. Several retrospective series have suggested a survival benefit to combined modality therapy. Gusberg, and Yannopoulos ( 10) in 1964 reported that the cure rate for early stage endometrial carcinoma was significantly improved in a nonrandomized but unselected series from Columbia Presbyterian Medical Center (p = 0.01) when preoperative radium was administered, compared to patients treated with surgery alone. This benefit appeared to be most pronounced in those with anaplastic tumors, enlarged uteri or cervical involvement. Patients with Stage
Disease Frw Survival; Grade 111
Dlserase
Free Survlval; Depth > 1/3
E
‘5 5
3
tn
c
0 8 P
0
20
??
Gmupl GroupP
1
- 0.
12’
2’4
3s
-
48
.
0
60
Time (Months) Fig. 5. Disease-free Group 2.
survival
for Grade
3 patients;
12
24 lime
Group
1 vs.
36
48
60
(Months)
Fig. 6. Disease-free survival for patients invasion: Group 1 vs. Group 2.
with deep myometrial
1016
1. J. Radiation Oncology 0 Biology 0 Physics
May 1990, Volume 18, Number 5
Dlsease Free Survival; High Risk
Disaa~. Fr.. Survlval; Low Risk IOol-~
r
a
r ._ .z L
60 -
8 E
40 -
$ P8
20 -
0
Gnnql Grorp2
??
o! 0
.
, 12
.
, 24 llme
Fig. 7. Disease-free depth > f); Group
.
, 36
.
, 46
.
0: 0
(
60
,
12
-
,
24
lime
(Months)
survival for high risk patients 1 vs. Group 2.
.
(Grade
3 or
1 wel1 differentiated tumors appeared to have an excellent prognosis with or without preoperative radium. The authors concluded that TAH-BSO was sufficient treatment for small, wel1 differentiated tumors but advocated combined preoperative radium and surgery for al1 tumors that were more advanced. Beiler et al. (3) reported in 1972 that there was a borderline statistically significant improved survival rate in early stage endometrial carcinoma patients treated by combined preoperative radium and surgery compared to surgery alone (91% VS79%, p = 0.07). This benefit was restricted to patients with medium to high grade lesions. Monson et al. ( 14) also noted a significant improvement in survival with combined modality therapy in women with Stage 1, high grade endometrial carcinomas compared with surgery alone. In that study, a total abdominal hysterectomy followed by radiation appeared to give the best results even though the authors felt these patients were adversely selected to receive radiotherapy. In women with Stage 1 poorly differentiated tumors, 5-year survival measured 92% in the postoperative radiation group compared to 59% in the surgery alone group. Our own series confirms that patients with Stage 1, grade 1 or 2 superficially invasive endometrial carcinomas have an excellent prognosis with or without preoperative intracavitary radiotherapy with a 97% 5-year actuarial disease-free survival. We believe the remaining “high risk” patients may have a better prognosis when treated with combined preoperative intravaginal and intrauterine cesium immediately before the hysterectomy. The mechanism responsible for benefit from preoperative intracavitary therapy is unknown. Perhaps preop-
Fig. 8. Disease-free depth < f); Group
.
,
36
.
,
46
.
1
60
(Months)
survival for low risk patients 1 vs. Group 2.
(Grade
1 or 2,
erative radiation reduces the possibility of tumor seeding at surgery or perhaps sterilizes microscopic metastasis in pelvic tissues. Although the policy of external beam pelvic radiotherapy directed to high risk patients was perfectly adequate in preventing pelvic failure with or without the addition of intracavitary cesium, the preoperative intracavitary insertion did reduce the overall risk of failure in high risk patients. Our high risk category included al1 patients with grade 3 lesions or deep myometrial invasion. Since it is impossible to preoperatively assess the amount of myometrial invasion, it is necessary to use histologie grade to predict which patients are likely to have deep invasion to decide who wil1 benefit from preoperative treatment. Our patients with Stage 1, grade 1 lesions had only an 11% risk of deep myometrial penetration and similar results have been reported by other authors. ( 1,9) These patients have such a low probability of falling into the high risk category that preoperative therapy would not seem to be routinely indicated. Our grade 2 and 3 patients had a 33% and 55% risk of deep myometrial invasion, respectively. These are the patients who would appear to be most likely to benefit from a routine preoperative intracavitary cesium or radium application to maximize the probability of long-term diseasefree survival. Like most reported studies, ours is a retrospective although nonselected series. We would like to see the value of a routine preoperative intracavitary application formally tested in a prospective randomized trial designed to include women with clinical Stage 1, moderately to poorly differentiated endometrial carcinomas.
REFERENCES 1. Aalders, J.; Abeler, V.; Kolstad, P.; Onsrud, M. Postoperative extemal irradiation and prognostic parameters in stage 1 endometrial carcinoma-clinical and histopathologie study of 540 patients. Obstet. Gynecol. 56:4 19-426; 1980. 2. Bedwinek, J.; Galakatos, A.; Gamel, M.; Ming-Shian, K.;
Stokes, S.; Perez, C. Stage 1, Grade 111adenocarcinoma of the endometrium treated with surgery and irradiation-sites of failure and correlation of failure rate with irradiation technique. Cancer 54~40-47; 1984. 3. Beiler, D. D.; Schmutz, D. A.; O’Rourke, T. L. Carcinoma
Cesium vs. external kam radiation 0 W. T. SAUSErf al. of the endometrium: radiation and surgery versus surgery alone. Radiology 102: 159- 164; 1972. 4. Brady, L. W.; Lewis, G. C.; Antoniades, J.; Prasasvinichai, S.: Torpie, R. J.; Asbell, S. 0.; Glassburn, J. R.; Schatanoff, D.: MacMurray, T. Evolution of radiotherapeutic techniques. Gynecol. Oncol. 2:314-323: 1974. 5. Cox, D. R. Regression models and life tables. J. R. Stat. Sec. 34: 187-220; 1972. 6. Eifel, P. J.; Ross, J.; Hendrickson, M.; Cox, R. S.: Kempson, R.; Martinez, A. Adenocarcinoma of the endometriumanalysis of 256 cases with disease limited to the uterine corpus: treatment comparisons. Cancer 52: 1026- 103 1; 1983. 1. Frick, H. C.: Munnell, E. W.; Richart, R. M.; Berger, A. P.: Lawry, M. F. Carcinoma ofthe endometrium. Am. J. Obstet. Gynecol. I 15663-672; 1973. 8. Graham, J. The value of preoperative or postoperative treatment by radium for carcinoma of the uterine body. Surg. Gynecol. Obstet. 132:855-860; May 1971. 9. Gusberg, S. B.; Jones, H. C.: Tovell. H. M. M. Selection of treatment for corpus cancer. Am. J. Obstet. Gynecol. 80: 374-380; 1960. D. Therapeutic decisions in 10. Gusberg. S. B.: Yannopoulos, corpus cancer. Am. J. Obstet. Gynecol. 88: 157-162; 1964. estimation from 11. Kaplan, E. L.; Meier P. Nonparametric incomplete observations. J. Am. Stat. Assoc. 53:457-48 1: 1958. 12. Landgren. R. D.; Fletcher, G. H.; Gallager, S.: Declos, L.; Wharton, J. T. Treatment failure sites according to irradiation technique and histology in patients with endometrial cancer. Cancer 40: 13 1-135: 1977.
1017
13. Mantel, N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother. Rep. 50:163-170: 1966. 14. Monson, R. R.; MacMahon, B.; Austin, J. H. Postoperative irradiation in carcinoma of the endometrium. Cancer 3 1: 630-632; 1973. 15. Piver, M. S. Stage 1 endometrial carcinoma: the role of adjunctive radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 6:367-368: 1980. L.; Tsukada, Y. A 16. Piver, M. S.; Yazigi, R.; Blumenson, prospective trial comparing hysterectomy, hysterectomy plus vagina1 radium, and uterine radium plus hysterectomy in stage 1 endometrial carcinoma. Obstet. Gynecol. 54:85-89: 1979. 17. Salazar, 0. M.; Feldstein, M. L.; DePapp, E. W.; Bonfiglio. T. A.; Keller. B. E.; Rubin. P.; Rudolph J. H. Endometrial carcinoma: analysis of failures with special emphasis on the use of initial preoperative external pelvic radiation. Int. J. Radiat. Oncol. Biol. Phys. 2: 1 10 1- 1 107; 1977. 18. Sall. S.: Sonnenblick, B.; Stone. M. L. Factors affecting surviva1 of patients with endometrial adenocarcinoma. Am. J. Obstet. Gynecol. 107:116-123: 1970. 19. Stokes. S.; Bedwinek. J.; Kao. M. S.; Camel. H. M.: Perez, C. A. Treatment of stage 1 adenocarcinoma of the endometrium by hysterectomy and adjuvant irradiation: a retrospective analysis of 304 patients. Int. J. Radiat. Oncol. Biol. Phys. 12:339-344; 1986. 20. Weigensberg, 1. J. Preoperative radiation therapy in stage 1 endometrial adenocarcinoma. Cancer 53:242-247; 1984. 21. Whetham, J. C. G.: Bean, J. L. M. Carcinoma of the endometrium. Am. J. Obstet. Gynecol. 1 I2:339-343; 1972.
Oncdogy Bid Phys Vol. 18. pp. 1011-1017 in the U.S.A. All rights reserved.
Copyright
0360-3016/90 $3.M) + .oO 0 1990 Pergamon Pres plc
0 Original Contribution ANALYSIS OF PREOPERATIVE INTRACAVITARY CESIUM APPLICATION VERSUS POSTOPERATIVE EXTERNAL BEAM RADIATION IN STAGE 1 ENDOMETRIAL CARCINOMA WILLIAM T. SAUSE,M.D., DONALD B. FULLER, M.D., W. GARY SMITH, M.D., GARY H. JOHNSON, M.D., HENRY P. PLENK, M.D. AND RONALD B. MENLOVE, PH.D. LDS Hospital,
Salt Lake City, Utah
Two groups of patients with surgical Stage 1 endometrial carcinoma treated at the LDS Hospita1 in Salt Lake City are analyzed. Group 1 comprises 112 patients treated from 1974 through 1976, during which time prewrative intracavitary cesium was routinely used in al1 patients. Group 2 comprises 117 patients treated 1981 through 1983
under the treatment policy of hysterectomy without preoperative cesium. High risk patients from each group (grade 3 and/or deep myometrial invasion) generally received similar postoperative external beam pelvic radiotherapy (4500-5000 cGy). While S-year actuarial disease-free survival rates were similar in each group (94% Group 1 VS 91% Group 2), multivariate analysis by the Cox Regression Method revealed that inclusion within treatment Group 2 carried independent adverse prognostic significante (p = 0.018). Other independent predictors of adverse 5-year disease-free survival included deep myometrial invasion and increasing histologie grade. Group 1 patients with grade 3 lesions had a superior 5-year actuarial disease-free survival(76% VS53%) compared to those from Group 2. Group 1 patients with deep myometrial invasion also had a superior 5-year disease-free survival(84% VS 69%). The remaining low risk patients (grade 1 or 2,
It has generally been acknowledged that combined therapy is associated with a significantly lower vagina1 and pelvic failure rate (1, 3, 4, 8, 9, 10, 12, 15, 16, 17). This benefit is principally limited to patients with adverse prognostic factors such as high grade histology or deep myometrial invasion (1, 3, 8). Even authors who favor the combined approach have expressed many varying opinions regarding optimal radiotherapeutic technique and timing (1, 2, 3, 4, 6, 8, 12, 15. 16, 17, 19). We report a series of 229 patients with FIGO Stage 1 endometrial carcinoma from the LDS Hospital in Salt Lake City treated during two different periods utilizing two different standard treatment polities. The first group of 112 patients was treated during the years 1974 through 1976 with a policy of routine preoperative intracavitary cesium application for al1 patients. The second group of 117 patients was treated during the years 198 1 through 1983 with the policy of no preoperative treatment.
INTRODUCTION Adenocarcinoma of the endometrium is now the most common gynecologic cancer and the fourth most common cancer overall among women. The majority of pa-
tients are postmenopausal and present with vagina1 bleeding early in their disease. Seventy percent or more have FIGO Stage 1 lesions at diagnosis (7). The optimal treatment method for Stage 1 endometrial carcinoma has never been precisely defined. Surgery is the mainstay of treatment and total abdominal hysterectomy with bilateral salpingo-oopherectomy (TAH-BSO) cures for most patients (6,8, 15, 16, 18,2 1). Many centers have utilized a combination to improve treatment results, domized trial documenting viva1 with this approach over reported.
of surgery and radiotherapy although a prospective ransignificantly improved sursurgery alone has never been
Acknowledgements-The authors wish to acknowledge the excellent secretarial work of Janet Lynn, and to thank Janet Gillette of the LDS Hospita1 Tumor Registry for providing the data base for this analysis. Accepted for publication 16 November 1989.
Presented at the October 1988 ASTRO Meeting in New Orleans, LA. Reprint requests to: William T. Sause, M.D., LDS Hospita1 Radiation Therapy Center, 8th Avenue & C Street, Salt Lake City, Utah 84143. Funded by a grant from the LDS Hospital Deseret Foundation. 1011
1012
1. J. Radiation Oncology 0 Biology0 Physics
METHODS Patients with surgical treated surgically during
AND MATERIALS
May 1990. Volume 18, Number 5 Table 1. Patient characteristics and distribution of prognostic factors according to treatment group
Stage 1 endometrial carcinoma the time petiods 1974 through
1976 and 1981 through 1983 were identified through the LDS Hospita1 Tumor Registry. These time periods were selected to provide a distinct separation between patients treated with two different standard treatment polities. The first group consisted of 118 patients treated during 1974 through 1976 of whom 6 were inevaluable because of sarcomatous histology. The second group consisted of 125 patients treated during 1981 through 1983 with 8 inevaluables due to sarcoma. The remaining 112 patients in group 1 and 117 patients in group 2 had surgically treated endometrial adenocarcinoma and formed the basis of this analysis. Our pathologists have used a standard grading criteria based on architecture throughout both periods of this analysis (wel1 differentiated = grade 1, moderately differentiated = grade 2 and poorly differentiated/anaplastic = grade 3). A minority of lesions (27 of 229, 12%) were ungraded. The treatment policy in effect for group 1 consisted of preoperative intrauterine and intravaginal cesium for virtually al1 patients. The intracavitary cesium was administered by a Fletcher-Suit after loading tandem and ovoids applicator and delivered a dose of 4500 to 5500 mg hr radium equivalent in one application. Extrafascial total abdominal hysterectomy and bilateral salpingo-oopherectomy were then performed within 24 to 48 hr following removal of the cesium application. Patients with grade 3 histology or deep (greater than i) myometrial invasion usually received supplemental postoperative pelvic externa1 beam radiation from an 8 MeV or 10 MeV linear accelerator to a parallel opposed or four field pelvic volume. The external beam dose was generally 4000 to 4500 cGy. Changes in the attending staff during the late 1970’s led to a new treatment policy for patients in group 2. These patients generally received no preoperative treatment. Those with grade 3 histology or deep myometrial invasion, usually received postoperative pelvic external beam radiotherapy to a dose of 4500 to 5000 cGy. Postoperative intravaginal cesium was rarely used and patients with low grade, superficially invasive lesions typically received no postoperative treatment. Some exceptions to the genera1 treatment polities occurred during both time periods but al1 patients are included in this analysis to avoid a potential source of selection bias. Table 1 summarizes prognostic factors and treatment given to patients in both treatment groups. Statistical methods Overall and disease-free survival curves were generated using the method of Kaplan and Meier. (11) Differences between survival curves were analyzed by the Mantel-
Group 1: 112 pts.
Mean age (yrs.) 64 Grade 1 2 3 Unk. Myometrial invasion <) >i Unk. depth Stage IA IB 1. Unk. Preopperative intracavity Cesium Proportion of high risk pts. receiving postoperative extemal pelvic irradiation
Group 2: 117 pts.
No.
(%)
No.
(%)
51 36 13 12
(46) (32) (12) (10)
83 21 9 4
(71) (18) (8) (3)
81 26 5
(72) (24) (4)
84 21 12
(72) (18) (10)
60 37 15 100
(54) (33) (13) (89)
68 44 5 5
(58) (38) (4) (4)
21130
(70)
19127
(70)
* High risk equals Grade 3 and/or deep myometrial invasion (>!).
Cox Method. (13) Multivariate analysis was performed by the Cox Proportional Hazards Method. (5) The median duration of follow-up was 130 months for group 1 and 55 months for group 2. Al1 patients followed longer than 60 months were censored at 60 months to allow for more accurate statistical comparison between the two groups. We fee1 this is valid because no tumor recurrences were seen later than 59 months after treatment and 8 1% were seen less than 36 months after treatment. To generate disease free survival curves, patients dying without evidente of cancer were treated as censored observations at the time of their death. RESULTS The 5-year overall and disease-free survival rates for al1 patients measured 84% and 92%, respectively (Fig. 1). The corresponding rates were 88% and 94%, respectively, for group 1 and 80% and 91% for group 2 (p = N.S., Fig. 2). Increasing histologie grade and deep (> 4) myometrial invasion were each associated with an adverse effect on disease-free and overall survival. Five-year disease-free survival rates measured 98% for grade 1, 9 1% for grade 2 and 67% for grade 3 lesions (p < 0.0000 1, Fig. 3). Diseasefree 5-year survival patients with superficial myometrial invasion measured 96%, compared to 69% for those with
Cesium vs. external beam radiation 0 W. T.
100.
E
E 5
s
a
SAUSE et al.
1013
_
60-
6040-
20-
-1
I
0
.
12
BO -omuw
Fig. 1. Overall and disease-free
deep invasion (p < 0.00001, Fig. 4). There was no difference in disease-free survival or overall survival in Stage IA when compared to Stage IB. Table 2 reveals a correlation between increasing histologic grade and probability of deep myometrial invasion. It is apparent that the risk of having deep myometrial invasion increased with increasing histologie grade (p < 0.000 1, chi square). To characterize risk factors more precisely, multivariate analysis was performed using the Cox Proportional Hazards Method. Variables entered included age, Stage (IA VSIB), histologie grade (1 VS2 VS3), depth of myometrial invasion (5: VS>$) and treatment group (1 VS2). Deep myometrial invasion (p < 0.0001) and increasing histologie grade (p = 0.007) each carried independent adverse prognostic significante. Inclusion within treatment group 1 carried independently favorable prognostic significante compared to group 2 for both 5-year disease-free (p = 0.0 18) and overall (p = 0.001) survival. Patient age and FIGO tumor Stage (IA VSIB) carried no independent prognostic significante. Further analysis of high risk patients revealed that those
ovonll
survival: al1 patients.
with grade 3 histology appeared to have superior diseasefree and overall survival during the period of routine preoperative intracavitary cesium application ( 1974 through 1976, group 1) compared to those treated under the more recent policy of no preoperative therapy (198 1 through 1983, group 2). Disease-free 5-year survival rates in these two groups measured 76% VS53%, respectively (Fig. 5). Because of relatively smal1 patient numbers in each subgroup, this finding did not achieve statistical significante. A similar trend towards improved disease-free and overall survival for patients with deep myometrial invasion was also noted in patient group 1. This group had an 84% 5year disease-free survival compared to 69% for corresponding patients treated under the group 2 policy (Fig. 6). Figure 7 reveals the benefit to routine preoperative cesium use (group 1) appeared to be restricted to “high risk” patients, that is, those with deep myometrial invasion and/ or grade 3 lesions. The remaining majority of “10~ risk” patients (grade 1 or 11,
§urvlvd
FNO Sunlval
Dbnn
-80 60
J
0 OAupl . uao2
;
w 40
: 20
0
12
24
36
m0(hmth8)
48
80 0
I
12
-
.
24
-
I 36
-W-W
Fig. 2. Overall and disease-free survival; Group 1 vs. Group 2.
-
I
48
-
+o
60
f I
1. J. Radiation Oncology 0 Biology 0 Physics
1014
Disease Free Survival
F
80 -
5 '5 3 cn
60 -
E 0
40 -
Pài
20 -
i
0 0
0 A
Gradel GradeII ?? Grade 111
1. 12
,* 24
ps 0.0001
I. 36
1. 46
I 60
Time (Months) Fig. 3. Disease-free
survival for al1 patients
according
to Grade.
Patterns offailure Local control was excellent in both patient groups. There were only five locoregional failures in the entire series of 229 patients (2%) and this did not vary between group 1 (2 of 112 patients, 1.7%) and group 2 (3 of 117 patients, 2.5%). A single, isolated vagina1 failure was observed in each group and in both patients this was noted in the distal vagina. There were no infield pelvic failures noted in the 40 high risk patients treated postoperatively with external beam radiotherapy, with or without a preoperative cesium application. Three of these 40 patients did have regional failure but two were in the distal vagina and one was in the labia, al1 outside the treated volume. Abdominal failure was the most common relapse pattem in our series, accounting for 12 of 2 1 total failures. This usually took the form of peritoneal carcinomatosis. The risk of abdominal failure for high risk patients was 12% (7 of 57 patients). Only 5 of the remaining 172 patients (2.9%) failed in the abdomen. DISCUSSION The 5-year actuarial disease-free and overall survival rates of 92% and 84% for our patients with surgically treated Stage 1 endometrial carcinoma are similar to most other reported series. (1, 2, 3, 4, 6, 8, 9, 10, 14, 15, 16, 18, 19) Our findings of adverse prognostic significante associated with increasing histologie grade (p < 0.0000 1) or deep myometrial invasion (p < 0.0001) have also been reported in multiple previous series. (1, 3, 7, 14, 15, 18, 19, 21) The unexpected finding in our analysis was that patients treated under a routine policy of an immediate properative tandem and ovoid intracavitary cesium application (group 1) fared better than the more recently treated group 2 patients (no preoperative therapy) according to the Cox Proportional Hazards Method. Superiority for 5-year disease-free survival (p = 0.018) and overall survival (p = 0.001) was noted in group 1 patients. This occurred even with high risk patients from both groups receiving
May 1990, Volume 18, Number 5
a similar volume, full dose postoperative extemal beam pelvic radiotherapy to 4500-5000 cGy. Inspection of Table 1 reveals that group 1 contained a smaller proportion of patients with grade 1 lesions than group 2 (46% VS 7 1%). Conversely, grade 2 or 3 lesions were more prevalent in group 1 compared to group 2 (44% VS 26%). Additionally, deep myometrial invasion was slightly more prevalent in group 1 (24% VS 18%). Slightly superior 5-year actuarial disease-free and overall survival rates were obtained in group 1 (Fig. 2) although this group contained a higher proportion of patients with unfavorable histologie features. Further analysis revealed that the improved prognosis in group 1 appeared to be restricted to high risk (grade 3 and/or > i myometrial invasion) patients (Fig. 5-8). Since the only major differente in treatment policy for group 1 was the routine administration of preoperative cesium and since similar proportions of high risk patients in each group received postoperative extemal beam pelvic radiotherapy, our data suggest that the preoperative cesium application may have been responsible for the decreased recurrence and death rate seen in group 1. Pelvic failure was extremely uncommon in either treatment group ( 1.7% group 1 VS2.5% group 2) and the predominant failure pattern was distant. Since the overall risk of recurrence was significantly lower in group 1, this implies that the preoperative cesium application may have exerted its favorable prognostic effect by devitalizing tumor cells and diminishing their ability to metastasize at the time of surgical manipulation. The advantage to performing the intracavitary cesium application immediately before surgery was that histopathologie findings in the resected uterus were not significantly altered and high risk patients could stil1 be selected for additional postoperative treatment. A review of literature on Stage 1 endometrial carcinoma reveals a wide diversity of opinion regarding optimal treatment for these patients. Approximately 80% of al1
Disease Free Survlval
Depltls1n cmpth>l)3
0 8
01
0
12
24 Tlme
36
ps o.oca
46
60
(Monthr)
Fig. 4. Disease-free survival for al1 patients of myometrial invasion.
according
to depth
Cesium vs. external beam radiation 0 W. T. SAUSEet al. Table 2. Correlation
between
1015
grade and depth of myometrial
invasion
Grade 1 Depth
3
Unk.
Total
No.
(W)
No.
((70)
No.
(%ó)
No.
@)
(85)
35
(61)
5
(23)
21
(78)
166
(72)
13 5
(11) (4)
193
(33) (6)
12 5
(22) (55)
3
(11)
47 16
(21) (7)
122
(100)
57
(100)
22
(100)
27
(100)
229
(100)
No.
@)
104 %k. q Total
2
Stage 1 endometrial carcinoma patients are cured by TAHBSO and some authors question the ability of combined modality therapy to improve these results. ( 19,2 1) Many other authors advocate the use of combined radiation and surgery in al1 patients or in selected patients with adverse prognostic features such as increased histologie grade or deep myometrial invasion. ( 1. 2, 3, 4, 6, 7. 8, 9, 10, 12. 14, 17, 18, 19) The principle benefit of combined modality therapy in most reported series has been a reduction in the vagina1 and pelvic failure rate. ( 1, 3, 4, 8, 9, 10, 15, 17) Brady et al. (4) reported that the rate of vagina1 persistente after hysterectomy alone for Stage 1 adenocarcinoma of the endometrium is 11% compared to 2% for combined preoperative radiation therapy and hysterectomy. Aalders et al. (1) showed in a large prospective randomized Norwegian trial that postoperative extemal pelvic radiotherapy significantly reduced the vagina1 and pelvic failure risk in Stage 1 endometrial adenocarcinoma when added to postoperative intravaginal radium application, compared to postoperative intravaginal radium alone. This benefit appeared to be most striking in patients with grade 3 histology and deep myometrial invasion in whom the addition of postoperative extemal beam pelvic radiotherapy reduced the risk of vagina1 and pelvic recurrence from 19.6% to 4.5%. Overall survival was not improved
in any patient subgroup, however, due to an increased incidence of distant metastases in the more aggressively treated group. Another prospective randomized trial in Stage 1 endometrial carcinoma reported by Graham (8) and updated by Piver et al. ( 15, 16) showed a slight but statistically insignificant survival improvement for either preoperative intrauterine or postoperative intravaginal radium application compared to hysterectomy alone at 5 and 10 years. Again, vagina1 failure rates were significantly reduced in the combined modality arms. Weigensburg (20) compared preoperative external beam radiation to preoperative intracavitary radiation in clinical Stage 1 carcinoma of the endometrium patients. He concluded that preoperative intracavity treatment was superior to extemal beam therapy. Several retrospective series have suggested a survival benefit to combined modality therapy. Gusberg, and Yannopoulos ( 10) in 1964 reported that the cure rate for early stage endometrial carcinoma was significantly improved in a nonrandomized but unselected series from Columbia Presbyterian Medical Center (p = 0.01) when preoperative radium was administered, compared to patients treated with surgery alone. This benefit appeared to be most pronounced in those with anaplastic tumors, enlarged uteri or cervical involvement. Patients with Stage
Disease Frw Survival; Grade 111
Dlserase
Free Survlval; Depth > 1/3
E
‘5 5
3
tn
c
0 8 P
0
20
??
Gmupl GroupP
1
- 0.
12’
2’4
3s
-
48
.
0
60
Time (Months) Fig. 5. Disease-free Group 2.
survival
for Grade
3 patients;
12
24 lime
Group
1 vs.
36
48
60
(Months)
Fig. 6. Disease-free survival for patients invasion: Group 1 vs. Group 2.
with deep myometrial
1016
1. J. Radiation Oncology 0 Biology 0 Physics
May 1990, Volume 18, Number 5
Dlsease Free Survival; High Risk
Disaa~. Fr.. Survlval; Low Risk IOol-~
r
a
r ._ .z L
60 -
8 E
40 -
$ P8
20 -
0
Gnnql Grorp2
??
o! 0
.
, 12
.
, 24 llme
Fig. 7. Disease-free depth > f); Group
.
, 36
.
, 46
.
0: 0
(
60
,
12
-
,
24
lime
(Months)
survival for high risk patients 1 vs. Group 2.
.
(Grade
3 or
1 wel1 differentiated tumors appeared to have an excellent prognosis with or without preoperative radium. The authors concluded that TAH-BSO was sufficient treatment for small, wel1 differentiated tumors but advocated combined preoperative radium and surgery for al1 tumors that were more advanced. Beiler et al. (3) reported in 1972 that there was a borderline statistically significant improved survival rate in early stage endometrial carcinoma patients treated by combined preoperative radium and surgery compared to surgery alone (91% VS79%, p = 0.07). This benefit was restricted to patients with medium to high grade lesions. Monson et al. ( 14) also noted a significant improvement in survival with combined modality therapy in women with Stage 1, high grade endometrial carcinomas compared with surgery alone. In that study, a total abdominal hysterectomy followed by radiation appeared to give the best results even though the authors felt these patients were adversely selected to receive radiotherapy. In women with Stage 1 poorly differentiated tumors, 5-year survival measured 92% in the postoperative radiation group compared to 59% in the surgery alone group. Our own series confirms that patients with Stage 1, grade 1 or 2 superficially invasive endometrial carcinomas have an excellent prognosis with or without preoperative intracavitary radiotherapy with a 97% 5-year actuarial disease-free survival. We believe the remaining “high risk” patients may have a better prognosis when treated with combined preoperative intravaginal and intrauterine cesium immediately before the hysterectomy. The mechanism responsible for benefit from preoperative intracavitary therapy is unknown. Perhaps preop-
Fig. 8. Disease-free depth < f); Group
.
,
36
.
,
46
.
1
60
(Months)
survival for low risk patients 1 vs. Group 2.
(Grade
1 or 2,
erative radiation reduces the possibility of tumor seeding at surgery or perhaps sterilizes microscopic metastasis in pelvic tissues. Although the policy of external beam pelvic radiotherapy directed to high risk patients was perfectly adequate in preventing pelvic failure with or without the addition of intracavitary cesium, the preoperative intracavitary insertion did reduce the overall risk of failure in high risk patients. Our high risk category included al1 patients with grade 3 lesions or deep myometrial invasion. Since it is impossible to preoperatively assess the amount of myometrial invasion, it is necessary to use histologie grade to predict which patients are likely to have deep invasion to decide who wil1 benefit from preoperative treatment. Our patients with Stage 1, grade 1 lesions had only an 11% risk of deep myometrial penetration and similar results have been reported by other authors. ( 1,9) These patients have such a low probability of falling into the high risk category that preoperative therapy would not seem to be routinely indicated. Our grade 2 and 3 patients had a 33% and 55% risk of deep myometrial invasion, respectively. These are the patients who would appear to be most likely to benefit from a routine preoperative intracavitary cesium or radium application to maximize the probability of long-term diseasefree survival. Like most reported studies, ours is a retrospective although nonselected series. We would like to see the value of a routine preoperative intracavitary application formally tested in a prospective randomized trial designed to include women with clinical Stage 1, moderately to poorly differentiated endometrial carcinomas.
REFERENCES 1. Aalders, J.; Abeler, V.; Kolstad, P.; Onsrud, M. Postoperative extemal irradiation and prognostic parameters in stage 1 endometrial carcinoma-clinical and histopathologie study of 540 patients. Obstet. Gynecol. 56:4 19-426; 1980. 2. Bedwinek, J.; Galakatos, A.; Gamel, M.; Ming-Shian, K.;
Stokes, S.; Perez, C. Stage 1, Grade 111adenocarcinoma of the endometrium treated with surgery and irradiation-sites of failure and correlation of failure rate with irradiation technique. Cancer 54~40-47; 1984. 3. Beiler, D. D.; Schmutz, D. A.; O’Rourke, T. L. Carcinoma
Cesium vs. external kam radiation 0 W. T. SAUSErf al. of the endometrium: radiation and surgery versus surgery alone. Radiology 102: 159- 164; 1972. 4. Brady, L. W.; Lewis, G. C.; Antoniades, J.; Prasasvinichai, S.: Torpie, R. J.; Asbell, S. 0.; Glassburn, J. R.; Schatanoff, D.: MacMurray, T. Evolution of radiotherapeutic techniques. Gynecol. Oncol. 2:314-323: 1974. 5. Cox, D. R. Regression models and life tables. J. R. Stat. Sec. 34: 187-220; 1972. 6. Eifel, P. J.; Ross, J.; Hendrickson, M.; Cox, R. S.: Kempson, R.; Martinez, A. Adenocarcinoma of the endometriumanalysis of 256 cases with disease limited to the uterine corpus: treatment comparisons. Cancer 52: 1026- 103 1; 1983. 1. Frick, H. C.: Munnell, E. W.; Richart, R. M.; Berger, A. P.: Lawry, M. F. Carcinoma ofthe endometrium. Am. J. Obstet. Gynecol. I 15663-672; 1973. 8. Graham, J. The value of preoperative or postoperative treatment by radium for carcinoma of the uterine body. Surg. Gynecol. Obstet. 132:855-860; May 1971. 9. Gusberg, S. B.; Jones, H. C.: Tovell. H. M. M. Selection of treatment for corpus cancer. Am. J. Obstet. Gynecol. 80: 374-380; 1960. D. Therapeutic decisions in 10. Gusberg. S. B.: Yannopoulos, corpus cancer. Am. J. Obstet. Gynecol. 88: 157-162; 1964. estimation from 11. Kaplan, E. L.; Meier P. Nonparametric incomplete observations. J. Am. Stat. Assoc. 53:457-48 1: 1958. 12. Landgren. R. D.; Fletcher, G. H.; Gallager, S.: Declos, L.; Wharton, J. T. Treatment failure sites according to irradiation technique and histology in patients with endometrial cancer. Cancer 40: 13 1-135: 1977.
1017
13. Mantel, N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother. Rep. 50:163-170: 1966. 14. Monson, R. R.; MacMahon, B.; Austin, J. H. Postoperative irradiation in carcinoma of the endometrium. Cancer 3 1: 630-632; 1973. 15. Piver, M. S. Stage 1 endometrial carcinoma: the role of adjunctive radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 6:367-368: 1980. L.; Tsukada, Y. A 16. Piver, M. S.; Yazigi, R.; Blumenson, prospective trial comparing hysterectomy, hysterectomy plus vagina1 radium, and uterine radium plus hysterectomy in stage 1 endometrial carcinoma. Obstet. Gynecol. 54:85-89: 1979. 17. Salazar, 0. M.; Feldstein, M. L.; DePapp, E. W.; Bonfiglio. T. A.; Keller. B. E.; Rubin. P.; Rudolph J. H. Endometrial carcinoma: analysis of failures with special emphasis on the use of initial preoperative external pelvic radiation. Int. J. Radiat. Oncol. Biol. Phys. 2: 1 10 1- 1 107; 1977. 18. Sall. S.: Sonnenblick, B.; Stone. M. L. Factors affecting surviva1 of patients with endometrial adenocarcinoma. Am. J. Obstet. Gynecol. 107:116-123: 1970. 19. Stokes. S.; Bedwinek. J.; Kao. M. S.; Camel. H. M.: Perez, C. A. Treatment of stage 1 adenocarcinoma of the endometrium by hysterectomy and adjuvant irradiation: a retrospective analysis of 304 patients. Int. J. Radiat. Oncol. Biol. Phys. 12:339-344; 1986. 20. Weigensberg, 1. J. Preoperative radiation therapy in stage 1 endometrial adenocarcinoma. Cancer 53:242-247; 1984. 21. Whetham, J. C. G.: Bean, J. L. M. Carcinoma of the endometrium. Am. J. Obstet. Gynecol. 1 I2:339-343; 1972.