- Email: [email protected]
Stage IB or IIA-B carcinoma of the intact uterine cervix ≥ 6 cm in diameter: Is adjuvant extrafascial hysterectomy beneficial?
Inf. J. Radiation Oncology Bid. Phys. Vol. 21, pp. WHO4 Printed in the U.S.A. All rights reserved.
Copyright
0360.3016/5’1 $3.00 + .M) 0 1991 Per&mm Press plc
??Original Contribution
STAGE IB OR IIA-B CARCINOMA OF THE INTACT UTERINE CERVIX 2 6 CM IN DIAMETER: IS ADJUVANT EXTRAFASCIAL HYSTERECTOMY BENEFICIAL? WILLIAM M. MENDENHALL, M.D.,* PATRICIA J. MCCARTY, B.A.,* LINDA S. MORGAN, M.D.,+ WELDON E. CHAFE, M.D.+ AND RODNEY R. MILLION, M.D.* Universityof FloridaCollegeof Medicine, Gainesville,FL This is an analysis of 150 patients with Stage IB or IIA-B carcinoma of the intact uterine cervix 2 6 cm in diameter treated with irradiation alone (75 patients) or irradiation followed by surgery (75 patients) at the University of Florida between October 1964 and June 1983. Minimum follow-up in this series was 5 years. There was no significant difference in the distribution of prognostic factors between the two treatment groups. The 5-year local control rate was 74% with irradiation alone and 76% with irradiation and surgery. The 5-year survival rates for irradiation alone versus irradiation plus surgery were as follows: cause specific, 62% and 5596, and absolute, 54% and 52%. The proportion of patients who developed treatment complications necessitating hospitalization or a second operation was 4/75 (5%) after irradiation alone and 12/75 (16%) after irradiation and surgery. The authors conclude that the routine use of adjuvant extrafascial hysterectomy is not warranted in this patient population. Cervix neoplasms, Radiotherapy, Surgery, Combined modality therapy. INTRODUCTION
followed by surgery (75 patients) at the University of Florida between October 1964 and June 1983. Patients with lesions < 6 cm in diameter were not included in the analysis as they were rarely treated with planned combined irradiation and surgery. Histopathology slides on all patients were reviewed by the Department of Pathology at the University of Florida at the time of referral. During the early part of the series, most patients were treated with irradiation alone, whereas irradiation plus adjunctive extrafascial hysterectomy was used for the majority of patients treated during the latter part of the series. Seventytwo patients developed recurrent disease: 83% at 12 years, 90% at 13 years, and 94% at 15 years. Four late (>5 years) recurrences were noted, including three local failures at 7,9, and 12 years, and one paraaortic node failure at 6.5 years. Four patients were lost to follow-up alive with no evidence of disease within 5 years of treatment: two patients treated with irradiation alone (at 1.5 and 4 years), and two patients treated with irradiation plus surgery (both at 2 years). One additional patient treated with irradiation and surgery was lost to follow-up with distant metastasis 7 months following treatment; she was coded as alive with distant metastasis alone at that time. The remaining patients had follow-up for at least 5 years or
The management of bulky or barrel-shaped Stage IB or IIA-B carcinoma of the cervix is controversial (5, 13, 16, 20). Although the FIG0 staging system does not take tumor size into account, increasing tumor diameter has been shown to affect prognosis adversely, prompting a search for more effective treatment (1-3, 11, 15). Most authors recommend treatment with either irradiation alone or irradiation followed by adjuvant extrafascial hysterectomy (3-5, 10-l 3, 17, 20). Primary radical hysterectomy is rarely indicated because many patients will have indications for postoperative irradiation, and the combination of these two modalities is associated with a relatively high risk of treatment complications. The purpose of this study is to evaluate the benefit of adding extrafascial hysterectomy after irradiation for Stage IB or IIA-B carcinoma of the intact uterine cervix r 6 cm in diameter.
METHODS
AND
MATERIALS
This is a retrospective analysis of 150 patients with carcinoma of the intact uterine cervix 2 6 cm in diameter managed with irradiation alone (75 patients) or irradiation
Presentedat the 32nd Annual Meeting of the AmericanSo-
Reprintrequeststo: WilliamM. Mendenhall,M.D., Dept. of Radiation Oncology, J-385, J. Hillis Miller Health Center, Gainesville, FL 32610-0385. Accepted for publication 25 January 199 1.
ciety for Therapeutic Radiology and Oncology, Miami Beach, FL, October 15-19, 1990. * Department of Radiation Oncology. +Department of Obstetrics and Gynecology. 899
I. J. Radiation Oncology 0 Biology 0 Physics
900
until death. The distribution of possible prognostic factors and treatment factors for both treatment groups is listed in Table 1. Only one patient presented with a tumor more than 9 cm in diameter; this was a IO-cm lesion and was managed with irradiation and surgery. Thirteen (8 1%) of 16 adenocarcinomas were managed with irradiation and surgery, reflecting the bias that, stage for stage, adenocarcinema was more radioresistant than squamous cell carcinoma. Most radiation oncologists now believe that there is no significant difference in radiation sensitivity between the two histologies. Older patients who were at high risk for complications of a major surgical procedure tended to be treated with irradiation alone. The parameter “percentage tumor regression at first radium application” was evaluated to determine if patients with lesions that exhibited minimal regression during the external-beam portion of the irradiation course were selected to undergo adjunctive hysterectomy. The distribution of tumor regression was similar in both treatment groups. No information on tumor regression was available for 13 patients treated with irradiation and surgery, including 5 patients for whom the information was not recorded and 8 patients who received preoperative radiation therapy consisting of external-beam irradiation alone. Table 1. Patient population
Parameter Stage IB IIA
IIB Tumor diameter 6-7 cm r8 cm Histology Squamous carcinoma Adenocarcinoma Adenosquamous carcinoma Age 545 years >45 years Pretreatment hematocrit No data 535 >35 Percentage tumor regression at first radium application No data 525% 26%-50% 251% Treatment Continuous-course irradiation Split-course irradiation
Irradiation alone (75 patients)
Irradiation and surgery (75 patients)
No. (%)
No. (%)
23 (31)
28 (37) 10 (13)
6 (8) 46 (61)
37 (50)
60 (80) 15 (20)
56 (75) 19 (25)
72 (96) 2 (3)
62 (83) 13 (17)
1 (1)
0 (0)
35 (47) 40 (53)
43 (57) 32 (43)
l(1) 35 (47) 39 (52)
5 (7) 25 (33) 45 (60)
1 (1) 19 (25) 50 (67) 5 (7)
13 (17) 12 (16) 44 (59) 6 (8)
60 (80) 15 (20)
65 (88) 9 (12)
September 1991, Volume 21, Number 4
Before treatment, patients routinely had a hemogram, blood chemistry profile, chest roentgenogram, and intravenous pyelogram. Selected patients underwent bipedal lymphangiogram and/or computed tomographic (CT) scan of the abdomen and pelvis. Patients were clinically staged according to the FIG0 staging system (8); no patient underwent surgical staging before radiation therapy. Pretreatment tumor diameter was assessed on the basis of rectal examination. The treatment techniques have been reported elsewhere ( 11, 19,20). Briefly, patients treated with irradiation alone received approximately 4000 cGy to the true pelvis, calculated at the central axis, using a four-field box at 170 to 180 cGy per fraction, one fraction per day, 5 days per week. Following external-beam irradiation, treatment was completed with two 46- to 48-hour intracavitary radium applications to deliver a total intracavitary dose of 5000 to 6000 mg-hours. Patients treated with irradiation plus surgery received approximately 4000 cGy to the true pelvis followed by one 69- to 72-hour intracavitary radium application to deliver approximately 4500 mg-hours. In the combined treatment group, one patient was treated with an intracavitary radium application alone followed by surgery and eight patients received external-beam irradiation alone before surgery. All patients receiving external-beam irradiation were treated with megavoltage equipment; 6oCo was used in 133 patients (89%). Extemalbeam irradiation was delivered using a continuous-course technique in 60 patients (80%) treated with irradiation alone and 65 patients (88%) treated with irradiation and surgery; the remainder were treated with split-course irradiation ( 14). The split-course technique was used between 1970 and 1974 and has since been discontinued (14). In the combined-treatment group, surgery was performed approximately 4 to 6 weeks following the completion of radiation therapy. The operations performed were extrafascial hysterectomy (57 patients), modified radical hysterectomy (1 patient), anterior exenteration (1 patient), total exenteration (2 patients), and laparotomy without resection ( 13 patients). Patients undergoing laparotomy only did not undergo resection for the following reasons: liver metastasis (2 patients), extensive lymph node metastasis ( 10 patients), and unresectable primary tumor (1 patient). The additional patient refused surgery following planned preoperative irradiation. The radiation therapy and hospital charts of all patients were reviewed and abstracted. Recurrent disease in the pelvis or extrapelvic sites was coded as such, regardless of whether it represented the first site of relapse or appeared subsequent to recurrent disease elsewhere. The rates of disease control, survival, and serious complications were calculated by the product-limit method (7, 18); significance levels between curves were calculated using the logrank test (9, 18). The four patients lost to follow-up within 5 years of treatment were censored from the analyses at the date of last follow-up. Multivariate analyses were performed using the forward stepwise log-rank tests of as-
901
Carcinoma of the cervix 0 W. M. MENLXWHALL er al.
sociation of covariates (6, 18). The parameters evaluated in the multivariate analyses were as follows: stage (IB-IIA vs IIB), size (6-7 cm vs r8 cm), histology (squamous cell carcinoma vs adenocarcinoma vs adenosquamous carcinoma), age at diagnosis (145 years vs >45 years), pretreatment hematocrit (135 vs >35), tumor regression at first radium application (~25% vs >25%), and treatment (irradiation vs irradiation plus surgery). In the multivariate analysis of distant metastasis, the impact of local control was also examined. RESULTS
Control of disease in the pelvis The rates of pelvic control for both treatment groups are depicted in Figure 1. There is no significant difference between the curves. The 5-year pelvic control rates for irradiation alone and irradiation plus surgery were as follows: IB-IIA, 75% and 76%; IIB, 74% and 76%; overall, 74% and 76%. Twenty-four patients treated with irradiation alone developed pelvic recurrence; three occurred at 7 years, 9 years, and 12 years following treatment and may represent second primary malignancies. Five patients (2 1%) underwent attempted surgical salvage after pelvic recurrence: radical hysterectomy ( 1 patient), anterior exenteration (3 patients), and laparotomy only (1 patient). Two patients (8%) were successfully salvaged and are alive and diseasefree 3 years and 13 years postoperatively. Thirty-one patients treated with irradiation and surgery developed a pelvic recurrence, and six (19%) underwent attempted surgical salvage procedures: total exenteration (3 patients) and laparotomy alone (3 patients). The salvage procedure was successful in one patient, who is alive and diseasefree 12 years following total exenteration. The results of the multivariate analysis for pelvic control are shown in Table 2. None of the parameters listed significantly affected the likelihood of pelvic control. In particular, the addition of surgery to irradiation did not reduce the risk of pelvic recurrence.
Table 2. Control of disease in the pelvis: multivariate analysis Significance level
Parameter
Rank order 1 2 3 4
Age Treatment Regression Size
0.1427 0.2407 0.5180 0.466 1
: 7
Hematocrit Stage Pathology
0.5808 0.7431 0.9892
Distant metastasis The incidence of distant metastasis (extrapelvic recurrence) for both treatment groups is depicted in Figure 2. At 5 years the probability of distant metastasis was 17% for the group treated with irradiation alone and 16% for the group treated with irradiation and surgery. Three patients treated with irradiation alone underwent attempted surgical salvage for a solitary lung metastasis noted at 1, 2, and 4 years following initial treatment. All three patients are alive and disease-free at 11, 13.5, and 19.5 years following the salvage procedure. One patient treated with irradiation and surgery underwent attempted salvage with surgery and irradiation for a solitary lung metastasis presenting 1 year after initial treatment. She is alive and disease-free 18 years following salvage treatment. The results of the multivariate analysis for distant metastasis are shown in Table 3. None of the parameters examined significantly influenced the risk of distant metastasis.
Cause-specificsurvival Cause-specific survival rates (censoring non-cancer-related deaths) for the two treatment groups are shown in Figure 3; there is no significant difference between the survival curves. The 5-year cause-specific survival rates were 62% for irradiation alone and 55% for irradiation
1oou
602
20%
1
RT
Ii
Alon*
-
RT . surpery
i8U
I
17x
,’
OUi( 0
Fig. 1. Control of disease in the pelvis (product-limit method). Date of pelvic recurrence was the end-point of the analysis; patients who never had pelvic recurrence or who were salvaged were censored at the time of last follow-up or the time of death.
1
2
3
4 TIME
5 6 (Yeare)
7
8
9
10
Fig. 2. Incidence of distant metastasis (product-limit method). Date of manifestation of distant metastasis was the end point of the analysis; patients who never had distant metastasis were censored at the time of last follow-up or the time of death. The resulting probabilities of freedom from distant metastasis were subtracted from 1 to calculate the probabilities of having distant metastasis.
902
I. J. Radiation Oncology 0 Biology 0 Physics
Table 3. Probability of distant metastasis: multivariate analysis
September 1991, Volume 21, Number 4 Table 4. Cause-specific
Rank order
Parameter
Significance level
1 2 3 4 5 6 7 8
Stage Treatment Size Regression Age Local control Hematocrit Pathology
0.1635 0.2774 0.3255 0.3358 0.3737 0.6276 0.868 1 0.9562
plus surgery. The results of the multivariate analysis for cause-specific survival are shown in Table 4. The only parameter that influenced cause-specific survival was tumor diameter; tumors 6-7 cm in diameter were associated with a significantly higher cause-specific survival rate than those 28 cm in diameter. Treatment did not exert a significant impact on this one point.
Absolute survival The absolute survival rates are depicted in Figure 4; the 5-year absolute survival rates were 54% for patients treated with irradiation alone and 52% for those treated with irradiation plus surgery. The results of the multivariate analysis for absolute survival are shown in Table 5. Pathology was of borderline significance, in favor of squamous cell carcinoma.
Complications Severe complications were defined as those resulting in hospitalization or necessitating an operation. Four patients (5%) treated with irradiation alone developed the following severe complications: rectosigmoiditis necessitating colostomy (2 patients), small bowel obstruction requiring
survival: multivariate analysis
Rank order
Parameter
Significance level
1
Size Regression Age Pathology Hematocrit Treatment Stage
0.0354 0.2094 0.2680 0.3554 0.650 1 0.6323 0.9326
2 3 4 5 6 7
surgical intervention (1 patient), and vesicaovaginal fistulae requiring an ileal conduit (1 patient). Severe complications were noted in 12 patients ( 16%) treated with irradiation and surgery: pyelonephritis necessitating hospitalization (1 patient), ureteral obstruction resulting in repeated pyelonephritis and renal insufficiency (1 patient), rectovaginal fistula necessitating colostomy (1 patient), abdominal wound abscess requiring hospitalization for incision and drainage (2 patients), small bowel obstruction necessitating hospitalization for conservative management (1 patient) or resulting in surgical intervention (2 patients), pyometra requiring hospitalization (1 patient) or surgical intervention (1 patient), pelvic abscess necessitating surgical exploration (1 patient), and ureterovaginal fistula resulting in ureteral obstruction and necessitating nephrectomy (1 patient). The incidence of severe complications (product-limit method) for patients treated with irradiation alone was 7% at 3.5 years and stable thereafter. For patients treated with irradiation and surgery, the incidence of severe complications was 15% at 1 year, 18% at 6 years, and stable thereafter. The difference between these curves is significant at the p = 0.027 level. DISCUSSION
In 1969, Durrance et al. (1) reported that, in patients with Stage II cervix cancer, most recurrences developed
60% 66%
RT *lone
0
1
2
3
4 TIME
-
RT . l”,p.ry
6 6 (mar*)
7
6
9
10
Fig. 3. Cause-specific survival (product-limit method). Date of death with cancer present was the end point of the analysis; patients not known to be dead or who did not have cancer present at death were censored at the time of last follow-up or time of death.
Fig. 4. Absolute survival (product-limit method). Death due to any cause was the end point of the analysis; patients not known to be dead were censored at the date of last follow-up.
Carcinoma of the cervix 0 W. M. MFSDENIWLL e? al.
Table 5. Absolute survival: multivariate
analysis
Rank order
Parameter
Significance level
1 2 3 4 5 6 7
Pathology Size Regression Treatment Stage Hematocrit Age
0.077 1 0.1239 0.3624 0.5797 0.7086 0.7830 0.8498
in patients with bulky or barrel-shaped endocervical carcinomas. Fletcher, in the 1970 Janeway Lecture (4), noted that the combination of irradiation and extrafascial hysterectomy resulted in a central pelvic recurrence rate of only 2% for patients with bulky or barrel-shaped Stage IB-II cervical carcinoma. Nelson et al. (12) subsequently reported that irradiation and extrafascial hysterectomy markedly reduced the risk of central pelvic recurrence and resulted in survival rates comparable to overall survival rates for similarly staged lesions. After those reports, various authors have confirmed ( 10, 13) or refuted (2, 15, 20) the benefit of irradiation plus extrafascial hysterectomy over irradiation alone. There are several possible explanations for these conflicting reports. First, “bulky or barrel-shaped” is a subjective term and does not precisely define the size of the lesion. For example, from the University of Kentucky, Gallion et al. (5) defined bulky or barrel-shaped Stage IB cervix cancer as >5 cm diameter, whereas Maruyama et al. (lo), from the same institution, included 27 patients with tumors 3 cm in diameter in a series of 80 patients
903
with “bulky or barrel-shaped” Stage IB cervical carcinomas. Second, irradiation plus surgery series often exclude patients who undergo operation but not resection because of advanced disease, thus biasing a comparison of the data against patients treated with irradiation alone. Third, some series contain a mixture of patients who are clinically and surgically staged. For example, a patient may be taken to surgery for a radical hysterectomy and found to have multiple positive nodes. If the resection is aborted and the patient treated with irradiation alone, inclusion of the patient in the “irradiation alone” group would create a bias against that treatment group, compared with patients treated with irradiation and surgery who are clinically staged. Fourth, some series exclude patients who were to undergo an extrafascial hysterectomy following irradiation, but in whom a more extensive surgical procedure was found to be necessary to extirpate the tumor completely. All of these factors may render a meaningful comparison between irradiation alone and irradiation plus extrafascial hysterectomy nearly impossible. The data presented in this paper demonstrate that irradiation alone provides pelvic control and survival rates similar to those obtained with irradiation and surgery for Stage IB and II lesions 16 cm in diameter. Additionally, the incidence of severe treatment complications is lower with irradiation alone. Our conclusion is that the routine use of extrafascial hysterectomy for bulky Stage IB or IIAB cervix carcinoma is not indicated. Extrafascial hysterectomy may be indicated following irradiation for the exceptional case in which the patient is pregnant, the geometry of the radium application is poor, or the uterus is enlarged because of possible tumor involvement.
REFERENCES 1. Durrance, F. Y.; Fletcher, G. H.; Rutledge, F. N. Analysis of central recurrent disease in Stages I and II squamous cell carcinomas of the cervix or intact uterus. Am. J. Roentgenol. 106:831-838; 1969. 2. Eifel, P. J.; Morris, M.; Oswald, M. J.; Wharton, J. T.; Delclos, L. Adenocarcinoma of the uterine cervix: prognosis and patterns of failure in 367 cases. Cancer 65:2507-25 14; 1990. 3. Einhom, N.; Patek, E.; Sjoberg, B. Outcome of different treatment modalities in cervix carcinoma Stage IB and IIA: observations in a well-defined Swedish population. Cancer 55:949-955; 1985. 4. Fletcher, G. H. Cancer of the uterine cervix: Janeway Lecture, 1970. Am. J. Roentgenol. 111:225-242; 1971. 5. Gallion, H. H.; van Nagelk Donaldson, E. S.; Hanson, M. B.; Powell, D. E.; Maruyama, Y.; Yoneda, J. Combined radiation therapy and extrafascial hysterectomy in the treatment of Stage IB barrel-shaped cervical cancer. Cancer 56:262-265; 1985. 6. Kalbfleisch, J. D.; Prentice, R. L. The statistical analysis of failure time data. New York: Wiley; 1980. 7. Kaplan, E. L.; Meier, P. Nonparametric estimation from incomplete observations. J. Am. Stat. Assoc. 53:457-481; 1958.
8. Kottmeier, H. L. Classification and staging of malignant tumors in the female pelvis. Int. J. Gynecol. Obstet. 9: 172179; 1971. 9. Lawless, J. E. Statistical models and methods for lifetime data. New York: Wiley; 1982:420-422. 10. Maruyama, Y.; van Nagell, J. R.; Yoneda, J.; Donaldson, E.; Gallion, H.; Higgins, R.; Powell, D.; Kryscio, R.; Bemer, B. Dose-response and failure pattern for bulky- or barrelshaped Stage IB cervical cancer treated by combined photon irradiation and extrafascial hysterectomy. Cancer 63:7076; 1989. 11. Mendenhall, W. M.; Thar, T. L.; Bova, F. J.; Marcus, R. B.; Morgan, L. S.; Million, R. R. Prognostic and treatment factors affecting pelvic control of Stages IB and IIA-B carcinomas of the intact uterine cervix treated with radiation therapy alone. Cancer 53:2649-2654; 1984. 12. Nelson, A. J.; Fletcher, G. H.; Wharton, J. T. Indications for adjunctive conservative extrafascial hysterectomy in selected cases of carcinoma of the uterine cervix. Am. J. Roentgenol. 123:91-99; 1975. 13. Q’Quinn, A. G.; Fletcher, G. H.; Wharton, J. T. Guidelines for conservative hysterectomy after irradiation. Gynecol. Oncol. 9:68-79; 1980. 14. Parsons, J. T.; Thar, T. L.; Bova, F. T.; Million, R. R. An
904
I. J. Radiation Oncology 0 Biology 0 Physics
evaluation of split-course irradiation for pelvic malignancies. Int. J. Radiat. Oncol. Biol. Phys. 6: 17% 18 1; 1980. 15. Perez, C. A.; Kao, M. S. Radiation therapy alone or combined with surgery in treatment of barrel-shaped carcinoma of the uterine cervix (Stages IB, IIA, IIB). Int. J. Radiat. Oncol. Biol. Phys. 11:1903-1909; 1985. 16. Rettenmaier, M. A.; Casanova, D. M.; Micha, J. P.; Moran, M. F.; Ramsanghani, N. S.; Syed, N. A.; Puthawala, A.; Disaia, P. J. Radical hysterectomy and tailored postoperative radiation therapy in the management of bulky-stage IB cervical cancer. Cancer 6312220-2223; 1989. 17. Rotman, M.; John, M. J.; Moon, S. H.; Choi, K. N.; Stowe, S. M.; Abitol, A.; Herskovic, T.; Sall, S. Limitations of ad-
September 1991, Volume 21, Number 4 junctive surgery in carcinoma of the cervix. Int. J. Radiat. Oncol. Biol. Phys. 5:317-332; 1979. 18. SAS Institute Inc. SAS technical report P-179, additional SAS/STAT procedures, release 6.03. Guy, NC: SAS Institute Inc.; 1988:49-89. 19. Thar, T. L.; Million, R. R.; Daly, J. W. Radiation treatment of carcinoma of the cervix. Semin. Oncol. 9:299-3 11; 1982. 20. Weems, D. H.; Mendenhall, W. M.; Bova, F. J.; Marcus, R. B.; Morgan, L. S.; Million, R. R. Carcinoma of the intact uterine cervix, Stages IB-IIA-B, ~6 cm in diameter: irradiation alone vs. preoperative irradiation and surgery. Int. J. Radiat. Oncol. Biol. Phys. 11:1911-1914; 1985.
Copyright
0360.3016/5’1 $3.00 + .M) 0 1991 Per&mm Press plc
??Original Contribution
STAGE IB OR IIA-B CARCINOMA OF THE INTACT UTERINE CERVIX 2 6 CM IN DIAMETER: IS ADJUVANT EXTRAFASCIAL HYSTERECTOMY BENEFICIAL? WILLIAM M. MENDENHALL, M.D.,* PATRICIA J. MCCARTY, B.A.,* LINDA S. MORGAN, M.D.,+ WELDON E. CHAFE, M.D.+ AND RODNEY R. MILLION, M.D.* Universityof FloridaCollegeof Medicine, Gainesville,FL This is an analysis of 150 patients with Stage IB or IIA-B carcinoma of the intact uterine cervix 2 6 cm in diameter treated with irradiation alone (75 patients) or irradiation followed by surgery (75 patients) at the University of Florida between October 1964 and June 1983. Minimum follow-up in this series was 5 years. There was no significant difference in the distribution of prognostic factors between the two treatment groups. The 5-year local control rate was 74% with irradiation alone and 76% with irradiation and surgery. The 5-year survival rates for irradiation alone versus irradiation plus surgery were as follows: cause specific, 62% and 5596, and absolute, 54% and 52%. The proportion of patients who developed treatment complications necessitating hospitalization or a second operation was 4/75 (5%) after irradiation alone and 12/75 (16%) after irradiation and surgery. The authors conclude that the routine use of adjuvant extrafascial hysterectomy is not warranted in this patient population. Cervix neoplasms, Radiotherapy, Surgery, Combined modality therapy. INTRODUCTION
followed by surgery (75 patients) at the University of Florida between October 1964 and June 1983. Patients with lesions < 6 cm in diameter were not included in the analysis as they were rarely treated with planned combined irradiation and surgery. Histopathology slides on all patients were reviewed by the Department of Pathology at the University of Florida at the time of referral. During the early part of the series, most patients were treated with irradiation alone, whereas irradiation plus adjunctive extrafascial hysterectomy was used for the majority of patients treated during the latter part of the series. Seventytwo patients developed recurrent disease: 83% at 12 years, 90% at 13 years, and 94% at 15 years. Four late (>5 years) recurrences were noted, including three local failures at 7,9, and 12 years, and one paraaortic node failure at 6.5 years. Four patients were lost to follow-up alive with no evidence of disease within 5 years of treatment: two patients treated with irradiation alone (at 1.5 and 4 years), and two patients treated with irradiation plus surgery (both at 2 years). One additional patient treated with irradiation and surgery was lost to follow-up with distant metastasis 7 months following treatment; she was coded as alive with distant metastasis alone at that time. The remaining patients had follow-up for at least 5 years or
The management of bulky or barrel-shaped Stage IB or IIA-B carcinoma of the cervix is controversial (5, 13, 16, 20). Although the FIG0 staging system does not take tumor size into account, increasing tumor diameter has been shown to affect prognosis adversely, prompting a search for more effective treatment (1-3, 11, 15). Most authors recommend treatment with either irradiation alone or irradiation followed by adjuvant extrafascial hysterectomy (3-5, 10-l 3, 17, 20). Primary radical hysterectomy is rarely indicated because many patients will have indications for postoperative irradiation, and the combination of these two modalities is associated with a relatively high risk of treatment complications. The purpose of this study is to evaluate the benefit of adding extrafascial hysterectomy after irradiation for Stage IB or IIA-B carcinoma of the intact uterine cervix r 6 cm in diameter.
METHODS
AND
MATERIALS
This is a retrospective analysis of 150 patients with carcinoma of the intact uterine cervix 2 6 cm in diameter managed with irradiation alone (75 patients) or irradiation
Presentedat the 32nd Annual Meeting of the AmericanSo-
Reprintrequeststo: WilliamM. Mendenhall,M.D., Dept. of Radiation Oncology, J-385, J. Hillis Miller Health Center, Gainesville, FL 32610-0385. Accepted for publication 25 January 199 1.
ciety for Therapeutic Radiology and Oncology, Miami Beach, FL, October 15-19, 1990. * Department of Radiation Oncology. +Department of Obstetrics and Gynecology. 899
I. J. Radiation Oncology 0 Biology 0 Physics
900
until death. The distribution of possible prognostic factors and treatment factors for both treatment groups is listed in Table 1. Only one patient presented with a tumor more than 9 cm in diameter; this was a IO-cm lesion and was managed with irradiation and surgery. Thirteen (8 1%) of 16 adenocarcinomas were managed with irradiation and surgery, reflecting the bias that, stage for stage, adenocarcinema was more radioresistant than squamous cell carcinoma. Most radiation oncologists now believe that there is no significant difference in radiation sensitivity between the two histologies. Older patients who were at high risk for complications of a major surgical procedure tended to be treated with irradiation alone. The parameter “percentage tumor regression at first radium application” was evaluated to determine if patients with lesions that exhibited minimal regression during the external-beam portion of the irradiation course were selected to undergo adjunctive hysterectomy. The distribution of tumor regression was similar in both treatment groups. No information on tumor regression was available for 13 patients treated with irradiation and surgery, including 5 patients for whom the information was not recorded and 8 patients who received preoperative radiation therapy consisting of external-beam irradiation alone. Table 1. Patient population
Parameter Stage IB IIA
IIB Tumor diameter 6-7 cm r8 cm Histology Squamous carcinoma Adenocarcinoma Adenosquamous carcinoma Age 545 years >45 years Pretreatment hematocrit No data 535 >35 Percentage tumor regression at first radium application No data 525% 26%-50% 251% Treatment Continuous-course irradiation Split-course irradiation
Irradiation alone (75 patients)
Irradiation and surgery (75 patients)
No. (%)
No. (%)
23 (31)
28 (37) 10 (13)
6 (8) 46 (61)
37 (50)
60 (80) 15 (20)
56 (75) 19 (25)
72 (96) 2 (3)
62 (83) 13 (17)
1 (1)
0 (0)
35 (47) 40 (53)
43 (57) 32 (43)
l(1) 35 (47) 39 (52)
5 (7) 25 (33) 45 (60)
1 (1) 19 (25) 50 (67) 5 (7)
13 (17) 12 (16) 44 (59) 6 (8)
60 (80) 15 (20)
65 (88) 9 (12)
September 1991, Volume 21, Number 4
Before treatment, patients routinely had a hemogram, blood chemistry profile, chest roentgenogram, and intravenous pyelogram. Selected patients underwent bipedal lymphangiogram and/or computed tomographic (CT) scan of the abdomen and pelvis. Patients were clinically staged according to the FIG0 staging system (8); no patient underwent surgical staging before radiation therapy. Pretreatment tumor diameter was assessed on the basis of rectal examination. The treatment techniques have been reported elsewhere ( 11, 19,20). Briefly, patients treated with irradiation alone received approximately 4000 cGy to the true pelvis, calculated at the central axis, using a four-field box at 170 to 180 cGy per fraction, one fraction per day, 5 days per week. Following external-beam irradiation, treatment was completed with two 46- to 48-hour intracavitary radium applications to deliver a total intracavitary dose of 5000 to 6000 mg-hours. Patients treated with irradiation plus surgery received approximately 4000 cGy to the true pelvis followed by one 69- to 72-hour intracavitary radium application to deliver approximately 4500 mg-hours. In the combined treatment group, one patient was treated with an intracavitary radium application alone followed by surgery and eight patients received external-beam irradiation alone before surgery. All patients receiving external-beam irradiation were treated with megavoltage equipment; 6oCo was used in 133 patients (89%). Extemalbeam irradiation was delivered using a continuous-course technique in 60 patients (80%) treated with irradiation alone and 65 patients (88%) treated with irradiation and surgery; the remainder were treated with split-course irradiation ( 14). The split-course technique was used between 1970 and 1974 and has since been discontinued (14). In the combined-treatment group, surgery was performed approximately 4 to 6 weeks following the completion of radiation therapy. The operations performed were extrafascial hysterectomy (57 patients), modified radical hysterectomy (1 patient), anterior exenteration (1 patient), total exenteration (2 patients), and laparotomy without resection ( 13 patients). Patients undergoing laparotomy only did not undergo resection for the following reasons: liver metastasis (2 patients), extensive lymph node metastasis ( 10 patients), and unresectable primary tumor (1 patient). The additional patient refused surgery following planned preoperative irradiation. The radiation therapy and hospital charts of all patients were reviewed and abstracted. Recurrent disease in the pelvis or extrapelvic sites was coded as such, regardless of whether it represented the first site of relapse or appeared subsequent to recurrent disease elsewhere. The rates of disease control, survival, and serious complications were calculated by the product-limit method (7, 18); significance levels between curves were calculated using the logrank test (9, 18). The four patients lost to follow-up within 5 years of treatment were censored from the analyses at the date of last follow-up. Multivariate analyses were performed using the forward stepwise log-rank tests of as-
901
Carcinoma of the cervix 0 W. M. MENLXWHALL er al.
sociation of covariates (6, 18). The parameters evaluated in the multivariate analyses were as follows: stage (IB-IIA vs IIB), size (6-7 cm vs r8 cm), histology (squamous cell carcinoma vs adenocarcinoma vs adenosquamous carcinoma), age at diagnosis (145 years vs >45 years), pretreatment hematocrit (135 vs >35), tumor regression at first radium application (~25% vs >25%), and treatment (irradiation vs irradiation plus surgery). In the multivariate analysis of distant metastasis, the impact of local control was also examined. RESULTS
Control of disease in the pelvis The rates of pelvic control for both treatment groups are depicted in Figure 1. There is no significant difference between the curves. The 5-year pelvic control rates for irradiation alone and irradiation plus surgery were as follows: IB-IIA, 75% and 76%; IIB, 74% and 76%; overall, 74% and 76%. Twenty-four patients treated with irradiation alone developed pelvic recurrence; three occurred at 7 years, 9 years, and 12 years following treatment and may represent second primary malignancies. Five patients (2 1%) underwent attempted surgical salvage after pelvic recurrence: radical hysterectomy ( 1 patient), anterior exenteration (3 patients), and laparotomy only (1 patient). Two patients (8%) were successfully salvaged and are alive and diseasefree 3 years and 13 years postoperatively. Thirty-one patients treated with irradiation and surgery developed a pelvic recurrence, and six (19%) underwent attempted surgical salvage procedures: total exenteration (3 patients) and laparotomy alone (3 patients). The salvage procedure was successful in one patient, who is alive and diseasefree 12 years following total exenteration. The results of the multivariate analysis for pelvic control are shown in Table 2. None of the parameters listed significantly affected the likelihood of pelvic control. In particular, the addition of surgery to irradiation did not reduce the risk of pelvic recurrence.
Table 2. Control of disease in the pelvis: multivariate analysis Significance level
Parameter
Rank order 1 2 3 4
Age Treatment Regression Size
0.1427 0.2407 0.5180 0.466 1
: 7
Hematocrit Stage Pathology
0.5808 0.7431 0.9892
Distant metastasis The incidence of distant metastasis (extrapelvic recurrence) for both treatment groups is depicted in Figure 2. At 5 years the probability of distant metastasis was 17% for the group treated with irradiation alone and 16% for the group treated with irradiation and surgery. Three patients treated with irradiation alone underwent attempted surgical salvage for a solitary lung metastasis noted at 1, 2, and 4 years following initial treatment. All three patients are alive and disease-free at 11, 13.5, and 19.5 years following the salvage procedure. One patient treated with irradiation and surgery underwent attempted salvage with surgery and irradiation for a solitary lung metastasis presenting 1 year after initial treatment. She is alive and disease-free 18 years following salvage treatment. The results of the multivariate analysis for distant metastasis are shown in Table 3. None of the parameters examined significantly influenced the risk of distant metastasis.
Cause-specificsurvival Cause-specific survival rates (censoring non-cancer-related deaths) for the two treatment groups are shown in Figure 3; there is no significant difference between the survival curves. The 5-year cause-specific survival rates were 62% for irradiation alone and 55% for irradiation
1oou
602
20%
1
RT
Ii
Alon*
-
RT . surpery
i8U
I
17x
,’
OUi( 0
Fig. 1. Control of disease in the pelvis (product-limit method). Date of pelvic recurrence was the end-point of the analysis; patients who never had pelvic recurrence or who were salvaged were censored at the time of last follow-up or the time of death.
1
2
3
4 TIME
5 6 (Yeare)
7
8
9
10
Fig. 2. Incidence of distant metastasis (product-limit method). Date of manifestation of distant metastasis was the end point of the analysis; patients who never had distant metastasis were censored at the time of last follow-up or the time of death. The resulting probabilities of freedom from distant metastasis were subtracted from 1 to calculate the probabilities of having distant metastasis.
902
I. J. Radiation Oncology 0 Biology 0 Physics
Table 3. Probability of distant metastasis: multivariate analysis
September 1991, Volume 21, Number 4 Table 4. Cause-specific
Rank order
Parameter
Significance level
1 2 3 4 5 6 7 8
Stage Treatment Size Regression Age Local control Hematocrit Pathology
0.1635 0.2774 0.3255 0.3358 0.3737 0.6276 0.868 1 0.9562
plus surgery. The results of the multivariate analysis for cause-specific survival are shown in Table 4. The only parameter that influenced cause-specific survival was tumor diameter; tumors 6-7 cm in diameter were associated with a significantly higher cause-specific survival rate than those 28 cm in diameter. Treatment did not exert a significant impact on this one point.
Absolute survival The absolute survival rates are depicted in Figure 4; the 5-year absolute survival rates were 54% for patients treated with irradiation alone and 52% for those treated with irradiation plus surgery. The results of the multivariate analysis for absolute survival are shown in Table 5. Pathology was of borderline significance, in favor of squamous cell carcinoma.
Complications Severe complications were defined as those resulting in hospitalization or necessitating an operation. Four patients (5%) treated with irradiation alone developed the following severe complications: rectosigmoiditis necessitating colostomy (2 patients), small bowel obstruction requiring
survival: multivariate analysis
Rank order
Parameter
Significance level
1
Size Regression Age Pathology Hematocrit Treatment Stage
0.0354 0.2094 0.2680 0.3554 0.650 1 0.6323 0.9326
2 3 4 5 6 7
surgical intervention (1 patient), and vesicaovaginal fistulae requiring an ileal conduit (1 patient). Severe complications were noted in 12 patients ( 16%) treated with irradiation and surgery: pyelonephritis necessitating hospitalization (1 patient), ureteral obstruction resulting in repeated pyelonephritis and renal insufficiency (1 patient), rectovaginal fistula necessitating colostomy (1 patient), abdominal wound abscess requiring hospitalization for incision and drainage (2 patients), small bowel obstruction necessitating hospitalization for conservative management (1 patient) or resulting in surgical intervention (2 patients), pyometra requiring hospitalization (1 patient) or surgical intervention (1 patient), pelvic abscess necessitating surgical exploration (1 patient), and ureterovaginal fistula resulting in ureteral obstruction and necessitating nephrectomy (1 patient). The incidence of severe complications (product-limit method) for patients treated with irradiation alone was 7% at 3.5 years and stable thereafter. For patients treated with irradiation and surgery, the incidence of severe complications was 15% at 1 year, 18% at 6 years, and stable thereafter. The difference between these curves is significant at the p = 0.027 level. DISCUSSION
In 1969, Durrance et al. (1) reported that, in patients with Stage II cervix cancer, most recurrences developed
60% 66%
RT *lone
0
1
2
3
4 TIME
-
RT . l”,p.ry
6 6 (mar*)
7
6
9
10
Fig. 3. Cause-specific survival (product-limit method). Date of death with cancer present was the end point of the analysis; patients not known to be dead or who did not have cancer present at death were censored at the time of last follow-up or time of death.
Fig. 4. Absolute survival (product-limit method). Death due to any cause was the end point of the analysis; patients not known to be dead were censored at the date of last follow-up.
Carcinoma of the cervix 0 W. M. MFSDENIWLL e? al.
Table 5. Absolute survival: multivariate
analysis
Rank order
Parameter
Significance level
1 2 3 4 5 6 7
Pathology Size Regression Treatment Stage Hematocrit Age
0.077 1 0.1239 0.3624 0.5797 0.7086 0.7830 0.8498
in patients with bulky or barrel-shaped endocervical carcinomas. Fletcher, in the 1970 Janeway Lecture (4), noted that the combination of irradiation and extrafascial hysterectomy resulted in a central pelvic recurrence rate of only 2% for patients with bulky or barrel-shaped Stage IB-II cervical carcinoma. Nelson et al. (12) subsequently reported that irradiation and extrafascial hysterectomy markedly reduced the risk of central pelvic recurrence and resulted in survival rates comparable to overall survival rates for similarly staged lesions. After those reports, various authors have confirmed ( 10, 13) or refuted (2, 15, 20) the benefit of irradiation plus extrafascial hysterectomy over irradiation alone. There are several possible explanations for these conflicting reports. First, “bulky or barrel-shaped” is a subjective term and does not precisely define the size of the lesion. For example, from the University of Kentucky, Gallion et al. (5) defined bulky or barrel-shaped Stage IB cervix cancer as >5 cm diameter, whereas Maruyama et al. (lo), from the same institution, included 27 patients with tumors 3 cm in diameter in a series of 80 patients
903
with “bulky or barrel-shaped” Stage IB cervical carcinomas. Second, irradiation plus surgery series often exclude patients who undergo operation but not resection because of advanced disease, thus biasing a comparison of the data against patients treated with irradiation alone. Third, some series contain a mixture of patients who are clinically and surgically staged. For example, a patient may be taken to surgery for a radical hysterectomy and found to have multiple positive nodes. If the resection is aborted and the patient treated with irradiation alone, inclusion of the patient in the “irradiation alone” group would create a bias against that treatment group, compared with patients treated with irradiation and surgery who are clinically staged. Fourth, some series exclude patients who were to undergo an extrafascial hysterectomy following irradiation, but in whom a more extensive surgical procedure was found to be necessary to extirpate the tumor completely. All of these factors may render a meaningful comparison between irradiation alone and irradiation plus extrafascial hysterectomy nearly impossible. The data presented in this paper demonstrate that irradiation alone provides pelvic control and survival rates similar to those obtained with irradiation and surgery for Stage IB and II lesions 16 cm in diameter. Additionally, the incidence of severe treatment complications is lower with irradiation alone. Our conclusion is that the routine use of extrafascial hysterectomy for bulky Stage IB or IIAB cervix carcinoma is not indicated. Extrafascial hysterectomy may be indicated following irradiation for the exceptional case in which the patient is pregnant, the geometry of the radium application is poor, or the uterus is enlarged because of possible tumor involvement.
REFERENCES 1. Durrance, F. Y.; Fletcher, G. H.; Rutledge, F. N. Analysis of central recurrent disease in Stages I and II squamous cell carcinomas of the cervix or intact uterus. Am. J. Roentgenol. 106:831-838; 1969. 2. Eifel, P. J.; Morris, M.; Oswald, M. J.; Wharton, J. T.; Delclos, L. Adenocarcinoma of the uterine cervix: prognosis and patterns of failure in 367 cases. Cancer 65:2507-25 14; 1990. 3. Einhom, N.; Patek, E.; Sjoberg, B. Outcome of different treatment modalities in cervix carcinoma Stage IB and IIA: observations in a well-defined Swedish population. Cancer 55:949-955; 1985. 4. Fletcher, G. H. Cancer of the uterine cervix: Janeway Lecture, 1970. Am. J. Roentgenol. 111:225-242; 1971. 5. Gallion, H. H.; van Nagelk Donaldson, E. S.; Hanson, M. B.; Powell, D. E.; Maruyama, Y.; Yoneda, J. Combined radiation therapy and extrafascial hysterectomy in the treatment of Stage IB barrel-shaped cervical cancer. Cancer 56:262-265; 1985. 6. Kalbfleisch, J. D.; Prentice, R. L. The statistical analysis of failure time data. New York: Wiley; 1980. 7. Kaplan, E. L.; Meier, P. Nonparametric estimation from incomplete observations. J. Am. Stat. Assoc. 53:457-481; 1958.
8. Kottmeier, H. L. Classification and staging of malignant tumors in the female pelvis. Int. J. Gynecol. Obstet. 9: 172179; 1971. 9. Lawless, J. E. Statistical models and methods for lifetime data. New York: Wiley; 1982:420-422. 10. Maruyama, Y.; van Nagell, J. R.; Yoneda, J.; Donaldson, E.; Gallion, H.; Higgins, R.; Powell, D.; Kryscio, R.; Bemer, B. Dose-response and failure pattern for bulky- or barrelshaped Stage IB cervical cancer treated by combined photon irradiation and extrafascial hysterectomy. Cancer 63:7076; 1989. 11. Mendenhall, W. M.; Thar, T. L.; Bova, F. J.; Marcus, R. B.; Morgan, L. S.; Million, R. R. Prognostic and treatment factors affecting pelvic control of Stages IB and IIA-B carcinomas of the intact uterine cervix treated with radiation therapy alone. Cancer 53:2649-2654; 1984. 12. Nelson, A. J.; Fletcher, G. H.; Wharton, J. T. Indications for adjunctive conservative extrafascial hysterectomy in selected cases of carcinoma of the uterine cervix. Am. J. Roentgenol. 123:91-99; 1975. 13. Q’Quinn, A. G.; Fletcher, G. H.; Wharton, J. T. Guidelines for conservative hysterectomy after irradiation. Gynecol. Oncol. 9:68-79; 1980. 14. Parsons, J. T.; Thar, T. L.; Bova, F. T.; Million, R. R. An
904
I. J. Radiation Oncology 0 Biology 0 Physics
evaluation of split-course irradiation for pelvic malignancies. Int. J. Radiat. Oncol. Biol. Phys. 6: 17% 18 1; 1980. 15. Perez, C. A.; Kao, M. S. Radiation therapy alone or combined with surgery in treatment of barrel-shaped carcinoma of the uterine cervix (Stages IB, IIA, IIB). Int. J. Radiat. Oncol. Biol. Phys. 11:1903-1909; 1985. 16. Rettenmaier, M. A.; Casanova, D. M.; Micha, J. P.; Moran, M. F.; Ramsanghani, N. S.; Syed, N. A.; Puthawala, A.; Disaia, P. J. Radical hysterectomy and tailored postoperative radiation therapy in the management of bulky-stage IB cervical cancer. Cancer 6312220-2223; 1989. 17. Rotman, M.; John, M. J.; Moon, S. H.; Choi, K. N.; Stowe, S. M.; Abitol, A.; Herskovic, T.; Sall, S. Limitations of ad-
September 1991, Volume 21, Number 4 junctive surgery in carcinoma of the cervix. Int. J. Radiat. Oncol. Biol. Phys. 5:317-332; 1979. 18. SAS Institute Inc. SAS technical report P-179, additional SAS/STAT procedures, release 6.03. Guy, NC: SAS Institute Inc.; 1988:49-89. 19. Thar, T. L.; Million, R. R.; Daly, J. W. Radiation treatment of carcinoma of the cervix. Semin. Oncol. 9:299-3 11; 1982. 20. Weems, D. H.; Mendenhall, W. M.; Bova, F. J.; Marcus, R. B.; Morgan, L. S.; Million, R. R. Carcinoma of the intact uterine cervix, Stages IB-IIA-B, ~6 cm in diameter: irradiation alone vs. preoperative irradiation and surgery. Int. J. Radiat. Oncol. Biol. Phys. 11:1911-1914; 1985.