E is treated by either radiation therapy or radical surgery.

Transcription

1 Radical Hysterectomy Versus Radiation Therapy for Stage B Squamous Cell Cancer of the Cervix Michael P. Hopkins, MD, and George W. Morley, MD Three hundred forty-five patients with Stage B squamous cell carcinoma of the cervix were treated at the University of Michigan Medical Center from 1970 to The overall cumulative 5-year survival rate was 89% and the mean age was 44.6 years. n 213 patients undergoing radical hysterectomy the cumulative 5-year survival rate was 92%; 14 patients were explored for radical hysterectomy that was not performed due to high risk features and their survival rate was 50%. Ninetyseven patients underwent radiation therapy as initial treatment and had a 5-year survival rate of 86%. There was no significant difference when radiation therapy was compared with radical hysterectomy (P = 0.098). The survival rates for lesions 3 cm or smaller were 94% for radical hysterectomy and 88% for radiation therapy. When the lesion was larger than 3 cm, the survival rates were 82% with radical surgery and 73% with radiation therapy. Metastatic disease to lymph nodes was present in 26 of the 213 patients undergoing radical hysterectomy. When 1 to 3 nodes were involved 16 of 19 patients survived and when 4 to 10 nodes were involved 3 of 7 patients survived. The addition of radiation therapy did not influence survival. Complications were similar in both treatment groups. Fistulas occurred in 4 of 213 patients undergoing radical hysterectomy and 1 of 111 undergoing radiation. Second surgery for a complication was required in 6 of 213 patients undergoing radical hysterectomy and 7 of 111 undergoing radiation. Survival and complication rates in early stage squamous cell carcinoma of the cervix are equal with either radical surgery or radiation therapy. Cancer 68: ; ARLY STAGE squamous cell carcinoma of the cervix E is treated by either radiation therapy or radical surgery. Most authors report one form of treatment or the other and often combine the adenocarcinoma with the squamous cell carcinomas. -' There are few comparative studies analyzing, within the same institution, a large number of patients treated by either radical surgery or radiation An alternating series that treated patients with radiation therapy or radical surgery was previously reported from our institution. The 5-year survival was reported through 1970 and no difference in survival From the Department of Obstetrics and Gynecology. University of Michigan Medical Center, Ann Arbor, Michigan. Address for reprints: Michael P. Hopkins, MD, Department of Obstetrics and Gynecology. Northeastern Ohio Universities, College of Medicine, 400 Wabash Avenue-A.C C., Akron, OH Accepted for publication April was found between radiation therapy and radical surgery. A discussion of that paper included questions concerning the influence of the number of lymph nodes on survival, the extended number of years over which the study was performed, and, therefore, the changing radiotherapeutic technique. Additionally, patients with adenocarcinomas were included.' This study was undertaken to determine if any survival difference existed for those patients with squamous cell carcinoma of the cervix treated by either radical surgery or radiation therapy. Material and Methods The clinical records, pathology reports, and gynecologic tumor conference notes for patients at the University of Michigan Medical Center with a diagnosis of carcinoma of the cervix for 1970 through 1985 were reviewed. All patients identified as having Stage B squamous cell car- 272

2 No. 2 RADCAL HYSTERECTOMY VERSUS RADATON THERAPY - Hopkins and Mor1e.v 273 cinoma of the cervix are reported here. Microinvasive squamous cell carcinoma (Stage A), as previously reported from our institution, was defined as 3 mm or less of invasion without angiolymphatic invasion." All patients with invasion greater than 3 mm or angiolymphatic invasion were considered to have Stage B disease. The original hematoxylin and eosin slides were reviewed for confirmation of the diagnosis. When the original slides were unavailable, the original pathology report and the gynecologic tumor conference notes were used. The information for the clinical presentation, medical history, treatment technique, and size of the cervical lesion was recorded. All patients were staged according to the nternational Federation of Gynecology and Obstetrics (FGO) guidelines. Radiation therapy during this period was given in a standardized fashion using the split-field technique. Therapy was planned to give 1980 cgy (180 cgy/d) to the entire pelvis followed by two intracavitary applications administering a total of 7000 to 8000 mg/h. This was then followed by external radiation with a midline block to bring the total dose to the lateral pelvic walls to 5000 cgy and the combined estimated dose at Point A to 8000 to 9000 cgy. Central boosts were not routinely given for larger lesions. External therapy was given by cobalt-60 with anterior-posterior ports or by linear accelerator (Cli- STAGE B SOUAMOUS CELL CERVX ALL PATENTS (Pz.007) RH (N=213) RT (N=l 11) FG. 1. Survival for all patients with Stage B disease comparing radical hysterectomy with all patients undergoing radiation therapy. TABLE. Why the Planned Radical Hysterectomy was not Performed at the Time of Surgical Exploration and the Patient Outcome Reason No. Outcome Bilateral positive nodes 6 5 DOD Unilateral positive nodes 2 2 NED Diverticulitis/celllulitis 2 NED, 1 DOD Parametrial involvement 3 3 NED Tumor penetration through cervix 1 1 DOD NED not evidence of disease: DOD: dead of disease. nac 18, 10-mV photons) with anterior-posterior and lateral ports. Brachytherapy was changed from a radium source to cesium in 1979 and external therapy was changed from cobalt to linear accelerator in 198. Statistical analysis was performed by the Department of Biostatistics, School of Public Health, University of Michigan. Survival was plotted with the life table method described by Kaplan and Meier." The Savage (Mantel- Cox) method was used to test differences in Results From 1970 to 1985, 345 patients were identified with Stage B disease. The overall cumulative 2-year survival rate was 95%, the 5-year survival rate was 89%, and the 10-year survival rate was 86%. The mean age of the patients was 44.6 years. The patients were treated as follows: radical hysterectomy (2 13 patients), radiation therapy ( 1 1 patients), or extrafascial hysterectomy plus radiation therapy (2 1 patients). The 2 1 patients treated by standard hysterectomy and radiation therapy are included in a larger group of patients previously reported and are not further analyzed in this report. l4 The 2 13 patients who underwent radical hysterectomy had a cumulative 5-year survival rate of 92%. One hundred eleven patients treated with primary radiation therapy had an 82% cumulative 5-year survival rate (P = 0.007) (Fig. 1). However, there were 14 patients who were initially explored for a radical hysterectomy that was not performed due to high-risk features. The patients are listed in Table 1 and this group had a 50% cumulative 5-year survival rate. When this group is eliminated from the radiation therapy group, the 97 patients who underwent radiation therapy as the initial planned treatment had an 86% cumulative 5-year survival rate; this does not reach significance compared with those treated by radical hysterectomy (P = 0.098) (Fig. 2). A comparison of radical surgery versus radiation therapy (excluding the 14 patients previously mentioned) was performed according to lesion size. Patients with a tumor size of 3 cm or less had a 94% cumulative 5-year survival rate when treated by radical surgery versus 88% when treated by radiation therapy (Fig.

3 2 74 CANCER July Vol. 68 C z 5 a =l v) c W z 0 L W a 1.o Y' 5 Yr 10 yr STAGE E SQUAMOUS CELL CERVX EXCLUDNG RHND (P1.098) RH (N=213) RT (N=97) FG. 2. Survival for all patients with Stage B disease treated with either radical hysterectomy or radiation therapy as the planned treatment (RHND-radical hysterectomy not done). 3). Patients with lesions larger than 3 cm had an 82% cumulative 5-year survival rate when treated by radical surgery and a 73% cumulative survival rate when treated by radiation therapy (Fig. 4). Patients who underwent radical hysterectomy with or without bilateral salpingooophorectomy were compared. There was a 97% survival rate for those patients where ovaries were conserved and an 88% survival rate where ovaries were removed. None of the removed ovaries contained metastatic disease. Patients undergoing radical hysterectomy had pelvic, common iliac, and low paraaortic node dissection. The mean number of lymph nodes removed was 24 (median, 24; range, 8 to 60). Radical hysterectomy was performed on 26 patients who had positive lymph nodes (Table 2). Nineteen patients had 3 or fewer positive nodes and 16 (84%) survived and were disease-free. Seven patients who underwent radical hysterectomy had more than three positive lymph nodes and three (43%) survived and were disease-free. The addition of radiation therapy in 14 of these patients did not improve survival (Fig. 5). The use of adjuvant radiation was analyzed according to the number of involved lymph nodes. When three or fewer lymph nodes were involved, nine of ten patients who received radiation and seven of nine who did not receive radiation remained disease-free. When more than three nodes were o.80 RH(N-173) RT (N=84).80 RH (N-40) RT (N-12) yr yr 5 yr 10 yr STAGE 18 SQUAMOUS CELL CERVX SZE 4cm, EXCLUDNG RHND (P=.03) FG. 3. Survival for all patients whose tumor size was 3 crn or less according to type of therapy RH: Radical hysterectomy; RT: radiation therapy; RHND: radical hysterectomy not done. STAGE 18 SaUAYOUS CELL CERVX SZE >3an, EXCLUMO RHUD (P8.87) FG. 4. Survival for patients whose tumor size was 3 cm or greater according to treatment. RH: radical hysterectomy; RT: radiation therapy; RHND: radical hysterectomy not done.

4 No. 2 RADCAL HYSTERECTOMY VERSUS RADATON THERAPY - Hopkins and Morley 275 TABLE 2. Number of Positive Lymph Nodes at Radical Hysterectomy and Outcome per Total Number of Patients With That Particular Number of Positive Nodes No. of positive nodes No. of patients DODJtotal Total 1/26 DOD: dead of disease. involved, one of four who received radiation and two of three who did not receive radiation remained disease-free. The following complications occurred in the 2 13 patients undergoing radical hysterectomy: wound infection ( 1 1 patients; 5%), pyelonephritis/urinary tract infection (8 patients), cuff cellulitis (4 patients), lymphedema (2 patients), bladder atony (9 patients), lymphocyst (3 patients), pneumonia (4 patients), ureteral obstruction ( l patient), vesicovaginal fistula (2 patients), and ureterovaginal fistula (2 patients). A ureterovaginal fistula in one patient healed with expectant management. The remaining four patients with a urologic complication required surgical correction. The overall fistula rate was 4 of 2 3 patients ( 1.8%). Thirty-one patients received radiation therapy after radical hysterectomy and 3 experienced complications including the following: lymphocyst ( 1 patient), vesicovaginal fistula (1 patient), and leg edema (1 patient). Complications for the 11 1 patients undergoing radiation therapy included bowel obstruction (7 patients), enteritis/proctitis (6 patients), lymphedema (2 patients), cystitis ( 1 patient), pyometra (1 patient), and rectovaginal fistula (1 patient). Second surgery for complications related to therapy was required for 6 of 213 patients (3%) after radical hysterectomy and 7 of patients (7%) after radiation therapy (no significant difference). Surgery after radical hysterectomy included drainage of the lymphocyst (two patients), correction of the urinary fistula (three patients), and correction of the urinary obstruction (one patient). Surgery after radiation therapy included five cases of bowel obstruction, one cecal augmentation for radiation-induced cystitis, and one colostomy for a rectovaginal fistula. There was no mortality related to radical hysterectomy and there was no mortality related to radiation therapy during the treatment phase of radiation therapy. One patient who had undergone radiation therapy experienced a bowel obstruction with perforation and peritonitis 4 years after treatment and died of this complication. Recurrences developed in 25 patients who were available for evaluation. The sites of recurrence in 14 patients treated by radical hysterectomy included distant metastases (6 patients), pelvic side wall involvement (4 patients), and vaginal cuff (4 patients). Eleven patients treated by radiation therapy experienced a recurrence; in 10 patients this was distant disease. One patient, who had a central recurrence and underwent pelvic exenteration, is alive and disease-free 10 years later. n these 25 patients, the sites of distant metastatic disease included the lung (10 patients), bone (spine) (7 patients), brain (3 patients), and supraclavicular nodes (2 patients); 4 patients died of renal failure secondary to ureteral obstruction from recurrent tumor. Discussion When new cases of Stage B squamous cell carcinoma of the cervix are diagnosed, the treatment options include either radical hysterectomy or radiation therapy. Many series combine the adenocarcinoma with squamous cell carcinoma and report only one treatment method, either radiation therapy or radical surgery. The reports that compare radiation therapy with radical surgery in the same institution are limited. Newton, in a series from 1956 to 1966, reported no difference in survival for o.a w RH (N-12) RH + RT (N-14) STAGE B SQUAMOUS CELL CERVX LYMPH NODES WSrrtvE, RADCAL HYSTERECTOMY (PS75) FG. 5. Survival when radiation therapy is added to radical hysterectomy when lymph nodes are positive.

5 276 CANCER July Vol. 68 patients treated by surgery or radiation.' He reported an 8 1 W survival rate for radical hysterectomy compared with 74% for radiation therapy. Morley and Seski reported the 5-year survival experience from the University of Michigan through 1970 and the results from our latest 15-year time period are similar. n the previously reported alternating series from the University of Michigan, no statistical difference was found between the 9 1.3% survival rate for radical hysterectomy compared with the 87.3% survival rate for radiation therapy.8 That survival also excluded patients who were explored for radical hysterectomy but who did not have it performed. The survival rates for both treatment techniques previously reported from our institution are remarkably similar to those reported here. The overall survival rate during the previous time period was 89% and this is identical to our report. The presence of metastatic disease to lymph nodes at the time of radical hysterectomy alters the prognosis. The beneficial effect of radiation therapy after lymph node dissection is unclear. Radiation therapy added for positive lymph nodes in this report did not improve survival. This is similar to the Society of Gynecologic Oncologists panel report that concluded that there was no measurable benefit when radiation therapy was added to radical hysterectomy for patients with three or fewer positive lymph nodes." Jobson et al. reported a 52% 5-year survival rate when radiation therapy was added to radical hysterectomy and this is decreased compared with the 80% 5-year survival rate reported here.16 This is likely due to the selection of patients rather than an effect of radiation therapy. Larson et af. concluded that adjuvant radiation may reduce pelvic recurrence and improve outcome in patients with node metastases." The risk of recurrence appears to escalate when more than three nodes are involved. This may represent the fact that systemic disease is present when this many lymph nodes are involved. ntraoperative findings will occasionally dictate that the radical surgery not be completed. Although this will hopefully be a rare occurrence, there were 14 of 227 patients (6%) who had their planned radical surgery aborted. When these 14 patients are evaluated, the overall survival is reasonable. However, those who survive have more extensive cervical disease present or features present placing them at a high risk for surgical complications. The survival was poor in three of eight patients (37%) when the surgery was aborted for node metastases. n a large series reported by Montana et al., the 5-year survival rate for 197 patients treated with radiation therapy was 83%.5 This is similar to our results. Montana et af. used the split-field technique and provided the same approximate dose to Point A as reported in our patients. n our report there was only 1 of 11 1 central recurrence, suggesting that the split-field technique with two brachy- therapy insertions provides excellent local control. t would seem that the larger lesions would be treated better with radiation therapy to avoid a difficult surgical procedure or a margin involved with cancer. n our experience, the larger lesions did not do any better with radiation therapy when compared with radical surgery. Homesley et af. reported a much higher recurrence rate when lesions treated by radiation therapy exceeded 4 cm in size. n their report, the survival rate decreased to 67% for lesions more than 4 cm in size." All of their patients were treated with radiation to a higher dose to Point A than those reported here. There has been a decrease in the fistula complications related to radical surgery. The complications related to fistula or obstruction in five patients (2.3%) during the current time period are half the 4.8% previously reported.8 n this report, the number of patients experiencing significant complications requiring surgery after radical hysterectomy (6 of 2 13) was approximately half that for radiation therapy (7 of 11 1). The radiation complication rates are similar to the severe complications encountered in 7 of 197 patients treated by Montana et af.' There was a low complication rate in our patients treated by radical hysterectomy and postoperative radiation therapy. This is in agreement with the report by Jobson et al. where 4 of 30 patients undergoing adjunct radiation therapy after radical hysterectomy had significant complications." Barter et af., however, reported a 30% complication rate, which is much higher." n patients with Stage B invasive squamous cell cancer of the cervix, excellent survival results can be obtained with either radiation therapy or radical hysterectomy. The fistula and major complication rates are low with either therapeutic technique. REFERENCES 1. Burghardt E, Pickel H, Haas J. Prognostic factors and operative treatment of stages B to B cervical cancer. Am J Obstei Gynecol 1987; Burke TW, Hoskins WJ. Heller PB, Shen MC, Weiser EB, Park RB. Clinical patterns of tumor recurrence after radical hysterectomy in stage B cervical carcinoma. Obstct Gvnecol 1987; 69: Creasman WT. Soper JT, Clarke-Pearson D. Radical hysterectomy as therapy for early carcinoma of the cervix. Am J Obsier Gynecol 1986; 55: Falk V, Lundgren N. Quarfordt L, Arstrom K. Primary surgical treatment of carcinoma stage ofthe uterine cervix. Aciu Obstei Gynecol Scund 1982; 61: Montana GS, Fowler WC. Varia MA, Walton LA, Mack Y. Analysis of results of radiation therapy for stage B carcinoma of the cervix. Cancer 1987: 6O: Sall S, Pineda AA, Calanog A, Heller P, Greenberg H. Surgical treatment of stages 9 and A invasive carcinoma of the cervix by radical abdominal hysterectomy. Am J Obsiet Gynecol 1979; 135: Underwood PB Jr, Wilson WC, Kreutner A, Miller MC 111, Murphy E. Radical hysterectomy: A critical review of twenty-two years' experience. Am J Obsiet Gynecol 1979; Morley GW, Seski JC. Radical pelvic surgery versus radiation ther-

6 No. 2 RADCAL HYSTERECTOMY VERSUS RADATON THERAPY * Hopkins and Morley 277 apy for stage carcinoma of the cervix (exclusive of microinvasion). Am J Obstet Gvnecol 1976: 126: Newton M. Radical hysterectomy or radiotherapy for Stage cervical cancer: A prospective comparison with 5 and 10 year follow-up. Am J Obstet Gynecol 1975; 123: Seski JC. Abell MR, Morley GW. Microinvasive squamous carcinoma of the cervix: Definition. histologic analysis, late results of treatment. Obstet Gynecol 1977; 50: Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J.4m Stat Assoc 1958; 53: Cox DR. Regression models and life tables. J R Stat Soc 1972; 34: Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cunccr Cllemother Rep 1966; 50: Hopkins MP. Peters WA, Andersen W. Morley GW. lnvasive cervical cancer treated initially by standard hysterectomy. G.vnecol Oncol 1989: 36: Morrow PC. Panel report: s pelvic radiation beneficial in the postoperative management of stage B squamous cell carcinoma of the cervix with pelvic node metastasis treated by radical hysterectomy and pelvic lymphadenectomy. Gynecol Oncol 1980; 10: Jobson VW, Girtanner RE, Averette HE. Therapy and survival of early invasive carcinoma of the cervix uteri with metastases to the pelvic nodes. Surg Gynecol Obstei 1980; 151 : Larson DM, Stringer CA, Copeland LJ, Gershenson DM, Malone JM. Rutledge FN. Stage B cervical carcinoma treated with radical hysterectomy and pelvic lymphadenectomy: Role of adjuvant radiotherapy. Obstet Gynecol 1987; 69: Homesley HD, Raen M, Blake DD el a/. Relationship of lesion size to survival in patients with stage B squamous cell carcinoma of the cervix uteri treated by radiation therapy. Surg Gynecol Obstet 1980; 150: Barter JF, Soong SJ, Shingleton HM, Hatch KD. Orr JW Jr. Complications of combined radical hysterectomy-postoperative radiation therapy in women with early stage cervical cancer. Gynecol Oncol 1989: