Preoperative irradiation for cancer of the rectum with extrarectal fixation

In,. 1. Rodiorron Oncology Btol Ph?s Vol. 21, pp. 1127-l 132 Printed I” the U.S.A. All rights reserved.

0360.3016/91 $3.00 + .W 0 1991 Pergamon Press plc

CopyrIght

??Clinical Original Contribution

PREOPERATIVE

ROBERT L. TOBIN, M.D.

IRRADIATION FOR CANCER OF THE RECTUM WITH EXTRARECTAL FIXATION ,* MOHAMMED MOHIUDDIN,

M.D.

* AND GERALD MARKS,

M.D.

t

Thomas Jefferson University Hospital, Philadelphia, PA 19107-5097 Tumor mobility of rectal cancer is well known to have prognostic significance for operative resection, local recurrence, and survival. Between 19761988, 220 patients have been consecutively treated with high dose preoperative radiotherapy at Thomas Jefferson University Hospital. During this time period, 134 patients were clinically determined by the surgeon and radiotherapist to have extrarectal tumor fixation as a primary indication for preoperative irradiation and are the subject of this review. The patient population can be further divided into two subgroups which include 49 patients with clinical tethering/partial fixation, and 85 patients with completely fixed tumors. Patients were treated with 4-field pelvic radiotherapy to 45 Gy in 25 fractions. Depending on location and degree of fixation, a localized boost dose was frequently delivered to the tumor for an additional 4.8-9.6 Gy using opposed high-energy lateral fields. Surgical resection was instituted 4-6 weeks post completion of radiotherapy. Significant tumor regression permitted sphincter preserving surgery in 105/ 134 (78%) of these patients. With a median follow-up of 37 months, the overall 5-year actuarial survival for our postradiation Stage A/B1 patients was 92% (n = 28), and 63% for Stage B2/C patients (n = 91). Local recurrence occurred in only 7% of the Stage A/B1 patients, and 18% in the Stage B2/C patients. Analyzed by pre-treatment clinical evaluation, the 5-year actuarial survival of these patients was 68% and 60% in the clinically tethered and fixed tumor subgroups, respectively @ = .51). Pelvic control was demonstrated in 86% of the patients in the tethered subgroup, and in 80% of the preoperative fixed patients. The combined treatment was well tolerated, with complications limited to 6% of the patient population. We conclude that preoperative radiotherapy for rectal carcinomas with clinical extrarectal fixation provides optimal presurgical cytoreduction and excellent survival. Furthermore, sphincter function can he maintained in a majority of patients with appropriate attention to patient selection. Rectal carcinoma,

Preoperative

irradiation,

Fixed rectal carcinoma.

Approximately 45,000 cases of adenocarcinoma of the rectum are diagnosed each year in the United States with an annual mortality rate of 7600 in 1990 (2). Historically, the majority of these patients have been treated with an abdomino-perineal resection and permanent colostomy. Despite the radical nature of this surgery, local recurrence rates average 30-45% in large series (3, 7) with little change in survival over the past 20 years. During this time period, a number of prognostic factors have been identified through large randomized trials. In 1984, the Working Group of the Medical Research Council (3) reported on the clinical pathological features of 824 rectal cancer patients treated in a prospective randomized trial with surgery alone or surgery preceded by low-dose

preoperative radiotherapy. Variables associated with lower survival rates were tumor fixation, location in the low rectum (< 8 cm), advanced circumferential configuration, high grade, and advancing Dukes’ stage. In this study of operable rectal tumors, both the overall and disease-free survival rates were found to be strongly related to features determined at the pretreatment assessment. Of these, the authors noted that the single most important factor was the mobility of the primary cancer. Recognizing the importance of preoperative clinical tumor assessment and treatment stratification, we initialized a prospective study of high dose (1 4500 cGy) preoperative radiotherapy for patients with unfavorable rectal carcinomas in 1976. Through 1988, 134 patients with clinically tethered or fixed rectal tumors were consecutively treated at the Bodine Cancer Center. Details of treatment, compli-

Presented at the 32nd Meeting of ASTRO, Miami, FL, 15-19 October 1990. * Department of Radiation Oncology and Nuclear Medicine. t Department of Surgery, Division of Colorectal Surgery and the Comprehensive Rectal Cancer Center. Reprint requests to: Robert L. Tobin, M.D., Thomas Jefferson University Hospital, Bodine Center, 111 S. 1 lth St., Philadel-

phia, PA 19107. Acknowledgements-The authors wish to thank Ms. Barbara Moscariello and Susan Pirolli for their invaluable assistance in preparation and typing of the text, and superb skills in the preparation of the figures and graphs used within. Dr. Tobin is a 1990 American Cancer Society Clinical Fellow. Accepted for publication 29 March 1991.

INTRODUCTION

1127

I. J. Radiation Oncology 0 Biology 0 Physics

1128

Table 1. Patient characteristics

(n = 1 34)

No. Median age

cations, and patterns stage are presented.

(39) (37) (24)

49 85

(37) (63)

METHODS

assessment

Number 5

Total

(%)

52 50 32

of failure by clinical

21,

Table 2. Type of surgery

62 29-8 1 73161

(range) Sex (M/F) Tumor level O-3 cm 3-6 cm 615 cm Tethered Partial fixation Fixed

October 1991, Volume

APR LAR CATS TATA FTLE Other

Tethered

29 26 39 31 5 4

5 10 15 16 3 -

Fixed 24 16 24 15 2 4

Sphincter preservation = 105/134 (78%); Miles resection = 29/134 (22%). APR = Abdomino-perineal resection; LAR = Low anterior resection; CATS = Combined abdominal trans-sacral resection; TATA = Transanal transabdominal resection; FILE = Full thickness local excision.

and

AND MATERIALS

Our goal was to improve the resectability, local control, and ultimate survival through a combined modality approach using high dose (45-55 Gy) preoperative radiotherapy followed in 4 to 6 weeks by surgical resection. All of our patients had significant regression of their tumors to allow for potentially curative resection. To the credit of the surgical efforts, three-quarters of these patients were able to undergo sphincter preserving surgery. Pretreatment evaluation included a thorough physical examination, proctosigmoidoscopy and biopsy, colonoscopy, ? barium enema. Radiographic studies included a chest X ray, CT scan of the abdomen and pelvis, and selective use of endorectal sonography for distal rectal cancers. Patient characteristics and staging Table 1 lists the characteristics of the treated population by age, sex, location, and extent of extrarectal tumor fixation. The median age of the treated population was 62 years (range 29-81). The vast majority of the tumors were located in the low rectum (86% between O-6 cm from the ano-rectal junction). All 134 patients were felt by clinical and radiographic methods to have full thickness tumor penetration through the bowel wall and were further characterized into categories of “partial” or “complete” fixation. The term partial fixation includes tumors with mobility in at least one direction with the sensation of tethering secondary to extra-rectal spread. Fixed tumors were immobile in all directions on digital exam with firm adherence to adjacent structures. Partial fixation or “tethering” was present in 49 patients (37%) and complete fixation noted in 85 (63%) of the cases. Postoperatively, patients were staged according to Astler-Coller staging criteria (1). Treatment technique h-radiation. After completion of the clinical evaluation and metastatic work-up, patients are treated with high dose

preoperative radiotherapy. The treatment portals used at TJUH include a four-field pelvis technique with all fields treated daily. Patients were treated on a 45 MV Betatron Linear Accelerator up to 1987, and have subsequently been treated on a 25 MV high energy linear accelerator.* Treatment volumes included shaped alloys to encompass the whole pelvis with a 1.5 cm margin on the pelvis inlet to cover iliac nodal lymphatics. The superior border was set at the L5/Sl junction, and the inferior border generally included the bottom of the ischial tuberosities. The field included the anal canal for tumors in the lower (below 7 cm) rectum. Barium and anal markers were used during simulation. The lateral fields included the external iliac nodal chain superiorly and the entire sacrum with margin posteriorly. Patients typically received 4500 cGy to the pelvis at 180-250 cGy fractions. Standard fractionation of 180 cGy was used in the majority of patients. In early 1986, a dose escalation of 5000-5500 cGy was employed using opposed high energy lateral fields centered on the tumor with a 2 cm margin. The boost field was typically 10 cm X 8 cm and also covered the presacrum, which, from our experience, is at high risk for tumor recurrence. Surgery. At 4 weeks postradiotherapy, the patient was reassessed by the surgeon to determine operability and best operative technique. Surgical resection was scheduled 4-6 weeks after the completion of radiotherapy (median 5.6 weeks) to allow maximal tumor regression. Despite the high incidence of unfavorable tumors by location and fixation, all patients underwent potentially curative surgical resection. The majority of patients, 105/136 or 78%, were able to have sphincter preserving surgery. The techniques used at TJUH have been described in previous publications (11, 14). Table 2 lists the patients by operative procedure and by pre-existing assessment. Miles resection (abdomino-perineal) was necessitated in only 22% (29/134) of our patients postirradiation. * Phillips SL-25, Crawley,

England.

Preoperative irradiation for rectal cancer 0 R. L. TOBINet al. CARCINOMA

OF THE RECTUM

TETHERED/FIXED

SURVIVAL

1129

Table 4. Patterns of failure by Astler-Coller

BY STAGE

100

‘Local + distant

Local

Post radiation Stage

No.

O,A,Bl

l/28

B2 C

5/52 3139

Stage

(%I

No.

(3) (10) (7)

l/28 6/52 5/39

Distant only

(%I

No.

(%I

(3) (12) (13)

- I28

(0) (17) (41)

9152 6139

Patterns of failure are listed by stage and are presented in Table 4. Local recurrence was noted in only 6% of the Stage A/B 1 patients and in 21% of the B 2/C patients.

Fig. 1. Actuarial survival by post irradiation Astler-Coller

Stage.

Follow-up and statistical methods. Postsurgery, all patients have been followed closely on a 3-month basis alternating visits with the radiation oncologist and the surgeon. Frequent endoscopic evaluation and serial laboratory studies have been performed. Follow-up radiographic studies were coordinated with the primary physicians. All patients have been followed for a median of 37 months (range 12-143 months). Actuarial survival curves were calculated for patients by clinical presentation and postirradiation stage. Stage D patients were included in the local recurrence and survival rate statistics in the “tethered” and “fixed” subgroups. Standard Kaplan Meier methods were used for survival and disease-free survival calculations. (8) The failure patterns listed are cumulative (i.e., not first site of failure). Local recurrence figures include all failures detected by clinical exam, radiographic studies, colonoscopy, or at surgical re-operation. RESULTS

“Tethered” vs. “fired” The 5-year actuarial survival by clinical presentation was 68% and 60% in “tethered” and “fixed” subgroups, respectively. Corresponding disease-free survival was 57% and 48%. The survival of the two groups is graphically presented in Figure 2. Post definitive radiation and surgery, no statistical differences in survival are noted between the two groups at 5 years (p = .51). Patterns of failure were also analyzed by degree of fixation. Local control was demonstrated in 86% of the “tethered’ ’ subgroup. Sites of failure included isolated local recurrence in 3 patients (6%), local plus distant metastasis in 4 patients (8%), and isolated distant metastasis in 11 patients (22%). For the 85 patients with “fixed” rectal tumors, local control was realized in 80%. Patterns of failure included isolated local recurrence in 6 patients (7%), local plus distant metastasis in 11 patients (13%), and isolated distant metastasis in 25 patients (29%). Complications In general, the combined treatment was well tolerated. Acute sequelae from the radiotherapy were limited to increased bowel and bladder frequency. This was easily con-

Post-irradiation stage Post-irradiation staging of the resected specimens revealed 21% to be Stage A/Bl, 39%, Stage B2/B3, 29%, Stage C, and 11% Stage D. Overall survival by stage is presented graphically in Figure 1 and listed in Table 3. The 5-year actuarial survival was 92% for Stage A/B 1, 70% for Stage B2 patients, and 51% for Stage C patients.

CANCER

OF THE RECTUM

PREOPERATIVE

RADIATION

TETHERED

THERAPY

VS FIXED

Table 3. Survival by Stage Post Radiation Stage O/A/B 1 B2lB3 CllC2lC3 D

No. of pts. No.

(%)

5 year DFS %

5 year O.S. %

28 52 39 15

(21) (39) (29) (11)

92 64 34 0

92 70 51 17

20

1

0-l 0

I

12

I

24

I

36

I

46

I

r

60

MONTHS

Fig. 2. Actuarial survival by pre-treatment tumor mobility.

1130

I. J. Radiation

Oncology

0 Biology 0 Physics

trolled with medications and diet adjustments. Treatment complications were seen in only 8 of 134 patients (6%). Minor complications included two cases of delayed wound healing and two patients who developed pelvic abscesses, which responded to appropriate antibiotics and drainage. Complications requiring surgery included two bowel obstructions with operative lysis of adhesions at 6 and 14 months, and two anastomotic strictures that did not respond to conservative dilation and subsequently necessitated a colostomy. No treatment mortality was observed.

DISCUSSION High dose preoperative radiotherapy has only recently been used in the adjunctive treatment of rectal cancer. Steadily, a number of small retrospective clinical trials have recently been reported demonstrating high respectability rates and low incidence of pelvic recurrences or complications using this format (5, 6, 10, 12, 13, 15, 16, 18, 20). Based on our own initial favorable experience with preoperative radiotherapy, we have been convinced that this approach offers appropriately selected patients the best chance for curative resection and long-term local recurrencefree survival. There are several advantages to using high dose preoperative radiotherapy in these patients: I. Adequate doses (2 4500 cGy) of radiation can sterilize the peripheral margins of disease. This is especially important within the confines of the lower bony pelvis, where surgical access to the tumor is technically difficult. 2. Preoperative irradiation also allows tumors to be resected with limited longitudinal margins and thereby extends the level at which sphincter sparing surgical resection can be safely performed in the distal rectum. 3. For patients with unresectable or borderline resectable tumors, full dose preoperative irradiation frequently shrinks the tumors sufficiently to allow for potentially curative resection. In addition to the experience at Jefferson, several institutions have recently reported dramatic improvements in local control and survival using this approach. At Washington University, Kodner et al. (10) reported on 90 patients who were believed to have transmurally invasive tumors received 4500 cGy/25 fractions preoperatively. After 4 to 6 weeks, surgical resection was performed which included sphincter preserving procedures in 37% of the patients. No anastomotic leaks were encountered. The authors noted considerable tumor shrinkage and the suggestion of pathologic downstaging compared with an earlier group of patients treated with low dose (2000 cGy) preoperative radiotherapy. A low incidence of nodal metastasis (22%) was noted postradiotherapy which was consistent with pathological findings from other trials using preoperative

October

1991, Volume 21, Number 5

irradiation compared to the surgical controls (3040% nodal metastasis) (6, 9, 17, 19). Excluding Stage D patients, the 5-year survival of these patients was 73% with 98% 5-year pelvic control. At the University of Florida, Mendenhall and colleagues (6) treated 71 patients with clinically respecable rectal carcinoma preoperatively to 3500-4500 cGy over 3l/2 to 5 weeks. On pathological review postradiotherapy, no tumor was found in 11% of the operative specimens. The local recurrence rate was only 8% versus 29% for historical controls treated with surgery (p < 0.02). The 5-year determinant disease-free survival was 71% for the irradiated patients versus 41% for the historically matched surgery only group (p < 0.01). Willet et al. (20) at Massachusetts General Hospital recently presented the outcome of 28 patients which, similar to this review, were selected on the basis of clinical extrarectal tumor extension. These 28 patients with clinically “tethered” rectal carcinomas were treated with pelvic radiotherapy to 5040 cGy over 5 weeks. Four to 6 weeks postirradiation, 23 patients were resected with an APR and 5 patients with a low anterior resection. At 5 years, local control and actuarial disease-free survival rates were 76% and 66%, respectively. Only two patients experienced complications, which included one small bowel obstruction and one case of delayed perineal wound healing. The European Organization for Research and Treatment of Cancer (EORTC) (7) recently completed a multi-institutional trial using preoperative radiotherapy (3450 cGy in 19 days, 230 cGy/fx). This was followed by immediate surgery, which included an abdomino-perineal resection in 81% of the patients. Of the 341 patients who were deemed to have undergone a curative resection with no distant spread at laparotomy, there was a suggestion of improved survival at 5 years in the preoperative radiotherapy arm (69% vs. 59%) with survival figures approaching statistical significance (p = 0.077). The most impressive survival difference was seen in patients less than 55 years of age who had a 5-year survival of 80% with preoperative irradiation versus 48% with surgery alone (p = 0.004). A statistically significant lower pelvic recurrence rate was also observed in the preoperative irradiation arm (15% vs. 35%) compared to surgery alone (p = 0.003). No significant increase in complications was observed in this study. Few studies have examined preoperative assessment of tumor mobility with the exception of the MRC trial (3) Approximately 45% of the patients were determined clinically to have preoperative tumor tethering or fixation with a 5-year local control rate and 5-year survival of only 37% and 29%, respectively. For mobile tumors, the respective local control and 5-year survival were 70% and 48%. Furthermore, patients with partially fixed or fixed cancers had only a 44% chance of being completely resected, compared with an 80% probability in the mobile cancers. In the present TJUH analysis, patients with “tethered” or fixed rectal tumors had similar survival and recurrence rates after high dose preoperative irradiation (5-year sur-

Preoperative

irradiation

for rectal cancer 0 R. L. TOBINet al.

viva1 68% vs 60%). Our results are vastly superior to the Medical Research Trial. The 5-year survival for the MRC patients was only 39% for tethered tumors and 25% for clinically fixed tumors, with dismal pelvic control with surgery alone or surgery plus low-dose preoperative radiation therapy (500 cGy X 1 or 2000 cGy/5fxs). Twenty-one percent of our initial 134 patients had Stage 0, A, Bl tumors at surgery, including 9 patients (7%) with no residual tumor identified in the pathological specimen. We are convinced that these patients had transmural invasion at initial assessment with radiation downstaging to a more favorable Astler-Coller stage. Similar findings have been noted after high doses of radiation by other recent investigators (10, 12, 18). Only one isolated local recurrence has been noted in this group of patients with a 5-year actuarial disease-free and overall survival of 92%. For our B2/C patients, again despite the advanced stage postirradiation and presence in the low rectum (76% below 6 cm level), the 5-year actuarial survival is 62%, which is considerably higher than the 4045% survival noted in surgery-only arms in randomized national trials (4, 14). Despite the high doses of radiation used at our institution, complications have been minimal. Sphincter preserving

1131

surgery was used in the majority of patients including 41/49 (84%) of the clinically “tethered” tumors, and 55/85 (65%) of the pre-radiotherapy fixed rectal cancers.

CONCLUSIONS Rectal cancer encompasses a wide spectrum of tumors with varying degrees of respectability based on local tumor extension and level within the rectum. With proper patient selection, we believe preoperative irradiation for rectal carcinoma with unfavorable characteristics provides optimal presurgical cytoreduction, sustained local control, and excellent survival. Furthermore, sphincter function can be maintained in a majority of patients with appropriate attention to patient selection and treatment techniques. We are presently exploring expanding our criteria for high dose (5000-5500 cGy) preoperative boost for unfavorable rectal carcinomas to enhance pelvic control and survival further. Further prospects for improving the resectability and survival of these patients include exploring the addition of chemotherapy (5_FU/Leukovorin) to clinically fixed or inoperable tumors in conjunction with radiation preoperatively

REFERENCES 1. Astler, V. S.; Coller, F. A. The prognostic significance of direct extension of carcinoma of the colon and rectum. Ann. Surg. 139:846-851; 1954. 2. Cancer Facts and Figures-1990. Atlanta: American Cancer Society 1990. 3. Duncan, W.; Smith, A. N.; Freedman, L. F.; Alderson, M. R.; Arnott, S. J.; Bleehen, N. M.; Bond, W. H.; Crowther, D.; Deeley, T. J.; Duthie, H. L.; Dykes, P. W.; Fielding, L. P.; Flatman, G. E.; Goligher, J. C.; Hawley, P. R.; Hughes, L. E.; Joslin, C. A. F.; Koriech, 0. M.; Morson, B. C.; Oates, G. D.; Peckham, M. J.; Sandland, M. R.; Schofield, P. F.; Slack, W.; Slaney, G.; Smith, J. A. R.; Scott, J. S.; Whitehouse, J. M. A.; Wrigley, P. F. M.; York-Mason, A. Clinico-pathological features of prognostic significance in operable rectal cancer in 17 centres in the U.K. Br. J. Cancer 50:435442; 1984. 4. Fisher, B.; Wolmark, N.; Rocketter, H.; Redmond, C.; Deutsch, M.; Wickerman, D. L.; Fischer, E. R.; Caplan, R.; Lemer, H.; Gordeon, P.; Feldman, M.; Cruz, A.; LegaultPoisson, S.; Jones, J. Postoperative adjuvant chemotherapy for rectal cancer; results from NSABP protocol R-01. JNCI 80:21-29; 1988. 5. Fortier, G. A.; Constable, W. C.; Meyers, H.; Wanebo, J. H. Preoperative radiation therapy for rectal cancer: an effective therapy in need of a clinical trial. Arch. Surg. 121:13801384; 1986. 6. Friedman, P.; Garb, J. L.; Park, W. C.; Stark, A.; DeConti, R. C.; Chabot, R.; Pace, 0. T. Survival following moderate dose preoperative radiation therapy for carcinoma of the rectum. Cancer 55:967-973; 1985. 7. Gerard, A.; Buyse, M.; Nordlinger, B.; Loygue, J.; Pene, F.; Kempf, P.; Bosset, J.; Gignoux, M.; Amaud, J.; Desaive, C.; Duez, N. Preoperative radiotherapy as adjuvant treatment in rectal cancer. Final results of a randomized study of the European Organization for Research and Treatment of

Cancer (EORTC). Ann. Surg. 606-614;

1988. estimations from incomplete observation. J. Stat. Assoc. 53:457463; 1958. Kligerman, M. M.; Urdaneta, N.; Knowlton, A. Preoperative irradiation of rectosigmoid carcinoma including its regional lymph nodes. Am. J. Roentgenol. 114:498-503; 1972. Kodner, I. J.; Shemesh, E.; Fry, R. D.; Walz, B. J.; Myerson, R.; Fleshman, J. W.; Schechtman, K. B. Preoperative irradiation for rectal cancer. Ann. Surg. 209:194-199; 1989. Marks, G.; Mohiuddin, M.; Borenstein, B. D. Preoperative Radiation therapy and sphincter preservation by the combined abdominotransacral technique for selected rectal cancers. Dis. Colon Rectum 28:565-571; 1985. Mendenhall, W. M.; Bland, K. I.; Rout, W. R.; Pfaff, W. W.; Million, R. R.; Copeland, E. M. Clinically resectable adenocarcinoma of the rectum treated with preoperative irradiation and surgery. Dis. Colon Rectum 31:287-290; 1988. Mendenhall, W. M.; Million, R. R.; Bland, K. I.; Pfaff, W. W.; Copeland, E. M. Preoperative radiation therapy for clinically resectable adenocarcinoma of the rectum. Ann. Surg. 202:215-222; 1985. Mohiuddin, M.; Marks, G. Cancer of the rectum. In: Mansfield, C., ed. Therapeutic radiology. Amsterdam: Elsevier 1989. Pilepich, M. V.; Munzemider, J. E.; Tak, W. K.; Miller, H. H. Preoperative irradiation of primarily unresectable colorectal carcinoma. Cancer 42:1077-1081; 1978. Reed, W. P.; Garb, J. L.; Park, W. C.; Stark, A. J.; Chabot, J. R.; Friedmann, P. Long-term results and complications of preoperative radiation in the treatment of rectal cancer. Surgery 103:161-167; 1988. Roswit, B.; Higgens, G. A.; Keehn, R. J. Preoperative irra diation for carcinoma of the rectum and rectosigmoid colon:

8. Kaplan, E.; Meier, P. Nonparametric 9.

10.

11.

12.

13.

14.

15.

16.

17.

1132

I. J. Radiation Oncology 0 Biology 0 Physics

report of a National Veterans Adminstration randomized study. Cancer 35:1597-1602; 1975. 18. Stevens, K. R.; Allen, C. V.; Fletcher, W. S. Preoperative radiotherapy for adenocaxinoma of the rectosigmoid. Cancer 37:2866-2874; 1976. 19. Wassif, S. B. The role of preoperative adjuvant therapy in

October 1991, Volume 21, Number 5 the management of borderline operability rectal cancer. Clin. Radiol. 33:353-358; 1982. 20. Willett, C.; Wood, W.; Donnelly, S.; Shellito, P.; Compton, C. Preoperative irradiation for tethered rectal carcinoma. Proc. of the 31st ASTRO Mtg. Int. J. Radiat. Oncol. Biol. Phys. 17(1):187; 1989.