Sphincter preservation with preoperative radiation therapy and coloanal anastomosis

Int. J. Radiation

Oncology

Biol.

Pergamon

Phys.. Vol. 31, No. 3. pp. 553-559. 1995 Copyright (0 1995 Elsevier Science Ltd Printed in the USA. All rights reserved 0360~3016/95 $9.50 + .OO

0360-3016(94)00375-O

SPHINCTER

PRESERVATION

WITH PREOPERATIVE COLOANAL ANASTOMOSIS

RADIATION

BRUCE D. MINSKY, M.D.,* ALFRED M. COHEN, WARREN E. ENKER, M.D.’ AND PHILIP PATY,

THERAPY

AND

M.D.,+ M.D.?

*Department of Radiation Oncology, +Colorectal Service. Department of Surgery. Memorial Sloan Kettering Cancer Center, NY Purpose: To determine if preoperative radiation therapy allows sphincter preservation in the treatment of rectal cancer. Methods and Materials: Thirty patients with the diagnosis of invasive, resectable, primary adenocarcinoma of the rectum limited to the pelvis were enrolled on a Phase I/II trial of preoperative radiation therapy plus low anterior resection/coloanal anastomosis. By preoperative assessment, all patients had invasive tumors (2: T2,28: T3) involving the distal half of the rectum and required an abdominoperineal resection. The median tumor size was 4 cm (range: 1.5-6 em) and the median distance from the anal verge was 4 cm (range: 3-7 cm). The whole pelvis received 46.8 Gy follows by a 3.60 Gy boost to the primary tumor bed. The median follow-up was 43 months (range: 6-82 months). Results: Of the 29 patients who underwent resection, 3 (10%) had a complete pathologic response and 24 (83%) were able to successfully undergo a low anterior resection/coloanal anastomosis. The incidence of local failure was crude: 17% and 4-year actuarial: 23%. The 4-year actuarial survival was 75%. One patient developed a partial disruption of the anastomosis and two developed recta1 stenosis. Analysis of sphincter function using a previously published scale was performed at the time of last follow-up in 22 of the 24 patients who underwent a low anterior resection/coloanal anastomosis. Function was good or excellent in 77%. The median number of bowel movements/ day was two (range: l-6). Conclusions: This technique may be an alternative to an abdominoperineai resection in selected patients. Continued follow-up is needed to determine if this approach ultimately has similar local control and survival rates as an a~ominoperineal resection. Rectal cancer, Adjuvant therapy, Organ preservation. INTRODUCTION

while achieving the same local control and survival rates as an APR and postoperative radiation therapy. We have previously reported our Phase I/II experience of this approach (12). This report updates our preliminary results with additional patients and foliow-up.

The standard surgical treatment for patients with clinically resectable, distal. invasive rectal cancer who are unable to undergo a low anterior resection is an abdominoperineal resection (APR). Although an APR offers good cure rates, it is associated with a permanent colostomy. Furthermore, when the tumor is transmural or involves the mesorectal or pelvic lymph nodes, the incidence of local failure is > 15% and adjuvant pelvic radiation therapy plus chemotherapy is recommended ( 1, 4, 6). Given the morbidity of an APR and the frequent need for postoperative adjuvant radiation therapy for selected patients with rectal cancer, the concept of preoperative radiation therapy plus low anterior resection/coloanal anastomosis as primary therapy is reasonable. The goal of this approach is to convert the surgical procedure from an APR to a low anterior resection/coloanal anastomosis

METHODS

AND

MATERIALS

Patient populution and eligibility criteria A total of 30 patients with the diagnosis of invasive, resectable, primary adenocarcinoma of the distal rectum limited to the pelvis were enrolled on this Phase I/i1 protocol. On preoperative assessment by the surgeon, all patients were judged clinically to require an APR. This Phase I/II protocol opened for accrual in December, 1987. The study was reviewed and approved by the Institutional Review Board of Memorial Sloan-Kettering Cancer Center.

Reprint requests to: Bruce D. Minsky, M.D., Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, 1275 York Ave., New York, NY 10021. ;4~~noM~le~~e~~~~.~-This work was supported in part by phil-

anthropic funds from the New York City District Council of Carpenters. Accepted for publication 1 July 1994. 553

554

I. J. Radiation Oncology 0 Biology l Physics

Of the 30, 25 received preoperative radiation therapy at Memorial Sloan-Kettering Hospital. Because of travel considerations, an additional five patients received their radiation therapy at other institutions. These patients were eligible for protocol and received radiation therapy according to protocol specifications. Because the radiation was not delivered at Memorial Sloan-Kettering Hospital, these five patients were excluded from the analysis of acute radiation toxicity. However, they were included in the analysis of downstaging, patterns of failure, survival, postoperative toxicity, and sphincter function. Pretreatment evaluation included a complete history and physical examination, proctoscopy and/or colonoscopy, transrectal ultrasound, CBC, platelet count, SMA20, CEA, chest x-ray, and abdominal/pelvic CT. Patients with prior nonrectal cancer [except noninvasive cervical carcinoma and skin cancer (excluding melanoma)], or who received prior pelvic radiation were excluded. All tumors were located 3-7 cm from the anal verge (at least f cm above the anorectal ring), as measured by digital exam, and were moderately differentiated adenocarcinomas. Hematolo~c eli~bility criteria included; WBC: 2 4.0 cells/~L, HGB: 2 10, PLT: r 150,000/$. All patients had a Karnofsky performance status of 90-100%. Of the 30 patients analyzed, 23 were male and 7 were female. The median age was 54 years (range: 33-75 years). The median follow-up was 43 months (range: 6-82 months). One patient was lost to follow-up at 17 months. Preoperative staging The preoperative T stage was determined clinically from physical exam, CT scan, and transrectal ultrasound. In the case of a discrepancy, the highest T stage was used. Clinically, all patients had invasive tumors (2: T2, 28: T3). There were no suspicious mesorectal or pelvic lymph nodes seen. Details of the T stage as well as nodal status were also obtained postoperatively from the pathologic specimens. However, because the patients received preoperative radiation therapy, this method may underestimate the true initial pathologic stage. The median tumor size prior to radiation therapy was 4 cm (range: 1.5-6 cm). The median distance from the anal verge was 4 cm (range: 3-7 cm). radiation therapy technical detains: e~~tern~l beam Patients received megavoltage radiation (2 6 Mv) and multiple field techniques f3 or 4 field (PA + laterals or PA:AP + laterals)] were used. All fields were treated each day. Port films were obtained weekly or more often if clinically indicated. Techniques to minimize the toxicity of pelvic radiation were used as previously described ( 10). In brief, at the time of simulation, a small bowel series was performed to help identify the small bowel. Patients were treated in the prone position and shaped blocks were used to help exclude normal tissues from the fields. Whole pe~v~c.~e~d.The lateral borders were 1.5 cm lateral to the widest bony margin of the true pelvic side walls.

Volume 3 I, Number 3, 1995

The distal border was 1 cm below the anus. The superior border was at the L5/Sl junction. The posterior field margin was a minimum of 1 cm behind the anterior bony sacral margin and blocks were used to spare posterior muscle and soft tissues. Because patients did not have clinical evidence of T4 tumors, the external iliac nodes were not included in the lateral radiation fields. Therefore, the anterior margin was at the most posterior aspect of the symphysis pubis. The anus was considered part of the target volume; therefore, it was included in the whole pelvic field. The whole pelvis plus the primary nodal groups at risk received 46.80 Gy. This was followed by a 3.60 Gy boost to the primary tumor bed. Boost jeld. The intent of the boost was to treat the primary tumor with a 3 cm margin and not to include the nodal groups. Therefore, the exact size was determined by the size and location of the primary tumor. In general, field sizes measured 10 X 10 or 12 X 12 cm and corner blocks were used if possible. Opposed lateral fields were used. The boost dose was 3.60 Gy; therefore, the total dose (pelvis + boost) was 50.40 Gy. Radiation therapy was delivered 5 days/week, once per day, at 18.0 Gy/day. Treatment planning was performed with computerized dosimetry and the dose was prescribed to the isodose line (95-100s) that surrounded the treatment volume at risk. Chemotherapy Patients with positive pelvic nodes or metastatic disease found at the time of surgery received postoperative 5-FU based chemotherapy. Because chemotherapy was not part of the protocol design, the agents and techniques were dependent on physician preference. Toxicity assessment during radiation therapy Patients were seen at least weekly during radiation therapy and 2-3 weeks following the completion of radiation therapy. A toxicity assessment and CBC were obtained. If clinically indicated, other blood tests or x-rays were obtained. The NC1 common toxicity criteria were used for most toxicity assessments. However, to provide a more accurate toxicity assessment, the criteria were modified for fatigue, diarrhea, tenesmus, bowel movements, dysuria, and erythema according to pre~ously published criteria ( 11, 14). An a~ominal/pelvic CT scan was performed l-2 weeks prior to surgery. Surgical procedure Four to 5 weeks following the completion of radiation therapy the patients underwent surgery. Surgery was performed in the Trendelenburg-lithotomy position with the patient in Lloyd-Davies stirrups. The entire left colon to the level of the middle colic artery was mobilized, with ligation of the inferior mesenteric artery and vein. The distal left colon was divided with a linear stapler at a level to ensure adequate length to reach to the pelvic floor. A radical resection of the rectum with sharp mobilization

Sphincter preservation 0 B. D. MINSKY

of the mesorectum on all sides and from the levators was performed from the abdominal incision. The rectosacral fascia was sharply incised posteriorly to mobilize the entire rectum to the level of the anorectal ring. Using the technique of Parks (17, 18), the mucosa was stripped from the dentate line to just above the levators. At the level of the anorectal ring the muscular rectal wall was divided by cautery and the specimen removed. The colon was brought into the anal canal, the staple line excised, and a direct anastomosis performed to the dentate line (including some internal sphincter muscle) with interrupted 2-O polyglycolic acid sutures. The colon was replaced into the sacral hollow. The pelvis was drained from above with two closed suction drains. A temporary transverse colostomy was brought to the skin, and primarily matured following abdominal closure. A petrolatum-impregnated gauze roll was placed in the anal canal to prevent “sideto-side” healing, and removed 4 to 5 days later. In general, the colostomy was closed 6-8 weeks after surgery and the patients were then kept on a regular diet with psyllium’ twice daily. A total of six patients had liver metastasis at the time of surgery; three with a single metastasis and three with multiple liver metastasis. Five of the six underwent simultaneous resection of the liver metastasis and primary tumor. One patient did not undergo resection of the primary tumor because the liver metastasis were unresectable. Although biopsy of the primary tumor site revealed no residual tumor, this patient was excluded from the analysis of downstaging because the nodes were not pathologically examined and the analysis of sphincter function because a low anterior resection/coloanal anastomosis was not performed. Toxicity assessment following surgery Patients were seen in routine follow-up (until progression of disease) at least every 3-4 months for the first 2 years, every 6 months for the next 3 years, then yearly. At each visit an interval history, physical examination, SMA-12, CEA, and, if possible, sigmoidoscopy was obtained. A chest x-ray and colonoscopy were obtained yearly. Abdominal/pelvic CT was obtained when indicated by history, examination, or CEA results. Sphincter function was assessed using a questionnaire administered by one physician (P.P.) and scored according to a previous published sphincter function criteria (Table 1 (12). Sphincter function at the time of the last followup was used. Soilage was defined as minimal leakage of mucus or liquid stool that occurred occasionally (one to two episodes per week = mild soilage) or more frequently (2 2 episodes per week = moderate soilage). These episodes were easily controlled by wearing a pad. Difficulty with evacuation was defined as the need, having left the toilet, to return for a second or multiple evacuation(s). ’ Metamucil.

555

et al.

Table 1. Memorial Sloan-Kettering anal sphincter function criteria Excellent: Good: Fair: Poor:

1-2 bowel movements/day, no soilage 3-4 bowel movements/day, and/or mild soilage Episodic > 4 bowel movements/day, and/or moderate soilage Incontinence

Eight patients were excluded from this analysis for the following reasons; five: underwent an APR, one: did not undergo resection of the primary tumor because of multiple unresectable liver metastasis, and two: dead of disease before an assessment by questionnaire was possible. Therefore, 22 of the 24 patients who underwent a low anterior resection/colonanal anastomosis were eligible for the analysis of sphincter function. Determination ofpatterns offailure Failure sites were determined by clinical exam, x-ray, or biopsy. None were determined by autopsy or by reoperation in asymptomatic patients. The one patient with unresectable liver metastasis who did not undergo resection of the primary was excluded from the patterns of failure analysis, leaving a total of 29 patients. Failure categories were expressed as the crude as well as the 4-year actuarial incidence of cumulative (total) failure. Local failure was defined as failure in the external beam pelvic field (tumor bed, pelvic nodes, anastomosis, drain site, or perineal scar). Abdominal failure was defined as failure in the liver, retroperitoneal nodes, or by the presence of peritoneal seeding. Patients who had a liver metastasis at the time of initial operation were scored as having abdominal failure. Distant failure included lung, bone, and brain. Statistical anal.vsis Analysis of the actuarial was performed using the 30 patients were included were no treatment-related intercurrent disease. The start of radiation therapy.

patterns of failure and survival Kaplan-Meier method (5). All in the survival analysis. There deaths and no patients died of results are calculated from the

RESULTS

Downstaging with preoperative radiation therapy Postradiation stages included; 3: TONOMO: 1: T 1NOMO; 9: T2NOMO;

4: T3NOMO;

3: T2Nl MO; 3: T3NlMO:

1:

T3N2MO; 1: T2NlMl; 1: T2N2Ml; 1: T3NOMl; 2: T3NlMl; 1: TXNXM 1. The superior and inferior margins of resection were negative. Excluding the one patient who did not undergo resection of the primary tumor due to multiple unresectable liver metastasis, 10% (3 out of 29) of patients had a complete pathologic response and

1. J. Radiation Oncology 0 Biology 0 Physics

556

83% (24 out of 29) were able to successfully undergo a low anterior resectionlcoloanal anastomosis. The remaining five patients required an APR. In the group of patients who underwent resection, 38% ( 11 out of 29) had positive pelvic nodes. Patterns ojfailure and survival The crude incidence of failure as a component of failure was local: 17%, abdominal: 24%, and distant: 10%. The 4-year actuarial incidence of failure as a component of failure was local: 23%, abdominal: 34%, and distant: 14%. The 4-year actuarial disease-free survival was 66% and the overall survival was 75% (Fig. 1). To-xicity during preoperative radiation therapy The toxicities during preoperative radiation therapy were primarily gastrointestinal, genitourinary, and skin. No patients required hospitalization for treatment-related toxicity. Nonhernatologic. Three separate components of gastrointestinal toxicity were examined (tenesmus. diarrhea, and the number of bowel movements). Diarrhea and frequent bowel movements were treated with diphenoxyIate/ atropine.* However, there was no adequate treatment for tenesmus. The incidence of toxicities is seen in Table 2. No patients developed grade 3-t toxicity. Radiation associated toxicity commonly began during the 3rd or 4th week of radiation therapy (27.0-37.8 Gy) and usually resolved within 2 weeks following the completion of radiation therapy. Hematologic. The median nadir counts were: WBC: 4.6 (range: 3.1-7.0), HGB: 13.2 (range: 7.1-15.1). and PLT (X1000): 218 (range: 144-366). Postoperative to”~i~it?~ and anal sphi~~ter.#il~~t~on Of the 22 patients eligible for sphincter function analysis, one (4%) developed a partial disruption of the anas-

0.9

0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1

I;--1 8757

o!

0

20

Fig. I. Actuarial 2 Lomotil.

40 60 TIME (Months)

80

survival for the total patient group.

1

Volume 3 1, Number 3, 1995 Table 2. ~onhematologic

toxicity during radiation

therapy*

Grade (%) Toxicity

1

2

3

4

Fatigue Nausea Vomiting Diarrhea Tenesmus Bowel Movements Dysuria E~thema

39 17

0 0

0 0

0 0

0

0

0

0

22 35 43 43 57

35 13 26 6 35

0 0 0 0 0

0 0 0 0 0

* Maxjmum grade per patient. tomosis and two (9%) developed rectal stenosis. The partial disruption healed within 2 months with conservative (nonoperative ) management. There was no fistula. Sphincter function was excellent: 36%, good: 41%, fair: 23%, and poor: 0%. Therefore, 77% (17 out of 22) of the evaluable patients had good or excellent sphincter function. Ali patients maintained complete continence of solid stool and no conversions to a permanent colostomy were required on the basis of poor function. Although minor soiling (32%) and difficulty with evacuation were common (50%), these problems were well tolerated. DISCUSSION

Surgical approuci?les.f~~llo~lting preoperative radiation therap! One of the advantages of preoperative radiation therapy is to possibly increase the chance of performing sphincter sparing surgery. The two sphincter-spa~ng surgical approaches have been used following preoperative pelvic radiation therapy for resectable rectal cancer are local excision and low anterior resection/coloanal anastomosis. Local excision. The use of local excision was reported by Otmezguine et al. from the CHU Henri Mondor Hospital. Patients received partial pelvic radiation therapy followed in 6-8 weeks by a local excision and a boost with afterloading Ir-192 (15). At the time of surgery, 8% of patients had a complete pathologic response. With a mean follow-up of 41 months. the local failure rate was 20% and three of five were salvaged with an APR. Of the six patients with positive margins, two deveioped local failure. The 20 patients with local control had normal sphincter function. The authors recommended that the use of this approach should be limited to patients with tumors OCcupying less than 50% of the circumference of the bowel wall. A similar approach was reported by Marks et al. from t-he Thomas Jefferson University. In their experience, 14 ptitients received full dose (45.0 Gy) preoperative pelvic radiation and, because they were medically unsuitable for

Sphincter preservation 0 B. D. MINSKY ei ai.

a low anterior resection, underwent a local excision (8). With a median follow-up of 31 months, the 3-year actuarial incidence of local failure was 23% and survival was 6 1%. Rectal reservoir function and sphincter control were considered good in 93%. Low anterior resection/coloanal anastumosis. The second surgical approach is a low anterior resection/coloanal anastomosis. The use of preoperative radiation therapy followed by low anterior rese~tion/coloanal anastomosis in patients who would otherwise require an APR has been reported by Papillon and Gerard from the Centre Leon Berard (16) and Marks et al. from the Thomas Jefferson University (9). At the Centre Leon Berard, 60 patients with T2-4 tumors received 30.0 Gy X 10 to the pelvis followed in 2 months by a low anterior resection/coloanal anastomosis (16). A complete pathologic response was seen in 17%. With a mean follow-up of 42 months, one patient (2%) developed a local failure and the 2-year actuarial diseasefree survival was 88%. Anastomotic leakages occurmd 13%. A similar approach has been reported by Marks et al. Seventy-four patients received preoperative radiation therapy and 4.5-8 weeks later, this was followed by a variety of sphincter preserving surgical techniques (9). Of the 74, 8 (11%) required an APR. At the time of surgery, 23% of patients had positive mesorectaI/pelvic nodes and 15% had a complete pathologic response, A subsequent manuscript reported the results of 52 patients with tumors I 3 cm from the anorectal ring who were treated in a similar fashion (7). It should be noted that in the Thomas Jefferson University series, distance is measured from the anorectal ring, whereas in our series, distance is measured from the anal verge, which is commonly 2-4 cm distal to the anorectal ring. The T stage was not mentioned. The crude local failure rate was 14% and the 5-year survival was 85%. In a separate update limited to the 38 patients who underwent preoperative radiation therapy followed by a transanal abdominal transanal proctosigmoidectomy and coloanal anastomosis (defined as a low anterior resection/ coloanal anastomosis in our series), the local failure rate was 13% and 5-year actuarial survival was 9 1%. Complications included: pelvic infection; 6%, anastomotic leak; 6%, small bowel obstruction not requiring surgery; 4%, and significant rectal stricture: 2%.

Low anterior resection/coloanal anastomosis vs. APR There are a number of endpoints when comparing the results of radiation therapy plus low anterior resection/ cotoanal anastomosis with an APR. These include downstaging, sphincter function, local control, and survival. The goal of sphincter preserving therapy is to obtain reasonable sphincter function without decreasing the ultimate local control and survival rates as compared with an APR. Downsraging. Following preoperative radiation therapy, 10% of our patients had a complete pathologic response,

557

which is similar to the experience of Papillon and Gerard (17%) (16), Mohiuddin and Marks (15%) (9), and Otmezguine et al. (8%) (15). The incidence of positive pelvic nodes seen at the time of surgery was higher in our series (38%) compared with the data from Mohuiddin and Marks (23%) (9). In our series, 83% of patients were able to undergo sphincter-sparing surgery, which is comparable to the 89% reported by Mohiuddin and Marks. It must be emphasized that the surgeon’s initial assessment of the type of surgery required is made during an office examination. Examination under anesthesia more closely simulates the improved mobility of the pelvic organs sometimes appreciated at the time of operation and may provide a more accurate assessment of the surgical procedure needed. Therefore, our downstaging data should be interpreted with caution. Sphincterjuzction. Sphincter function is an important endpoint in the assessment of the results of conservative treatment of rectal cancer. If a poor functional result is achieved by sphincter preservation, then an APR would be the preferred treatment. In our series, with a median follow-up of 43 months, sphincter function was good to excellent in 77% of patients. Otmezguine et al. reported normal sphincter function in all locally controlled patients (15). Marks et al. reported rectal reservoir function and sphincter control as good in 93% of patients who underwent local excision (8) and normal in 86% who unde~ent low anterior resection/coloanal anastomosis (9). Anastomotic leakages occurred in 13% of patients in the experience of Papillon and Gerard (16), which is slightly greater than the incidence of partiai disruption of the anastomosis in both the Jefferson (6%) and our series (4%). It is difficult to accurately compare the sphincter functional results among the series because all use different scales. Furthermore, the change in sphincter function with longer follow-up cannot be predicted. ~anomet~c testing of sphincter tone may offer a more objective assessment. Overall, the combined experience indicates that most patients treated with preoperative radiation therapy and low anterior resection/coloanal anastomosis will have acceptable sphincter function that is preferable to a permanent colostomy. The most common problems are frequent stools, minor leakage, and incomplete evacuation Local control and survival. Because preoperative radiation therapy downstages rectal cancer, it is not possible to accurately compare, stage for stage, the results of preoperative radiation therapy plus low anterior resection/ coloanal anastomosis with an APR plus postoperative radiation therapy. Furthermore, there is an adverse selection bias against the preoperative approach. This is because patients with unsuspected metastasis found at the time of operation are excluded from postoperative adjuvant therapy, whereas they are included in preoperative adjuvant trials. The Intergroup RTOG 94-01 trial of preoperative vs. postoperative combined modality therapy for resectable rectal cancer will address this issue. The crude local failure rates with preoperative radiation

5.58

1.J. Radiation Oncology 0 Biology 0 Physics

therapy and sphincter sparing surgery range from 2-23%: Papillon and Gerard; 2% (f 16), Otmezguine; 6% ( I5), Marks and Mohiuddin; 13-23% (8, 9), and 17% in the present series. The differences between the local failure rates may be caused, in part, by differences in patient selection and surgical procedures. Because the median follow-up in our series is only 43 months, further followup is needed to perform an accurate comparison of our results with other series. The role of c~ernot~er~~.~ There are few data examining the use of adjuvant chemotherapy in patients who undergo sphincter preserving surgery and radiation therapy. However, given the statistically significant impact of chemotherapy on survival in patients with resectable rectal cancer reported in the randomized adjuvant postoperative trials ( 1,3,4,6), we now recommend the use of preoperative combined modality therapy followed by postoperative chemotherapy in patients whose tumors have transmural penetration (11). CONCLUSIONS

In summa~, the standard surgical treatment for patients with clinically resectable, distal invasive rectal cancer who are unable to undergo a low anterior resection is an APR. We report the use of preoperative radiation therapy as a technique to convert the surgical procedure from an APR to a low anterior resection/coloanal anastomosis.

Volume 31, Number 3, 1995

The preliminary data reveal reasonable local control, survival, and functional results with this approach. The conservative management of invasive rectal cancer with preoperative radiation therapy plus low anterior resection/ coloanal anastomosis may be an alternative to an APR in selected patients. However, we must emphasize that the local failure rate is higher compared with some surgical series (2). Additional follow-up is needed to determine if this approach ultimately has similar local control and survival rates as an APR. The issue of whether to perform a local excision (in those selected patients whom it is technically possible) or to recommend a low anterior resection/coloanal anastomosis following preoperative radiation therapy remains unanswered. Our present approach to sphincter preservation with the use of combined modality therapy in patients with resectable rectal cancer who would otherwise require an APR is dependent on the T stage. Following transrectal ultrasound, patients with a clinical T2 tumor amenable to a full thickness local excision receive postoperative radiation therapy plus 5-FU based chemotherapy following local excision if the margins are negative and tumor is either pa~ologically T2 or if it is a T 1 with adverse pathologic features (13). If by transrectal ultrasound there is transmural penetration, patients receive combined preoperative radiation therapy and 5-FU and low dose leucovorin followed by a low anterior resection or low anterior resection/coloanal anastomosis and additional postoperative chemotherapy (11).

REFERENCES

1. Douglass, H. 0.; Moertel, C. G.; Mayer, R. J.; Thomas, P. R.; Lindblad, A. S.; Mittleman, A.; Stablein, D. M.; Bruckner, H. W. Survival after postoperative combination treatment of rectal cancer. N. Engl. J. Med. 3 15:1294- 1295;

8.

1986.

2. Enker, W. E,; Heilweil, M. L.; Hertz, R. E. L.; Pilipshen, S. J.; Stearns, M. W., Jr.; Janov, A. J.; Sternberg, S. S. En bloc pelvic lyphadenopathy and sphincter preservation in the surgical management of rectal cancer. Ann. Surg. 203: 426-433: 1986, 3. Fisher, B.; Wolmark,

N.: Rockette, H.; Redmond, C.; Deutsch, M.; et al. Postoperative adjuvant chemotherapy or radiation therapy for rectal cancer: Results from NSABP protocol R-01. J. Natl. Cancer Inst. 80:21-29; 1988. 4. Gas~ointestinal Tumor Study Group: Prolongation of the disease-freeinterval in surgically treated rectal carcinoma. N. En&J. Med. 312:14&j-1472; 1985. 5. Kaplan, E. L.; Meier, P. Nonparametric estimation for incomplete observation. J. Am. Stat. Assoc. 53:457-481; 1958. 6. Krook, J. E.; Moertel, C. G.: Gunderson, L. L.; Wieand, H. S.; Collins, R. T.: Beart, R. W.; Kubista, T. P.; Poon, M. A.; Meyers, W. C.; Mailliard, J. A.; Twito, D. I.; Morton, R. F.; Veeder, M. H.; Witzig, T. E.; Cha, S.; Vidyarthi, S. Effective surgical adjuvant therapy for high-risk rectal carcinoma. N. En& J. Med. 324:709-715: 1991. 7. Marks, G.; Mohiuddin, M.; Masoni, L. The reality of radical sphincter preservation surgery for cancer of the distal 3 cm

9. 10. 11.

12.

13. 14.

of the rectum following high-dose radiation. Int. J. Radiat. Oncol. Biol. Phys. 27:779-783; 1993. Marks, G.; Mohiuddin, M. M.; Masoni, L.; Pecchioli, L. High-dose preoperative radiation and full-thickness local excision-A new option for patients with select cancers of the rectum. Dis. Colon Rectum 33:735-740: 1990. Marks, J.; Mohiuddin, M.; Kakinic, J. New hope and promise for sphincter preservation in the management of cancer of the rectum. Semin. Oncol. 18:388-398; 199 1. Minsky, B. D. Pelvic radiation therapy in rectal cancer: Technical considerations. Semin. Radiat. Oncol. 3:42-47: 1993. Minsky, B. D.: Cohen, A. M.; Enker, W. E.; Kelsen, D. P.; Kemeny, N.; flson, D.; Guilkm, J.; Saltz, L.; Franked, J.; Conti, J. Preoperative S-FU, low dose leucovorin, and concurrent radiation therapy for rectal cancer. Cancer 73:273278; 1994. Minsky, B. I).; Cohen, A. M.; Enker. W. E.; Sigurdson, E. Phase J/II trial of preoperative radiation therapy and COloanal anastomosis in distal invasive resectable rectal cancer. Int. J. Radiat. Oncol. Biol. Phys. 23:387-392; 1992. Minsky, B. D.; Enker, W. E.; Cohen, A. M.; Lauwers, G. Local excision and postoperative radiation therapy for rectal cancer. Am. J. Clin. Oncol. (in press). Minsky, B. D.; Kemeny, N.; Cohen, A. M.; Enker, W. E.; Kelsen, D. P.; Reichman, B.; Saltz, L.; Sigurdson, E. R.; Frankel, 3. Preoperative high-dose leucovo~n/5-fluorouracil

Sphincter

preservation

and radiation therapy for unresectable rectal cancer. Cancer 67:2859-2866; 1991. 1.5. Otmezguine, Y.; Grimard, L.; Calitchi, E.; Despretz, J.; Mazeron, J. J.; LeBourgeois, J. P.; Pierquin, B.; Julien, M. A new combined approach in the conservative management of rectal cancer. Int. J. Radiat. Oncol. Biol. Phys. 17:539545; 1989. 16. Papillon, J.; Gerard, J. P. Role of radiotherapy in anal pres-

l B. D. ib%INSKY ttf ai.

559

ervation for cancers of the lower third of the rectum. Int. J. Radiat. Oncol. Biol. Phys. 19: 1219-1220; 1990. 17. Parks, A. G. Perianal anastomosis. World J. Surg. 6:531538; 1982. 18. Pollett, W. G.; Nicholls, R. J. The relationship between the extent of distal clearance and survival and local recurrence rates after curative anterior resection for carcinoma of the rectum. Ann. Surg. 198:159-163; 1984.