Inl. J. Radration Onco/ogy Rio/, Phys., Vol. 17, p. 663-668 0360-3016/89 $3.00 + .OO Printed in the IJSA. All rights rewved. Copyright 0 1989 Pergamon Press plc

??Brief Communication

REMOTE AFTERLOADING INTRALUMINAL BRACHYTHERAPY IN THE TREATMENT OF RECTAL, RECTOSIGMOID, AND ANAL CANCER:

A FEASIBILITY STUDY

NATHAN KAUFMAN, M.D.,* DATTATREYUDU NORI, M.D.,* BRENDA SHANK, M.D.,* LUIS LINARES, M.D.,* LOUIS HARRISON, M.D.,* DANIEL FASS, M.D.*

AND WARREN ENKER, M.D.?

Memorial Sloan-Kettering Cancer Center, 1275 York Ave., NY, NY 10021

From 1981 to 1986, 28 patients (27 evaluable) were treated with intraluminal brachytherapy (ILBT) using a remote afterloading technique for persistent or recurrent anal, rectal and rectosigmoid cancers. Eighty-nine percent underwent previous surgery for colorectal cancer. Seventy-seven percent of the patients received external beam irradiation (ERT) as a part of the present treatment. Intraluminal brachytherapy was given with a 2 cm diameter cylinder and the dose per fraction ranged from 440 cGy to 840 cGy at 0.5 cm from the surface of the cylinder. Follow-up ranged from 1 to 74 months with a median of 12 months. Patients were divided into two groups. Group I consisted of 15 patients receiving elective ILBT, Group II: 13 patients with recurrent disease. Seventy-one percent of the patients in Group I and 39% of the patients in Group II achieved local control. The majority of patients tolerated treatment well with only transient reactions. However, three patients (11%) developed grade 3 (GJ) complications requiring surgical intervention. Eight patients developed moderate complications-grade 2 (G+requiring only conservative treatment. This study has identified several factors which appear to influence the risk of developing complications with this combined treatment, using remote afterloading apparatus, among which are technique of previous external beam irradiation, treatment length, anatomical location, intraluminal brachytherapy fractionation, and total cu- mulative dose (ERT + ILBT). This experience suggests that intraluminal brachytherapy appears to be an acceptable form of treatment, as a boost to external beam radiation therapy, in the management of rectal and colorectal cancers.

Intraluminal brachytherapy, Rectal cancer, Remote afterloading brachytherapy.

INTRODUCTION

In the last 15 to 20 years, Radiation Therapy has gained a continually increasing role in the treatment of colorectal cancer. External beam radiotherapy (ERT) is commonly used as an adjuvant to potentially curative surgery, as primary therapy in unresectable or medically inoperable cases, and in treatment of recurrent disease. Electron beam therapy and brachytherapy are used intraoperatively. In- terstitial and intracavitary brachytherapy have been used with significant success for treatment of early, accessible lesions in carefully selected patients.

We have added remote afterloading intraluminal brachytherapy (ILBT) to this armamentarium. Intralu- minal brachytherapy with remote afterloading technique has been in use and development at Memorial Hospital since 1963 (6). Our experience has shown that ILBT af- fords a number of treatment advantages in terms of ra-

diation protection, dosimetry, and technical factors. This technique has been successfully implemented in treatment of gynecological cancers, esophageal cancers, and bron- chial tumors (9).

The current report is a retrospective analysis of our experience with ILBT in the treatment of persistent and recurrent rectal, rectosigmoid, and anal cancers. We pres- ent the techniques and results of treatment, as well as an analysis of complications and recommendations for use of this modality.

METHODS AND MATERIALS

Patient characteristics (Table 1) From January 198 1 to April 1986, 28 patients were

treated with remote afterloading ILBT for rectal, recto- sigmoid, and anal cancers, either for persistence, recur-

* Department of Radiation Oncology. t Colorectal Service, Department of Surgery. Reprint requests to: Dattatreyudu Nori, M.D.

Acknowledgements-The authors thank Marek Rimmel for as-

sistance with data analysis, and Natalie Masino for manuscript preparation.

Accepted for publication 15 March 1989.

663

664 1. J. Radiation Oncology 0 Biology 0 Physics September 1989, Volume 17, Number 3 I

Table I.

Category Indications

Group 1 Elective P.O. Boost to

Primary ERT Group II Recurrent

disease

ILBT alone without ERT ILBT + conventional ERT ILBT + unconventional ERT

No. of patients %

12 43

3 11

13 46

6 22 12 44 9 33

loo* --

rence, or as an adjunct to current therapy. Twenty-seven cases were evaluable (one patient was lost to follow-up after the treatment): 14 (52%) had rectal cancer, 10 (37%) had rectosigmoid cancer, and 3 (11%) had anal cancer. There were 16 males and 11 females. The median age was 66 years, ranging from 27 to 88 years.

80 - f” I

60- 1 Overall Actuarial Survival % I

I ?? 0

40 - b---~-_c_-.__*

Overall Local DFS 20 -

I I I I I 12 3 4 5

Indications for ILBT were as follows: 13 patients were treated for recurrent disease, and the treatment was given postoperatively in 12 patients (4 for gross residual disease, and 8 following suboptimal surgery but without gross re- sidual). Three patients were treated with ILBT as a boost to primary ERT.

Years Fig. I. Actuarial survival and actuarial disease-free survival of the 27 evaluable patients; the values at 5 years being 50% and 36%, respectively.

Prior treatment consisted of surgery plus ERT in 18 patients, surgery alone in 6 patients, and ERT alone in 3 patients. Of the 2 1 patients who received prior ERT, con- ventional fractionation of 200 cGy/fraction was used in 12 patients, and 9 had unconventional fractionation (300 cGy/fraction-7 patients; 250 cGy/fraction-2 patients). Total ERT doses ranged from 2700 cGy to 6000 cGy. Almost all patients were treated with 3 or 4 fields. A variety of surgical procedures were performed in these patients, including low anterior resection, abdomino-perineal re- section (1 patient was treated with ILBT via a fistulous tract recurrence), total and subtotal excisions, and ful- guration. Eight patients received systemic chemotherapy at some point in their treatment course.

lengths varied from 3 to 10 cm. Appropriate shielding was added in selected cases. The dose rate ranged from 100 to 200 cGy/minute. Dose per fraction ranged from 440 cGy to 840 cGy. Number of fractions ranged from 1 to 5, given at weekly intervals. These treatments were delivered on an out patient basis, using topical anesthesia.

RESULTS

Treatment technique

Patients were followed at regular intervals with physical exam and appropriate radiologic and endoscopic studies. Follow-up ranged from 1 to 74 months, with a median of 12 months. Actuarial survival and local disease-free survival were determined using the life-table method. Sta- tistical significance was determined using the G-statistic test for independence.

All patients were treated with a remote afterloader*. Figure 1 shows the survival and local disease-free sur- Details regarding the use of this apparatus have been de- vival of the 27 evaluable patients. Overall actuarial 5-year scribed previously (9) and are also discussed in a subse- survival was 50% and overall local disease-free 5-year sur- quent section of this report. The treatment is delivered vival was 36%. Local control was further analyzed ac- via a single cable mounted high activity 19’Ir source. This cording to indications for treatment (Table 1). Figure 2 is driven from the safe in a sequential manner by a step- shows the actuarial local disease-free survivals for the two ping-motor which is guided by a microcomputer. Standard groups. At 5 years, it was 43% for Group I and 30% for applicator cylinders are used, and in some cases are cus- Group II. The median local disease-free survival was 24 tom made, according to the anatomical location and ex- months for Group I and 12 months for Group II. In ab- tent of the target volume. Cylinders of 2 cm diameter solute terms, 7 1% (10 of 14) of Group I achieved local were used in all patients for ILBT, with doses prescribed control while 39% (5 of 13) of Group II achieved local to 0.5 cm from the surface of the applicator. Treatment control. Of the ten patients in Group 1 achieving local

* Gamma Med II, Mick Radionuclear Instruments, Bronx, N.Y.

Treatment of rectal, rectosigmoid and anal cancer 0 N. KAUFMAN et al. 665

INTRALUM~NAL BRACHYTHERAI’Y patients with evaluable symptoms achieved palliation of IN REtI’NMMHD, RECTAL AND ANAL CANCER those symptoms (median duration, 8 months).

100 -- I

80 n

I C- -)-- a---._-d

20 - Group II Actuarial Local DFS

1 2 3 4 5

Years Fig. 2. Actuarial local disease-free survival for Group I and Group II; the values at 5 years being 43% and 30%, respectively.

control, two were among the four patients who presented with gross residual disease postoperatively, and eight were those who had undergone suboptimal surgery but with no gross residual. Seven of these eight maintained local control at follow-up ranging from 5 to 60 months. Overall, there were 15 complete responses (56%), 10 partial re- sponses (37%), and 2 (7%) minimal responses.

Four patients have achieved long-term disease-free sur- vival (Table 2). Three of these were Group I patients and one was in Group II. Patient 1 had anal carcinoma, achieved a complete response subsequent to 3000 cGy of pelvic ERT and concomittant 5-FU plus Mitomycin C, and received adjuvant ILBT instead of a previously planned abdominal perineal resection (APR). She had persistent refractory proctitis which ultimately resolved 5 years after ILBT.

The symptoms of recurrence in Group II included pain, rectal bleeding, and obstipation/obstruction. Nine of ten

Complications Patients were followed at regular intervals to monitor

development of complications as well as to determine local control. Complications were carefully analyzed for all pa- tients. Overall, 13 patients (48%) had no complications, three (11%) had transient treatment related symptoms not requiring any treatment (G, complication), eight (30%) had chronic symptoms requiring conservative treatment (G2 complication), and three ( 11%) patients had compli- cations requiring surgical intervention (G3 complication). Complications included bleeding, pain, diarrhea, proctitis, ulcer, tenesmus and stricture.

Subsequent discussion of complications refers to G2 and G3 complications. Analysis of complications yielded the following relevant parameters (Table 3): ILBT frac- tionation, total cumulative ILBT plus ERT dose, ERT fractionation, ILBT treatment length, anatomical loca- tion. Total dose and number of fields used for ERT could not be shown to have any relation to development of complications. Treatment with systemic chemotherapy could not be shown to relate to complications and neither could type of previous surgery.

Seven of fifteen (48%) patients receiving less than three ILBT fractions had complications and 4 of 12 (33%) pa- tients receiving more than two fractions had complica- tions.

Figure 3 illustrates the relationship between total cu- mulative radiation dose and complications. Ten of 17 (59%) patients receiving greater than 4000 cGy cumulative dose had complications. One of ten (10%) patients re- ceiving 4000 cGy or less cumulative dose had complica- tions.

Three of 12 (25%) patients who received conventional ERT developed complications, compared with eight of nine (89%) patients who received unconventional ERT and developed complications. Six patients had no ERT, only postoperative ILBT; none of these developed com- plications.

Table 2. Intraluminal brachytherapy in rectosigmoid, rectal and anal cancer: profile of long term survivors

Age Sex Site Primary F$ and date

Current F$

Elective Recurrence

Current F+

ERT ILBT

Complica- tions

Grade I$

status at LFU

(months)

Patient 1 62 F Anal 5Fu + MitoC 2182

Patient 2 58 F Rectal LAR l/83

Patient 3 54 M Rectal LAR 10179

Patient 4 70 F Rectal Local excision 618 1

v’ 200 x 1.5 3182

J 300x 10 2183

J 300 x 10 1 O/80

r/ 300 x 9 8/81

700 X 1 G2 - NED (60+) OS/IO cm

500 X 4 G2 - NED (29+) OS/10 cm

700 x 2 Gz - NED (29+) 0.518 cm

700 X 2 G3 APR NED (54+) OS/10 cm

666 I. J. Radiation Oncology 0 Biology 0 Physics September 1989, Volume 17, Number 3

Table 3. Intraluminal brachytherapy in rectosigmoid, rectal and anal cancer: factors influencing complications (G2 + Gg)

Category

Fractionation ILBT <2Fx >2Fx

ERT conventional Fx ERT unconventional Fx

Treatment length 3-7cm 8-10cm

Total cumulative dose (ERT + ILBT)

4000 cGy 4-7000 cGy

Anatomical site

Total no. of patients

15 12

12 9

8 19

9 18

No. of patients with complications

7 4

3 8

0 11

1 10

% complications

41.5 = 33 1 p N.S.

25 89 1 p < 0.005

< 58 Ol p 0.005

11 55 1 p < 0.025

Rectal Rectosigmoid Anal

14 10 3

7 1 3

50 3 p < 0.05 p < 0.01

ILBT treatment length seemed to have a significant effect on complication rate. None of the patients treated with a treatment length of less than 7 cm had complica- tions compared with 58% who developed complications when the treatment length exceeded 7 cm.

Anatomical location of the tumor site appeared to con- tribute to complications. All patients treated for anal can- cer (three out of three) had complications. Note that one

INTRALUBNNAL BBACHYTHERAPY IN REfJTWfGMOfD, RECTAL AND ANAL CANCER

Analysis of 62 G3 Cm@kations

7000

1 6000 -

5000 - Total

Cu;$ive 4ooo _

WY) 3000 -

0

0

??Compl~catlons

0 No Compkatlons

0

0. *

0 0

0" 0

I I I I 1 1

1 2 3 4 5 6

# of Fractions

Fig. 3. GzGs complications as a function of total cumulative ILBT + ERT dose, and as a function of ILBT fractionation. There was only one complication at a total cumulative dose less than 4000 cGy.

of the rectal carcinomas was 0.3 cm from the anal verge and this patient developed a GZ complication as well.

Table 4 describes the profile of the patients who de- veloped GS complications. At least one of the above men- tioned parameters are relevant in each of the three cases.

DI!XXJSSION

It had been a long held view that colorectal cancer was minimally radiosensitive. Among the reasons for this opinion were the technical limitations of radiotherapy in past years and because the majority of the experience was with advanced, residual, or recurrent cancer (15, 19). Management of colorectal cancer in patients with persis- tent or recurrent disease remains a difficult problem. While palliation is achieved in a great majority of patients when treated with radiation therapy, local control rates have been unsatisfactory and cures have been infrequent. Wang and Schulz ( 19) treated 111 patients with persistent or recurrent disease, with ERT, achieving palliation in 84%. Williams (20) reported an 85% rate of pain relief and a 5.8% 5-year survival. The reported experience in this review of 90% palliation also corroborates the expe- rience in the literature. Results for recurrences are often poorer than for persistent disease, possibly because most recurrences occur in tissue whose vasculature has been compromised by previous surgery (13). The differences between our Group I and Group II are in keeping with these observations. Although the number of patients is small, it is notable that the 30%, 5-year actuarial disease- free survival for our Group II (recurrent cancer) compares very favorably with the reports of ERT alone in the treat- ment of recurrent colorectal cancer (13, 15, 19,20).

Rich et al. (id), treated 26 patients with small cancers limited to the lower $ of the rectum with local excision or fulguration and local ERT. The indications for this

Treatment of rectal, rectosigmoid and anal cancer 0 N. KAUFMAN et al. 667

Table 4. Intraluminal brachytherapy in rectosigmoid, rectal and anal cancer: profile of patients with G, complications

Status at ERT ILBT Interval since Treatment of LFU

No. Age Sex Site Type of surgery (CGY) (GY) Complication ILBT complication (months)

I 83 M Rectal Fulguration 300 X 10 440 X 4 at 0.5 Rectal ulcer 6 months Surgical excision NED (IO+) cm weekly interval Q length 10 cm

II 64 M Anal Excision 300 x 10 700 x 2 Stricture 12 months Colostomy (17) +MitoC 4 weeks apart died of DM

+5Fu Q length 10 cm (locally NED) III 70 F Rectal Local excision 300 x 9 700 X 2 at 0.5 Severe proctitis 9 months APR NED (55+)

(+margins) cm weekly with pain (nests of tumor interval F$ and bleeding cells) length 10 cm

treatment were similar to those for our Group I patients. ERT following surgery with no gross residual disease re- sulted in 94% local control rate. ERT following a surgical procedure leaving gross residual disease resulted in a 56% local control rate. While these data are almost identical to those reported here, note that many of the patients presented in this ILBT review had advanced disease at presentation, and therefore are not exactly comparable.

Addition of interstitial brachytherapy with 19’Ir to ERT has led to improved local control rates (7, 17). Fourquet et al. (2) reported 1 -year local control of 64% in 55 patients with recurrences whose disease was unresectable at time of surgery. A similar aggressive combined approach has been reported by Gunderson et al. (3) and Tepper et al. ( 18) using intraoperative electron beam radiotherapy in- stead of interstitial brachytherapy, with relatively long- term local control ranging from 40% to 90% in selected groups of patients.

Primary radiation therapy is not the treatment of choice in the great majority of colorectal cancers (1, 9, 14, 20). However, in carefully selected early rectal and anal can- cers, Syed et al. (17) and Jackson (7), using combined ERT and interstitial brachytherapy, along with Papillon ( 11, 12) and Sischy ( 16), using intracavitary contact ra- diotherapy, have reported excellent local control rates with minimal side effects and complications.

Remote afterloading technique such as that used in this report has an obvious advantage over brachytherapy af- terloading techniques in that staff radiation exposure is completely eliminated. Other favorable aspects include: short treatment times, outpatient treatments, utilization of small sources which allows intraluminal radiation in otherwise inaccessible sites, ability to tailor treatment and dose distribution according to site and extent of disease, relative ease and convenience of administration of treat- ment, and high acceptance rate by the patients.

Biologic effects of sparsely ionizing radiation are sig- nificantly affected by dose rate (4). Radiobiologic mani- festations of dose rate are related to SLD repair, cell pro- liferation, and reassortment in the cell cycle. Both labo-

ratory and clinical studies of high dose rate effects are not conclusive (5). Although it may be assumed that quali- tative differences in effects on both tumors and normal tissues are incurred with different fractionation schedules, there are no conclusive experimental data to confirm this assumption. The NSD concept has been useful, and Orton ( 10) and Liversage (8) observed that total prescribed doses should be reduced by 20 to 50% in going from low to high dose rate treatments. Our previous experience has shown that increased fractionation plays a significant role in lowering treatment related side effects (9).

We use as standard treatment parameters a dose per fraction of 500 cGy prescribed to 0.5 cm from the surface of the applicator. With a 2 cm cylinder this yields a 909 cGy surface dose and with a 3 cm diameter cylinder this yields a 770 cGy surface dose. With 2 cm diameter cyl- inders, the dose rates of 0.5 cm depth vary from 100 to 200 cGy/minute, and at the surface, vary from 200 to 400 cGy/minute. Before embarking on a treatment course with ILBT, depth dose and surface dose must be thor- oughly considered.

In analyzing the data from this study, it is instructive to compare them to those of an analogous technique es- tablished by Papillon ( 12) and also reported on by Sischy ( 15). The treatment is via proctoscope using a contact 50 KeV orthovoltage unit. The dose rate is 1000 to 2000 cGy/minute prescribed to the surface. Three to four frac- tions are administered at intervals of 2 to 3 weeks with total doses of 10,000 to 15,000 cGy.

Lesions only up to 5 cm by 3 cm can be treated, and 5 cm lesions require two overlapping fields. For primary therapy, additional considerations are that lesions be low grade, exophytic, and no further than 11 to 12 cm from the anal verge. In patients meeting these requirements treated for cure, Sischy reported local control in 95% of 94 patients, and Papillon reported 78% 5-year disease- free survival. While none of our patients could be de- scribed as being in a favorable risk category, 7 1% of the Group I patients did achieve local control. Sischy treated 73 patients whose lesions did not meet the primary criteria

668 I. J. Radiation Oncology ??Biology 0 Physics

due to size, circumferential extent, or thick ulcerative morphology, and reported 3 1 (42%) of these patients as achieving local control. Papillon reported 35% local con- trol in 16 patients who had ulcerative lesions located near the sphincter-for these patients he recommended ERT plus brachytherapy. In our Group II patients, local control was achieved in 39%. There are high rates of complications with both techniques when treating lesions near the anal sphincter. Apparently the transitional mucosa is quite sensitive to radiation and the underlying sphincter muscle is prone to necrosis.

SUMMARY AND CONCLUSIONS

Twenty-eight patients who had previously undergone surgery and/or ERT received ILBT with remote after- loading technique from 198 1 to 1986. Local control was achieved in 7 1% of Group 1 and 39% of Group II patients,

September 1989, Volume 17, Number 3 ,

with seven of eight patients treated adju&ively main- taining local control. Several factors related to increased complications were identified: ERT fractionation, total cumulative ERT plus ILBT dose, ILBT fractionation, ILBT treatment length, anatomic location.

The rationale for treating colorectal cancer with remote afterloading ILBT is explained principalIy by the favorable characteristics of the technique itself. It is simpler and more convenient than interstitial brachytherapy or intra- operative electron beam therapy, and does not require the strict selection criteria of the Papillon technique. The ex- perience is limited, but it appears that ILBT is a feasible form of treatment, either alone or in combination with ERT. Combined with standard ERT, ILBT using multiple fractions, with treatment lengths no greater than 7 cm, appears to result in satisfactory local control and accept- able morbidity in the treatment of colorectal cancer.

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

Cummings, B. J.; Rider, W. D.; Howard, A. R.; Keane, T. J.; Thomas, G. M. External beam radiation therapy for adenocarcinoma of the rectum. Dis. Colon Rectum 26:30; 1983. Fourquet, A.; Enker, W. E.; Shank, B.; Nori, D.; Hopfan, S.; Anderson, L. L.; Hilaris, B. S. The value of interstitial radiation in advanced and recurrent colorectal cancer. En- docuriether. Hyperther. Oncol. 1: 113- 117; 1985. Gunderson, L. L.; Cohen, A. M.; Dosoretz, D.; Shipley, W. U.; Hedberg, S. E.; Wood, W. C.; Rodkey, G. V.; Suit, H. D. Residual, unresectable or recurrent colorectal cancer: External beam irradiation and intraoperative electron beam boost + resection. Int. J. Radiat. Oncol. Biol. Phys. 9: 1597; 1983. Hall, E. J. Time-dose relationships in brachytherapy. In: Hi&is, ed. Afterloading: 20 years of experience, 1955- 1975. New York, N.Y.: Memorial Sloan-Kettering Cancer Center; 1975: 35-40. Hall, E. J. The biological basis of endocurietherapy: the Henschke Memorial Lecture, 1984. Endocuriether. Hy- perther. Oncol. 1:141-151; 1985. Henschke, U. K.; Hilaris, B. S.; Mahan, G. D. Remote af- terloading with intracavitary applicators. Radiology 83:344- 345; 1964. Jackson, B. R. Iridium implants in treatment of anorectal carcinoma. Dis. Col. Rect. 23: 145- 150; 1980. Liversage, W. A comparison of the predictions of the CRS, TDF and Liversage formula with clinical experience. Br. J. Radiol. Spec. Rept. No. 17; 198 1. Nori, D.; Hilaris, B. S.; Chadha, M.; Bains, M.; Jain, S.; Hopfan, S.; Anderson, L. L. Clinical applications of a remote afterloader. Endocuxiether. Hyperther. 1: 193-200; 1985. Orton, C. G. Radiobiological dose rate considerations with

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

remote afterloading. In: Shearer, D. R., ed. Recent advances in brachytherapy physics. New York, N.Y.: American In- stitute of Physics; 198 1: 190-200. Papillon, J. Intracavitary irradiation of early rectal cancer for cure: a series of 186 cases. Cancer 36696-701; 1975. Papillon, J. Rectal and anal cancers: conservative treatment by irradiation-an alternative to radical surgery. New York, N.Y.: Springer-Verlag; 1982. Rao, A. R.; Kagan, A. R.; Chan, P. Y. M.; Gilbert, H. A.; Nussbaum, H. Effectiveness of local radiotherapy in colo- rectal carcinoma. Cancer 42:1082-1086; 1978. Rich, T. A.; Weiss, D. A.; Mies, C.; Fitzgerald, T. J.; Chaffey, J. T. Sphincter preservation in patients with low rectal cancer treated with radiation therapy with or without local excision or fulguration. Radiology 156:527-531; 1985. Sischy, B. The place of radiotherapy in the management of rectal adenocarcinoma. Cancer 50~263 l-2637; I982. Sischy, B. Endocavitary irradiation for adenocarcinoma of the rectum. CA 34333-339; 1984. Syed, A. M. N.; Puthawala, A.; Neblett, D.; George III, F. W.; Mynt, S.; Lipsette, J. S.; Jackson, B. J.; Flemming, P. A. Primary treatment of carcinoma of the lower rectum and anal canal by a combination of external irradiation and interstitial implant. Radiology 128: 199-203; 1978. Tepper, J. E.; Cohen, A. M.; Wood, W. C.; Hedberg, S. E.; Orlow, E. Intraoperative electron beam radiotherapy in the treatment of unresectable rectal cancer. Arch. Surg. 12 1: 421-423; 1986. Wang, C. C.; Schulz, M. D. The role of radiation therapy in the management of carcinoma of the sigmoid, rectosig- moid and rectum. Radiology 79: l-5; 1962. Williams, I. G. Radiotherapy of carcinoma of the rectum. Edinburgh: Churchill-Livingstone; 1960:2 1 O-2 19.