Treatment of cervical carcinoma employing a template for transperineal interstitial Ir192 brachytherapy

hf. 1. Radiation Oncology Biol. Phys.. Vol. 9. pp. 819-827 Rinted in the U.S.A. All rights reserved.

036&3016/83/060819-09$03.00/0 tipyri8ht 0 1983 Per8amon Press Ltd.

??Original Contribution TREATMENT OF CERVICAL CARCINOMA EMPLOYING A TEMPLATE FOR TRANSPERINEAL INTERSTITIAL Ir19*BRACHYTHERAPY OTIS

GADDIS JR., M.D.,’ C. PAUL MORROW, M.D.,’ VACLAV KLEMENT, M.D.,* JOHN B. SCHLAERTH, M.D.’ AND RICHARD H. NALICK, M.D.’

‘Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Los Angeles County, University of Southern California Medical Center, Women’s Hospital; *Department of Radiology, Division of Radiation Oncology, Los Angeles County, University of Southern California Medical Center, General Hospital The development of a template technique at this institution for transperineal interstitial-intracavitary

bracbytherapy employing Ir”* wire has previously been reported. In this paper we report the results of radiation treatment of 84 women with fresh, primary squamous carcinoma of the cervix admitted to tbe Los AngelesCounty-Universityof Southern California Medical Center from April, 1975 to September, 1979 who received at least one transperineal template implant as part of their initial treatment. The 75 evaluable patients were followed 3 to 60 months, with a median of 17 months. Recurrence rates in the pelvic treatment field by clinical (FIGO)stage groupingwere 35.7% (5/14) StageIB; 0% (O/8)Stage HA; 20% (S/25) Stage IIB; 46.2 % (12/26) Stage III; and 0% (O/2) Stage IVA. The overall failure rate within the treatment field was 29.3% (22/75). The non-tumor associated rectovaginal and vesicovaginal fistula rate was 14.3% (2/14) in Stage IB; 0% (O/8) in Stage IIA; 16.0% (4/25) in Stage IIB; 15.4% (4/26) in Stage III; and 0% (O/2) in Stage WA. The non-tumor associated fistula rate for al1 stages was 13.3% (10/75). Severe or grade III no&tulous, delayed adverse effects (proctosigmoiditis, cystitis, vault necrosis) occurred in an additional 6 patients. Thus, 21.3% (16/75) of all evaluable patients experienced severe adverse radiation effects during the follow-up period. he-radiation staging laparotomy was performed on 31 patients. It bad no obvious effect on the pattern or rate of radiation complications. The role of the interstitial-intracavitary template in the treatment of primary cervical carcinoma is discussed. Cervical Carcinoma, radiation effects.

Interstitial brachytherapy, __ Treatment failure, Non-tumor associated fistulae, Severe adverse

INTRODUCIION

The potential advantages of this perineal template system are: 1) a more homogenous dose distribution even in the presence of poor pelvic geometry (vaginal stenosis, tumor); 2) more flexibility in dose distribution permitting a higher dose to areas of advanced tumor volume such as the parametria and pelvic sidewalls without the concomitant parallel increase in dose to the vulnerable bladder and rectum.‘*’ If these hypotheses are correct, the template should increase the local control rate in patients with large volume disease without a proportional increase in severe adverse effects. The purpose of this report is to analyze our experience with this template brachytherapy in the treatment of primary carcinoma of the cervix with special emphasis on tumor control and severe, adverse irradiation effects.

Since the advent of radiation therapy as a primary treatment modality in the control and cure of cervical

carcinoma, clinicians have relentlessly searched for improvements to increase survival, without a parallel increase in treatment related complications. Local resistance in patients with large primary lesions continues to be a major source of treatment failure, partly because the distorted anatomy prohibits the optimal use of conventional brachytherapy. In addition the nearby bladder and rectum limit the dose that can be safely administered. A brachytherapy system addressing these shortcomings was developed at the Los Angeles County-University of Southern California Medical Center and put into clinical use in 1974.4 This system employs a plastic block with central perforation for placement of a vaginal cylinder. Around the central fenestration are 44 smaller, symmetrical placed perforations through which guide needles are inserted into the parametrial tissues. The guide needles are designed for after-loading with Ir19’wire. The details of the applicator have been presented elsewhere.4*s*‘4

METHODS AND MATERIALS From April, 1975 to September, 1979 105 patients with pelvic malignancy who were evaluated and treated at the Women’s Hospital, Los Angeles County-University of Southern California Medical Center, had one or two

Reprint requests to: Otis Gaddis Jr., M.D. Women’s Hospital, 1240 N. Mission Road, L-903, Los Angeles, CA 90033.

Accepted for publication 20 January 1983. 819

820

Radiation

Oncology

0 Biology 0 Physics

transperineal template implants as part of their therapy. Eighty-four of these 105 patients had fresh, histologically confirmed primary cervical squamous carcinoma. These 84 women accounted for approximately one half of all cases of cervical carcinoma treated in the LAC/USC Department of Radiation Oncology during the study period. Nine of the 84 patients were excluded from analysis because they underwent hysterectomy post radiation as part of their planned treatment (1 patient), or were lost to follow-up within 3 months after completing their radiation therapy (8 patients). Thus, the final study group consists of 75 patients. Pretreatment evaluation included history and physical examination, chest X ray, intravenous pyelography, cystoscopy, urine cytology, urinalysis, proctosigmoidoscopy, complete blood counts, electrolyte panel, liver panel, biochemical profile, serology and clotting screen. When clinical findings dictated, examination under anesthesia, radionuclide scan (liver, brain, bone), computerized tomography (brain, chest, abdomen, pelvis) or biopsy of any suspicious soft tissue lesion outside the pelvis was also performed. The 31 patients who qualified for and consented to approved Gynecologic Oncology Group protocols underwent pretreatment staging laparotomy, which consisted of examination under anesthesia followed by exploratory laparotomy via a midline abdominal incision. Manual and visual exploration of the abdominal and pelvic viscera were performed with intraoperative biopsies of areas suspicious for tumor. The paravesical and pararectal spaces were developed and the cardinal ligaments palpated for tumor infiltration. A selective pelvic and para-aortic lymph node dissection was then accomplished. Following histological diagnosis and metastatic evaluation, each patient was presented to the combined Gynecologic Oncology-Radiation Oncology Tumor Board. The patients received a combined departmental clinical FIG0 (International Federation of Gynecology and Obstetrics) staging’ and treatment profile was jointly agreed upon. All patients were to receive 5000 rad whole pelvis megavoltage teletherapy (Linac; Cobalt’j”) with or without a midline block at 4000 rad depending on central tumor volume and tumor response. Treatment dosage was 180200 rad per day via 15 cm x 15 cm PA/AP portals. Portals were extended to the common iliac or aortic nodes only when staging laparotomy documented the presence of disease in these areas (12 patients). An occasional patient received two or three initial fractions of 250-500 rad to control severe vaginal hemorrhage. Whether the Fletcher-Suit or perineal template brachytherapy system was chosen depended upon the local anatomy at the time of the implant. Patients with poor geometry, a large central tumor or parametrial infiltration received one or two perineal template applications. All others received the Fletcher-Suit tandem and ovoids. The brachytherapy was administered in two sessions at two week intervals after completion of the external beam therapy.

June 1983, Volume 9, Number

6

Dosimetry plots for the perineal template applications used in this study were based on an ideal geometric model. This model stored within the computer predicts individual source location assuming that the source guides are straight, perpendicular to the template and placed at an equal tissue depth. Following a template application, anterior-posterior and lateral radiographs of the pelvis are taken. The X-Y-Z coordinates of the template system are determined and the external cervical OSis designated the geometric origin. The computer is then given the following information: 1) number of source guides in the template and their location; 2) the number of Iridium”’ sources (seeds) per guide, along with the source activity and source separation (usually 1 cm); and 3) source (radium or cesium) activity and separation for the intrauterine tandem. With all appropriate information provided, the computer calculates a dose rate at any desired geometric point within the tumor mass by summating the dose contributions from all the Iridium’92 source points. The calculation incorporates the inverse square law and a correction factor for attenuation attributable to medium and scatter. The contribution of dose rate from tandem sources is calculated from tabular date as provided by Young and Batho.“s’9 Tod’s points A and B doses are considered to be of insignificant value when calculating dose rates for the template.“V’s*‘6 The clinic and hospital charts of all 84 patients were scrutinized with attentiveness to age, gravidity, general health (incapacitating illness), tumor size, tumor histology, remote and recent pelvic surgery, irradiation treatment profile, duration of follow-up, treatment failure, fistula formation, and non-fistulous significant adverse irradiation effects. The appearance of the latter three items were correlated with post-treatment follow-up. Patients were examined bi-weekly post completion of treatment until resolution of pelvic tenderness. Thereafter, examinations were performed quarterly for one year, then every four months for two years, and then biannually. The median age of the patients in the study group was 50 years (range 3&87 yrs) and the mean gravidity 5.3 (range O-14). The histology was squamous in 73 patients and adenosquamous in 2. Seventy-one percent (53/75) of the patients in the study group enjoyed good health at the time of diagnosis, although three were treated in the past for a potentially incapacitating illness (tuberculosis, syphilis, ethanol abuse). Of the remaining 22 patients, 13 (17.3%) presented with cardiovascular related illness and/or diabetes mellitus (only one of 13 was debilitated-this patient had diabetic neuropathy); one patient had significant chronic obstructive pulmonary disease; five patients presented with tumor related renal failure and three with tumor related anemia. Patients were considered disease free when 1) the cervical-vaginal tumor bed had completely healed; 2) all pelvic tenderness had abated following completion of radiation therapy; 3) the pelvic examination was normal

Interstitial brachytherapy ??0.

except for radiation related changes and 4) the pap smear had reverted to normal. With the reoccurrence of any of the above parameters in association with weight loss, sciatic pain, lower extremity edema or deterioration of the intravenous pyelogram, an investigation to ferret out recurrent carcinoma was undertaken. While some patients had clinical symptoms, signs, and X ray studies suggestive of recurrence, only those with biopsy verification of recurrent squamous carcinoma in the pelvis were considered to have recurrence in the treatment field for the purpose of this study. Fistulae were considered non-tumor associated if 1) tumor was absent in the site prior to commencing therapy and, 2) if tumor was absent in the area of the fistula at the time the fistula was diagnosed. The grading of bladder, vaginal, rectal and sigmoid adverse effects was modified from Strockbrine er ~1.‘~ Grade 2 adverse effects were those persisting greater than 12 months, requiring prolonged narcotic use, frequent clinic visits, transfusions and/or showing abnormal X ray changes in the involved brgan. Grade 3 adverse effects were those requiring prolonged hospitalization, resuscitation with more than 3 units of blood for a single bleeding episode, and/or corrective major surgery. Grade 1 adverse effects (moderate, intermittent symptoms with minimal X ray changes) were not considered significant for this report. RESULTS The 75 patients received 116 template and 30 FletcherSuit applications for a total of 146 endocurie treatments (Table 1). Seventy-one patients had two applications, either 2 templates (41 patients) or one template and one Fletcher-Suit (30 patients). Four patients received a single template application as their only brachytherapy. Recorded tumor response post teletherapy and prior to the first brachytherapy application ranged from 40-70%, but this information was not consistently obtainable. All patients were followed a minimum of 3 months after the last brachytherapy treatment. The median follow-up was 17 months, with a range of 3-60 months. Recurrences The overall recurrence rate was 41.3% (3 l/75; Tables 2 and 3). Twenty-nine percent (22/75) of the study group Table 1. Method of brachytherapy

GADDIS

JR. et al.

recurred within the treatment field. Six of 14 patients (42.7%) with Stage IB disease recurred, five (35.7%) in the treatment field of which four (28.6%) were central failures. The recurrence rate for Stage IIB patients was 32% (8/25), of whom 5 (20%) recurred in the treatment field. For Stage III patients, the overall recurrence rate was 61.5% (16/26), with 46.2% (12/26) failing in the treatment field. Seventy-seven percent (24/31) of the recurrences were diagnosed within 12 months post therapy, 84% (26/31) within 18 months, and 94% (29/31) within 24 months (Fig. 1). Median recurrence time was 6 months, with a range of l-28 months. Five patients recurred outside the median follow-up period of 17 months. Non-tumor associatedfistulae A total of 17 fistulae occurred in the study group of which 12 were non-tumor associated. In Stage IB the rate was 14.3% (2/14), Stage IIB 16% (4/25) and Stage III 15.4% (4/26). No fistulae were recorded in Stage IIA. Both Stage IVA patients had fistulae, but these were considered tumor associated. The overall fistula rate was 13.3% (10/75). When the absolute number of fistulae rather than the number of patients with fistulae is used for calculations, the fistula rate for Stage IIB increases to 24% (6/25) and the overall fistula rate becomes 16% (12/75). Ten of 12 (83.3%) fistulae were recta-vaginal and two of 12 (16.7%) were vesicovaginal. The defect size of eight fistulae was recorded. The range was 2 cm x 1 cm to 3 cm x 4 cm with an average size of 2.1 cm x 2.3 cm. Four recta-vaginal fistulae were located at the vaginal apex, three in the middle third of the vagina, and one in the distal third of the vagina. The location of two fistulae was not recorded. The mean tumor diameter in patients developing a fistula post radiation therapy was 7.3 cm. The range was 4-8 cm. Figure 2 analyzes the fistulae versus time of follow-up. The median time post treatment to fistulae diagnosis was 11 months, with a range of 5 to 19 months. The median follow-up of all patients in the study group was 17 months. Three of the 12 non-tumor associated fistulae (25%) occurred beyond the median follow-up time. Eight of 10 patients (80.0%) with non-tumor associated fistulae Table 2. Frequency and location of recurrent cancer in the study group

in study group

Recurrent in treatment

Brachytherapy FIG0 stage

IB IIA IIB III (A + B) IVA Total

All recurrent

Total

patients

Template

14 8 25 26 2 75

22 11 38 41 4 116

821

Fletcher-Suit 6 4 9 11 0 30

FIG0 stage IB IIA IIB III A + B IVA Total

Total patients 14 8 25 26 2 75

No.

(%)

6 1 8 16 0 31

(42.7) (12.5) (32.0) (61.5) (0.0) (41.3)

field No.

@Jo)

5 0

(35.7) (0.0) (20.0) (46.2) (0.0) (29.3)

1: 2;

822

Radiation Oncology 0 Biology 0 Physics

June 1983, Volume

Table 3. Site of recurrence

FIG0 stage IB IIA IIB III (A - B) IVA Total

14 8 25 26 2 15

*Only site of recurrence

Number 6

for the study group

Total patients recurrent

Total patients

9,

Site of recurrence Central

6 1 8 16 0 31

Pelvic wall

4* 0 3* 6 0 13 (17.3%)

Distant

4 0 4* 10t 0 18 (24.0%)

5 1 5 7 0 18 (24.0%)

in 1 patient.

tOnly site of recurrence in 3 patients.

were treated with two template applications. ing two patients with non-tumor associated received one template and one Fletcher-Suit

adverse radiation effects, and 16 patients (21.3%) experienced 18 Grade 3 effects. Because of group crossover, 16 of 75 patients (21.3% of the study group) had a total of 30 incapacitating injuries (12 fistulae and 18 Grade 3 severe adverse effects). Overall there were 43 serious complications of treatment recorded in 16 patients. None of these 43 complications resulted in a treatment related mortality. Site analysis (Table 5) denotes that 66.6% (12/ 18) of the severe adverse effects were rectosigmoid, 16.7% (3/ 18) bladder, and 16.7% (3/ 18) vaginal necrosis. Only two patients having significant adverse effects and fistulae

The remainfistulae each application.

Non-fistulous adverse e$ects Table 4 demonstrates an even stage distribution of non-fistulous Grade 3 adverse radiation effects with four for Stages IB/IIA, and six each for Stage IIB and Stage III. The percent of patients injured was 18.2 (4/22) for Stages IB/IIA, 24 (6/25) for Stages IIB and 23 (6/26) for Stages III. Twelve patients (16%) had 13 Grade 2

FIQURE TIME

TO

1

RECURRENCE

POST-THERAPY TIME RANGE

N-70

RANGE

N-40

N-56 N-4

1

N-31 \ N-29

N-34 11

3

3

TOTAL

(b)

NUMBER

3

NUMBER

BEGINNINa

IN BAR

CARCINOMA

N-27

5

8

12

OF OF

IS

15

PATIENTS EACH

ACTUAL

DIAQNOSED

18

REMAINING

3 MONTH

NUMBER DURING

21

IN THE

24

STUDY

OBSERVATION

OF

PATIENTS

A 3 MONTH

WITH

27

GROUP

30

AT

PERIOD

RECURRENT

OBSERVATION

PERIOD

MONTHS 6 MONTHS

OF

ALL

17

a N-75

THE

l-28

3-60

MEDIAN

N -

RECURRENCE

MEDIAN

FOLLOW-UP

(a)

TO

PATIENTS MONTHS MONTHS

Interstitial brachytherapy 0 0. FIQURE TIME TO FISTULA

GADDIS

823

JR. et of. TIME TO FISTULA

2 POS-i

RANGE

THERAPY

5-l@

MEDIAN

11 MONTHS

FOLLOW-UP RANGE

OF ALL PATIENTS 3-60

MEDIAN

N-65

MONTHS

MONTHS

17 MONTHS

N-45

N-71

N-54

N-75 1 N-75

I

1

:

I

3

9

12

3

N-50

15

18

21

24

MONTHS POST TREATMENT N -

TOTAL

PATIENTS

IN GROUP

NUMBER IN EAR IS NUMBER OF FISTULAE

had remote pelvic surgery (1 oophorectomy in 1947; 1 subtotal hysterectomy and adnexectomy in 1950). Investigation of the brachytherapy method used in the severe adverse effects group disclosed that the 54.7% (41/75) of the study group who received two template applications accounted for 69% (1 l/16) of the patients with severe adverse effects, while 40.0% (30/75) of the patients who received one perineal template and one Fletcher-Suit application accounted for 25% (4/16) of the patients with severe adverse effects from radiation

Table 4. Frequency

Stage IB IIA IIB III IVA Total

Total patients 14 8 25 26 2 75

therapy. Thirty-one of the 75 (41.3%) evaluable patients had a pretreatment staging laparotomy (Table 6). Fifteen Grade 3 adverse effects and fistulae (SAEF) were noted in eight patients of each group. In the staging laparotomy group six of 25 Stage IIB and III patients had 12 SAEF. In the nonstaging laparotomy group, five of 26 patients with Stage IIB and III had 10 SAEF. The major difference in the two groups was a larger number of Stage I cases in the no laparotomy group (11 cases) compared to the laparotomy group (3 cases).

of Grade 2, Grade 3 and fistulous radiation

Patients with significant adverse effects 3 1 6 6 0 16

*Includes proctosigmoiditis, cystitis and vaginal necrosis. tSix fistulae in four patients.

adverse effects Adverse effects

Grade 2*

Grade 3 *

Fistulae

Total

2 3 5 3 0 13

5 1 6 6 0 18

2 0 6t 4 0 12

9 4 17 13 0 43

824

Radiation

Oncology

0 Biology 0 Physics

June 1983, Volume 9, Number 6

Table 5. Organ site of Grade 3 radiation

adverse effects and non-tumor

associated

fistulae

Location of injury Fistulae Stage IB IIA IIB III A + B IVA Total

Total patients

Patients with SAEF*

14 8 25 26 2 75

Sigmoid

3 1 6 6 0 16

Bladder

Vaginal necrosis

RV

1 0 0 2 0 3

1 0 1 1 0 3

2 0 4 4 0 10

3 1 5 3 0 12

vv

2

2

*Significant adverse effects and fistulae. RV = Rectovaginal; VV = Vesico-vaginal.

DISCUSSION The proximity of the bladder and rectum to the pelvic genitalia and their relative sensitivity to radiation injury are the primary determinants of the maximum dose of radiation which can be delivered to the cervix and medial parametrial tissues. Even within the dose constraints imposed by these organs, optimal results often can not be achieved in the treatment of cervical carcinoma because the tumor is large or the local conditions are unsuitable to obtain satisfactory dosimetry using conventional intracavitary applicators. The insertion of multiple interstitial radiation sources into the parametria through a perineal template helps to overcome the dosimetric limitations resulting from poor geometry. In addition, because of its flexibility this method allows a more favorable parametrial to bladder and rectum dose distribution through differential loading (number, location, source activity) and unloading of the needles and the central tandem. This clinical trial was undertaken to see if these apparent advantages could be translated into an improvement in treatment results. Patients were selected for the interstitial template therapy because they had large cervical malignancies, regardless of stage, and/or an anatomic configuration unsuited to the Fletcher-Suit tandem and ovoids. Over the course of the trial, the judgment about the need for interstitial therapy vis a vis conventional intracavitary therapy became less restrictive and by the end of the study period nearly all patients undergoing

Table 6. Non-tumor

associated

Total cases

IB IIA IIB III (A + B) IVA Total

3 3 12 13 0 31

related to pre-radiation

LAP

Total patients with SAEF 1 1 3 3 0 8

cancer

received one or two

Survival Although the target patients of this study have a somewhat poorer prognosis stage for stage than the cervical cancer population in general, evaluation of the results nevertheless requires comparison to other reported series. The survival data from the M.D. Anderson Hospital and Tumor Institute (MDAH) can be used as a yardstick because the series is large, contemporary, carefully analyzed and their treatment methods have become a norm. Based on a total of 1705 patients, in 1971 Fletcher and Rutledge reported the five year survival rates of patients with carcinoma of the cervix on the intact uterus (Table 7).6 Stages IIB and III can be compared more directly because patient selection is less likely to bias treatment outcome than in the earlier stages. The MDAH five year survival rate for Stage IIB is 66.5%, while the percent of patients alive and disease free at two years in our perineal template series is 68%. (For Stage III patients, the MDAH five year survival rate is 41% while 38.5% of the Stage III patients in our study were alive and disease free at one year post therapy.) It also seems appropriate to compare our current results with earlier data from this institution by Kurohara et al.” The five year survival rate was 50% for 83 Stage III cervical cancer patients treated between 1968 and 1974 employing megavoltage teletherapy and the MDAH brachyther-

Grade 3 adverse effects and fistulae (SAEF) Staging

FIG0 stage

radiation therapy for cervical interstitial template implants.

staging laparotomy

No staging LAP

Total SAEF

Total cases

Total patients with SAEF

2 1 5 I 0 15

11 5 13 13 2 44

3 0 3 2 0 8

Total SAEF 5 0 7 3 0 15

Interstitial brachytherapy 0 0.

Table 7. Five year corrected survival squamous carcinoma of the cervix at the M.D. Anderson Hospital FIG0 stage I 11

Number of patients

% survival

407 618

91.5 75.4

A B III IV

327 291 599 81

83.5 66.5 41.0 14.0

From Fletcher, page 660.6

apy schedule.” The clinical characteristics of the study group of Kurohara et al. is very similar to the present study group in age, race and economic status. It appears that, while the survival rates achieved in this study are good, there is no obvious improvement over results reported from MDAH and our own preceeding experience. Local control A group of patients with “early lesions” requiring special treatment ingenuity are those with large Stage IB lesions including the barrel cervix. The difficulty in curing these patients has been documented by Fletcher,6 who pointed out the significant incidence of central failure in patients with large, non-parametrial disease following whole pelvis external beam radiation and intensive brachytherapy. In a later paper from MDAH7, only 2% (l/50) of patients with an endocervical carcinoma >6 cm in diameter, who received adjuvant extrafascial hysterectomy as part of their therapy, experienced a central recurrence. A similar group of patients not receiving hysterectomy failed centrally in 17.5% (18/103) of cases. There was also a substantial difference in survival between the two groups of patients. The five year survival rate for the extrafascial hysterectomy group was 6 1% compared to 50% in the non-hysterectomy group. Others have also documented the impact of tumor volume on local control and survival. Holmesley et al.* reported a 95% five year cure rate for radiation treated Stage IB cervical cancer patients with lesions less than 4 cm in diameter. A similar group with a lesion size of r4 cm had a cure rate of only 67%. The central failure rate for the two groups was 5% (l/21) and 13% (3/22) respectively. All patients received whole pelvis and two intracavitary implants using Manchester or Fletcher-Suit applicators. They concluded that adjuvant hysterectomy should be performed in patients with Stage IB cervical cancer when the lesion size is greater than 4 cm in maximum diameter. The tumor size of the 22 patients with non-parametrial disease (Stage IB and IIA) in our study group ranged from l-10 cm with an average of 5 cm. All four of the patients (19%) failing in the cervix had Stage IB disease. Excluding the five patients with lesions t4 cm in diame-

GADDIS

JR. et al.

825

ter, the central failure rate was 23.5% (4/17), a figure similar to the non-hysterectomy group reported from MDAH. We conclude that the technique used in our study will not eliminate the need for adjuvant simple hysterectomy.

Complications That a positive correlation between increasing treatment dose and treatment related, non-tumor associated fistulae and other adverse effects exists is a basic principle Numerous other factors can also of radiation therapy. ‘*3*9 influence the overall complication rate, the most important of which are the dose rate, treatment time, tumor size, field size, prior surgery and certain medical problems. Thus, direct comparison of complication rates between series is always difficult. The overall radiotherapy related fistula rates reported in the literature range from 1.4% to 5.3%~.~**‘~~‘Again ~*‘* looking at results in more advanced patients because the selection factors will be minimized, Strockbine et al.13 at the MDAH found 6 non-tumor associated fistulae in 215 Stage IIB patients (2.8%), all of whom were treated by external beam radiation and Fletcher-Suit brachytherapy. They reported 21 non-tumor associated fistulae among 443 Stage III patients, a fistula rate of 4.7%. By comparison, in this study, 10 of 75 patients developed 12 non-tumor associated rectovaginal and vesicovaginal fistulae, for an overall fistula rate of 13.3%. Twenty-five percent (3/12) of the non-tumor associated fistulae occurred beyond the median follow-up time of 17 months. By extrapolation the actual fistula rate should lie between a maximum of 22.2% (10/45) and the observed rate of 13.3% (10/75). The non-tumor associated fistula rate for Stage IIB patients was 16.0% (4/25); for Stage III patients it was 15.4% (4/26). Non-fistulous Grade 3 adverse effects were noted in 6.4% (22/346) of the radiation only study patients reported by Peckham et al.” A similar rate of 6.7% (53/784) was reported by Strockbine et aLI3 However, all 53 Grade 3 non-fistulous adverse effects in their study occurred in Stage IIB and Stage III (A & B) patients, a rate of 8.1%.13 In this study group, the overall non-fistulous Grade 3 radiation adverse effect rate was 21.3% (16/75). Seventyfive percent (12/ 16) of those incurring Grade 3 injuries were Stage IIB and III patients, giving a rate of severe adverse effects of 23.5% for these patients. Eighteen percent of Stage IB/IIA patients experienced Grade 3 adverse effects. However, these figures cannot be fairly compared to complication rates usually reported for Stages IB and IIA patients because the average tumor size in these combined “early” stages was 5 cm. Clearly the rate of treatment related complications in this study exceeded that expected from conventional standards of radiation therapy. Although a preliminary review of results using this

826

Radiation Oncology 0 Biology 0 Physics

template system suggested that survival in patients with advanced cervical carcinoma would be superior to results obtained by conventional external beam-intracavitary brachytherapy methods,4 current analysis indicates that no survival advantage has been achieved by comparison to a historical control group from our own institution” and by comparison to results reported from other institutions.6.7V’3Furthermore, it appears that the use of interstitial template brachytherapy to treat early stage disease with a large tumor volume is inferior to the results obtained by conventional radiotherapy combined with an adjuvant simple hysterectomy as advocated by Fletcher and Rutledge at the MDAH. The data from this study, unfortunately, do not permit conclusions regarding the efficacy of parametrial interstitial therapy versus intracavitary therapy in controlling lateral parametrial and pelvic nodal disease. It is precisely in these areas that this techniques should prove to be most advantageous. The frequency of serious complications, especially fistulae, is substantially higher than that noted in the preliminary report.4 Even considering that the patients were selected on the bases of poor local geometry and large tumor volume the complication rate is surprisingly high, far exceeding the adverse effects rates previously reported from our own and also from other institutions. Pre-treatment staging laparotomy did not appear to be a major factor, but the use of two templates versus one template (plus one Fletcher-Suit) application did increase the risk for complications. Most of the severe adverse effects were a result of rectal or sigmoid injuries, at least some of which may be related to remedial technical problems. For example, the 0600 needle at the periphery of the vaginal cylinder sometimes entered the rectum. This needle is no longer used unless the cancer extends onto the vagina posteriorly, and it is never placed through the rectal mucosa. In reviewing the dosimetry data, different patients were noted to have 1) doses to point A in excess of 12,000 rad, 2) bladder and rectal doses at times in excess of 8,600 rad or 3) dose rates to point A of up to

June 1983, Volume 9, Number 6

200 rad/hr. Other probable contributing factors to the predominantly posterior orientation of the radiation injuries observed in this study group is the linear arrangement of the sources’3 and, perhaps, the posterior slant of the needles necessitated by the normal pelvic anatomy. These preliminary data on fistula formation and Grade 3 adverse effects do not support the hypothesis that differential sparing of the rectum and bladder will safely permit a substantially higher parametrial and pelvic wall dose. What actually resulted was a high complication rate without the benefit of improved local control or survival. Peckham et al.‘* using scintillation counter to monitor bladder and rectal doses, similarly gave unrestricted doses to the parametria and disregarded the point A dose of Tod and Meredith.‘5v’6 They concluded that the complication rate was unacceptable and no significant additional survival benefit was observed over the control dose method. Previously published data on template dosimetry suggested that the classical anatomical points of Tod’s A and B are insignificant for dose calculation using this perineal template”; instead, consideration is given to the homogenous maximum, minimum and average tumor dose within a given tumor volume. However, a literature search suggests that to date no published guidelines are established indicating dose limitations for the bladder, rectum or medial parametrium; nor is there recorded a maximum or average homogeneous tumor dose per unit tumor volume beyond which an intolerable complication rate exists. The initial assumption that parametrial and pelvic wall doses could be pushed beyond established guidelines because of the small treatment volume, as long as the bladder and rectal doses limitations were observed, has proved to be incorrect. Since recognition of these treatment related difficulties, template brachytherapy at LAC-USC Women’s Hospital has and is continuing to undergo corrective adjustments in patient selection, loading techniques, seed activity selection and dose rate control.

REFERENCES 1. Buchler, D.A., Kline, J.C., Peckham, B.M., Boone, M.L.M., Carr, W.F.: Radiation reaction in cervical cancer therapy. Am. J. Obstet. Gynecol. 111: 745-750, 1971. 2. Classification and staging of malignant tumors in the female pelvis. Acta. Obstet. Gynec. Stand. SO: 1-7, 197 1. 3. Einhorn, N.: Frequency of severe complications after radiation therapy for cervical carcinoma. Acta. Rad. Ther. 14: 42-48,1975. 4. Feder, B.H., Syed, A.M.N., Neblett, D.: Treatment of extensive carcinoma of the cervix with the “transperineal parametrial butterfly”--A preliminary report on the revival of Waterman’s approach. Int. J. Radiat. Oncol. Biol. Phys. 4: 735-742,1978. 5. Fleming, P., Syed, A.M.N., Neblett, D., Puthawala, A., George III, F.W., Townsend, D.: Description of an aftercavitary technique in the loading Ir19* interstitial-intra treatment of carcinoma of the vagina. Obstet. Gynecol. 55: 525-530.1980.

6. Fletcher, G.H.: Textbook of Radiotherapy. 2nd edition. Philadelphia, Lea and Febiger, 1973. 7. Fletcher, G.H., Wharton, J.T.: Principle of irradiation therapy for gynecologic malignancies. Curr. Prob. Ob/Gyn. 2: 2-43, 1978. 8. Homesley, H.D., Raben, M., Blake, D.D., Ferree, C.R., Bullock, M.S., Linton, E.B., Greiss, F.C., Jr., Rhyne, A.L.: Relationship of lesion size to survival in patients with stage IB squamous cell carcinoma of the cervix uteri treated by radiation therapy. Surg. Gynecol. Obstet. 150: 529-531, 1980. 9. Kagan, A.R., Nussbaum, H., Gilbert, H., Chan, P.Y.M., Rao, A., Saltz, A., Tawa, K., Wollins, M.: A new staging system for irradiation injuries following treatment for cancer of cervix uteri. Gynecol. Oncol. 7: 166-l 75, 1979. 10. Kurohara, S.S., Di Saia, P., Kurohara, J., Grossman, I., George III, F.W., Morrow, C.P.: Uterine cervical cancer: Treatment with megavoltage radiation results and after-

Interstitial brachytherapy 0 0.

11.

12.

13.

14.

loading intracavitary techniques. Am. J. Roentgenol. 133: 293-297, 1979. Neblett, D.L., Harrop, R., Haymond, H.R., Koval, J.M., Syed, A.M.N.: Interstitial-intracavitary (Syed/Neblet applicator) dose-rate distribution plots. LAC/USC Med. Ctr. USCSch. o/Med. Publ. 1: 1-24, 1978. Peckham, B.M., Kline, J.C., Schultz, A.E., Cameron, J.R., Vermund, H.: Radiation dosage and complications in cervical cancer therapy. Am. J. Obstet. Gynecol. 104: 485-494, 1969. Strockbine, M.F., Hancock, J.E., Fletcher, G.H.: Complications in 83 1 patients with squamous cell carcinoma of the intact uterine cervix treated with 3,000 rads or more whole pelvis irradiation. Am. J. Roentgenol. 108~293-304, 1970. Syed, A.M.N., Feder. B.H.: Technique of after-loading interstitial implants. Radiol. Clin. 46: 458-475, 1977.

GADDIS

JR. et al.

827

15. Ted, M.C., Meredith, W.J.: A dosage system for use in the treatment of cancer of the uterine cervix. Brit. J. Rad. 11: 809-824,1938. 16. Tod, M., Meredith, W.J.: Treatment of cancer of the cervix uteri-A revised “Manchester Method.” Brit. J. Rad. 26: 252-257,1953. 17. Van Nagell, J.R., Jr., Parker, J.C., Jr., Maruyama, Y., Utley, J., Luckett, P.: Bladder or rectal injury following radiation therapy for cervical cancer. Am. J. Obstet. Gynecol. 11% 727-732,1974. 18. Villasanta, U.: Complications of radiotherapy for carcinoma of the uterine cervix. Am. J. Obstet. Gynecol. 114: 717-726,1972. 19. Young, M.E.J., Batho, H.F.: Dose tables for linear radium sources calculated by an electronic computer. Brit. J. Rad. 37: 38&44,1964.