Radiation therapy alone in the treatment of carcinoma of uterine cervix I. Analysis of tumor...

Radiation Therapy Alone in the Treatment of Carcinoma of the Uterine Cervix

11. Analysis of Complications

CARLOS A. PEREZ, MD,* SHERRY BREAUX, MPH,' JOHN M. BEDWINEK, MD,' HYWEL MADOCJONES, MD, PHD,t H. MARVIN CAMEL, MD,+

JAMES A. PURDY, PHD." AND BRUCE J. WALZ, MD*

A retrospective analysis was carried out on 81 1 patients with histologically proven invasive carcinoma of the uterine cervix treated with irradiation alone. A correlation was made of the doses of irradiation delivered to the pelvic organs with external beam and intracavitary insertions. Approximately 3% of the patients exhibited grade 2 gastrointestinal complications, and 2% developed grade 2 urinary complications; 5% of the patients developed grade 3 gastrointestinal complications, and 3% developed grade 3 urinary complications. Other types of complications, primarily grade 2, such as vaginal necrosis, pelvic abscess, thrombophlebitis, etc, were seen in approximately 5% of the patients. Thus, the total percentage of patients developing grade 2 complications was 10% and grade 3 complications, approximately 8%. About 25% of the patients who had complications showed more than one sequela. The most frequently observed grade 2 complications were proctitis, cystitis, vaginal stenosis, and partial small bowel obstruction which were treated with conservative management. Grade 3 complications required surgical treatment and consisted most frequently of ureteral stricture, vesicovaginal fistula, rectovaginal fistula, sigmoid stricture, small bowel obstruction, proctitis, and large rectal ulcers. The most significant factor affecting the appearance of complications was the total dose of irradiation delivered to the pelvic organs by the whole pelvis external irradiation and intracavitary insertions. With maximum total doses up to 8000 rad the incidence of grade 2 and 3 complications was less than 5%. However, with higher doses the incidence of complications increased to 10% to 15%. In patients receiving total doses of 6000 rad to the bladder or rectum, more complications were noted when only one intracavitary insertion was performed, as compared with two or three. Eighty percent of the rectosigmoid complications occurred within 30 months of initial therapy, in contrast to 48 months for the urinary complications. Patients who developed complications had survival rates comparable to those without complications. This underscores the need to rapidly institute treatment on patients who have severe injury after radiation therapy. Even though it is difficult to determine the exact total dose delivered to a specific volume within the pelvis, the current study strongly indicates that dose calculations to specific anatomical points may be reliable parameters to use in modifying treatment techniques to deliver doses of irradiation that will not exceed tolerance limits for the pelvic structures, when treating patients with carcinoma of the uterine cervix with irradiation alone.

Cancer 54:235-246, 1984.

RRADIATION is the main modality of therapy in patients with Stage 113, 111, and IV carcinoma of the uterine

cervix; for those with Stage I and IIA, comparable results are obtained with either irradiation or radical hysterectomy. It is imperative that the effectiveness of a treatment

modality be assessed not only by its ability to control the tumor but also by its toxicity.

This report evaluates the incidence and types of complications in 8 1 1 patients with invasive carcinoma of the uterine cervix treated with definitive radiotherapy alone

I

From the *Division of Radiation Oncology, Mallinckrcdt Institute of Radiology, Washington University School of Medicine, St. Louis, Missouri, Whe Department of Therapeutic Radiology, Tufts New England Medical Center, Boston, Massachusetts, and the $Department of Ob- stetrics and Gynecology, Washington University School of Medicine, St. Louis, Missouri.

Address for reprints: Carlos A. Perez, MD, 45 I 1 Forest Park, Suite 41 1, St. Louis, MO 63108.

The authors thank the following persons for their valuable conin- butions: the staff of the Department of Obstetrics and Gynecology for participation in treatment of patients and providing follow-up information, Ms. Mary Ann Hederman for data collection, Ms. Alice Becker for preparation of the manuscript, and Ms. Leslie MacConnell-Clubbs for graphic work.

Accepted for publication April 29, 1983.

235

236 CANCER July I5 1984 Vol. 54

at the Mallinckrodt Institute of Radiology between Jan- uary 1959 and December 1977. The complications are analyzed relative to techniques and doses of irradiation.

Materials and Methods

The authors camed out a retrospective analysis of the records for all patients with histologically proven carcinoma of the uterine cervix, Stages I, 11, 111, and IVA treated with definitive radiation therapy. Six patients ex- plored in preparation for a radical hysterectomy, who were found to have metastatic periaortic lymph nodes, were treated by irradiation alone and are included in this analysis.

All patients were initially evaluated with a physical and pelvic examination, routine blood counts, blood chemistry profile, chest x-ray, and intravenous pyelogram. Multiple biopsies of the cervix were done on all patients, and a dilatation and curettage was frequently carried out. In patients with stages more advanced than IIA or in those with a history of gastrointestinal disturbances a barium enema was performed. Lymphangiogram was done on selected patients with Stages IB and IIA and on most patients with more advanced stages. All patients were staged jointly by the staffs of the Division of Radiation Oncology and Department of Gynecological Oncology according to the FIG0 classification.' When there was a disagreement, the patient was assigned to the less advanced stage. Patients with barrel-shaped cervix, endometrial extension of cervical carcinoma or periaortic lymph node metastases were classified without modification of the clinical stage ascertained in pelvic examination.

Patients were treated with a combination of external irradiation and two (sometimes three) intracavitary insertions. All patients were followed for a minimum of 3 years (until death or until lost to follow-up). Median follow-up time is 8.8 years. Information was obtained in 97% of the patients either from our own records or by letter or telephone contact with the referring physicians and occasionally through direct communication with the patients or relatives.

All complications of treatment were carefully assessed. They were classified according to degree of severity as:

Grade 1: Minor symptoms, self-limited or responding to simple outpatient management, mainly diar- rhea and bladder and rectal irritation which developed during treatment and subsided within 4 weeks of completion of therapy.

Grade 2: Major symptoms, repeated occurrences which often required hospitalization for diagnosis and for nonsurgical management. These included: moderate proctitis or sigmoiditis, rectal ulcer, small bowel obstruction treated medically, malabsorption syndrome, chronic cystitis, partial ureteral stricture, etc.

Grade 3: Complictions which required major surgery for correction or were life threatening, such as pulmonary embolus. These included: rectovaginal fistula, sigmoid or small bowel perforation, rectal, sigmoid or small bowel obstruction treated surgically, vesicovaginal or ureterovaginal fistula, rectal or bladder ulcer, hemorrhagic cystitis, or ureteral stricture requiring surgical correction.

The radiation therapy and other hospital records were reviewed for all patients, and the data were coded on computer-compatible forms. The radiation therapy doses, including external beam and intracavitary contributions, were recalculated in all patients. In 50% of the patients all the implant films and isodose curves of the radioactive source insertions were available for review and the doses at several points of interest were determined. In the re- maining patients on whom the original implant films were not available for review, the doses at the same points of interest were estimated using the records in the chart and dose levels from computer-generated isodose distributions with the loading and geometry similar to that described for the insertions performed in the individual patients (Figs. 1A and 1B). The correlation of grade 2 and 3 complications with irradiation doses were separately analyzed in the two groups of patients, depending on the completeness of the dosimetry data calculated or estimated tumor doses. Since the incidence of complications was approximately the same, the two groups were merged and correlations of doses of irradiation with complications were generated in the total population, according to the clinical stage.

The maximum dose of irradiation contributed by the external beam and the intracavitary insertions to the bladder and rectum was calculated on a sagittal plane at two points (Fig. 2):

1. The bladder dose at the level of trigone, as indicated by a Foley catheter placed in the bladder at the time of insertion or 2 cm anterior to the tandem, 1 or 2 cm above the vaginal fornices.

2. The rectal dose was calculated 2 cm posterior to the midportion of the tandem, and 1 cm above the vaginal ovoids.

We are aware that from a biological standpoint there is a difference in cellular effects between continuous low- dose-rate brachytherapy and high-dose-rate fractionated external irradiation used in the treatment of these patients. (With low-dose-rate brachytherapy the oxygen enhance- ment ratio is lower; there is more interference with repair of sublethal damage, and probably more cells are in ra- diosensitive phases of the cell cycle.) However, since all of our patients were treated with basically the same techniques, we believe it is reasonable to add the contribution from both modalities without further corrections.

No. 2 IRRADIATION COMPLICATIONS IN CERVICAL CA - Perez et al. 237

FIGS. 1A AND IB. AP (A, left) and lateral (B, right) radiographs of the pelvis with afterloading Fletcher tandem and ovoids in place, showing the isodose distribution in coronal and sagittal planes.

Actuarial survival was calculated according to the life- table method of Cutler and Ederer.’ Patients lost to follow- up were withdrawn at that time from the risk group.

All data were processed and analyzed on a VAX 1 1/780 computer using the BMDP-79, Biomedical Com- puter Programs, P-Series (Digital Corporation, May- nard, MA).3

Methods of Treatment The policies of treatment at Washington University

were previously described in detail.4 Patients with Stage IB and IIA were treated with a combination of external irradiation (1000-2000 rad whole pelvis and an additional parametrial dose with a midline block to deliver a total of 4000-5000 rad to the lateral parametria) and two intracavitary insertions for a total of 6000 to 7500 mgh. Patients with Stages IIB and I11 were treated with a combination of external irradiation (2000-4000 rad whole pelvis and additional parametrial dose to delive: 5000- 6000 rad to the external iliac nodes) and two intracavitary insertions for a total of 7000 to 8000 mgh (Figs. 3A and 3B). Patients with Stage IVA were treated with 4000 rad whole pelvis and an additional 2000 rad to the parametria combined with one or two intracavitary insertions for 5000 to 6000 mgh.

External irradiation was administered with 22 or 25 MV photon beams through 15 X 15 or 15 X 18 cm anteroposterior (AP) and posteroanterior (PA) ports. In

September 1974 the field size was changed to 15 X 18 cm for Stages IIB, 111, and IVA to include a greater volume of the common iliac lymph nodes. Central axis doses were calculated at the midplane of the pelvis. The daily dose was 180 to 200 rad daily, five fractions per week.

In the majority of patients the first intracavitary insertion was done after 1000 to 2000 rad tumor dose (TD)

BLADDER

FIG. 2. Schematic representation of points at which dose to bladder or rectum was calculated utilizing sagittal dose distributions. Bladder and rectum dose = mgh X 0.8.

238 CANCER July 15 1984 Vol. 54

PLANE THROUGH LOWEST TANDEM SOURCE x 1000 RAD

x 1000 RADS

Rcs. 3A AND 3B. Cross section of the pelvis demonstrating dose distributions obtained with combination of intracavitary insertions and external irradiation (whole pelvis and split fields with midline shield).

were delivered to the pelvis; the second insertion was performed 2 weeks later; when necessary, a third insertion was carried out at the completion of the external therapy.

Until 1965 the Washington University Ter-Pogossian applicator was used, but this was replaced by the Fletcher and later the Fletcher-Suit afterloading applicator. Cobalt 60 sources were used for the intracavitary insertions until 1973, at which time they were replaced by 13’cesium. Intracavitary doses were routinely calculated in milligram hours (mgh) radium equivalent. The dose rate at 2 cm from the sources was approximately 60 to 80 rad/hour.

Table 1 summarizes the average doses delivered by external irradiation or intracavitary insertions to various anatomic sites in the pelvis. There was a trend toward somewhat higher doses delivered with external irradiation, particularly to the lateral pelvic wall. The small increases in the average dose to the bladder or rectum in each treatment period is not statistically significant.

After whole pelvis irradiation was completed, a step wedge block constructed according to the configuration of the intracavitary dose distribution was used to shield the midline in two thirds of the patients. In the remainder of the patients a 3-cm rectangular midline block was em- ployed.

From 1959 to 1964 there was a period of development of new techniques. Initially, a step wedge was designed following the configuration of an intracavitary insertion with tandem and vaginal ovoids. The lower portion of the wedge was truncated (Fig. 4A), and later it became apparent that the tissues receiving the entire dose of external irradiation in addition to the contribution from the intracavitary insertion were being subjected to higher doses than initially intended. This caused an increased incidence of complications until 1963, at which time the technique was modified to include a lower block to shield the bladder and the rectum (Fig. 4B). In 1965 and in 1970 additional wedges were designed to conform more closely to the configuration of the intracavitary insertions (Fig. 4C).

Because of this variation in technique, the complications of therapy were analyzed separating these periods. The overall incidence of grade 3 complications in the period 1959 to 1964 was 12.9%; in 1965 to 1969, 13.6%; and in the period from 1970 through 1977, 5.4%. This difference is statistically significant (P = 0.006) and could be related to greater sophistication in treatment techniques, better integration of the intracavitary, and external therapy and increasing individualization of therapy. Ad- mittedly, some of these patients have been followed for only 3 years, and other complications may appear later. However, 80% of the rectosigmoid complications developed within 3 years from initial treatment and 80% of the urinary complications within four years. Thus, a longer

TABLE 1 . Carcinoma of the Uterine Cervix: Irradiation Alone in All Stages

Yr of treatment (mean doses in rad)

Site of dose calculation 1959-1964t 1965-1969g 1970-1977 I(

Whole pelvis 1320 1492 2014 Lateral pelvic wall 309 1 3904 5083

6366 6733 6784 6623 * 1492 7042 k 1322 7387 f 1403

Mgh Bladder*.? Rectum*.? 6583 ? 1526 6969 f 1291 7568 * 1298

Mallinckrodt Institute of Radiology, 1959-1977. * Mean f 1 standard deviation.

radiation and brachytherapy (rad). Values include dose contributions from whole pelvis external ir-

$ 194 patients. 6 227 patients. 11 390 patients. Mgh: milligram hours.

No. 2 IRRADIATION COMPLICATIONS IN CERVICAL CA . Perez et al. 239

FIGS. 4A-4C. (A, left) Initial step wedge used at MIR to shield midportion of pelvis. The truncated base allowed external irradiation to be delivered to tissues where high dose from intracavitary sources was received. (B, center) Truncated step wedge with rectangular block to shield the midline structures. (C, right) Revised design of step wedge, covering the previously unshielded area in the initial wedge.

follow-up should not result in a significantly higher incidence of complications.

Since grade 1 injuries are considered predictable re- actions of the pelvic organs to irradiation and practically all these symptoms subside with conservative therapy in a matter of weeks, only grade 2 and 3 complications are analyzed in detail.

Results

Incidence and Type of Complication

plications in the various tumor stage groups. Approxi- mately 3% of the patients exhibited a grade 2 gastrointestinal complication, 2% developed a grade 2 urinary complication, and approximately 5% developed other types. About 5% of the patients developed grade 3 gastrointestinal complications, and 3% developed grade 3 urinary complications.

Seventy-five percent of the patients developing complications had only one organ affected, whereas in the other 25% two or even three complications were noted (for instance, combination of vesicovaginal and recto- -

Table 2 summarizes the percentage of patients developing gastrointestinal, urinary, or other types of com-

vaginal fistulae). Thus, the number of complications is higher than the number of patients.

TABLE 2. Carcinoma of the Uterine Cervix: Irradiation Alone and Complications of Treatment (Grades 2 and 3)

No. of patients with

tions* complica- Incidence of complications

Intestinal Urinary Other No. of Seventy

patients grade Severity grade Severity grade Severity grade Stage treated 2 2 3 2 3 2 3

IA 26 3 1 1 (3.8%) 1 (3.8%) - - 2 (7.7%) -

IIA 86 7 5 1 (1.2%) 5 (5.8%) 2 (2.3%) 2 (2.3%) 6 (7%) -

IIIA 10 1 - - - - - 1 (10%) -

IVA 14 1 1 - - - 2 (14.3%) 1 (7.1%) - Total 81 1 64 56 23 (2.8%) 42 (5.2%) 16 (2%) 26 (3.2%) 42 (5.2%) 9 ( 1 . 1 % )

IB 277 22 18 6 (2.2%) 10 (3.6%) 2 (0.7%) 10 (3.6%) 20 (7.2%) 2 (0.7%)

IIB 215 14 21 7 (3.3%) 19 (8.8%) 9 (4.2%) 7 (3.3%) 4 (1.9%) 6 (2.8%)

IIIB 183 16 10 8 (4.4%) 1(3.8%) 3 (1.6%) 5 (2.7%) 8 (4.4%) 1 (0.5%)

~ ~~

Mallinckrodt Institute of Radiology, 1959-1977. * Patients with both grade 2 and grade 3 complications are counted

in grade 3 column. Stage IB, 5 ; Stage IIA, 1; Stage IIB, 4; Stage IIIB, 2.

240 CANCER July 15 1984 VOl. 54

TABLE 3. Carcinoma of the Uterine Cervix: Irradiation Alone and Type of Grade 2 Complications

Stage

IA IB IIA IIB IIIA IIIB IVA

Total no. of patients treated No. of complications Percent

Intestinal Proctitis Rectal ulcer Sigmoid stricture Small bowel obstruction Malabsorption

Chronic cystitis Ureteral stricture Incontinence Ureteral stricture

Urinary

Other Pelvic abscess Pulmonary embolus Vaginal vault necrosis Leg edema Vaginal stenosis Thrombophlebitis

26 277 3 28

1 1 10.1

I 4

3

2 17

86 215 9 20

10.5 9.3

5

1 1

1

2 6 1 1 I

1 2 1

1 4

1

10 I83 14 1 19 1

10 10.4 7. I

3

1

2

1 5 1

Mallinckrodt Institute of Radiology, 1959- 1977.

Grade 2 complications are summarized in Table 3. They were noted in approximately 10% of the patients. The most frequent complications of this type were proctitis, which was noted in 13 patients, chronic cystitis seen in 13 patients, and vaginal stenosis observed in 30 patients. Partial small bowel obstruction which improved with conservative management, including decompression with a nasogastric tube but requiring no surgical procedure, was present in six patients.

Grade 3 complications are summarized in Table 4. They were observed in about 8% of the patients. The most frequent injuries were ureteral stricture (12), rectovaginal fistula (7), sigmoid stricture (8), small bowel obstruction (9), vesicovaginal fistula (9), proctitis (6), and rectal ulcers (3).

Factors Afecting the Incidence of Complications

Age: Age showed no significant impact in the incidence of gastrointestinal or urinary complications. This may be partially related to the fact that patients older than 65 years of age were treated generally with slightly lower doses (5%-10%) of intracavitary and external irradiation.

Previous pelvic surgery or pelvic inflammatory disease: In our experience the history of previous pelvic surgery did not significantly correlate with the incidence of complications. Likewise, patients with a history of pelvic inflammatory disease had a 15% incidence of grade 2 or 3

complications in comparison with 15.2% in those patients without such a history. The differences are not statistically significant. Again, this may be related to individualization of therapy, since the total dose was reduced by 5%-10% in the patients with a “high risk history” for complications.

Dose of external and intracavitary irradiation: Tables 5 and 6 show a correlation of three different dose levels to the rectum or bladder with the incidence of grade 2 and 3 complications in the rectosigmoid and urinary tract. There is a slight increase in the incidence of complications in patients receiving whole pelvis doses in the range of 3000 to 3500 rad when the total dose to the pelvic organs was less than 6000 rad. However, the difference is not statistically significant. Patients treated with higher dose levels showed no significant impact of increasing dose of whole pelvis irradiation in the incidence of complications. Further, it should be noted that those patients receiving over 8000 rad total dose to the bladder or rectum have a higher incidence of complications than those treated with lower doses, irrespective of the whole pelvis dose. Patients receiving over 3500 rad to the whole pelvis showed one complication, but as noted in the footnote of the tables, this could be due to the fact that few of these patients survived for over 2 years.

Intracavitary insertions: The performance of only one insertion in the patients receiving doses lower than 6000 rad to the rectum or bladder correlated with a higher incidence of complications (10% in rectosigmoid and 4%

No. 2 IRRADIATION COMPLICATIONS IN CERVICAL CA Perez et al. 24 1

TABLE 4. Carcinoma of the Uterine Cervix: Irradiation Alone and Type of Grade 3 Complications

Stage

IA IB IIA IIB IIIA IIIB IVA

Total no. of patients treated 26 277 86 215 10 I83 14

Percent 3.8 7.9 8.1 14.9 - 7.1 14.3 No. of complications 1 22 7 32 - 13 2

Intestinal Rectovaginal fistula Sigmoid perforation Small bowel perforation Proctitis Rectal ulcer Sigmoid stricture Small bowel obstruction Other GI

Cystitis Bladder ulcer Vesicovaginal fistula Ureteral stricture Other GU

Urinary

2 5 1 3

1 2 2

3 3 I 3 1 4

1 1 2

1 2 1 3 2 4 2 4

3 2

Other Pulmonary embolus Pelvic hemorrhage 1 Pelvic abscess 1 1 Arteriosclerosis 2 Other 1 2

Mallinckrodt Institute of Radiology, 1959- 1977.

in the bladder or ureter). Less than 0.5% of the patients treated with two or three insertions and the lower dose developed complications. This difference is of borderline statistical significance for the rectosigmoid complications (P = 0.08) and for urinary complications (P = 0.1). No differences were observed with higher doses when one or more insertions were performed.

Types of intracavitary insertions: The use of tandem and ovoids, tandem with protruding vaginal source or interstitial therapy did not significantly affect the incidence

TABLE 5. Carcinoma of the Uterine Cervix: Irradiation Alone and Correlation of Grade 2 and 3 Rectosigmoid

Complications With Doses of Irradiation

G I gastrointestinal; GU: genitourinary.

1

of complications in the various stages. The total doses were decreased 10% to 15% in patients treated with tandem and protruding vaginal source or needles; this may account for the lack of correlation of these parameters.

The quality of the intracavitary insertion (position of applicators) was found to have no impact on the incidence of complications.

Type of midline shielding block: Step-wedge blocks were used in approximately 50% of the patients and 3 cm (on the tray) wide rectangular blocks were used in the other

TABLE 6. Carcinoma of the Uterine Cervix: Irradiation Alone and Correlation of Grade 2 and 3 Urinary Complications

With Doses of Irradiation

Total dose to rectum (rad-cGy) Whole-pelvis

dose <6000 6000-8000 >8000

Total dose to bladder (rad-cGy) Whole-pelvis

dose <6000 6000-8000 > 8000

0 0130 018 o/ 1 <1500 3/52 (5.8%) 71153 (4.6%) 3/27 (11.1%)

1501-2500 4/84 (4.8%) 12/257 (4.7%) 14/116 (12.1%) 2501-3500 111 I (9.1%) 1/22 (4.5%) 3/13 (23.1%)

>3500 0/7* 1/15 (6.7%)t 0/14$

Mallinckrodt Institute of Radiology, 1959- 1977. Irradiation: external whole-pelvis and intracavitary.

* Five died before 2 yr. t Four died before 2 yr.

Eight died before 2 yr; two died at 3-4 yr.

0 0126 011 1 012 <1500 2/58 (3.4%) 6/153 (3.9%) 3/21 (14.3%)

1501-2500 1/99 (1%) 12/244 (4.9%) 15/114 (13.2%) 2501-3500 117 (14.3%) 1/27 (3.7%) 1/12 (8.3%)

>3500 0182 1112t 01 16t

Mallinckrodt Institute of Radiology, 1959- 1977. Irradiation: external whole-pelvis and intracavitary.

* Six died before 2 yr.; one died at 2 yr.; 6 mo. t Three died before 2 yr. $ Eight died before 2 yr.; two died at 3 yr.


H Rectosigmid Complications *--* GU Complications .- -. small Bowel compllcntlons

/ /

a

2 3 W

FIG. 5. Graph depicting chronological appearance of complications in the rectosigmoid, bladder, or ureter and the small intestines (grades 2 and 3).

half of the patients. The incidence of grade 2 and 3 complications is comparable in both groups (approximately 15%).

Time of Appearance of Complications

Figure 5 depicts the time of appearance of grade 2 and 3 rectosigmoid or urinary complications in all patients. (Other complications are not included.) Eighty percent of the rectosigmoid complications were observed within 30 months from initial therapy as opposed to 48 months for the urinary complications. The grade 3 complications appeared earlier (4-6 months) than the grade 2 which may be related to the higher doses of irradiation delivered to the damaged organs.

Impact of Appearance of Complications on Incidence of Pelvic Recurrence and Survival

Patients who developed no complications after treatment had a slightly higher incidence of pelvic recurrences (except in Stage IIA), perhaps reflecting lower tumor doses. The differences however are not statistically significant (Table 7).

Correlation of Complications With Actuarial Survival

Figures 6A and 6B illustrate the actuarial survival in patients without complications (Fig. 6A) or with grade 2-to-3 complications (Fig. 6B). In Stages IB, IIA, and IIB, the 5-year actuarial survival is approximately the same in both groups (differences are not statistically significant). However, in Stage 111, 60% of 28 patients with grade 2- to-3 complications survived 5 years in contrast to only 35% of 155 patients without complications. This may be related to the higher doses of irradiation, better tumor control, and longer survival observed in patients developing major complications.

Correlation of Total Doses of Irradiation and Incidence of Major Complications

The correlation of doses of irradiation to specific organs in the pelvis and complications is somewhat uncertain, because of the difficulty in superimposing the doses of irradiation over the entire organ (volume) and in assessing the exact dose of irradiation (homogeneity of dose) that was delivered to the damaged portion of the pelvic viscera. However, an attempt was made to correlate the maximum dose delivered over an area larger than 12 cm2 in the rectum or the bladder and the incidence of complications. Also, it was assumed that the dose delivered to the distal ureter was approximately the same as that of the trigone of the urinary bladder. The small bowel was presumed to receive a dose equal to that of the lateral parametrium. Obviously, if there was fixation of small bowel loops to the pelvis, the contribution from the intracavitary therapy would be higher. Thus, the following correlations of dose and complications must be evaluated cautiously, keeping in mind the limitations of the methods of computation available.

Figure 7 shows a correlation of the maximum dose of irradiation calculated in the bladder or the rectosigmoid and incidence of complications. It is noted that with doses below 8000 rad about 5% of the patients developed either rectosigmoid or genitourinary complications. However,

TABLE 7. Carcinoma of the Uterine Cervix: Irradiation Alone and Correlation of Frequency of Pelvic Recurrences and Presence of Complications of Treatment

Incidence of pelvic recurrences

IA IB IIA IIB IIIA IIIB IVA ~~ ~

No complications 0122 15/237 (6.3%) 6/74 (8.1%) 30/180 (16.7%) 319 (33.3%) 63/157 (40.1%) 10112 (83.3%)

Complications 1/18 215 312 I - 1/10 o/ 1

2/14 O/ 1 6/16 o/ 1

complications 0/4 1/40 (2.5%) 3/12 (25%) 5/35 (14.3%) 011 7/26 (26.9%) 012

Grade 3 01 1 Grade 2 013 0122 117

Total no. of patients with

Mallinckrodt Institute of Radiology, 1959- 1977. Differences are not statistically significant.


' " O F

(22 Patients)

(236 Patients)

6/40 - GU Complicaiions 9

15( --- Reciosigmoid Complications

* = = = = = = -

(74 Patients)

(180 Patients1

a UJ -A--

I- Z I 't (155 Patients) I 4ii99 W

n 2ot L (9 Patients)

I I I I I 7000 8000 9000 ?I0000 56000

DOSE TO ORGAN (Radl

FIG. 7. Incidence of GU (bladder and ureter) or rectosigmoid complications as a hnction of dose delivered to those organs (grades 2 and 3). 3

OO 1 2 3 4 5 YEARS AFTER INITIAL THERAPY

dality, it is important to determine not only the probability of tumor control but also the morbidity of the therapy.

For many years a great deal of effort has been applied to determine the doses to the bladder and rectum in an effort to decrease the incidence of complications.

Strockbine and associates5 stated that because of the inverse square law, the dose delivered by brachytherapy cannot be simplified by calculating it at a given point. From our retrospective study and other reports it appears that doses that are calculated at specific points, utilizing computed isodose curves may be useful in determining the management of individual patients, and they appear to show a strong correlation with incidence of pelvic recurrences and complications. It is obvious that the total volume receiving a certain dose cannot be ascertained, but evolving technology will allow for tridimensional dose computations in the foreseeable future. Increasing use of computed isodose distributions has permitted a better understanding of the doses of irradiation delivered to specific volumes. A variety of planar or point calculations have been designed to assess the doses received by the pelvic structures.

(40 Patients)

[L 3 UJ $ * O t

(4 Patients)

(I2 Patients)

(26 Patients) (35 Potients)

a

H Stage I A H Stoge IB H Stage nA

V A-A Stage EB H Stage I U A '

n B-W SlageIIIB

1 I 1 1 1 1 2 3 4 5

YEARS AFTER INITIAL THERAPY

FIGS. 6A AND 6B. (A, top) Actuarial overall suMval of patients without complications. (B, bottom) Actuarial overall survival of patients who developed grade 2 or 3 complications.

with higher doses the incidence of complications increases to about 15%. This difference is statistically significant (chi-square = 8.76; degree of freedom [dfl = 1, P = 0.003).

The dose of irradiation to the lateral pelvic wall is correlated with incidence of major complications of the small bowel in Figure 8. With doses below 5000 rad the complication rate is about 1%, but it is somewhat higher with larger doses.

H GRADE 2-3

Discussion

It is well known that irradiation injury of normal tissues depends on the organ under consideration, the dose of irradiation, fractionation, and the volume treated. In assessing the effectiveness of a particular therapeutic mo-

W a 01 I I I 0 15000 6000 27000

DOSE LATERAL PELVIC NODES (RAD)

FIG. 8. Grade 3 (closed squares) and total complications of grades 2 to 3 (closed circles) of the small intestines as a function of dose.

244 CANCER July 15 1984 Vol. 54

Fletcher and coworkers6 reported initially on techniques directly measuring the radiation with dosimeters placed in the bladder and the rectum. He discontinued this technique because the measurements did not correlate well with the incidence of injuries. The measured doses were highly dependent on the position and the sensitivity of the probe in relation to the radioactive sources. Cun- ningham and associates’ described a technique inserting a plastic tube containing thermoluminescent dosimeters (TLD) in the rectum with the applicators and sources in treatment position. The authors pointed out that the un- certainty in the localization of the lithium fluoride dosimeters on the radiographs and the difficulty of inserting the plastic tube into the high-dose region limited the application of this direct dosimetry on a routine basis. Dis- crepancies of 10% to 20% were found between the in vivo rectal dosimetry and the computer-calculated doses.

Further, Hall8 and Ellis and Sorensen’ have pointed out that it is difficult to add the biological effects of radiation when it is delivered with brachytherapy (2-3 rad per minute) or external irradiation (100-500 rad per minute). In our analyses, since the patients were treated with similar techniques the tables and graphs indicate the doses ofirradiation in rad delivered with either modality without consideration for possible differences in the biological effect of dose rate.

Many authors have identified patients with a higher probability of developing complications following radiation therapy. These include elderly patients, and those with a history of pelvic inflammatory disease or previous pelvic surgery. Powell-Smith” reported in 18% incidence of GI injury in 12 1 patients with previous pelvic inflammatory disease or prior surgery in contrast to 7% in 197 patients with neither history. Unal and coworkers” observed 17% bowel complications in 4 I patients with pelvic inflammatory disease or previous pelvic surgery, but only to 2.3% in 2 13 patients without “high r isk” factors.

Kottmeier and Gray” in an analyses of 500 patients with invasive carcinoma of the uterine cervix treated with radiotherapy noted that serious bladder injuries were present in 1 1.2% of the patients with a history of salpin- gitis, in contrast to 6.2% of patients without such a history. However, there was no correlation between bladder or rectal injuries with retroflexion of the uterus or previous history of pelvic surgery. They pointed out that this ab- sence of differences in complications may have been ob- scured by the fact that in high risk patients the radium dosage was routinely decreased. Similar observations are present in our analysis, since in general we have reduced the doses of irradiation by 5% to 10% in patients in high- risk groups.

Furthermore, technical factors, such as the position of the intracavitary applicators in the pelvis or the size of the patient have been reported to influence the incidence

of major complications. Powell-Smith ‘O observed a higher incidence of complications in patients treated with large pelvic fields (greater than 13 X 20 cm) or in patients measuring over 20 cm in thickness. On the other hand, Graham and Villalba” and van Nagell and others14 have reported an increased incidence of small bowel complications in underweight patients. It should be emphasized that Powell-Smith’o reported on patients treated with Co- balt-60 units (55 or 80 cm SSD), and therefore the maximum doses were relatively high. However, with high energy photon beams there is better dose distribution, and less exposure (maximum) doses are required to achieve the desired tumor dose in the mid-pelvis. This was corroborated by a randomized study reported by Allt” in patients with Stage IIB and IT1 carcinoma ofthe uterine cervix. Patients receiving external beam irradiation with betatron x-rays had a 3% incidence of major complications as opposed to 15% complications in a comparable group receiving similar doses with Cobalt-60.

Kottmeier and GrayI2 indicated that the frequency of rectal or bladder injury was better correlated with total dosage than with the radium dose alone. These authors observed a 7% to 8% incidence of grade 2 and 3 rectal injuries in patients receiving less than 6000 rad, about 18% in patients receiving 6000 to 8000 rad, and 26% in 23 patients receiving higher doses. The bladder grade 2 and 3 injuries were about 6% to 7% in patients receiving less than 8000 rad, 9.5% in those receiving up to 9000 rad and 29% in patients treated with higher doses. Even though their dosimetry techniques were rudimentary, they included direct readings at the time of the application with a 0.3 ml Sievert ionization chamber facing the biad- der and rectum.

Lee and coworkers16 and Kagan and associate^'^ have correlated high-dose areas in the rectum, bladder, or vagina with the development of clinically significant injuries in these organs. They suggested that these complications can be prevented by modifying the loading of the applicators with the radioactive sources and by changing the time of the intracavitary applications. Maruyama and others’* and Fletcher’’ have advocated a practice which has been used at our institution for the last 15 years, consisting of the use of blank sources in the tandem or differential loading of the tandem or ovoids depending on the dose distribution shown by computer calculations.

Cunningham and associates’ failed to demonstrate a close correlation between the number of milligram hours of intracavitary therapy administered and the incidence of major injuries to the pelvic organs.

Total doses contributed by both the external and intracavitary irradiation to a given organ have a greater predictive value in correlation with frequency of major complications than the dose delivered by either modality alone. This is an important concept since one must adjust


TABLE 8. Carcinoma of the Uterine Cervix: Policies of Treatment With Irradiation

External irradiation (rad)*

Tumor stage Tumor extent

Additional Brachytherapy parametrial dose (2 insertions)

Whole pelvis (midline shield) (mgh)t

IB (small) Superficial ulceration, less than 2 cm in diameter - 4500 6500-7000

IB (large) Four-quadrant involvement, no endocervical 1000 4000 7000-7500 or involving less than two quadrants

component or significant expansion IIA Not barrel-shape type 2000 3000 7500 IB-IIA (bulky),$ IIB, Barrel-shape cervix, parametrial extension 2000 4000 8000

IIB, IIIB, IV Poor pelvic anatomy, patients not readily treated 4000 2000 6500 IIIA-IIIB

with intracavitary insertions (barrel-shape cervix not regressing, inability to locate external 0s)

Policies of the Mallinckrodt Institute of Radiology. * 180 rad/day, 5 weekly fractions, utilizing 18 or higher million volts

photon beams. t 60-80 rad/hr at point A. In patients over 65 yr, or with history of

previous pelvic inflammatory disease or pelvic surgery, reduce doses

the external beam or brachytherapy doses in order to deliver a tolerable dose of irradiation to sensitive structures.

Hamberger and coworkers" and Unal and associates" respectively analyzed 325 and 254 patients with carcinoma of uterine cervix treated with either intracavitary radium and parametrial irradiation or increasing doses of whole pelvic irradiation combined with intracavitary radium. These authors reported significant increase in complications in the bladder or rectum with higher doses of intracavitary radium and increasing whole pelvis external irradiation doses. They observed an 8.1% incidence of severe complications in patients receiving 4000 rad whole pelvis irradiation plus radium, 10% serious complications in patients receiving 5000 rad whole pelvis irradiation plus radium and 15% complications in patients receiving 6000 rad to the whole pelvis. Unal et al. ( 1 1) described about 10% complications in patients receiving 3500 rad parametrial dose and over 9000 mgh or 2000 rad whole pelvis, over 1500 rad to the parametrium and more than 7500 mgh.

Pourquier and others2' reviewed the dosimetry of external and intracavitary irradiation in 41 rectal and rectosigmoid complications in a group of 287 patients treated. Calculating the maximum and mean rectal doses on the anterior surface of this organ, they concluded that complications increased sharply with cumulative doses above 7000 to 7500 rad. As in our experience, they concluded that the critical dose and threshold region varied as a proportion of the external and intracavitary contributions of the total dose. They also emphasized that external irradiation to the whole pelvis above 4500 rad significantly increased the risk of complications.

by 10%. $ In stage IB and IIA, if complete regression not obtained, do extra-

fascia1 conservative hysterectomy (reduce brachytherapy dose to 6000 milligram hour).

From our past experience, we recommend the policies of treatment outlined in Table 8.

Fletcher'' and Strockbine and associates5 have indicated that fractionating the intracavitary insertions over several weeks improves the tolerance of the normal tissues and allows for regression of central tumors; this confirms our experience with patients treated to a total parametrial dose of less than 6000 rad.

Recurrences after treatment of carcinoma of the uterine cervix, except for a few central ones in patients with Stage IB or IIA disease are practically always fatal. Therefore, it is imperative to deliver doses of irradiation that will provide the maximum tumor control in the pelvis with an acceptable number of complications. It should be pointed out that in our study patients with major complications had fewer pelvic recurrences (although not statistically significant) and exhibited higher survival than patients without complications. This may reflect the better tumor control in patients developing complications, since in general they received higher doses. Further, patients who live longer may have a greater probability of developing complications. This observation refutes the concept that major sequelae of therapy will result in decreased overall survival.

The good prognosis of the patients developing complications underscores the need for rapid and definitive treatment of these unfortunate sequelae, particularly if a surgical procedure is indicated.

REFERENCES

1. Kottmeier H-L. Classification and staging of malignant tumors in

2. Cutler SJ, Ederer F. Maximum utilization of the life table method the female pelvis. J Znt Fed Gynecol Obstet 197 1 ; 9: 172- 179.

in analyzing survival. J Chronic Dis 1958; 8:699-71 I .


3. Dixon WJ, Brown MB, eds. BMDP-79: Biomedical Computer Programs, P-Series. Los Angeles, University of California Press, 1979.

4. Perez CA, Breaux S , Madoc-Jones H ef al. Radiation therapy alone in the treatment of carcinoma of uterine cervix: I. Analysis of tumor recurrence. Cancer 1983; 51:1393-1402.

5 . Strockbine MF, Hancock JE, Fletcher GH. Complications in 831 patients with squamous cell carcinoma of the intact uterine cervix treated with 3000 rad or more whole pelvis irradiation. A m J Roentgenol Radium Ther Nucl Med 1970; 108:293-304.

6. Fletcher GH, Brown TC, Rutledge FN. Clinical significance of rectal and bladder dose measurements in radium therapy of cancer of uterine cervix. Am J Roentgenol Radium Ther Nucl Med 1958; 79:42 1 - 450.

7. Cunningham DE, Stryker JA, Velkley DE, Chung CK. Routine clinical estimation of rectal, rectosigmoidal, and bladder doses from intracavitary brachytherapy in the treatment of carcinoma of the cervix. Int J Radiat Oncol Biol Phys 1981: 7:653-660.

8. Hall ELI. Radiation dose-rate: A factor of importance in radiobiology and radiotherapy. Br J Radiol 1972; 45:81-97.

9. Ellis F, Sorensen A. Method of estimating biological effect of combined intracavitary low dose rate radiation with external radiation in carcinoma of cervix uteri. Radiology 1974; 110:681-686.

10. Powell-Smith C. Factors influencing the incidence of radiation injury in cancer of the cervix. J Can Assoc Radiol 1965; 14:132-137.

11. Unal A, Hamberger AD, Seski JC, Fletcher GH. An analysis of the severe complications of irradiation of carcinoma of the uterine cervix: Treatment with intracavitary radium and parametnal irradiation. Znt J Radiat Oncol Biol Phys 1981; 7:999-1004.

12. Kottmeier H-L, Gray MJ. Rectal and bladder injuries in relation

to radiation dosage in carcinoma of the cervix. A m J Obstet Gynecol 1961; 82:74-82.

13. Graham JB, Villalba RJ. Damage to the small intestine by radiotherapy. Surg Gynecol Obstet 1963; 116:665-668.

14. van Nagell JR Jr, Maruyama Y , Parker JC Jr, Calton WL. Small bowel injury following radiation therapy for cervical cancer. Am J Obstet Gynecol 1974; 118:163-167.

15. Allt WEC. Supervoltage radiation treatment in advanced cancer of the uterine cervix: A preliminary report. Can Med Assoc J 1969;

16. Lee M, Kagan AR, Nussbaum H, Wollin M, Winkely JH, Nor- man A. Analysis of dose, dose-rate, and treatment time in the production of injuries by radium treatment for cancer of the uterine cervix. Br J Radiol 1976; 49:430-440.

17. Kagan AR, DiSaia PJ, Wollin M, Nussbaum H, Tawa K. The narrow vagina, the antecedent for irradiation injury. Gyneco/ Oncol

18. Maruyama Y , van Nagell JR, Wrede DE, Coffey C, Utley JF, Avila J. Approaches to optimization of dose in radiation therapy of cervix carcinoma. Radiology 1976; 120:389-398.

19. Fletcher GH. Squamous cell carcinoma of the uterine cervix. In: Fletcher GH, ed. Textbook of Radiotherapy, ed. 3. Philadelphia: Lea & Febiger, 1980; 720-773.

20. Hamberger AD, Unal A, Gershenson DM, Fletcher GH. Analysis of the severe complications of irradiation of carcinoma of the cervix: Whole pelvis irradiation and intracavitary radium. Int J Radiat Oncol Biol Phys 1983; 9:367-371.

2 1. Pourquier H, Dubois JB, Delard R. Cancer of the uterine cervix: Dosimetric guidelines for prevention of late rectal and rectosigmoid complications as a result of radiotherapeutic treatment. Int J Radiat Oncol Biol Phys 1982; 8:1887-1895.

100:792-797.

1976; 4~291-298.

Radiation therapy alone in the treatment of carcinoma of uterine cervix I. Analysis of tumor...

Documents

Transcript of Radiation therapy alone in the treatment of carcinoma of uterine cervix I. Analysis of tumor...