ABSTRACT

THE VARIABILITY OF APPLICATOR POSITION AMONG HIGH DOSE RATE (HDR) INTRACAVITARY BRACHYTHERAPY APPLICATIONS IN LOCALLY ADVANCED CERVICAL CANCER PATIENTS WHO WERE TREATED WITH RING & TANDEM APPLICATORS

Cuneyt Ebruli; Aysenur Demiral; Ferhat Eyiler; Riza Cetingoz; Munir Kinay
Dokuz Eylül University Hospital; Radiation Oncology Dept.; IZMIR; TURKEY

Aim: HDR applicators are frequently used in the brachytherapy of cervical cancer. The reproducibility of brachytherapy applications among fractions is very important in terms of optimisation of treatment. In this study; the interindividual and intraindividual applicator position variability was evaluated retrospectively in HDR intracavitary brachytherapy applications performed in locally advanced cervical cancer treatment using ring and tandem (R & T) applicators.

Patients and Method: Eight patients with locally advanced cervical cancer formed the study population who had been treated in Dokuz Eylül University Department of Radiation Oncology between the years 2000 and 2005 with HDR brachytherapy using R & T applicators. Patients were treated with 45 Gy external radiotherapy (RT) using pelvic box technique and then they were given intracavitary brachytherapy in 3 fractions of 8 Gy (to point A). The 3–dimensional geometrical variation of applicator center in craniocaudal; mediolateral and anteroposterior directions was determined on the basis of bony reference points in 24 pairs of orthogonal films obtained in the conventional simulator. Then the following evaluations were performed: (1) The applicator position variability in all applications (interindividual variability); (2) The applicator position variability of other two applications relative to the 1st application in a single patient (intraindividual variability relative to the 1st application); (3) The applicator position variability of 3 applications relative to the mean of 3 applications in a single patient (intraindividual variability relative to the mean of 3 applications). In the lateral simulation films; the distances between applicator center and bladder and rectum reference points were also measured. Among the potential factors that could influence the reproducibility of R & T applications; age; stage; the period between external RT and brachytherapy were evaluated with univariate analysis.

Results: The range of R & T applicator center’s position was 13;08 mm; 1;66 mm; and 15 mm in craniocaudal; mediolateral and anteroposterior directions; respectively. Standard deviation (SD) of interindividual applicator variability was 3;83 mm in craniocaudal; 0;39 mm in mediolateral and 2;86 mm in anteroposterior directions. SD of intraindividual variability relative to 1st application was 1;91 mm in craniocaudal; 0;4 mm in mediolateral; and 4;26 mm in anteroposterior directions. And SD of intraindividual variability relative to the mean of 3 applications was 0;95 mm in craniocaudal; 1;86 mm in mediolateral; and 1;24 mm in anteroposterior directions. The lateral angulation of R & T applicator (α) was 10° to the right and 0° to the left while anteroposterior angulation (β) was 30° to the anterior direction and 0° to the posterior direction. SD of the distance between applicator center and bladder and rectum reference points was 1;19 mm and 0;5 mm; respectively. In the univariate analysis; there was no factor influencing applicator position variability.


Conclusion: According to the results of this study; the applicator position variability in R & T applications was relatively less compared to other studies in the literature. In order to extract definitive conclusions about factors affecting positional reproducibility of R & T applicators; studies are needed that evaluate more parameters and that include higher number of patients.

Key words: cervical cancer; radiotherapy; brachytherapy; ring and tandem; applicator position variability; reproducibility of brachytherapy applications.





I. OBJECTIVE

In gynecological brachytherapy; in order to protect the adjacent normal tissue while high dose is applied to the primary tumor site; the convenience of the location of the intracavitary applicator and its immobility is very important.

In various studies (1; 2; 3; 4); it has been reported that the applicator moved and rotated significantly in anteroposterior; mediolateral and craniocaudal direction relative to the bony pelvis and bladder/rectum. Due to this activity occurring in-fraction and between the fractions; the dose of the reference point A and the doses taken up by the normal tissues can change.

Since usually 3 or more applications have been carried out in high dose rate (HDR) brachytherapy; geometric reproducibility of the location of applicator position has a great importance. The studies on geometric mobility of the ring and tandem (R&T) applicators are limited in number (1; 2).

In the present study; it was aimed to analyze the geometric differences between the applications retrospectively investigating the applicator position changes in intracavitary brachytherapy (ICBT) applications of the cases with local regional advanced cervix carcinoma to whom external RT and ICBT were applied according to the cervix cancer therapy protocol prepared by the Dokuz Eylül University Gynecologic Oncology Group (DEJOG) and were treated by using ring and tandem applicators in their ICBTs.


IV. PATIENTS AND METHOD

It was detected that there were 8 cases among the patients with local regional advanced cervix carcinoma who were applied HDR brachytherapy with ring and tandem (R&T) applicators following external RT in the Radiation Oncology Clinic of Dokuz Eylül University (DEU) between 2000 and 2005. Following the termination of external RT (4500 cGy); R&T applicators were placed to the cases in the brachytherapy room; they were taken to the simulator room and their orthogonal X-rays were taken. Following this procedure; the cases were carried to the brachytherapy room again. Brachytherapy applications were carried out in 3 fractions with weekly intervals using R&T applicators with a Nucletron remote control after-loading MicroSelectron® HDR device giving 8 Gy to the reference point A in the Manchester system.

ICBT with R&T applicators was always carried out under epidural anesthesia. Following placement of the applicator; the patient was taken from the lithotomy position to supine position while her legs were extended; it was ensured that she stayed in this position while simulation X-rays were being taken and during the therapy.

Patient and tumor characteristics are shown in Table 1.
Table 1. Patient and Tumor Characteristics
Age
Median 57
Interval ( 36 – 71)

Stage (FIGO)
Stage IIa % 12;5 (n= 1)
Stage IIb % 37;5 (n= 3)
Stage IIIb % 50 (n= 4)

Histological type
Squamous % 100 (n= 8)
Other % 0 (n= 0)
Grade
Grade 1 % 0 (n= 0)
Grade 2 % 100 (n= 8)
Grade 3 % 0 (n= 0)

Method
24 pairs of orthogonal X-rays of ICBT applications of the 8 cases were detected. These simulation X-rays were taken with concentric X-ray device (simulator) in neutral supine position as AP (0°) and lateral (90°); different from placement position.

It was observed that of the ring applicators with a diameter of 26; 30 and 34 mm in the ring applicator set; only the applicators having a diameter of 30 and 34 mm have been used. It was found out that of the intrauterine tandems fixed to the ring applicator; the ones with 60°C gradient and 20; 40 and 60 mm length were used. In all cases; rectal tractors having a gradient of 60°C were used. In all cases; after ring; tandem and rectal tractors were combined; they were fixed to the therapy table by a metal fixing mechanism (Figure 1).

Figure 1. The system that fixates R&T applicators and rectal tractor to the therapy table

Target volume has been determined to include the whole cervix; proximal corpus as far as possible; 1/3 upper part of the vagina and various parts of the parametrium according to the tumor extension; in line with the recommendations of ICRU 38; after considering the patient and tumor anatomy. After the R&T applicators with convenient size and gradient have been applied; reference point A in Manchester system is taken as the reference point where the dose is defined. While the therapy plan to be applied to the patient was being selected; source site and time optimization brought by the “stepping source” facility of the Microselectron® device was called upon as well.

In the present study; a three-dimensional reference system has been defined according to the apparent bony structures seen in the X-rays of ICBT applications in order to evaluate the applicator positions. 24 pairs of X-rays from 8 cases were examined and the central point of applicator has been determined by the lines defined below.

Craniocaudal: A virtual line that joins the vertices of femur heads in the AP (0°) X-ray was created to detect applicator movement in this direction; this line was defined as the reference level to detect movement in craniocaudal direction. If femur heads were not on the same plane; virtual line that passes through the midpoint of the lines that pass from the vertices of both femur heads parallel to the ground plane; were taken as the reference level (Figure 2).

Figure 2. The line that passes through the femur heads



Mediolateral: In order to detect the movement in this direction; middle line that passes through the exact midpoint of pubic symphysis and sacrum of the bony pelvis in the AP (0°) X-rays; was used (Figure 3).

Figure 3. The line that joins middle line of sacrum and pubic symphysis

Anteroposterior: In order to detect the changes in this direction; perpendicular line that descends to the ground plane from the exact midpoint of sacrum promontory and pubic symphysis in the lateral (90°) X-rays; was used (Figure 4).

Figure 4. The line that descends perpendicularly to the ground plane from the midpoint of sacrum promontory and pubic symphysis


In order to detect the applicator position changes in each of the three planes; after the lines were determined in craniocaudal; lateral and anteroposterior planes; the midpoint where long axis of the ring applicator intersects with the tandem was defined as the central point of applicator since it was concordant with the cervical orifice (Figure 5).

Figure 5. Central point of the applicator

The distances between craniocaudal; lateral and anteroposterior lines and central point were recorded in millimeters according to the three dimensional Cartesian coordinate system. While this recording process was going on; the measurement results that are cranial to the line that passes through the femur head vertices were evaluated as (+); while the results caudal to the line were evaluated as (-). Results right to the midline were evaluated as (+); while results to the left were evaluated as (-). As for lateral X-rays; since all the results are posterior to the line; they were evaluated as (-).

In order to detect the angular changes of R&T applicator in lateral direction; the angle between the axis that joins the tip of the applicator with the central point (applicator axis) and sagittal pelvic line [=alfa(a)] is determined in the AP X-ray (Figure 6).

Figure 6. Alpha angle

In order to detect the angular changes of R&T applicator in anteroposterior direction; the angle between the perpendicular line that descends from the middle of the line that joins the sacral promontory and pubic symphysis and applicator axis [=beta(b)]is determined in the lateral (90°) X-ray (Figure 7).

Figure 7. Beta angle

Data on rectum and bladder reference doses for each application were recorded as well. In addition; in lateral simulation X-rays of ICBT applications; the distances between ring central point and rectum (R) and bladder (B) points were measured and recoded in millimeters (Figure 8).

Figure 8. The distance between applicator center and rectum and bladder reference points

Examining all the applications of every patient; the ranges in which the parameters related to the anatomical position of the applicator in the pelvis (cranial/caudal; lateral; anterior/posterior distances according to bony reference points with lateral and anterior-posterior angles) changed and average values (median; mean and standard deviation) in the R&T application were found.
Inter- and intraindividual variability of the applicator center was examined. In order to evaluate interindividual variability; average anatomical applicator position for all applications was normalized as “0 position”. Later on; in all applications according to this “0 position”; maximum; minimum; median; mean (with ±%95 confidence interval) change and standard deviation was determined in craniocaudal; lateral and anteroposterior directions.

In order to detect intraindividual applicator center variability; the applicator position in the initial application was taken as “0 position”. Later on; maximum; minimum; median; mean (with ±%95 confidence interval) change and standard deviation of the patients’ other applications were calculated according to the initial application.

Intraindividual applicator center variability was also found out by calculating the difference between all applications relative to each other; using “0 position”; the average of 3 applications. Variability calculated according to this was again stated as maximum; minimum; median; mean (with ±%95 confidence interval) and standard deviation in craniocaudal; lateral and anteroposterior directions.

In addition; in which ranges the distance of the applicator center to bladder and rectum reference points change and average values (median; mean and standard deviation) were found out.

Finally; possible factors that effect variability of the position of applicator center were examined. For this aim; single variable analysis was performed by age (≤ 60; > 60); stage (≤ 2B; > 2B); duration between the external RT and first brachytherapy (≤8 days; >8 days) using Mann-Whitney U test; investigating whether the position of applicator center effects intraindividual variability compared to the first application. Statistical analyses were performed by computer using SPSS® v.13 software.


V. FINDINGS

V.a. CHANGES OF THE ANATOMICAL POSITION OF R&T APPLICATOR IN THE PELVIS

It was observed that change of R&T applicator in the pelvis in longitudinal direction according to bony reference points was in the range of 3.64 mm in cranial direction and 9.44 mm in caudal direction when the line that passes through the femur heads was taken as reference. Change in mediolateral direction was in the range of 0.64 mm at the left side and 1.02 mm at the right side when the line that passes through the midpoint of pubic symphysis and sacrum was taken as reference. The anteroposterior change was in the range of 0-15 mm towards posterior when the line that passes through the midpoint of the line between sacrum promontory and pubic symphysis was taken as reference (Table 2).

It was seen that the angular change of R&T applicator in the lateral direction (a angle) was in the range of 10° to the right and 0° to the left. It was observed that the angular change of R&T applicator in the anteroposterior direction (b angle) was in the range of 30° to the anterior and 0° to the posterior (Table 2).

Table 2. R&T Applicator Position Change According to the Bony Reference System

Position Change According to the Reference Lines (mm)

Angle Change (°)



Craniocaudal

Mediolateral

Anteroposterior

Lateral
(α)

Anteroposterior
(β)

Variability
13;08
1;66
15
10
30
Cranial
3;64
-
-
-
-
Caudal
-9;44
-
-
-
-
Right
-
-1;02
-
0
-
Left
-
0;64
-
10
-
Anterior
-
-
0
-
30
Posterior

-
15
-
0
Median
-2;49
0
-9;9
4
26
Mean
-2;89
- 0;02
-9;85
5;25
23;25
Standard
Deviation

3;83

0;39

2;86

3;42

8;35



V.b. INTERINDIVIDUAL AND INTRAINDIVIDUAL CHANGES OF THE R&T APPLICATOR IN THE PELVIS

It was found that standard deviations of the interindividual applicator variability were 3.83 mm in the longitudinal plane; 0.39 mm in the mediolateral plane and 2.86 mm in the anteroposterior plane (Figure 9). Standard deviations of the intraindividual applicator variability according to the initial R&T application were 1.91 mm in the longitudinal plane; 0.4 mm in the mediolateral plane and 4.26 mm in the anteroposterior plane. When standard deviations of the intraindividual applicator variability were reviewed according to average of the three applications; they were 0.95 mm in the longitudinal plane; 1.86 in the mediolateral plane and 1.24 mm in the anteroposterior plane. Data related to these results are shown in the table below (Table 3; Figure 9).
Table 3. Interindividual and Intraindividual Changes of the R&T Applicator

(mm)

Interindividual Variability
Intraindividual Variability According to the First Application

Intraindividual Variability
Craniocaudal




Maximum
0;74
1;32
0;88
Minimum
-12;34
- 5;74
- 3;3
Median
- 5;39
- 0;26
- 0;1
Mean
- 5;79
- 0;92
- 0;35
Standard Deviation
3;83
1;91
0;95
±%95 Confidence Interval
- 7;41 and- 4;17
- 1;94 and 0;08
- 0;75 and 0;05
Mediolateral
Maximum
0;62
0;77
8;2
Minimum
-1;04
- 0;65
-1;62
Median
- 0;02
0
0;06
Mean
- 0;04
0;11
0;52
Standard Deviation
0;39
0;40
1;86
±%95 Confidence Interval
- 0;20 and 0;12
- 0;10 and 0;32
- 0;25 and 1;31
Antero
posterior
Maximum
- 9;82
1;58
0;83
Minimum
- 24;82
-13;1
- 4;9
Median
-19;71
- 0;28
- 0;02
Mean
-19;67
-1;78
- 0;32
Standard Deviation
2;86
4;26
1;24
±%95 Confidence Interval
-20;88 and -18;46
- 4;05 and 0;49
- 0;85 and 0;19



Figure 9. Graph that shows the interindividual changes

V.c. CHANGE IN THE DISTANCE OF THE BLADDER AND RECTUM REFERENCE POINTS TO THE APPLICATOR CENTRAL POINT

Standard deviations of the distance between applicator central point and bladder and rectum were found as 1.19 mm and 0.5 mm; respectively. These data are shown in Table 4.







Table 4. Change In the Distance Of the Bladder And Rectum Reference Points To the Applicator Central Point



Bladder
Reference
Point
(mm)
Rectum
Reference
Point
(mm)
Maximum
8;81
5;92
Minimum
4;47
4;21
Median
6;05
4;93
Mean
6;26
4;86
Standard Deviation
1;19
0;5



V.d. DATA RELATED TO THE DOSES TAKEN UP BY THE BLADDER AND RECTUM REFERENCE POINTS

In brachytherapy application of the 8 cases by 24 R&T applicators; mean dose taken up by the bladder reference point was 409.5 cGy (standard deviation 218.15 cGy); while the dose taken up by the rectum reference point was 442.91 cGy (standard deviation 109.75 cGy).

V.e. POSSIBLE FACTORS THAT EFFECT VARIABILITY OF THE POSITION OF APPLICATOR CENTER

In the single variable analysis that was performed; factors related to the patient (age); disease (stage) and RT (the time between external RT and brachytherapy) which were considered to have an effect on the position of applicator center; were shown to have no effect on the changes of applicator position (Table 5).


Table 5. Results of the Single Variable Analysis That Investigated the Effect of the Factors Related to Patient; Disease and Radiotherapy on Variability of the Position of Aplicator Center

n
p
Craniocaudal
Age
≤ 60
5

0;456
> 60
3
Stage
≤ 2B
4

0;149
> 2B
4
Time between external RT and ICBT
≤ 8 days
4

0;773
> 8 days
4
Mediolateral
Age
≤ 60
5

0;297
> 60
3
Stage
≤ 2B
4

0;386
> 2B
4
Time between external RT and ICBT
≤ 8 days
4