- Email: [email protected]
Assessment of skin dose and its relation to cosmesis in conservative treatment of early breast cancer
Inr. J Radralion Oncolo~.v Biol Phy.7, Vol. 14, PP. 29 l-296 Printed in the U.S.A. All rights reserved.
Copyright
0
0360-3016/88 1988 Pergamon
$3.00 + .OO Journals Ltd.
0 Original Contribution ASSESSMENT OF SKIN DOSE AND ITS RELATION TO COSMESIS IN THE CONSERVATIVE TREATMENT OF EARLY BREAST CANCER F. HABIBOLLAHI,
DMRT,
P. J. WINTER,
M. PHIL.,* FRCR,$
H. M. 0. MAYLES, D. TONG,
M. A. CHAUDARY,
FRCS*
FRCR,$
M.Sc.,-f-
W. P. M. MAYLES,
I. S. FENTIMAN,
AND J. L. HAYWARD,
PH.D.,?
FRCS,*
FRCS*
Guy’s Hospital, London SE 1 9RT A conservation technique has been developed for the treatment of early breast cancer which involved removal of the tumor, axillary clearance, tumor site implantation with Iridium-192 wires for a boost dose and subsequent treatment of the breast with radical megavoltage external beam therapy. Although the cosmetic results were satisfactory in the majority of the patients, for some it was rated as fair or poor. One variable factor which could have carried some morbidity was the dose of radiation received by the skin. In 51 patients, doses were measured at several points over the treated breast using Thermoluminescent Dosimetry (TLD) at the time of the iridium implant and during the subsequent external beam therapy. Development of skin pigmentation, oedema, and fibrosis were unrelated to the dose received by the skin but the findings suggested that doses greater than 50 Gy to the skin increased the possibility of late (>24 months) telangiectasia over the boosted area. Treatment of tumors in the lower half of the breast, or in large breasts, was associated with a higher incidence of poor cosmesis. This may have been the result of varying posture on the interstitial dose distribution from the Iridium-192 wires and comparison of dose distribution in both supine and erect positions was carried out. Skin, Dosimetry,
Iridium-192,
Breast cancer, Cosmesis.
METHODS
INTRODUCTION
AND
MATERIALS
Radiation dose to the skin over the treated breast was measured in 5 1 patients undergoing conservative treatment between May 1983 and May 1985. These patients were part of a larger prospective randomized study comparing conservative treatment and mastectomy in women with early breast cancer (EORTC Trial 10801). The details of the technique and criteria for patient selection have been published earlier.’ Patients with tumors measured clinically to be 4 cm or less and under the age of 70 were accepted and treated with local excision, total axillary clearance, and a boost dose with iridium implant to the tumor site of 20 Gy over 2 days. Finally external beam therapy was given to the breast to 46 Gy over 44 weeks using a 4 MeV linear accelerator (Fig. 1). The interstitial implant was performed with flexible plastic tubes using the after-loading technique as described by Paine ( 1972).9 Dose measurements were made with Lithium Fluoride (LiF) microrods of type TLD-700. Skin doses during the iridium implant and external beam radiotherapy were measured at 14 points (Fig. 2). For each set of measurements 3 groups of LiF rods were used, each
The mortality rate from breast cancer has shown no improvement in 50 years and little gain in survival has been noted since 1949.‘,8 For many years the controversy about the surgical treatment of early breast cancer has continued. Radical surgery, even in node negative cases, offers no more than a 70% chance of survival at 10 years and this figure has not changed over the last 3 decades. The failure of radical surgery to prolong survival has resulted in the increasing popularity of conservative surgery followed by radiotherapy, particularly in Europe. The results of some recent studies have demonstrated that, in selected groups of patients, conservative treatment is a valid alternative to mastectomy.5~‘2~‘4 The combination of excision biopsy and radiation therapy requires close collaboration between the surgeon and the radiotherapist if a good cosmetic result is to be achieved. Both the extent of the surgical procedure and the radiotherapy technique influence the final cosmetic result.2.4,6.’‘,I3 This paper reports the cosmetic result in a defined group of patients treated for early breast cancer and its relationship to skin radiation dose.
Reprint requests to: I. S. Fentiman, FRCS. Accepted for publication 28 August 1987.
* ICRF, Department of Clinical Oncology. t Department of Clinical Physics and Bioengineering. $ Department of Radiotherapy. 291
1.J. Radiation Oncology 0 Biology 0 Physics
292
300
February 1988, Volume 14, Number 2
r
L’
wedge -____
Wedge ----_.
-
Fig. 1. External beam treatment of the breast with 4 Mev linear accelerator.
housed in a thin plastic tube with the individual rods separated from each other by nylon spacers. One tube containing five rods spaced at 30 mm intervals was placed in a sagittal plane on the surface of the breast with the central rod located at the nipple. The second tube was placed perpendicular to the first (in the transverse plane). The rods were again separated by 30 mm but the center rod (at the nipple) was omitted. In the third tube the rods were separated by 10 mm and this tube was placed along the scar. For the iridium implants IO mm copper markers were included in the tubes on either side of each rod so that the position could be determined from radiographs and the dose calculated (Fig. 3). Doses were measured during the implant and on 2 consecutive days dur-
Colour
ing external doses for the (Table 1 and To evaluate the following
beam therapy. From these measurements full course of therapy could be calculated Fig. 4). the cosmetic outcome of the treated breast criteria were used:
Excellent: The treated breast looked essentially the same as the opposite breast. Good: There were minimal, but identifiable changes compared with the opposite breast. Fair: Obvious difference between the two sides but without major distortion. Poor: Major aesthetic changes in the treated breast. The first two groups were counted
code scar
Colour code
Fig. 2. Position of TLD’s over the breast.
as favorable
or satis-
Skin dose and cosmesis 0 F. HABIBOLLAHI et al
293
Colour code _-_-------
_--.-
I--.-_-_._
I
Nylon tube
-
1 cm Metal . . . . 1 cm Nylon IU LIF Fig. 3. TLD housing.
factory cosmesis and the last two as unfavorable or unsatisfactory. Breast size was assessed on the basis of bra cup size as follows: small (Size A), medium (Size B), and large (Size C and D). RESULTS Patients were aged between 24 and 69 years and the median duration of follow-up for this group of patients was twenty-seven months (range 19-4 1). The average skin dose to the whole breast was obtained by taking the average of the nine dosimeters in the transverse and sagittal tubes with the nipple dose being counted twice. The mean for all patients was 30.0 * 2.3 Gy. Similarly the scar dose was taken as the average of the 5 dosimeters placed on the scar. The mean for all patients was 44.4 * 4.5 Gy. The contribution of the implant to these mean doses was 4.4 + 1.7 Gy and 8.9 + 2.3 Gy respectively.
Table 1. TLD measurement Iridium Point number
Measured
during the implant
The cosmetic results were recorded as very good in 14 cases (27%), good in 27 cases (53%), fair in 9 cases (18%) and poor in only one case (2%). The primary tumor was located in the upper half of the breast in 39 patients (77%) and in this group only 3 patients (8%) had unfavorable cosmesis. Only 12 patients (23%) had tumors in the lower half of the breast and in this group 7 patients (58%) were recorded with unsatisfactory cosmesis (X2, Yates correction = 11.89 p < 0.00 1). Thus, there was a highly significant reduction in cosmetic outcome when the tumor was in the lower half of the breast. However, the dose received by the scar was not significantly different in the upper or lower half of the breast (Table 2). Similarly, when those patients with poor cosmetic results were analyzed separately there were no differences between skin dosages received by those with upper half or lower half tumors (Table 3). The relation between the size of the breast and an unfavorable cosmetic outcome was also recorded in
and subsequent
external
doses
Computed
beam radiotherapy
External
beam
Result
Day 1
Day 2
Mean
Total
1 2 3 4
224 531 301 141
216 605 317 157
224 531 301 141
3674 347 1 3826 3770
3785 3659 3759 3639
3730 3565 3792 3705
3954 4096 4093 3846
2
814 114
674 116
814* 114
3067 3495
3096 3680
308 1 3587
3895 3701
: 9 10 11 12 13 14
520 131 69 957 817 734 691 464
464 150 77 944 759 699 601 477
520 131 69 957 817 734 691 464
3534 3406 2870 3131 3318 3415 3633 4045
3358 3457 3139 2978 3312 3444 3643 3918
3432 3446 3004 3054 3315 3430 3638 3981
3952 3577 3073 4011 4132 4164 4329 4445
3513 3484 3981
3828 4216 4445
Average-Whole Average-Scar: Maximum:
breast:
315 733 957
Note: Patient number: 27/85; Dose given: 1999 cGy; Paris System dose: 1999 cGy. * Disagreement between results excessive (+ 15%).
294
I.
J.
Radiation
Oncology 0
Biology 0 Physics
February 1988. Volume 14, Number 2
w/ >)
;Y
CGY
CGY
30 -
DO-
DOTotal
Total
>O-
Total
R External
DO-
:z!?&
zz?
DO-
DO-
Iridium
Iridium
OK?-+-+ 1 2 Superior
3 Nipple
4
h?-?,IY-? 1 2
5 Inferior
Lateral
3 4 5 Nipple Medial
1
2
3 Scar
4
5
Fig. 4. Visual display of figures in Table 1.
this group of patients and no association was found (Table 4). Relative skin dose measurements in relation to breast size are given in Table 5. Although the mean iridium contribution to the scar dose was less in women with larger breasts, this difference was not significant. There were 1 1 patients with slight and 2 with moderate residual oedema. Six patients had large breasts, 6 were medium, and 1 was small. The tumor was in the upper half of the breast in 10 patients and in 3 others was in the lower half. There was no major difference in the recorded skin dose of women with breast oedema compared with those without swelling. Residual oedema was detectable mainly in patients with less than 2 years of follow-up. Fibrosis was detected in 14 patients but in only two cases was this moderate. Eight patients had large breasts, 5 were medium, and 1 was small. In 10 patients the tumor was in the upper half of the breast and in 4 others it was in the lower half. No significant skin dose differences were noted but the majority of these patients had more than
Table 2. Skin dose measurements
and location of implant
Mean total scar dose Gy
Mean iridium contribution to the scar dose, Gy
Upper half tumors (39 patients)
43.8 + 4.6
8.7 & 2.2
Lower half tumors ( 12 patients)
46.2 ? 3.7
9.9 + 2.6
2 years of follow-up. Three patients developed a small amount of telangiectasia over the boosted area and in two the average total scar dose was more than 50 Gy with over 24 months of follow-up. There were another 3 patients with similar high scar dose with no telangiectasia and shorter follow-up period of under 24 months. One patient also developed a skin recurrence immediately above the scar and it was noted that this patient received less than the average skin dose. In this case the average whole breast skin dose was 33.8 Gy and the average scar dose was also recorded as 33.8 Gy. This patient had large breasts with a deep-seated tumor. In comparing the computer calculated doses with the TLD measured doses it was noticed that the TLD measured dose at the most inferior point on the breast (point 5) was consistently higher than the computer calculated dose. This is the point at which the skin dose would be most affected by a change in position of the patient; the computer calculated doses used radiographs taken with the patient in a supine position, and did not account for
Table 3. Skin dose measurements cosmetic
of patients results
Mean total scar dose Gy Upper half tumors (3 patients) Lower half tumors (7 patients)
with unfavorable
Mean iridium contribution to the scar dose, Gy
39.1 f 5.4
8.7 I!z2.0
46.8 f 3.6
10.0 ? 3.0
Skin dose and
cosmesis 0
Table 4. Breast size and cosmetic outcome
Small breasts, 7 patients ( 14%) Medium breasts, 22 patients (43%) Large breasts, 22 natients (43%)
Fair cosmesis
Poor cosmesis
2 2 5
0 A
F. HABIBOLLAHI et al.
outcome. TLD measurements were also compared with computer calculated doses for similar points. The following factors limit the accuracy of the TLD measurements: 1. The rods in their plastic
2.
Table 5. Breast size and skin dose
Small breasts Medium breasts Large breasts
Mean total scar dose Gy
Mean iridium contribution to the scar dose, Gy
46.6 t 1.7 44.7 ? 4.6 44.3 t 4.8
10.4 t 2.8 9.1 + 2.2 8.4 f 2.1
3.
4. the fact that in practice the patient might spend half the treatment time sitting up. To determine whether the patient’s position might have an effect on delivered skin dose and implant shape, two sets of radiographs were taken on four patients, one with the patient supine and the other with the patient erect. The two sets of calculated skin doses were compared and in all cases these showed that the skin dose rate at point 5 was lowered when the patient was in an erect position. What was more interesting however, was the effect on the implant shape, the Paris Isodose, and alteration in dose distribution within the implant volume. This can be seen from Figures 5a and 5b, for the case of an implant in the lower half of the breast.
DISCUSSION The aim of this study was to measure skin dose with thermoluminescent dosimetry (TLD) during the iridium 192 implant and subsequent external beam radiotherapy and study the relation between skin dose and cosmetic
295
5.
6.
housing must stay attached to the skin firmly during the time of measurement. During the iridium implant the rods are separated from the skin surface by the thickness of the nylon tube. Comparisons on a phantom with the iridium wire at a representative distance from the surface indicated that this reduced the dose measured by about 4% and this amount was therefore added to the measured doses. The amount of backscatter is less for rods situated above the skin compared to the skin surface. No allowance was made for this. The rod has a finite size and no account is taken of dose variations along its 6 mm length. The dose gradients in an interstitial implant are large and consequently the doses measured may not be representative of the average, for example, a very high dose may be recorded close to an iridium wire. In addition, the TLD measurements doses from the iridium implant were calculated with a computer; it was found that calculated doses were on average 12% higher than those measured. Phantom measurements suggested that this difference could be accounted for by the lack of backscattering material on the skin surface--the computer calculations being valid only for a homogeneous medium.
In this series of patients an observer-based method with a four scale grading was used and in each case during their follow-up period patients were examined by the same observer and cosmesis recorded independently from the previous observation. There were also a series of large color photographs available but this technique has some limitations as Pezner et al.” suggested. Beadle et ~1.~reviewed the cosmetic results of 239 patients treated by primary radiation without adjuvant che-
Skin Marker Skin Marker
Fig. 5. (a) Dose distribution of erect films.
calculated on the basis of supine films. (b) Dose distribution
calculated
on the basis
296
1. J. Radiation Oncology 0 Biology 0 Physics
motherapy. They noted that the age of the patient and size or site of the primary tumor did not significantly influence the overall cosmetic result. In addition they found that the unfavorable cosmesis correlated closely with the development of moderate or severe breast retraction either alone or with telangiectasia. We did not observe any relation between the age of the patient and their cosmesis, but larger tumors were associated with less favorable cosmetic outcome. There was no significant relation between the size of the breast and cosmesis, although larger breasts simply retained oedema for longer periods of time. When the site of the primary tumor was taken into consideration, significant differences were noted and cosmesis was less satisfactory in patients in whom the tumor was in the inferior part of the breast. Some of the poor cosmetic results were due to the effect of slight surgical defects in biopsy technique which were more noticeable in the lower half of the breast. In addition, residual oedema remained detectable for longer periods of time. However, the data in Table 2 shows that, while the mean total scar dose for lower half tumors was slightly higher than that for upper half tumors, the difference is not significant. In view of the change in the position of the patient during the iridium implant, some alteration of dosimetry may take place.
February 1988, Volume 14, Number 2
Thus, the dose received by the breast tissue may be higher than anticipated and prescribed. The basis of computer calculations was the supine films but when erect views were obtained small changes in the implant volume were noted. This was more evident when the implant was situated in the lower half of the breast. However. the excess doses were not high enough to account for the differences noted between the cosmetic outcome in the upper and lower half tumors. It seems reasonable to assume that possibly all of the above factors, up to a certain point, were responsible for the significant cosmetic difference between the upper and lower half tumors. This requires further studies, especially with rigid metal implants, which may change less in position compared with flexible plastic tubing.
CONCLUSIONS Within the radiation dose range used in the group of patients in this study. no major morbidity was noted and the treatment was well tolerated with a satisfactory cosmetic outcome in the majority of cases. No relationship was demonstrated between skin dose (as measured by TLD) and the end cosmetic result, although a significant reduction in aesthetic outcome was found in patients with inferiorly-placed primary tumours.
REFERENCES I. Adair, F., Berg, J., Joubert.
L., Robbins, G.F.: Long-term of breast cancer patients: The 30 year report. Cuncer33: 1145-l 150,1974. Ashford, R.F.U., Philhphs, R.H., Coe, M.A., Pickering, D.G.L., Bailey, B.. Ellis, H.: Conservative excision and radiotherapy for early breast cancer. An acceptable alternative? C/in. Oncol. 10: 45-58, 1984. Beadle, G.F., Silver, B.. Botnick, L., Hellman, S.. Harris. J.R.: Cosmetic results following primary radiation therapy for early breast cancer. Cuncer 54: 29 11-29 18, 1984. Bedwinek, J.: Treatment of Stage I and II adenocarcinoma of the breast by tumour excision and irradiation. Int. J. Radiat. Oncol. Biol. Phys. 7: 1553-1559, 1981. Fisher, B.. Bauer, M., Margolese, R., Poisson, R., Pilch, Y., Redmond, C., Fisher, C., Wolmark, N., Deutsch, M., Montague, E., Saffer, E., Wickerham, L., Lerner, H., Glass, A., Shibata, H., Deckers, P., Ketcham, A., Oishi, R.. Russell, 1.: Five year results of a randomised clinical trial comparing total mastectomy and segmental mastectomy with or without radiation in the treatment of breast cancer. N. Eng. J. Med. 312: 665-673, 1985. Harris, J.R., Levene, M.B., Svensson, G., Hellman, S.: Analysis of cosmetic results following primary radiation therapy for Stage I and II carcinoma of the breast. Int. J. follow-up
2.
3.
4.
5.
6.
Rudiut. Oncol. Biol. Phys. 5: 257-26 1, 1979. 7. Hayward, J.L., Winter, P.J., Tong, D., Rubens,
R.D., Payne, J.G., Chaudary, M.A., Habibollahi, F.: A new approach to the conservative treatment of early breast cancer. Surgery 95(3): 270-274, 1984. 8. Myers, M.H., Hankey, B.F., Steinhorn, S.C., Flannery,
9.
10.
I I.
12.
13.
J.T.: Inter-relationship of incidence, survival and mortality and their effects on time trend. In Trends in Cancer Incidence. Magnus. K. (Ed.). New York, Hemisphere Pub. 1982. pp. 79-87. Paine, C.H.: Modern after-loading methods for interstitial radiotherapy. Clin. Radiol. 23: 263-272, 1972. Pezner, R.D., Lipsett, J.A., Vora, N.L., Desai, K.R.: Limited usefulness of observer-based cosmesis scales employed to evaluate patients treated conservatively for breast cancer. Int. J. Rudiut. Oncol. Biol. Phys. 11: 1117-l 119, 1985. Pierquin, B., Otmezguine, Y., Lobo, P.A.: Conservative management of breast carcinoma. The Creteil experience. Actu Radio/. Oncol. 22: 101-107, 1983. Sarrazin, D., Tubiana. M., Le, M., Fontaine, F., Arriagada, R.: Conservative treatment of minimal breast cancer. In Breast Cuncer. Experimental und Clinical Aspects, Mouridsen, H.T., Palshof, T. (Eds.). Oxford, Pergamon. 1980, pp. 25 l-253. Schwartz, G.F., Danoff, S.A., Feig, H., Rosenblum, H.M., Bloss, J.D., Gall, J.R.: The selective treatment of “early” carcinoma of the breast by lumpectomy, level I axillary dissection and radiotherapy. Recent Rex Cuncer Rex 90:
133-140,1984. 14. Veronesi, U., Saccozzi, R., Del Vecchio, M., Banfi, A., Clemente. C., De Lena, M., Gallus, G., Grecco, M., Luini, A., Marubini, E., Muscolino, G., Pilke, F., Salvadori, B., Zecchini. A., Zucali, R.: Comparing radical mastectomy with quadrantectomy, axillary dissection and radiotherapy in patients with small cancers ofthe breast. N. Eng. J. Med. 305:6-Il. 1981.
Copyright
0
0360-3016/88 1988 Pergamon
$3.00 + .OO Journals Ltd.
0 Original Contribution ASSESSMENT OF SKIN DOSE AND ITS RELATION TO COSMESIS IN THE CONSERVATIVE TREATMENT OF EARLY BREAST CANCER F. HABIBOLLAHI,
DMRT,
P. J. WINTER,
M. PHIL.,* FRCR,$
H. M. 0. MAYLES, D. TONG,
M. A. CHAUDARY,
FRCS*
FRCR,$
M.Sc.,-f-
W. P. M. MAYLES,
I. S. FENTIMAN,
AND J. L. HAYWARD,
PH.D.,?
FRCS,*
FRCS*
Guy’s Hospital, London SE 1 9RT A conservation technique has been developed for the treatment of early breast cancer which involved removal of the tumor, axillary clearance, tumor site implantation with Iridium-192 wires for a boost dose and subsequent treatment of the breast with radical megavoltage external beam therapy. Although the cosmetic results were satisfactory in the majority of the patients, for some it was rated as fair or poor. One variable factor which could have carried some morbidity was the dose of radiation received by the skin. In 51 patients, doses were measured at several points over the treated breast using Thermoluminescent Dosimetry (TLD) at the time of the iridium implant and during the subsequent external beam therapy. Development of skin pigmentation, oedema, and fibrosis were unrelated to the dose received by the skin but the findings suggested that doses greater than 50 Gy to the skin increased the possibility of late (>24 months) telangiectasia over the boosted area. Treatment of tumors in the lower half of the breast, or in large breasts, was associated with a higher incidence of poor cosmesis. This may have been the result of varying posture on the interstitial dose distribution from the Iridium-192 wires and comparison of dose distribution in both supine and erect positions was carried out. Skin, Dosimetry,
Iridium-192,
Breast cancer, Cosmesis.
METHODS
INTRODUCTION
AND
MATERIALS
Radiation dose to the skin over the treated breast was measured in 5 1 patients undergoing conservative treatment between May 1983 and May 1985. These patients were part of a larger prospective randomized study comparing conservative treatment and mastectomy in women with early breast cancer (EORTC Trial 10801). The details of the technique and criteria for patient selection have been published earlier.’ Patients with tumors measured clinically to be 4 cm or less and under the age of 70 were accepted and treated with local excision, total axillary clearance, and a boost dose with iridium implant to the tumor site of 20 Gy over 2 days. Finally external beam therapy was given to the breast to 46 Gy over 44 weeks using a 4 MeV linear accelerator (Fig. 1). The interstitial implant was performed with flexible plastic tubes using the after-loading technique as described by Paine ( 1972).9 Dose measurements were made with Lithium Fluoride (LiF) microrods of type TLD-700. Skin doses during the iridium implant and external beam radiotherapy were measured at 14 points (Fig. 2). For each set of measurements 3 groups of LiF rods were used, each
The mortality rate from breast cancer has shown no improvement in 50 years and little gain in survival has been noted since 1949.‘,8 For many years the controversy about the surgical treatment of early breast cancer has continued. Radical surgery, even in node negative cases, offers no more than a 70% chance of survival at 10 years and this figure has not changed over the last 3 decades. The failure of radical surgery to prolong survival has resulted in the increasing popularity of conservative surgery followed by radiotherapy, particularly in Europe. The results of some recent studies have demonstrated that, in selected groups of patients, conservative treatment is a valid alternative to mastectomy.5~‘2~‘4 The combination of excision biopsy and radiation therapy requires close collaboration between the surgeon and the radiotherapist if a good cosmetic result is to be achieved. Both the extent of the surgical procedure and the radiotherapy technique influence the final cosmetic result.2.4,6.’‘,I3 This paper reports the cosmetic result in a defined group of patients treated for early breast cancer and its relationship to skin radiation dose.
Reprint requests to: I. S. Fentiman, FRCS. Accepted for publication 28 August 1987.
* ICRF, Department of Clinical Oncology. t Department of Clinical Physics and Bioengineering. $ Department of Radiotherapy. 291
1.J. Radiation Oncology 0 Biology 0 Physics
292
300
February 1988, Volume 14, Number 2
r
L’
wedge -____
Wedge ----_.
-
Fig. 1. External beam treatment of the breast with 4 Mev linear accelerator.
housed in a thin plastic tube with the individual rods separated from each other by nylon spacers. One tube containing five rods spaced at 30 mm intervals was placed in a sagittal plane on the surface of the breast with the central rod located at the nipple. The second tube was placed perpendicular to the first (in the transverse plane). The rods were again separated by 30 mm but the center rod (at the nipple) was omitted. In the third tube the rods were separated by 10 mm and this tube was placed along the scar. For the iridium implants IO mm copper markers were included in the tubes on either side of each rod so that the position could be determined from radiographs and the dose calculated (Fig. 3). Doses were measured during the implant and on 2 consecutive days dur-
Colour
ing external doses for the (Table 1 and To evaluate the following
beam therapy. From these measurements full course of therapy could be calculated Fig. 4). the cosmetic outcome of the treated breast criteria were used:
Excellent: The treated breast looked essentially the same as the opposite breast. Good: There were minimal, but identifiable changes compared with the opposite breast. Fair: Obvious difference between the two sides but without major distortion. Poor: Major aesthetic changes in the treated breast. The first two groups were counted
code scar
Colour code
Fig. 2. Position of TLD’s over the breast.
as favorable
or satis-
Skin dose and cosmesis 0 F. HABIBOLLAHI et al
293
Colour code _-_-------
_--.-
I--.-_-_._
I
Nylon tube
-
1 cm Metal . . . . 1 cm Nylon IU LIF Fig. 3. TLD housing.
factory cosmesis and the last two as unfavorable or unsatisfactory. Breast size was assessed on the basis of bra cup size as follows: small (Size A), medium (Size B), and large (Size C and D). RESULTS Patients were aged between 24 and 69 years and the median duration of follow-up for this group of patients was twenty-seven months (range 19-4 1). The average skin dose to the whole breast was obtained by taking the average of the nine dosimeters in the transverse and sagittal tubes with the nipple dose being counted twice. The mean for all patients was 30.0 * 2.3 Gy. Similarly the scar dose was taken as the average of the 5 dosimeters placed on the scar. The mean for all patients was 44.4 * 4.5 Gy. The contribution of the implant to these mean doses was 4.4 + 1.7 Gy and 8.9 + 2.3 Gy respectively.
Table 1. TLD measurement Iridium Point number
Measured
during the implant
The cosmetic results were recorded as very good in 14 cases (27%), good in 27 cases (53%), fair in 9 cases (18%) and poor in only one case (2%). The primary tumor was located in the upper half of the breast in 39 patients (77%) and in this group only 3 patients (8%) had unfavorable cosmesis. Only 12 patients (23%) had tumors in the lower half of the breast and in this group 7 patients (58%) were recorded with unsatisfactory cosmesis (X2, Yates correction = 11.89 p < 0.00 1). Thus, there was a highly significant reduction in cosmetic outcome when the tumor was in the lower half of the breast. However, the dose received by the scar was not significantly different in the upper or lower half of the breast (Table 2). Similarly, when those patients with poor cosmetic results were analyzed separately there were no differences between skin dosages received by those with upper half or lower half tumors (Table 3). The relation between the size of the breast and an unfavorable cosmetic outcome was also recorded in
and subsequent
external
doses
Computed
beam radiotherapy
External
beam
Result
Day 1
Day 2
Mean
Total
1 2 3 4
224 531 301 141
216 605 317 157
224 531 301 141
3674 347 1 3826 3770
3785 3659 3759 3639
3730 3565 3792 3705
3954 4096 4093 3846
2
814 114
674 116
814* 114
3067 3495
3096 3680
308 1 3587
3895 3701
: 9 10 11 12 13 14
520 131 69 957 817 734 691 464
464 150 77 944 759 699 601 477
520 131 69 957 817 734 691 464
3534 3406 2870 3131 3318 3415 3633 4045
3358 3457 3139 2978 3312 3444 3643 3918
3432 3446 3004 3054 3315 3430 3638 3981
3952 3577 3073 4011 4132 4164 4329 4445
3513 3484 3981
3828 4216 4445
Average-Whole Average-Scar: Maximum:
breast:
315 733 957
Note: Patient number: 27/85; Dose given: 1999 cGy; Paris System dose: 1999 cGy. * Disagreement between results excessive (+ 15%).
294
I.
J.
Radiation
Oncology 0
Biology 0 Physics
February 1988. Volume 14, Number 2
w/ >)
;Y
CGY
CGY
30 -
DO-
DOTotal
Total
>O-
Total
R External
DO-
:z!?&
zz?
DO-
DO-
Iridium
Iridium
OK?-+-+ 1 2 Superior
3 Nipple
4
h?-?,IY-? 1 2
5 Inferior
Lateral
3 4 5 Nipple Medial
1
2
3 Scar
4
5
Fig. 4. Visual display of figures in Table 1.
this group of patients and no association was found (Table 4). Relative skin dose measurements in relation to breast size are given in Table 5. Although the mean iridium contribution to the scar dose was less in women with larger breasts, this difference was not significant. There were 1 1 patients with slight and 2 with moderate residual oedema. Six patients had large breasts, 6 were medium, and 1 was small. The tumor was in the upper half of the breast in 10 patients and in 3 others was in the lower half. There was no major difference in the recorded skin dose of women with breast oedema compared with those without swelling. Residual oedema was detectable mainly in patients with less than 2 years of follow-up. Fibrosis was detected in 14 patients but in only two cases was this moderate. Eight patients had large breasts, 5 were medium, and 1 was small. In 10 patients the tumor was in the upper half of the breast and in 4 others it was in the lower half. No significant skin dose differences were noted but the majority of these patients had more than
Table 2. Skin dose measurements
and location of implant
Mean total scar dose Gy
Mean iridium contribution to the scar dose, Gy
Upper half tumors (39 patients)
43.8 + 4.6
8.7 & 2.2
Lower half tumors ( 12 patients)
46.2 ? 3.7
9.9 + 2.6
2 years of follow-up. Three patients developed a small amount of telangiectasia over the boosted area and in two the average total scar dose was more than 50 Gy with over 24 months of follow-up. There were another 3 patients with similar high scar dose with no telangiectasia and shorter follow-up period of under 24 months. One patient also developed a skin recurrence immediately above the scar and it was noted that this patient received less than the average skin dose. In this case the average whole breast skin dose was 33.8 Gy and the average scar dose was also recorded as 33.8 Gy. This patient had large breasts with a deep-seated tumor. In comparing the computer calculated doses with the TLD measured doses it was noticed that the TLD measured dose at the most inferior point on the breast (point 5) was consistently higher than the computer calculated dose. This is the point at which the skin dose would be most affected by a change in position of the patient; the computer calculated doses used radiographs taken with the patient in a supine position, and did not account for
Table 3. Skin dose measurements cosmetic
of patients results
Mean total scar dose Gy Upper half tumors (3 patients) Lower half tumors (7 patients)
with unfavorable
Mean iridium contribution to the scar dose, Gy
39.1 f 5.4
8.7 I!z2.0
46.8 f 3.6
10.0 ? 3.0
Skin dose and
cosmesis 0
Table 4. Breast size and cosmetic outcome
Small breasts, 7 patients ( 14%) Medium breasts, 22 patients (43%) Large breasts, 22 natients (43%)
Fair cosmesis
Poor cosmesis
2 2 5
0 A
F. HABIBOLLAHI et al.
outcome. TLD measurements were also compared with computer calculated doses for similar points. The following factors limit the accuracy of the TLD measurements: 1. The rods in their plastic
2.
Table 5. Breast size and skin dose
Small breasts Medium breasts Large breasts
Mean total scar dose Gy
Mean iridium contribution to the scar dose, Gy
46.6 t 1.7 44.7 ? 4.6 44.3 t 4.8
10.4 t 2.8 9.1 + 2.2 8.4 f 2.1
3.
4. the fact that in practice the patient might spend half the treatment time sitting up. To determine whether the patient’s position might have an effect on delivered skin dose and implant shape, two sets of radiographs were taken on four patients, one with the patient supine and the other with the patient erect. The two sets of calculated skin doses were compared and in all cases these showed that the skin dose rate at point 5 was lowered when the patient was in an erect position. What was more interesting however, was the effect on the implant shape, the Paris Isodose, and alteration in dose distribution within the implant volume. This can be seen from Figures 5a and 5b, for the case of an implant in the lower half of the breast.
DISCUSSION The aim of this study was to measure skin dose with thermoluminescent dosimetry (TLD) during the iridium 192 implant and subsequent external beam radiotherapy and study the relation between skin dose and cosmetic
295
5.
6.
housing must stay attached to the skin firmly during the time of measurement. During the iridium implant the rods are separated from the skin surface by the thickness of the nylon tube. Comparisons on a phantom with the iridium wire at a representative distance from the surface indicated that this reduced the dose measured by about 4% and this amount was therefore added to the measured doses. The amount of backscatter is less for rods situated above the skin compared to the skin surface. No allowance was made for this. The rod has a finite size and no account is taken of dose variations along its 6 mm length. The dose gradients in an interstitial implant are large and consequently the doses measured may not be representative of the average, for example, a very high dose may be recorded close to an iridium wire. In addition, the TLD measurements doses from the iridium implant were calculated with a computer; it was found that calculated doses were on average 12% higher than those measured. Phantom measurements suggested that this difference could be accounted for by the lack of backscattering material on the skin surface--the computer calculations being valid only for a homogeneous medium.
In this series of patients an observer-based method with a four scale grading was used and in each case during their follow-up period patients were examined by the same observer and cosmesis recorded independently from the previous observation. There were also a series of large color photographs available but this technique has some limitations as Pezner et al.” suggested. Beadle et ~1.~reviewed the cosmetic results of 239 patients treated by primary radiation without adjuvant che-
Skin Marker Skin Marker
Fig. 5. (a) Dose distribution of erect films.
calculated on the basis of supine films. (b) Dose distribution
calculated
on the basis
296
1. J. Radiation Oncology 0 Biology 0 Physics
motherapy. They noted that the age of the patient and size or site of the primary tumor did not significantly influence the overall cosmetic result. In addition they found that the unfavorable cosmesis correlated closely with the development of moderate or severe breast retraction either alone or with telangiectasia. We did not observe any relation between the age of the patient and their cosmesis, but larger tumors were associated with less favorable cosmetic outcome. There was no significant relation between the size of the breast and cosmesis, although larger breasts simply retained oedema for longer periods of time. When the site of the primary tumor was taken into consideration, significant differences were noted and cosmesis was less satisfactory in patients in whom the tumor was in the inferior part of the breast. Some of the poor cosmetic results were due to the effect of slight surgical defects in biopsy technique which were more noticeable in the lower half of the breast. In addition, residual oedema remained detectable for longer periods of time. However, the data in Table 2 shows that, while the mean total scar dose for lower half tumors was slightly higher than that for upper half tumors, the difference is not significant. In view of the change in the position of the patient during the iridium implant, some alteration of dosimetry may take place.
February 1988, Volume 14, Number 2
Thus, the dose received by the breast tissue may be higher than anticipated and prescribed. The basis of computer calculations was the supine films but when erect views were obtained small changes in the implant volume were noted. This was more evident when the implant was situated in the lower half of the breast. However. the excess doses were not high enough to account for the differences noted between the cosmetic outcome in the upper and lower half tumors. It seems reasonable to assume that possibly all of the above factors, up to a certain point, were responsible for the significant cosmetic difference between the upper and lower half tumors. This requires further studies, especially with rigid metal implants, which may change less in position compared with flexible plastic tubing.
CONCLUSIONS Within the radiation dose range used in the group of patients in this study. no major morbidity was noted and the treatment was well tolerated with a satisfactory cosmetic outcome in the majority of cases. No relationship was demonstrated between skin dose (as measured by TLD) and the end cosmetic result, although a significant reduction in aesthetic outcome was found in patients with inferiorly-placed primary tumours.
REFERENCES I. Adair, F., Berg, J., Joubert.
L., Robbins, G.F.: Long-term of breast cancer patients: The 30 year report. Cuncer33: 1145-l 150,1974. Ashford, R.F.U., Philhphs, R.H., Coe, M.A., Pickering, D.G.L., Bailey, B.. Ellis, H.: Conservative excision and radiotherapy for early breast cancer. An acceptable alternative? C/in. Oncol. 10: 45-58, 1984. Beadle, G.F., Silver, B.. Botnick, L., Hellman, S.. Harris. J.R.: Cosmetic results following primary radiation therapy for early breast cancer. Cuncer 54: 29 11-29 18, 1984. Bedwinek, J.: Treatment of Stage I and II adenocarcinoma of the breast by tumour excision and irradiation. Int. J. Radiat. Oncol. Biol. Phys. 7: 1553-1559, 1981. Fisher, B.. Bauer, M., Margolese, R., Poisson, R., Pilch, Y., Redmond, C., Fisher, C., Wolmark, N., Deutsch, M., Montague, E., Saffer, E., Wickerham, L., Lerner, H., Glass, A., Shibata, H., Deckers, P., Ketcham, A., Oishi, R.. Russell, 1.: Five year results of a randomised clinical trial comparing total mastectomy and segmental mastectomy with or without radiation in the treatment of breast cancer. N. Eng. J. Med. 312: 665-673, 1985. Harris, J.R., Levene, M.B., Svensson, G., Hellman, S.: Analysis of cosmetic results following primary radiation therapy for Stage I and II carcinoma of the breast. Int. J. follow-up
2.
3.
4.
5.
6.
Rudiut. Oncol. Biol. Phys. 5: 257-26 1, 1979. 7. Hayward, J.L., Winter, P.J., Tong, D., Rubens,
R.D., Payne, J.G., Chaudary, M.A., Habibollahi, F.: A new approach to the conservative treatment of early breast cancer. Surgery 95(3): 270-274, 1984. 8. Myers, M.H., Hankey, B.F., Steinhorn, S.C., Flannery,
9.
10.
I I.
12.
13.
J.T.: Inter-relationship of incidence, survival and mortality and their effects on time trend. In Trends in Cancer Incidence. Magnus. K. (Ed.). New York, Hemisphere Pub. 1982. pp. 79-87. Paine, C.H.: Modern after-loading methods for interstitial radiotherapy. Clin. Radiol. 23: 263-272, 1972. Pezner, R.D., Lipsett, J.A., Vora, N.L., Desai, K.R.: Limited usefulness of observer-based cosmesis scales employed to evaluate patients treated conservatively for breast cancer. Int. J. Rudiut. Oncol. Biol. Phys. 11: 1117-l 119, 1985. Pierquin, B., Otmezguine, Y., Lobo, P.A.: Conservative management of breast carcinoma. The Creteil experience. Actu Radio/. Oncol. 22: 101-107, 1983. Sarrazin, D., Tubiana. M., Le, M., Fontaine, F., Arriagada, R.: Conservative treatment of minimal breast cancer. In Breast Cuncer. Experimental und Clinical Aspects, Mouridsen, H.T., Palshof, T. (Eds.). Oxford, Pergamon. 1980, pp. 25 l-253. Schwartz, G.F., Danoff, S.A., Feig, H., Rosenblum, H.M., Bloss, J.D., Gall, J.R.: The selective treatment of “early” carcinoma of the breast by lumpectomy, level I axillary dissection and radiotherapy. Recent Rex Cuncer Rex 90:
133-140,1984. 14. Veronesi, U., Saccozzi, R., Del Vecchio, M., Banfi, A., Clemente. C., De Lena, M., Gallus, G., Grecco, M., Luini, A., Marubini, E., Muscolino, G., Pilke, F., Salvadori, B., Zecchini. A., Zucali, R.: Comparing radical mastectomy with quadrantectomy, axillary dissection and radiotherapy in patients with small cancers ofthe breast. N. Eng. J. Med. 305:6-Il. 1981.