Ini J. Rediorion Onmlo~.v Em/. Ph.w Vol. Printed in the U.S.A. All rights reserved.
23, pp.
925-93
I
0360-3016/92 $5.00 + .oO Copyright 0 1992 Pergamon Press Ltd.
??Clinical Original Contribution
ARE COSMETIC RESULTS FOLLOWING CONSERVATIVE SURGERY AND RADIATION THERAPY FOR EARLY BREAST CANCER DEPENDENT ON TECHNIQUE? ANNE DE LA ROCHEFORDI~RE, FRANK VICINI, M.D.,
M.D., ANTHONY L. ABNER, M.D., BARBARA SILVER, B.A., ABRAM RECHT, M.D. AND JAY R. HARRIS, M.D.
Joint Center for Radiation Therapy, Department of Radiation Oncology, Harvard Medical School, Boston, MA To assess the cosmetic results in relation to treatment technique, we retrospectively reviewed the results for 1159 Stage I-II breast cancer patients treated with conservative surgery and radiotherapy between 1970-1985. All patients underwent gross excision followed by radiation therapy including au implant or electron beam boost. The total dose to the primary site was 2 60 Gy. Because of technical modifications introduced over time after 1981, the population was divided arbitrarily into two cohorts: 504 patients treated through 1981 and 655 treated between 1982-1985. Median follow-up times for surviving patients in the two cohorts were 107 months and 67 months, respectively. Cosmetic outcome was evaluated by the examining physician and scored as excellent, good, fair or poor. Excellent results at 5 years were scored in 59% of early cohort patients and 74% of the latter cohort (p = 0.002). Acceptable results (either good or excellent) were seen in 84% and 948, respectively (p = 0.02). In the latter cohort, the likelihood of achieving an excellent result, but not an acceptable result, was significantly related to the volume of resected breast tissue and the use of chemotherapy. The number of fields (three-field technique, provided that fields are precisely matched, compared to tangents only) and boost type (implant vs electrons) did not influence the cosmetic outcome. We conclude that our current technique using breast RT to 45-
46 Gy and a boost to the primary site of 16-18 Gy is associated with a high likelihood of acceptable cosmetic results and that this likelihood is not diminished by the use of adjuvant chemotherapy, a large breast resection, the use of a third field, or boost type. Breast cancer, Conservative Surgery, Radiotherapy, Cosmesis. INTRODUCIION
precise matching of the tangential fields and the anterior field to treat the axillary and/or supraclavicular region, and the use of a large volume boost or a boost dose greater than 18 Gy. Based on these findings, a number of treatment modifications were introduced over time at our institution beginning in 1982. Sufficient time has now transpired to allow us to assess the results of these modifications. Thus, in this article, we divided arbitrarily the entire population into two cohorts of patients. We compare the cosmetic results in patients treated prior to 1982 to those in patients treated between 1982 and 1985, and assess treatment factors which are now associated with a worsening of the cosmetic appearance given these treatment modifications. This information, along with information on factors associated with local recurrence and complications, may be useful in providing patients with an improved technique of breast-conserving treatment.
Both retrospective (2, 6, 10, 16, 21, 24) and randomized clinical trials (9, 25,27) have established that breast-conserving treatment (consisting of conservative surgery and radiation therapy) and mastectomy result in equivalent survival rates. Given this, the principal goals of breastconserving treatment are to use techniques of surgery and radiation therapy which will provide a high level of local tumor control, a low incidence of complications, and preservation of a satisfactory cosmetic appearance. Studies from our institution (3, 13, 18) and others (5, 7, 8, 11, 22,29) have analyzed the results in treated patients in an attempt to better achieve these goals. In particular, we have previously identified a number of radiation treatment factors associated with a worsening of the cosmetic result. These include the use of doses to the whole breast greater than 50 Gy, fraction sizes greater than 2 Gy per day, im-
Presented at the 33rd Annual Meeting of the American Society for Therapeutic Radiology and Oncology, Washington, November 199 1.
Reprint requests to: Jay R. Harris, M.D., Joint Center for Radiation Therapy, 50 Binney St., Boston, MA 02 115. Accepted for publication 2 1 February 1992.
DC, 925
926
I. J. Radiation Oncology 0 Biology 0 Physics
METHODS
AND MATERIALS
Between 1968 and 1985, a total of 1624 women with UICC-AJCC clinical Stage I or II unilateral invasive breast carcinoma were treated with conservative surgery (CS) and radiation therapy (RT) at the Joint Center for Radiation Therapy (JCRT). Of these, 1360 underwent gross excision of the tumor and received a total dose to the primary tumor site of greater than or equal to 60 Gy using either an interstitial implant or an electron boost. Two hundred and one patients were excluded from the cosmetic analysis because cosmetic follow-up was not undertaken at this institution. The study population thus consisted of 1159 patients, all of whom were treated after January 1970. A total of 504 patients were treated through 1981 (“earlier cohort”) and 655 were treated between I982 and 1985 (“latter cohort”). The patient characteristics and treatment factors for patients treated in the two time periods are shown in Table 1. In aggregate, there were 679 T 1 tumors (59%) and 480 T2 tumors (410/o),and this distribution was similar in the two time intervals. All patients underwent a complete gross tumor excision, but the total amount of tissue removed at surgery varied from an excisional biopsy, removing the tumor surrounded by a small margin of normal tissue, to a wide local excision, sometimes achieved by a re-excision. Thirty-three patients (7%) had a re-excision in the earlier cohort whereas 245 patients (37%) underwent re-excision in the latter cohort (p < 0.0001). The volume of breast tissue resected was quantified for 44 1 of the 655 patients (67%) in the latter cohort, by multiplying the length by the width by the depth of the resected breast specimen as recorded on the pathology report. If more than one specimen was received, the volumes were calculated separately and then summed. Patients were divided into three groups of equal numbers based on the volume of breast tissue resected (I 35 cm’, 36-85 cm3, and 2 86 cm3). The median volume of breast tissue resected was 40 cm3 (range, 0.4-443 cm3> for Tl tumors, 60 cm3 (range, 4.5-651 cm3) for T2 tumors less than or equal to 3 cm, and 85 cm3 (range, 8-315 cm”) for T2 tumors greater than 3 cm.
Table 1. Treatment 1970-1981 No. patients Axill. diss. Reexcision Adj. CT 6orSMV 3-field Implant Whole breast dose 18 Gy
characteristics 1982-1985
p value
504 77% 7% 24% 4% 85% 95%
655 79% 37% 27% 27% 55% 47%)
NS < 0.000 NS < 0.000 < 0.000 i 0.000
67% 85%
95% 42%
< 0.000 1 < 0.000 1
Axill. diss. = Axillary dissection; motherapy.
Adj. CT = Adjuvant
1
1 I 1
che-
Volume 23. Number 5. 1992
Axillary dissection was performed in 77% of patients in the earlier cohort and 79% of the latter cohort (p = ns). The dissections, in general, were limited to removal of level I and II lymph nodes. Positive nodes were found in 35% of patients (135/390) in the earlier group and 38% of patients ( 197/520) in the latter group (p = ns). During the earlier time interval, 38 1 patients received RT alone and 123 patients (24%) were given adjuvant chemotherapy. During the latter period, 481 women received RT alone, and 174 patients (27%) underwent RT and adjuvant chemotherapy. In 8 1 cases (42 in the earlier cohort and 39 in the latter one), radiation and chemotherapy were delivered concurrently. Since we had demonstrated in a previous study that concurrent chemotherapy has an adverse effect on cosmetic outcome (12) those patients were excluded from the cosmetic analysis. Between 1970 and 198 1,483 patients were treated with a 4 MV linear accelerator and 21 women (4%) on a 6 or 8 MV machine. Between 1982 and 1985, 467 women were treated with 4 MV and 169 (27%) with 6 or 8 MV. The machine energy was not recorded for 19 patients in the latter group. The volume irradiated with external beam treatment varied considerably during the 15 years of the study. Between 1970 and 198 1, 429 patients (85%) received irradiation to the breast and regional nodes, usually with a three-field technique, irrespective of whether an axillary dissection was performed or not. Between 1982 and 1985, patients less commonly had lymph node irradiation. In general, patients without metastases to axillary nodes were treated with tangential breast irradiation only. In addition, in this period, the three-field technique was modified to provide a more accurate match between the opposed tangential fields which treated the breast and the direct anterior field which treated the nodes (26). The median dose to the breast was 46 Gy (range, 3852 Gy) in the time-period 1970-81. One hundred sixtyseven patients (33%) received a breast dose 2 50 Gy. Between 1982 and 1985, the median dose to the breast was 46 Gy (range, 38-51 Gy), but only 32 patients (5%) received a dose r 50 Gy. Also, in the latter time interval, patients treated with adjuvant therapy more commonly received 1.8 Gy 5 times per week for a total dose of 45 Gy, rather than 2 Gy 5 times per week for a total dose of 46 Gy. All patients received an additional dose (“boost”) to the primary site. Iridium 192 interstitial implant was used for 95% of patients in the early period and for 47% of patients treated between 1982 and 1985. Other patients received an electron beam boost with energies ranging from 7 to 11 MeV selected so that the 80% isodose curve would encompass the boost volume. No patient in this series received both interstitial and electron beam boost. The median boost (electron or implant) dose was 20.80 Gy in the earlier cohort (range, 10-27.6 Gy) and 18 Gy in the latter cohort (range, lo-24 Gy). Of note, 85% of patients (430/504) treated in the earlier time interval received more than 18 Gy by boost, compared to 42%
927
Cosmetic results following CS and RT 0 A. DE LA R~CHEFORDI~RE etal.
of patients (272/655) treated in the second period (p < 0.0001). The median implant dose was 20.7 Gy in the earlier group (range, 10-27.6 Gy) and 20 Gy (range, 11.1-24 Gy) in the latter group. The implant boost dose was greater than or equal to 24 Gy for 5% of patients treated in the earlier period and 0.3% of patients in the later period. Follow-up was generally performed at 6-month intervals after completion of the treatment. Patients were censored from cosmetic follow-up at the time of local or distant failure or the diagnosis of opposite breast cancer. The median follow-up time for surviving patients was 107 months (range, 52-2 19) for the earlier cohort and 67 months (range, 12-94) for the latter one. Overall cosmetic results were classified as excellent, good, fair, or poor by the examining physician. An “excellent” score was assigned when the treated breast looked essentially the same as the opposite breast. A “good” cosmetic score was used to describe minimal but identifiable effects of treatment. An “acceptable” result was defined as either an “excellent” or a “good” result. A “fair” cosmetic score was given when significant effects of radiation on the breast were noted. A “poor” cosmetic score was used to describe severe normal tissue sequelae. An “unacceptable” result was defined as either a “fair” or a “poor” score. The cosmetic results were plotted graphically by grouping observations into 12-month intervals. If more than one observation was made on the same patient during a time-interval, the worst observation was scored. The midpoint of each time interval was used in all curves, thus individual observations ranged up to 6 months on either side of each time point. Analysis and comparisons between subgroups were performed at the specific time points of 3 and 5 years because we have previously demonstrated that results tend to stabilize at about 3 years (3). Statistical
comparisons between groups and proportions were calculated at fixed points using the Fisher exact test, and p values of 0.05 or less were considered statistically significant. RESULTS
We compared the overall cosmetic outcome during the two treatment time-periods ( 1970-8 1 and 1982-85). The results were significantly better at both 3 and 5 years for the patients treated with the techniques used in the latter period. An excellent result at 3 years was scored for 58% of the 330 patients in the earlier cohort evaluated at that time-point, compared with 73% of the 352 patients in the latter cohort (p < 0.0001). An acceptable result (either excellent or good) was scored in 86% and 96% of patients, respectively (p < 0.0001). Fair results were seen in 10% and 3.5%, and poor results in 4% and 0.5%, respectively (p < 0.000 1). At 5 years, excellent results were scored for 59% of the 204 patients evaluated in the earlier cohort compared with 74% of the 221 patients evaluated in the latter group (p = 0.002). Acceptable results were found in 84% and 94% (p = 0.002) (Fig. I), fair results in 11% and 5% (p = 0.04), and poor results in 5% and l%, respectively (p = 0.03). Of note, the 5-year actuarial rate of recurrence in the treated breast was 10% for patients treated between 1970-8 1 and 8% for patients treated between 1982-85 (p = ns). We then focused on the patients treated in the more recent period to determine the influence of various treatment factors on the cosmetic outcome (Table 2). Patients in the 1982-85 cohort who received adjuvant chemotherapy (given sequentially) were less likely to have excellent cosmetic scores than patients who did not receive adjuvant chemotherapy (Fig. 2). At 3 years, 77% of patients (214/279) who did not receive chemotherapy had
04
I
12
24
Fig. 1. Percentage of excellent or good cosmetic p < 0.000 1. At 5 years: p < 0.002.
36 MONTHS
48
60
72
results for patients treated during the two time-periods.
At 3 years:
928
1. J. Radiation Oncology 0 Biology 0 Physics
Table 2. Factors influencing
outcome
at 3 years (1982- 1985)
% excellent
279 57
77 65
0.09
97 95
NS
85 78
86 78
0.0009
96 97
NS
70 141 45
51 84 68
0.03
94 99 91
NS
21
62
TTV I 35 cm3 TTV = 36-85 cm3 TTV 2 86 cm3 T
23, pp.
925-93
I
0360-3016/92 $5.00 + .oO Copyright 0 1992 Pergamon Press Ltd.
??Clinical Original Contribution
ARE COSMETIC RESULTS FOLLOWING CONSERVATIVE SURGERY AND RADIATION THERAPY FOR EARLY BREAST CANCER DEPENDENT ON TECHNIQUE? ANNE DE LA ROCHEFORDI~RE, FRANK VICINI, M.D.,
M.D., ANTHONY L. ABNER, M.D., BARBARA SILVER, B.A., ABRAM RECHT, M.D. AND JAY R. HARRIS, M.D.
Joint Center for Radiation Therapy, Department of Radiation Oncology, Harvard Medical School, Boston, MA To assess the cosmetic results in relation to treatment technique, we retrospectively reviewed the results for 1159 Stage I-II breast cancer patients treated with conservative surgery and radiotherapy between 1970-1985. All patients underwent gross excision followed by radiation therapy including au implant or electron beam boost. The total dose to the primary site was 2 60 Gy. Because of technical modifications introduced over time after 1981, the population was divided arbitrarily into two cohorts: 504 patients treated through 1981 and 655 treated between 1982-1985. Median follow-up times for surviving patients in the two cohorts were 107 months and 67 months, respectively. Cosmetic outcome was evaluated by the examining physician and scored as excellent, good, fair or poor. Excellent results at 5 years were scored in 59% of early cohort patients and 74% of the latter cohort (p = 0.002). Acceptable results (either good or excellent) were seen in 84% and 948, respectively (p = 0.02). In the latter cohort, the likelihood of achieving an excellent result, but not an acceptable result, was significantly related to the volume of resected breast tissue and the use of chemotherapy. The number of fields (three-field technique, provided that fields are precisely matched, compared to tangents only) and boost type (implant vs electrons) did not influence the cosmetic outcome. We conclude that our current technique using breast RT to 45-
46 Gy and a boost to the primary site of 16-18 Gy is associated with a high likelihood of acceptable cosmetic results and that this likelihood is not diminished by the use of adjuvant chemotherapy, a large breast resection, the use of a third field, or boost type. Breast cancer, Conservative Surgery, Radiotherapy, Cosmesis. INTRODUCIION
precise matching of the tangential fields and the anterior field to treat the axillary and/or supraclavicular region, and the use of a large volume boost or a boost dose greater than 18 Gy. Based on these findings, a number of treatment modifications were introduced over time at our institution beginning in 1982. Sufficient time has now transpired to allow us to assess the results of these modifications. Thus, in this article, we divided arbitrarily the entire population into two cohorts of patients. We compare the cosmetic results in patients treated prior to 1982 to those in patients treated between 1982 and 1985, and assess treatment factors which are now associated with a worsening of the cosmetic appearance given these treatment modifications. This information, along with information on factors associated with local recurrence and complications, may be useful in providing patients with an improved technique of breast-conserving treatment.
Both retrospective (2, 6, 10, 16, 21, 24) and randomized clinical trials (9, 25,27) have established that breast-conserving treatment (consisting of conservative surgery and radiation therapy) and mastectomy result in equivalent survival rates. Given this, the principal goals of breastconserving treatment are to use techniques of surgery and radiation therapy which will provide a high level of local tumor control, a low incidence of complications, and preservation of a satisfactory cosmetic appearance. Studies from our institution (3, 13, 18) and others (5, 7, 8, 11, 22,29) have analyzed the results in treated patients in an attempt to better achieve these goals. In particular, we have previously identified a number of radiation treatment factors associated with a worsening of the cosmetic result. These include the use of doses to the whole breast greater than 50 Gy, fraction sizes greater than 2 Gy per day, im-
Presented at the 33rd Annual Meeting of the American Society for Therapeutic Radiology and Oncology, Washington, November 199 1.
Reprint requests to: Jay R. Harris, M.D., Joint Center for Radiation Therapy, 50 Binney St., Boston, MA 02 115. Accepted for publication 2 1 February 1992.
DC, 925
926
I. J. Radiation Oncology 0 Biology 0 Physics
METHODS
AND MATERIALS
Between 1968 and 1985, a total of 1624 women with UICC-AJCC clinical Stage I or II unilateral invasive breast carcinoma were treated with conservative surgery (CS) and radiation therapy (RT) at the Joint Center for Radiation Therapy (JCRT). Of these, 1360 underwent gross excision of the tumor and received a total dose to the primary tumor site of greater than or equal to 60 Gy using either an interstitial implant or an electron boost. Two hundred and one patients were excluded from the cosmetic analysis because cosmetic follow-up was not undertaken at this institution. The study population thus consisted of 1159 patients, all of whom were treated after January 1970. A total of 504 patients were treated through 1981 (“earlier cohort”) and 655 were treated between I982 and 1985 (“latter cohort”). The patient characteristics and treatment factors for patients treated in the two time periods are shown in Table 1. In aggregate, there were 679 T 1 tumors (59%) and 480 T2 tumors (410/o),and this distribution was similar in the two time intervals. All patients underwent a complete gross tumor excision, but the total amount of tissue removed at surgery varied from an excisional biopsy, removing the tumor surrounded by a small margin of normal tissue, to a wide local excision, sometimes achieved by a re-excision. Thirty-three patients (7%) had a re-excision in the earlier cohort whereas 245 patients (37%) underwent re-excision in the latter cohort (p < 0.0001). The volume of breast tissue resected was quantified for 44 1 of the 655 patients (67%) in the latter cohort, by multiplying the length by the width by the depth of the resected breast specimen as recorded on the pathology report. If more than one specimen was received, the volumes were calculated separately and then summed. Patients were divided into three groups of equal numbers based on the volume of breast tissue resected (I 35 cm’, 36-85 cm3, and 2 86 cm3). The median volume of breast tissue resected was 40 cm3 (range, 0.4-443 cm3> for Tl tumors, 60 cm3 (range, 4.5-651 cm3) for T2 tumors less than or equal to 3 cm, and 85 cm3 (range, 8-315 cm”) for T2 tumors greater than 3 cm.
Table 1. Treatment 1970-1981 No. patients Axill. diss. Reexcision Adj. CT 6orSMV 3-field Implant Whole breast dose 18 Gy
characteristics 1982-1985
p value
504 77% 7% 24% 4% 85% 95%
655 79% 37% 27% 27% 55% 47%)
NS < 0.000 NS < 0.000 < 0.000 i 0.000
67% 85%
95% 42%
< 0.000 1 < 0.000 1
Axill. diss. = Axillary dissection; motherapy.
Adj. CT = Adjuvant
1
1 I 1
che-
Volume 23. Number 5. 1992
Axillary dissection was performed in 77% of patients in the earlier cohort and 79% of the latter cohort (p = ns). The dissections, in general, were limited to removal of level I and II lymph nodes. Positive nodes were found in 35% of patients (135/390) in the earlier group and 38% of patients ( 197/520) in the latter group (p = ns). During the earlier time interval, 38 1 patients received RT alone and 123 patients (24%) were given adjuvant chemotherapy. During the latter period, 481 women received RT alone, and 174 patients (27%) underwent RT and adjuvant chemotherapy. In 8 1 cases (42 in the earlier cohort and 39 in the latter one), radiation and chemotherapy were delivered concurrently. Since we had demonstrated in a previous study that concurrent chemotherapy has an adverse effect on cosmetic outcome (12) those patients were excluded from the cosmetic analysis. Between 1970 and 198 1,483 patients were treated with a 4 MV linear accelerator and 21 women (4%) on a 6 or 8 MV machine. Between 1982 and 1985, 467 women were treated with 4 MV and 169 (27%) with 6 or 8 MV. The machine energy was not recorded for 19 patients in the latter group. The volume irradiated with external beam treatment varied considerably during the 15 years of the study. Between 1970 and 198 1, 429 patients (85%) received irradiation to the breast and regional nodes, usually with a three-field technique, irrespective of whether an axillary dissection was performed or not. Between 1982 and 1985, patients less commonly had lymph node irradiation. In general, patients without metastases to axillary nodes were treated with tangential breast irradiation only. In addition, in this period, the three-field technique was modified to provide a more accurate match between the opposed tangential fields which treated the breast and the direct anterior field which treated the nodes (26). The median dose to the breast was 46 Gy (range, 3852 Gy) in the time-period 1970-81. One hundred sixtyseven patients (33%) received a breast dose 2 50 Gy. Between 1982 and 1985, the median dose to the breast was 46 Gy (range, 38-51 Gy), but only 32 patients (5%) received a dose r 50 Gy. Also, in the latter time interval, patients treated with adjuvant therapy more commonly received 1.8 Gy 5 times per week for a total dose of 45 Gy, rather than 2 Gy 5 times per week for a total dose of 46 Gy. All patients received an additional dose (“boost”) to the primary site. Iridium 192 interstitial implant was used for 95% of patients in the early period and for 47% of patients treated between 1982 and 1985. Other patients received an electron beam boost with energies ranging from 7 to 11 MeV selected so that the 80% isodose curve would encompass the boost volume. No patient in this series received both interstitial and electron beam boost. The median boost (electron or implant) dose was 20.80 Gy in the earlier cohort (range, 10-27.6 Gy) and 18 Gy in the latter cohort (range, lo-24 Gy). Of note, 85% of patients (430/504) treated in the earlier time interval received more than 18 Gy by boost, compared to 42%
927
Cosmetic results following CS and RT 0 A. DE LA R~CHEFORDI~RE etal.
of patients (272/655) treated in the second period (p < 0.0001). The median implant dose was 20.7 Gy in the earlier group (range, 10-27.6 Gy) and 20 Gy (range, 11.1-24 Gy) in the latter group. The implant boost dose was greater than or equal to 24 Gy for 5% of patients treated in the earlier period and 0.3% of patients in the later period. Follow-up was generally performed at 6-month intervals after completion of the treatment. Patients were censored from cosmetic follow-up at the time of local or distant failure or the diagnosis of opposite breast cancer. The median follow-up time for surviving patients was 107 months (range, 52-2 19) for the earlier cohort and 67 months (range, 12-94) for the latter one. Overall cosmetic results were classified as excellent, good, fair, or poor by the examining physician. An “excellent” score was assigned when the treated breast looked essentially the same as the opposite breast. A “good” cosmetic score was used to describe minimal but identifiable effects of treatment. An “acceptable” result was defined as either an “excellent” or a “good” result. A “fair” cosmetic score was given when significant effects of radiation on the breast were noted. A “poor” cosmetic score was used to describe severe normal tissue sequelae. An “unacceptable” result was defined as either a “fair” or a “poor” score. The cosmetic results were plotted graphically by grouping observations into 12-month intervals. If more than one observation was made on the same patient during a time-interval, the worst observation was scored. The midpoint of each time interval was used in all curves, thus individual observations ranged up to 6 months on either side of each time point. Analysis and comparisons between subgroups were performed at the specific time points of 3 and 5 years because we have previously demonstrated that results tend to stabilize at about 3 years (3). Statistical
comparisons between groups and proportions were calculated at fixed points using the Fisher exact test, and p values of 0.05 or less were considered statistically significant. RESULTS
We compared the overall cosmetic outcome during the two treatment time-periods ( 1970-8 1 and 1982-85). The results were significantly better at both 3 and 5 years for the patients treated with the techniques used in the latter period. An excellent result at 3 years was scored for 58% of the 330 patients in the earlier cohort evaluated at that time-point, compared with 73% of the 352 patients in the latter cohort (p < 0.0001). An acceptable result (either excellent or good) was scored in 86% and 96% of patients, respectively (p < 0.0001). Fair results were seen in 10% and 3.5%, and poor results in 4% and 0.5%, respectively (p < 0.000 1). At 5 years, excellent results were scored for 59% of the 204 patients evaluated in the earlier cohort compared with 74% of the 221 patients evaluated in the latter group (p = 0.002). Acceptable results were found in 84% and 94% (p = 0.002) (Fig. I), fair results in 11% and 5% (p = 0.04), and poor results in 5% and l%, respectively (p = 0.03). Of note, the 5-year actuarial rate of recurrence in the treated breast was 10% for patients treated between 1970-8 1 and 8% for patients treated between 1982-85 (p = ns). We then focused on the patients treated in the more recent period to determine the influence of various treatment factors on the cosmetic outcome (Table 2). Patients in the 1982-85 cohort who received adjuvant chemotherapy (given sequentially) were less likely to have excellent cosmetic scores than patients who did not receive adjuvant chemotherapy (Fig. 2). At 3 years, 77% of patients (214/279) who did not receive chemotherapy had
04
I
12
24
Fig. 1. Percentage of excellent or good cosmetic p < 0.000 1. At 5 years: p < 0.002.
36 MONTHS
48
60
72
results for patients treated during the two time-periods.
At 3 years:
928
1. J. Radiation Oncology 0 Biology 0 Physics
Table 2. Factors influencing
outcome
at 3 years (1982- 1985)
% excellent
279 57
77 65
0.09
97 95
NS
85 78
86 78
0.0009
96 97
NS
70 141 45
51 84 68
0.03
94 99 91
NS
21
62
TTV I 35 cm3 TTV = 36-85 cm3 TTV 2 86 cm3 T
