Fax: (248) 551-0089
Long-term cosmetic results and toxicity after accelerated partial-breast irradiation
A method of radiation delivery by interstitial brachytherapy for the treatment of early-stage breast carcinoma
Article first published online: 18 JAN 2006
Copyright © 2006 American Cancer Society
Volume 106, Issue 5, pages 991–999, 1 March 2006
How to Cite
Chen, P. Y., Vicini, F. A., Benitez, P., Kestin, L. L., Wallace, M., Mitchell, C., Pettinga, J. and Martinez, A. A. (2006), Long-term cosmetic results and toxicity after accelerated partial-breast irradiation. Cancer, 106: 991–999. doi: 10.1002/cncr.21681
- Issue published online: 17 FEB 2006
- Article first published online: 18 JAN 2006
- Manuscript Accepted: 20 SEP 2005
- Manuscript Revised: 18 AUG 2005
- Manuscript Received: 20 MAY 2005
- accelerated partial breast irradiation;
The objective of this study was to assess the cosmesis and toxicities in patients with early-stage breast carcinoma who received treatment with accelerated partial breast irradiation (APBI) using interstitial brachytherapy.
From April 1993 to December 2001, 199 patients with Stage I–II breast carcinoma received breast-conserving therapy with APBI to the tumor bed alone through a low-dose-rate (LDR) or high-dose-rate (HDR) implant. A template guide was used. The LDR dose was 50 Gray (Gy) over 96 hours; the outpatient HDR implant delivered 32 Gy in 8-Gy or 34 Gy in 10-Gy twice-daily fractions. Cosmesis (Harvard criteria) and toxicities (Radiation Therapy Oncology Group guidelines) were assessed at ≤ 6 months, 2 years, and 5 years.
The median follow up was 6.4 years. Breast pain, edema, erythema, and hyperpigmentation all diminished over time. Breast fibrosis and hypopigmentation increased until the 2-year mark and then stabilized. Fat necrosis and telangiectasia increased over time, with a fat necrosis rate of 11% at 5 years. Nearly all telangiectasias (34% at 5 yrs) were Grade 1 (< 2 mm). The remaining toxicities were Grade 1 at all follow-up intervals. Infections (11%) occurred predominantly within the first month after treatment. Good-to-excellent cosmetic outcomes were noted in 95–99% of patients over time; cosmetic results stabilized at 2 years with excellent results increased out to 5 years.
APBI with interstitial brachytherapy resulted in mild chronic toxicities, the majority of which diminished or reached a plateau over time. Long-term cosmesis was good to excellent in 95–99% of patients and stabilized at 2 years. Cancer 2006. © 2006 American Cancer Society.
Breast-conserving therapy (BCT) has produced survival equivalent to mastectomy in the treatment of patients with early-stage invasive breast carcinoma in several randomized Phase III clinical trials.1, 2 However, the necessity to treat the whole breast in women who are eligible for BCT has never been investigated thoroughly. Conceptually, whole-breast irradiation has been given to address presumed microscopic occult disease.3 Such an assumption leads to the standard dose-fractionation scheme of treatment over 6.0–6.5 weeks, a time commitment that may dissuade some otherwise eligible women to not undergo BCT. In addition, many of the toxicities associated with BCT are related directly to treatment of the entire breast.
To circumvent both the protracted time course and toxicities of whole-breast irradiation, investigators have studied the use of partial breast irradiation in an accelerated hypofractionation scheme as an alternative form of treatment for selected patients.4–14 With such accelerated treatment delivered over 4–5 days, early published results of accelerated partial breast irradiation (APBI) from 2 United States institutions established its relative safety.4–6, 11 More centers began treating appropriately selected patients with such hypofractionated accelerated treatments as interest in APBI increased. Aside from the advantage gained from the obvious time compression of APBI treatments, the limited volume of breast irradiated has dosimetric advantages of delivering minimal to no dose to normal surrounding structures, such as the underlying lung, and, for left-sided lesions, little to no dose to the heart. The 1-week perioperative treatment with APBI also may eliminate the delay in starting systemic adjuvant therapy among patients who are deemed appropriate candidates for chemotherapy.
The efficacy of such APBI techniques has been reported from several institutions with median follow-up > 5 years reported from some centers.8, 11, 14 If local control and survival are comparable between standard BCT and APBI, then long-term cosmetic outcomes and toxicities after APBI increasingly will become important. The objective of the current study was to assess the early and late toxicities as well as cosmetic outcomes at specified time points in patients with early-stage breast carcinoma who received treatment with interstitial brachytherapy in an accelerated hypofractionation scheme.
MATERIALS AND METHODS
One hundred ninety-nine consecutive patients with invasive early-stage breast carcinoma were treated prospectively at William Beaumont Hospital (WBH) with hypofractionated APBI using interstitial brachytherapy directed only at the region of the tumor bed as part of their BCT from 1993 to 2001. The study population included 158 women who were enrolled prospectively to undergo one of three Institutional Review Board-approved partial breast irradiation protocols. All patients signed informed consent forms prior to treatment. All eligible patients had Stage I–II breast carcinoma as defined by the International Union Against Cancer fifth edition guidelines and had undergone macroscopic total resection of the primary tumor. Reexcision to achieve negative surgical margins was performed as needed to obtain margins ≥ 2 mm; 111 patients (56%) underwent such reexcisional surgery. Eligibility criteria for the 3 partial breast irradiation protocols included infiltrating ductal carcinomas that measured < 3 cm in greatest dimension, negative surgical margins ≥ 2 mm, older than age 40 years, and surgically staged axilla with ≤ 3 positive lymph nodes; in 1997, the last criterion was changed to negative lymph nodes based on the documented survival benefit of regional radiotherapy (RT) along with local RT in lymph node-positive, postmastectomy patients from the Danish and British Columbia Trials.15, 16 Patients who had an extensive intraductal component, infiltrating lobular histology, ductal carcinoma in situ (DCIS), or clinically significant areas of lobular carcinoma in situ were excluded.
An additional 41 patients who did not meet all eligibility criteria for the partial breast irradiation protocol also were included in our analysis. These patients were given the opportunity to be treated with partial breast irradiation for compassionate reasons, such as refusing mastectomy and/or refusing conventional whole-breast RT. Reasons for not meeting the eligibility criteria for partial breast irradiation protocols in these 41 patients included negative but close margins (0–2 mm; 4 patients; 2%), noninfiltrating ductal histology (21 patients; 11%), timing of partial breast irradiation after lumpectomy (4 patients; 2%), participation in other studies (8 patients; 4%), or unspecified reasons (4 patients; 2%). These additional nonprotocol patients were added to this analysis, because the reasons for their ineligibility were minor, and their inclusion likely would not affect the recurrence rate.
Patients were followed every 3 months by their radiation oncologist or surgeon during the first 2 years after treatment and every 6 months thereafter. Baseline mammography was performed at 6 months after the completion of partial breast irradiation and yearly thereafter. Follow-up was complete through October 2004.
All patients were treated with template-guided radiation techniques according to the guidelines of one of three different protocols for APBI. In total, 120 patients (60%) were treated as inpatients with low-dose-rate (LDR) iodine-125 implants that delivered 50 grays (Gy) over 96 hours at 0.52 Gy per hour. Seventy-nine patients (40%) were treated as outpatients with a high-dose-rate (HDR) implant that delivered either 32 Gy in 8 fractions (separated by 6 hours; 71 patients) or 34 Gy in 10 fractions (separated by 6 hours; 8 patients). Regardless of the brachytherapy technique used, every implant was designed to irradiate the lumpectomy cavity with at least a 1–2 cm margin. Quality control criteria for these patients have been reported previously.17, 18 In total, 139 patients (70%) received adjuvant therapy after they completed APBI: Twenty-five patients (13%) received adjuvant systemic chemotherapy, and 114 patients (57%) received adjuvant tamoxifen therapy.
The toxicities and cosmesis were assessed at 3 defined time points: at ≤ 6 months, at 2 years, and at 5 years of follow-up. The toxicity parameters examined included the following: breast edema, erythema, fibrosis, hyperpigmentation, hypopigmentation, breast pain, breast infection, telangiectasia, and fat necrosis. Toxicities were graded by using the Radiation Therapy Oncology Group (RTOG)/Eastern Cooperative Oncology Group (ECOG) late radiation morbidity scoring scheme19 for skin, subcutaneous tissues, pain, radiation dermatitis, and dermatology/skin from the Common Toxicity Criteria, version 2.0.20 Breast edema, erythema, pigmentary changes, and telangiectasia fell under the domain of radiation dermatitis and skin-late RT morbidity scoring. Breast fibrosis and breast pain were under the domains of subcutaneous tissues-late RT morbidity scoring and pain due to radiation, respectively. Breast infections and fat necrosis were either present or not and were noted accordingly. In accordance with the guidelines of Common Toxicity Criteria, version 2.0, toxicities were graded by using the acute/chronic radiation morbidity scale: Grade 0, no observable radiation effects; Grade 1, mild radiation effects; Grade 2, moderate radiation effects; and Grade 3, severe radiation effects. Cosmetic evaluation was based on the standards set forth by the Harvard criteria.21 An excellent score was given when the treated breast looked essentially the same as the contralateral, untreated breast. A good score was assigned for minimal but identifiable radiation effects of the treated breast. Scoring a fair result meant significant radiation effects readily observable. A poor score was used for severe sequelae of normal tissue. The treating physician at a regularly scheduled follow-up visit scored the cosmetic result. No patient-reported scoring of cosmetic outcome was done. Likewise, the treating radiation oncologist did all toxicity scoring for each patient.
For this study, patient charts were reviewed retrospectively for radiation-related morbidities and cosmetic results. Data for both toxicities and cosmesis were recorded and analyzed at the most recent follow-up visit and at more remote visits corresponding to the 3 intervals of ≤ 6 months, 2 years, and 5 years.
The statistical method employed for the incidence/severity of toxicities and cosmetic outcome with various parameters was Pearson chi-square analysis stratified for no toxicity versus any toxicity.
Patient and treatment-related characteristics of the patients who received hypofractionated APBI are displayed in Table 1. The median follow-up was 6.4 years (range, 0.3–10.6 yrs) for all patients. Sixty-two patients (31%) were followed for at least 8 years, 84 patients (42%) were followed for at least 7 years, 134 patients (67%) were followed for at least 5 years, and 190 patients (95%) were followed for at least 2 years. No patient was lost to follow-up.
|Characteristic||No. of patients (%)|
|Mammographically detected||139 (70)|
|< 10 mm||78 (39)|
|< 20 mm||183 (92)|
|Lymph node status (n = 195 patients)|
|Grade 1–2||157 (79)|
|Grade 3||42 (21)|
|Margins of excision (n = 188 patients)|
|From ≥ 2 mm to < 5 mm||51 (27)|
|From ≥ 5 mm to < 10 mm||26 (14)|
|> 10 mm||111 (59)|
|Adjuvant chemotherapy||25 (13)|
|Adjuvant tamoxifen therapy||114 (57)|
In reference to toxicities, breast pain, edema, erythema, and hyperpigmentation all diminished in frequency over time (Table 2). At the 6-month, 2-year, and 5-year time points, breast pain, nearly all of which was mild (Grade 1), diminished from 27% at 6 months down to 8% by 5 years. Breast edema likewise was predominately mild and went from 50% (6 mos) to 12% (2 yrs) to 6% by the 5-year assessment time. Similarly, mild (Grade 1) erythema decreased from 35% to 11% at 2 years with stabilization thereafter. Mild hyperpigmentation (Grade 1) followed a similar downward trend in frequency: from 67% at 6 months to 37% at 5 years. All of these trends were analyzed statistically by using Pearson chi-square analysis, and it was determined that they were not chance occurrences.
|Toxicity||Interval (% of patients)|
|At ≤ 6 mos (n = 165 patients)||At 2 yrs (n = 128 patients)||Follow-up ≥ 5 yrs (n = 79)|
|Grade 1||Grade 2||Grade 3||Grade 1||Grade 2||Grade 3||Grade 1||Grade 2||Grade 3|
Sequelae that increased until the 2-year mark and later stabilized included breast fibrosis (from 22% at 6 mos, to 48% at 2 yrs, to 46% at 5 yrs) and hypopigmentation (from 18%, to 34%, to 38%, respectively). It is noteworthy that any slight degree of periscar induration was scored as mild fibrosis, regardless of whether or not postsurgical changes may have contributed. Nearly all pigmentary changes, whether hyperpigmentation or hypopigmentation, were mild and pinpoint rather than diffuse, corresponding to the sites where the LDR catheters or HDR needles had been placed. Likewise, chi-square analysis verified these trends. The time-course trend of hypopigmentation followed that of fibrosis with an increase in frequency out to 2 years with subsequent stabilization.
Fat necrosis and telangiectasia increased with the passage of time, with fat necrosis increasing from 1% at ≤ 6 months to 9% at 2 years and 11% at 5 years. The median time to occurrence of fat necrosis was 5.5 years (range, 0.25–8.2 yrs) (Table 3). Most fat necroses were asymptomatic (32 of 41 patients; 78%) and were detected mammographically (28 of 41 patients; 68%), usually by the presence of coarse, large, dystrophic calcifications or circumscribed, fat-containing masses associated in some patients with coarse, peripheral calcifications. Of the 9 patients who had symptomatic fat necrosis, 3 patients (7.3%) underwent surgical excision, and 1 patient (2.4%) received steroids. Telangiectasias, nearly all of which were Grade 1, were distributed evenly between LDR and HDR treatment modalities at 5 years (34% for both LDR and HDR; P = 0.983). Typical telangiectasias are documented in Figure 1 and barely were perceptible, usually measuring ≤ 2 mm. Similarly, the remaining toxicities were Grade 1 at all follow-up intervals.
|Toxicity||Interval (% of patients)|
|At ≤ 6 mos (n = 165 patients)||At 2 yrs (n =128 patients)||Follow-up ≥ 5 yrs (n = 79 patients)|
|Grade 1||Grade 2||Grade 3||Grade 1||Grade 2||Grade 3||Grade 1||Grade 2||Grade 3|
|Fat necrosis (% of all patients)a||1||9||11|
Infections (Table 4) that required outpatient antimicrobial therapy were seen in 11% of the patients: Seven percent were acute infections (< 1 mo), and in 4% were delayed infections (> 1 mo). All infections were confined to the breast with no axillary infections. In addition to antibiotics, 5 of 22 patients who had infections (2.5%) required incision and drainage or debridement. Those patients who underwent implant placement with an open cavity had a statistically significant increase in the rate of infection compared with patients who underwent implant placement with a closed cavity (P = 0.003).
|Variable||No. of patients (%)||P value|
|Overall infection rate|
|Entire group||22 (11.0)||0.199|
|LDR group||16 (8.0)|
|HDR group||6 (3.0)|
|Time of occurrence|
|< 1 Mo||14 (7.0)||0.240|
|> 1 Mo||8 (4.0)|
|Open cavity||17 (8.5)||0.003a|
|Closed cavity||5 (2.5)|
Good-to-excellent cosmetic outcomes were noted in 95–99% of patients, depending on the time of assessment (Table 5, Figs 2, 3). Early assessment at 6 months revealed a large percentage of good scores (85%). However, between 6 months and 2 years, an increasing number of excellent scores from 10% to 29% were noted. Comparison of cosmetic results at the 2-year and 5-year intervals demonstrated stabilization of scores, and the percentage of excellent results increased at 5 years. Throughout all time points of cosmetic assessment, the rate of good-to-excellent scores was never < 95%.
|Cosmetic outcome||Percentage of patients|
|At < 6 mos (n = 165 patients)a|
|Total (Good + Excellent)||95|
|At 2 yrs (n = 128 patients)a|
|Total (Good + Excellent)||97|
|At > 5 yrs (n = 79 patients)|
|Total (Good + Excellent)||99|
No statistical difference was noted in the incidence/severity of any toxicity or cosmetic outcome with the following parameters: tamoxifen, type of brachytherapy (LDR vs. HDR), and tumor classification (T1 vs. T2 [Pearson chi-square analysis]). However, for patients who received chemotherapy, the incidence of breast erythema at 2 and 5 years and the rates of delayed infections were higher compared with patients who did not receive chemotherapy (P = 0.015, P = 0.016, and P = 0.003, respectively). In addition, cosmetic assessment at 6 months demonstrated statistically more excellent/good results in patients who did not receive chemotherapy compared with patients who received chemotherapy (100% vs. 94%; P = 0.005).
Several institutions have reported that results with hypofractionated APBI have yielded excellent local control rates, disease-free survival rates, and overall survival rates.8, 11, 12, 14, 22 Two institutions have analyzed their prospectively treated patients with accelerated brachytherapy alone and compared them with a matched group of patients who were eligible for brachytherapy alone but who received conventional external beam RT (EBRT).8, 11
In the experience of the Ochsner Clinic from January 1992 through October 1993, 50 patients with 51 breast carcinomas received either an LDR implant (45 Gy over 3.5–6.0 days) or an HDR implant (32 Gy in 8 fractions over 4 days in twice-daily fractions). All patients had tumors that measured < 4 cm in greatest dimension with negative margins.11 Patients who had negative or positive lymph node status were eligible in addition to patients who had DCIS. With a median follow-up of 75 months, 1 in-breast recurrence (2%) and 3 regional lymph node failures (6%) were noted for an 8% local regional recurrence rate. Compared with a contemporary group of 94 patients who qualified for brachytherapy alone but who received traditional whole-breast EBRT instead, the outcome for the 50 patients with 51 breast carcinomas who underwent brachytherapy was similar to the outcome for the EBRT patients, who had a local-regional recurrence rate of 5%.
Similarly, the WBH experience of 199 patients who received brachytherapy alone (120 patients with LDR brachytherapy and 79 patients with HDR brachytherapy), with a median follow-up of 6.4 years, yielded a 5-year actuarial local tumor bed recurrence rate of 1.2%; the elsewhere breast failure rate was 6%.8 To compare potential outcome differences based on the volume of breast irradiated, the patients who received interstitial brachytherapy alone were matched with 199 patients who received whole-breast RT. Match criteria included tumor size, lymph node status, patient age, margins of excision, estrogen receptor status, and use of tamoxifen. The rate of local recurrence did not differ statistically between the 2 patient groups: The whole-breast RT group demonstrated a local recurrence rate of 1% compared with patients who received partial-breast RT, who had a similar 1% risk of local failure (P = 0.65). No significant statistical differences were seen in the 5-year actuarial cause-specific survival rate of 97% versus 97% (P = 0.34) and the overall survival rate of 93% versus 87% (P = 0.23) between the patients who received whole-breast RT and the patients who received partial-breast RT alone, respectively.
The only published Phase III trial that compares APBI by interstitial brachytherapy with whole-breast EBRT is the Hungarian randomized study by Polgar et al.14 At 7 years, their results in terms of ipsilateral breast recurrences did not differ statistically between APBI and whole-breast EBRT. In addition, those authors noted no statistical difference in the 7-year survival endpoints for both treatment groups.
With such data substantiating comparable local control and survival between standard BCT and APBI, long-term toxicities and cosmetic outcomes after APBI become more important. We were able to draw some significant inferences and conclusions from reported cosmetic and toxicity results in patients who received BCT with standard whole-breast, tangential, external-beam fields.
Our 5-year results of cosmesis after APBI document good-to-excellent results in 99% of patients at 5 years with stabilization of cosmetic results at 2 years. It is noteworthy that between years 2 and 5, the actual percentage of excellent cosmesis continues to increase. With regard to toxicity, the majority of episodes are Grade 1, and most such toxicities resolve or stabilize with the passage of time. In the interpretation of the number of telangiectasias seen, we note that coding of such late skin changes included near microscopic, very mild telangiectasias, the majority of which were fine pinpoint changes that were seen at the site where the interstitial needles or catheters had been placed (see Fig. 1, which documents a typical mild case of telangiectasias). Although interobserver interpretation may vary in denoting such changes, we decided to err on the side of conservatism and scored telangiectasias for any, even minute changes.
In reviewing our cosmetic results with those reported for standard BCT, consisting of 6.5 weeks of EBRT, our results corroborate the finding that there is stabilization of cosmesis at the 2–3-year point, with little change out beyond 10 years.9, 21, 23 Considering the stabilization of cosmetic scores with the percentage of excellent results improving over time, the trend suggests that, beyond 10 years, there would be no significant deterioration in the cosmetic outcome.
The toxicity profile of our series is comparable to that reported for conventional BCT delivered over 6.0–6.5 weeks.8 The caveat is that most reported series qualitatively limit toxicity assessment to breast edema, retraction, telangiectasia, and ipsilateral arm edema.21, 23, 24 In the series by Rose et al.,21 those authors noted that the major morbidities of breast edema, retraction, and telangiectasia evolved with different time courses. Breast edema occurred early (within the first yr) and tended to resolve slowly. Breast retraction and telangiectasia occurred somewhat later but did not change significantly between the 3-year and 7-year marks. Other investigators25 have demonstrated an increase in the incidence of telangiectasia with lengthening follow-up. However, in all reports,21, 25–27 most toxicities were low grade and resolved or stabilized with the passage of time, similar to what we observed in the current series.
Various patient-related and treatment-related factors have been correlated with cosmetic and morbidity results in BCT with whole-breast, tangential treatments. The most important and often cited treatment-related factor that leads to both morbidity and compromised cosmesis is the extent of breast surgery required to achieve appropriate margins.21, 28–30 Other treatment-related factors that contribute to morbidity and/or worse cosmetic outcomes are related to parameters that are a function directly of whole-breast external-beam treatments. The number of radiation fields,30 the use of a direct anterior electron field (for boosting),24 greater separation between the tangential fields28 and total whole-breast dose > 50 Gy30–35 are just a few of the factors that lead to greater morbidity. All of these factors are obviated with the brachytherapy approach to treatment. Various reports are conflicting with respect to whether or not chemotherapy, especially given concurrently, affects cosmesis and morbidity.21, 24, 34, 35 In our series, because the brachytherapy treatment was given in a single week with an accelerated fractionation scheme, the use of concurrent chemotherapy and sequencing issues about when chemotherapy should be given became a moot point.
Detailed toxicity and cosmetic outcome data from centers that perform APBI with interstitial brachytherapy are not prevalent. The data from our series were collected retrospectively from a review of patient charts by a single investigator. The results of toxicity, along with cosmetic evaluations, were tabulated from dictated notes that were made by the treating physician for each patient. Toxicities were scored according to the RTOG/ECOG late radiation morbidity scoring scheme19 with the Common Toxicity Criteria, version 2.0.20 This toxicity scoring scheme has been incorporated by other investigators who employ APBI in the treatment of selected patients with early-stage breast carcinoma.13, 14, 22, 36
In assessing toxicities and cosmesis, interpretations of changes like fibrosis and cosmetic outcomes are not entirely objective.27, 37, 38 In our study, grading of fibrosis was based on the degree of induration palpated at the time of each follow-up visit. Because induration dimensions were not always recorded, the grading by the examining clinician became the basis on which the degree of fibrosis was assessed. Fibrotic changes can be difficult to differentiate between sequelae from postsurgical changes and sequelae from radiation effects.12, 27, 39 Indeed, with our reexcision rate of 56%, such second surgical procedures likely contributed to breast fibrosis/induration. Thus, fibrosis is a continuum and a morbidity of both surgical excision and a late radiation effect. It would be difficult to determine the proportional contribution of surgery versus the contribution of radiation that leads to fibrosis. However, we conservatively assigned any degree of induration under subcutaneous tissue-late RT morbidity scoring (fibrosis) solely related to a late radiation sequela.
Based on the constellation of all toxicity factors that we analyzed, the treating clinician for each patient made the assessment of cosmesis. This was done based on the Harvard criteria21; such criteria have been applied similarly by other institutions that use APBI.11, 14, 22 Since late 1999, interval photography has been performed routinely during scheduled outpatient visits with the objective of collecting a library of patient pictures to allow for third-party assessments of cosmetic outcomes.
More insight may be gained from the collective published experiences of selected centers as data from institutes that practice APBI mature. King et al. reported that, among patients who received brachytherapy alone, they observed a 22% incidence of Grade 1 and 2 treatment complications, which consisted of a combination of skin erythema, moist desquamation, telangiectasia, pain, and fibrosis.11 Grade 3 toxicities requiring surgical intervention were uncommon (8%). Of the Grade 3 morbidities, 4% (2 of 51 episodes) consisted of fat necrosis, and both of those patients underwent surgery. Although the temporal course of these toxicities reported by King et al. was not stated, the vast majority (22%) were low-grade episodes, similar to our reported morbidities, which were predominantly Grade 1. Additional published APBI toxicity profiles were limited to descriptions of fibrosis, telangiectasias, and fat necrosis. In two studies, mild-to-moderate, in-field fibrosis was observed in 29.5–42.2% of patients who received APBI, although no time frame was defined for the appearance of fibrosis12, 14; however, severe Grade 3 fibrosis that led to alteration of the cosmetic score was a less frequent event in both studies (9.1% and 2.2%, respectively). Those results of fibrosis found with interstitial APBI are very comparable to our rates of mild fibrosis over time, which varied from 22% (at 6 mos) to 46% (at 5 yrs) with only 1% severe late fibrosis at 5 years. Rates of telangiectasia have ranged from 3.7% to 22.7%, but cosmetic-altering, dense, severe telangiectasias are a rare occurrence with a reported incidence of 4.5%.12 Our 5-year rate of mild, barely perceptible telangiectasias (< 2 mm) was 34%, and we observed no severe angiogenic changes that led to a change in cosmetic scoring. The reported incidence of fat necrosis resulting from APBI has ranged from 2.3% to 24.2%.12, 13 The vast majority of those patients showed resolution with expectant observation. This also was true in our current series, with an 11% 5-year incidence of fat necrosis in which 90% of the affected patients were managed conservatively with observation.
Cosmesis has been reported to be good to excellent in most patients, ranging from 75% (median follow-up for cosmesis, 20 mos11), to 84.4% (7-yr actuarial data14), to 90% for patients who received HDR implants alone (median follow-up, 42 mos12). Those cosmetic results were very similar to our 95–99% good-to-excellent cosmesis observed over time with stabilization at 2 years.
With the successful documentation of the efficacy of APBI with interstitial brachytherapy, other methods of delivering partial-breast irradiation are being explored, such as with the MammoSite RTS® brachytherapy catheter, approved for clinical use by the United States Food and Drug Administration in May 2002.40, 41 In addition, centers are exploring whether brachytherapy interstitial implants (which technically are operator-dependent, requiring the intraoperative placement of temporary after-loaded needles or catheters into the involved breast) can be replaced with 3-dimensional, conformal EBRT (3-D CRT) delivered over 5 days42, 43 or 10 days.44, 45 Such new conformal technology has been investigated by the Radiation Therapy Oncology Group (RTOG) in a Phase I/II trial (RTOG 0319) on partial-breast irradiation using 3-D CRT that completed accrual in late April 2004.
In conclusion, these extended time-interval morbidity and outcome data from 199 selected patients who received treatment at our institution with APBI by interstitial brachytherapy document mild long-term toxicities, the majority of which diminish or reach a plateau over time. Long-term cosmetic results are good to excellent in > 95% of patients and stabilize at 2 years, with continued improvement at extended follow-up intervals beyond treatment completion. Local control, cause-specific survival, and overall survival are comparable to matched controls treated with EBRT. These results of patients who received APBI with interstitial brachytherapy closely parallel the results from several single-institutional experiences along with a Phase III randomized trial, all of which demonstrated its 5-year efficacy in selected patients. Collectively, the reported APBI outcomes from these centers that employ interstitial brachytherapy document that the toxicity profile is safe, the control rates are excellent, and the cosmetic results are good to excellent in the vast majority of patients. Such outcome results have added impetus for a United States-based, randomized, Phase III clinical trial sponsored jointly by the National Surgical Adjuvant Breast Project and the RTOG (which began accrual in March 2005) to provide definitive, Class I evidence of the efficacy of APBI.46
The authors thank Ms. Margaret Calhoun for her assistance in the preparation of this article.
- 9Long-term cosmetic results and toxicity with accelerated partial breast irradiation (APBI) utilizing interstitial brachytherapy. Int J Radiat Oncol Biol Phys. 2003; 57: S309–S310., , , et al.
- 28Patient characteristics and treatment factors affecting cosmesis following lumpectomy and breast irradiation. Am J Clin Oncol. 2003; 26; 4: 350–353., .
- 46Radiation Therapy Oncology Group. RTOG 0413 broadcast, NSAPB B-39/RTOG 0413: “a randomized Phase III study of conventional whole breast irradiation (WBI) versus partial breast irradiation (PBI) for women with Stage 0, I or II breast cancer,” activated March 21, 2005. Available from URL: http://www.rtog.org/members/protocols/0413/0413broadcast.html [accessed April 5, 2005].