- Email: [email protected]
Planning the breast tumor bed boost: Changes in the excision cavity volume and surgical scar location after breast-conserving surgery and whole-breast irradiation
Int. J. Radiation Oncology Biol. Phys., Vol. 66, No. 3, pp. 680 – 686, 2006 Copyright © 2006 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/06/$–see front matter
doi:10.1016/j.ijrobp.2006.04.042
CLINICAL INVESTIGATION
Breast
PLANNING THE BREAST TUMOR BED BOOST: CHANGES IN THE EXCISION CAVITY VOLUME AND SURGICAL SCAR LOCATION AFTER BREAST-CONSERVING SURGERY AND WHOLE-BREAST IRRADIATION KEVIN S. OH, M.D.,* FENG-MING KONG, M.D., PH.D.,* KENT A. GRIFFITH, M.P.H., M.S.,† BETH YANKE, C.M.D.,* AND LORI J. PIERCE, M.D.* *Department of Radiation Oncology, University of Michigan Medical School, Ann Arbor, MI; and †Biostatistics Unit of the University of Michigan Comprehensive Cancer Center, Ann Arbor, MI Purpose: The aims of this study were to determine the changes in breast and excision cavity volumes after whole-breast irradiation and the adequacy of using the surgical scar to guide boost planning. Methods and Materials: A total of 30 women consecutively treated for 31 breast cancers were included in this study. Simulation CT scans were performed before and after whole-breast irradiation. CT breast volumes were delineated using clinically defined borders. Excision cavity volumes were contoured based on surgical clips, the presence of a hematoma, and/or other surgical changes. Hypothetical electron boost plans were generated using the surgical scar with a 3-cm margin and analyzed for coverage. Results: The mean CT breast volumes were 774 and 761 cc ( p ⴝ 0.22), and the excision cavity volumes were 32.1 and 25.1 cc ( p < 0.0001), before and after 40 Gy (39 – 42 Gy) of whole-breast irradiation, respectively. The volume reduction in the excision cavity was inversely correlated with time elapsed since surgery (R ⴝ 0.46, p < 0.01) and body weight (R ⴝ 0.50, p < 0.01). The scar-guided hypothetical plans failed to cover the excision cavity adequately in 62% and 53.8% of cases using the pretreatment and postradiation CTs, respectively. Per the hypothetical plans, the minimum dose to the excision cavity was significantly lower for tumors located in the inner vs. outer quadrants ( p ⴝ 0.02) and for cavities >20 cc vs. <20 cc ( p ⴝ 0.01). Conclusions: This study demonstrates a significant reduction in the volume of the excision cavity during whole-breast irradiation. Scar-guided boost plans provide inadequate coverage of the excision cavity in the majority of cases. © 2006 Elsevier Inc. Breast cancer, Radiotherapy, Excision cavity, Surgical scar, Boost.
INTRODUCTION
recurrence was significantly lower in women receiving a boost (4.3% vs. 7.3%, p ⬍ 0.001). Although the use of boost irradiation is a recommended practice, no standard technique has been established. In a recent Patterns of Care Study for early-stage breast cancer, the boost volume was determined using CT in 11.7%, a clinical fluoroscopic simulator in 43.9%, and clinically alone in 37.2% (3). With recent technologic advancement, CT-based planning has become more popular in practice over the last several years (4). CT-based planning often uses the excision cavity identified on simulation CT scans acquired before the initiation of whole-breast irradiation, approximately 4 to 6 weeks before initiation of the boost. In the interim between CT simulation and boost treatment, changes in the breast contour such as swelling secondary to edema may occur from whole-breast irradiation. These
The benefit of boost irradiation in women undergoing breast-conserving therapy has been supported by two randomized trials to date. A French trial (1) randomized 1,024 women with invasive cancers ⱕ3 cm after excision and 50 Gy to the whole breast to either a 10-Gy boost or no further treatment to the tumor bed. The reported 5-year actuarial rate of local recurrence was 3.6% (with boost) vs. 4.5% (no further treatment, p ⫽ 0.044). The European Organization for Research and Treatment of Cancer (EORTC) conducted a similar trial (2) randomizing patients with Stage I or II breast cancer after margin-negative breast-conserving surgery and whole-breast irradiation to 50 Gy to receive either a 16-Gy boost (2,661 patients) or no further treatment (2,657 patients). Again, the 5-year actuarial rate of local
Presented in part at the 46th Annual Meeting of the American Society for Therapeutic Radiology and Oncology (ASTRO), Atlanta, GA, October 3–7, 2004. Received Feb 3, 2006, and in revised form April 11, 2006. Accepted for publication April 16, 2006.
Note—An online CME test for this article can be taken at www.astro.org under Education and Meetings. Reprint requests to: Lori J. Pierce, M.D., Department of Radiation Oncology, University of Michigan Medical School, Ann Arbor, MI 48109. Tel: (734) 936-7810; Fax: (734) 763-7370; E-mail: [email protected] 680
Excision cavity and surgical scar in planning breast boost irradiation
changes may affect the localization of both the tumor bed and surgical scar. The extent of these changes and their effect on boost planning are unknown. Clinically based boost planning is two-dimensional, with the field largely determined by the surgical scar. Although clinical and operative notes are often considered, many clinicians simply use the surgical scar with a 2- to 3-cm margin as the primary landmark for the electron boost field. Multiple reports (5–11) have indicated that this technique is prone to creating “geographical misses.” However, none of these studies investigated these inaccuracies using CT-derived data gathered after whole-breast irradiation just before the boost. The objectives of this study were (1) to examine the changes in breast and excision cavity volumes after whole-breast irradiation, and (2) to compare the adequacy of using the surgical scar to guide boost field placement using data derived from CT scans performed before and near the completion of wholebreast irradiation.
METHODS AND MATERIALS Study population After approval from the Institutional Review Board of the University of Michigan Medical School, the CT scans from 30 consecutively treated patients with breast cancer were reviewed. Patients had Stages 0 to III breast cancer (12) and underwent breast-conserving surgery and RT. One patient had bilateral breast cancer, and thus a total of 31 breast volumes were included.
Radiation planning Delineation of breast and excision cavity. CT planning of the breast and excision cavity was performed in two sessions. The first CT simulation (CT1) was performed for treatment planning of the whole breast. The second CT simulation (CT2) was performed within 1 week before the start of the boost after delivering a mean dose of 40.6 Gy (95% CI ⫽ 39.2– 41.9 Gy) to the whole breast. Breast volumes were contoured on CT1 based on clinical borders, using CT-compatible catheters placed 1.0 to 1.5 cm beyond palpable breast tissue. Using image fusion tools, these catheters were marked as reference lines and superimposed on CT2. Contouring of excision cavities on both CT1 and CT2 was guided by the presence of surgical clips, hematoma, and/or other surgical changes, using the changes that resulted in the greatest volume. All contouring was performed by a single physician (KO) and approved by attending radiation oncologists (FK and LP). Hypothetical treatment planing. To design the scar-based boost field, 26 hypothetical electron boost plans were generated on both CT1 and CT2 scans by applying a constant margin of 3 cm around all points along the surgical scar. In the remaining five breasts, the excision cavity depth was not amenable to boost planning with electrons. The selection of electron energy was based on the depth of the excision cavity, with the dose prescribed to the 90% isodose surface at the deepest edge of the cavity as defined by CT. Dose–volume histograms for the excision cavities were generated for each case and analyzed for adequacy of coverage. Analysis included minimum and mean doses received by the excision cavity, as well as the percentage of cavity volume included within 50%, 80%, and 90% isodose lines.
● K. S. OH et al.
681
Data analysis Significant differences in the data gathered from CT1 and CT2 were assessed using Student’s paired t-test with a two-tailed distribution. The Pearson correlation coefficient was used to evaluate possible associations between changes in excision cavity volumes and clinical factors, including baseline breast volume, initial excision cavity volume, weight, age, time elapsed between surgery and RT, and use of chemotherapy. For all statistical tests, p values ⱕ5% were considered significant.
RESULTS Patient and tumor characteristics Patient and tumor characteristics are summarized in Table 1. Body weights were highly variable, with a median weight of 72.7 kg and a range of 44.9 to 119.8 kg. The most common histology was invasive ductal carcinoma. The most common tumor locations were the outer quadrants. A total of 14 patients received chemotherapy before RT, all of whom received a doxorubicin-based regimen. Of these, 7 underwent additional chemotherapy with a taxane before RT. Patients who received any chemotherapy were combined into one group for subsequent analyses. Of 30 patients, 24 (80%) received hormonal therapy as part of their care; all patients received hormonal therapy after RT. Thus,
Table 1. Patient and tumor characteristics Age (y) at RT initiation Median Range Weight (kg) Median Range T stage Tis T1 T2 Histology DCIS IDC ILC Mixed invasive Location Inner Outer Central Time (weeks) elapsed between surgery and RT start 0–4 4.1–8 8.1–12 12.1–16 ⬎16 Preradiation chemotherapy Yes No
55.4 34.5–80.5 72.7 44.9–119.8 n (%) 8 (26) 19 (61) 4 (13) 8 (26) 20 (65) 1 (3) 2 (6) 11 (35) 17 (55) 3 (10) 3 (10) 11 (35) 4 (13) 4 (13) 9 (29) 14 (45) 17 (55)
Abbreviations: DCIS ⫽ ductal carcinoma in situ; IDC ⫽ invasive ductal carcinoma; ILC ⫽ invasive lobular carcinoma. Several patients required reexcision of the tumor bed for positive margins. In all cases, our analyses are based on the time elapsed between radiation start and the most recent surgery.
682
I. J. Radiation Oncology
● Biology ● Physics
Volume 66, Number 3, 2006
Table 3. Relationship between body weight and extent of change in biopsy cavity volume
Table 2. Volumetric changes in breast and excision cavity after whole-breast irradiation
Breast 95% CI Excision cavity 95% CI
CT1 (cc)
CT2 (cc)
% change
p value
774 596–951 32.1 25.1–39.2
761 591–930 25.1 18.8–31.5
⫹0.11%
0.22
Body weight
⬍30 cc
⬎30 cc
All
⫺22.50%
⬍0.0001
ⱕ72 kg ⬎72 kg
⫺29.8% ⫺13.3%
⫺42.0% ⫺12.7%
⫺34.1% ⫺12.9% p ⫽ 0.0208
Abbreviations: CT1 ⫽ CT obtained before whole-breast irradiation; CT2 ⫽ CT obtained within 1 week before boost irradiation.
the effect of concurrent hormonal therapy and RT upon breast and excision cavity volume could not be studied presenting this analysis. The median time elapsed between surgery and CT1 was 8.7 weeks (range, 2.9 –37.0 weeks). The median time elapsed between surgery and start of RT was 9.3 weeks (range, 3.1–37.9 weeks). Breast volume after radiation The mean breast volumes before and after whole-breast irradiation were 774 cc and 761 cc, respectively, representing a mean change of 0.11% ( p ⫽ 0.22). Change in breast volume was not significantly different when patients pretreated with chemotherapy (⫹1.0% change) were compared with those who were not (⫺0.60% change, p ⫽ 0.58). These changes are summarized in Table 2 and Fig. 1. Excision cavity volume after radiation The mean excision cavity volumes before and after whole-breast irradiation were 32.1 cc and 25.1 cc, respec-
Initial biopsy cavity volume
tively, representing a mean reduction of 22.5% ( p ⬍ 0.0001). Nearly 94% (29/31) of excision cavities were decreased in size after whole-breast irradiation. The relative reduction in excision cavity volume with RT was inversely proportional to the time elapsed since most recent surgical excision (R ⫽ 0.46, p ⬍ 0.01) and body weight (R ⫽ 0.50, p ⬍ 0.01). This reduction was not significantly correlated with age (R ⫽ ⫺0.06, p ⫽ 0.76), baseline breast volume (R ⫽ 0.29, p ⫽ 0.11), and initial excision cavity volume (R ⫽ 0.06, p ⫽ 0.77). Reduction in excision cavity volume was not significantly different when patients pretreated with chemotherapy (17.8% reduction) were compared with those who were not (26.4% reduction, p ⫽ 0.31). These changes are summarized in Tables 2 and 3 and Figs. 2 to 4. Adequacy of scar-based boost treatment planning When boost plans were planned on CT1 using the surgical scar, the mean percentages of excision cavity covered Change in Biopsy Cavity Volume Following Whole Breast Irradiation
Change in Breast Volume Following Whole Breast Irradiation 80 2000
70
60 1500
Volume (cc)
Volume (cc)
50
1000
40
30
20
500
10
0
CT1
CT2
Fig. 1. Change in breast volume after whole-breast irradiation. Each line represents the change of a single patient’s breast volume before (left) and after (right) whole-breast irradiation. In the vast majority of cases, the breast volume remained essentially unchanged. The average change was ⫹0.11%, p ⫽ 0.22. CT1 ⫽ first computed tomographic simulation; CT2 ⫽ second computed tomographic simulation.
0
CT1
CT2
Fig. 2. Change in biopsy cavity volume after whole-breast irradiation. Each line represents the change of a single patient’s excision cavity volume before (left) and after (right) whole-breast irradiation. Of 31 excision cavities, 29 (93.5%) were reduced in volume. The average change was ⫺22.5%, p ⬍ 0.0001. CT1 ⫽ first computed tomographic simulation; CT2 ⫽ second computed tomographic simulation.
Excision cavity and surgical scar in planning breast boost irradiation
● K. S. OH et al.
683
% Volume Change in Biopsy Cavity
40% 20% 0% 0
50
100
150
200
250
300
-20% -40% -60% -80% -100% Days Elapsed Between Surgery and Whole Breast RT
Fig. 4. Scatter graph demonstrating a trend of excision cavity shrinkage as time elapses from surgery. Each point represents a single excision cavity. Of the 31 patients, 20 required reexcision of the tumor bed for positive margins. In all cases, analysis is based on the time elapsed between the start of radiotherapy (RT) and the most recent surgery. The relative reduction in excision cavity volume demonstrates an inversely proportional trend when compared with time elapsed since surgery (R ⫽ 0.46).
dose, 16 of 26 (62%) of scar-based plans were found to result in a geographical miss of the excision cavity based on CT1 plans. Using CT2 plans, 14 of 26 (53.8%) of scar-based plans had a geographical miss (Fig. 5). The minimum dose (described as percentage of prescription) to the excision cavity was significantly lower if the cavity was located in the inner quadrants (35.5% minimum dose) compared with the outer quadrants (67.8% minimum dose, p ⫽ 0.016) of the breast. Larger excision cavities (⬎20 cc) received significantly lower minimum doses than cavities ⱕ20 cc (35.7% vs. 67.4%, respectively, p ⫽ 0.011) (Table 5). DISCUSSION
Fig. 3. Comparison of an excision cavity before and after wholebreast irradiation (40 Gy). Over the course of radiotherapy (RT), the seroma appears to have been contracted and replaced with soft tissue density. Subsequently, the surgical clip has moved in the medial direction.
within 90%, 80%, and 50% isodose lines were 77.7%, 87.7%, and 95.3%, respectively. When planned on CT2 using the surgical scar, the corresponding percentages of excision cavity covered were 76.7%, 88.3%, and 95.3%, respectively (Table 4). When comparing plans generated from CT1 vs. CT2, there were no statistically significant differences in coverage ( p ⫽ 0.68 for V90, p ⫽ 0.64 for V80, and p ⫽ 0.99 for V50). When a “geographical miss” was defined as any portion of the excision cavity receiving ⬍50% of the prescribed
After breast-conserving surgery, for a boost field to offer maximum tumor control, the treatment field must accurately encompass the postoperative excision cavity. This study of 30 women with 31 Stage 0 to III breast cancers assessed the adequacy of common methods of boost planning to cover the tumor bed by following changes in the breast volume and excision cavity during adjuvant radiation therapy. In this study, the whole-breast volume did not significantly change during the course of whole-breast irradiation. However, the volume of the excision cavities significantly decreased by an average of 22.5%. The extent of this reduction was inversely associated with the time interval between surgery and radiation start, but was not correlated with Table 4. Volumes encompassed within 90%, 80%, and 50% isodose lines on scar-based treatment planning (CT2) Isodose
Mean % of volume covered
95% CI
90% 80% 50%
76.7% 88.3% 95.3%
67.1–86.3% 81.5–95.1% 90.6–99.9%
684
I. J. Radiation Oncology
● Biology ● Physics
Fig. 5. Digitally reconstructed radiograph depicting the scar (red), excision cavity (light blue), and scar-based hypothetical field border (green oval). In this case, the excision cavity extends far beyond an electron boost field based on the surgical scar.
initial breast or excision cavity volume or exposure to chemotherapy. Furthermore, the use of the surgical scar to guide placement of the boost field was shown to be grossly inaccurate when measured on CT scans performed before and near the completion of whole-breast irradiation. The traditional clinical method of boost planning is guided primarily by the location of the surgical scar, physical examination, clinical and operative notes, and patient recollection. Although these are all important considerations in the design of the boost field, this method is limited by the lack of direct visualization of the surgical tumor bed. The aforementioned randomized trials (1, 2) prescribed the boost irradiation to the center of the tumor bed without clearly visualizing the excision cavity. Some centers have advocated the use of ultrasound to confirm the excision cavity within the breast (13–15). Multiple other reports have used plain film radiographs to assess the accuracy of clinical boost field planning, and many have concluded that this method results in an inaccurate definition of the tumor bed (5–11). In a study of 27 patients, Denham et al. (5) compared the adequacy of plans derived clinically (from scar location, patient recollection, and surgical notes) to those defined by surgical clips. In this series, 10/24 (42%) of clinically derived plans resulted in incomplete coverage of the excision cavity as demarcated by clips. In a study of 35 patients, Bedwinek (6) demonstrated that, when boost plans were generated using the surgical scar (with 2-cm to 3-cm margin), the surgical clips were outside the boost field in 19 of 35 (54%) of cases. Harrington et al. (7) included 50 patients in a study demonstrating that clinically derived boost plans were inaccurate in 68% of cases. The largest study of this kind was performed by Machtay et al. (9). This was a retrospective analysis of 316 cases in which hypothetical fields were derived using multiple planning techniques (based on surgical scar alone) and compared against
Volume 66, Number 3, 2006
the actual location of the excision cavity as defined by surgical clips. When wide (3– 4 cm) margins were used, inadequate coverage of clips occurred in 10% to 36% of cases. Regine et al. (11) compared CT planning of the electron boost with clinically derived plans and found that only five of 17 (29%) clinical plans adequately included the tumor bed within the treatment volume, whereas the rest (71%) were geographical misses. In a similar study of 45 patients, Messer et al. (10) compared CT-based plans with clinically derived plans (defined by physical examination and pre- and postsurgical mammography). After review of isodose distributions, 80% of clinically derived plans required major modifications. Our study used a CT scan obtained just before the start of the boost and confirmed that scar-guided field placement still does not adequately determine the excision cavity despite the reduction in the tumor bed during whole-breast irradiation. Using this method, 14 of 26 (53.8%) scar-based plans were found to result in geographical misses of the clinical target volume (CTV) as defined on CT2. Scar-based planning was especially inadequate for excision cavities of large volume (⬎20 cc). One possible explanation is that the clinical boost plans were guided by a constant 3-cm margin around the surgical scar. Therefore, using this technique, the additional volume around the perimeter of the scar was proportionally more generous for smaller-scale excisions, that is, shorter scars. If a different technique had been used, such as that described by Machtay et al. (9) using margins equal to one-half of the scar length, then perhaps small and large cavities would have been covered with equivalent adequacy. When analyzed by location, excision cavities located within the inner quadrants were also found to be at significantly higher risk of being inadequately covered by scar-guided fields. In these inner quadrant cases, the surgical scars often did not extend over the medial portion of the excision cavities. This phenomenon is likely caused by technical constraints of the excision itself as well as the vascular supply to lesions approaching the midline. For these reasons, we believe that using the scar alone to guide treatment planning is especially poor for larger and inner quadrant tumors. In the current era, CT-based planning (4, 10, 11, 16) and placement of radiopaque surgical clips assist in defining the shape of the field and the depth of the tumor bed. Conceptually the CTV is the region encompassing microscopic Table 5. Adequacy of scar-based treatment planning (CT2) Population
Minimum dose received (% of prescribed dose)
All GTV ⱕ 20 cc GTV ⬎ 20 cc Outer quadrants Inner quadrants
51.6% 67.4% 35.7% 67.8% 35.5%
p value
0.011 0.016
Abbreviation: GTV ⫽ gross tumor volume. Minimum dose received is expressed as a percentage of the prescribed dose.
Excision cavity and surgical scar in planning breast boost irradiation
tumor spread and thus the volume at highest risk for local recurrence. In boost planning, the CTV is the postoperative excision cavity plus a margin that is not uniformly defined (17). Most centers use a single pretreatment CT from which plans are generated for both whole-breast and boost irradiation (18). Our initial concern was that significant changes in the breast contour (e.g., swelling) could occur during the course of whole-breast irradiation and inadvertently extend the highest-risk regions outside of the boost planned with the initially defined excision cavity (from CT1). However, contrary to our concerns, this study demonstrated minimal changes in the breast contour with whole-breast irradiation and a significant reduction in excision cavity volume in almost all (29/31) cases during whole-breast irradiation. The reduction in excision cavity volume was greater when radiation therapy was started shortly after surgery, suggesting that this change may be related to cicatricial contraction or replacement by normal tissue in the postoperative period. Weed et al. recently reported a similar relationship between the excision cavity volume and time elapsed from surgery but found that the reduction is minimal after 40 days postoperatively (19). In contrast, our findings demonstrate continued reduction in the excision cavity volume past 150 days postoperatively (Fig. 2). Although it is difficult to make a direct comparision between the two studies, the majority (20/31) of our patients started RT ⬎40 days from surgery, whereas the majority of the patients in the Weed et al. report started RT ⬍40 days from surgery. Therefore, because of the disparity in treatment schedules, it is difficult to compare our findings directly. Our findings support that the boost plan generated based on a CT simulation obtained before the initiation of breast radiation adequately covers the excision cavity, and that a separate boost simulation at the end
● K. S. OH et al.
685
of whole-breast irradiation is not needed for adequate coverage of the above-mentioned excision cavity. Our study has demonstrated the inadequacy of scarbased boost planning despite a reduction in biopsy cavity volume during intact breast irradiation. However, we acknowledge the limitations of this study and the interpretation of the findings. First, the major endpoint of this study was dosimetric coverage of the excision cavity and not local control. Although the studies previously discussed presumed that improved dosimetry leads to improved local control, this has not been a universal finding (20). Second, we presumed that the excision cavity is the best surrogate for the true location of the primary tumor, as it was logistically difficult to correlate the excision cavity with the pathologic location of the tumor itself. To minimize further uncertainty, we chose to plan all hypothetical scar-based fields by using a uniform 3-cm margin around the scar. However, the reliability of the excision cavity as a proxy for tumor location is currently unknown. Clearly, more study is needed to answer these challenging questions. CONCLUSION In summary, our findings support the practice of planning the boost from a CT scan performed before whole-breast irradiation. Scar-based hypothetical plans underdosed the excision cavity in the majority of cases studied, regardless of the time at which the CT was obtained. Inasmuch as the boost plans of the French (1) and EORTC (2) trials were performed without the use of CT-based planning, further improvement in local control may be expected as we continue to improve our ability to accurately localize and treat the excision cavity.
REFERENCES 1. Romestaing P, Lehingue Y, Carrie C, et al. Role of a 10-Gy boost in the conservative treatment of early breast cancer: results of a randomized clinical trial in Lyons, France. J Clin Oncol 1997;15:963–968. 2. Bartelink H, Horiot JC, Poortmans P, et al. Recurrence rates after treatment of breast cancer with standard radiotherapy with or without additional radiation. N Engl J Med 2001;345: 1378 –1387. 3. Pierce LJ, Moughan J, White J. 1998 –1999 Patterns of Care Study process study of national practice patterns using breastconserving surgery and radiotherapy in the management of Stage I–II breast cancer. Int J Radiat Oncol Biol Phys 2005; 62:183–192. 4. Benda RK, Yasuda G, Sethi A, et al. Breast boost: are we missing the target? Cancer 2003;97:905–909. 5. Denham JW, Sillar RW, Clarke D. Boost dosage to the excision site following conservative surgery for breast cancer: it’s easy to miss! Clin Oncol 1991;3:257–261. 6. Bedwinek J. Breast-conserving surgery and irradiation: the importance of demarcating the excision cavity with surgical clips. Int J Radiat Oncol Biol Phys 1993;26:675– 679. 7. Harrington KJ, Harrison M, Bayle P, et al. Surgical clips in planning the electron boost in breast cancer: a qualitative and quantitative evaluation. Int J Radiat Oncol Biol Phys 1996;34:579. 8. Hunter MA, McFall TA, Hehr KA. Breast-conserving surgery
9.
10. 11. 12. 13.
14. 15.
for primary breast cancer: necessity for surgical clips to define the tumor bed for radiation planning. Radiology 1996; 200:281. Machtay M, Lanciano R, Hoffman J, et al. Inaccuracies in using the lumpectomy scar for planning electron boosts in primary breast carcinoma. Int J Radiat Oncol Biol Phys 1994; 30:43– 48. Messer PM, Kirikuta IC, Bratengeier K, et al. CT planning of boost irradiation in radiotherapy of breast cancer after conservative surgery. Radiother Oncol 1997;42:239. Regine WF, Ayyangar KM, Komarnickyk LT, et al. ComputerCT planning of the electron boost in definitive breast irradiation. Int J Radiat Oncol Biol Phys 1991;20:121–125. Greene FL, Page DL, Fleming ID, et al., editors. AJCC cancer staging manual. 6th ed. New York: Springer-Verlag; 2002. Debiose DA, Horwitz EM, Martinez AA, et al. The use of ultrasonography in the localization of the lumpectomy cavity for interstitial brachytherapy of the breast. Int J Radiat Oncol Biol Phys 1997;38:755. Gilligan D, Hendry JA, Yarnold JR. The use of ultrasound to measure breast thickness to select electron energies for breast boost therapy. Radiother Oncol 1994;32:265. Leonard C, Harlow CL, Coffin C, et al. Use of ultrasound to guide radiation boost planning following lumpectomy for carcinoma of the breast. Int J Radiat Oncol Biol Phys 1993;27:1193–1197.
686
I. J. Radiation Oncology
● Biology ● Physics
16. Solin LJ, Danoff BF, Schwartz GF, et al. A practical technique for the localization of the tumour volume in definitive irradiation of the breast. Int J Radiat Oncol Biol Phys 1985;11: 1215–1220. 17. Vicini FA, Kestin LL, Goldstein NS. Defining the clinical target volume for patients with early-stage breast cancer treated with lumpectomy and accelerated partial breast irradiation: a pathologic analysis. Int J Radiat Oncol Biol Phys 2004; 60:722–730. 18. Benda RK, Mendenhall NP, Lind DS, et al. Breast-conserving
Volume 66, Number 3, 2006
therapy (BCT) for early-stage breast cancer. J Surg Oncol 2004;85:14 –27. 19. Weed DW, Yan D, Martinez AA, et al. The validity of surgical clips as a radiographic surrogate for the lumpectomy cavity in image-guided accelerated partial breast irradiation. Int J Radiat Oncol Biol Phys 2004;60:484 – 492. 20. Fein DA, Fowble BL, Hanlon AL, et al. Does the placement of surgical clips within the excision cavity influence local control for patients treated with breast-conserving surgery and irradiation? Int J Radiat Oncol Biol Phys 1996;34:1009 –1017.
doi:10.1016/j.ijrobp.2006.04.042
CLINICAL INVESTIGATION
Breast
PLANNING THE BREAST TUMOR BED BOOST: CHANGES IN THE EXCISION CAVITY VOLUME AND SURGICAL SCAR LOCATION AFTER BREAST-CONSERVING SURGERY AND WHOLE-BREAST IRRADIATION KEVIN S. OH, M.D.,* FENG-MING KONG, M.D., PH.D.,* KENT A. GRIFFITH, M.P.H., M.S.,† BETH YANKE, C.M.D.,* AND LORI J. PIERCE, M.D.* *Department of Radiation Oncology, University of Michigan Medical School, Ann Arbor, MI; and †Biostatistics Unit of the University of Michigan Comprehensive Cancer Center, Ann Arbor, MI Purpose: The aims of this study were to determine the changes in breast and excision cavity volumes after whole-breast irradiation and the adequacy of using the surgical scar to guide boost planning. Methods and Materials: A total of 30 women consecutively treated for 31 breast cancers were included in this study. Simulation CT scans were performed before and after whole-breast irradiation. CT breast volumes were delineated using clinically defined borders. Excision cavity volumes were contoured based on surgical clips, the presence of a hematoma, and/or other surgical changes. Hypothetical electron boost plans were generated using the surgical scar with a 3-cm margin and analyzed for coverage. Results: The mean CT breast volumes were 774 and 761 cc ( p ⴝ 0.22), and the excision cavity volumes were 32.1 and 25.1 cc ( p < 0.0001), before and after 40 Gy (39 – 42 Gy) of whole-breast irradiation, respectively. The volume reduction in the excision cavity was inversely correlated with time elapsed since surgery (R ⴝ 0.46, p < 0.01) and body weight (R ⴝ 0.50, p < 0.01). The scar-guided hypothetical plans failed to cover the excision cavity adequately in 62% and 53.8% of cases using the pretreatment and postradiation CTs, respectively. Per the hypothetical plans, the minimum dose to the excision cavity was significantly lower for tumors located in the inner vs. outer quadrants ( p ⴝ 0.02) and for cavities >20 cc vs. <20 cc ( p ⴝ 0.01). Conclusions: This study demonstrates a significant reduction in the volume of the excision cavity during whole-breast irradiation. Scar-guided boost plans provide inadequate coverage of the excision cavity in the majority of cases. © 2006 Elsevier Inc. Breast cancer, Radiotherapy, Excision cavity, Surgical scar, Boost.
INTRODUCTION
recurrence was significantly lower in women receiving a boost (4.3% vs. 7.3%, p ⬍ 0.001). Although the use of boost irradiation is a recommended practice, no standard technique has been established. In a recent Patterns of Care Study for early-stage breast cancer, the boost volume was determined using CT in 11.7%, a clinical fluoroscopic simulator in 43.9%, and clinically alone in 37.2% (3). With recent technologic advancement, CT-based planning has become more popular in practice over the last several years (4). CT-based planning often uses the excision cavity identified on simulation CT scans acquired before the initiation of whole-breast irradiation, approximately 4 to 6 weeks before initiation of the boost. In the interim between CT simulation and boost treatment, changes in the breast contour such as swelling secondary to edema may occur from whole-breast irradiation. These
The benefit of boost irradiation in women undergoing breast-conserving therapy has been supported by two randomized trials to date. A French trial (1) randomized 1,024 women with invasive cancers ⱕ3 cm after excision and 50 Gy to the whole breast to either a 10-Gy boost or no further treatment to the tumor bed. The reported 5-year actuarial rate of local recurrence was 3.6% (with boost) vs. 4.5% (no further treatment, p ⫽ 0.044). The European Organization for Research and Treatment of Cancer (EORTC) conducted a similar trial (2) randomizing patients with Stage I or II breast cancer after margin-negative breast-conserving surgery and whole-breast irradiation to 50 Gy to receive either a 16-Gy boost (2,661 patients) or no further treatment (2,657 patients). Again, the 5-year actuarial rate of local
Presented in part at the 46th Annual Meeting of the American Society for Therapeutic Radiology and Oncology (ASTRO), Atlanta, GA, October 3–7, 2004. Received Feb 3, 2006, and in revised form April 11, 2006. Accepted for publication April 16, 2006.
Note—An online CME test for this article can be taken at www.astro.org under Education and Meetings. Reprint requests to: Lori J. Pierce, M.D., Department of Radiation Oncology, University of Michigan Medical School, Ann Arbor, MI 48109. Tel: (734) 936-7810; Fax: (734) 763-7370; E-mail: [email protected] 680
Excision cavity and surgical scar in planning breast boost irradiation
changes may affect the localization of both the tumor bed and surgical scar. The extent of these changes and their effect on boost planning are unknown. Clinically based boost planning is two-dimensional, with the field largely determined by the surgical scar. Although clinical and operative notes are often considered, many clinicians simply use the surgical scar with a 2- to 3-cm margin as the primary landmark for the electron boost field. Multiple reports (5–11) have indicated that this technique is prone to creating “geographical misses.” However, none of these studies investigated these inaccuracies using CT-derived data gathered after whole-breast irradiation just before the boost. The objectives of this study were (1) to examine the changes in breast and excision cavity volumes after whole-breast irradiation, and (2) to compare the adequacy of using the surgical scar to guide boost field placement using data derived from CT scans performed before and near the completion of wholebreast irradiation.
METHODS AND MATERIALS Study population After approval from the Institutional Review Board of the University of Michigan Medical School, the CT scans from 30 consecutively treated patients with breast cancer were reviewed. Patients had Stages 0 to III breast cancer (12) and underwent breast-conserving surgery and RT. One patient had bilateral breast cancer, and thus a total of 31 breast volumes were included.
Radiation planning Delineation of breast and excision cavity. CT planning of the breast and excision cavity was performed in two sessions. The first CT simulation (CT1) was performed for treatment planning of the whole breast. The second CT simulation (CT2) was performed within 1 week before the start of the boost after delivering a mean dose of 40.6 Gy (95% CI ⫽ 39.2– 41.9 Gy) to the whole breast. Breast volumes were contoured on CT1 based on clinical borders, using CT-compatible catheters placed 1.0 to 1.5 cm beyond palpable breast tissue. Using image fusion tools, these catheters were marked as reference lines and superimposed on CT2. Contouring of excision cavities on both CT1 and CT2 was guided by the presence of surgical clips, hematoma, and/or other surgical changes, using the changes that resulted in the greatest volume. All contouring was performed by a single physician (KO) and approved by attending radiation oncologists (FK and LP). Hypothetical treatment planing. To design the scar-based boost field, 26 hypothetical electron boost plans were generated on both CT1 and CT2 scans by applying a constant margin of 3 cm around all points along the surgical scar. In the remaining five breasts, the excision cavity depth was not amenable to boost planning with electrons. The selection of electron energy was based on the depth of the excision cavity, with the dose prescribed to the 90% isodose surface at the deepest edge of the cavity as defined by CT. Dose–volume histograms for the excision cavities were generated for each case and analyzed for adequacy of coverage. Analysis included minimum and mean doses received by the excision cavity, as well as the percentage of cavity volume included within 50%, 80%, and 90% isodose lines.
● K. S. OH et al.
681
Data analysis Significant differences in the data gathered from CT1 and CT2 were assessed using Student’s paired t-test with a two-tailed distribution. The Pearson correlation coefficient was used to evaluate possible associations between changes in excision cavity volumes and clinical factors, including baseline breast volume, initial excision cavity volume, weight, age, time elapsed between surgery and RT, and use of chemotherapy. For all statistical tests, p values ⱕ5% were considered significant.
RESULTS Patient and tumor characteristics Patient and tumor characteristics are summarized in Table 1. Body weights were highly variable, with a median weight of 72.7 kg and a range of 44.9 to 119.8 kg. The most common histology was invasive ductal carcinoma. The most common tumor locations were the outer quadrants. A total of 14 patients received chemotherapy before RT, all of whom received a doxorubicin-based regimen. Of these, 7 underwent additional chemotherapy with a taxane before RT. Patients who received any chemotherapy were combined into one group for subsequent analyses. Of 30 patients, 24 (80%) received hormonal therapy as part of their care; all patients received hormonal therapy after RT. Thus,
Table 1. Patient and tumor characteristics Age (y) at RT initiation Median Range Weight (kg) Median Range T stage Tis T1 T2 Histology DCIS IDC ILC Mixed invasive Location Inner Outer Central Time (weeks) elapsed between surgery and RT start 0–4 4.1–8 8.1–12 12.1–16 ⬎16 Preradiation chemotherapy Yes No
55.4 34.5–80.5 72.7 44.9–119.8 n (%) 8 (26) 19 (61) 4 (13) 8 (26) 20 (65) 1 (3) 2 (6) 11 (35) 17 (55) 3 (10) 3 (10) 11 (35) 4 (13) 4 (13) 9 (29) 14 (45) 17 (55)
Abbreviations: DCIS ⫽ ductal carcinoma in situ; IDC ⫽ invasive ductal carcinoma; ILC ⫽ invasive lobular carcinoma. Several patients required reexcision of the tumor bed for positive margins. In all cases, our analyses are based on the time elapsed between radiation start and the most recent surgery.
682
I. J. Radiation Oncology
● Biology ● Physics
Volume 66, Number 3, 2006
Table 3. Relationship between body weight and extent of change in biopsy cavity volume
Table 2. Volumetric changes in breast and excision cavity after whole-breast irradiation
Breast 95% CI Excision cavity 95% CI
CT1 (cc)
CT2 (cc)
% change
p value
774 596–951 32.1 25.1–39.2
761 591–930 25.1 18.8–31.5
⫹0.11%
0.22
Body weight
⬍30 cc
⬎30 cc
All
⫺22.50%
⬍0.0001
ⱕ72 kg ⬎72 kg
⫺29.8% ⫺13.3%
⫺42.0% ⫺12.7%
⫺34.1% ⫺12.9% p ⫽ 0.0208
Abbreviations: CT1 ⫽ CT obtained before whole-breast irradiation; CT2 ⫽ CT obtained within 1 week before boost irradiation.
the effect of concurrent hormonal therapy and RT upon breast and excision cavity volume could not be studied presenting this analysis. The median time elapsed between surgery and CT1 was 8.7 weeks (range, 2.9 –37.0 weeks). The median time elapsed between surgery and start of RT was 9.3 weeks (range, 3.1–37.9 weeks). Breast volume after radiation The mean breast volumes before and after whole-breast irradiation were 774 cc and 761 cc, respectively, representing a mean change of 0.11% ( p ⫽ 0.22). Change in breast volume was not significantly different when patients pretreated with chemotherapy (⫹1.0% change) were compared with those who were not (⫺0.60% change, p ⫽ 0.58). These changes are summarized in Table 2 and Fig. 1. Excision cavity volume after radiation The mean excision cavity volumes before and after whole-breast irradiation were 32.1 cc and 25.1 cc, respec-
Initial biopsy cavity volume
tively, representing a mean reduction of 22.5% ( p ⬍ 0.0001). Nearly 94% (29/31) of excision cavities were decreased in size after whole-breast irradiation. The relative reduction in excision cavity volume with RT was inversely proportional to the time elapsed since most recent surgical excision (R ⫽ 0.46, p ⬍ 0.01) and body weight (R ⫽ 0.50, p ⬍ 0.01). This reduction was not significantly correlated with age (R ⫽ ⫺0.06, p ⫽ 0.76), baseline breast volume (R ⫽ 0.29, p ⫽ 0.11), and initial excision cavity volume (R ⫽ 0.06, p ⫽ 0.77). Reduction in excision cavity volume was not significantly different when patients pretreated with chemotherapy (17.8% reduction) were compared with those who were not (26.4% reduction, p ⫽ 0.31). These changes are summarized in Tables 2 and 3 and Figs. 2 to 4. Adequacy of scar-based boost treatment planning When boost plans were planned on CT1 using the surgical scar, the mean percentages of excision cavity covered Change in Biopsy Cavity Volume Following Whole Breast Irradiation
Change in Breast Volume Following Whole Breast Irradiation 80 2000
70
60 1500
Volume (cc)
Volume (cc)
50
1000
40
30
20
500
10
0
CT1
CT2
Fig. 1. Change in breast volume after whole-breast irradiation. Each line represents the change of a single patient’s breast volume before (left) and after (right) whole-breast irradiation. In the vast majority of cases, the breast volume remained essentially unchanged. The average change was ⫹0.11%, p ⫽ 0.22. CT1 ⫽ first computed tomographic simulation; CT2 ⫽ second computed tomographic simulation.
0
CT1
CT2
Fig. 2. Change in biopsy cavity volume after whole-breast irradiation. Each line represents the change of a single patient’s excision cavity volume before (left) and after (right) whole-breast irradiation. Of 31 excision cavities, 29 (93.5%) were reduced in volume. The average change was ⫺22.5%, p ⬍ 0.0001. CT1 ⫽ first computed tomographic simulation; CT2 ⫽ second computed tomographic simulation.
Excision cavity and surgical scar in planning breast boost irradiation
● K. S. OH et al.
683
% Volume Change in Biopsy Cavity
40% 20% 0% 0
50
100
150
200
250
300
-20% -40% -60% -80% -100% Days Elapsed Between Surgery and Whole Breast RT
Fig. 4. Scatter graph demonstrating a trend of excision cavity shrinkage as time elapses from surgery. Each point represents a single excision cavity. Of the 31 patients, 20 required reexcision of the tumor bed for positive margins. In all cases, analysis is based on the time elapsed between the start of radiotherapy (RT) and the most recent surgery. The relative reduction in excision cavity volume demonstrates an inversely proportional trend when compared with time elapsed since surgery (R ⫽ 0.46).
dose, 16 of 26 (62%) of scar-based plans were found to result in a geographical miss of the excision cavity based on CT1 plans. Using CT2 plans, 14 of 26 (53.8%) of scar-based plans had a geographical miss (Fig. 5). The minimum dose (described as percentage of prescription) to the excision cavity was significantly lower if the cavity was located in the inner quadrants (35.5% minimum dose) compared with the outer quadrants (67.8% minimum dose, p ⫽ 0.016) of the breast. Larger excision cavities (⬎20 cc) received significantly lower minimum doses than cavities ⱕ20 cc (35.7% vs. 67.4%, respectively, p ⫽ 0.011) (Table 5). DISCUSSION
Fig. 3. Comparison of an excision cavity before and after wholebreast irradiation (40 Gy). Over the course of radiotherapy (RT), the seroma appears to have been contracted and replaced with soft tissue density. Subsequently, the surgical clip has moved in the medial direction.
within 90%, 80%, and 50% isodose lines were 77.7%, 87.7%, and 95.3%, respectively. When planned on CT2 using the surgical scar, the corresponding percentages of excision cavity covered were 76.7%, 88.3%, and 95.3%, respectively (Table 4). When comparing plans generated from CT1 vs. CT2, there were no statistically significant differences in coverage ( p ⫽ 0.68 for V90, p ⫽ 0.64 for V80, and p ⫽ 0.99 for V50). When a “geographical miss” was defined as any portion of the excision cavity receiving ⬍50% of the prescribed
After breast-conserving surgery, for a boost field to offer maximum tumor control, the treatment field must accurately encompass the postoperative excision cavity. This study of 30 women with 31 Stage 0 to III breast cancers assessed the adequacy of common methods of boost planning to cover the tumor bed by following changes in the breast volume and excision cavity during adjuvant radiation therapy. In this study, the whole-breast volume did not significantly change during the course of whole-breast irradiation. However, the volume of the excision cavities significantly decreased by an average of 22.5%. The extent of this reduction was inversely associated with the time interval between surgery and radiation start, but was not correlated with Table 4. Volumes encompassed within 90%, 80%, and 50% isodose lines on scar-based treatment planning (CT2) Isodose
Mean % of volume covered
95% CI
90% 80% 50%
76.7% 88.3% 95.3%
67.1–86.3% 81.5–95.1% 90.6–99.9%
684
I. J. Radiation Oncology
● Biology ● Physics
Fig. 5. Digitally reconstructed radiograph depicting the scar (red), excision cavity (light blue), and scar-based hypothetical field border (green oval). In this case, the excision cavity extends far beyond an electron boost field based on the surgical scar.
initial breast or excision cavity volume or exposure to chemotherapy. Furthermore, the use of the surgical scar to guide placement of the boost field was shown to be grossly inaccurate when measured on CT scans performed before and near the completion of whole-breast irradiation. The traditional clinical method of boost planning is guided primarily by the location of the surgical scar, physical examination, clinical and operative notes, and patient recollection. Although these are all important considerations in the design of the boost field, this method is limited by the lack of direct visualization of the surgical tumor bed. The aforementioned randomized trials (1, 2) prescribed the boost irradiation to the center of the tumor bed without clearly visualizing the excision cavity. Some centers have advocated the use of ultrasound to confirm the excision cavity within the breast (13–15). Multiple other reports have used plain film radiographs to assess the accuracy of clinical boost field planning, and many have concluded that this method results in an inaccurate definition of the tumor bed (5–11). In a study of 27 patients, Denham et al. (5) compared the adequacy of plans derived clinically (from scar location, patient recollection, and surgical notes) to those defined by surgical clips. In this series, 10/24 (42%) of clinically derived plans resulted in incomplete coverage of the excision cavity as demarcated by clips. In a study of 35 patients, Bedwinek (6) demonstrated that, when boost plans were generated using the surgical scar (with 2-cm to 3-cm margin), the surgical clips were outside the boost field in 19 of 35 (54%) of cases. Harrington et al. (7) included 50 patients in a study demonstrating that clinically derived boost plans were inaccurate in 68% of cases. The largest study of this kind was performed by Machtay et al. (9). This was a retrospective analysis of 316 cases in which hypothetical fields were derived using multiple planning techniques (based on surgical scar alone) and compared against
Volume 66, Number 3, 2006
the actual location of the excision cavity as defined by surgical clips. When wide (3– 4 cm) margins were used, inadequate coverage of clips occurred in 10% to 36% of cases. Regine et al. (11) compared CT planning of the electron boost with clinically derived plans and found that only five of 17 (29%) clinical plans adequately included the tumor bed within the treatment volume, whereas the rest (71%) were geographical misses. In a similar study of 45 patients, Messer et al. (10) compared CT-based plans with clinically derived plans (defined by physical examination and pre- and postsurgical mammography). After review of isodose distributions, 80% of clinically derived plans required major modifications. Our study used a CT scan obtained just before the start of the boost and confirmed that scar-guided field placement still does not adequately determine the excision cavity despite the reduction in the tumor bed during whole-breast irradiation. Using this method, 14 of 26 (53.8%) scar-based plans were found to result in geographical misses of the clinical target volume (CTV) as defined on CT2. Scar-based planning was especially inadequate for excision cavities of large volume (⬎20 cc). One possible explanation is that the clinical boost plans were guided by a constant 3-cm margin around the surgical scar. Therefore, using this technique, the additional volume around the perimeter of the scar was proportionally more generous for smaller-scale excisions, that is, shorter scars. If a different technique had been used, such as that described by Machtay et al. (9) using margins equal to one-half of the scar length, then perhaps small and large cavities would have been covered with equivalent adequacy. When analyzed by location, excision cavities located within the inner quadrants were also found to be at significantly higher risk of being inadequately covered by scar-guided fields. In these inner quadrant cases, the surgical scars often did not extend over the medial portion of the excision cavities. This phenomenon is likely caused by technical constraints of the excision itself as well as the vascular supply to lesions approaching the midline. For these reasons, we believe that using the scar alone to guide treatment planning is especially poor for larger and inner quadrant tumors. In the current era, CT-based planning (4, 10, 11, 16) and placement of radiopaque surgical clips assist in defining the shape of the field and the depth of the tumor bed. Conceptually the CTV is the region encompassing microscopic Table 5. Adequacy of scar-based treatment planning (CT2) Population
Minimum dose received (% of prescribed dose)
All GTV ⱕ 20 cc GTV ⬎ 20 cc Outer quadrants Inner quadrants
51.6% 67.4% 35.7% 67.8% 35.5%
p value
0.011 0.016
Abbreviation: GTV ⫽ gross tumor volume. Minimum dose received is expressed as a percentage of the prescribed dose.
Excision cavity and surgical scar in planning breast boost irradiation
tumor spread and thus the volume at highest risk for local recurrence. In boost planning, the CTV is the postoperative excision cavity plus a margin that is not uniformly defined (17). Most centers use a single pretreatment CT from which plans are generated for both whole-breast and boost irradiation (18). Our initial concern was that significant changes in the breast contour (e.g., swelling) could occur during the course of whole-breast irradiation and inadvertently extend the highest-risk regions outside of the boost planned with the initially defined excision cavity (from CT1). However, contrary to our concerns, this study demonstrated minimal changes in the breast contour with whole-breast irradiation and a significant reduction in excision cavity volume in almost all (29/31) cases during whole-breast irradiation. The reduction in excision cavity volume was greater when radiation therapy was started shortly after surgery, suggesting that this change may be related to cicatricial contraction or replacement by normal tissue in the postoperative period. Weed et al. recently reported a similar relationship between the excision cavity volume and time elapsed from surgery but found that the reduction is minimal after 40 days postoperatively (19). In contrast, our findings demonstrate continued reduction in the excision cavity volume past 150 days postoperatively (Fig. 2). Although it is difficult to make a direct comparision between the two studies, the majority (20/31) of our patients started RT ⬎40 days from surgery, whereas the majority of the patients in the Weed et al. report started RT ⬍40 days from surgery. Therefore, because of the disparity in treatment schedules, it is difficult to compare our findings directly. Our findings support that the boost plan generated based on a CT simulation obtained before the initiation of breast radiation adequately covers the excision cavity, and that a separate boost simulation at the end
● K. S. OH et al.
685
of whole-breast irradiation is not needed for adequate coverage of the above-mentioned excision cavity. Our study has demonstrated the inadequacy of scarbased boost planning despite a reduction in biopsy cavity volume during intact breast irradiation. However, we acknowledge the limitations of this study and the interpretation of the findings. First, the major endpoint of this study was dosimetric coverage of the excision cavity and not local control. Although the studies previously discussed presumed that improved dosimetry leads to improved local control, this has not been a universal finding (20). Second, we presumed that the excision cavity is the best surrogate for the true location of the primary tumor, as it was logistically difficult to correlate the excision cavity with the pathologic location of the tumor itself. To minimize further uncertainty, we chose to plan all hypothetical scar-based fields by using a uniform 3-cm margin around the scar. However, the reliability of the excision cavity as a proxy for tumor location is currently unknown. Clearly, more study is needed to answer these challenging questions. CONCLUSION In summary, our findings support the practice of planning the boost from a CT scan performed before whole-breast irradiation. Scar-based hypothetical plans underdosed the excision cavity in the majority of cases studied, regardless of the time at which the CT was obtained. Inasmuch as the boost plans of the French (1) and EORTC (2) trials were performed without the use of CT-based planning, further improvement in local control may be expected as we continue to improve our ability to accurately localize and treat the excision cavity.
REFERENCES 1. Romestaing P, Lehingue Y, Carrie C, et al. Role of a 10-Gy boost in the conservative treatment of early breast cancer: results of a randomized clinical trial in Lyons, France. J Clin Oncol 1997;15:963–968. 2. Bartelink H, Horiot JC, Poortmans P, et al. Recurrence rates after treatment of breast cancer with standard radiotherapy with or without additional radiation. N Engl J Med 2001;345: 1378 –1387. 3. Pierce LJ, Moughan J, White J. 1998 –1999 Patterns of Care Study process study of national practice patterns using breastconserving surgery and radiotherapy in the management of Stage I–II breast cancer. Int J Radiat Oncol Biol Phys 2005; 62:183–192. 4. Benda RK, Yasuda G, Sethi A, et al. Breast boost: are we missing the target? Cancer 2003;97:905–909. 5. Denham JW, Sillar RW, Clarke D. Boost dosage to the excision site following conservative surgery for breast cancer: it’s easy to miss! Clin Oncol 1991;3:257–261. 6. Bedwinek J. Breast-conserving surgery and irradiation: the importance of demarcating the excision cavity with surgical clips. Int J Radiat Oncol Biol Phys 1993;26:675– 679. 7. Harrington KJ, Harrison M, Bayle P, et al. Surgical clips in planning the electron boost in breast cancer: a qualitative and quantitative evaluation. Int J Radiat Oncol Biol Phys 1996;34:579. 8. Hunter MA, McFall TA, Hehr KA. Breast-conserving surgery
9.
10. 11. 12. 13.
14. 15.
for primary breast cancer: necessity for surgical clips to define the tumor bed for radiation planning. Radiology 1996; 200:281. Machtay M, Lanciano R, Hoffman J, et al. Inaccuracies in using the lumpectomy scar for planning electron boosts in primary breast carcinoma. Int J Radiat Oncol Biol Phys 1994; 30:43– 48. Messer PM, Kirikuta IC, Bratengeier K, et al. CT planning of boost irradiation in radiotherapy of breast cancer after conservative surgery. Radiother Oncol 1997;42:239. Regine WF, Ayyangar KM, Komarnickyk LT, et al. ComputerCT planning of the electron boost in definitive breast irradiation. Int J Radiat Oncol Biol Phys 1991;20:121–125. Greene FL, Page DL, Fleming ID, et al., editors. AJCC cancer staging manual. 6th ed. New York: Springer-Verlag; 2002. Debiose DA, Horwitz EM, Martinez AA, et al. The use of ultrasonography in the localization of the lumpectomy cavity for interstitial brachytherapy of the breast. Int J Radiat Oncol Biol Phys 1997;38:755. Gilligan D, Hendry JA, Yarnold JR. The use of ultrasound to measure breast thickness to select electron energies for breast boost therapy. Radiother Oncol 1994;32:265. Leonard C, Harlow CL, Coffin C, et al. Use of ultrasound to guide radiation boost planning following lumpectomy for carcinoma of the breast. Int J Radiat Oncol Biol Phys 1993;27:1193–1197.
686
I. J. Radiation Oncology
● Biology ● Physics
16. Solin LJ, Danoff BF, Schwartz GF, et al. A practical technique for the localization of the tumour volume in definitive irradiation of the breast. Int J Radiat Oncol Biol Phys 1985;11: 1215–1220. 17. Vicini FA, Kestin LL, Goldstein NS. Defining the clinical target volume for patients with early-stage breast cancer treated with lumpectomy and accelerated partial breast irradiation: a pathologic analysis. Int J Radiat Oncol Biol Phys 2004; 60:722–730. 18. Benda RK, Mendenhall NP, Lind DS, et al. Breast-conserving
Volume 66, Number 3, 2006
therapy (BCT) for early-stage breast cancer. J Surg Oncol 2004;85:14 –27. 19. Weed DW, Yan D, Martinez AA, et al. The validity of surgical clips as a radiographic surrogate for the lumpectomy cavity in image-guided accelerated partial breast irradiation. Int J Radiat Oncol Biol Phys 2004;60:484 – 492. 20. Fein DA, Fowble BL, Hanlon AL, et al. Does the placement of surgical clips within the excision cavity influence local control for patients treated with breast-conserving surgery and irradiation? Int J Radiat Oncol Biol Phys 1996;34:1009 –1017.