Secondary Logo

Journal Logo

Articles

Impact of Boost Technique on Outcome in Early-Stage Breast Cancer Patients Treated With Breast-Conserving Therapy

Frazier, Robert C., M.D.; Kestin, Larry L., M.D.; Kini, Vijay, M.D.; Martinez, Alvaro A., M.D., F.A.C.R.; Chen, Peter Y., M.D.; Baglan, Kathy L., M.D.; Vicini, Frank A., M.D.

Author Information
American Journal of Clinical Oncology: February 2001 - Volume 24 - Issue 1 - p 26-32
  • Free

Abstract

Although breast-conserving therapy (BCT) has become an accepted option in the overall management of patients with early-stage breast cancer, the optimal radiotherapy technique remains controversial. Traditional treatment after lumpectomy generally consists of whole breast irradiation at a dose of up to 45 to 50 Gy followed by a supplemental boost to the tumor bed of up to ≥60 Gy. Despite the widespread application of this treatment approach, the necessity for a boost to the tumor bed is uncertain. For example, results from the National Surgical Adjuvent Breast Project (NSABP) B-06 trial comparing breast conservation to mastectomy showed excellent rates of local control using only whole breast radiation therapy at a dose of up to 50 Gy. 1 Conversely, Romenstaing et al. 2 recently published a randomized trial with 1,024 patients that documented an improvement in local control with the use of a boost to the tumor bed. Although the difference in local control with a boost in this study was only 1% to 2%, the results were statistically significant suggesting that some subsets of patients may benefit from a higher dose of radiation. Considering that patients who develop a local recurrence after BCT may be at a higher risk for distant failure, 3,4 it is likely that a boost to the tumor bed will remain a standard treatment approach for most patients. Since it is uncertain if the type of boost (e.g., electrons, interstitial implants, or photons) may affect outcome, we retrospectively analyzed our institution’s experience treating patients with early-stage breast cancer with BCT. Differences in local recurrence were analyzed based on the type of boost used.

MATERIALS AND METHODS

Between January 1, 1980 and December 31, 1989, a total of 613 patients with stage I and II breast carcinoma were treated with BCT at William Beaumont Hospital, Royal Oak, MI. Eleven patients presented with bilateral breast carcinoma (two synchronous and nine metachronous). Sixty-four patients were excluded from this analysis having received a tumor bed dose less than 60 Gy. Therefore, the study population consisted of 563 carcinomas in 552 women. The 1992 American Joint Commission on Cancer staging system was used. The median patient age at diagnosis was 55 years (range, 25–85 years). Two hundred six patients (36%) were 50 years or younger, and 357 (63%) were older than 50 years.

Surgery consisted of at least an excisional biopsy in all patients, defined as a gross total resection of the primary tumor. Three hundred seventy-two patients (66%) underwent a reexcision of the tumor bed at the discretion of the radiation oncologist or surgeon. Indications for reexcision included positive margins (tumor at the edge of the pathologic specimen), close margins (tumor ≤2 mm from the edge of the pathologic specimen), or uncertain margins. Inking of pathologic specimens was not routinely performed prior to 1987. Residual carcinoma was found in 184 reexcision specimens (48%). Five hundred thirty-five patients (95%) underwent an axillary lymph node dissection generally limited to levels I and II. A mean of 15 lymph nodes were removed. (range, 0–39). All pathology information was obtained from patient charts.

Our radiation therapy treatment technique has been described previously. 5 Radiotherapy was initiated a median of 4 weeks (range, 1–41 weeks) after the final surgical procedure. All patients received whole breast irradiation to a total dose of 45 Gy to 50 Gy over 5 weeks using 4 MV, 6 MV, or 10 MV photons via opposed tangents with a posterior beam splitter or unopposed tangents (wedges were routinely used). This was followed by a boost to the tumor bed in all patients to a dose of at least 60 Gy using electrons (232 patients), photons (15 patients) or an interstitial implant (316 patients) with either 192Ir (194 patients) or 125I (122 patients). The breast alone was treated in 461 patients. The remaining 91 patients were treated to the breast, supraclavicular fossa, and axilla (median axillary and supraclavicular dose, 49 Gy).

Adjuvant chemotherapy was administered to 106 patients (19%) in various sequencing with radiotherapy. Eighty-five of the 134 women with positive lymph nodes received adjuvant chemotherapy (63%). Twenty-one of 401 patients (5%) with negative axillary lymph nodes received chemotherapy. The chemotherapy regimens varied during this treatment time. However, most patients were treated with cyclophosphamide, methotrexate, and 5-fluorouracil (5-FU) (CMF) or with cyclophosphamide and doxorubicin with or without 5-FU (CA/CAF) (84 patients). Tamoxifen was administered to 104 patients (20%).

Implant Technique

Our implant technique has been reported previously. 6 To summarize, the target volume was outlined on the skin of the breast. Stainless steel 17-gauge trocar needles were then inserted in the breast. After needle placement was complete, modified afterloading plastic tubes were pulled through the breast as the trocar needles were removed. Barium buttons were then positioned on the afterloading tubes to secure the implant.

Dosimetry Using 125I Implants

Postoperatively, patients were transported to the radiation oncology department for localization films. A film normal to the implant planes was taken to define the active length. An orthogonal pair (generally an AP and lateral) was used for computerized reconstruction. Isodose distributions were calculated using the Nucletron Planning System. The standard policy after August 1986, was to deliver a minimal tumor dose of 1500 cGy at 62.5 cGy/hr over 24 hr. Large volume implants covering an entire quadrant and the nipple-areolar complex were performed in most patients, particularly those with an extensive intraductal component (EIC).

Since 125I ribbons are prepared by loading loose seeds (separated by spacers) in hollow ribbons, seed spacing was adjusted to deliver the desired dose rate to the target isodose line. In addition, individualized ribbon construction allowed the interseed separation to be adjusted to optimize the dose distribution within a given region of an implant in cases with deviated catheters. Most commonly, seed spacing was adjusted in a single catheter to insure that the 100% isodose line encompassed the entire implant volume. Dose optimization was performed in 40% of patients. The ribbons were afterloaded in the patient’s room and were secured with Henschke buttons. Thin flexible lead rubber was taped over the implant for shielding. Generally, shielding requirements consisted of 1 to 1.5 mm of lead rubber (0.08–0.125 mm lead). Additional layers of thin lead rubber were applied if necessary to reduce the exposure rates by 99%.

Dosimetry Using 192Ir Implants

With 192Ir implants, the volume of the implant was preplanned and ribbons of specified active length were ordered prior to the implant procedure. After implant construction, the ribbons were cut as needed at 1-cm intervals to correspond to the implant geometry. Since seed spacing is fixed with 192Ir ribbons, the dose distribution could not be optimized as with the 125I implants. The ribbons were afterloaded in the patients’ rooms and were secured with Henschke buttons. Shielding consisted of 5.0 cm of lead mounted on a portable rolling bedside shield.

After 24 hours the implants and ribbons were removed and a survey of the patient and room completed. Implant removal was well tolerated and rarely required postimplant analgesia. Overall, the hospital stay was usually less than 36 hours.

Patients were followed up at 3 to 6 month intervals by their radiation oncologist or surgeon. Baseline mammography was performed 6 to 12 months after completion of radiation therapy and repeated yearly thereafter. The median follow-up for all patients was 116 months (range, 2–196 months). The median follow-up for the 421 surviving patients was 122 months (range, 2–196 months). Four hundred eleven surviving patients (97%) have been followed for 5 years, 239 (57%) for 10 years, and 51 (10%) for 13 years. Thirty-two patients (6%) were lost to follow-up a median of 73 months (range, 2–178 months) after diagnosis and were censored at their last follow-up date. Follow-up was complete through June 31,1997.

A local recurrence was defined as the detection of cancer in the parenchyma and/or skin of the treated breast occurring before or simultaneously with the discovery of a regional failure or distant metastasis. Recurrences were defined according to the criteria established at the Joint Center for Radiation Therapy, Harvard Medical School. 7 Recurrences were classified by their location in the breast relative to the primary tumor and boost volume. A true recurrence (TR) occurred within the area of the boost. A marginal miss (MM) occurred adjacent to the boost. An elsewhere failure (E) occurred several centimeters from the primary site and was generally considered a new primary breast cancer. A skin or unclassifiable recurrence was defined as “other.” All recurrences were confirmed histologically. Regional recurrences were defined as the appearance of cancer in the regional lymph nodes before or simultaneously with metastasis. Overall survival (OS) reflects all deaths, and disease-free survival (DFS) was based on deaths attributed to breast carcinoma.

Actuarial results for local control, regional recurrence, distant metastases, and survival were estimated by the Kaplan-Meier method. 8 The association of clinical, pathologic, and treatment-related variables with any given event was analyzed using Fisher’s Exact Test (two-tailed) for categorical variables and logistic regression for continuous variables. Multivariate analysis of the covariates was performed with the Cox proportional hazards model. 9 A p-value of ≤0.05 was considered significant. The 563 cases of breast carcinoma were used for local control and regional analysis. The rates of distant metastasis, DFS, and OS were analyzed for all 549 patients. The bilateral breast carcinomas were analyzed as follows: in synchronous patients one breast was randomized in the analysis, and in metachronous patients only the first breast carcinoma was analyzed. All time intervals were calculated from the date of completion of radiation treatment. Statistical analysis was performed using the SAS software version 6.12 (SAS Institute, Cary, NC, U.S.A.).

RESULTS

Clinical, pathologic, and treatment-related characteristics of the entire patient population are listed in Table 1. There have been 41 recurrences in the 563 treated breasts, for 5-, 10-, and 13-year actuarial local recurrence rates of 2.8%, 7.5%, and 11.2%, respectively. The median time to local recurrence was 73 months (range, 12–147 months). There were 15 recurrences (37%) within 5 years of follow-up and 26 (63%) occurring thereafter. Thirty-seven local recurrences (78%) were isolated. Two presented simultaneously with distant metastases, and two presented simultaneously with regional failure. The location of the local recurrence was TR/MM in 31 cases (76%) and E in 10 cases (24%). Thirty-three recurrences (78%) were invasive and 8 (20%) were noninvasive. Of the 41 local recurrences, 36 were salvaged with mastectomy. The median follow-up after a local recurrence was 51 months (range, 1–104 months). Of the 41 local recurrences, there have been 12 deaths, for a 5-year actuarial overall survival rate of 72%. The 5-year DFS after a local recurrence was 65%.

Table 1
Table 1:
Clinical, pathologic, and treatment-related characteristics of all 563 cases

Table 2 lists multiple clinical, pathologic, and treatment-related factors analyzed on univariate analysis for an association with local recurrence. Complete data for the factors analyzed were unavailable in some cases. The analysis was therefore restricted to those for whom complete data were available. Of these, only patient age ≤35 years and the presence of unknown/positive margins were associated with an increased risk of local recurrence and TR/MM failure.

Table 2
Table 2:
Univariate analysis (p-values) of selected factors potentially associated with local recurrence

The incidence of tumor recurrence in the breast was compared between patients boosted with electrons, photons, or an interstitial implant (Table 3). No significant differences in the 13-year actuarial rates of local recurrence were detected between patients boosted with either 125I, 192Ir, electrons, or photons (p = 0.08). The location of the local recurrences (e.g., TR/MM vs. E) was also compared between each boost technique. Again, no statistically significant differences in the location of failures in the breast were noted between patients boosted with electrons, photons, or an interstitial implant.

Table 3
Table 3:
Thirteen-year actuarial analysis for each boost technique

Table 4 lists clinical and treatment-related characteristics of the study population in relation to the type of boost technique used. To determine if differences in patient or treatment characteristics could account for differences in treatment outcome, the frequency of various risk factors were analyzed according to boost type. Patients boosted with photons more frequently had positive/uncertain margins (46%) (p < 0.01), higher T-stage (53%) (p = 0.01), and higher pathologic N stage (54%) (p = 0.02).

Table 4
Table 4:
Clinical and treatment-related characteristics versus boost technique

On multivariate analysis, younger patient age and positive/uncertain margin status remained independently associated with ipsilateral breast failure (Table 5). The type of boost technique did not significantly affect local failure (p = 0.72).

Table 5
Table 5:
Cox Proportional Hazards model

For the entire group, 22 patients experienced a regional failure, for an actuarial rate of 3.7%, 3.9%, and 4.5% at 5-, 10- and 13 years (Table 6). Eighteen of the 22 regional failures were isolated, and 4 occurred concomitantly with distant metastases. Eighty-two of the 549 patients developed distant metastases for an overall distant failure rate of 11.2%, 15.6%, and 16.3% at 5-, 10- and 13 years, respectively.

Table 6
Table 6:
Actuarial results for all cases at 5, 10, and 13 years

Of the 421 surviving patients, 9 are alive with evidence of disease. For the entire group, the DFS rates at 5-, 10-, and 13 years were 82.4%, 70.2%, and 58.6%, respectively, and the OS rates at 5-, 10-, and 13 years were 87.7%, 76.4%, and 67.0%, respectively. For stage I patients, the OS rates at 5-, 10-, and 13-years were 92.8%, 83.6%, and 72.9%, respectively. For stage II patients, the OS rates at 5-, 10-, and 13 years were 80.7%, 66.2%, 59.9%, respectively.

A total of 488 patients had sufficient information available to evaluate cosmetic outcome. An excellent/good result was noted in 440 patients (89%), fair in 31 (8%), and poor in 17 (3%). The impact of boost technique on cosmetic outcome was evaluated (Table 7). With at least 106 months of median follow-up in all subsets of patients studied, no significant differences in the percentage of patients obtaining good/excellent cosmetic results were noted between patients boosted with either electrons (89%), photons (89%), 125I (94%), or 192Ir (89%) (p = 0.45).

Table 7
Table 7:
Cosmetic outcome versus boost technique

Arm edema was the most common complication after BCT, developing in a total of 41 patients (8%). This was mild in 30 patients (6%) and moderate or severe in 11 patients (2%). All patients who suffered arm edema had an axillary lymph node dissection. Thirty-two patients received radiation to the breast alone and the remaining nine had regional radiation as well. A rib fracture developed after treatment in 6 patients (1%). No significant differences in the incidence of arm edema or rib fracture were seen based on type of boost. No statistically significant differences in the incidence of fat necrosis, or breast fibrosis were seen between patients boosted with electrons, photons, or implants. No patients developed symptomatic pneumonitis or brachial plexopathy.

DISCUSSION

In the current study, we retrospectively analyzed our institution’s long-term experience treating a large group of early-stage breast cancer patients with BCT to determine if the type of boost affected outcome. With a median follow-up of 116 months, no statistically significant differences in the 5-, 10-, and 13-year actuarial rates of local recurrence were noted based on boost technique. On multivariate analysis (including boost technique, age, margin status, T stage, and nodal status) only age and margin status were associated with local recurrence. In addition, both the cosmetic results and rates of complications were similar regardless of the boost method employed. These results suggest that patients judged to be candidates for tumor bed boosts can be effectively managed with either electron beam irradiation or an interstitial implant.

The necessity to deliver a higher dose of radiation to the lumpectomy cavity after whole breast irradiation has remained the subject of speculation for many years. Arguments for and against a boost stem from the fact that several large trials have produced conflicting results. Proponents against a boost cite older data from the NSABP B-06 trial where excellent rates of local control were achieved with only whole breast irradiation to 50 Gy. 1 Advocates for a boost point out that the only randomized trial addressing this issue demonstrated a small but statistically significant improvement in local control. 2 Unfortunately, it is impossible to objectively compare these data since all risk factors for local recurrence may not be balanced equally between trials. Clearly, the effect of patient age, margin status, the use of chemotherapy, and the extent of surgery must be controlled when analyzing the impact of dose on outcome. Until such data are available, the true efficacy of a boost will remain undefined.

Assuming a boost to the tumor bed is required, it is also uncertain at the present time if the type of boost can affect results. Our previous work with fewer patients and shorter follow-up showed that either electrons, photons, or an interstitial implant could produce an acceptable outcome. 10 Our current study, with greater numbers of patients and longer follow-up confirms our previous findings establishing the long-term equivalence of the different treatment approaches. Other studies (Table 8) have also addressed this important issue. 2,10–15 Perez et al. 13 showed no difference in outcome or cosmesis for 701 patients treated with either an electron beam boost or interstitial boost. Conversely, Fourquet et al. 11 published a trial comparing 192Ir implant boost to 60Co external beam irradiation boost in 255 patients. The 192Ir group had a 24% local recurrence rate versus a 39% local recurrence rate in the 60Co group. Wazer et al. 16 reported a prospective trial of 509 patients treated with an escalating tumor bed boost dose based on final surgical margin status. The boost was delivered with electrons, photons, or 192Ir implants. With a median follow-up of 72 months, patients with positive surgical margins had a higher rate of local failure compared with other patients even though this group received the highest doses of radiation (70 Gy). Again, it is impossible to directly compare these data since multiple factors can affect recurrence rates after BCT. As a result, no definite conclusions can be reached supporting the superiority of any boost approach.

Table 8
Table 8:
Outcome with electron, photon, or interstitial boosts in breast-conserving therapy

Regardless of the type of boost chosen, it is critical that accurate tumor bed localization be obtained prior to irradiation delivery. Multiple studies clearly show that clinical estimates of the tumor bed are grossly inaccurate leading to potentially inadequate and incomplete coverage. 17–22 For example, Machtay et al. 17 reported on 316 cases of early-stage breast carcinoma treated with whole breast irradiation followed by an electron beam boost. Inadequate coverage of the lumpectomy cavity (defined by surgical clips) was noted when using the scar to define the lumpectomy cavity. Bedwinek 18 reviewed 35 patients with surgical clips placed at the time of surgery. The author reported inadequate coverage of the surgical clips in 54% with the use of a clinical setup (palpation of postoperative induration to define the boost field margins). These data clearly indicate that accurate localization of the tumor bed is not intuitive and that radiographic verification (whether with computed tomography scans, ultrasound, or surgical clips) is critical.

Although some radiobiologic data suggest that certain types of brachytherapy boosts (e.g., low dose rate) may provide a more efficacious outcome, the available clinical information is simply too weak to generate firm conclusions. Our current data do not provide any additional insight in this debate. Although patients treated with implants generally had worse prognostic factors, patient numbers were still too small to establish clear differences in outcome based on the type of boost. Krishnan et al. 23 reported that patients with an EIC achieved excellent rates of local control with an interstitial boost. The authors felt that these results were due to the superiority of interstitial brachytherapy in this setting. Again, these retrospective reports are clouded by the lack of a control group of patients with similar pretreatment prognostic factors. As a result, the advantages of brachytherapy in this setting remain uncertain.

Since recent data 3,4 suggest an increase in distant failure in patients who develop a local recurrence after BCT, the issues of tumor bed dose and outcome need to be further explored. In addition, since margins of excision appear to also affect long-term local control, 24,25 defining the interrelationships between margins of excision, total tumor bed dose, and the extent of resection is critical. Until such data are available, it seems logical to consider adding a boost to the tumor bed, particularly in patients with adverse findings.

CONCLUSIONS

Patients with stage I and II breast cancer undergoing BCT and judged to be candidates for tumor bed boosts can be effectively managed with electron beam, photon beam, or interstitial brachytherapy boosts. Long-term local control and cosmetic outcome are excellent regardless of which boost technique is used.

Acknowledgment: The authors thank Mamtha Balasubramaniam, M.P.H., for her assistance with statistical analysis and Vicky A. Dykes for her secretarial assistance with manuscript preparation.

REFERENCES

1. Fisher B, Anderson S, Redmond C, et al. Reanalysis and results after 12 years of follow-up in a randomized clinical trial comparing total mastectomy with lumpectomy with or without irradiation in the treatment of breast cancer. N Engl J Med 1995; 333: 1456–61.
2. Romenstaing P, Lehingu Y, Carrie C, et al. Role of a 10 Gy boost in the conservative treatment of early breast cancer: results of a randomized trial in Lyon, France. J Clin Oncol 1997; 15: 963–8.
3. Elkhuizen P, van de Vijver M, Hermanns J, et al. Local recurrence after breast-conserving therapy for invasive breast cancer: high incidence in young patients and association with poor survival. Int J Radiat Oncol Biol Phys 1998; 40: 859–67.
4. Koscielny S, Tubiana M. The link between local recurrence and distant metastases in human breast cancer. Int J Radiat Oncol Biol Phys 1999; 43: 11–24.
5. Kini V, Vicini F, Frazier R, et al. Mammographic, pathologic, and treatment-related factors associated with local recurrence in patients with early-stage breast cancer treated with breast conserving therapy. Int J Radiat Oncol Biol Phys 1996; 43: 341–6.
6. Vicini F, White J, Gustafson G, et al. The use of iodine-125 seeds as a substitute for iridium-192 seeds in temporary interstitial breast implants. Int J Radiat Oncol Biol Phys 1993; 27: 561–6.
7. Rose M, Olivetta I, Cady B, et al. Conservative surgery and radiation therapy for early breast cancer, long-term cosmetic results. Arch Surg 1989; 124: 153–7.
8. Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–9.
9. Cox DR. Regression models and life-tables. J R Stat Soc B 1972; 34: 187–220.
10. Vicini F, Horwitz E, Lacerna M, et al. Long-term outcome with interstitial brachytherapy in the management of patients with early-stage breast cancer treated with breast conserving therapy. Int J Radiat Oncol Biol Phys 1997; 37: 845–52.
11. Fourquet A, Campana F, Mosseri V, et al. Iridium-192 versus cobalt-60 boost in 3–7 cm breast cancer treated by irradiation alone: final results of a randomized trial. Radiother Oncol 1994; 34: 114–20.
12. Krishnan L, Jewell W, Mansfield C, et al. Early-stage breast cancer: local control after conservative surgery and radiation therapy with immediate interstitial boost. Radiology 1993; 187: 95–8.
13. Perez C, Taylor M, Halverson K, et al. Brachytherapy or electron beam boost in the conservation therapy of carcinoma of the breast: a nonrandomized comparison. Int J Radiat Oncol Biol Phys 1996; 34: 995–1007.
14. Mansfield C, Komarnicky L, Schwartz G, et al. Ten-year results in 1070 patients with stages I and II breast cancer treated conservative surgery and radiation therapy. Cancer 1995; 75: 2328–36.
15. Touboul E, Belkacemi Y, Lefranc J, et al. Early breast cancer: influence of type of boost (electrons vs iridium-192 implant) on local control and cosmesis after conservative therapy and radiation therapy. Radiother Oncol 1995; 34: 105–13.
16. Wazer D, Schmidt-Ullrich R, Ruthazer R, et al. Factors determining outcome for breast-conserving irradiation with margin-directed dose escalation to the tumor bed. Int J Radiat Oncol Biol Phys 1998; 40: 851–8.
17. 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: 43–8.
18. 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: 675–9.
19. Debiose D, Horwitz E, Martinez A, et al. The use of ultrasonography in the localization of the lumpectomy cavity for interstitial brachytherapy of the breast. Int J Radiat Biol Phys 1997; 38: 755–9.
20. Denham J, Sillar R, Clarke D, et al. Boost damage to the excision site following conservative surgery for breast cancer: it’s easy to miss! Clin Oncol 1991; 3: 257–61.
21. Harrington K, Harrison M, Bayle P, et al. Surgical clips in planning the electron boost in breast cancer: a qualitative and quantitative evaluation. Int J Radiat Biol Phys 1996; 34: 579–84.
22. Sedlmayer F, Hassan B, Dieter K, et al. Quality assurance in breast cancer brachytherapy: geographic miss in the interstitial boost treatment of the tumor bed. Int J Radiat Oncol Biol Phys 1996; 34: 1133–9.
23. Krishnan L, Jewell W, Krishnan E, et al. Breast cancer with extensive intraductal component: treatment with immediate interstitial boost irradiation. Radiology 1992; 183: 273–6.
24. Freedman G, Fowble B, Hanlon A., et al. Patients with early stage invasive cancer with close or positive margins treated with conservative surgery and radiation have an increased risk of breast recurrence that is delayed by adjuvant systemic therapy. Int J Radiat Biol Phys 1999; 44: 1005–15.
25. Schnitt S, Abner A, Gelman R. The relationship between microscopic margins or resection and the risk of local recurrence in patients with breast cancer treated with breast-conserving surgery and radiation therapy. Cancer 1994; 74: 1746–51.
Keywords:

Breast neoplasms; Radiotherapy; Breast implants; Interstitial implants; Breast-conserving therapy; Mammary neoplasms; Segmental resections; Lumpectomy

© 2001 Lippincott Williams & Wilkins, Inc.