Current Application and Research Directions for Partial-Breast Irradiation

MammoSite Balloon Catheter Brachytherapy

The MammoSite balloon brachytherapy device (MammoSite Radiation Therapy System; Cytyc Corp; Marlboro, Mass), introduced in 2002, is a form of intracavitary brachytherapy that is simpler than interstitial brachytherapy in terms of technique and treatment planning. The apparatus consists of a double-lumen catheter that is 15 cm in length and 6 mm in diameter. The catheter contains a central lumen that allows for a high-dose-rate Ir-192 source, while the balloon is located distally. The balloon is available in two sizes when inflated—either 4-5 cm or 5-6 cm in diameter—for variability in the dimensions of a lumpectomy cavity.

The catheter is implanted after lumpectomy either at the time of surgery directly into the cavity, after surgery under ultrasound guidance through a small, separate incision, or after surgery directly into the cavity through the healing lumpectomy wound. After catheter placement, computed tomography (CT) of the breast is performed prior to initiation of treatment to determine whether the balloon-to-skin distance is ≥ 5 mm, the conformity of the balloon to the walls of the lumpectomy cavity is > 90%, and symmetry exists between the balloon and the center shaft of the catheter (Figure 2). These guidelines were developed by the ASBS to ensure proper patient selection for this technique.

The ASBS published the outcomes of insertion techniques in a registry trial of 1,403 patients who received MammoSite breast brachytherapy.[51] The trial, initiated in May 2002 by the manufacturer (who relinquished control of the trial in November 2003), accrued patients from 87 institutions over 30 months. Patients were enrolled per the ABS eligibility criteria listed earlier. A total of 1,237 patients (87%) received APBI via MammoSite, 43 (3%) received a boost via MammoSite, and 123 (9%) underwent catheter explantation. Explantation was performed because of poor skin spacing (35%), irregular cavity (28%), positive margins (9%), and balloon failure (9%).

A subset analysis of 198 patients with ductal carcinoma in situ with a median follow-up of 15 months was presented at the ASTRO annual meeting in November 2006.[52] These patients received 34.0 Gy in 10 fractions over 5 days. They reported a 1% (n = 1) 2-year actuarial rate of ipsilateral breast tumor recurrence. At 6, 12, 24, and 36 months, the cosmetic results were reported as good or excellent in 94% (n = 180), 97% (n = 103), 89% (n = 28), and 90% (n = 10) of patients. They found that skin spacing greater than 6 mm, larger balloon-to-skin distance, and closed cavity placement were all associated with a good/excellent cosmetic result. Infection occurred in 7% of patients and did not reach statistical significance in terms of cosmetic outcome.

At the ASTRO annual meeting in November 2006, Cuttino et al presented a pooled analysis of patients with stage 0, I, and II breast carcinoma treated with MammoSite at nine institutions between 2000 and 2004.[53] All 483 patients received 34 Gy in 10 fractions over 5 days. The median follow-up was 2 years, and all patients had a minimum follow-up of 1 year. They found an in-breast failure rate of 1.2% (n = 6), but only 0.4% of all patients experienced a failure that was characterized as a true recurrence or marginal miss. Cosmetic results were reported as good/excellent in 91% of patients. Administration of prophylactic antibiotics, skin spacing > 5 mm, and use of multiple dwell positions contributed to less dermatologic toxicity in terms of severe acute skin reactions, severe hyperpigmentation, and grade 3/4 acute skin reactions, respectively.

A multi-institutional phase II clinical trial was conducted between May 2003 and January 2006 to evaluate the utility of MammoSite in patients with ductal carcinoma in situ.[54] Eligibility criteria included the following: age ≥ 45 years, unicentric pure DCIS, ≥ 1-mm margins, tumor size ≤ 5 cm, clinically node-negative, and a postlumpectomy mammogram showing complete resolution of any suspicious microcalcifications. A total of 133 patients were enrolled, 117 of whom received the MammoSite implant. Seventeen patients underwent removal of the implant for various reasons, including suboptimal skin distance, positive margins, and irregular cavity. Thus, 100 patients completed treatment, with a median follow-up period of 9 months. Two patients experienced an ipsilateral breast recurrence, with one being a true recurrence/marginal miss. Approximately 98% of patients reported a good/excellent cosmetic result, and the researchers found a 4% infection rate, consistent with other series. Details of these studies can be found in Table 3.

3D Conformal External-Beam Radiotherapy

Although the most widely used form of radiation therapy to treat carcinomas of all types, this technique is associated with the least amount of data supporting its role in APBI. Three-dimensional conformal EBRT (3D-CRT) is a noninvasive method of delivering APBI that provides increased dose homogeneity, leading to the theoretical potential for better cosmetic outcomes compared to other techniques (Figure 3). Furthermore, 3D-CRT is less operator-dependent than interstitial brachytherapy and patients do not have to undergo explantation of brachytherapy catheters, as do a proportion of patients treated with MammoSite. In addition, cost-analysis investigations have indicated that 3D-CRT may be less expensive than brachytherapy techniques that do not require an extra surgical procedure or inpatient hospitalization.[55,56] With the emergence of computed tomography (CT)-based simulation, easier identification of the tumor bed and decreased doses to critical normal structures (in addition to the above reasons) have led to an interest in delivering APBI using 3D-CRT.

• William Beaumont Hospital Study—Vicini and colleagues at the William Beaumont Hospital first used 3D-CRT to deliver APBI in a select group of patients using active breathing control to account for movement of the breast secondary to respiration.[57] They found this technique to be feasible and initiated a phase I/II trial further investigating the role of 3D-CRT in patients who met the eligibility criteria for RTOG 95-17.[58] Thirty-one patients were enrolled and underwent CT-based planning. The clinical target volume (CTV) was defined as the lumpectomy cavity plus a 1- to 1.5-cm margin, limited by the skin surface and chest wall. A 1-cm margin was added to form the planning-target volume (PTV). The first five patients received 34 Gy over 10 twice-daily fractions, while the remainder received 38.5 Gy over 10 twice-daily fractions. With a median follow-up of 10 months, the study revealed no recurrences and a 100% rating of good/excellent cosmesis. The technical aspects of this study were found to be feasible and easily reproducible.

• RTOG Study—The RTOG conducted a phase I/II trial to evaluate the feasibility and reproducibility of 3D-CRT in delivering APBI.[59] They enrolled 58 patients with stage I or II invasive ductal carcinoma involving lesions ≤ 3 cm and negative surgical margins. After lumpectomy, the surgical cavity was defined on CT scan, and this was denoted as the gross tumor volume (GTV). An expansion of 1.0 to 1.5 cm was added to form the CTV. The CTV was restricted to within 5 mm of the skin surface and lung-chest wall interface. An additional margin of > 1.0 cm was provided to calculate the PTV, to account for penumbra. However, a separate PTV structure was defined to exclude this volume to within 5 mm of the skin surface and lung-chest wall interface and was used for the dose-volume histogram (DVH) analysis.

A total of 38.5 Gy was delivered in 10 fractions over 5 days.[59] Patients were not treated using active breathing control. Port films and orthogonal pair films were taken four times during the course of therapy. The dose-volume constraints were as follows: < 50% of the ipsilateral breast should receive < 50% of the prescribed dose and 25% of the ipsilateral whole breast should receive the prescribed dose; the contralateral breast received < 3% of the prescribed dose; < 10% of the ipsilateral lung could receive 30% of the prescribed dose; < 10% of the contralateral lung could receive 5% of the prescribed dose; < 5% of the heart could receive a maximum of 5% of the prescribed dose for right-sided lesions, and for left sided lesions, the volume of lung receiving 5% of the dose should be less than conventional WBI. Finally, the maximum dose to the thyroid could be 3% of the prescribed dose.

The primary endpoint of the study was to determine whether 3D-CRT was reproducible. Indeed, the investigators confirmed the technique's, as there were only four cases with major variations in the first 42 evaluable plans. All four of these major variations arose from the strict DVH constraints on the ipsilateral lung. The results of this study served as the foundation for the external-beam arm of the NSABP B-39/RTOG 0413 clinical trial that opened in 2005.

• NYU Study—Formenti and colleagues at New York University designed a phase I/II study in 2000 evaluating the role of APBI delivered, using 3D-CRT, to patients laying in the prone position. Advocates of this approach maintain that the prone position reduces normal tissue motion secondary to respiration and cardiac systole, and further allows for the removal of the heart and lungs from the treatment field. This group treated 78 patients with a median follow-up of 28 months—the longest follow-up of APBI using 3D-CRT to date.[60] As of this writing, there have been no recurrences, and cosmesis was rated as good/excellent in 92% of patients. Details on the experience with 3D-CRT are listed in Table 4.

Intraoperative Radiotherapy

Intraoperative radiotherapy (IORT) is a method of delivering a single dose of radiation directly to the tumor bed or to the exposed tumor at the time of surgery. For most tumor sites, it can be used adjuvantly after surgery or as a boost, to be followed by fractionated EBRT for either palliative or curative intent. The objective of IORT is to deliver a high single dose of electrons or low-energy photons to the exposed target volume while displacing critical, dose-limiting structures.

Traditionally, patients were transported from the operating room to the radiotherapy suite during surgery, or surgery was performed in the radiotherapy suite. However, several devices that function as mobile IORT machines are now available. For example, the Intrabeam (Carl Zeiss AG, Oberkochen, Germany) produces 50-kVp x-rays. Other such devices include the Mobetron System (Oncology Care Systems Group of Siemens Medical Systems, Intraop Medical Inc, Santa Clara, Calif) and the Novac 7 System (Hitesys spA, Aprilia, Italy), both of which produce electrons between 4 and 12 MeV.

While the use of IORT has been most extensively studied in sarcomas, abdominal, genitourinary, and gynecologic malignancies, the experience in breast cancer is limited. Recently, it has been actively explored in Europe as a component of BCT, in light of longer treatment delays secondary to rising costs and poor access to health care. Advocates of IORT affirm that a geographic miss, which may occur with standard EBRT, is avoided with this technique, since the surgical bed is visualized at the time of radiation therapy. The limitation of radiotherapy solely to the tumor bed is dependent on histology, accurate patient selection, and the skill of the surgeon in obtaining negative surgical margins. This is crucial to the success of IORT.

The theoretical benefit of IORT is the delivery of a large single dose of radiation, either with x-rays or electrons. Large single doses of radiation are thought to be more effective in "late-responding" tissues such as the lung and spinal cord, which have a low alpha/beta ratio (2.0-6.3 and 1.7-4.9, respectively).[61] Breast tissue and tumors are thought to have an alpha/beta ratio of 10. The linear-quadratic model, which uses alpha and beta values to determine the relative effectiveness of a fractionation scheme on early- and late-responding tissues, may not be reliable for use with single-fraction sizes greater than 6 to 8 Gy.[62] Given this limitation, estimates of single-fraction sizes comparable to the standard 60 Gy delivered over 30 fractions are thought to be 20 to 22 Gy.[62,63] However, the late effect on breast tissue using large single-fraction sizes is unknown.

• Prospective Studies—Several prospective series of patients have been studied, in which IORT was used to deliver radiotherapy as a part of BCT. A study by Veronesi at the European Institute of Oncology examined 237 patients with tumors < 2 cm who received quandrantectomy followed by immediate IORT using the Novac 7, with 222 patients receiving 21 Gy using 3- to 9-MeV electrons.[63] With a mean follow-up of 19 months, 2.5% of patients developed adverse effects secondary to IORT and only 1.3% patients developed a recurrence in the ipsilateral breast, all of which were outside the treatment field. The European Institute of Oncology has met its goal of accruing 824 patients randomized to either WBI plus boost or a single dose of 21 Gy using IORT. The results from this trial are eagerly awaited.

Another study conducted by the University College of London examined 185 patients with early-stage breast cancer who received IORT using the Intrabeam system.[62,64] The prescribed dose was one 5- to 20-Gy fraction at a depth of 1 and 0.2 cm, respectively. Twenty-two patients received IORT alone, while 163 patients received it as a boost prior to WBI. Their data has not been reported in full, but only two recurrences were noted and cosmetic outcomes were described as good. This group is also accruing 1,600 patients to a randomized study designed to test equivalence comparing WBI plus boost to single-fraction IORT.