For women with ductal carcinoma in situ (DCIS), should radiation therapy (RT) be recommended in all cases, or are there situations in which RT can be omitted? These were among the questions facing the 26-member Breast Cancer Guidelines Panel of the National Comprehensive Cancer Network (NCCN). In addition to the controversial role of RT in the treatment of DCIS, other topics highlighted in the updated 2008 guidelines were the use of lymph node surgery in patients with DCIS, breast reconstruction following mastectomy, a multigene assay for guiding treatment recommendations in particular patient populations, new agents such as ixabepilone (Ixempra) for metastatic disease, and inflammatory breast cancer.
One major change in the breast cancer guidelines was incorporation of the option of lumpectomy without lymph node surgery and RT for essentially all women with DCIS, which received a category 2B recommendation (lower quality of evidence and nonuniform consensus). For the primary treatment of these women, it joins lumpectomy without lymph node surgery with whole-breast RT (category 1) and total mastectomy with or without sentinel node biopsy (SNB).
Data from two large randomized clinical trials—NSABP
(National Surgical Adjuvant Breast and Bowel Project) B-17 (Fisher B et al.
J Clin Oncol 1998;16:441−452) and an EORTC (European Organization for
Research on the Treatment of Cancer) study (Bijker N et al. J Clin Oncol
2006;24:3381–3387)—support the use of RT in the treatment of patients with
DCIS. Both trials showed reduced local failure rates in women treated with
breast-conserving surgery (with negative margins) and RT compared with women
undergoing breast-conserving surgery alone, with about a 50% reduction in the
risk of local failure.
However, these benefits did not extend to overall survival. “I would challenge anyone to find any data to show that the treatment of DCIS with radiation or without radiation, by mastectomy versus breast-conserving surgery, with or without hormonal therapy provides any survival advantage,” stated Stephen B. Edge, MD, of Roswell Park Cancer Institute.
In both of these trials, the researchers “were unable to identify any subset of women where there was no reduction in the risk of local recurrence with RT,” Dr. Edge reported. However, many groups have tried to define subsets of women with “low- risk” DCIS for whom the risk of local failure appears to be low enough to omit RT, he asserted. In two trials evaluating the use of lumpectomy without RT to treat women with either small, low-grade disease with wide margins or negative margins following re-excision (Wong JS et al. J Clin Oncol 2006;24:1031−1036) or women with either low- or intermediate-grade DCIS < 2.5 cm or high-grade DCIS < 1 cm (Hughes L et al. 2006 San Antonio Breast Cancer Symposium. December 14–17, 2006; San Antonio, Texas. Abstract 29), local recurrence rates of 12% in the former trial and 6.8% and 13.7%, respectively, for the two groups in the latter trial were reported.
“The question is whether these numbers represent figures that are low enough to justify omitting RT,” Dr. Edge remarked. A scoring system to predict the likelihood of local recurrence for women with DCIS based on tumor size, margin width, patient age, and pathologic tumor characteristics has been developed (Silverstein MJ, Lagios MD. J Surg Oncol 2007;95:605−609), although Dr. Edge cautioned that these results are based on a retrospective analysis.
According to the NCCN breast cancer outcomes database for women treated for DCIS from 2003 to 2006, 81% received RT. “It’s pretty obvious that a large fraction of women in the United States who were treated with breast-conserving surgery did not receive RT,” Dr. Edge admitted. Therefore, the updated NCCN guidelines include a statement reflecting the uncertainty involved in selecting women with DCIS who can be appropriately treated by lumpectomy alone, saying that whole-breast RT following lumpectomy reduces recurrence rates in DCIS by about 50%. It mentions factors that affect the risk of recurrence, indicates that some low-risk patients may be treated by excision alone, and declares that there appears to be no difference in patient survival among the local treatments.
“So the updated recommendations really place the onus back on the physician to have an appropriate discussion with the patient with DCIS when making decisions regarding the use of RT,” Dr. Edge summarized.
On the issue of axillary nodal staging in patients with DCIS, Dr. Edge said that he was “surprised by how many people do SNB for all women with DCIS in the community.” A study evaluating results of two trials from the NSABP (B-17 and B-24; Julian TB et al. Ann Surg Oncol 2007;14:2202−2208), which addressed whether SNB is advised for these women, does not support the use of SNB in patients with conservatively treated, localized DCIS, he said.
“The large majority of patients with DCIS who have positive nodes have disease detectable only by immunohistochemistry (IHC),” remarked Dr. Edge, and studies have shown that overall disease-specific survival did not differ for patients with DCIS who had IHC-detected nodes versus those who did not (El-Tamer M et al. Ann Surg Oncol 2005;12:254–259). Based on these and other data, the 2008 NCCN guidelines recommend that women with DCIS being treated with lumpectomy do not undergo lymph node surgery; however, the recommendation for women with DCIS being treated by mastectomy includes SNB. The latter recommendation is based on the unlikelihood that a subsequent SNB could be performed if invasive cancer was detected after mastectomy, said Dr. Edge.
Another change in the 2008 guidelines regarding RT was made for women with invasive breast cancer. Based on evidence from randomized trials showing a survival advantage with RT after mastectomy for women with early-stage invasive breast cancer characterized by one to three positive axillary lymph nodes, the NCCN Breast Cancer Guidelines Panel strengthened the recommendation for the use of RT in this group of patients. The guidelines now state that postmastectomy RT to the chest wall and supraclavicular area should be “strongly considered” for these women. A recommendation for consideration of RT to internal mammary nodes is also included in the guidelines for patients with one to three positive nodes following mastectomy, although this suggestion was more controversial.
“A shortcoming of our guidelines over the past years has been the absence of a guideline covering techniques of RT,” Dr. Edge explained. So, for the first time, the 2008 NCCN guidelines offer an outline on the principles of RT, including best practices, doses, the fields, and mention of the investigational use of partial breast irradiation.
Principles of breast reconstruction are described in another new guideline included this year. It focuses on reconstruction following mastectomy, including information on skin-sparing mastectomy and the sequencing of reconstruction with respect to RT. Dr. Edge emphasized the importance of a multidisciplinary evaluation within a breast program on the use of reconstruction and of including patients in the discussion with careful counseling.
Furthermore, “reconstruction has the potential for affecting the delivery of RT,” he said. For instance, according to an M. D. Anderson Cancer Center series (Motwani SB et al. Int J Radiat Oncol Biol Phys 2006;66:76−82), 52% of women who underwent reconstruction and received RT had “compromised” RT, either in terms of delivery of radiation or dosing to underlying structures. In addition, women considering reconstruction with autologous tissue should strongly “consider delaying reconstruction until after RT,” Dr. Edge suggested, since pre-irradiation reconstruction may lead to a worse cosmetic outcome. However, pre-irradiation breast reconstruction is preferred if an implant is used, he added, because it can allow sparing of skin and expansion of non-radiated skin.
In terms of systemic adjuvant therapy for invasive breast cancer, the NCCN guidelines divide the usual histologies of breast cancer into four biologically important subsets for making treatment decisions, based on hormone receptor status (estrogen, progesterone, and HER2 [human epidermal growth factor receptor 2]), said Robert W. Carlson, MD, of Stanford Comprehensive Cancer Center and Chair of the NCCN Breast Cancer Guidelines Panel. The four subsets then are divided further by classic histopathologic features, Dr. Carlson explained, that help to predict further risk of recurrence of disease.
In the updated guidelines, for the first time, the panel has incorporated the option of using a multigene assay (Oncotype DX) in making treatment recommendations for selected patients, stated Dr. Carlson. The Oncotype DX assay provides an estimate of risk of recurrence termed “a recurrence score,” which can then be used to place a patient into a low-risk (score of < 18), intermediate-risk (18–30), or high-risk (³ 31) group. However, the recurrence score is “really a continuous variable,” he said.
Dr. Carlson discussed the evidence from retrospective analyses of data from the placebo-controlled NSABP B-14 and B-20 trials, which demonstrated that recurrence score was predictive of benefit from tamoxifen (B-14 trial; Paik S et al. N Engl J Med 2004;351:2817–2826) and from the addition of chemotherapy to tamoxifen (B-20 trial; Paik S et al. J Clin Oncol 2006;24:3726–3734) for patients with node-negative, hormone receptor-positive breast cancer, thereby providing support for inclusion of this assay in the guidelines. In the placebo arm of the NSABP B-14 study, “women with intermediate to high recurrence scores did have a higher risk of death at 10 years than patients with a low recurrence score,” reported Dr. Carlson, whereas the NSABP B-20 study demonstrated that the most marked benefit of chemotherapy was seen in patients with high recurrence scores.
Systemic adjuvant treatment recommendations based on results of Oncotype DX 21-gene reverse transcriptase-polymerase chain reaction (RT-PCR) assay
To clarify the clinical implications of using this assay, Dr. Carlson concluded that this test is limited to women with hormone receptor-positive, HER2-negative, node-negative disease and has been validated only in tamoxifen-treated patients who have received first-generation chemotherapy. For women with a low recurrence score, adjuvant endocrine therapy alone is recommended. For women with an intermediate recurrence score, adjuvant endocrine therapy with or without chemotherapy is recommended. For women with a high recurrence score, adjuvant endocrine therapy plus adjuvant chemotherapy is recommended.
All of these assay-based recommendations are deemed category 2B, indicating there was a lower level of evidence and a lack of uniform agreement among the panel.
A new epothilone analog, ixabepilone (Ixempra), has been added to the 2008 guidelines. It has demonstrated activity as monotherapy in phase II trials of patients with metastatic breast cancer who have received prior treatment with an anthracycline; a taxane; or an anthracycline, a taxane, and capecitabine (Xeloda), reported Dr. Carlson. In a phase III randomized trial evaluating capecitabine with or without ixabepilone for patients with locally advanced or metastatic breast cancer resistant to anthracycline and taxane therapy (Thomas ES et al. J Clin Oncol 2007;25:5210−5217), the hazard ratio for disease progression favored the combination of drugs (5.8 months vs 4.2 months; hazard ratio = 0.75 [0.64–0.88]; P = 0.0003). “This is a highly statistically significant difference,” he added; “however, it does translate into only a 1.6-month difference in median progression-free survival, and there have been no survival data reported from this trial.”
Concerns about liver function with the use of ixabepilone were raised in this study. Approximately 30% of patients with grade 2 or greater liver dysfunction who were treated with capecitabine plus ixabepilone died of neutropenia-related complications. “Ixabepilone should not be administered to patients with significant liver function abnormalities,” warned Dr. Carlson. Furthermore, the combination of these two agents resulted in higher rates of grade 3/4 myelosuppression and peripheral neuropathy than did capecitabine alone.
In light of these study findings, the combination of ixabepilone and capecitabine for recurrent or metastatic breast cancer received a category 2B recommendation in the updated guidelines. The limited experience with ixabepilone, the substantial toxicity observed with this drug alone and in combination with capecitabine, and the absence of any survival data are the reasons for the nonuniform agreement of the panel, Dr. Carlson stated.
One last new addition to the 2008 NCCN guidelines centers on inflammatory breast cancer (IBC). “A consistent criticism of our guidelines over the past few years has been that we have included IBC treatment as part of our locally advanced breast cancer guidelines,” Dr. Carlson revealed. Now, the panel has created a separate guideline addressing IBC. Historically linked to an unfavorable prognosis, IBC is usually associated with dermal lymphatic involvement, he stated. “Clinically, this looks like cellulitis of the breast, and any cellulitis of the breast that occurs in a nongravid, nonlactating woman should be assumed to be inflammatory breast cancer” until proved otherwise.
All women with IBC should be treated with preoperative chemotherapy, classically anthracycline-based chemotherapy regimens. If the patient has HER2-positive disease, trastuzumab (Herceptin) should be incorporated. For the vast majority of women who respond to neoadjuvant chemotherapy, total mastectomy with axillary dissection plus RT is recommended, Dr. Carlson said. For women who do not respond to neoadjuvant chemotherapy, some alternative systemic chemotherapy may be appropriate, he concluded.