- © 2012 by American Society of Clinical Oncology
Should Response to Preoperative Chemotherapy Affect Radiotherapy Recommendations After Mastectomy for Stage II Breast Cancer?
- Corresponding author: Harold J. Burstein, MD, PhD, Dana-Farber Cancer Institute, 450 Brookline Ave, Boston, MA 02215; e-mail: hburstein{at}partners.org.
A healthy 38-year-old woman presents with a palpable mass at the two o'clock position of the left breast. Mammography discloses a 3.2-cm spiculated lesion corresponding with the palpable abnormality, and ultrasonography confirms a solid mass in the breast and an enlarged, 2.1-cm lymph node with a thickened cortex in the ipsilateral axilla. Ultrasound-guided core needle biopsy of the breast mass diagnoses an invasive ductal carcinoma, poorly differentiated (grade 3), with lymphovascular invasion. Lymph node fine-needle aspiration is positive for malignant cells. Immunohistochemical studies on the breast specimen indicate that the tumor is triple negative, lacking estrogen and progesterone receptors and human epidermal growth factor 2 expression. The patient receives preoperative chemotherapy with doxorubicin and cyclophosphamide followed by paclitaxel. She has a dramatic clinical response; the breast and axillary masses shrink rapidly with chemotherapy. At the time of mastectomy, she is found to have several foci of residual invasive cancer in the breast (largest focus, 0.3 cm), located in a 3-cm tumor bed showing treatment effect (Figs 1A to 1C). The margins are negative, as are all 11 axillary lymph nodes. Her pathologic response to neoadjuvant treatment is judged Miller-Payne grade 4. She is referred for consideration of postmastectomy irradiation.
Baseline and post-treatment pathology. Photomicrographs showing (A) high-power magnification of high-grade invasive breast cancer at initial diagnosis and (B, C) results after neoadjuvant chemotherapy. (B) Low-power view showing minimal residual carcinoma and substantial treatment effect. (C) High-power view showing a 3-mm focus of residual high-grade carcinoma. These postpreoperative pathologic results are best classified as a Payne-Miller grade 4 (residual cancer burden, 1) response.
Challenges in Diagnosis and Management
Postmastectomy radiotherapy dramatically lowers the risk of locoregional recurrence1,2 and, in many patients, improves overall survival.3 When patients undergo surgery before adjuvant chemotherapy, the risk factors for locoregional recurrence and the indications for postmastectomy radiotherapy are reasonably well established.1,2,4–6 Known risk factors include younger patient age, baseline stage including size of the primary tumor and extent of nodal involvement, presence of lymphovascular invasion, and high tumor grade. Current guidelines recommend consideration of postmastectomy radiotherapy for tumors > 5 cm and/or with ≥ four positive nodes and strongly suggest consideration of radiotherapy for women with one to three positive nodes.7
In addition to providing effective systemic therapy, the goal of preoperative/neoadjuvant therapy is to improve locoregional treatment options. Downstaging the breast and axilla can render a previously inoperable patient operable and can enable breast-conserving surgery in some women for whom mastectomy would otherwise have been required. Of note, women who initially present with inoperable or locally advanced breast cancer, as well as all women who undergo breast-conserving surgery after preoperative chemotherapy, receive postoperative radiation therapy as a standard part of their multidisciplinary treatment program.
Understanding the extent to which initial clinical stage and final pathologic stage after preoperative chemotherapy contribute to the risk of locoregional recurrence has been a challenge, along with the related question of how response to chemotherapy in the preoperative setting should affect radiation therapy recommendations after mastectomy. The clinical criteria for decision making are unclear, because there are no data from randomized trials, and thresholds for when radiation therapy is warranted are not uniformly agreed on. The real debate has been whether an excellent response to preoperative chemotherapy is a sufficiently favorable prognostic finding that radiation therapy can be omitted. The stakes are high. Such patients may be able to avoid irradiation and its associated adverse effects, inconvenience, and cost. But the consequences of omitting radiation therapy might be significant. The Early Breast Cancer Trialists' Collaborative Group3 showed that postmastectomy radiation therapy in patients at high risk for locoregional recurrence not only decreased locoregional recurrence but also improved overall survival.
Summary of the Relevant Literature
Retrospective analyses have explored interactions between treatment response and the need for radiation therapy after preoperative chemotherapy and mastectomy. Investigators at MD Anderson Cancer Center have shown that radiation therapy after preoperative chemotherapy and mastectomy improves outcomes for patients with locally advanced breast cancer.8 Stage at both baseline and after preoperative chemotherapy predict for locoregional recurrence.9,10 Those patients presenting with locally advanced breast cancer who achieve a complete pathologic response still seem to benefit from postmastectomy radiation therapy.11 A goal of retrospective subset analyses has been to identify a cohort of patients with so favorable a prognosis that additional treatment in the way of radiation therapy is unlikely to improve meaningfully their long-term prognosis. Along these lines, investigators from MD Anderson identified a small subset of patients with clinical stage II breast cancer at baseline who had a complete pathologic response or had limited residual disease in lymph nodes at the time of completion axillary dissection. This cadre of patients had a low risk of subsequent regional tumor recurrence (5%; n = 42) in the absence of postmastectomy radiation therapy.12
Given the limited publications on using chemotherapy response to determine whether to add radiation therapy after mastectomy, the data from the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-18 and B-27 studies published in the report accompanying this article are most welcome.13 The NSABP data offer both larger patient numbers, mature follow-up, and the systematic omission of radiation therapy after mastectomy, minimizing selection bias. The vast majority of patients presented with clinical stage II breast cancer, and thus this data set complements the MD Anderson series, which was dominated by patients with stage III disease. Among the 1,071 patients treated with mastectomy, multivariable analysis showed that both clinical and pathologic features were significant contributors to the risk of locoregional recurrence. Pathologic response in the breast and nodes (residual disease in the breast and nodes negative v complete pathologic response: hazard ratio [HR], 2.21; residual positive nodes v complete pathologic response: HR, 4.48) was a more powerful predictor of locoregional recurrence than initial nodal status (clinically node positive v clinically node negative: HR, 1.53) or tumor size (< 5 v > 5 cm: HR, 1.58). Notably, all patients with residual nodal involvement by cancer had a > 10% risk of locoregional recurrence.
The NSABP data did not address the importance of lymphovascular space invasion, molecular subtype, or gradation of response as predictors of recurrence. Other investigators found that lymphovascular space invasion increases the risk of locoregional recurrence in the adjuvant setting5,14,15 and after preoperative systemic therapy.16 Patients with triple-negative tumors may have a higher-than-expected risk of locoregional recurrence.17–19 Interestingly, this risk factor is modified dramatically by response to preoperative systemic therapy. Caudle et al20 reported that patients with triple-negative tumors with a pathologic complete response achieved locoregional control rates of 99%; in contrast, persistent disease in the breast or lymph nodes was particularly ominous in this group. This finding is concordant with the risk of distant recurrence after preoperative chemotherapy in women with triple-negative tumors, where again pathologic complete response is a favorable prognostic factor. Finally, the NSABP data do not assess gradations of response aside from yes or no in their characterization of residual tumor. Other investigators have suggested that near-complete response with evidence for dramatic treatment effect and minimal residual cancer burden may carry a better prognosis than seen in patients with less overt tumor change in response to preoperative therapy.
Suggested Approaches to Management
The clinical impact of the NSABP data resides most clearly in the finding of cohorts at low risk of locoregional recurrence after mastectomy such that radiation therapy is not likely to be valuable. Precise numeric thresholds are not established; however, a 10% risk of locoregional recurrence is a threshold above which many clinicians recommend consideration of radiation therapy. This threshold for treatment reflects the likely gains with radiation therapy. Given a 10% risk of locoregional recurrence, a reduction of one half to two thirds using radiation therapy might achieve an absolute 5% to 7% decrease in event rate for patients, which in turn might translate into measurable survival gains. Table 1 summarizes the rates of locoregional recurrence by baseline and postpreoperative chemotherapy stage. This experience suggests that patients presenting with clinical stage II breast cancer who experience complete pathologic response with preoperative chemotherapy have a low risk of locoregional recurrence in the absence of postmastectomy radiation therapy.
Rates of Locoregional Recurrence After Mastectomy in NSABP B-18 and B-27
In our patient, the presence of residual disease in the breast seems likely to increase her risk of locoregional recurrence from trace to approximately 10% through 10 years of follow-up. In addition, her presentation includes other adverse prognostic risk factors for locoregional recurrence such as triple-negative histology and young age. Thus, despite an excellent response to induction chemotherapy, we would offer postmastectomy radiation therapy (Table 2). Radiation therapy generally reduces the risk of locoregional recurrence by approximately two thirds,1,2 which in this case would decrease a risk of 10% to 12% to < 4%. We acknowledge that any associated improvement in overall survival would be extremely small. Whether this small benefit is worthwhile to the patient would need to be discussed in light of the adverse effects of radiation treatment, its potential impact on the timing and nature of reconstruction options, and the perceived value of preventing locoregional recurrence. The chest wall would be included routinely in the treatment fields of patients receiving postmastectomy radiation therapy, because it is the site of the majority of locoregional recurrences. Radiation to the regional lymph nodes after mastectomy is standard in the adjuvant setting. However, after preoperative chemotherapy, it is less clear whether axillary dissection lowers the risk of nodal recurrence such that specific regional nodal irradiation could be omitted. Daveau et al21 analyzed the impact of regional nodal irradiation after preoperative chemotherapy in a cohort of women who had breast-conserving surgery and, in their retrospective series, found no improvement in rates of locoregional recurrence or survival for nodal irradiation. The NSABP data would suggest that the risk of nodal recurrence is rather low, < 10%. However, because of the small additional morbidity of regional nodal irradiation coupled with the low chance of salvage treatment should there be a recurrence, we would favor radiation to the level III axillary and supraclavicular nodes in our patient. We would not include the full axilla in the radiation field, because the risk of axillary recurrence is low after a completion axillary dissection.4,22 The NSABP data coupled with other retrospective series suggest a low (≤ 5%) risk of regional recurrence in women with stage II breast cancer who achieve complete pathologic response with neoadjuvant chemotherapy, and had our patient experienced that with preoperative treatment, we would not have recommended postmastectomy radiation therapy despite the presence of a clinically positive axillary lymph node at presentation.
Treatment Recommendations for Postmastectomy Radiation Therapy in Women Receiving Preoperative Chemotherapy
Preoperative systemic therapy can safely convert some patients with large tumors who may, at diagnosis, require mastectomy to breast-conservation candidates.23 Despite having triple-negative disease, had our patient desired breast conservation and achieved negative surgical margins, this would also have been a reasonable option. There is no evidence that mastectomy confers a benefit over breast-conserving therapy in patients with early-stage, triple-negative breast cancers.24
The original impetus for neoadjuvant chemotherapy was to improve survival in women with breast cancer beyond the benefits seen with adjuvant therapy. To date, preoperative treatment has not achieved that goal.25 However, modification of locoregional therapy has emerged as a clear and compelling benefit from preoperative chemotherapy. Not only can preoperative treatment permit breast conservation, but it can also potentially enable the selection of a favorable subgroup of patients who may do well with less aggressive locoregional therapy. These additional data from the NSABP provide an important step toward the goal of more tailored radiation therapy options.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Provision of study materials or patients: All authors
Manuscript writing: All authors
Final approval of manuscript: All authors
Acknowledgment
We thank our colleagues Andrea Richardson, MD, PhD, for providing pathology images and Jay Harris, MD, for thoughtful review of the manuscript.
