Multidisciplinary considerations in the treatment of triple-negative breast cancer.

Abstract

A healthy, premenopausal woman aged 45 years discovered a lump in her right breast and her right axilla. Physical examination disclosed a 3-cm mass with ill-defined borders in the upper outer quadrant (UOQ) of the right breast, without skin changes, and a 2-cm mobile right axillary lymph node. There was no cervical or supraclavicular adenopathy. Bilateral diagnostic mammography demonstrated a cluster of masses adjacent to the palpable abnormality in the UOQ of the right breast and a prominent right axillary lymph node; ultrasound also demonstrated a cluster of masses in the UOQ of the right breast measuring 1.6 cm and ≥2 enlarged axillary lymph nodes with thickened cortices. An ultrasound-guided core biopsy of the breast mass with clip placement confirmed a high-grade, invasive ductal carcinoma with associated ductal carcinoma in situ. The tumor lacked estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) expression on immunohistochemistry (IHC) and was HER2-negative by FISH (ie, triple-negative). A fine-needle aspiration biopsy of an abnormal axillary lymph node, with placement of a high-visibility Q-clip, confirmed metastatic carcinoma. Her disease was staged as clinical T2N1 based on the size of the palpable mass and axillary adenopathy: American Joint Committee on Cancer clinical prognostic stage IIB.1 Her family history included breast cancer in 2 paternal cousins, a paternal aunt, and a paternal grandmother. Multigene panel testing for hereditary cancer showed no deleterious mutations. She was evaluated by a multidisciplinary (surgical, medical, and radiation oncology) team. With a triple-negative cancer involving the axillary lymph nodes, she warranted either neoadjuvant (preoperative) or adjuvant (postoperative) chemotherapy, and the discussion centered on the sequencing of treatments, the surgical options of breast-conservation therapy (BCT) versus mastectomy, the extent of axillary surgery, and the need for postmastectomy radiation. On the basis of the tumor size-to-breast size ratio, she was only a borderline candidate for BCT and, with palpable adenopathy, she would require an axillary lymph node dissection (ALND) were surgery to be pursued as the first treatment modality. A magnetic resonance imaging scan obtained to clarify the extent of disease demonstrated a heterogeneously enhancing mass, representing the biopsy-proven malignancy, and an adjacent nonmass-like enhancement measuring 2.9 × 2.0 × 1.6 cm occupying the UOQ of the right breast. Axillary adenopathy was also seen, with the largest node measuring 2.1 cm (Fig. 1). The team recommended neoadjuvant chemotherapy (NAC) with the goal of surgical downstaging to permit breast-conserving surgery and also possibly to avoid ALND. She received preoperative, dose-dense doxorubicin plus cyclophosphamide followed by paclitaxel (ddACT). Her course was complicated by alopecia and the onset of amenorrhea with menopausal symptoms. After NAC, her breast and axillary examinations normalized. Repeat magnetic resonance imaging demonstrated a residual enhancing mass now measuring only 1.5 cm and normal-appearing axillary nodes. After considering surgical options, the patient chose mastectomy for management of the index lesion and contralateral prophylactic mastectomy for future risk reduction; she also pursued bilateral tissue-expander reconstruction. Because she no longer had palpable disease in the axilla, she was offered sentinel lymph node (SLN) biopsy with seed localization of the clipped axillary node for axillary staging. Intraoperatively, the clipped node was identified as one of three SLNs, and all were negative for malignancy on intraoperative touch preparation evaluation. Final pathology showed 9 mm of residual invasive carcinoma in the breast, persistently high-grade, without ER, PR, or HER2 expression, and the clipped SLN had a 2-mm focus of carcinoma (Fig. 2); thus the pathologic stage was ypT1bN1mi (where ypT1b indicates the size of the residual invasive cancer after NAC and N1mi indicates residual micrometastatic nodal disease). The residual cancer burden (RCB) index, which is based on the extent of residual tumor, the overall cancer cellularity (45%), and the status and size of any lymph node metastases, was RCB III.2 She was returned to the operating room for completion ALND and had 10 lymph nodes removed, one of which contained a 4-mm focus of cancer, yielding a final pathological stage of ypT1bN1. She then received adjuvant radiation therapy (RT) to the chest wall, axillary apex, supraclavicular nodes, and internal mammary nodes in the first 3 intercostal spaces (Fig. 3) followed by adjuvant capecitabine chemotherapy. This patient has stage II, triple-negative breast cancer (TNBC) and warrants chemotherapy in either the adjuvant or neoadjuvant setting. For several reasons, NAC is the preferred approach (Fig. 4). The canonical trial, National Surgical Adjuvant Breast and Bowel Project (NSAPB) B-18, compared NAC versus adjuvant chemotherapy for operable breast cancer and established important clinical principles.3 Neoadjuvant treatment achieves the same long-term tumor control and survival as adjuvant therapy; neoadjuvant treatment can downstage the tumor, facilitating breast-conserving surgery in many women who required mastectomy at presentation; and the extent of response to neoadjuvant treatment—particularly a pathological complete response (pCR)—serves as a powerful prognostic marker for recurrence risk. These findings have subsequently been validated through innumerable trials and meta-analyses.4, 5 In many respects, this patient is the ideal candidate for NAC. Tumors that are high-grade, triple-negative, or HER2-positive are most likely to have dramatic responses to NAC. With a tumor between 2 and 5 cm in size, creating a borderline situation for breast-conserving surgery, treatment response is likely to alter her surgical options. NAC is also increasingly recognized as a strategy for downstaging the axilla. Despite palpable axillary adenopathy, were she to experience a good clinical response to NAC, she could be a candidate for SLN biopsy to stage the axilla and ultimately may avoid ALND if she achieves a pCR in the SLNs.6 Anthracycline-based, alkylator-based, and taxane-based chemotherapy regimens such as ddACT remain the gold standard as either adjuvant or neoadjuvant treatment for node-positive, TNBC.7 Trials have examined whether additional neoadjuvant treatments could improve outcomes in TNBC. The addition of carboplatin to standard chemotherapy regimens consistently improves the rate of pCR.8-10 However, its impact on other clinical endpoints, including rates of breast-conserving surgery and the risk of recurrence, is less clear compared with standard anthracycline-based, taxane-based, and alkylator-based chemotherapy.8, 11 Checkpoint inhibitors such as pembrolizumab and atezolizumab augment pCR rates when added to chemotherapy in the neoadjuvant treatment of TNBC12, 13; however, the long-term effect of immunotherapy agents on recurrence or survival remains to be determined, and these approaches are not yet routine in clinical care. Thus ddACT (or related variations), without the addition of platinum salts or checkpoint inhibitors, remains the preferred regimen for neoadjuvant therapy in TNBC, although maturating data from clinical trials may redefine the role of these additional agents. NAC downsizes tumors, allowing patients thought to require mastectomy to become BCT candidates. In a meta-analysis that included 11 studies and approximately 5000 patients, Mieog et al reported a 17% decrease in the mastectomy rate for patients receiving NAC.14 This underestimates the effect of NAC on converting patients to BCT candidates, as most trials lacked a requirement to declare which patients were BCT candidates at diagnosis. In NSABP B-18, surgeons indicated whether a patient was a candidate for BCT before initiation of NAC. For those with tumors ≥5 cm, a population generally not considered optimal candidates for BCT, a lumpectomy was proposed in 3% of patients but was successfully performed in 22% after receipt of NAC.15 For women with tumors measuring from 2 to 5 cm, NAC improved the BCT rate by 10%. There have been questions about the risk of local-regional recurrence (LRR) after NAC. In NSABP B-18, when controlling for age and clinical tumor size, there was no difference in ipsilateral breast tumor recurrence rates comparing patients who were initially thought to be lumpectomy candidates with those who became candidates after NAC.15 A subsequent meta-analysis found no difference in the incidence of LRR in patients who received NAC and underwent BCT compared with those who underwent upfront surgery as long as both surgery and RT were performed as part of local-regional therapy.14 In addition, there were no differences in LRR rates for patients receiving NAC versus adjuvant chemotherapy when stratified by surgery type (mastectomy vs BCT). Thus BCT provides excellent local-regional control in appropriately selected patients after NAC. In this particular patient scenario, despite her good response to NAC and candidacy for BCT, the patient opted to undergo bilateral mastectomy, a choice informed by her young age and family history. Although rates of local recurrence for the index lesion are equivalent to those of BCT or mastectomy (as described above), over the last decade, we have witnessed many patients considering bilateral mastectomy for the management of unilateral breast cancer, citing decreased worry and peace of mind as the rationale for this decision. Among patients with breast cancer, contralateral prophylactic mastectomy (CPM) rates have increased from 2% in 1998 to 12.7% in 2012.16 There is also evidence that genetic testing itself leads to increased utilization of CPM. CPM rates are higher among patients who are tested and have negative results than among patients who are untested.17 This is an important area of ongoing investigation, as qualitative studies of women undergoing CPM reveal that women without a pathogenic germline mutation overestimate their risk of contralateral breast cancer by as much as 5 times,18, 19 and there is limited evidence that CPM results in decreased worry or improved cancer-specific distress.20, 21 It is clear that the ultimate decision to undergo CPM varies by patient and surgeon characteristics and is largely driven by patient preference.22-24 Historically, all patients with breast cancer, regardless of clinical node status or sequencing of treatments, underwent ALND. For patients presenting with clinically node-negative disease, it is well documented that the SLN biopsy procedure provides reliable nodal staging with less morbidity than ALND in patients undergoing upfront surgery or NAC.25-28 Among those presenting with clinically node-positive disease (cN1), increased use of targeted therapies in the neoadjuvant setting has resulted in increasing rates of axillary pCR, leading investigators from the American College of Surgeons Oncology Group (ACOSOG) to conduct the Z1071 trial (Clinical Trials.gov identifier NCT00881361)29 to gauge the utility of SLN mapping after NAC. Patients with positive nodes who became clinically node-negative after NAC underwent SLN biopsy with a planned ALND, allowing for determination of the false-negative rate (FNR) of the SLN biopsy procedure. In 663 evaluable patients, the FNR was 12.6%, which exceeded the prespecified threshold of 10%; however, that FNR was comparable to the FNR of 14.2% in the SENTINA trial30 and 12.3% in the Sentinel Node Biopsy After Neoadjuvant Chemotherapy in Biopsy-Proven Node-Positive Breast Cancer (SN-FNAC) trial (ClinicalTrials.gov identifier NCT00909441),31 which were similar studies conducted concurrently in Europe and Canada, respectively. Additional details of these 3 trials are provided in Table 1.29-31 Subsequent analyses of these and other data sets have identified strategies for performing the SLN biopsy procedure that maximize identification rates and minimize the FNR, thus allowing appropriate patients to safely undergo SLN mapping and avoid ALND after NAC. In the ACOSOG Z1071 trial, if a dual-tracer technique with both technetium and blue dye was used, the FNR was 10.8%,29 and, if 3 SLNs were identified, the FNR was 9.1%.29 Similarly, in the SENTINA trial, the FNR was 24.3% if a single SLN was identified but only 7.3% if 3 SLNs were found.30 Investigators in the SN FNAC trial reported an FNR of 8.4% when IHC was routinely used to evaluate SLNs, and any IHC-positive node was considered positive.31 As discussed further below, this is likely critical as even low-volume disease carries prognostic significance. Placing a clip in the suspicious lymph node at the time of fine-needle aspiration or core biopsy before NAC allows for a much lower FNR when conducting SLN biopsy after NAC. In ACOSOG Z1071, when the clipped lymph node was within the SLN specimen, the FNR was 6.8%.32 In a series of 191 patients with clipped biopsy-positive lymph nodes who received NAC and then underwent SLN biopsy with a planned ALND, investigators at The University of Texas MD Anderson Cancer Center reported an FNR of 4.2%. If they removed the clipped node and any additional SLNs, the FNR was 1.4%.33 These experiences suggest that SLN biopsy is an appropriate surgical option after NAC when there is no residual tumor in the axilla clinically or pathologically and when ≥3 nodes are excised or the clipped SLN is negative (Fig. 4). However, ALND remains the standard management of the axilla in the setting of residual disease after NAC, even with low-volume residual nodal disease, micrometastases, and isolated tumor cells (ITCs) because the size of the SLN metastasis after NAC has not been correlated with the likelihood of additional nodal disease. Among 702 patients treated with NAC at Memorial Sloan Kettering Cancer Center, 17% with ITCs in an SLN and 64% with micrometastases in an SLN had additional positive non-SLNs identified at completion ALND,34 identical to the rate (62%) of additional non-SLN node-positive disease in patients with macrometastases in an SLN. There is long-term prognostic significance of even limited volume residual nodal disease after NAC. In a National Cancer Database study of 35,536 patients who had cT1-cT4N0-N1 breast cancer treated with NAC and surgery, those who had ITCs and micrometastases in the axillary lymph nodes had 5-year overall survival (OS) rates of 82.8% and 79.5%, respectively, which were significantly worse than the rate for those who had ypN0 disease (88.9%; P < .001).35 Residual nodal disease is also a marker for increased risk of LRR in the NSABP B18 and NSABP B27 NAC trials.36 Given these risks, complete surgical management of the axilla after NAC in patients with any volume of residual nodal disease is recommended to both identify and quantify the amount of residual nodal disease for local-regional control and subsequent treatment planning. It remains unclear whether axillary RT (AxRT) can substitute for an axillary dissection in women with residual nodal involvement after NAC. The After Mapping of the Axilla: Radiotherapy or Surgery (AMAROS) trial (ClinicalTrials.gov identifier NCT00014612) demonstrated the equivalence of AxRT as an alternative to ALND in women with a positive SLN before systemic therapy. These results cannot be directly extrapolated to patients who received NAC and have residual disease, as their risk of local recurrence is greater because they have had tumor cells identified that essentially are resistant to initial chemotherapy. The role for AxRT as an alternative to ALND in the setting of residual nodal cancer after NAC is being explored in Alliance A011202 (ClinicalTrials.gov identifier NCT01901094). Eligible patients in that trial have received NAC yet remain SLN-positive and are randomized to ALND or AxRT. With residual micrometastasis in the SLN, this patient would be eligible for the trial. Most data on the benefit of RT after mastectomy come from trials of women who did not receive NAC. A meta-analysis conducted by the Early Breast Cancer Trialists' Collaborative Group showed a lower risk of recurrence and better OS with the addition of RT for patients with node-positive cancer after mastectomy.37 Pathologic features, including grade,38 lymphovascular invasion,39 and tumor biologic subtype,40 combined with the extent of nodal involvement,41 modulate the risk of recurrence and thus RT benefit. Among women treated with NAC, these same factors—especially the extent of nodal involvement—are prognostic for LRR. In a study of patients undergoing mastectomy but not radiation after preoperative doxorubicin or paclitaxel chemotherapy at The University of Texas MD Anderson Cancer Center, 5-year LRR was correlated with clinical factors at presentation (clinical tumor [T] classification) as well as pathologic factors at the time of surgery, including the number of involved lymph nodes and the size of residual tumor.42 The most comprehensive analysis of LRR after preoperative systemic therapy comes from the 3088 patients who participated in the NSABP B-18 and B-27 trials,36 as patients undergoing mastectomy in those trials did not receive RT. On multivariate analysis of 1947 patients who underwent mastectomy, predictors of LRR included both clinical factors before surgery (tumor >5 cm: hazard ratio, 1.58; P = .0095; clinically involved nodes: hazard ratio, 1.53; P = .017) and pathological factors at the time of surgery (residual disease in the breast vs complete response in the breast and lymph nodes: hazard ratio, 2.21; P < .001; residual disease in the lymph nodes vs complete response in the breast and lymph nodes: hazard ratio, 4.46; P < .001). Through 10 years of follow-up, there were no local recurrences among patients who underwent mastectomy with node-positive disease at diagnosis who had a pCR in the breast and lymph nodes after NAC. By contrast, patients with persistent nodal involvement despite NAC had a 10-year LRR rate of 17% to 22%, depending on clinical tumor size. The negative impact of residual disease after NAC is most pronounced in patients with TNBC. A retrospective analysis from Memorial Sloan Kettering Cancer Center found that the risk of LRR for women with residual tumor despite NAC, mastectomy, and radiation was 20% in those with TNBC compared with 4% in women with hormone receptor-positive cancers. Based largely on NSABP B18 and B27 data, patients who have a pCR to NAC are predicted to have low rates of LRR despite initial positive nodes and, consequentially, have a minimal benefit from RT. To test this hypothesis prospectively, the NSABP is currently conducting a randomized phase 3 trial (ClinicalTrials.gov identifier NCT01872975) of postmastectomy radiation or no adjuvant RT among patients who undergo mastectomy and receive breast radiation alone or breast and nodal radiation among breast-conservation patients who have negative axillary nodes after preoperative chemotherapy. All patients must have T1-T3 disease at presentation, with biopsy-proven positive axillary nodes. With an accrual goal of 1636, this stands to be the largest and most definitive trial testing the prognostic value of in vivo response to NAC. Until the mature results of this trial are available, radiation is generally considered for patients who have positive axillary nodes at diagnosis, even with a complete response to NAC. The current patient has TNBC with axillary involvement at diagnosis and residual disease in both the breast and axilla after NAC. Her risk of LRR is 20% to 30% over the next 10 years. RT decreases this risk by approximately two-thirds43 and confers a survival benefit. Regional nodal irradiation including the chest wall, supraclavicular, and internal mammary lymph nodes is appropriate in this case (Fig. 4). The risk of recurrence in the axilla after completion dissection is low,44 and radiation to the axilla can be largely avoided to minimize the chance of lymphedema.45 To minimize cardiac exposure, which is a concern even with right-sided irradiation when including the internal mammary lymph nodes, radiation was delivered during maximal inspiration.46 Common side effects of radiation include fatigue, skin erythema and irritation, and possibly desquamation during treatment. With modern treatment planning, the risk of symptomatic radiation pneumonitis47 is <1%, as is the risk of radiation-induced secondary malignancy.48 The patient underwent breast reconstruction with subpectoral tissue expanders placed at the time of mastectomy. These were inflated before radiation treatment planning, although the contralateral side was subsequently deflated to optimize beam geometry. She underwent reinflation of the contralateral side shortly after RT and replacement of the expanders for permanent implants 6 months after RT. Waiting a minimum of 6 months to complete reconstruction has been associated with a lower rate of capsular contracture.49 There are concerns that the radiation dose to the heart and lung could be compromised by immediate reconstruction,50 although modern 3-dimensional treatment planning, including deep inspiration breath holding, minimizes these risks. RT takes a toll on the cosmetic outcome of immediate implant-based reconstructions. In one of the few prospective series studying RT and immediate implant-based reconstruction, only two-thirds of patients had an intact implant with good or excellent cosmetic results at 2 years,51 a much lower percentage than typically seen without RT. A meta-analysis of 20 studies looking at reconstruction success among 2348 patients undergoing implant-based reconstruction and RT showed a 17.6% rate of implant failure and a 37.5% risk of Baker grade 3 (breast mound is firm but appears normal) or 4 (breast mound is hard, painful, and appears abnormal52) capsular contracture. There are few studies of long-term reconstructive success and patient satisfaction. The presence of residual tumor in the breast and/or lymph node puts the patient at high risk of recurrence despite intensive NAC, surgery, and radiation treatment. As with LRR, the extent of residual disease in the breast and axilla, the magnitude of treatment effect, and the tumor subtype are all factors in estimating distant disease-free survival. In contemporary studies of patients who had TNBC with residual nodal involvement after standard NAC, 5-year to 10-year recurrence risks remain between 40% and 60%.5, 53, 54 That persistent risk has prompted interest in studies to determine whether additional therapy would improve the long-term prognosis. The CREATE-X trial (Japanese University Hospital Medical Information Network [UMIN] Clinical Trials Registry number UMIN000000843) randomized women with residual invasive cancer after NAC to receive either no further treatment or additional adjuvant therapy with the oral chemotherapy agent capecitabine.55 Among women with TNBC, capecitabine substantially reduced the risk of recurrence by approximately 15% in absolute terms and improved OS. On the basis of these data, we recommended additional therapy with 6 cycles of capecitabine after the completion of RT (Fig. 4). This adaptive approach to TNBC, including NAC followed by surgery and the development of a postsurgical systemic and local-regional therapy plan informed by the extent of residual cancer, has proven to be an effective strategy for individualizing therapy and optimizing long-term outcomes. It serves as a new paradigm for treatment and for drug development in TNBC. Building on this paradigm, multiple National Cancer Institute-supported cooperative groups have embarked on clinical trials for women with residual TNBC after standard neoadjuvant treatment, including Eastern Cooperative Oncology Group-American College of Radiology Imaging Network (ECOG-ACRIN) trial EA1131 (ClinicalTrials.gov identifier NCT02445391),56 which compares capecitabine versus carboplatin, Southwest Oncology Group trial S1418/NRG BR00 (ClinicalTrials.gov identifier NCT02954874),57 which randomizes patients to receive pembrolizumab or not, and NSABP B-55/BIG 6-13(ClinicalTrials.gov identifier NCT02032823),58 which is open to women with known BRCA1/BRCA2 mutations who are randomized to receive either adjuvant olaparib or placebo. These complex, multidisciplinary recommendations, including discussions on ALND, postmastectomy RT, and additional adjuvant chemotherapy, reflect how the treatment of TNBC is tailored to both the initial clinical stage and the effects of NAC. NAC serves not only as effective systemic therapy with the prospect of enhancing breast preservation but also as a strategy for identifying patients at greater or lesser risk of recurrence and altering their treatment program based on treatment response. The pathological outcomes at the end of NAC are critical for determining the optimal surgical approach (mastectomy vs BCT, SLN vs ALND) and the need for RT (regional nodes, postmastectomy RT) and/or additional systemic treatment. This adaptive decision making, informed by the patient's experience, and particularly the tumor response, provides powerful justification for neoadjuvant therapy as the preferred approach in stage II and III TNBC (Fig. 4). Despite this intensive course of multimodality therapy, the patient has a considerable risk of recurrence in the future, particularly in the first 5 years after diagnosis. There are relatively few late recurrences of TNBC after 5 years.59 She should be seen every 6 months for the first 2 years, and then annually, by a member of her treatment team for a routine history and physical examination. After bilateral mastectomy and reconstruction, there is no role for routine breast or chest imaging. Similarly, there is no role for routine laboratory testing or radiological studies in the absence of concerning symptoms. Changes in the chest wall area, new constitutional symptoms, or other physical examination findings warrant appropriate evaluation. Patients need social and emotional support to make informed decisions. This patient is facing decisions about her postsurgical treatment and longer term follow-up. She also must find a way to fully engage in shared decision making with her providers. For many patients, this could mean having a friend or family member present at critical medical visits where anxiety is likely to be high and could interfere with the ability to listen and understand. If she has no one available, is the hospital able to provide a more experienced patient or other volunteer to accompany her? For patients lacking a strong social network, bringing social work into the team could be extremely helpful. Has she been asked about her personal situation, her values, and her treatment priorities? It is the provider's job, not the patient's, to bring up these subjects. These can touch on sensitive issues but are critically important factors in helping her make the right decisions and they might not surface unless she is asked. Patients have different views on the amount of information they want as well as the amount of treatment they are willing to tolerate. Her treatment team has recommended RT. Are there clinical trials in the adjuvant setting that could be considered? Some patients might be familiar with and interested in clinical trials, whereas others might never have considered them and/or may hold misconceptions. If available, patients should be introduced to clinical trials as a possible treatment strategy, regardless of their age, background, or financial circumstances. The patient is undergoing reconstruction, which will lengthen and possibly complicate her treatment. It is especially important that she understands the possible range of cosmetic and sensory outcomes of each reconstructive option. An informed decision should include the opportunity to view successful and less successful photos of prior patients. The possible rate of complications and their implications for additional surgery and time must be clearly explained. This patient's high risk of breast cancer recurrence places great importance on her posttreatment follow-up. In addition, she is at risk for lymphedema and, while unlikely, has a small risk of long-term cardiac disease. She needs to be fully informed about the importance of follow-up as well as other things she can do to possibly reduce the risk of a recurrence. Programs or clinical trials focused on diet, weight loss, smoking cessation, or mindfulness might be of interest as well. This is a good time to make sure she is aware of the educational, supportive, financial, and lifestyle-related resources available (Table 2). Although helpful resources are likely to exist both at the hospital and in the community, it is not always easy for patients to learn about them. Resources that offer support in navigating this complexity can help to ensure a successful outcome. Patients who are the sole financial support of their family unit might find it helpful to know about financial counseling. Patients who feel socially isolated or would like to talk about the psychological impact of breast cancer should be introduced to individual and group services provided by social work. For patients who would prefer to seek support outside of their cancer center, knowing about local breast cancer organizations would be helpful. Finally, lifestyle/behavior programs that focus on more holistic topics like mindfulness, sleep, intimacy, exercise, or nutrition can go a long way toward improving a patient's ability to tolerate treatment and to transition more smoothly to a new normal. Elizabeth A. Mittendorf reports personal fees from Merck, Exact Sciences (formerly known as Genomic Health), and Roche/Genentech and grants from GlaxoSmithKline, outside the submitted work. Tari A. King reports personal fees from Exact Sciences (formerly known as Genomic Health), outside the submitted work. Jennifer R. Bellon, Harold J. Burstein, and Elizabeth S. Frank had no disclosures to report. Elizabeth A. Mittendorf acknowledges the Rob and Karen Hale Distinguished Chair in Surgical Oncology for support.