eTable 1. Dermatology/Skin/Nail Assessment (from NCI Common Toxicity Criteria for Adverse Events, version 3.0, with modifications)
eTable 2. Expression of Candidate Proteins at Baseline and Week 2 and Change in Expression of Candidate Proteins between Baseline and Week 2 by Patient pCR Status
eTable 3. Genes Significantly Changed in Patient Samples between Before and After Panitumumab Treatment
eFigure 1. Trial Design
eFigure 2. Panitumumab Dose Modification Algorithm for Toxicity
eFigure 3. Disease-free Survival (A) and Overall Survival (B) Estimates for 37 Evaluable Patients
eFigure 4. Differentially Expressed Genes Regulated by Panitumumab
Customize your JAMA Network experience by selecting one or more topics from the list below.
Matsuda N, Wang X, Lim B, et al. Safety and Efficacy of Panitumumab Plus Neoadjuvant Chemotherapy in Patients With Primary HER2-Negative Inflammatory Breast Cancer. JAMA Oncol. 2018;4(9):1207–1213. doi:10.1001/jamaoncol.2018.1436
Is the combination of the anti–epidermal growth factor receptor antibody panitumumab and neoadjuvant chemotherapy safe and effective in patients with primary human epidermal growth factor receptor 2 (HER2)-negative inflammatory breast cancer (IBC)?
In this single-arm, open-label trial, the combination of panitumumab and neoadjuvant chemotherapy produced pathologic complete response rates of 14% in patients with HER2-negative, hormone receptor–positive disease and 42% in patients with triple-negative IBC. Treatment-related hematological and dermatological toxic effects were substantial but transient, and there were no treatment-related deaths.
The combination of panitumumab and neoadjuvant chemotherapy for primary HER2-negative IBC had significant efficacy, particularly in patients with triple-negative IBC.
Combining conventional chemotherapy with targeted therapy has been proposed to improve the pathologic complete response (pCR) rate in patients with inflammatory breast cancer (IBC). Epidermal growth factor receptor (EGFR) expression is an independent predictor of low overall survival in patients with IBC.
To evaluate the safety and efficacy of the anti-EGFR antibody panitumumab plus neoadjuvant chemotherapy in patients with primary human epidermal growth factor receptor 2 (HER2)-negative IBC.
Design, Setting, and Participants
Women with primary HER2-negative IBC were enrolled from 2010 to 2015 and received panitumumab plus neoadjuvant chemotherapy. Median follow-up time was 19.3 months. Tumor tissues collected before and after the first dose of panitumumab were subjected to immunohistochemical staining and RNA sequencing analysis to identify biomarkers predictive of pCR.
Patients received 1 dose of panitumumab (2.5 mg/kg) followed by 4 cycles of panitumumab (2.5 mg/kg), nab-paclitaxel (100 mg/m2), and carboplatin weekly and then 4 cycles of fluorouracil (500 mg/m2), epirubicin (100 mg/m2), and cyclophosphamide (500 mg/m2) every 3 weeks.
Main Outcomes and Measures
The primary end point was pCR rate; the secondary end point was safety. The exploratory objective was to identify biomarkers predictive of pCR.
Forty-seven patients were accrued; 7 were ineligible. The 40 enrolled women had a median age of 57 (range, 23-68) years; 29 (72%) were postmenopausal. Three patients did not complete therapy because of toxic effects (n = 2) or distant metastasis (n = 1). Nineteen patients had triple-negative and 21 had hormone receptor–positive IBC. The pCR and pCR rates were overall, 11 of 40 (28%; 95% CI, 15%-44%); triple-negative IBC, 8 of 19 (42%; 95% CI, 20%-66%); and hormone receptor–positive/HER2-negative IBC, 3 of 21 (14%; 95% CI, 3%-36%). During treatment with panitumumab, nab-paclitaxel, and carboplatin, 10 patients were hospitalized for treatment-related toxic effects, including 5 with neutropenia-related events. There were no treatment-related deaths. The most frequent nonhematologic adverse event was skin rash. Several potential predictors of pCR were identified, including pEGFR expression and COX-2 expression.
Conclusions and Relevance
This combination of panitumumab and chemotherapy showed the highest pCR rate ever reported in triple-negative IBC. A randomized phase 2 study is ongoing to determine the role of panitumumab in patients with triple-negative IBC and to further validate predictive biomarkers.
ClinicalTrials.gov Identifier: NCT01036087
Inflammatory breast cancer (IBC), which accounts for 3% to 5% of all breast cancers, is characterized by aggressive progression and metastasis. Pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) is associated with improved progression-free and overall survival; however, the rate of pCR to NAC is low for IBC (15.2%).1
Evidence from preclinical and clinical studies indicates that epidermal growth factor receptor (EGFR) might be a promising therapeutic target for IBC. EGFR is overexpressed in 18% of breast cancer cases and up to 50% of IBC cases.2-4 Preclinical studies showed that suppression of EGFR signaling in breast cancer controls tumor growth by enhancing apoptosis5,6 and suppressing the cancer stem cell population.7,8 Targeting the EGFR pathway shifted the phenotype of IBC cells from mesenchymal to epithelial and inhibited tumor growth and metastatic progression.9 Furthermore, we have shown that EGFR regulates IBC cells expressing cancer stem cell markers through COX-2 and Nodal has been identified as a potential driver of EGFR/COX-2–mediated cancer stem cell regulation in IBC cells.10 Indeed, in patients with IBC, EGFR expression is independently associated with a high rate of recurrence and shorter survival.2
To our knowledge, before this report, no study had investigated NAC including an anti-EGFR antibody for IBC. To determine the efficacy and safety of such therapy, we conducted a phase 2 trial of NAC with the anti-EGFR monoclonal antibody panitumumab, nab-paclitaxel, and carboplatin (PNC) followed by an anthracycline-containing regimen in patients with primary human epidermal growth factor receptor 2 (HER2)-negative IBC.
This single-arm, open-label phase 2 trial was designed to assess the efficacy and safety of PNC followed by fluorouracil, epirubicin hydrochloride, and cyclophosphamide (FEC) as NAC in patients with operable HER2-negative IBC. The study was approved by the Institutional Review Board of The University of Texas MD Anderson Cancer Center. Written informed consent was obtained from all patients. The trial protocol is available in Supplement 1.
Eligibility criteria included age at least 18 years, histologic confirmation of breast carcinoma, clinical diagnosis of HER2-negative IBC,11 Eastern Cooperative Oncology Group performance status score of 0 or 1, and adequate hematologic, hepatic, renal, and cardiac functions. Patients with distant metastases amenable to radiotherapy and/or surgery based on consultation with our IBC multidisciplinary team were eligible. Pregnant and lactating women were excluded, as were patients with previous radiotherapy or chemotherapy, HER2-positive breast carcinoma, recurrent breast cancer, history of other malignant neoplasm, positive test for HIV infection or acute or chronic active hepatitis B or C virus infection, history of extensive interstitial lung disease, other significant medical or psychiatric condition that would make assessment of toxic effects or efficacy difficult, uncontrolled intercurrent illness, or peripheral neuropathy more than grade 2. Hormone receptor (HR) positivity was defined as positive staining in at least 10% of cells.
The first 17 patients received 1 dose of panitumumab (2.5 mg/kg) on day 1 and, starting 1 week later, 4 cycles of PNC (panitumumab, 2.5 mg/kg; paclitaxel-protein bound [nab-paclitaxel], 5 mg/mL intravenous piggyback 100 mg/m2; and carboplatin intravenous piggyback at a dose of 2.0 area under the concentration time curve) weekly for 3 weeks. As a result of unacceptable hematological toxic effects, the PNC cycle for the remaining 23 patients was changed to 3 weeks of weekly treatment followed by 1 week off to allow bone marrow recovery. During cycle 4, patients received panitumumab weekly for 2 weeks and carboplatin and nab-paclitaxel weekly for 3 weeks. After PNC, all patients received 4 cycles of FEC consisting of fluorouracil (500 mg/m2), epirubicin hydrochloride (100 mg/m2), and cyclophosphamide (500 mg/m2) weekly for 3 weeks (eFigure 1 in Supplement 2). For correlative studies, tumor core needle biopsy was performed before and after the first dose of panitumumab (at the beginning of week 2 of PNC). Radiologists attempted to collect the pre- and postpanitumumab tumor samples from neighboring areas. Biopsy was optional but recommended to all patients, especially during the second half of study accrual. After NAC, all patients underwent modified radical mastectomy with axillary node dissection followed by irradiation of the chest wall and draining lymphatics.
Clinical breast examination and adverse events assessment were performed before each cycle of PNC and FEC.12 Toxic effects were graded according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events, version 3.0. Dose modifications were made according to the organ system with the highest-grade of toxic effects; dose modifications were specified for skin- and nail-related toxic effects (eTable 1 in Supplement 2). When toxic effects occurred, treatment delay was allowed for up to 3 weeks for resolution of acute toxic effects and 4 weeks for resolution of chronic toxic effects (except alopecia, skin changes, and anemia). Dose modifications for panitumumab were as described in eFigure 2 in Supplement 2. Dose modifications for FEC were based on standard practice.
To identify potential biomarkers for pCR, we examined protein and gene expression in tumor core needle biopsy samples obtained before and after the first dose of panitumumab and analyzed the relationship between (1) protein expression before and after first dose of panitumumab and pCR and (2) change in protein and gene expression and pCR.
Immunohistochemical staining of EGFR, pEGFR, E-cadherin, vimentin, COX-2, and Nodal in tumor biopsy samples was performed at the MD Anderson Research Histopathology Facility. The staining was interpreted and recorded by S.K. Protein expression levels were scored by multiplying the staining intensity score (1, weak; 2, moderate; or 3, strong) by the percentage of epithelial tumor cells with positive staining. Individuals performing immunostaining and scoring were blinded to clinical outcome. Matched pairs of tumor biopsy samples obtained before and after first panitumumab dose were identified. The change in protein expression for each pair was calculated as postpanitumumab level of expression minus prepanitumumab level of expression.
RNA was extracted from matched pairs of samples and sequenced on an Illumina HiSeq2000 at the MD Anderson Sequencing and Microarray Facility. Sequencing reads of samples that passed FastQC checks were aligned with tophat2 to the UCSC hg19 reference genome (Illumina iGenomes repository).13 The concordant pair alignment rate of all samples was greater than 80%. Gene-level read counts were summarized using htseq-count.
Pathologic complete response was defined as absence of invasive carcinoma in the breast, axillary lymph nodes, and skin and absence of tumor emboli within the surgical field.14 Patient characteristics examined for possible association with pCR included race, age, menopausal status, clinical N category, TNM stage at presentation, nuclear grade, and primary tumor subtype. Continuous variables were summarized using mean (standard deviation) and median (range). Categorical variables were summarized in frequency tables. χ2 test or Fisher exact test was used to compare categorical variables between patients with and without pCR. Wilcoxon rank sum test was used to compare continuous variables between patients with and without pCR. Disease-free survival was defined as the time from surgery to the first disease recurrence or death. Overall survival was defined as the time from date of surgery to death or last follow-up. The Kaplan-Meier method was used to estimate the distributions of disease-free and overall survival. The log-rank test was used to compare survival distributions between patients with and without pCR. The intent-to-treat population included all patients who received at least 1 dose of PNC. The evaluable population included all patients who underwent NAC plus surgery, postoperative radiotherapy, and, for HR-positive patients, adjuvant hormonal therapy.
Associations between prepanitumumab protein expression and pCR status, protein expression at week 2 and pCR status, and change in protein expression and pCR status were assessed using Wilcoxon rank sum test. Statistical significance was defined as 2-sided P < .05. SAS, version 9.4 (SAS Institute, Inc), was used for data analysis.
For RNA sequencing analysis, to compare expression across samples, the log-transformed gene-level read counts of each sample were first normalized to total library size and then fitted with a generalized linear model with an identity using the R package edgeR.15 A random effect for each participant was used to accommodate dependency between samples from the same participant. Groups were compared using likelihood ratio test adjusted by the Benjamini-Hochberg procedure.
From November 2010 through July 2015, 47 patients were accrued and assessed for eligibility. Seven patients were ineligible and excluded (3 had HER2-positive disease, 2 had metastatic disease, and 2 did not receive financial approval for treatment). Of the 40 patients who started PNC, 3 discontinued treatment early because of adverse events (n = 2) or distant metastasis (n = 1) (Figure). Baseline characteristics of the 40 enrolled patients are presented in Table 1. The median age was 57 years (range, 23-68 years). Twenty-one patients (53%) had HR-positive/HER2-negative IBC, and 19 (47%) had triple-negative (TN)-IBC.
A pCR was achieved by 11 of 40 enrolled patients (28%; 95% CI, 15%-44%), 8 of 19 with TN-IBC (42%; 95% CI, 20%-66%), and 3 of 21 with HR-positive/HER2-negative IBC (14%; 95% CI, 3%-36%). The association between pCR and subtype was significant (χ2P = .05) (Table 1). We observed no significant associations between pCR and age, menopausal status, clinical N category, or nuclear grade (Table 1). A pCR was achieved by 11 of 37 evaluable patients (30%; 95% CI, 16%-47%), 8 of 17 with TN-IBC (47%; 95% CI, 23%-72%), and 3 of 20 with HR-positive/HER2-negative IBC (15%; 95% CI, 3%-38%).
Treatment-related toxic effects are summarized in Table 2. Among the first 17 patients treated with PNC weekly regimen, 9 developed grade 3 or higher neutropenia, of whom 4 received granulocyte–colony-stimulating factor, and 4 developed grade 3 or higher thrombocytopenia, of whom 3 required platelet transfusion. Therefore, the remaining 23 patients were treated with 3 weeks of PNC followed by 1 week off, which had less intense hematological toxic effects compared with the PNC weekly regimen.
Among all 40 patients treated with PNC, 12 (30%) had grade 4 neutropenia, 6 (15%) had grade 3 or 4 thrombocytopenia, and 3 (8%) had grade 3 or 4 hypomagnesemia. The major nonhematological toxic effect was skin toxic effects; 6 patients (15%) had grade 3 skin rash. All patients’ symptoms were reversible with oral and topical antibiotics.
Two patients were unable to complete PNC therapy because of toxic effects. One required hospitalization for grade 3 neutropenia and severe mucositis; the other left the study during cycle 3 of PNC because of neutropenia and thrombocytopenia that delayed reinitiation of treatment. During PNC therapy, 10 patients were hospitalized for treatment-related toxic effects: 3 for neutropenia, 2 for febrile neutropenia, and 1 each for diarrhea, pulmonary embolism, bleeding from the rectum (ischemic colitis), fever without neutropenia, and confusion of unknown origin. One patient developed possible therapy-related myelodysplastic syndrome. There were no treatment-related deaths.
The median follow-up time for enrolled patients was 19.3 months (95% CI, 13.8-29.5 months). Fourteen patients experienced recurrence (distant recurrence in 13; recurrence in the contralateral breast in 1), and 9 died. Of the 9 patients who died, 1 had initial stage IV disease, and 7 had a high burden of disease at diagnosis and quickly developed distant metastasis. The lung was the most common site of recurrence (n = 4), followed by bone (n = 3) and brain (n = 3). The median follow-up time for the 31 survivors was 19.2 months (range, 0.3-52.2 months). Among the 37 evaluable patients, there was no significant difference in disease-free (hazard ratio, 0.66; 95% CI, 0.18-2.40; P = .53) and overall survival (hazard ratio, 0.69; 95% CI, 0.14-3.32; P = .64) between patients with and without pCR (eFigure 3 in Supplement 2).
Comparison of the protein expression of EGFR, pEGFR, E-cadherin, vimentin, COX-2, and Nodal in prepanitumumab tissues between patients with and without a pCR showed that expression of pEGFR (P = .05) and COX-2 (P = .05), but not other proteins, was correlated with pCR outcome (eTable 2 in Supplement 2). Comparison of changes in protein expression between prepanitumumab and postpanitumumab samples in patients with and without a pCR showed no significant correlations with pCR outcome.
RNA sequencing analysis of 13 pairs of matched tumor samples collected before and after first dose of panitumumab identified no genes with significant expression changes after panitumumab treatment. We divided these samples into TN-IBC (8 pairs) and HR-positive/HER2-negative IBC (5 pairs) and analyzed differential gene expression.
In the TN-IBC samples, 2 genes (POU3F3 and EGR1) were significantly downregulated and 4 genes (BBOX1, GLYATL2, MUCL1, and LCN2) were significantly upregulated after panitumumab treatment (eFigure 4A and eTable 3 in Supplement 2). We identified no gene whose change in expression after panitumumab treatment predicted pCR status of patients with TN-IBC.
In the HR-positive/HER2-negative samples, 19 genes were significantly downregulated and 10 genes were significantly upregulated after panitumumab treatment (eTable 3 in Supplement 2). Canonical pathways analysis showed that these genes contribute to metastasis, cellular immune response, inflammation, and retinoid signaling–regulated biological function (eFigure 4B in Supplement 2).
In this single-arm phase 2 study, we evaluated the efficacy and safety of PNC followed by FEC for patients with primary HER2-negative IBC. The intent-to-treat pCR rate was 28% in all evaluable patients, 42% in patients with TN-IBC, and 14% in patients with HR-positive/HER2-negative IBC. These pCR rates compare favorably with rates previously reported for patients with IBC. Costa et al16 reported that the pCR rate in 93 patients with IBC treated with docetaxel (75 mg/m2), doxorubicin (50 mg/m2), and cyclophosphamide (500 mg/m2) was only 8.6%. Our group retrospectively reviewed the outcomes of 527 patients with primary stage III IBC treated with NAC and found that the pCR rate was 15.2% overall and 12.4% in the 139 patients with TN-IBC.1 Most of the patients in this retrospective study received an anthracycline-containing regimen. To our knowledge, the overall pCR rate in the present study, 42%, is the highest reported to date for patients with TN-IBC.
The present study is the first to indicate that an anti-EGFR antibody, panitumumab, may enhance sensitivity to NAC, as indicated by a significantly better pCR rate than the pCR rate in historical series, especially in TN-IBC. In previous reports of EGFR-targeted therapy in other types of breast cancer, the efficacy was moderate. Our findings agree with those of Nabholtz et al,17 who reported that panitumumab plus FEC followed by docetaxel resulted in a pCR rate of 46.8% (95% CI, 32.5%-61.1%) in operable TNBC (n = 47). We acknowledge that the use of nab-paclitaxel in our study may add a layer of complexity to the interpretation of our findings. The GeparSepto (German Breast Group 69) trial showed that compared with traditional paclitaxel, nab-paclitaxel significantly increased the rate of pCR following anthracycline-based chemotherapy: the rate was 38.4% (233 of 606; 95% CI, 34.6%-42.3%) in the nab-paclitaxel group vs 29.0% (174 of 600; 95% CI, 25.4%-32.6%) in the conventional paclitaxel group.18,19 However, whereas our study was limited to patients with IBC, the GeparSepto trial included few patients with IBC. We recognize that the nonrandomized nature of our study limits assessment of the direct contribution of panitumumab to pCR, especially because panitumumab was given in combination with carboplatin and nab-paclitaxel.20,21 However, our study suggests a positive impact of EGFR-targeted therapy for TN-IBC patients, and we are conducting a randomized phase 2 study (NCT02876107) to definitively determine the role of panitumumab.
The initial panitumumab-containing regimen used in our study had significant hematological toxic effects. After the schedule was modified to days 1, 8, and 15 every 28 days, toxic effects was substantially reduced. The most frequent nonhematological toxic effect was rash, which is expected with panitumumab therapy. Although 15% of patients had grade 3 skin toxic effects, symptoms were manageable. We speculate that the grade 4 hematologic and grade 3 nonhematologic adverse events, such as neuropathy, may have been caused by carboplatin. In the GeparSixto trial,20 which assessed an anthracycline-taxane-bevacizumab combination with or without carboplatin, hematological toxic effects, including grade 3 or 4 neutropenia, were significantly more common with carboplatin (65% vs 27%), and the carboplatin dose was decreased.20 Similar rates of hematologic toxic effects, particularly neutropenia, were observed in phase 2 studies of panitumumab in combination with either gemcitabine and carboplatin or paclitaxel and carboplatin in metastatic triple-negative breast cancer.22,23 The fact that 1 patient in our study developed possible therapy-related myelodysplastic syndrome underscores the need for future validation of the hematological toxic effects and optimization of the dose and schedule of chemotherapy in combination with anti-EGFR drugs.
In our comprehensive correlative analysis to determine predictive biomarkers for pCR, neither EGFR expression before panitumumab treatment nor its change after the first dose of panitumumab was correlated with pCR status, in line with previous reports that EGFR expression does not predict responsiveness to cetuximab in colorectal cancer or gefitinib in non–small cell lung cancer.20,24,25 Our findings suggest that expression of pEGFR, an indicator of activated EGFR pathway, and COX-2, a critical molecule in inflammatory response and EGFR-regulated cancer stemlike cell population,10 are potential predictors of response to panitumumab. Our RNA sequencing analysis identified other candidate predictors of pCR to panitumumab-containing NAC in TN-IBC. These candidates have been shown to contribute to the regulation of EGFR tyrosine kinase activity,26 promotion of epithelial-mesenchymal transition, tumor growth, and metastasis,27-29 and regulation of resistance to EGFR-targeted therapy in breast cancer and other cancers.30,31 However, given the small number of patients in our study, molecular predictors of response could not be robustly assessed. Thus, changes in expression of these candidate genes need to be validated in IBC cells, xenografts, and more patient tumor samples collected in our ongoing randomized phase 2 trial (NCT02876107).
Because this was a single-arm phase 2 study, the contribution of panitumumab to pCR could not be definitely determined. Because of the small number of patients, it was difficult to determine the molecular predictors of response via comparison between the pCR and non-pCR groups.
Treatment with PNC followed by FEC as NAC for primary HER2-negative IBC causes substantial but acceptable hematological and dermatological toxic effects and has significant efficacy, particularly in patients with TN-IBC. We are conducting a randomized phase 2 trial of the same carboplatin-containing chemotherapy with and without panitumumab in patients with TN-IBC to determine the efficacy of panitumumab (NCT02876107). Our findings warrant further validation of predictive biomarkers for EGFR-targeted therapy.
Accepted for Publication: March 3, 2018.
Corresponding Author: Naoto T. Ueno, MD, PhD, Morgan Welch Inflammatory Breast Cancer Research Program and Clinic, Section of Translational Breast Cancer Research, Department of Breast Medical Oncology, University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Unit 1354, Houston, TX 77030 (firstname.lastname@example.org).
Published Online: June 7, 2018. doi:10.1001/jamaoncol.2018.1436
Author Contributions: Dr Ueno had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Matsuda, Wang, and Lim contributed equally to this work and are co–first authors.
Study concept and design: Wang, Lim, Babiera, Woodward, Le-Petross, Valero, Ueno.
Acquisition, analysis, or interpretation of data: Matsuda, Wang, Lim, Krishnamurthy, Alvarez, Willey, Parker, Song, Shen, Hu, Wu, Li, Babiera, Murray, Arun, Brewster, Reuben, Stauder, Barnett, Woodward, Le-Petross, Lucci, DeSnyder, Tripathy, Ueno.
Drafting of the manuscript: Matsuda, Wang, Lim, Krishnamurthy, Parker, Hu, Wu, Tripathy, Ueno.
Critical revision of the manuscript for important intellectual content: Wang, Lim, Krishnamurthy, Alvarez, Willey, Song, Shen, Li, Babiera, Murray, Arun, Brewster, Reuben, Stauder, Barnett, Woodward, Le-Petross, Lucci, DeSnyder, Tripathy, Valero, Ueno.
Statistical analysis: Lim, Song, Shen, Hu, Wu, Li, Ueno.
Obtained funding: Wang, Ueno.
Administrative, technical, or material support: Lim, Willey, Reuben, Stauder, Barnett, Le-Petross, Lucci, DeSnyder, Tripathy, Ueno.
Study supervision: Lim, Krishnamurthy, Willey, Shen, Hu, Brewster, Woodward, Ueno.
Conflict of Interest Disclosures: Dr Lim receives research funding from Merck, Genentech, Puma Biotechnology, and Takeda. Ms Parker has stock from Amgen. Dr Arun receives research funding from Abbvie and PharmaMar and had travel expenses covered by AstraZeneca in May 2017. Dr Reuben is a consultant for Angle plc. Dr Stauder receives research funding from EMD Serono and is a member of the speakers bureau and had travel expenses covered by Varian Medical Systems. Dr Barnett was a consultant for and served as an advisory board member for Pfizer. Dr Lucci is a member of the speakers bureau for Genomic Health Inc. Dr DeSnyder receives research funding from Impedimed. Dr Varelo has received honoraria from Amgen. Dr Ueno receives research funding from Amgen, Celgene, Daiichi-Sankyo, Bayer, Genentech, Novartis, the National Institutes of Health, and Breast Cancer Research Foundation. No other disclosures are reported.
Funding/Support: This work was supported by National Institutes of Health grant 1R01CA205043-01A1 (Dr Ueno), Breast Cancer Research Foundation BCRF-17-161 (Dr Ueno), the Morgan Welch Inflammatory Breast Cancer Research Program (Dr Ueno), a State of Texas Rare and Aggressive Breast Cancer Research Program grant (Dr Ueno), Amgen (Dr Ueno), Celgene (Dr Ueno), and National Cancer Institute Cancer Center Support Grants P30CA016672 and CA016672, which support MD Anderson’s Biostatistics Shared Resource and Sequencing and Microarray Facility.
Role of the Funder/Sponsor: The National Institutes of Health and the State of Texas Rare and Aggressive Breast Cancer Research Program had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Amgen and Celgene had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation of the manuscript. Amgen and Celgene reviewed and approved the manuscript and agreed with the decision to submit the manuscript for publication.
Additional Contributions: We thank the patients and their families for participating in this study. Stephanie P. Deming, BA, and Sunita C. Patterson, BA, (Department of Scientific Publications, The University of Texas MD Anderson Cancer Center) provided expert editorial assistance. They did not receive any compensation from a funding sponsor for such contributions.