Background

Triple negative breast cancer (TNBC) is categorized by a lack in estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor 2 receptor (HER2). TNBC is known to be more aggressive and lethal than other types of breast cancer because of their highly invasive and migratory ability. However, the mechanisms and main contributors of their metastatic ability are still unclear. ETS transcription factors are related with tumorigenesis in many tissues including breast epithelium. Among them, ELF3 (ESX/ESE1) is an epithelial-specific gene that is specifically associated with breast cancer and has been amplified in early breast cancer. The potential role of ELF3 in cytoplasm is presented, but the mechanism of ability to regulate tumor-associated gene expression and breast cell survival for ELF3 are not yet known. Several studies have suggested that the role of EMT is important in the aggressiveness and metastasis of TNBC. It is also known to play an important role in the formation of tumors in most invasive cancer. The main functions of EMT, such as down-regulation of E-cadherin and up-regulation of MMP, are known to be regulated by transcription factors including Snail, Slug and ZEB1.

Methods

For investigation of ELF3 ability in TNBC, we performed a series of assays; western blot, wound healing, invasion, soft-agar colony formation, flow cytometry, anoikis, CAM and immunofluorescence assays.

Results

In this study, we found that ELF3, an ETS transcription factor, was expressed low and high in mesenchymal and epithelial type of TNBC cell lines, respectively. Remarkably, overexpressed ELF3 in the highly invasive TNBC cell lines, MDA-MB-231 and BT549, suppressed EMT by attenuating the expression of several EMT-associated proteins (Vimentin, Slug and MMPs). Moreover, ELF3 overexpression reduced the tumor growth of BT549 in chick embryo chorioallantoic membrane (CAM) model.

Conclusions

Although high ELF3 expression has been known to be associated with tumorigenesis of other breast cancer types, overexpression of ELF3 in TNBC suppressed the metastatic potential of invasive TNBC cells. Our results suggest the important role for ELF3 in metastasis of TNBC.

Legal entity responsible for the study

Ewha Womans University

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

Recently several new chemotherapeutic agents have been developed and indicated for treatment of recurrent breast cancer. One of them, eribulin have improved overall survival (OS) in EMBRACE trials. On the other hand, bevacizumab has improved progression-free survival (PFS) in several clinical studies, but not extended OS in them. And there are few reports estimating these agents’ effects on OS and PFS of advanced or recurrent breast cancer patients in a real world setting.

Methods

Recurrent breast cancer patients who received chemotherapies in Toyama Breast Cancer Research Group (TBCRG) group institutes from January 2013 to March 2015 were reviewed. Kaplan-Meier method was utilized to estimate OS or PFS, and log-rank test was used to compare OS or PFS. Univariate and multivariate analyses were preformed to find significant factor(s) concerning OS.

Results

Of 208 patients who received chemotherapies in the period mentioned, there were 157 patients who received chemotherapies using both or either of eribulin (Eri), and bevacizumab (Bev). Median age of each group using Eri/Bev were 52.4/54.6. Disease status of Eri/Bev were as follows; 68.8%/79.0% ER positive, 18.8%/11.3% HER2 positive. Metastasis of each group using Eri/Bev were observed at; 51.0%/55.2% bones, 7.3%/8.1% CNS, 56.3%/47.2% lung, and 33.3%/49.6% liver. Response rates and PFS from the starting period of Eri/Bev were 21%/75%, and 387days/297days. Multivariate COX regression analysis disclosed negative HER2 status in Eri group, and liver involvement in Bev group influenced the response rates significantly. Earlier administration of Eri shows better survival durations from primary chemotherapies (1470 days), comparison with those of Bev (1076 days). Response rate/disease control rate of pre-line chemotherapies in Eri group were 32%/57%, and not significantly diffierent from 29%/65% in Bev group, but those of post-line chemotherapies were 21%/63% in Eri group and 6%/28% in Bev group.

Conclusions

According to our retrospective observation analyses of group data, eribulin did not reduce the effects of subsequent chemotherapies, and improved OS of advanced or recurrent breast cancer patients. We have to reassess the value of new chemotherapeutic agents continuously.

Disclosure

All authors have declared no conflicts of interest.

Background

In India, breast cancer is the second most common malignant condition among women. It is hereditary in 10% of cases, the majority related to mutations in the BRCA1 and BRCA2 genes. The presence of BRCA germline mutations in breast/ovarian carcinomas has been shown to have prognostic and therapeutic significance. Identification of tumours with BRCA defects has therapeutic and prognostic implications. Our objective was to assess whether immunohistochemical analysis (IHC) for BRCA is an effective method for the detection of BRCA dysfunction in hereditary breast carcinoma or not.

Methods

We have selected 231 patients for BRCA1/2 mutation detection during Aug.2010 to Oct.2015 from the breast cancer patients attended our hospital, NCRI. After taking written consent 4-5ml peripheral blood and/or operated tissue (where possible) were collected from BC patients. BRCA1/2 mutations were identified by ARMS-PCR, DNA sequencing and whole genome sequencing. We performed IHC staining with BRCA1 and BRCA2 antibody to distinguish tumour status between patients according to their indication of BRCA1 or BRCA2 mutation.

Results

Average age of the patients was 45.87±1.57 yrs. BRCA1/2 mutations were identified in 24 (10.38%) patientsthrough above mentioned methods. Tumour samples fromnine BRCA positive cases and 15BRCA negative cases were chosen for investigation of IHC. Out of 9 samples with BRCA mutations, 7were BRCA immunostainingnegative, with absence of nuclear or cytoplasmic staining. On the otherhand, from the 15 patients negative for mutations in both genes, 12 were positive for BRCA1immunostaining with a clear nuclear immunoreactivity in tumour cells. From ROCcurve analysis, it was found that IHC negativity (area- 0.211, CI: 0.011-0.411) isassociated (p = 0.02) with BRCA positivity.

Conclusions

In conclusion, we observed a high specificity for the prediction of BRCA1/2 carriers withimmunohistochemistry using BRCA antibody. Validation of this assay, using a larger sample, will allow using immunohistochemistry to decide which high-risk patients should be screenedfirst for the BRCA mutation gene.

Clinical trial identification

NA

Legal entity responsible for the study

Netaji Suibhas Chandra Bose Cancer Research Institue

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

It is unclear whether perioperative fluoropyrimidine (FP) use impacts the efficacy of capecitabine in advanced breast cancer treatment.

Methods

Medical records of patients with advanced breast cancer who received capecitabine between 2008 and 2016 at National Cancer Center Hospital (Tokyo, Japan) were reviewed. Progression-free survival (PFS), overall survival (OS), tumor response, and adverse events (AEs) were compared between a FP group (prior perioperative FP use) and a non-FP group (no prior FP use). To evaluate the effect of prior perioperative FP use on survival outcomes, hazard ratios (HRs) for PFS and OS were estimated for the FP group compared with the non-FP group.

Results

A total of 289 patients (FP group: n = 106; non-FP group: n = 183) were analyzed. Patient characteristics were similar between the two groups. The median recurrence-free interval (RFI) was 3.94 (range: 0.27-20.11) years in the FP group and 4.24 (range: 0.27-27.07) years in the non-FP group (p = 0.402). The FP group had poorer PFS than the non-FP group (univariate HR: 1.33; 95% confidence interval [CI]: 1.03-1.72, p = 0.028; multivariate HR: 1.33; 95% CI: 1.02-1.73; p = 0.034). However, OS was similar between the groups (univariate HR: 1.16; 95% CI: 0.84-1.62; p = 0.368; multivariate HR: 1.06; 95% CI: 0.74-1.51; p = 0.755). Multivariate HRs for PFS for the FP group with short RFI and long RFI (cutoff: RFI=4 years) separately were 1.56 (95% CI: 1.06-2.28; p = 0.025) and 1.20 (95% CI: 0.84-1.70; p = 0.326), respectively. With different cutoffs (RFI=3, 4, and 5 years), the ranges of adjusted HRs for PFS were 1.32-1.67 with short RFI, and 1.00-1.25 with long RFI. A trend for larger HR for the FP group with short RFI than with long RFI was also seen for OS with different cutoffs of RFI. The response rate (FP group vs. non-FP group) was 21.0% vs. 14.8% (p = 0.306), and the disease control rate was 59.9% vs. 54.5% (p = 0.422). There was no significant difference in AEs between the two groups.

Conclusions

Capecitabine can be used for patients with advanced breast cancer with FP use history, as OS does not correlate with prior FP use. For patients with FP use history, RFI may be a relevant factor for treatment selection.

Legal entity responsible for the study

National Cancer Center Hospital

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

In this study, we observe the patterns initial palliative treatment for premenopausal patients with HR-positive/HER2-negative MBC and determine if non-adherence to clinical guidelines are associated with worse clinical outcomes in terms of progression-free survival (PFS) and overall survival (OS) in the South Korean population.

Methods

A retrospective review was performed for premenopausal patients diagnosed with HR-positive MBC between October 1997 and May 2016 who received palliative systemic treatments at a large tertiary medical center. Survival outcomes were analyzed according to the palliative treatment received prior to disease progression.

Results

The review identified a total of 272 premenopausal patients with HR-positive/HER2-negative MBC. The patients were young (median age, 39 years), as per the premenopausal criteria. Chemotherapy alone was first-line palliative treatment in 78 patients, with endocrine therapy as the initial treatment in 133 patients. In 57 patients, the first line treatment was switched from chemotherapy to endocrine treatment prior to any disease progression. Both progression free survival and overall survival were significantly longer for chemotherapy-endocrine therapy (median PFS 18.2 months and OS 85.2 months) than for chemotherapy-alone (median PFS 12.6 months and OS 45.5 months) or endocrine therapy-alone (median PFS 7.0 months and OS 57.3 months) (all P values < 0.01). In multivariate analysis, chemotherapy-endocrine therapy was an independent predictive value for improved PFS and OS (hazard ratio [HR] 0.33, 95% CI 0.20-0.52, P < 0.001; HR 0.38, 95% CI 0.19-0.73, P = 0.004).

Conclusions

In our study population, chemotherapy alone was not objectively inferior to endocrine therapy as the initial palliative treatment. In addition, chemotherapy followed by endocrine therapy was associated with objective higher response rate than endocrine therapy alone. Further studies should explore the relationship between non-adherent treatment patterns and patient outcomes across the largely premenopausal breast cancer populations across Asian countries.

Legal entity responsible for the study

Prof. Yeon Hee Park

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

There are many reports suggesting that the levels of bone turnover markers (BTM) might be correlated with skeletal-related-events and disease progression in breast cancer patients with bone metastases (BM). We therefore evaluated the correlations OS with the changes in urinary N-telopeptide of type I collagen (u-NTX) in breast cancer patients with BM treated with denosumab (Dmab).

Methods

34 patients were enrolled in this study. All patients received Dmab 120 mg subcutaneously every month and u-NTX was checked at baseline and 1 month after Dmab was administered. We calculated the percentage change of u-NTX level (u-NTX%) [(baseline u-NTX level – 1-month u-NTX level)/baseline u-NTX level] and established a cut-off value by using receiver operating characteristic analysis. OS was defined as the time interval from Dmab administration to the date of death. The patients were divided into two groups by using the cut-off value, and OS was evaluated using the Kaplan-Meier method and analyzed by a log-rank test.

Results

According to the immunohistochemical analysis, estrogen receptor positive and human epidermal growth factor receptor 2 negative (ER+/HER2-) profile was observed in 22 patients, while ER+/HER2+, ER-/HER2+, and ER-/HER2- profiles were observed in 4, 2, and 6 patients, respectively. 2 patients had oligo-BM and 32 patients had multiple-BM. The mean u-NTX% was 0.63 and the established cut-off value was 0.669. The u-NTX% in 19 patients was >0.669 (high u-NTX group) while the remaining 15 patients had u-NTX% under 0.669 (low u-NTX group). OS in the high u-NTX group (n = 19) was significantly longer than in the low u-NTX group (n = 15) (24 months(M) vs 15 months(M), p = 0.005). In 22 ER+/HER2- patients, OS in the high u-NTX group (n = 12) was significantly longer than in the low group (n = 10) (34 M vs 16 M, p = 0.002). Moreover, in 20 patients without other organs metastases, OS in the high u-NTX group (n = 12) was significantly longer than in the low group (n = 8) (24 M vs 15 M, p < 0.001).

Conclusions

High u-NTX% in early phase of treatment could be a promising finding for OS in breast cancer patients with BM treated with Dmab.

Legal entity responsible for the study

Department of Breast Surgery, Yokkaichi Municipal Hospital

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

Differential efficacy of newly registered therapies in subgroups of metastatic breast cancer (mBC) is an important consideration for their subsequent use in clinical practice. In a systematic literature review, we evaluated differences in outcome regarding progression free survival (PFS), time to progression (TTP), overall survival (OS) and visceral versus non-visceral disease. The impact of HER2- and hormone receptor-status was also considered.

Methods

A systematic literature search (6362 hits) in the meta-Database PubMed was performed for the last 20 years. 257 studies (n = 126,291) were included for further analysis. 69 studies had published data for visceral vs non visceral disease including phase III trials. Out of these 69 studies we selected n = 16 studies (n = 13,083) which looked at the endpoints mentioned above. In order to achieve comparability, we extracted the information of hazard ratios (HR), confidence intervals (CI) and times in weeks (if available) for PFS, TTP, OS of the entire study population, which was divided into three groups: HER2-positive, HER2-negative, unknown HER2 status.

Results

No statistically significant difference in treatment response was found in mBC patients looking at HRs and CIs. Relevant, yet not statistically significant differences were found in the specific response of visceral metastases to modern combination therapies, especially in HER2-positive breast cancer: There was a benefit regarding OS using lapatinib combined with trastuzumab or trastuzumab and docetaxel combined with pertuzumab. Additionally, in two chemotherapy trials, there was a numerical difference between therapy response in visceral vs. non-visceral metastases regarding PFS in the unknown HER2 group, and regarding OS in the HER2 negative group.

Conclusions

In the subgroup analyses, we did not find any significant differences in response rates for visceral vs. non-visceral metastasis. There seems to be a beneficial effect of combination therapies regarding OS in visceral disease. At the present time, metastasis localization should not be used as a predictive marker for choice of systemic therapy in mBC.

Legal entity responsible for the study

Tahir Mehmood

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

The standard therapy for primary treatment of HER2-positive metastatic breast cancer (MBC) is combination therapy of pertuzumab (PER), trastuzumab (HER) and docetaxel (DTX). Although the effectiveness of trastuzumab emtansine (T-DM1) after HER treatment has been reported, there are few reports on the effectiveness of T-DM1 for patients treated with PER. We retrospectively investigated the effectiveness of T-DM1 on HER2-positive MBC previously treated with PER.

Methods

Between October 2013 and June 2017, 79 patients with HER2-positive MBC were treated with PER. 44 patients were investigated the subsequent treatment. 34 patients received T-DM1, and 10 patients received treatment other than T-DM1 after PER treatment.

Results

Median treatment line was 3.0 (1-9) vs 4.0 (1-9) in the T-DM1 treatment and other than T-DM1 treatment, respectively. The response rate was CR 0% vs 0%, PR 36.0% vs 25%, SD 32.0% vs 62.5%, PD 32.0% vs 12.5%, respectively. The objective response rate was 36.0% vs 20.0%. The clinical benefit rate was 48.0% vs 50.0%. Median time to treatment failure was 6.6 months vs 2.9 months, respectively. There was a significant difference in median overall survival; median not reached vs 19.6 months (p = 0.04).

Conclusions

OS was significantly better with administration of T-DM1 after PER treatment. Based on the results of this study, it was confirmed that efficacy of T-DM1 in patients with HER2-positive metastatic breast cancer previously treated with PER.

Legal entity responsible for the study

Koshi Matsui

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

The efficacy and tolerability of palbociclib (P) + endocrine therapy (ET) in ABC patients (pts) was previously established; here, we report results in the Asia-Pacific (AP) region (Australia, Japan, Korea, Taiwan).

Methods

In PALOMA-2, postmenopausal women untreated for their HR+/HER2- ABC were randomized 2:1 to P (125 mg/d [3/1 schedule]) + letrozole (L; 2.5 mg/d) or placebo (PB)+L. In PALOMA-3, women of any menopausal status with HR+/HER2- ABC, whose disease had progressed after previous ET, were randomized 2:1 to P + fulvestrant (F; 500 mg) or PB+F.

Results

Of 666 PALOMA-2 pts, 92 (14%) were from the AP region (P+L, 64; PB+L, 28). Baseline characteristics: median age (61 y) younger than overall population; white (20%), Asian (79%); ≥2 disease sites (70%); visceral disease (59%); prior (neo)adjuvant ET (64%); disease-free interval >12 months since prior (neo)adjuvant therapy (58%). Median PFS (mPFS): 22 mo (95% CI, 19-26) for P+L vs 14 mo (7-22) for PB+L (HR, 0.49; 1-sided P=0.007). All-grade treatment-emergent adverse events (TEAEs) (P+L/PB+L) occurred in 100%/96% of pts; the most common AE in the P+L arm was neutropenia (91%). 55% of P+L pts required a dose reduction due to AEs. Of 521 PALOMA-3 pts, 114 (22%) were from the AP region (P+F, 78; PB+F, 36). Baseline characteristics: younger (median age 53 y) and more premenopausal pts (38%) vs overall population; white (23%), Asian (75%); ≥2 disease sites (66%); visceral disease (57%); prior ET (100%); prior chemotherapy for ABC (70%). mPFS: 13 mo (95% CI, 9-16) for P+F vs 6 (4-9) for PB+F (HR, 0.51; 1-sided P=0.002). All-grade TEAEs (P+F/PB+F) occurred in 100%/92% of pts; the most common AE in the P+F arm was neutropenia (95%). 51% of P+F pts required a dose reduction due to AEs.

Conclusions

Because PALOMA-2 only enrolled postmenopausal pts and percentages of premenopausal pts in AP vs other regions was higher, the proportion of AP pts was higher in PALOMA-3 vs -2. P+ET showed clinically meaningful improvement in mPFS vs PB+ET and a tolerable safety profile in AP pts with HR+/HER2- ABC in the 1^st-line and later-lines of therapy, regardless of menopausal status.

Clinical trial identification

Pfizer (NCT01740427; NCT01942135).

Legal entity responsible for the study

Pfizer Inc

Funding

Pfizer Inc.

Disclosure

S-A. Im: Consulting or advisory role for AstraZeneca, Novartis, Hanmi Corp, and Spectrum, N. Masuda: Honoraria from Chugai and AstraZeneca, K. Inoue: Research funding from Pfizer, Lilly, Chugai, Daiichi-Sankyo, Taiho, MSD, Parexel (Puma), S-B. Kim: Research funding from Novartis, Sanofi-Aventis, Kyowa-Kirin Inc, and Dongkook Pharma Co., Ltd. A. Redfern: Advisory board for Pfizer Australia, J. Lombard: Honoraria from Roche and AstraZeneca, D. Lu, K. Puyana Theall, E.G. Gauthier: Employee and owns stock in Pfizer, H. Mukai: Honoraria from AstraZeneca, Eisai, Novartis Pharma and Taiho Pharmaceutical, and received research funds from Chugai Pharmaceutical, Nippon Kayaku, Novartis Pharma, Pfizer Japan, and Sanofi

All other authors have declared no conflicts of interest.

Background

MONALEESASIA is an ongoing Phase Ib dose-escalation/-expansion study evaluating RIB + LET in a subset of Asian pts from Hong Kong and Singapore with HR+, HER2–ABC.

Methods

Postmenopausal pts from Hong Kong and Singapore with HR+, HER2– ABC and no prior systemic therapy for ABC received RIB (400 or 600 mg/day; 3-weeks-on/1-week-off) + LET (2.5 mg/day; continuous). Dose escalation used a Bayesian Logistic Regression Model with overdose control. The primary endpoint was to determine the maximum tolerated dose (MTD)/recommended Phase II dose (RP2D); secondary endpoints included safety and pharmacokinetic (PK) profile. Blood samples for PK non-compartmental analysis were collected on Days 1 and 21 of Cycle 1 from pre-dose to 24 h post-dose.

Results

At data cut-off (Jan 16, 2017), 26 pts were enrolled (RIB 400 mg: n = 6; 600 mg: n = 20). One dose-limiting toxicity was observed at 600 mg (Grade 3 increased alanine transaminase [ALT]); the MTD/RP2D was RIB 600 mg/day (3-weeks-on/1-week-off) + LET 2.5 mg/day (continuous). Grade 3/4 treatment-related adverse events (TRAEs) occurred in 4/6 pts at 400 mg and 17/20 pts at 600 mg during the dose-escalation and -expansion parts. The most common Grade 3/4 TRAEs (≥20% in either cohort; 400 mg vs 600 mg) were decreased neutrophil count (n = 3 vs n = 7), neutropenia (n = 1 vs n = 6), increased ALT (n = 1 vs n = 4), and increased aspartate transaminase (n = 0 vs n = 4). After a single dose (Cycle 1 Day 1) and at steady state (Cycle 1 Day 21), absorption of both RIB and LET was rapid (Table); increases in RIB exposure were dose dependent.

Conclusions

The MTD/RP2D was declared as RIB 600 mg/day (3-weeks-on/1-week-off) + LET 2.5 mg/day (continuous) in Asian pts from Hong Kong and Singapore with HR+, HER2– ABC. Preliminary safety and PK data are consistent with prior observations in non-Asian pts.

Clinical trial identification

CLEE011A2115C/NCT02333370.

Legal entity responsible for the study

Novartis Pharmaceuticals Corporation

Funding

Novartis Pharmaceuticals Corporation

Disclosure

Y-S. Yap: Received honoraria and provided consultancy for Novartis. Y. Ito: Yoshinori Ito reports grants from MSD, AstraZeneca, Novartis, Parexel, Chugai, and Lilly. O. Bornstein: Orna Bornstein is an employee of Novartis Pharmaceuticals Corporation. Y. Han: Yu Han is an employee at Novartis Pharmaceuticals Corporation. T. Samant: Tanay Samant is an employee of Novartis Pharmaceuticals Corporation and owns stocks/shares in Novartis Pharmaceuticals Corporation. X. Liu: Xiaochun Liu is an employee of Novartis Pharmaceuticals Corporation and owns stocks in Novartis Pharmaceuticals Corporation. J. Chiu: Dr. Chiu served on advisory boards for Novartis and Pfizer.Table: 94O

Summary of RIB and LET PK profiles

Background

In Asia, efficient metastatic breast cancer (mBC) diagnosis, treatment and care is hindered by cultural beliefs and the stigmatization of breast cancer (BC). Despite increasing recognition and efforts from policymakers, advocacy groups and the wider healthcare community, there is urgent need for targeted action and stakeholder collaboration to improve patient outcomes.

Methods

A comprehensive analysis of National Cancer Control Plans (NCCPs), policies and programs was conducted in Japan and South Korea, two developed Asian healthcare systems. Policy components were aligned to the BC/mBC patient journey, and evaluated using standardized criteria on adoption of NCCP goals, and BC/mBC policies and programs. Advocacy initiatives were identified in the policy analysis in Japan and South Korea and through an advocacy promising practice implemented in China.

Results

There has been considerable BC/mBC policy development in Asia but gaps persist across all areas of the patient journey. The analysis finds that cultural beliefs act as barriers to diagnosis and treatment, and deter policy development. For instance, BC/mBC stigmatization is not efficiently tackled due to low levels of trained primary care healthcare professionals (HCPs). This deficiency is also reflected in limited patient awareness and disease prevention, inefficient care coordination, disproportionate emphasis on surgery versus treatment, and use of complementary and alternative medicines. Similarly, access to palliative care and rehabilitative support remain important prevalent needs. Advocacy efforts identified in Japan, South Korea and China sought to fill policy gaps. In China, a model aimed at strengthening primary HCPs through a culturally-adapted BC education toolkit showed promising results for replication in other settings.

Conclusions

Engaging with stakeholders across the patient journey is critical to address cultural barriers and unmet needs of BC/mBC patients. Policy initiatives and promising practices in this research exemplify successful multi-stakeholder engagement to inform further advocacy and policy development.

Legal entity responsible for the study

Susan G. Komen; Pfizer, Inc.

Funding

Pfizer, Inc.

Disclosure

F. Cardoso: Consultant: Astellas/Medivation AstraZeneca Celgene Daiichi-Sankyo Eisai GE Oncology Genentech GlaxoSmithKline (GSK) Macrogenics Merck-Sharp Merus BV Mylan Novartis Pfizer Pierre-Fabre Roche Sanofi Teva. M. Thrift-Perry, K. Faircloth: Pfizer, Inc. All other authors have declared no conflicts of interest.