W Fraser Symmans

Okayama University, Okayama, Okayama, Japan

Are you W Fraser Symmans?

Claim your profile

Publications (374)

  • [Show abstract] [Hide abstract] ABSTRACT: Background: Low-molecular-weight-cyclin E (LMW-E) detected by Western blot, predicts for reduced breast cancer survival, however, it is impractical for clinical use. LMW-E lacks a nuclear localization signal which leads to accumulation in the cytoplasm that can be detected by immunohistochemistry. We tested the hypothesis that cytoplasmic staining of cyclin E can be used as a predictor of poor outcome in different subtypes of breast cancer using patient cohorts with distinct clinical and pathologic features. Methods: We evaluated the subcellular localization of cyclin E in breast cancer specimens from 2,494 patients from 4 different cohorts: 303 from a prospective study and 2,191 from retrospective cohorts (National Cancer Institute [NCI], MD Anderson Cancer Center [MDA] and the United Kingdom [UK]). Median follow-up times were 8.0, 10.1, 13.5, and 5.7 years, respectively. Results: Subcellular localization of cyclin E on immunohistochemistry was associated with full-length (nuclear) and low molecular weight isoforms (cytoplasmic) of cyclin E on Western blot analysis. In multivariable analysis, cytoplasmic cyclin E staining was associated with the greatest risk of recurrence compared with other prognostic factors across all subtypes in three (NCI, MDA and UK) of the cohorts. In the MDA cohort, cytoplasmic cyclin E staining outperformed Ki67 and all other variables as prognostic factors. Conclusion: Cytoplasmic cyclin E, identifies patients with the highest likelihood of recurrence consistently across different patient cohorts and subtypes. These patients may benefit from alternative therapies targeting the oncogenic isoforms of cyclin E.
    Article · Nov 2016 · Clinical Cancer Research
  • Lars Johan Sandberg · Mark W. Clemens · W. F. Symmans · [...] · Steven J. Kronowitz
    [Show abstract] [Hide abstract] ABSTRACT: Background: Molecular profiling using breast cancer subtype has an increasing role in the multidisciplinary care of the breast cancer patient. We sought to determine the role of breast cancer subtyping in breast reconstruction and specifically if breast cancer subtyping can determine the need for postmastectomy radiation therapy (PMRT) and predict recurrence-free survival (RFS) to plan for the timing and technique of breast reconstruction. Methods: We reviewed prospectively collected data from 1931 reconstructed breasts in breast cancer patients who underwent mastectomy between November 1999 and December 2012. Reconstructed breasts were grouped by breast cancer subtype and examined for covariates predictive of RFS and need for PMRT. Results: Of the reconstructed breasts, 753 (39%) were luminal A, 538 (27.9%) luminal B, 224 (11.6%) luminal HER2, 143 (7.4%) HER2 enriched, and 267 (13.8%) TNBC. PMRT was delivered in 69 (48.3%) HER2 enriched patients, 94 (42%) luminal HER2, 200 (37.2%) luminal B, 99 (37.1) TNBC, and 222 (29.5%) luminal A (p < 0.0001). Luminal A cases had better RFS than HER2 enriched cases, and TNBC cases had worse RFS than HER2 enriched cases. Luminal B and Luminal HER2 cases had RFS similar to that for HER2 enriched cases. Luminal A subtype was associated with the best RFS. Subtyping may have improved the breast surgery planning for 33.1% of delayed reconstructions that did not require PMRT, and 37% of immediate reconstructions that did require PMRT. Conclusion: This study is the first publication in the literature to evaluate breast cancer subtype to stratify risk for decision making in breast reconstruction.
    Article · Nov 2016 · Plastic & Reconstructive Surgery
  • Article · Jul 2016 · Cancer Research
  • Article · Jul 2016 · Cancer Research
  • [Show abstract] [Hide abstract] ABSTRACT: BACKGROUND The genetic and clinical heterogeneity of breast cancer makes the identification of effective therapies challenging. We designed I-SPY 2, a phase 2, multicenter, adaptively randomized trial to screen multiple experimental regimens in combination with standard neoadjuvant chemotherapy for breast cancer. The goal is to match experimental regimens with responding cancer subtypes. We report results for veliparib, a poly(ADP-ribose) polymerase (PARP) inhibitor, combined with carboplatin. METHODS In this ongoing trial, women are eligible for participation if they have stage II or III breast cancer with a tumor 2.5 cm or larger in diameter; cancers are categorized into eight biomarker subtypes on the basis of status with regard to human epidermal growth factor receptor 2 (HER2), hormone receptors, and a 70-gene assay. Patients undergo adaptive randomization within each biomarker subtype to receive regimens that have better performance than the standard therapy. Regimens are evaluated within 10 biomarker signatures (i.e., prospectively defined combinations of biomarker subtypes). Veliparib-carboplatin plus standard therapy was considered for HER2-negative tumors and was therefore evaluated in 3 signatures. The primary end point is pathological complete response. Tumor volume changes measured by magnetic resonance imaging during treatment are used to predict whether a patient will have a pathological complete response. Regimens move on from phase 2 if and when they have a high Bayesian predictive probability of success in a subsequent phase 3 neoadjuvant trial within the biomarker signature in which they performed well. RESULTS With regard to triple-negative breast cancer, veliparib-carboplatin had an 88% predicted probability of success in a phase 3 trial. A total of 72 patients were randomly assigned to receive veliparib-carboplatin, and 44 patients were concurrently assigned to receive control therapy; at the completion of chemotherapy, the estimated rates of pathological complete response in the triple-negative population were 51% (95% Bayesian probability interval [PI], 36 to 66%) in the veliparib-carboplatin group versus 26% (95% PI, 9 to 43%) in the control group. The toxicity of veliparib-carboplatin was greater than that of the control. CONCLUSIONS The process used in our trial showed that veliparib-carboplatin added to standard therapy resulted in higher rates of pathological complete response than standard therapy alone specifically in triple-negative breast cancer.
    Article · Jul 2016 · New England Journal of Medicine
  • [Show abstract] [Hide abstract] ABSTRACT: BACKGROUND The heterogeneity of breast cancer makes identifying effective therapies challenging. The I-SPY 2 trial, a multicenter, adaptive phase 2 trial of neoadjuvant therapy for high-risk clinical stage II or III breast cancer, evaluated multiple new agents added to standard chemotherapy to assess the effects on rates of pathological complete response (i.e., absence of residual cancer in the breast or lymph nodes at the time of surgery). METHODS We used adaptive randomization to compare standard neoadjuvant chemotherapy plus the tyrosine kinase inhibitor neratinib with control. Eligible women were categorized according to eight biomarker subtypes on the basis of human epidermal growth factor receptor 2 (HER2) status, hormone-receptor status, and risk according to a 70-gene profile. Neratinib was evaluated against control with regard to 10 biomarker signatures (prospectively defined combinations of subtypes). The primary end point was pathological complete response. Volume changes on serial magnetic resonance imaging were used to assess the likelihood of such a response in each patient. Adaptive assignment to experimental groups within each disease subtype was based on Bayesian probabilities of the superiority of the treatment over control. Enrollment in the experimental group was stopped when the 85% Bayesian predictive probability of success in a confirmatory phase 3 trial of neoadjuvant therapy reached a pre-specified threshold for any biomarker signature ("graduation"). Enrollment was stopped for futility if the probability fell to below 10% for every biomarker signature. RESULTS Neratinib reached the prespecified efficacy threshold with regard to the HER2-positive, hormone-receptor-negative signature. Among patients with HER2-positive, hormone-receptor-negative cancer, the mean estimated rate of pathological complete response was 56% (95% Bayesian probability interval [PI], 37 to 73%) among 115 patients in the neratinib group, as compared with 33% among 78 controls (95% PI, 11 to 54%). The final predictive probability of success in phase 3 testing was 79%. CONCLUSIONS Neratinib added to standard therapy was highly likely to result in higher rates of pathological complete response than standard chemotherapy with trastuzumab among patients with HER2-positive, hormone-receptor-negative breast cancer.
    Article · Jul 2016 · New England Journal of Medicine
  • Veerle Bossuyt · W. Fraser Symmans
    [Show abstract] [Hide abstract] ABSTRACT: The use of neoadjuvant systemic therapy for the treatment of breast cancer patients is increasing. Pathologic response in the form of pathologic complete response (pCR) and grading systems of partial response, such as the residual cancer burden (RCB) system, gives valuable prognostic information for patients and is used as a primary endpoint in clinical trials. The breast cancer and pathology communities are responding with efforts to standardize pathology in patients receiving neoadjuvant chemotherapy. In this review, we summarize the challenges that postneoadjuvant systemic therapy surgical specimens pose and how pathologists and the multidisciplinary team can work together to optimize handling of these specimens. Multidisciplinary communication is essential. A single, standardized approach to macroscopic and microscopic pathologic examination makes it possible to provide reliable response information. This approach employs a map of tissue sections to correlate clinical, gross, microscopic, and imaging findings in order to report the presence of pCR (ypT0 ypN0 and ypT0/is ypN0) versus residual disease, the ypT and ypN stage using the current AJCC/UICC staging system, and the RCB.
    Article · Jul 2016 · Annals of Surgical Oncology
  • [Show abstract] [Hide abstract] ABSTRACT: Importance: We previously described and validated a breast cancer staging system (CPS+EG, clinical-pathologic scoring system incorporating estrogen receptor-negative disease and nuclear grade 3 tumor pathology) for assessing prognosis after neoadjuvant chemotherapy using pretreatment clinical stage, posttreatment pathologic stage, estrogen receptor (ER) status, and grade. Development of the CPS+EG staging system predated routine administration of trastuzumab in patients with ERBB2-positive disease (formerly HER2 or HER2/neu). Objective: To validate the CPS+EG staging system using the new definition of ER positivity (≥1%) and to develop an updated staging system (Neo-Bioscore) that incorporates ERBB2 status into the previously developed CPS+EG. Design, setting, and participants: Retrospective review of data collected prospectively from January 2005 through December 2012 on patients with breast cancer treated with neoadjuvant chemotherapy at The University of Texas MD Anderson Cancer Center. Main outcomes and measure: Prognostic scores were computed using 2 versions of the CPS+EG staging system, one with ER considered positive if it measured 10% or higher, the other with ER considered positive if it measured 1% or higher. Fits of the Cox proportional hazards model for the 2 sets of prognostic scores were compared using the Akaike Information Criterion (AIC). Status of ERBB2 was added to the model, and the likelihood ratio test was used to determine improvement in fit. Results: A total of 2377 patients were included; all were women (median age, 50 years [range, 21-87 years]); ER status was less than 1% in 28.9%, 1% to 9% in 8.3%, and 10% or higher in 62.8%; 591 patients were ERBB2 positive. Median follow-up was 4.2 years (range, 0.5-11.7 years). Five-year disease-specific survival was 89% (95% CI, 87%-90%). Using 1% or higher as the cutoff for ER positivity, 5-year disease-specific survival estimates determined using the CPS+EG stage ranged from 52% to 98%, thereby validating our previous finding that the CPS+EG score facilitates more refined categorization into prognostic subgroups than clinical or final pathologic stage alone. The AIC value for this model was 3333.06, while for a model using 10% or higher as the cutoff for ER positivity, it was 3333.38, indicating that the model fits were nearly identical. The improvement in fit of the model when ERBB2 status was added was highly significant, with 5-year disease-specific survival estimates ranging from 48% to 99% (P < .001). Incorporating ERBB2 into the staging system defined the Neo-Bioscore, which provided improved stratification of patients with respect to prognosis. Conclusions and relevance: The Neo-Bioscore improves our previously validated staging system and allows its application in ERBB2-positive patients. We recommend that treatment response and biologic markers be incorporated into the American Joint Committee on Cancer staging system.
    Article · Mar 2016
  • Source
    Lei Huo · Yan Wang · Yun Gong · [...] · Zhuang Zuo
    [Show abstract] [Hide abstract] ABSTRACT: Inflammatory breast cancer is the most aggressive form of breast cancer. Identifying new biomarkers to be used as therapeutic targets is in urgent need. Messenger RNA expression profiling studies have indicated that inflammatory breast cancer is a transcriptionally heterogeneous disease, and specific molecular targets for inflammatory breast cancer have not been well established. We performed microRNA expression profiling in inflammatory breast cancer in comparison with locally advanced noninflammatory breast cancer in this study. Although many microRNAs were differentially expressed between normal breast tissue and tumor tissue, most of them did not show differential expression between inflammatory and noninflammatory tumor samples. However, by microarray analysis, quantitative reverse transcription PCR, and in situ hybridization, we showed that microRNA-205 expression was decreased not only in tumor compared with normal breast tissue, but also in inflammatory breast cancer compared with noninflammatory breast cancer. Lower expression of microRNA-205 correlated with worse distant metastasis-free survival and overall survival in our cohort. A small-scale immunohistochemistry analysis showed coexistence of decreased microRNA-205 expression and decreased E-cadherin expression in some ductal tumors. MicroRNA-205 may serve as a therapeutic target in advanced breast cancer including inflammatory breast cancer.Modern Pathology advance online publication, 26 February 2016; doi:10.1038/modpathol.2016.38.
    Full-text available · Article · Feb 2016 · Modern Pathology
  • Article · Feb 2016 · Cancer Research
  • Article · Feb 2016 · Cancer Research
  • E Mayer · A DeMichele · P Dubsky · [...] · M Gnant
    Article · Feb 2016 · Cancer Research
  • MC Liu · WF Symmans · C Yau · [...] · A DeMichele
    Article · Feb 2016 · Cancer Research
  • [Show abstract] [Hide abstract] ABSTRACT: Importance The long-term effect of axillary pathologic complete response (pCR) on survival among women with breast cancer treated with primary systemic chemotherapy (PST) is unknown.Objective To assess the long-term effect of axillary pCR on relapse-free survival (RFS) and overall survival (OS) in women with breast cancer with cytologically confirmed axillary lymph node metastases treated with PST.Design, Setting, and Participants We retrospectively analyzed the effect of axillary pCR on 10-year OS and RFS among all women who received a diagnosis of breast cancer stages II to III with cytologically confirmed axillary metastases between 1989 and 2007 who received PST at a large US comprehensive cancer center. Women were stratified by post-PST axillary status, and survival outcomes were estimated and compared according to response in the breast and axilla.Main Outcomes and Measures Outcomes of interest were RFS and OS.Results Of 1600 women treated, median (range) age at diagnisis was 49 (21-86) years. A total of 454 (28.4%) achieved axillary pCR. These patients were more likely to have human epidermal growth factor receptor 2 (HER2)-positive and triple-negative disease (P < .001), pCR in the breast (P < .001), high-grade tumors (P < .001), and lower clinical and pathologic T stage (P = .002). Ten-year OS rates were 84% (95% CI, 79%-88%) and 57% (95% CI, 54%-61%) (P < .001) and 10-year RFS rates 79% (95% CI, 74%-83%) and 50% (95% CI, 46%-53%) (P < .001) for patients with axillary pCR and residual axillary disease, respectively. For patients with axillary pCR, 10-year OS rates were 90% (95% CI, 84%-94%) for those with breast pCR and 72% (95% CI, 61%-80%) for those with residual breast disease (P < .001). For patients with residual axillary disease, 10-year OS rates were 66% (95% CI, 56%-74%) for patients with and 56% (95% CI, 52%-60%) for patients without breast pCR (P = .02). Of patients receiving HER2-targeted therapy for HER2-positive disease, 67.1% (100 of 149) achieved axillary pCR; 10-year OS rates were 92% (95% CI, 84%-96%) and 57% (95% CI, 20%-82%) (P = .003) and 10-year RFS rates 89% (95% CI, 81%-94%) and 44% (95% CI, 18%-68%) (P < .001) for those with axillary pCR and residual axillary disease, respectively.Conclusions and Relevance Axillary pCR was associated with improved 10-year OS and RFS. Patients with axillary and breast pCR after PST had superior long-term survival outcomes. Patients undergoing HER2-targeted therapy for HER2-positive disease had high rates of axillary pCR, and those with axillary pCR had excellent 10-year OS.
    Article · Dec 2015
  • Source
    Christos Hatzis · W Fraser Symmans · Ya Zhang · [...] · Lajos Pusztai
    [Show abstract] [Hide abstract] ABSTRACT: Pathologic complete response (pCR) to neoadjuvant chemotherapy reflects the cytotoxic efficacy of a drug, but patient survival is influenced by many other factors. The purpose of this study was to assess the relationship between increased pCR rate and trial-level survival benefit in triple negative breast cancer (TNBC). We used bootstrap resampling from a neoadjuvant trial to simulate trials with different pCR rates. We used estimates from AdjuvantOnline to simulate trial populations with different baseline prognosis and estimated survival improvements associated with changes in pCR rate. Assuming that survival is similar for patients with pCR regardless of treatment arm, a linear relationship exists between increasing pCR rate and increasing RFS. The slope is equal to the difference in survival between those with pCR and RD which in turn is influenced by (i) the baseline prognosis of the trial population, (ii) interactions between prognostic variables and pCR, and (iii) the efficacy of the post-neoadjuvant therapies. For example, if the pCR rates are 30% and 60% (OR=3.5) and the 10-year RFS of the control arm is 0.74, the trial would require 3550 patients per arm, whereas if the RFS is 0.54 the trial would only require 425 patients per arm to detect significant survival benefit. We provide a framework for understanding the relationship between pCR and overall survival benefit that can help inform the design of neoadjuvant trials aiming to demonstrate improved survival from a regimen that results in higher pCR rate. Copyright © 2015, American Association for Cancer Research.
    Full-text available · Article · Aug 2015 · Clinical Cancer Research
  • Lajos Pusztai · Roman Rouzier · W Fraser Symmans
    [Show abstract] [Hide abstract] ABSTRACT: The article by Rouzier and colleagues, published in the August 15, 2005, issue of Clinical Cancer Research, demonstrated that different molecular subtypes of breast cancer have different degrees of sensitivity to chemotherapy, but the extent of response to neoadjuvant therapy has a different meaning by subtype. Several molecular subtype-specific clinical trials are under way to maximize pathologic complete response rates in triple-negative breast cancer and HER2-positive cancers, and to provide adjuvant treatment options for patients with residual invasive disease. Clin Cancer Res; 21(16); 3575-7. ©2015 AACR.See related article by Rouzier et al., Clin Cancer Res 2005;11(16) Aug 15, 2005;5678-85. ©2015 American Association for Cancer Research.
    Article · Aug 2015 · Clinical Cancer Research
  • Elena Provenzano · Veerle Bossuyt · Giuseppe Viale · [...] · W Fraser Symmans
    [Show abstract] [Hide abstract] ABSTRACT: Neoadjuvant systemic therapy is being used increasingly in the treatment of early-stage breast cancer. Response, in the form of pathological complete response, is a validated and evaluable surrogate end point of survival after neoadjuvant therapy. Thus, pathological complete response has become a primary end point for clinical trials. However, there is a current lack of uniformity in the definition of pathological complete response. A review of standard operating procedures used by 28 major neoadjuvant breast cancer trials and/or 25 sites involved in such trials identified marked variability in specimen handling and histologic reporting. An international working group was convened to develop practical recommendations for the pathologic assessment of residual disease in neoadjuvant clinical trials of breast cancer and information expected from pathology reports. Systematic sampling of areas identified by informed mapping of the specimen and close correlation with radiological findings is preferable to overly exhaustive sampling, and permits taking tissue samples for translational research. Controversial areas are discussed, including measurement of lesion size, reporting of lymphovascular space invasion and the presence of isolated tumor cells in lymph nodes after neoadjuvant therapy, and retesting of markers after treatment. If there has been a pathological complete response, this must be clearly stated, and the presence/absence of residual ductal carcinoma in situ must be described. When there is residual invasive carcinoma, a comment must be made as to the presence/absence of chemotherapy effect in the breast and lymph nodes. The Residual Cancer Burden is the preferred method for quantifying residual disease in neoadjuvant clinical trials in breast cancer; other methods can be included per trial protocols and regional preference. Posttreatment tumor staging using the Tumor-Node-Metastasis system should be included. These recommendations for standardized pathological evaluation and reporting of neoadjuvant breast cancer specimens should improve prognostication for individual patients and allow comparison of treatment outcomes within and across clinical trials.Modern Pathology advance online publication, 24 July 2015; doi:10.1038/modpathol.2015.74.
    Article · Jul 2015 · Modern Pathology
  • Source
    [Show abstract] [Hide abstract] ABSTRACT: To provide recommendations on the appropriate use of breast tumor biomarker assay results to guide decisions on systemic therapy for metastatic breast cancer. A literature search and prospectively defined study selection identified systematic reviews, meta-analyses, randomized controlled trials (RCTs), prospective-retrospective studies, and prospective comparative observational studies published from 2006 through September 2014. The literature search revealed 17 articles that met criteria for further review: 11 studies reporting discordances between primary tumors and metastases in expression of hormone receptors or human epidermal growth factor receptor 2 (HER2), one RCT that addressed the use of a biomarker to decide whether to change or continue a treatment regimen, and five prospective-retrospective studies that evaluated the clinical utility of biomarkers. In patients with accessible metastases, biopsy for confirmation of disease process and retesting of estrogen receptor, progesterone receptor, and HER2 status should be offered, but evidence is lacking to determine whether changing anticancer treatment on the basis of change in receptor status affects clinical outcomes. With discordance of results between primary and metastatic tissues, the Panel consensus is to use preferentially the estrogen receptor, progesterone receptor, and HER2 status of the metastasis to direct therapy if supported by the clinical scenario and patient's goals for care. Carcinoembryonic antigen, cancer antigen 15-3, and cancer antigen 27-29 may be used as adjunctive assessments, but not alone, to contribute to decisions regarding therapy. Recommendations for tumor rebiopsy and use of circulating tumor markers are based on clinical experience and Panel informal consensus in the absence of studies designed to evaluate the clinical utility of the markers. As such, it is also reasonable for clinicians to not use these markers as adjunctive assessments. © 2015 by American Society of Clinical Oncology.
    Full-text available · Article · Jul 2015 · Journal of Clinical Oncology
  • Source
    [Show abstract] [Hide abstract] ABSTRACT: Nucleotide alterations detected by next generation sequencing are not always true biological changes but could represent sequencing errors. Even highly accurate methods can yield substantial error rates when applied to millions of nucleotides. In this study, we examined the reproducibility of nucleotide variant calls in replicate sequencing experiments of the same genomic DNA. We performed targeted sequencing of all known human protein kinase genes (kinome) (~3.2 Mb) using the SOLiD v4 platform. Seventeen breast cancer samples were sequenced in duplicate (n=14) or triplicate (n=3) to assess concordance of all calls and single nucleotide variant (SNV) calls. The concordance rates over the entire sequenced region were >99.99%, while the concordance rates for SNVs were 54.3-75.5%. There was substantial variation in basic sequencing metrics from experiment to experiment. The type of nucleotide substitution and genomic location of the variant had little impact on concordance but concordance increased with coverage level, variant allele count (VAC), variant allele frequency (VAF), variant allele quality and p-value of SNV-call. The most important determinants of concordance were VAC and VAF. Even using the highest stringency of QC metrics the reproducibility of SNV calls was around 80% suggesting that erroneous variant calling can be as high as 20-40% in a single experiment. The sequence data have been deposited into the European Genome-phenome Archive (EGA) with accession number EGAS00001000826.
    Full-text available · Article · Jul 2015 · PLoS ONE
  • V Bossuyt · E Provenzano · W F Symmans · [...] · D Cameron
    [Show abstract] [Hide abstract] ABSTRACT: Neoadjuvant systemic therapy (NAST) provides the unique opportunity to assess response to treatment after months rather than years of follow-up. However, significant variability exists in methods of pathologic assessment of response to NAST, and thus its interpretation for subsequent clinical decisions. Our international multidisciplinary working group was convened by the Breast International Group-North American Breast Cancer Group (BIG-NABCG) collaboration and tasked to recommend practical methods for standardized evaluation of the post-NAST surgical breast cancer specimen for clinical trials that promote accurate and reliable designation of pathologic complete response (pCR) and meaningful characterization of residual disease. Recommendations include multidisciplinary communication; clinical marking of the tumor site (clips); and radiologic, photographic, or pictorial imaging of the sliced specimen, to map the tissue sections and reconcile macroscopic and microscopic findings. The information required to define pCR (ypT0/is ypN0 or ypT0 yp N0), residual ypT and ypN stage using the current AJCC/UICC system, and the Residual Cancer Burden system were recommended for quantification of residual disease in clinical trials. © The Author 2015. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: journals.permissions@oup.com.
    Article · May 2015 · Annals of Oncology

Publication Stats

14k Citations


  • 2010
    • Okayama University
      Okayama, Okayama, Japan
  • 2009
    • University of California, Davis
      • Department of Pathology and Laboratory Medicine
      Davis, California, United States
    • University of Pittsburgh
      Pittsburgh, Pennsylvania, United States
  • 2005-2006
    • University of Houston
      Houston, Texas, United States
    • University of Oxford
      Oxford, England, United Kingdom
  • 2002-2003
    • University of Miami
      كورال غيبلز، فلوريدا, Florida, United States
  • 1998
    • NYU Langone Medical Center
      New York, New York, United States