Helmut Friess

Technische Universität München, München, Bavaria, Germany

Are you Helmut Friess?

Claim your profile

Publications (536)2427.78 Total impact

  • W-H Tang, H Friess
    Techniques in Coloproctology 09/2012; · 1.54 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVE:: We have developed a multifactorial histopathological prognostic score (PRSC) for patients with gastric cancer treated with neoadjuvant chemotherapy before surgery for the accurate discrimination of patient subgroups with differing outcomes. BACKGROUND:: For patients with gastric cancer who undergo multimodal treatment, the postoperative staging classifications used for nontreated tumors may not accurately predict patient prognosis. METHODS:: We evaluated 428 gastric carcinoma specimens after a cisplatin-based chemotherapy. The factors for the Union for International Cancer Control/American Joint Committee on Cancer (UICC/AJCC) ypT-category, ypN-category, and histopathological tumor regression were assigned a value from 1 to 3 as follows: ypT0 to 2 = 1 point; ypT3 = 2 points; ypT4 = 3 points; ypN0 = 1 point; ypN1 to 2 = 2 points; ypN3a to 3b = 3 points; less than 10% residual tumor per tumor bed = 1 point; 10% to 50% residual tumor per tumor bed = 2 points; and greater than 50% residual tumor per tumor bed = 3 points. A 3-tiered PRSC based on the sum value was established (group A: 34 points; group B: 5-7 points; group C: 8-9 points) and was found to correlate with patient prognosis. RESULTS:: The PRSC showed a clear discrimination of 3 significantly different prognostic groups (group A: 76 patients; group B: 210 patients; group C: 142 patients; P < 0.001). In multivariate analyses, including the completeness of resection, tumor diameter, lymphatic vessel invasion, tumor grading, and Lauren classification, the PRSC was the only independent prognostic factor for overall survival (hazard ratio [HR] = 2.03; 95% confidence intervals [CI], 1.49-2.78; P < 0.001). It was slightly superior to the UICC/AJCC staging system (HR = 1.66; 95% CI, 1.20-2.27; P = 0.002) when analyzed with tumor regression as an additional independent factor (HR = 1.27; 95% CI, 1.01-1.62; P = 0.044) included in the analysis. CONCLUSIONS:: The proposed PRSC reveals the most accurate prediction of survival for patients with gastric carcinoma after neoadjuvant chemotherapy followed by surgery. The PRSC clearly identifies 3 subgroups with different prognoses and may be helpful for therapeutic decisions.
    Annals of surgery 09/2012; · 7.90 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Introduction: The glial-cell-line-derived neurotrophic factor (GDNF) family member neurturin (NRTN) and its receptor GFRα2 play a deciding role in the normal development of pancreatic parasympathetic innervation. In this study, we aimed at investigating the role of NRTN/GFRα2 axis in pancreatic neuropathy in human chronic pancreatitis (CP). Methods: Expression of NRTN/GFRα2 was compared between normal human pancreas (NP) and CP tissues via immunohistochemistry, immunoblotting and QRT-PCR and correlated to abdominal pain sensation. To elucidate the impact of NRTN in pancreatic neuroplasticity, neuronal phenotype and glial density were quantified via an in vitro neuroplasticity assay in dissociated newborn rat dorsal root ganglia (DRG) cultured 1) in CP tissue extracts depleted from NRTN, 2) in NP, 3) in untreated CP tissue extracts and 4) CP extracts in which nerve growth factor (NGF), glial-cell-derived-neurotrophic factor (GDNF) or transforming growth factor-β1 (TGFβ1) was depleted Results: NRTN and GFRα2 were highly upregulated in CP, especially in intrapancreatic nerves and the extracellular matrix. CP tissue demonstrated increased amounts of mature multimeric NRTN and elevated levels of GFRα2. The noticeable neurotrophic effect of CP tissue extracts upon DRG neurons was diminished upon blockade of NRTN from these extracts, However, blockade of NRTN from CP extracts did not influence the density of DRG glia cells. Conclusion: The NRTN/GFRα2 axis is activated during the course of CP and represents a major key player in the reactive neural alterations in CP. This is the first study to provide functional evidence for the contribution of neurotrophic factors to neuroplasticity in CP.
    AJP Gastrointestinal and Liver Physiology 09/2012; · 3.65 Impact Factor
  • Source
    Martin Loos, Helmut Friess
    [Show abstract] [Hide abstract]
    ABSTRACT: Advances in surgical technique and better perioperative management have significantly improved patient outcomes after liver surgery. Even major hepatectomy can be performed safely with low morbidity and mortality. Post-resection liver failure is among the most feared complications after extended hepatectomy. In order to increase the future liver remnant (FLR) and to expand the pool of candidates for surgical resection, Schnitzbauer et al recently presented a new 2-stage surgical approach which combines right portal vein ligation (rPVL) with in situ splitting (ISS) of the liver parenchyma. In comparison to other current strategies, such as interventional portal vein embolization, hypertrophy of the FLR was more pronounced (median volume increase = 74%; range: 21%-192%) and more rapid (after a median of 9 d; range: 5-28 d) after rPVL and ISS. In this commentary, we discuss the technical aspects and clinical impact of rPVL combined with ISS. Based on the reported data, this new 2-stage therapeutic approach represents a promising new strategy for patients with locally advanced liver disease, previously regarded as marginally resectable or even unresectable, potentially enabling curative resection. However, morbidity is significant and mortality not negligible.
    World journal of gastrointestinal surgery. 07/2012; 4(7):163-5.
  • Journal of Clinical Oncology 07/2012; 30(23):e209-12. · 18.04 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The small actin-binding protein destrin is one of the key regulators involved in remodeling of the actin cytoskeleton, a process crucial for cytokinesis, cell migration and polarized cell growth as well as for cancer cell migration and invasion. A novel ex vivo nerve invasion model mirroring perineural cancer cell invasion as a key feature of pancreatic ductal adenocarcinoma has been previously established. Using this model, highly nerve-invasive clones of human pancreatic cancer cell lines have been obtained. Genome-wide transcriptional analyses of these cells revealed up-regulation of destrin in highly versus lowly nerve-invasive pancreatic cancer cells. Increased expression of destrin in these nerve-invasive cells was validated using quantitative RT-PCR and immunoblotting; concomitant changes in cell morphology were demonstrated using immunofluorescence analysis. Silencing of destrin by two specific siRNA oligonucleotides in Panc-1 pancreatic cancer cells decreased invasiveness and migration, and reduced proliferation of these cells. Destrin is upregulated in nerve-invasive pancreatic cancer cells and its expression might be related to perineural invasiveness.
    Pancreatology 07/2012; 12(4):350-7. · 2.04 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Pancreatic ductal adenocarcinoma (PDAC) is one of the five most lethal malignancies worldwide and survival has not improved substantially in the past 30 years. Desmoplasia (abundant fibrotic stroma) is a typical feature of PDAC in humans, and stromal activation commonly starts around precancerous lesions. It is becoming clear that this stromal tissue is not a bystander in disease progression. Cancer-stroma interactions effect tumorigenesis, angiogenesis, therapy resistance and possibly the metastatic spread of tumour cells. Therefore, targeting the tumour stroma, in combination with chemotherapy, is a promising new option for the treatment of PDAC. In this Review, we focus on four issues. First, how can stromal activity be used to detect early steps of pancreatic carcinogenesis? Second, what is the effect of perpetual pancreatic stellate cell activity on angiogenesis and tissue perfusion? Third, what are the (experimental) antifibrotic therapy options in PDAC? Fourth, what lessons can be learned from Langton's Ant (a simple mathematical model) regarding the unpredictability of genetically engineered mouse models?
    Nature Reviews Gastroenterology &#38 Hepatology 06/2012; 9(8):454-67. · 10.43 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background/Aims: Measurement of serum lactate remains the most frequently applied laboratory investigation to diagnose acute mesenteric (intestinal) ischemia. The present review aims at critically questioning the widespread measurement of serum lactate to diagnose acute mesenteric ischemia in clinical practice and at drawing attention to more novel markers of intestinal ischemia. Methods: An electronic search of multiple databases was performed with the key words 'lactate', 'marker', 'mesenteric', 'intestinal' and 'ischemia' to detect all relevant studies. Additionally, the references of published articles were also reviewed. Results: Serum lactate is an unspecific marker of tissue hypoperfusion and undergoes significant elevation only after advanced mesenteric damage. While L-lactate is the routinely measured stereoisomer of lactate, the other stereoisomer, D-lactate, has been shown to bear a somewhat higher specificity, which is still not comparable to the extremely specific nature of ischemia markers from other organs (e.g. cardiac ischemia). Larger studies are currently lacking to reliably advocate the routine clinical usage of novel markers like mucosal damage markers such as intestinal fatty acid-binding protein. Conclusion: Based on current evidence, the level of no single serum marker, including serum lactate, is elevated early and specifically enough in the serum to diagnose acute mesenteric ischemia.
    Digestive surgery 06/2012; 29(3):226-35. · 1.37 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: BACKGROUND: The optimal strategy for abdominal wall closure has been an issue of ongoing debate. Available studies do not specifically enroll patients who undergo emergency laparotomy and thus do not consider the distinct biological characteristics of these patients. The present randomized controlled trial evaluates the efficacy and safety of two commonly applied abdominal wall closure strategies in patients undergoing primary emergency midline laparotomy. Methods/design The CONTINT trial is a multicenter, open label, randomized controlled trial with a twogroup parallel design. Patients undergoing a primary emergency midline laparotomy are enrolled in the trial. The two most commonly applied strategies of abdominal wall closure after midline laparotomy are compared: the continuous, all-layer suture technique using slowly absorbable monofilament material (two Monoplus(R) loops) and the interrupted suture technique using rapidly absorbable braided material (Vicryl(R) sutures). The primary endpoint within the CONTINT trial is an incisional hernia within 12 months or a burst abdomen within 30 days after surgery. As reliable data on this primary endpoint is not available for patients undergoing emergency surgery, an adaptive interim analysis will be conducted after the inclusion of 80 patients, allowing early termination of the trial if necessary or modification of design characteristics such as recalculation of sample size. DISCUSSION: This is a randomized controlled multicenter trial with a two-group parallel design to assess the efficacy and safety of two commonly applied abdominal wall closure strategies in patients undergoing primary emergency midline laparotomy. Trial registration NCT00544583.
    Trials 05/2012; 13(1):72. · 2.21 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: BACKGROUND: Postoperative surgical site infections cause substantial morbidity, prolonged hospitalization, costs and even mortality and remain one of the most frequent surgical complications. Approximately 14% to 30% of all patients undergoing elective open abdominal surgery are affected and methods to reduce surgical site infection rates warrant further investigation and evaluation in randomized controlled trials. METHODS: To investigate whether the application of a circular plastic wound protector reduces the rate of surgical site infections in general and visceral surgical patients that undergo midline or transverse laparotomy by 50%. BaFO is a randomized, controlled, patient-blinded and observer-blinded multicenter clinical trial with two parallel surgical groups. The primary outcome measure will be the rate of surgical site infections within 45 days postoperative assessed according to the definition of the Center for Disease Control. Statistical analysis of the primary endpoint will be based on the intention-to-treat population. The global level of significance is set at 5% (2 sided) and sample size (n = 258 per group) is determined to assure a power of 80% with a planned interim analysis for the primary endpoint after the inclusion of 340 patients. DISCUSSION: The BaFO trial will explore if the rate of surgical site infections can be reduced by a single, simple, inexpensive intervention in patients undergoing open elective abdominal surgery. Its pragmatic design guarantees high external validity and clinical relevance. Trial registration http://www.clinicaltrials.gov NCT01181206. Date of registration: 11 August 2010; date of first patient randomized: 8 September 2010.
    Trials 05/2012; 13(1):57. · 2.21 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Predicting the clinical outcome of cancer patients based on the expression of marker genes in their tumors has received increasing interest in the past decade. Accurate predictors of outcome and response to therapy could be used to personalize and thereby improve therapy. However, state of the art methods used so far often found marker genes with limited prediction accuracy, limited reproducibility, and unclear biological relevance. To address this problem, we developed a novel computational approach to identify genes prognostic for outcome that couples gene expression measurements from primary tumor samples with a network of known relationships between the genes. Our approach ranks genes according to their prognostic relevance using both expression and network information in a manner similar to Google's PageRank. We applied this method to gene expression profiles which we obtained from 30 patients with pancreatic cancer, and identified seven candidate marker genes prognostic for outcome. Compared to genes found with state of the art methods, such as Pearson correlation of gene expression with survival time, we improve the prediction accuracy by up to 7%. Accuracies were assessed using support vector machine classifiers and Monte Carlo cross-validation. We then validated the prognostic value of our seven candidate markers using immunohistochemistry on an independent set of 412 pancreatic cancer samples. Notably, signatures derived from our candidate markers were independently predictive of outcome and superior to established clinical prognostic factors such as grade, tumor size, and nodal status. As the amount of genomic data of individual tumors grows rapidly, our algorithm meets the need for powerful computational approaches that are key to exploit these data for personalized cancer therapies in clinical practice.
    PLoS Computational Biology 05/2012; 8(5):e1002511. · 4.87 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Predicting the clinical outcome of cancer patients based on the expression of marker genes in their tumors has received increasing interest in the past decade. Accurate predictors of outcome and response to therapy could be used to personalize and thereby improve therapy. However, state of the art methods used so far often found marker genes with limited prediction accuracy, limited reproducibility, and unclear biological relevance. To address this problem, we developed a novel computational approach to identify genes prognostic for outcome that couples gene expression measurements from primary tumor samples with a network of known relationships between the genes. Our approach ranks genes according to their prognostic relevance using both expression and network information in a manner similar to Google's PageRank. We applied this method to gene expression profiles which we obtained from 30 patients with pancreatic cancer, and identified seven candidate marker genes prognostic for outcome. Compared to genes found with state of the art methods, such as Pearson correlation of gene expression with survival time, we improve the prediction accuracy by up to 7%. Accuracies were assessed using support vector machine classifiers and Monte Carlo cross-validation. We then validated the prognostic value of our seven candidate markers using immunohistochemistry on an independent set of 412 pancreatic cancer samples. Notably, signatures derived from our candidate markers were independently predictive of outcome and superior to established clinical prognostic factors such as grade, tumor size, and nodal status. As the amount of genomic data of individual tumors grows rapidly, our algorithm meets the need for powerful computational approaches that are key to exploit these data for personalized cancer therapies in clinical practice.
    PLoS Computational Biology 05/2012; 8(5):e1002511. · 4.87 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We investigated the prevalence of HIT II in liver transplant recipients and analysed associated factors. In recipients with clinically suspected HIT II in the 4Ts pretest clinical scoring system HIPA-assay was performed. Next, 37 clinical variables were analysed retrospectively for their association with HIT II. Factors significantly correlated to our findings in univariate analysis were included in a multivariate model and binary logistic regression analysis. Among 46 recipients 21 patients were suspicious in the 4Ts pretest and 14 of them (30.4%) were diagnosed HIT-antibody positive. Patient's age (P = 0.001), postoperative dialysis (P = 0.028), and postoperative hospital stay (P = 0.035) were significantly associated with development of HIT-antibodies in univariate analysis. Postoperative dialysis and postoperative hospital stay turned out as epiphenomena of patient's age, the only independent predictor (P = 0.021). Using multiple χ(2) -testing, a cut-off could be calculated, assigning patients younger than 59 years to a low risk group and patients of 59 years and older to a high risk group. High incidence of peri-operative HIT II seroconversion in liver transplant recipients is not associated with factors known to induce thrombocyte activation, like blood products or cell-saver. Only patients' age was identified as independent predictor.
    Transplant International 04/2012; 25(7):739-47. · 3.16 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: INTRODUCTION: The chemokine fractalkine/CX3CL1 and its highly selective receptor CX3CR1 mediate critical physiological events during inflammatory responses. The fractalkine/CX3CR1 axis has been shown to play a key role in the pathogenesis and the progression of a large number of diseases in which imbalance of the immune response is frequently seen. Since our last review published in early 2010, the fractalkine/CX3CR1 axis has gained vast attention as a potential therapeutic target in the scientific community, which can be clearly seen in the large number of studies that have been published on this issue since then. AREAS COVERED: A Medline/PubMed search was performed to detect all recently published studies on the role of the fractalkine/CX3CR1 axis as a therapeutic target in a wide range of clinical diseases. EXPERT OPINION: Recently published studies further underline the high potential of the fractalkine/CX3CR1 axis as a major target for future treatment of pain, inflammation and cancer. However, no clinical trials on novel therapeutics targeting fractalkine or CX3CR1 have been initiated so far, so that the fractalkine/CX3CR1 axis does still not find application in daily clinical practice.
    Expert Opinion on Therapeutic Targets 04/2012; 16(6):613-8. · 4.90 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Following the first successful gastric resection for gastric cancer by Theodor Billroth in 1881 surgery has made tremendous progress leading to improved surgical mortality and morbidity. However, while treatment of early gastric cancer is frequently curative, 5-year survival rates for advanced gastric cancer remain dismal despite the application of perioperative multimodal treatment concepts. In this article we will outline key clinical trials that have lead to an improvement in treatment of gastric cancer patients with specific emphasis on the last 20 years. We will then outline recent concepts and key clinical trials that are currently being conducted in the field. Finally we will outline open questions that remain to be elucidated in the future. J. Surg. Oncol © 2012 Wiley Periodicals, Inc.
    Journal of Surgical Oncology 04/2012; · 2.64 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We have identified syndecan-2 as a protein potentially involved in perineural invasion of pancreatic adenocarcinoma (PDAC) cells. Syndecan-2 (SDC-2) expression was analyzed in human normal pancreas, chronic pancreatitis and PDAC tissues. Functional in vitro assays were carried out to determine its role in invasion, migration and signaling. SDC-2 was expressed in the majority of the tested pancreatic cancer cell lines while it was upregulated in nerve-invasive PDAC cell clones. There were 2 distinct expression patterns of SDC-2 in PDAC tissue samples: SDC-2 positivity in the cancer cell cytoplasm and a peritumoral expression. Though SDC-2 silencing (using specific siRNA oligonucleotides) did not affect anchorage-dependent growth, it significantly reduced cell motility and invasiveness in the pancreatic cancer cell lines T3M4 and Su8686. On the transcriptional level, migration-and invasion-associated genes were down-regulated following SDC-2 RNAi. Furthermore, SDC-2 silencing reduced K-ras activity, phosphorylation of Src and--further downstream--phosphorylation of ERK2 while levels of the putative SDC-2 signal transducer p120GAP remained unaltered. SDC-2 is a novel (perineural) invasion-associated gene in PDAC which cooperates with K-ras to induce a more invasive phenotype.
    Molecular Cancer 04/2012; 11:19. · 5.13 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This review outlines the important multimodal treatment issues associated with locally advanced rectal cancer. Changes to chemotherapy and radiation schema, as well as modern surgical approaches, have led to a revolution in the management of this disease but the morbidity and mortality remains high. Adequate treatment is dependent on precise preoperative staging modalities. Advances in staging via endorectal ultrasound, computed tomography, MRI and PET have improved pretreatment triage and management. Important prognostic factors and their impact for this disease are under investigation. Here we discuss the different treatment options including modern tumor-related surgical approaches, neoadjuvant as well as adjuvant therapies. Further clinical progress will largely depend on the broader implementation of multidisciplinary treatment strategies following the principles of evidence-based medicine.
    Expert Review of Anti-infective Therapy 04/2012; 12(4):481-94. · 2.07 Impact Factor
  • Source
    03/2012; , ISBN: 978-953-51-0394-3
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Regional lymph node metastasis negatively affects prognosis in colon cancer patients. The molecular processes leading to regional lymph node metastasis are only partially understood and proteomic markers for metastasis are still scarce. Therefore, a tissue-based proteomic approach was undertaken for identifying proteins associated with regional lymph node metastasis. Two complementary tissue-based proteomic methods have been employed. MALDI imaging was used for identifying small proteins (≤25 kDa) in situ and label-free quantitative proteomics was used for identifying larger proteins. A tissue cohort comprising primary colon tumours without metastasis (UICC II, pN0, n = 21) and with lymph node metastasis (UICC III, pN2, n = 33) was analysed. Subsequent validation of identified proteins was done by immunohistochemical staining on an independent tissue cohort consisting of primary colon tumour specimens (n = 168). MALDI imaging yielded ten discriminating m/z species, and label-free quantitative proteomics 28 proteins. Two MALDI imaging-derived candidate proteins (FXYD3 and S100A11) and one from the label-free quantitative proteomics (GSTM3) were validated on the independent tissue cohort. All three markers correlated significantly with regional lymph node metastasis: FXYD3 (p = 0.0110), S100A11 (p = 0.0071), and GSTM3 (p = 0.0173). FXYD3 and S100A11 were more highly expressed in UICC II patient tumour tissues. GSTM3 was more highly expressed in UICC III patient tumour tissues. By our tissue-based proteomic approach, we could identify a large panel of proteins which are associated with regional lymph node metastasis and which have not been described so far. Here we show that novel markers for regional lymph metastasis can be identified by MALDI imaging or label-free quantitative proteomics and subsequently validated on an independent tissue cohort. Copyright © 2012 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.
    The Journal of Pathology 03/2012; · 7.59 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In clinical diagnostics, it is of outmost importance to correctly identify the source of a metastatic tumor, especially if no apparent primary tumor is present. Tissue-based proteomics might allow correct tumor classification. As a result, we performed MALDI imaging to generate proteomic signatures for different tumors. These signatures were used to classify common cancer types. At first, a cohort comprised of tissue samples from six adenocarcinoma entities located at different organ sites (esophagus, breast, colon, liver, stomach, thyroid gland, n = 171) was classified using two algorithms for a training and test set. For the test set, Support Vector Machine and Random Forest yielded overall accuracies of 82.74 and 81.18%, respectively. Then, colon cancer liver metastasis samples (n = 19) were introduced into the classification. The liver metastasis samples could be discriminated with high accuracy from primary tumors of colon cancer and hepatocellular carcinoma. Additionally, colon cancer liver metastasis samples could be successfully classified by using colon cancer primary tumor samples for the training of the classifier. These findings demonstrate that MALDI imaging-derived proteomic classifiers can discriminate between different tumor types at different organ sites and in the same site.
    Journal of Proteome Research 03/2012; 11(3):1996-2003. · 5.06 Impact Factor

Publication Stats

11k Citations
2,427.78 Total Impact Points

Institutions

  • 2007–2014
    • Technische Universität München
      • Clinic and Polyclinic for Surgery
      München, Bavaria, Germany
    • Otto-von-Guericke-Universität Magdeburg
      Magdeburg, Saxony-Anhalt, Germany
  • 2013
    • Leiden University Medical Centre
      Leyden, South Holland, Netherlands
  • 2012
    • Helmholtz Zentrum München
      • Institut für Pathologie
      München, Bavaria, Germany
    • Massachusetts General Hospital
      Boston, Massachusetts, United States
  • 2009–2012
    • Deutsches Herzzentrum München
      München, Bavaria, Germany
  • 2002–2012
    • Southeast University (China)
      Nan-ching-hsü, Jiangxi Sheng, China
  • 2010–2011
    • Heinrich-Heine-Universität Düsseldorf
      Düsseldorf, North Rhine-Westphalia, Germany
  • 2002–2011
    • Universität Heidelberg
      • Institute of Pathology (Mannheim)
      Heidelberg, Baden-Wuerttemberg, Germany
    • Heidelberg University
      Tiffin, Ohio, United States
  • 2002–2010
    • University of Liverpool
      • Cancer Research Centre
      Liverpool, ENG, United Kingdom
  • 2007–2009
    • Beth Israel Deaconess Medical Center
      • Center for Vascular Biology Research
      Boston, MA, United States
  • 2006–2009
    • German Cancer Research Center
      • Division of Functional Genome Analysis
      Heidelberg, Baden-Wuerttemberg, Germany
  • 2004–2007
    • Geisel School of Medicine at Dartmouth
      • Department of Medicine
      Hanover, New Hampshire, United States
  • 2003–2007
    • University of California, Los Angeles
      • Department of Surgery
      Los Angeles, CA, United States
    • University of Münster
      Muenster, North Rhine-Westphalia, Germany
    • Peking Union Medical College Hospital
      • Department of General Surgery
      Beijing, Beijing Shi, China
  • 2002–2005
    • Northwestern University
      • Department of Surgery
      Evanston, IL, United States
  • 1994–2003
    • University of California, Irvine
      • • Department of Medicine
      • • Division of Endocrinology
      Irvine, CA, United States
  • 1995–2002
    • Universität Bern
      • Visceral and Transplantation Surgery Research Group
      Bern, BE, Switzerland
  • 2001
    • Inselspital, Universitätsspital Bern
      • Department of Visceral Surgery and Medicine
      Bern, BE, Switzerland
    • General Hospital of Shenyang Military Region
      Feng-t’ien, Liaoning, China
  • 1991–1996
    • Universität Ulm
      • Institute of General Medicine
      Ulm, Baden-Württemberg, Germany