Shuji Ogino, MD, PhD, MS (Epidemiology)
Professor, Department of Pathology, Harvard Medical School
Professor, Department of Epidemiology, Harvard School Of Public Health
Professor of Pathology, Medical Oncology, Dana-Farber Cancer Institute
Professor, Pathology, Brigham And Women's Hospital
DF/HCC Program AffiliationCancer Epidemiology
Lab WebsiteThe Ogino MPE LaboratoryThe MPE Lab
Research AbstractI am Professor of Pathology, HMS, BWH and DFCI, and soon to be Professor in the Department of Epidemiology at Harvard T.H. Chan School of Public Health (SPH), leading the transdisciplinary and multi-institutional Molecular Pathological Epidemiology (MPE) Laboratory (The MPE Laboratory). Notably, I am the only Harvard faculty member with appointments in both pathology and epidemiology, and likely I am the first and only one in the history of Harvard University. I am the only pathologist among faculty members of Master of Science in Computational Biology and Quantitative Genetics (CBQG) Program at Harvard Chan SPH.
The relationship between exposures and molecular changes in tumors has been examined for decades (eg, smoking -> KRAS mutation), usually under the umbrella of "molecular epidemiology". However, this type of analysis needs a consideration of disease heterogeneity (eg, KRAS mutation present vs. absent), which necessitates a paradigm shift from conventional epidemiology. Conventional epidemiology is based on the premise that individuals with a given disease (by name) are homogeneous and can be lumped together to analyze exposure associations.
Considering this paradigm shift, “Molecular Pathological Epidemiology (MPE)” (or Molecular Pathologic Epidemiology) has been coined for integrative (interdisciplinary and transdisciplinary) science (Ogino et al. J Natl Cancer Inst 2010; Ogino et al. Gut 2011; Ogino et al. Nat Rev Clin Oncol 2011; Ogino et al. Int J Epidemiol 2012; Ogino et al. Am J Epidemiol 2012). MPE is conceptually defined as "epidemiology of molecular pathology, pathogenesis and heterogeneity of disease" (Ogino et al. Oncogene 2013). MPE encompasses ALL HUMAN DISEASES (Ogino et al. Mod Pathol 2013), and MPE research can be done in any epidemiology cohorts. MPE of non-neoplastic diseases (such as obesity) is a possible future research area (Field et al. JAMA 2013). I have been advancing the MPE field as the first “Molecular Pathological Epidemiologist (MPE-ist)” (Ogino et al. Am J Epidemiol 2012). The power and promise of MPE has been well attested by our recent studies (eg, Morikawa et al. JAMA 2011; Liao et al. N Engl J Med 2012; Nishihara et al. JAMA 2013; Nishihara et al. N Engl J Med 2013, etc.). As I am a unique faculty in pathology (Harvard Medical School) and epidemiology (Harvard School of Public Health), my long-term goal is to transform pathology and epidemiology into an integrative science for both education and research.
In MPE, we try to dissect complex interrelationship between; (1) environmental, dietary, lifestyle and genetic factors; (2) alterations in the tumor microenvironment and cellular molecules; and (3) disease pathways and evolution. MPE encompasses all human diseases (most of which are complex multifactorial diseases), and takes into account genome, epigenome, methylome, transcriptome, proteome, metabolome, microbiome, metagenome, reactome, and interactome. MPE is based on the "Unique Tumor Principle" (Ogino et al. Expert Rev Mol Diagn 2012) and more broadly, the “Unique Disease Principle” (Ogino et al. Mod Pathol 2013). MPE design can be used as the next step of genome-wide association study (GWAS) (“GWAS-MPE Approach”; Ogino et al. Gut 2011; Garcia-Albeniz et al. Carcinogenesis 2013; Nan et al. J Natl Cancer Inst in press). Other new concepts related to MPE include "Colorectal Continuum Paradigm / Theory / Hypothesis" (Yamauchi, Morikawa, et al. Gut 2012; Yamauchi, Lochhead, et al. Gut 2012), which underscores the importance of interplay of gut microbiota, host factors (diet, immunity, inflammation, etc.), and carcinogenesis. This novel “Colorectal Continuum Paradigm” has a considerable impact on gastrointestinal research and clinical practice. The “Etiologic Field Effect” (EFE) concept (Lochhead et al. Mod Pathol 2015) is the expanded paradigm of field effect, considering contribution of the exposome from environment as well as the interactome in the tumor microenvironment. EFE is applicable to all steps of cancer evolution. "The lifecourse-MPE model" is an integrative science of lifecourse epidemiology and MPE (A Nishi et al. Am J Prev Med 2015). "Immuno-MPE" is an integrative science of cancer immunology, molecular pathology and epidemiology, and the first study examined plasma vitamin D and risk of colorectal carcinoma according to level of lymphocytic or T-cell infiltrate in tumor (M Song et al. Gut 2015).
The MPE paradigm has been in widespread use; e.g., Curtin et al. Pathol Res Int 2011; Hughes et al. PLoS ONE 2011; Hughes et al. Int J Epidemiol 2012; Ku et al. Mod Pathol 2012; Rex et al. Am J Gastroenterol 2012; Koshiol et al. Ann Epidemiol 2012; Chia et al. Nat Rev Clin Oncol 2012; Spitz et al. Cancer Discovery 2012; Galon et al. J Tranl Med 2012; Rosty et al. Mod Pathol 2013; Lam et al. Cancer Epidemiol Biomarkers Prev 2013; Weijenberg et al. Curr Nutr Rep 2013; Buchanan et al. Cancer Epidemiol Biomarkers Prev 2013; Burnett-Hartman et al. Cancer Res 2013; Alvarez et al. World J Gastroenterol 2013; Hagland et al. Dig Surg 2013; Zaidi et al. Prog Lipid Res 2013; Hughes et al. Cancer Res 2013; Abbenhardt et al. Int J Mol Epidemiol Genet 2013; J Bae et al. Brit J Cancer 2013; Amirian et al. Infectious Agents and Cancer 2013; Ikramuddin and Livingston. JAMA 2013; Hoffmeister et al. Cancer Epidemiol Biomarkers Prev 2013; RF Araujo et al. Pathol Res Pract 2013; Y Zhu et al. Brit J Cancer 2014; NJ Shaheen Am J Gastroenterol 2014; HR Hagland et al. Cancer Letters 2014; J Brandstedt et al. BMC Cancer 2014; F Coppedè. Expert Rev Gastroenterol Hepatol 2014; F Bishehsari et al. World J Gastroenterol 2014.
We have been utilizing comprehensive MPE databases of two U.S. nationwide prospective cohort studies, the Nurses’ Health Study (N=121,000 followed since 1976) and the Health Professionals Follow-up Study (N=51,500 followed since 1986), as well as CALGB trials (Alliance for Clinical Trials in Oncology). Our discoveries by the MPE approach include (to mention just several); influence of the tumor microenvironment on tumor phenotype (Straussman et al. Nature 2012); YAP1 (yes-associated protein 1; or YAP) in colorectal cancer (Barry et al. Nature 2013); Interaction between aspirin use and PIK3CA mutation in colorectal cancer (Liao et al. N Engl J Med 2012, which is replicated by Domingo et al. J Clin Oncol 2013); interactions between aspirin use and PTGS2 (cyclooxygenase-2, COX-2) expression in colorectal cancer (Chan et al. N Engl J Med 2007; Chan et al. JAMA 2009); aspirin use and BRAF mutation in colorectal cancer (Nishihara et al. JAMA 2013); endoscopy screening and post-colonoscopy colorectal cancer with CIMP and microsatellite instability MSI (Nishihara et al. N Engl J Med 2013); interactions between obesity (host energetics) and FASN (fatty acid synthase) expression in colorectal cancer (Ogino et al. J Clin Oncol 2008; Kuchiba et al. J Natl Cancer Inst 2012); interactions between host energetics and CTNNB1 (-catenin) activation in colorectal cancer (Morikawa et al. JAMA 2011; Morikawa et al. Cancer Res 2013); LINE-1 (long interspersed nucleotide element-1) hypomethylation and colon cancer aggression (Ogino et al. J Natl Cancer Inst 2008) and colorectal cancer family history (Ogino et al. J Natl Cancer Inst 2013); one-carbon nutrients, alcohol and colon cancer risk according to TP53 (p53) status and LINE-1 methylation level (Schernhammer et al. Gastroenterology 2008; Schernhammer et al. Gut 2010); fusobacterium (microbiota) in colorectal cancer (Kostic et al. Genome Res 2012); CDK8 in colorectal cancer (Firestein et al. Nature 2008; Firestein et al. Int J Cancer 2010); VTI1A-TCF7L2 fusion (translocation) in colorectal cancer (Bass et al. Nat Genet 2011). In summary, a better understanding of heterogeneity of carcinogenic processes and influences of exogenous and endogenous factors will further contribute to personalized prevention and treatment strategies.
To transform pathology and epidemiology by the integrative MPE field and concept, I recently launched three programs. (1) The MPE Working Group (MPE WG) aims to establish standardized methodologies in MPE research. The MPE WG currently consists of researchers in the MPE areas, including biostatistics experts such as Drs. Donna Spiegelman, Molin Wang and Bernard Rosner. (2) International MPE Meeting (Molecular Pathological Epidemiology Meeting Meeting) aims to gather selected experts and discuss ways to advance this transdisciplinary science. The first meeting in Boston on April 24, 2013 (my birthday) went very successfully. The Second International MPE Meeting (the first open international MPE meeting was held in Boston on December 4-5, 2014, was a resounding success, hosting a larger group of epidemiologists, biostatisticians and pathologists. The Third International MPE Meeting will be held in Boston on May 12-13, 2016 at Boston Red Sox Jimmy Fund Auditorium. (3) The “STROBE-MPE” initiative (Ogino et al. Am J Epidemiol 2012) is an extension of STROBE, which stands for “STrengthening of the Reporting of OBservational Epidemiology” (von Elm et al. PLoS Med 2007), and is an international guideline for epidemiology research. I always call for collaboration in this STROBE-MPE international initiative. In addition, (4) I am a member of Education and Professional Development Committee (chaired by Dr. Jay Kaufman) of Society for Epidemiologic Research (SER) to address paucity of interdisciplinary education opportunities.
Another focus of my research (which is very much related to MPE) is epigenetics and epigenomics of colorectal cancer. We have found that LINE-1 hypomethylation in colorectal cancer is associated with shorter survival (Ogino et al. J Natl Cancer Inst 2008), family history of colorectal cancer (Ogino et al. J Natl Cancer Inst 2013) and young age of onset (Baba et al. Mol Cancer 2010); and that it is preventable by high folate (leafy vegetables) and low/no alcohol intakes (Schernhammer et al. Gut 2010). I have been characterizing the CpG island methylator phenotype (CIMP), a unique molecular phenotype in colorectal cancer. My investigation led to the discovery of “CIMP-low” associated with KRAS mutation (Ogino et al. J Mol Diagn 2006), which has been confirmed by other investigators including Peter Laird’s group (Hinoue et al. Genome Res 2012). We were the first to apply "structural equation modeling (SEM)" to epigenetic and genetic correlation structure analysis (Tanaka et al. Am J Pathol 2010). We have been deciphering prognostic roles of CIMP, MSI, KRAS, BRAF and PIK3CA mutations, and immune reactions in colorectal cancer (Ogino et al. J Clin Oncol 2009; Ogino et al. Gut 2009; Ogino et al. Clin Cancer Res 2009; Nosho et al. J Pathol 2010; Ogino et al. Clin Cancer Res 2012; Liao et al. Clin Cancer Res 2012; Imamura et al. Clin Cancer Res 2012; Lochhead et al. J Natl Cancer Inst 2013; Ogino et al. J Natl Cancer Inst 2013).