We appreciate Dr. Kuller's perspective (1) on our proposal to integrate pathology and epidemiology education for current and future epidemiology, in which molecular pathological epidemiology (MPE) plays an increasingly major role (2).
MPE differs from conventional molecular epidemiology in its conceptual framework
MPE is based on the concept that tumors or diseases that are usually lumped together as a single entity consist of a heterogeneous group of conditions (3). MPE subcategorizes patients according to molecular pathology to decipher pathogenesis (4–6), whereas conventional molecular epidemiology is largely confined to traditional disease classification, without much attention to taking disease heterogeneity into account. Many of the studies mentioned by Dr. Kuller in his commentary (1) were conventional molecular epidemiology studies that analyzed certain adverse cardiovascular events as a single entity. The use of diagnostic pathology in conventional molecular epidemiology (primarily, to diagnose a disease as a traditional entity) is quite different from the way molecular pathology functions in MPE.
The integration of molecular pathology and epidemiology, a hallmark of MPE, is relatively new (4–6). MPE research has provided not only unique opportunities to dissect causal mechanisms but also unique challenges, which have been discussed in detail elsewhere (6). Thus, MPE should be regarded as a distinct subspecialty in epidemiology, separate from conventional molecular epidemiology. The distinct concept of MPE has been widely adopted (7–20).
Insufficient acknowledgement of disease heterogeneity is a limitation of conventional epidemiology
Dr. Kuller states that epidemiology can still reduce disease morbidity and mortality without MPE-type pathobiologic analysis (1). To illustrate the importance of appreciating disease heterogeneity, we discuss colorectal cancer here as a prototypical example of complex multifactorial disease (21); nevertheless, similar principles apply to other cancers and to noncancerous multifactorial diseases. Because pathologic and biologic features differ according to tumor molecular characteristics, including microsatellite instability (MSI) in colorectal cancer (7, 8, 21, 22), etiologic factors probably influence cancer development and progression differentially by disease subtype (4–6). MPE has the advantage of being able to refine risk estimates for specific disease subtypes and helping to establish causal associations (4–6). MPE studies have suggested that obesity increases risk of colorectal cancer with non-MSI-high (23–25) or low-level FASN expression (26) and that low intake of folate and high consumption of alcohol increase the risks of LINE-1 (long interspersed nucleotide element 1) hypomethylated colon cancer (27). Moreover, MPE studies may help in guiding cancer patient management. For example, aspirin may not be recommended to all colorectal cancer patients, but it may be more effective for patients with PTGS2-positive tumors than for those with PTGS2-negative tumors (28). As another example, physical activity can be recommended to reduce mortality in CTNNB1-negative cancer patients, while physical activity may not be as effective in promoting the survival of CTNNB1-positive cancer patients (29).
Subspecialization is an advancement, not a peril
Dr. Kuller raises a concern about the risk of further compartmentalization of epidemiology into specific areas of disease expertise (1). However, as our understanding of human diseases improves, detailed subject knowledge and expertise will become crucial in epidemiology. We believe that subspecialization is beneficial, while core epidemiologic methods and principles should be common features across subspecialties.
Epidemiologists should be leaders of MPE, as epidemiology transforms into an integrated molecular and population-level science
Change and advancement are fundamental properties of science. As paradigms and concepts change over time, new disciplines will replace obsolete ones. While epidemiology will last as long as there are human diseases, its methods, paradigms, and philosophy will change, in keeping with the advancement of other biomedical sciences. Transformation of epidemiology and epidemiologists is needed on an ongoing basis. We believe that integration of molecular pathology should be a fundamental basis of epidemiology education and research in the 21st century. We need to nurture the next generation of molecular pathological epidemiologists with expertise in molecular pathology and epidemiology. Molecular pathological epidemiologists can lead both MPE and the future of epidemiology.
Author affiliations: Cancer Epidemiology Program, Dana-Farber/Harvard Cancer Center, Boston, Massachusetts (Shuji Ogino, Andrew H. Beck, Edward Giovannucci); Department of Pathology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts (Shuji Ogino, Emily E. King, Danny A. Milner); Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts (Shuji Ogino, Edward Giovannucci); Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts (Shuji Ogino); Department of Pathology, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts (Andrew H. Beck); Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, Maryland (Mark E. Sherman); Department of Immunology and Infectious Diseases, Harvard School of Public Health, Boston, Massachusetts (Danny A. Milner); Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts (Edward Giovannucci); and Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts (Edward Giovannucci).
This work was supported in part by National Institutes of Health grants (grant R01 CA151993 to Shuji Ogino, grant K23 AI072033 to Danny A. Milner, grant P01 CA87969 to Susan E. Hankinson, and grant P01 CA55075 to Walter C. Willett) and in part by the Intramural Research Program of the National Cancer Institute, National Institutes of Health.
The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The funding agencies did not have any role in the decision to submit the manuscript for publication or the writing of the manuscript.
Conflict of interest: none declared.