January 31, 2007

NIH Campus, Building 31, Room 4C32
Bethesda, Maryland

Elizabeth Gretz, PhD, NIAMS
Stephen I. Katz, MD, PhD, NIAMS
Jill Buyon, MD, New York University

The NIAMS roundtable on Gender and Sex Factors in Inflammation and Immune-Mediated Diseases was organized as a follow-up to a workshop on the influence of gender and sex in the regulation of inflammatory responses that was sponsored by the NIH Office of Research on Women's Health (ORWH). Participants were familiar with the ORWH meeting's recommendations, which served as a starting point for discussion of how gender and sex factors may play a role in the pathogenesis and response to treatment in autoimmune and inflammatory diseases of interest to the NIAMS.

The roundtable agenda was organized around four topics-the roles of estrogen and estrogen receptors in tolerance and autoimmunity; the influence of gender and sex factors in development of inflammation and perception of pain; the interplay of gender and sex factors with genetic and environmental causes of disease; and clinical and translational research opportunities. As participants discussed research specific to each topic, certain points were repeated throughout the day:

  • Elucidating the mechanisms by which sex and gender influence disease susceptibility, manifestation, and progression will yield important clues about disease variations in both sexes.
  • Sex and gender differences in autoimmune diseases are due to many factors besides estrogen and lymphocytes. Epithelial cells, endothelium, and adipocytes were mentioned as examples of cells involved in autoimmune or autoinflammatory diseases that are likely to function in sex-specific ways.
  • The research community needs to increase its efforts to share resources, data sets, and information because major advances in autoimmunity may come from laboratories studying other topics. The NIH must continue to facilitate collaboration, and to encourage use of animal resources developed by other investigators.
  • Better models are needed to understand the pathophysiology of autoimmune diseases, and methods must be developed to facilitate translation of findings. Researchers not only have difficulty translating results from animal models to humans; translation from in vitro experiments into animal models and from patients into new, hypothesis-driven laboratory experiments also is challenging. Although identification of sex-specific factors that influence autoimmune diseases in humans will aid greatly in establishing appropriate models, model development and use is hampered because many researchers do not provide sufficient information about their cell culture methods (including the sex of cells studied) when reporting results.


A primary point of the discussion about Estrogen and Estrogen Receptors in Tolerance and Autoimmunity was that sex differences are caused by more than just estrogen levels. Participants noted that autoimmune diseases are more prevalent in women than in men even though estrogen suppresses immune function. Therefore, efforts to reconcile estrogen-mediated suppression of immunity with the increased disease burden in women have promise for advancing the medical community's understanding of autoimmune disease development in both sexes.

When considering the influence of estrogen on biologic processes, researchers must remember that local concentrations of estrogen are likely to differ throughout an organism. Local levels of other factors—such as estrogen metabolites and various estrogen receptors—also are likely to play a role in disease manifestation. Estrogen metabolites, for example, can cause or reduce inflammation. Differential expression of estrogen receptors in various cell and tissue types (e.g., immune cells, endothelium) and under different conditions (e.g., age, adiposity) were noted as understudied areas of importance. Little is known about estrogen-receptor trafficking in a variety of cells, including components of the immune system. Because sex hormone-mediated signaling also is important for cell differentiation and turnover, elucidating hormonal effects at a cellular level will likely have relevance to numerous human diseases.

Cell culture studies of sex hormones are particularly challenging, however. Fetal calf serum contains high levels of estrogen and, furthermore, it is not standardized for calf gestational age and sex. Although some laboratories try to circumvent lot-dependent variations by removing estrogen, insulin, and other ingredients from the media, discussants recognized that the existing solutions are not ideal. They also noted that freshly cultured cells have estrogen bound to surface receptors, the expression of which is influenced by the extracellular environment. Additionally, many laboratories do not have any knowledge about the age or sex of their cell source, although age and sex may be important variables affecting interpretation and generalizability of results (groups using cells derived from fetal foreskin were a notable exception). Participants suggested that the research community would be more cognizant of the influence that sex may have on tissue culture lines if reviewers and editors began to emphasize to authors that cell lineage should be noted.

Hormonal variations that occur on a whole-organism level with developmental stages (e.g., puberty, fertile adulthood), chronobiologic cycles (e.g., circadian, menstrual, or seasonal cycles), and reproductive stages (e.g., pregnancy, postpartum) are widely recognized, and researchers from numerous fields are exploring environmental conditions that can influence levels of sex hormones. Considerably less is known, however, about hormonal influences on an organism's response to its environment. Just as the development of puberty in young girls can be triggered by environmental exposures that do not have noticeable effects on the hormonal status of adult women, discussants indicated that a person's developmental stage (and associated levels of gonadal hormones) may render the individual more susceptible to disease when exposed to pathogens that, at a different time, might have no effect.

Susceptibility to environmental triggers also was discussed in-depth in the session on Genetics and the Environment. Across the United States, researchers are conducting epidemiologic studies to assess associations between young girls' environmental exposures and breast cancer risk. These cohorts could be a valuable resource that also could be used to study environmental factors associated with autoimmune diseases.

Many chemicals thought to contribute to autoimmune diseases are found in low levels in the environment. Although men and women might differ in their responses to chemical exposures, and such variations may affect disease manifestation, little evidence has been accumulated from human or animal studies to support or refute these hypotheses. Discrepancies between men and women in susceptibility to disease due to low-level exposures are thought to be due to differences in metabolism or storage that cause otherwise harmless compounds to accumulate to disease-inducing levels, but more research is needed in this area.

During the discussion on Inflammation and Pain, participants noted that sex differences with respect to pain perception and analgesic response had been discussed at length in the 2001 Institute of Medicine report Exploring the Biological Contributions to Human Health—Does Sex Matter? (www.nap.edu/books/0309072816/html/). In contrast to a decade ago, the research community now recognizes sex differences as being significant. Researchers have yet to dissect, however, which features of pain processing are different and the mechanisms by which hormones play a role in pain processing. Participants debated whether women are more responsive to pain than men; if the two sexes differ in their reactions to somatic and visceral (or acute and chronic) pain; and if pain sensitivity differences observed between men and women depend on the anatomic regions involved (e.g., extremities, face, abdomen).

They agreed that much needs to be resolved, and emphasized that studies must be designed to control for societal, as well as biologic, confounders. Although MRI studies have revealed differences in brain activity between men and women, researchers have not been able to interpret what they mean. Electrophysiological measurements in rats and mice have demonstrated sex differences in pain processing, but results vary with the types of pain (acute vs. inflammatory) and the species studied.

Current efforts are directed toward breeding mice that have heightened or blunted pain responses. Such animals will be useful not only for elucidating the molecular underpinnings of pain, but also when studying different human conditions. Women who have temporomandibular joint disorder or fibromyalgia, for example, tend to have a hypersensitivity to somatic pain relative to healthy women. Participants with experience in studying pain stated that researchers need to improve the tools they use to measure pain. They also must develop methods for collecting information about pain over time so that disease progression and changes in symptoms can be quantified.

The influence of pain on inflammation was mentioned as another important research question. Because inflammation can be an appropriate response to adverse stimuli, questions about how inflammation is triggered when needed—and how it is down-regulated when no longer useful—are not unique to investigators studying diseases of interest to the NIAMS. Participants noted that the definition of "inflammation" varies with the disease or tissue being studied, and stated that the absence of a common definition for inflammation is impeding research on how inflammatory responses are regulated.

Not all cells of the same tissue have identical responses to insults. Epigenetic mosaicism, caused by differential expressions of genes from maternal and paternal X chromosomes, adds a level of complexity in females that is not observed in males. More research on the mechanisms regulating X-chromosome inactivation and gene dosing would be helpful for understanding differences in disease manifestation both between men and women, and within a single sex.

Studies of karyotypes other than the standard XX and XY also will provide valuable information about sex factors that influence disease. Karyotypes associated with increased incidence of autoimmune diseases include Klinefelter syndrome (47,XXY), in which men have low levels of testosterone and are at increased risk of lupus. Girls in whom Turner syndrome is caused by a defective X chromosome (XQ) have a high incidence of thyroid disease, but this increased risk is not observed in those who have Turner syndrome due to monosomy X (XO). Karyotype variation within an organism also is associated with disease; some female patients who have scleroderma or autoimmune thyroid disease have been identified as having a subset of B cells that lack an X chromosome (monosomy X B cells).

During the session on Genetics and the Environment, scientists voiced concerns that results from genetic studies will have ramifications beyond improving health. Researchers can protect only the personal information of study participants-once a genetic test has been developed, any insurance company or employer who has access to DNA samples can use it and possibly deny coverage or employment to people who carry certain mutations. The NIH National Human Genome Research Institute is studying this and related issues, and the U.S. Congress is considering legislation to prevent genetic discrimination.

In the absence of definitive tests, health care providers often have difficulty diagnosing autoimmune disease which, in turn, causes difficulty for researchers who study patients. Lupus was mentioned as an example because the phenotype of lupus can vary: a patient must meet 4 of 11 diagnostic criteria to have lupus. Although racial and sex differences in symptom manifestation have been reported, participants wondered about the extent to which societal influences affect how men and women perceive symptoms (e.g., whether men are more likely than women to ignore symptoms such as photosensitivity or alopecia, rather than experiencing those symptoms less often than women do). Furthermore, the phenotype for some autoimmune diseases can vary over time—some lupus patients are initially diagnosed as having rheumatoid arthritis or scleroderma. Researchers also have a difficult time ascertaining at what age patients developed lupus, as people do not get diagnosed immediately after they begin to exhibit symptoms.

Other people have autoantibodies associated with disease, but are asymptomatic. This observation is challenging the dogma that autoantibodies cause autoimmune diseases. Many now believe that, for some diseases, they are simply a useful biomarker—a paradigm shift that is reopening the question of what, exactly, is an autoimmune disease.

The importance of collaboration and the paucity of animal models had been discussed during the morning session on Estrogen and Estrogen Receptors in Tolerance and Autoimmunity and received additional consideration during discussion of Clinical and Translational Research. While participants debated the merits of various animal models, much of their discussion focused on mice. They concluded that, regardless of the animal model adopted, studies that use animals of both sexes are likely to yield information about the role of sex and sex differences in disease and health that cannot be obtained by single-sex animal studies.

Several participants seemed unaware that NIH is cataloging mice and rats that are available as resources, and supports several repositories including the Mutant Mouse Regional Resource Centers, the Knockout Mouse Project, and the Rat Resource and Research Center. Staff emphasized that mice and cell lines are available, but the investigators are responsible for identifying research questions that they will address.

Some participants were enthusiastic about developing long-term epidemiologic studies (similar to the Children's Health Study) that may provide information about the role of environmental exposures in disease. Another proposal was for investigators to plan long-term, comprehensive studies with the intention of mentors passing down these studies to the next generation of scientists (an approach has been implemented by the Framingham Heart Study, for example). Although existing medical records databases can be used for some studies, investigators identified several variables that may be important for research, but about which information is not collected (e.g., dosage of birth control pills or hormone therapy and length of use).

The NIH should encourage scientists who are proposing clinical studies on sex differences in autoimmune diseases also to propose research that would provide insights into associated mechanisms. Because the people who run clinical trials are not necessarily interested in conducting experiments that would answer basic biologic questions, answers to those important questions are not pursued. Participants seemed to agree that basic researchers who will be performing ancillary studies should be included in the earliest phases of study development. They also speculated that, as more is understood about the mechanisms underlying sex differences in autoimmune diseases, researchers from other areas may become more interested in identifying and understanding differences in their male and female patients.

In their closing remarks, several participants noted that the research community must make concerted efforts to collaborate on projects and to share information, data sets, and resources. If groups studying sex differences in separate organ or functional systems would pool their areas of expertise (e.g., if neurologists and immunologists would collaborate), researchers could make major advances toward understanding sex differences with greater efficiency. Although discussants did not propose ways by which groups would reach consensus on experimental methods or agree on standardized definitions of phenotypes, they emphasized that bringing investigators together for discussions like the NIAMS roundtable was an essential first step.

Specifically, the NIH can play a leadership role to advance the field by

  • Improving communication among research groups and enhancing the sharing of information and resources.
  • Developing research infrastructure and resources and ensuring that the research community knows that they are available.
  • Encouraging mechanistic studies that would accompany NIH-funded clinical trials.
  • Emphasizing collaborative efforts and multi-disciplinary and interdisciplinary studies in the review process.

Outside Participants

BUYON, Jill, M.D.
Associate Director, Division of Rheumatology
Professor, Department of Medicine
New York University School of Medicine

CLEGG, Deborah J., Ph.D.
Assistant Professor, Department of Psychiatry
Genome Research Institute, University of Cincinnati

COOPER, Glinda, Ph.D.
Epidemiologist, National Center for Environmental Assessment
United States Environmental Protection Agency

Professor, Division of Rheumatology and Immunology
Medical University of South Carolina

Professor, Department of Biomedical Sciences
University of Maryland, Baltimore

GRIMALDI, Christine, Ph.D.
Associate Professor, Department of Medicine - Rheumatology
Columbia University

KOVATS, Susan, Ph.D.
Assistant Member, Arthritis and Immunology Research Program
Oklahoma Medical Research Foundation

LOCKSHIN, Michael, M.D.
Director, Barbara Volcker Center for Women and Rheumatic Disease
Co-Director, Mary Kirkland Center for Lupus Research
Hospital for Special Surgery

MILLER, Virginia, Ph.D.
Professor, Surgery and Physiology
Mayo Clinic College of Medicine

NELSON, J. Lee, M.D.
Professor of Medicine, Division of Rheumatology
Fred Hutchinson Cancer Research Center

OFFNER, Halina, Ph.D.
Research Immunologist and Professor, Department of Neurology
Oregon Health and Science University

Associate Member, Department of Arthritis and Immunology
Oklahoma Medical Research Foundation

President and CEO
National Women's Health Resource Center

Last Reviewed: