Based on new evidence and a review of prior studies, the committee for Update 2014 did not find any new significant associations between the relevant exposures and immune outcomes. Current evidence supports the findings of earlier studies that
- No specific diseases involving immune suppression, allergy, autoimmunity, or inflammation had sufficient evidence of an association with the chemicals of interest.
As in Veterans and Agent Orange: Update 20101 (IOM, 2012, hereafter referred to as Update 2010), in this volume immune-system disorders are addressed in a separate chapter that precedes the chapters on other adverse health outcomes. In Veterans and Agent Orange (VAO) reports prior to Update 2010—Veterans and Agent Orange: Health Effects of Herbicides Used in Vietnam, hereafter referred to as VAO (IOM, 1994), Veterans and Agent Orange: Update 1996 (IOM, 1996), Update 1998 (IOM, 1999), Update 2000 (IOM, 2001), Update 2002 (IOM, 2003), Update 2004 (IOM, 2005), Update 2006 (IOM, 2007), and Update 2008 (IOM, 2009)—the possible adverse health outcomes arising from disruptions of the immune system were included in the “Other Health Effects” chapter. The current
1Despite loose usage of “Agent Orange” by many people, in numerous publications, and even in the title of this series, this committee uses “herbicides” to refer to the full range of herbicide exposures experienced in Vietnam, while “Agent Orange” is reserved for a specific one of the mixtures sprayed in Vietnam.
committee elected to revisit comprehensively the limited epidemiologic evidence concerning the association of immune disease with herbicide exposure in light of the substantial volume of toxicologic evidence of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) impairment of the immune systems of laboratory animals.
This chapter opens with an overview of the various types of health problems that can arise from the malfunctioning of the human immune system. The standard VAO sections leading to the committee’s assignment of a health outcome to a category of association follow and include a new tabulation of all the immune-related epidemiologic information that has been considered in this series and a synopsis of the information that is new in this update. The next section discusses factors that may lead the immune responses of animals exposed to the chemicals of interest (COIs) to be much more pronounced than any observed to date in humans. The chapter closes with the committee’s thoughts regarding research on the possibility that immune perturbations in humans function as a mechanistic step in the development of disease processes in other organ systems.
The immune system plays three important roles in the body:
- It defends the body against infection by viruses, bacteria, and other disease-producing microorganisms, known as pathogens.
- It defends against cancers by destroying mutated cells that might otherwise develop into tumors and by providing immunity against tumors.
- It provides resident immune cells that are specially adapted for different tissues and organs (such as microglia in the central nervous system and Kupffer cells in the liver) that help to regulate the functional activity and integrity of those tissues.
To recognize the wide array of pathogens in the environment, the immune system relies on many cell types that operate together to generate immune responses. These cells arise from stem cells in the bone marrow, are found in lymphoid tissues throughout the body, and circulate in the blood as white blood cells (WBCs). The main types of WBCs are granulocytes, monocytes, and lymphocytes. Each type has many specialized cell populations that are responsible for specific functions connected to the production of specific mediators, such as immune hormones, cytokines, and other secreted factors. Imbalances in those specialized populations or in their level of functional activity can result in inadequate or improper immune responses, which may lead to pathologic outcomes. Diseases arising from immune dysfunction may be apparent immediately or observed only after an organism encounters an environmental challenge that causes the immune cells to respond (such as an infection).
There are four major categories of immune dysfunction, which are not mutually exclusive: immune suppression, allergy, autoimmunity, and inflammatory
dysfunction (inappropriate or misdirected inflammation). Immune suppression usually manifests itself as an increased incidence of infections or an increased risk of neoplasatic, allergic, autoimmune, and inflammatory disorders can be manifested as diseases that affect virtually any tissue. It is often difficult to diagnose such diseases, so they may or may not be medically categorized as immune disorders.
The suppression of immune responses can reduce resistance to infectious disease and increase the risk of cancer. Infection with the human immunodeficiency virus (HIV) is a well-recognized example of an acquired immune deficiency in which a specific type of lymphocyte (CD4+ T cell) is the target of the virus. The decline in the number of CD4+ T cells after HIV infection correlates with an increased incidence of infectious diseases, including fatal opportunistic infections, and with an increased incidence of several types of cancer. The treatment of cancer patients with toxic chemotherapeutic drugs suppresses the immune system by inhibiting the generation of new WBCs by the bone marrow and blocking proliferation of lymphocytes during an immune response. Both of those examples represent severe immune suppression in which the adverse outcome is easily detected with clinical measurements.
Immune suppression can also result from exposure to chemicals in the workplace or in the environment and manifest as recurrent infections, opportunistic infections, a higher incidence of a specific category of infections, or a higher incidence of many forms of cancer. However, unless the immune suppression is severe, it is often difficult to obtain clinical evidence that directly links chemically induced changes in immune function to increases in infectious diseases or cancers, because many confounding factors can influence a person’s ability to combat infection. Such confounders include age, vaccination status, the virulence of the pathogen, the presence of other diseases (such as diabetes), stress, smoking, and the use of drugs or alcohol. Therefore, immunotoxicology studies are often conducted in laboratory animals to understand the scope and mechanism of chemical-induced immune suppression. The results of such studies can be used to develop biomarkers to assess the effects in human populations. Infectious-disease models in animals can also be used to determine whether the pattern of disease changes with chemical exposure.
The immune system sometimes responds to a foreign substance that is not pathogenic. Such immunogenic substances are called allergens. Like most immune-based diseases, allergic diseases have both environmental and genetic risk factors. Their prevalence has increased in many countries in recent decades (CDC, 2004b;
Linneberg et al., 2000; Simpson et al., 2008; Sly, 1999). Major forms of allergic diseases are asthma, allergic rhinitis, atopic dermatitis, and gastrointestinal responses. In immediate hypersensitivity, the response to some allergens, such as pollen and bee venom, results in the production of immunoglobulin E (IgE) antibodies. Once produced, IgE antibodies bind to mast cells, which are specialized cells that occur in tissues throughout the body such as lung airways, the intestinal wall, and blood-vessel walls. When a person is exposed to the allergen again, it binds to the antibodies on the mast cells and causes them to release histamine and leukotrienes, which produce the symptoms associated with an allergic response. In delayed-type hypersensitivity (DTH) reactions, also known as cell-mediated immunity, other allergens, such as poison ivy and nickel, activate allergen-specific lymphocytes (memory T-cells) at the site of contact (usually the skin) that release substances that cause inflammation and tissue damage. Some allergic responses, such as those to food allergens, may involve a combination of allergen-specific lymphocyte-driven and IgE-driven inflammation. Allergic responses may be manifested in specific tissues (such as skin, eyes, airways, and gastrointestinal tract) or may result in a system-wide response called anaphylaxis.
The National Institutes of Health’s Autoimmune Disease Coordinating Committee recognizes 80 different autoimmune diseases and conditions which affect the cardiovascular, respiratory, nervous, endocrine, dermal, gastrointestinal, hepatic, and excretory systems (NIH Autoimmune Diseases Coordinating Committee, 2005). These diseases affect both men and women, but most of them affect more women than men (Fairweather et al., 2008). Genetic predisposition, age, hormone status, and environmental factors, such as the presence of infectious diseases and stress, are known to affect the risk of developing autoimmune diseases, and different autoimmune diseases tend to occur in the same person and to cluster in families. The existence of some autoimmune diseases is also a risk factor for the development of other immune-related diseases, such as some types of cancer (Landgren et al., 2010).
Autoimmune disease is an example of the immune system’s causing rather than preventing disease: The immune system attacks the body’s own cells and tissues as though they are foreign. Inappropriate immune responses that result in autoimmune disease can be promoted by different components of the immune system (such as antibodies and lymphocytes) and can be directed against a wide variety of tissues or organs. For example, the autoimmune reaction in multiple sclerosis is directed against the myelin sheath of the nervous system; in Crohn disease, the intestine is the target of attack; in type 1 diabetes mellitus, the insulin-producing cells of the pancreas are destroyed by the immune response; and rheumatoid arthritis arises from an immune attack on the joints, although it can also involve the lung, heart, and additional organs.
More generalized forms of autoimmune diseases also occur. Systemic lupus erythematosus (SLE) is an autoimmune disease in which multiple organs are targeted by immune attack. In such a case, patients have a variety of symptoms that often occur in other diseases, which makes diagnosis difficult. A characteristic rash across the cheeks and nose and a sensitivity to sunlight are common symptoms of SLE; oral ulcers, arthritis, pleurisy, proteinuria, and neurologic disorders may also be present. Almost all people who have SLE test positive for antinuclear antibodies in the absence of drugs known to induce them. The causes of SLE are unknown, but environmental and genetic factors have been implicated. Some of the environmental factors that may trigger it are infections, antibiotics (especially those in the sulfa and penicillin groups) and some other drugs, ultraviolet radiation, extreme stress, and hormones. Occupational exposures to such chemicals as crystalline silica, solvents, and pesticides have also been associated with SLE (Cooper and Parks, 2004; Parks and Cooper, 2005).
Inflammatory diseases (also referred to as auto-inflammatory diseases) make up a more recently identified category of immune-related disorders and are characterized by exaggerated, excessively prolonged, or misdirected dysfunctional inflammatory responses (usually involving immune cells). Tissue disease can result from this inappropriate inflammation, which can affect virtually any organ. Examples of the diseases and other conditions that are most often included in other disease categories but are also considered to be inflammatory diseases are coronary arterial disease, asthma, eczema, chronic sinusitis, hepatic steatosis, psoriasis, celiac disease, and prostatitis. Inflammatory diseases often occur with one another, which has resulted in the categorizing of different but linked inflammatory diseases together as a single chronic inflammatory disorder (Borensztajn et al., 2011); among these inflammatory disorders are atherosclerosis and chronic pulmonary obstructive disease. Inappropriate inflammation also appears to play a role in promoting the growth of neoplasms (Bornschein et al., 2010; Hillegass et al., 2010; Landgren et al., 2010; Porta et al., 2011; Winans et al., 2010); examples can be seen in the higher prevalence of specific cancers in patients who have such inflammatory diseases as inflammatory bowel disease (Lucas et al., 2010; Viennot et al., 2009; Westbrook et al., 2010), prostatitis (Sandhu, 2008; Wang W et al., 2009), and psoriasis (Ji et al., 2009).
Ordinarily, inflammation can be advantageous in fighting infectious diseases. It is one component of the normal host response to infection and is mediated by innate immune cells. Inflammatory responses have evolved to speed the movement of macrophages, granulocytes, and some lymphocytes to the area of infection, where they produce toxic metabolites that kill pathogens. Interactions among innate immune cells and epithelial and endothelial cells are important in regulating the magnitude of inflammation, and improperly regulated inflammation
can contribute to diseases that arise in non-lymphoid tissues, such as the lungs, skin, nervous system, endocrine system, and reproductive system.
The following comments are restricted to findings related to the immune system that occur after adult human exposures. For a discussion of potential effects on the immune system arising from early-life (such as perinatal) exposures (which would not be directly applicable to the Vietnam veterans who are the target of this report), see Chapters 4 and 9. Studies that served as the basis of prior updates of VAO are shown in Table 7-1.
A handful of the direct studies of veterans listed in Table 7-1 reported a statistically significant difference in a single immune measure (Kim HA et al., 2003; Michalek et al., 1999b). But invariably the same effect was not found in other studies of Vietnam veterans, nor was support for the effect found in epidemiologic studies of other populations. Thus, there were no consistent findings indicative of immunosuppression, increased risk of autoimmunity (usually as measured with autoantibodies), or biomarkers of atopy or allergy (such as increased IgE concentrations). Much of the focus of the studies was on measuring T4:T8 ratios. The T4:T8 ratio is an effective biomarker of the progression of HIV-induced AIDS, but the TCDD-exposure animal data indicate that it is not an immunologic index that is expected to be altered. The results of a survey of Australian Vietnam veterans (O’Toole et al., 2009) included purportedly significant increases in the prevalence of a number of conditions in which immune function may play a prominent role, but the study’s methods were deemed unreliable.
The occupational-exposure studies shown in Table 7-1 evaluated the concentrations of lymphoid populations in circulation, such as CD4, CD8 (and the ratio of the two), and natural killer (NK) cells; cell-mediated immunity (the delayed-hypersensitivity response); serum concentrations of immunoglobulins, such as IgM, IgG, and IgA; concentrations of complement, such as C3 and C4; and concentrations of cytokines, such as IL-1, IL-2, interferon-gamma, IL-4, IL-6, and tumor necrosis factor (TNF)-alpha. A few studies also included disease or condition end points, such as rheumatoid arthritis, SLE, immune suppression, and sensitivity to fungal infection. Ex vivo analyses included measures of NK activity, lymphoid mitogen-induced proliferation, and the mixed lymphocyte response (MLR) against allogeneic cells. Some studies identified one or more dioxin-related shifts in immune measures, but many reported no significant differences
|US Air Force Health Study—Ranch Hand veterans vs SEA veterans||All COIs|
|Participants in 1997 examination cycle, Ranch Hands vs comparisons (incidence)||No change in surface markers for B and T cells, no change in serum Ig, no change in autoantibodies (antinuclear antibody, smooth muscle autoantibody, parietal cell autoantibody, rheumatoid factor, and monoclonal immunoglobulins), and no dose-related change in DTH response||Michalek et al., 1999b|
|Participants in 1987 examination cycle, Ranch Hands vs comparisons (morbidity)||No change in surface markers for B and T cells||Wolfe et al., 1990|
|Participants in 1985 examination cycle, Ranch Hands vs comparisons (morbidity and mortality)||No change in surface markers for B and T cells||Wolfe et al., 1985|
|US CDC Vietnam Experience Study—||All COIs|
|Cross-sectional study, with medical examinations, of Army veterans: 9,324 deployed vs 8,989 non-deployed|
|Morbidity—Deployed vs non-deployed||No differences in infections, no changes in B and T cell-surface markers, WBC counts, or circulating serum Ig||CDC, 1988b|
|Mortality (1965–2000)||No suggestion of excess deaths due to immune-system disorders (ICD-9 240–279, which covers endocrine, nutritional, metabolic, and immunity disorders).||Boehmer et al., 2004|
|US VA Cohort of Monozygotic Twins||All COIs|
|Physical health—morbidity||Increase in skin conditions of unknown etiology, no increase in blood disorders||Eisen et al., 1991|
|US American Legion Cohort||All COIs|
|Physical health and reproductive outcomes||Increase in skin conditions and arthritis||Stellman SD et al., 1988b|
|State Studies of US Vietnam Veterans||All COIs|
|Michigan Vietnam Veterans (deployed vs non-deployed)||Increased mortality from infectious (including parasitic) diseases||Visintainer et al., 1995|
|New Jersey Agent Orange Commission||Depressed response to tetanus in DTH tests, decrease in CD4 and SmIg+ B cells||Kahn et al., 1992b|
|Texas Agent Orange Advisory Committee||Increase in percentage of active T rosette-forming cells||Newell, 1984|
|Sample of 1,000 Male Australian Vietnam Veterans–prevelance||All COIs|
|Australian Vietnam Veterans—longitudinal cohort study of 67 conditions in randomly selected Vietnam veterans vs general population||Increase in hay fever, increases in infectious and parasitic diseases, increase in arthritis||O’Toole et al., 2009|
|Australian Conscripted Army National Service (18,940 deployed vs 24,642 non-deployed)||All COIs|
|1983–1985—Australian Vietnam Veterans—longitudinal cohort study of 67 conditions in randomly selected Vietnam veterans vs general population||Increase in hay fever, increases in infectious and parasitic diseases, increase in arthritis||CDVA, 1997b|
|Korean Vietnam Veterans||All COIs|
|Immunotoxicologic study||Increase in IgE and IL-4, decrease in IgG1 and IFN-gamma, no change in lymphocyte counts||Kim et al., 2003|
|Vietnamese Vietnam Veterans||All COIs|
|Antinuclear and sperm autoantibodies||No change in autoantibodies to sperm, antinuclear bodies||Chinh et al., 1996|
|IARC Phenoxy Herbicide Cohort—Dutch workers from 2 plants that produced and formulated chlorophenoxy herbicides (Plant A, n = 1,167; Plant B, n = 1,143).||Chlorophenoxy herbicides: Negative correlation between TCDD exposure and markers of humoral immunity, except perhaps for C4||Saberi Hosnijeh et al., 2011|
|IARC Phenoxy Herbicide Cohort—||Chlorophenoxy herbicides:|
|Subset of Dutch workers (n = 85) from 2 plants that produced and formulated chlorophenoxy herbicides (high exposure = 47, low exposure = 38); serum collected 30 yrs after exposure||General reduction in most analyte levels with the strongest effects for fractalkine, fibroblast growth factor (FGF2), and transforming growth factor alpha (TGF-α)||Saberi Hosnijeh et al., 2012a|
|High vs low: CD4/CD8 ratio increased (p = 0.05); no difference for other cell counts and lymphocyte subsets Decrease in B cells with increasing serum TCDD||Saberi Hosnijeh et al., 2012b|
|Soluble CD27 and CD30 levels not related to TCDD levels; With exclusion of chronically ill subjects, IL1RA decreased with increasing TCDD levels||Saberi Hosnijeh et al., 2013a|
|IARC Phenoxy Herbicide Cohort—||Dioxins, phenoxy herbicides|
|German production workers (2,479 workers at 4 plants, in IARC as of 1997)|
|Cross-sectional study of 153 male workers in six chemical plants in Germany||TCDD (during production of TCP): DTH responses not correlated with dioxin concentration; slight decrease in IgM was reported with increasing dioxin exposure; overall lymphoid counts not different||Benner et al., 1994|
|German production workers at BASF Ludwigshafen Plant—BASF cleanup workers from 1953 accident (n = 247); 114 with chloracne, 13 more with erythema; serum TCDD levels (not part of IARC)||Focus on TCDD|
|138 surviving workers from a larger cohort of 254 exposed workers after an accident in a BASF TCP production facility||TCDD: Among 14 immune measures; regression analysis of TCDD concentration suggested marginal positive associations with IgG, IgA, C3, and C4; marginal reductions in some lymphocyte population were also reported||Ott et al., 1994|
|IARC Phenoxy Herbicide Cohort—||Dioxins, 2,4,5-T; 2,5-DCP; 2,4,5-TCP|
|German production workers at Boehringer-Ingelheim Plant in Hamburg (1,144 men working > 1 month in 1952–1984; generation of TCDD reduced after chloracne outbreak in 1954)|
|Updated and expanded evaluation of 158 workers in a German chemical plant with differing exposure studied in two trials||TCDD (or “TCDD toxic equivalents” from PCDD/PCDF): No differences in serum Ig or cytokine (IL1, IL6, TNF-alpha)||Neubert et al., 2000|
|19 highly exposed chemical workers vs 28 unexposed controls in two chemical plants in Hamburg, Germany||TCDD (in chemical plant): In subset of leukocytes, increase in CD8+ memory T cells and decrease in naïve T cells (CD45RA+) after TCDD exposure, as was stimulated IFN-gamma production from whole blood cultures associated with TCDD exposure||Ernst et al., 1998|
|192 workers in a German pesticide plant, including 29 highly exposed and 28 controls compared for immune functional tests||TCDD (or TEQs from PCDD/PCDF exposure): No significant changes in TCDD and lymphocyte subsets, antibody responses to vaccination, lymphocyte proliferation, or autoantibody production; decrease in chromate resistance of PHA-stimulated lymphocytes in highest exposure group||Jung et al., 1998|
|Comparison of 11 2,4,5-trichlorophenol production workers 20 years after exposure vs 10 unexposed age-matched workers in the same company||TCDD: No differences in any lymphoid subset or in mitogen-induced proliferation; TCDD exposure was associated with decreases in MLR response and in stimulation with IL-2 in vitro||Tonn et al., 1996|
|Examination of eight trichlorophenol production workers who developed chloracne and were re-examined 15–25 yrs after initial exposure||TCDD: Reduced gamma globulins in the most-exposed workers; no significant effects on T4, T8 ratios||Jansing and Korff, 1994|
|89 volunteers involved in decontamination work at a chemical plant in Hamburg, Germany; no control population||TCDD (or equivalents via PCDD/PCDF exposure): Potentially complicated by age differences among the compared groups; only subtle, clinically nonsignificant changes were seen among immune-cell surface markers in a comparison of higher exposed vs low-exposed to moderately exposed workers||Neubert et al., 1993, 1994|
|NIOSH Cohort (current and former workers from chemical plants in New Jersey and Missouri, 2 of the 12 plants included in the NIOSH Mortality Study)||Dioxins, phenoxy herbicides|
|Cross-sectional study of 259 TCDD-exposed 2,4,5-trichlorophenate (and its derivatives) workers (mean serum TCDD, 223 ppt) and 243 unexposed residential controls (mean serum TCDD, 6ppt)||TCDD (exposure in a chemical plant): No significant changes in serum Ig or major leukocyte categories; TCDD associated with decreased circulating CD26 cells (activated T cells)||Halperin et al., 1998|
|1987 cross-sectional study of 281 chemical-plant workers in NJ and MO at least 15 yrs after exposure vs 260 unexposed controls||TCDD (as a contaminant in chemical production): Increase in TCDD associated with a decrease in CD3/Ta1 (helper lymphocytes) cells||Sweeney et al., 1997/1998|
|Other Studies of Industrial Workers (not related to IARC or NIOSH phenoxy cohorts)|
|EUROPIT Study—Prospective multicenter cohort study (Bulgaria, Finland, Italy, The Netherlands) of 238 pesticide-exposed workers vs 198 unexposed workers||Pesticide factories (not specifically TCDD): Reduced antibody responses to hepatitis B vaccination among exposed workers carrying a specific IL-1 allele||Baranska et al., 2008|
|OCCUPATIONAl—HERBICIDE-USING WORKERS (not related to IARC sprayer cohorts)|
|Agricultural Health Study (AHS)—prospective study of licensed pesticide sprayers in Iowa and North Carolina: commercial (n = 4,916 men), private/farmers (n = 52,395, 97.4% men), and spouses of private sprayers (n = 32,347, 0.007% men), enrolled 1993–1997; follow-ups with CATIs 1999–2003 and 2005–2010||Pesticides/herbicides|
|Comparison from the AHS of 534 cases of self-reported physician-diagnosed depression vs 17,051 controls||Both high-level acute pesticide exposure (OR = 2.6, 95% CI 1.7–3.8) and cumulative pesticide exposure (OR = 1.5, 95% CI 1.2–2.0) were positively associated with increase in depression||Beseler et al., 2008|
|29,074 female spouses of pesticide applicators in the AHS||Depression was significantly associated with pesticide poisoning (OR = 3.3, 95% CI 1.7–6.2) but not with lower cumulative exposure||Beseler et al., 2006|
|Nested case-control study of rheumatoid arthritis in agricultural families (57,000 pesticide applicators and their spouses).||No strong risk factors were identified for pesticide mixing or application or for any specific class of pesticides in the AHS of rheumatoid arthritis.||De Roos et al., 2005b|
|Other Studies of Herbicide-Using Workers|
|Longitudinal study of 10 farmers during 1994 within 7 days before and 1–12 days and 50–70 days after exposure||2,4-D and MCPA formulations: Decreases in percentages of CD4, CD8, CTL, CD8-DR, and NK cells and in NK activity and mitogen-stimulated lymphoproliferation; CD4/CD8 ratio was unaltered; CD3 and CD8 percentages had recovered by the second assessment period; no significant correlations between immune changes and amount of pesticides applied||Faustini et al., 1996|
|Seveso Cleanup Workers Prospective study using analysis of samples from 36 cleanup workers (divided into three groups based on time spent in the contamination area); pre-employment samples and samples after 9 months were analyzed for comparison with samples from 31 unexposed workers||TCDD No differences in WBC counts and platelet counts||Ghezzi et al., 1982|
|Seveso, Italy Residential Cohort—Industrial accident July 10, 1976 (723 residents Zone A; 4,821 Zone B; 31,643 Zone R; 181,574 local reference group)||TCDD|
|Study of 101 chloracne cases vs 211 controls 20 years after the accident; relatively low statistical power was available because the study examined the occurrence of individual diseases||Persistent increase in TCDD in chloracne cases; younger people seemed to be more susceptible; no major trends in disease occurrence||Baccarelli et al., 2005a|
|Study of 62 people from a highly exposed zone and 53 from noncontaminated areas 20 yrs after the accident||Plasma concentration of TCDD was determined; multivariate regression analysis showed significant decrease in plasma IgG with increasing TCDD concentration and no changes in IgM, IgA, or C3||Baccarelli et al., 2002|
|45 children (3–7 yrs of age) living in exposed areas vs 45 unexposed children as controls||No differences in serum IG, mitogen responses of lymphocytes (PHA and pokeweed), or percentage of rosette-forming lymphocytes||Pocchiari et al., 1979|
|Times Beach (MO) Cohort||TCDD|
|Regression analysis used for comparisons among 41 exposed people for adipose-tissue, TCDD vs immune measures; three exposed groups defined by tissue dioxin||No TCDD–DTH response relationships were reported; no change in mitogen responsiveness; some serum markers (A/G ratio and serum IgG) were affected||Webb et al., 1989|
|82 people in more highly contaminated areas vs 40 in low-risk exposure areas as controls||No differences in DTH response or T-cell subsets (T4/T8)||Webb et al., 1987|
|80 people in highly contaminated areas vs 40 controls in lower-risk areas Pilot study of small numbers of people; for comparisons, people were assigned to two environmental-exposure groups: those in high-risk areas (27 men, 23 women, and 15 children) and those in low-risk areas (12 men, 10 women, and 8 children)||No differences in DTH induration or T-cell subset analysis (T4/T8) Multitest DTH evaluation to seven recall antigens was performed, no statistical differences were reported, and only trends were noted; no statistical differences were reported for T-cell markers (T3, T4, and T8) or mitogen-induced lymphocyte proliferation (PHA, Con A, and pokeweed mitogen), and only trends were noted||Stehr et al., 1986 Knutsen, 1984|
|Quail Run Mobile Home Park (MO) Cohort||TCDD|
|A subset of the previously anergic persons in the Stehr-Green et al. (1987) study were re-evaluated in the DTH test with a higher DTH test dose and highly trained, blinded readers||Retesting of DTH failed to produce the differences observed initially||Evans et al., 1988|
|Small (ill-defined) samples were used; comparisons of residents of the Quail Run Mobile Home Park with residents of St. Louis–area trailer parks as controls||DTH suppression in the exposed group was reported, but data from two of four readers were discarded; no differences in T-cell mitogen stimulation; decreases in percentages of T3, T4, and T11 cells in the exposed group||Knutsen et al., 1987|
|154 people in highly contaminated area vs 155 in three low–environmental-contamination areas as controls||Increase in anergy and decrease in induration for DTH in exposed group; data from some readers were excluded; decrease in percentages of T3, T4, and T11 cells, but no difference in cell number of T4/T8 ratio||Stehr-Green et al., 1987|
|80 people in a high–exposure risk group vs 40 controls||Decreases in DTH indurations, number of positive reactors, and percentages of T3, T4, and T11 cells in the exposed group||Andrews et al., 1986|
|154 people in the exposed area vs 155 unexposed people in an uncontaminated area||Recall antigen multitest for DTH, increase in percentage of anergy and decrease in duration in exposed group; data from two of four readers were excluded||Hoffman et al., 1986|
|Other Environmental Studies Belgium (Flanders)—200 people 17–18 yrs of age in three areas of Flanders (Belgium); TEQ values were calculated from serum dioxin–like PCB concentrations, and relationships with immune measures were examined||Dioxins and PCBs: Decreases in eosinophil and NK-cell counts with increasing TEQ; IgE concentrations; history of upper airway allergy, and odds of a positive RAST test correlated negatively with serum TEQ; IgA concentrations correlated positively with TEQ||Van den Heuvel et al., 2002|
|Finland—123 men and 132 women from high–fish consumption group||TEQ for dioxins, furans, and PCBs: CRP was not associated with overall TEQ for men (p = 0.29) or women (p = 0.94)||Tururen et al., 2012|
|Germany—Cross-sectional study of 221 teachers who worked in German day-care centers treated with wood preservatives vs 189 teachers who worked in untreated facilities||Dioxin in wood preservatives, exposure primarily via inhalation: No effects of inhaled dioxin were seen on T4 or T8 cell numbers or on the ratio; some evidence of a dose–response relationship was seen for risk of anergy (or hypoergy) in the DTH assay||Wolf and Karmaus, 1995|
|Japan—1,063 men and 1,201 women without occupational dioxin exposure from 125 areas||All WHO 2005 DlCs: Self-reported asthma not associated with DLCs; marginal association of atopic dermatitis and allergic rhinitis with DLCs||Nakamoto et al., 2013|
|US (Seattle)—109 postmenopausal women tested for immune function at start and after 1 yr exercise program||Mono-ortho PCBs 105, 118, 156: PHA-induced T-lymphocyte proliferation decreased with PCB levels after 1 yr, but not at start; NK cytotoxicity not associated with PCBs at either time||Spector et al., 2014|
|US (NHANES)||Dioxin-like PCBs|
|1,721 adults assessed for serum dioxin–like PCBs and self-reported arthritis||Association between serum dioxin–like PCBs and prevalence of arthritis particularly among women||Lee et al., 2007a|
|632 women and 670 men assessed for dioxin-like PCBs and serum antinuclear antibodies||In women only, TEQ for PCBs associated with positivity for antinuclear antibodies (p < 0.001)||Gallagher et al., 2013|
|Norway—blood samples from 24 Norwegian hobby fishermen were compared with those of 10 male referents as controls||PCDD, exposure from food: The study generally lacks experimental details; no differences in an NK cell marker or in NK activity were seen; apparently, some effects on lymphoid markers were observed but specific details are lacking||Lovik et al., 1996|
|Sweden—23 high consumers of fatty fish from the Baltic Sea (containing low concentrations of PCDD) vs 20 low consumers or nonconsumers of fish as controls||PCDD, exposure from food: Blood PCDDs were significantly different between the groups; mercury concentrations also differed; NK cells correlated negatively with blood concentrations of persistent organic chemicals; no other||Svensson et al., 1994|
|South Korea (Ansan)—comparison of immune measures in 31 waste-incineration workers vs 84 controls||TCDD (via waste incineration): Lymphoid subsets, IFN-gamma, and Ig not statistically different; decrease in IL-4 and increase in T-cell activation (measured as combined CD3 and CD69 markers) associated with TCDD exposure||Oh et al., 2005|
|United Kingdom (Derbyshire)—18 chemical workers in a 2,4,5-T in the Coalite Oils and Chemical, Ltd. factory exposed as a result of an industrial accident 17 yrs before study vs 15 matched controls||TCDD: No changes in serum Ig classes, increases in antinuclear antibodies and immune complexes, and increase in circulating NK cells (Leu7+) in exposed workers||Jennings et al., 1988|
|United States (California)—telephone interviews concerning environmental and occupational chemical exposures were conducted with 50 AIDS patients (with Kaposi sarcoma) and 50 homosexual men as controls||Chemical exposures, including pesticides, and Agent Orange: No significant differences were reported in a small study that generally lacked focus||Hardell et al., 1987|
ABBREVIATIONS: 2,4-D, 2,4-dichlorophenoxyacetic acid; 2,4,5-T, 2,4,5-trichlorophenoxyacetic acid; AHS, Agricultural Health Study; CATI, computer-assisted telephone interview; CDC, Centers for Disease Control and Prevention; CI, confidence interval; COI, chemical of interest; Con A, concanavalin A; CRP, C-reactive protein; DLC, dioxin-like compound; DTH, delayed-type hypersensitivity; IARC, International Agency for Research on Cancer; ICD, International Classification of Diseases; IFN-gamma, interferon-gamma; Ig, immunoglobulin; IL, interleukin; IL1RA, interleukin one receptor agonist; MCPA, methyl-4-chlorophenoxyacetic acid; MLR, mixed lymphocyte response; MO, Missouri; NHANES, National Health and Nutrition Examination Survey; NIOSH, National Institute for Occupational Safety and Health; NK, natural killer; OR, odds ratio; PCB, polychlorinated biphenyl; PCDD, polychlorinated dibenzo-p-dioxin (highly chlorinated, if four or more chlorines); PCDF, polychlorinated dibenzofurans; PHA, phytohemagglutinin; RAST, radioallergosorbent; SEA, Southeast Asia; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; TCP, trichlorophenol; TEQ, total toxic equivalent; TNF, tumor necrosis factor; VA, Department of Veterans Affairs; WBC, white blood cell; WHO, World Health Organization.
in the same measures. Saberi Hosnijeh et al. (2012a) reported a positive correlation between plasma TCDD concentrations and decreased levels of cytokines, chemokines, and growth factors. However, this correlation was not supported by other studies. That is particularly true of the study by Neubert et al. (2000), which measured toxicity equivalents (TEQs) for dioxin but found no immunoglobulin or cytokine alterations. In general, the various occupational-exposure findings do not provide a consistent or clear picture of alterations in immune measures that could be extrapolated to an increased risk of a single disease or even a broader category of diseases. The exception may be observations of pesticide-associated autoimmunity and immune suppression. Immune suppression was rather consistently associated with very high pesticide exposures or pesticide poisonings. However, because the studies generally concerned broad categories of pesticide exposure, their relevance to herbicide exposures in Vietnam is not clear.
Several environmental-exposure studies reported alterations, but the findings were inconsistent among the studies (see Table 7-1). Some studies reported alterations in immune measures associated with TEQs for dioxin. For example, Van den Heuvel et al. (2002) reported that IgE, positive radioallergosorbent (RAST) tests in response to specific allergens, eosinophil counts, and NK-cell counts correlated negatively with dioxin TEQs but that IgA increased; these alterations, however, were not seen consistently in other studies. Baccarelli et al. (2002) found no changes in IgA but saw changes in IgG in the Seveso population. Svensson et al. (1994) found that NK-cell numbers were reduced with increasing concentrations of persistent organic chemicals, but Lovik et al. (1996) found no difference in NK numbers or activity. Similarly, the occupational-exposure studies (see Table 7-1) that examined NK concentrations reported the full spectrum of results: no alterations (Halperin et al., 1998), a decrease (Faustini et al., 1996), and even an increase in NK numbers (Jennings et al., 1988) in dioxin-exposed people.
As seen in Table 7-1, some early studies of the Quail Run Mobile Home Park population exposures reported that dioxin exposure was associated with a reduced cell-mediated immune response, the DTH response (Andrews et al., 1986; Hoffman et al., 1986; Knutsen et al., 1987; Stehr-Green et al., 1987). But some of those studies had technical problems in assessment and in the follow-up analyses. Dioxin-associated changes were not confirmed (Evans et al., 1988; Webb et al., 1989). In addition, several studies of the Times Beach population did not find any alteration of the DTH response in dioxin-exposed populations (Knutsen, 1984; Stehr et al., 1986; Webb et al., 1987).
Analysis of National Health and Nutrition Examination Survey (NHANES) data found that exposure to dioxin-like polychlorinated biphenyls (PCBs) was associated with an increase in self-reported arthritis (Lee et al., 2007a), but De Roos et al. (2005b) found no such association in their study.
Prior VAO updates have concluded that human data were either insufficient or inconsistent with respect to an increased risk of immunosuppression, allergic disease, or autoimmune disease.
Vietnam-Veteran and Case-Control Studies
No new case-control studies or studies of Vietnam veterans exposed to the COIs and adverse immunologic conditions have been published since Update 2010.
Since Update 2012, several additional relevant occupational studies have been reported. In a publication reviewed in Update 2012, Saberi Hosnijeh et al. (2012a) examined serum cytokine concentrations in a subsample of 47 highly TCDD-exposed workers and 38 low-exposed workers selected from those in the Dutch subcohort (Bueno de Mesquita et al., 1993) of the IARC cohort who were alive in 2006 at the end of follow-up by Boers et al. (2010). Highly exposed workers were matched to low-exposed workers by factory, sex, age, and residence at the time of study. Having generated results consistent with immune suppression being associated with TCDD exposure, Saberi Hosnijeh et al. (2012b, 2013a) continued with this group to assess TCDD levels with respect to several other immunological parameters. Comparing the high- and low-exposure groups, Saberi Hosnijeh et al. (2012b) found no differences in hematologic measurements other than an increase in the CD4/CD8 lymphocyte ratio (p = 0.05). With adjustment for age, body mass index (BMI), drinking, smoking, medication, and chronic, inflammatory, or recent infectious disease, the WBC subsets generally decreased with increasing TCDD levels, but only for B lymphocytes was this tendency significant. Finally, Saberi Hosnijeh et al. (2013a) reported on the levels of interleukin 1 receptor agonist (IL1RA) as well as of the soluble forms of CD27 and CD30, immunomodulatory members of the TNF receptor superfamily. Here they found no association of TCDD level with CD27 or CD30; however, IL1RA was significantly decreased in those with higher TCDD levels after adjusting for concurrent chronic disease. This result is also consistent with a degree of immune system impairment being associated with high exposure to TCDD.
Several additional studies focused on immunological changes after exposure to the COIs in environmental settings. Spector et al. (2014) assessed immune
function in 109 postmenopausal women who participated in a year-long study of exercise and health in Seattle. Blood samples gathered at baseline and at 1 year were analyzed for levels of dioxin-like PCBs and tested for NK cell cytotoxicity and for phytohemagglutinin-induced T-lymphocyte proliferation. At baseline, the concentration of mono-ortho PCBs 105, 118, and 156 were not associated with lymphocyte proliferation; after a year, however, a decrease with increased levels of this set of PCBs (p = 0.039) was observed. No association of NK cytotoxicity with PCB levels was observed.
In a cross-sectional study of 1,063 men and 1,201 women living throughout Japan (who had not been occupationally exposed to dioxins), from 2002 to 2010, Nakamoto et al. (2013) gathered fasting blood samples for an assessment of environmental exposure to dioxin-like compounds (DLCs). Blood levels and corresponding TEQs were determined for dioxin-like polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), and PCBs. A logistic regression that adjusted for age, sex, smoking habit, drinking habit, regional block, and survey year estimated the odds of self-reported histories of several allergic-like diseases by quartiles (picogram per gram [pg/g] lipid) for PCDDs/PCDFs, for PCBs, and for all DLCs. Self-reported asthma was not associated with any of these three groupings, while atopic dermatitis showed a marginally significant trend over quartiles for PCDDs/PCDFs (p = 0.04) and for all DLCs (p = 0.02), and allergic rhinitis did for PCDDs/PCDFs (p = 0.04), but not for all DLCs (p = 0.14). These results showed an association with exposure to dioxin and DLCs and a significantly decreased incidence of reported atopic dermatitis.
One study examined the link between environmental exposure to the COIs and autoimmunity. In 632 women and 670 men who participated in NHANES in 2003–2004, Gallagher et al. (2013) investigated the association of either non–dioxin-like or and dioxin-like PCBs and the levels of serum antinuclear antibodies (ANA), which are associated with autoimmune disorders. Among the women, after adjustments for mercury blood level, race, menopausal status, diet, and BMI, total TEQs for PCBs were significantly associated with positivity for ANA (intensity ≥ 3) for each higher quartile compared with the lowest and for overall trend (p < 0.001). This pattern was observed neither in women for non-dioxin–like PCBs nor in men for either type of PCBs.
Turunen et al. (2012) derived total TEQs for 17 PCDD/F and 37 PCB congeners in blood samples from 123 men and 132 women from a population with high fish consumption and analyzed their relationship with C-reactive protein (CRP), an indicator of inflammation. No evidence of a trend across the tertiles of overall TEQ concentration was seen for either men (0.29) or women (p = 0.94).
There is an extensive body of evidence from experimental studies in animal-model systems indicating that TCDD, other dioxins, and several DLCs are
immunotoxic (Kerkvliet, 2009, 2012). Studies in laboratory mice show that the immunotoxicity of TCDD and DLCs depends on activation of the aryl hydrocarbon receptor (AHR). As most of the cell types involved in the immune system express the AHR, there are many potential pathways to immunotoxicity. TCDD-induced immunotoxicity is due primarily to changes in adaptive immune responses resulting in the suppression of both antibody-mediated and cell-mediated immunity. Dioxin and other AHR agonists may also reduce the clearance of infections and promote tumor growth through alterations in immune function. TCDD exposure alters macrophages and neutrophils such that it exacerbates some types of inflammation during infections and may contribute to the development of chronic inflammatory lung disease (Teske et al., 2005; Wong PS et al., 2010). Although there are many examples of dioxin and DLCs having immunosuppressive effects, these compounds also appear to influence autoimmune diseases, which are viewed as an inappropriate increase in immune function. Therefore, these compounds may be best described as immunomodulatory. Although the mechanisms of this immunomodulatory effect are not entirely clear, recently the tryptophan catabolite, kynurenine, was recognized as a ligand for the AHR and shown to function as a suppressor of allogenic T-cell proliferation (Opitz et al., 2011), providing a direct link between the AHR pathway and normal immune function.
TCDD has been shown to be a potent immunosuppressive chemical in laboratory animals and cell culture models. The relative potencies of given DLCs based on induction of hepatic enzymes—their toxicity equivalence factors (TEFs)—appear to predict the degree of immunosuppression induced (Smialowicz et al., 2008). The exposure of animals to dioxin not only suppresses some adaptive immune responses, but also has been shown to increase the incidence and severity of various infectious diseases and to increase the development of cancers (Choi et al., 2003; Elizondo et al., 2011; Fiorito et al., 2010, 2011; Head and Lawrence, 2009; Jin et al., 2010; Sanchez et al., 2010). It is consistent with its immunosuppressive effects that TCDD exposure suppresses the allergic immune response of rodents; this in turn results in decreased allergen-associated pathologic lung conditions and has been shown to suppress the development of experimental autoimmune disease (Quintana et al., 2008), to induce the suppression of autoimmune uveoretinitis (Zhang L et al., 2010), and to affect colitis (Takamura et al., 2011), arthritis (Nakahama et al., 2011), and inflammatory lung diseases, such as silicosis (Beamer et al., 2012).
Some current reports indicate that the AHR pathway plays an integral role in B-cell maturation, and that TCDD and DLC exposure may alter the B-cell and result in critical changes in the immune response (Baba et al., 2012; Sibilano et al., 2012; Simones and Shephard, 2011; Singh et al., 2011). Working with human B cells in vitro, Allan and Sherr (2010) demonstrated a new AHR-dependent mechanism by which exposure to environmental polycyclic aromatic hydrocarbons could suppress humoral immunity by blocking differentiation of B cells
into plasma cells. Recently, this finding was confirmed by data from human hemopoietic stem cells (HSCs) and knockout AHR mouse models showing that the AHR is critical in HSC maturation and differentiation (Fracchiolla et al., 2011; Singh et al., 2011a). Recently, using a novel pluripotent stem cell–based culture system, Smith et al. (2013) demonstrated that AHR expression and activity can direct human hematopoietic progenitor cell proliferation and differentiation. These data show that pluripotent hematopoietic human cells express AHR and that AHR agonists enhance erythroid differentiation, whereas antagonism of AHR favors the expansion of megakaryocyte cells. This finding is supportive of previous work indicating that B-cell activation results in increased AHR expression and that exposure of B-cells to B[a]P suppresses B-cell differentiation (Allan and Sherr, 2010). Lu H et al. (2010) demonstrated that although human B cells appeared less responsive to TCDD in increasing expression of AHR battery genes, TCDD’s ability to decrease IgM production was similar in both mouse and human B cells. Data from Zhang et al. (2013) suggest that this decrease in IgM production is the result of a TCDD-mediated decrease in B-cell terminal differentiation, resulting in fewer IgM-producing cells. TCDD alters not only HSC maturation but also alters proliferation and migration in vivo and in vitro (Casado et al., 2011), which indicates that exposure may have multiple effects on immune-cell function.
Cellular immunity, mediated by the thymus and T cells, is also a target of TCDD/dioxin exposure and the AHR pathway. Early evidence indicated that dioxin and DLC alter cellular immunity, because it was observed that exposure to the chemicals resulted in thymic involution and suppressed cytotoxic T-lymphocyte activity (Hanieh, 2014). Recent attention has focused on the ability of the AHR to induce regulatory T cells, or Tregs (Kerkvliet, 2012; Marshall and Kerkvliet, 2010). Tregs have potent suppressive activity in the immune system, and their inappropriate induction by TCDD could account for much of the immune suppression. AHR activation in dendritic cells has also been shown to promote the development of Tregs by inducing tryptophan metabolism. AHR activation in B cells can directly disrupt the production of antibodies (Sulentic and Kaminski, 2011). The recent demonstration that AHR activation by TCDD leads to the development of Tregs helps explain the diversity of effects seen after exposure to TCDD (Funatake et al., 2008; Kerkvliet, 2012; Marshall et al., 2008; Quintana et al., 2008; Stockinger et al., 2011; Yamamoto and Shlomchik, 2010).
One ultimate effect of dysregulation of the immune system is an alteration of autoimmunity. Data from animal models and cell culture indicate that exposure to dioxin and DLCs alters the development of autoimmune disorders. For example, antagonism of the AHR represses the expression of cytokines and chemokines in primary human synovial fibroblasts (Lahoti et al., 2013), indicating a potential contribution to the inflammatory process of rheumatoid arthritis. Nguyen et al. (2013) have hypothesized that may occur when AHR stimulation of IL-17 production in Th17 cells overwhelms the immune suppressive effects of inhibition
of Treg differentiation. TCDD has also been shown to induce apoptosis in rabbit chondrocytes, which supports a potential role of TCDD in contributing in a novel way to arthritis (Yang and Lee, 2010). Exposure to TCDD was also shown to induce the reactivation of the latent form of the Epstein barr virus (EBV) in 19 patients with Sjogren’s syndrome, an autoimmune disease, when compared to activation in 19 healthy patients (Inoue et al., 2012). Furthermore, a study of 18 people who had allergic asthma, 17 people whose asthma was controlled, and 12 controls showed that the plasma concentrations of IL-22 and the expression of the AHR in peripheral blood mononuclear cells was associated with the severity of allergic asthma; this finding strengthened the possibility that the AHR is involved in allergic asthma, thereby implying a role for dioxin exposure in this condition (Zhu et al., 2011). Thus, depending on the disease, TCDD exposure could exacerbate or ameliorate symptoms.
Very few studies of humans and exposure to the COIs have addressed outcomes that would be considered disease states primarily due to perturbation of immune function. In the 30-year follow-up study of the Vietnam Experience Study, Boehmer et al. (2004) found no suggestion of excess deaths that could be attributed to immune-system disorders (risk ratio [RR] = 1.32, 95% confidence interval [CI] 0.50–3.47, for International Classification of Diseases, Revision 9 [ICD-9] 240–279, which covers endocrine, nutritional, metabolic, and immunity disorders).
One would expect exposure to substantial doses of TCDD to result in immune suppression in Vietnam veterans. However, several studies of various measures of human immune function failed to reveal consistent correlations with TCDD exposure, probably because the exposures were inadequate to produce immune suppression or because the characteristics measured were not among those most relevant with respect to biologic plausibility. No clear pattern of an increase in infectious disease has been documented in the studies of veterans exposed to TCDD or to the herbicides used in Vietnam. However, three occupational-exposure studies provide some support for the idea that exposure to TCDD may result in an altered immune response to some exposures and an increased frequency of infections. The study of a single highly exposed person (Brembilla et al., 2011) confirmed TCDD-associated changes in immune measures that may not be applicable to people whose exposure was considerably lower. Immune alteration and the frequency and duration of specific types of infections should therefore be a focus of future studies. Suppression of the immune response by TCDD might
increase the risk of some kinds of cancer in Vietnam veterans, but there is no evidence to support such an association.
Allergic and Autoimmune Diseases
Epidemiologic studies have been inconsistent with regard to TCDD’s influence on IgE production in humans. No human studies have specifically addressed the influence of TCDD on autoimmune disease, but several animal studies have shown that TCDD suppresses the development of autoimmune diseases. The study of people who had allergic asthma or controlled asthma strengthened the data and suggested that the AHR (and thus dioxin exposure) is involved in the disease (Zhu et al., 2011). More studies are needed to determine the mechanism of TCDD-induced allergic and autoimmune disease, including rheumatoid arthritis.
Few effects of phenoxy herbicide or cacodylic acid exposure on the immune system have been reported in animals or humans, and no clear association between such exposure and autoimmune or allergic disease has been found. The exposure of laboratory animals to phenoxy herbicides or cacodylic acid has not been associated with immunotoxicity.
Lee et al. (2007a) found a significant association between concentrations of dioxin-like PCBs and the prevalence of arthritis in women, but not in men. There is no experimental evidence to support that finding, but increased inflammatory responses could be involved. There are no other human data on the potential for dioxin or the herbicides of interest to induce dysregulation of inflammation that could contribute to an increased risk of inflammation-associated diseases.
Possible associations involving infectious or inflammation-related diseases should be a focus for the future. Examples of earlier studies whose results support the occurrence of such adverse outcomes are Baccarelli et al. (2002), Baranska et al. (2008), Beseler et al. (2008), Oh et al. (2005), O’Toole et al. (2009), Tonn et al. (1996), and Visintainer et al. (1995).
On the basis of the evidence reviewed here and in previous VAO reports, the present committee concludes that there is inadequate or insufficient evidence to determine whether there is an association between exposure to the COIs and any specific diseases involving immune suppression, allergy, autoimmunity, or inflammation.
Animal studies and in vitro studies with human cells and cell lines are important ways of trying to understand the underlying biologic mechanisms associated with immunotoxic and other responses to xenobiotics (“foreign” substances that do not normally occur in biologic systems). However, as discussed above, despite the vast array of data supporting the immunotoxicity of TCDD in laboratory animals, little evidence from studies of Vietnam veterans or other human populations suggests that exposure to TCDD or the herbicides of concern produce immune alterations that have directly observable and predictable functional consequences. Many factors must be considered in examining the relevance of animal and in vitro studies to human disease and disease progression, and they are discussed in Chapter 4. Here, we present the factors that are important in considering differences between the results of laboratory studies and the findings of observational epidemiologic studies.
Magnitude and Timing of Exposure
In general, the TCDD exposures used in animal studies have been orders of magnitude higher than the exposures that Vietnam veterans are likely to have received during military service. It is well known that the immune system is highly susceptible to xenobiotic exposure during critical stages of development, such as gestation, and that primary immune responses are easier to alter than secondary immune responses. In vivo studies show that exposure to antigens may be important, so the timing of antigen exposure relative to TCDD exposures may be an important variable.
Human immune diseases are likely to have complex etiologies and to be under the influence of numerous genes and gene–environment interactions (Dietert et al., 2010). Differences in AHR affinity between species may be a factor in animal-to-human extrapolation. For example, many strains of mice (AHRb) are known to exhibit greater susceptibility of CYP1A1 induction and immune suppression than other strains (AHRd). In contrast, a simple single-haplotype difference in susceptibility to TCDD has not been observed in humans. Rats appear to be more similar to the resistant AHRd phenotype of mice in their sensitivity to TCDD. Indeed, it is difficult to produce immune suppression in rats with TCDD because of that, and there probably are other genetic reasons as well.
There are well-known differences in susceptibility to xenobiotic exposures between male and female animals. There are probably multiple reasons for the differences, some of which may pertain to immunomodulation by sex steroids. Similarly, evidence suggests that specific immune-based health risks in humans have important sex differences. For example, women generally are much more susceptible than men to the development of several autoimmune diseases; such differences in humans may result from a combination of genetic factors and environmental exposures. One simple example of this is the fact that the gene associated with control (at least in part) of the T-regulatory immune cells (these cells can suppress some of the immune response) is located on the X chromosome. Hence, there are (at least conceptually) different mechanisms through which the immune system could be altered in men and women; the incomplete silencing of one X chromosome could alter the suppressive immune environment in women. This has ramifications for future studies. In considering the potential effects of the COIs on the immune system and the risk of disease, sex-based differences in chemically induced adverse immune outcomes need to be investigated. Future studies should ensure that—whether in animal models or in human studies—gene-specific or sex-specific immune effects are able to be evaluated with sufficient statistical power to support distinctions.
Stress is a well-known modifier of human immune responses. It is an ever-present variable that is difficult to assess or control for in epidemiologic studies. Stress, of course, is ever present in combatants and thus likely to play an important role in their immune response, which would be extremely difficult to estimate or to study.