Psychoneuroimmunology

Nicholas Cohen,¹ Howard Kehrl,² Birgitta Berglund,³ Ann O'Leary,⁴ Gerald Ross,⁵ James Seltzer,⁶ and Clifford Weisel⁷

¹ Department of Microbiology and Immunology, University of Rochester, Rochester, New York
² U.S. Environmental Protection Agency, Research Triangle Park, North Carolina
³ Department of Psychology, Stockholm University, Stockholm, Sweden
⁴ Department of Psychology, Rutgers, The State University of New Jersey, Piscataway, New Jersey
⁵ Environmental Health Center-Dallas, Dallas, Texas
^{6 6} Indoor Hygienic Technologies, San Diego, CA
⁷ EOHSI, Rutgers, The State University of New Jersey, Piscataway, New Jersey

• Introduction
• Experimental Questions Relevant to the Relationship between Exposure and Symptoms in Chemical Sensitivities
• Hypotheses
  • Subject Selection
  • Controlled Exposure Studies
  • Response Measures

Key words: immunology, multiple chemical sensitivities, stress, psychoneuroimmunology, inflammatory response

This paper is based on a work group discussion at the Conference on Experimental Approaches to Chemical Sensitivity held 20-22 September 1995 in Princeton, New Jersey. Manuscript received at EHP 12 August 1996; manuscript accepted 27 November 1996.

Abbreviations used: MCS, multiple chemical sensitivity; NK, natural killer; VOC, volatile organic compounds.

Introduction

Experimental Questions Relevant to the Relationship between Exposure and Symptoms in Chemical Sensitivities

• In situations in which a sensitizing chemical may be immunomodulatory, do environmental cues temporally associated with the sensitizing exposure serve as conditioned stimuli (4)? If so, can such cues, with or without subthreshold amounts of the chemical, elicit a chemical sensitivity response? Can the chemical sensitizer itself, provide both unconditioned and conditioned stimuli?

• Are there significant immunological, neuroendocrine, and psychosocial deviations from normalcy shortly before, at the time of, or consequent to, the development of MCS symptomatology? If physiological changes are noted consequent to psychosocial changes, are those psychosocial changes implicated in the development of MCS? If so, could psychological/behavioral intervention be of therapeutic value, at least in some subset of patients (3)?

• What is the interval between the initiating event and the development of early symptoms of MCS? What is the interval between the initial trigger and the so-called increase in sensitivities to diverse stimulants?

• Is the duration of these intervals related to psychosocial factors (e.g., stress) or to personality/sex/health status (e.g., allergic status, autoimmunity or autoimmune predisposition) of the subject, or the chemical nature of the initiator? Do events perceived as stressful in the recent or past history of the individual play roles in the onset and/or progression of MCS?

• What do chemical initiators of MCS and initiators of MCS such as a car accident or childbirth have in common? Which factor is more important in the traumatic initiation: the exposure to chemicals associated with the traumatic event, and/or the stressful experience associated with the event? Can a stressor precipitate MCS in a chemically sensitized individual who is not displaying overt symptoms of MCS at the time of stressor exposure?

• Why is there variability in the development and severity of MCS? Is there any relationship between major histocompatibility complex phenotype and MCS? Is there a detectable humoral or cellular immune response to the initiating chemical? Is an immune response causal or the result of the MCS process(es)?

Hypotheses

        Study 1: Population exposed to a chemical initiating event--for example, the installation of new carpets (including glue) in the workplace.

        Study 2: Population exposed to a man-made chemical spill.

        Study 3: Population exposed to a natural disaster such as an earthquake, flood, or hurricane.

Subject Selection
An assumption common to each study population is that some percentage of the individuals in each heterogeneous population will within a predetermined period (e.g., 1 year) develop MCS. These subjects will comprise the first study group. A second group will consist of those individuals who have been exposed to the same stimuli but do not develop MCS within the same predetermined time period. To determine whether individuals in this second group exhibit any immunological consequences of the exposure that might have a psychosocial attribution, a group of demographically matched individuals that have not been exposed could be studied. Given the assumption that the cost of performing these studies is of no consequence, all subjects in each study population will be evaluated, by health status, by a full battery of immunological tests. Some examples of immunologic assessments are serum immunoglobulin levels (all isotypes); complement levels; mitogen responses; TH1- and TH2-derived cytokine levels; proinflammatory cytokine (e.g., IL-1, IL-6, TNF-) levels; soluble and cell-bound cytokine receptors; natural killer (NK) cell numbers and activity; substance P and proinflammatory cytokines in nasal lavage; quantitative cytokine mRNA; antibody titers to volatile organic compounds (VOCs); antibodies to a benign antigen or flu vaccine (immunized during study); and skin testing with standard recall antigens and allergens. Psychometric instruments will also be used to evaluate such factors as stress, loneliness, coping, helplessness, and suggestibility. The following are examples of psychometric instruments: the Profile of Mood States questionnaire; the Minnesota Multiphasic Personality Inventory; the Reaction Scheme Test; and the State and Trait Anxiety Inventory. These procedures will be done before and periodically after (e.g., 1, 3, 6, 9, 12 months) the putative initiating event. Variables to be considered (and controlled for) as ways of separating (stratifying) the subject population, include but are not restricted to the typical demographic issues of age, sex, socioeconomic status, and ethnicity. The hypothesis to be tested is that: some psychosocial factor(s) will distinguish individuals who develop MCS from those who do not.

Total VOC concentration and selected VOCs will be measured in the air and biological fluids (blood) of the subjects to determine the actual exposure received in study 1. Estimates of the exposures in study 2 would rely on measurements in biological fluid (blood or urine) of the suspected compounds of exposure or their metabolites. Consideration must be given to the time interval between when the exposure occurred and when the biological sample was collected and the biological half-life of the species measured.

Controlled Exposure Studies
There are multiple design problems inherent in conducting clinical studies of MCS patients under controlled exposure conditions. These include selection of the appropriate exposure conditions such as routes of delivery, issues regarding blinding or masking, and identification and utilization of relevant response measures. With respect to relevant response measures, most MCS patients report acute symptoms that occur within minutes to hours of the exposure. The time course of this symptom-exposure relationship makes mechanisms involving neoantibody production highly suspect and focuses attention on local inflammatory responses; the production/release of proinfammatory cytokines such as IL-1, IL-6, and TNF-; the release of acute phase proteins; and immediate hypersensitivity reactions.

One design problem associated with these studies is selecting the time after the initiating event when individuals who develop MCS in the study populations should undergo controlled exposure testing. This could be a fixed time (e.g., 1 year) after the initiating event or it could be a predetermined time after development of MCS that is independent of the time of the initiating event. If the latter time is used, then selection of the individual from the exposed group who does not develop MCS at the time of a controlled exposure becomes a problem (i.e., that individual could, in principle, still develop MCS after the predetermined time).

There are several approaches to performing controlled exposure studies that may be applicable to the MCS question. Perhaps the most valid approach to controlled subject testing is a double-blind study design in which the subject is unable, from usual sensory perceptions, to detect the exposure condition. If the controlled exposure is conducted within a longitudinal study as described, it would be useful to compare the psychosocial variables collected at initiation of the studies for those subjects with high and low hit rates for detecting chemicals in subthreshold concentrations.

Although olfactory maskers were successfully used in one study, the odors and irritant effects of volatile organic chemicals make blinding extremely difficult. One can consider using subthreshold exposure levels where individual subject perceptual (olfactory, sensory irritation) thresholds are determined and exposure conditions are set at some magnitude below this level. However, in considering the complexities and controversies of MCS, such studies likely would come under intense criticism if performed without allowing for the adaptation/masking phenomena described by MCS proponents. Another approach in reevaluating the MCS phenomenon is to utilize above-threshold exposures in an attempt to identify physiological or biochemical parameters that provide biologic plausibility for the reported symptoms; such work would be exploratory. Using recognizable above-threshold exposures would not be blinded and would not differentiate between conditioning and other underlying mechanisms. However, such an approach possibly would provide a measure other than symptoms that could then be examined further with regard to dose-response, routes of exposure, and blinding. This approach would require studying subjects who report the symptoms of interest.

Response Measures
In addition to measuring symptoms of MCS, preference should be given to measures that can be made during or in close proximity to exposures. The measures selected must have biologic plausibility for the symptoms reported by the MCS patients. Most MCS patients report acute symptoms that occur within minutes to hours following an exposure. The time course of this symptom-exposure relationship makes mechanisms that involve antibody production or cell-mediated immunity highly suspect and focuses attention on local (e.g., nasal) inflammatory responses and the local and systemic production of proinflammatory cytokines such as IL-1, IL-6 and TNF- measured pre- and postexposure. Additional analyses of endocrine activation (stress hormones such as salivary cortisol) and autonomic nervous system arousal (heart rate, skin conductance, pupilography, catecholamines) before and after deliberate exposure should also be conducted. In summary, the following should be examined during or in close proximity to challenges: systemic inflammatory markers (e.g., proinfammatory cytokines) (relevant symptoms: fatigue, arthralgias, myalgias); nasal inflammatory markers--neuropeptides, proinflammatory cytokines, histamine/tryptase, cells, albumin (relevant symptoms: nasal symptoms); autonomic nervous system--heart rate, skin conductance, pupilography (relevant symptoms: dizziness, palpitations, flushing, cold extremities, anxiety); endocrine system--stress hormones (e.g., salivary cortisol, prolactin); stress measures--heart rate, respiratory rate, end-tidal CO₂ (relevant symptoms: anxiety, palpitations, dyspnea).

Psychoneuroimmunological Consequences of MCS
Regardless of whether there is a psychoneuroimmunological component to the development of MCS, it is reasonable to hypothesize that there are behavioral and immunological consequences of MCS that may be attributable to the stress of a chronic disability (3). Thus, it might prove useful to longitudinally follow individuals who develop MCS and periodically monitor their psychosocial status, immune function, and general health. An important related question, at least from the perspective of the psychoneuroimmunologist, is whether abnormalities in any of these parameters are associated, at least correlationally, with an altered incidence of microbial diseases, autoimmune diseases, and cancer. If so, perhaps appropriate psychosocial interventions can be developed. In terms of intervention, it would be informative to determine the extent to which MCS patients are consumers of alternative (complementary) medical therapies, and whether such treatments have beneficial effects.

4. Ader R, Cohen N. Psychoneuroimmunology: conditioning and stress. Ann Rev Psychol 44:53-85 (1993).