In this issue, Weyand and colleagues [3] report a study designed to determine whether specific cytokine patterns distinguish giant cell arteritis from polymyalgia rheumatica. That the research team found evidence of macrophage and T-cell activation in giant cell arteritis is neither new nor unexpected. However, the investigators were able to distinguish between temporal artery biopsy specimens from patients with polymyalgia rheumatica and those from patients with giant cell arteritis based on their lymphokine profiles. This is an important new observation. Further, based on the patterns of T-cell derived cytokines, the authors postulated that TH1 cells are critically involved in the granulomatous infiltrate of giant cell arteritis. This conclusion is based on the finding that interferon- is present only in patients with giant cell arteritis (albeit in only 67% of the samples). Pathologic studies also found that the macrophage component of the inflammatory response is shared between polymyalgia rheumatica and giant cell arteritis and that tissue samples from patients with polymyalgia rheumatica without histologic evidence of inflammation contain proinflammatory cytokines. These are also important observations.

The sensitivity of the polymerase chain reaction is revolutionizing our understanding of disease. This technology can detect messenger RNA for a specific product from a single cell. By applying such a sensitive test, Weyand and colleagues provide support for the contention that giant cell arteritis and polymyalgia rheumatica may be a continuum of the same disease process. This contention has already been substantiated by clinical descriptions and laboratory comparisons [4].

These findings can be considered in light of our current understanding of the biological effects of cytokines. Cytokines have long been recognized as being involved in a host of normal and disease processes. These proteins regulate virtually all of the important biological processes such as cell growth, cell activation, inflammation, immunity, tissue repair, fibrosis, and morphogenesis. The families of cytokines are subdivided into groups, such as interferons, interleukins, and tumor necrosis factor, based primarily on their degree of homology [5]. Cytokines may be classified as proinflammatory (for example, interferon-) or anti-inflammatory (for example, tumor growth factor). The presence or absence of a specific cytokine in a given disease may allow one to draw inferences about the cause of the process when cell sources and main target cells of these cytokines have been identified. Subpopulations of T lymphocytes can be distinguished on the basis of the cytokines that they produce or to which they respond [5]. These populations have been divided into three major divisions.

1. TH1 cells that produce interleukin-2 and interferon-. These cells are believed to be of primary importance in mediating cellular immunity and inducing autoimmune reactions.

2. TH2 cells that produce interleukin-4, interleukin-5, interleukin-10, and granulocyte-macrophage-colony stimulating factor. These cells are believed to be of primary importance in regulating the pathogenic effect of TH1 cells and in controlling the synthesis of IgE and a substantial percentage of IgG.

3. TH0 cells that produce both patterns of cytokines are believed to be precursors of TH1 and TH2 cells.

Does giant cell arteritis differ from polymyalgia rheumatica because of more effective activation of immunopathogenic cells or a failure to regulate the activities of these cells? This is the critical enigma of autoimmune disease. Vasculitic lesions may be produced by interactions of subpopulations of T cells (TH1 and TH2 cells) and the inflammatory cytokines that they produce [6]. The finding by Weyand and colleagues [3] of TH1 cells in the inflammatory lesions of giant cell arteritis may suggest that these cells are responsible for the generation of the lesions. However, several confounding factors exist. For example, what was the original stimulus for the attraction of T cells to the site of the lesions? What antigens might have been responsible for attracting T cells? Because T cells can have different properties at different states of activation, what is the state of activation of T cells within the lesions? For example, do T cells express the high-affinity interleukin-2 receptor, which indicates T-cell activation [7]? Were other indicators of T-cell activation, such as HLA-DR, transferrin receptors, and interleukin-2 receptors [8, 9], present to allow one to infer that T cells are responsible for generation of the lesion? Were co-signals of activation present such as the B7-CD28 signal [10] and the appropriate adhesion molecules that localize the topography and regulate the intensity of the response [11]? If not, T cells may not be responsible for generating the lesion. None of these factors was addressed in the study by Weyand and coworkers. They must be elucidated to understand the dynamic generation of the inflammatory lesions.

What is the event that incites activation from polymyalgia rheumatica to full-blown giant cell arteritis? Is there a failure in primary activation of immunopathogenic cells in giant cell arteritis? Alternatively, do giant cell arteritis and polymyalgia rheumatica differ because of less effective immunoregulation of immunopathogenic cells in giant cell arteritis? Several investigators have shown monocyte-derived macrophages and CD4⁺ T cells in giant cell arteritis lesions by immunohistochemical techniques [4, 8, 11-15]. Granulomatous vasculitic lesions are regulated by interactions of TH1 and TH2 cells [5, 16]. The orderly transition from TH1 to TH2 cell dominance has been shown not only at the level of the intact host but also within the individual evolving granuloma [17]. This switch to TH2 dominance may lead to suppression of TH1 responses at the systemic level [18]. TH2 cells may suppress TH1 responses through interleukin-10 production [19]. Conversely, TH1 cells produce interferon- that suppresses TH2 cell development [6]. Thus, finding relatively high levels of interferon- in the vasculitic lesions of patients with giant cell arteritis may explain the failure to observe substantial numbers of TH2 cells [15] or TH2 cell-derived cytokines in giant cell arteritis. Other regulatory events may include antigen-driven apoptosis (cell death) [20] and production of suppressive T cells or the soluble factors that they produce [21].

In this context, it is attractive to speculate that the granulomatous lesion may be a localized immunoregulatory organelle [16]. Initially, it is the nidus of immunologic interactions and serves to support local and systemic sensitization. Later, it serves to entrap immunologically competent cells, perhaps generated systemically, and to inactivate these cells within the lesions, thus limiting the pathologic response.

Humoral factors may also be important in vasculitis. Previous studies have implicated complement activation in giant cell arteritis and have also shown the absence of TH2 cells, which are required for antibody synthesis. Arterial specimens from patients with giant cell arteritis show evidence of classic complement, alternative complement, and lytic complex activation, whereas specimens from patients with polymyalgia rheumatica show classic complement and lytic complement activation but no evidence of alternative pathway activation. Therefore, arterial wall destruction may be arrested at the lytic complement phase [15]. The interplay of cellular and humoral factors remains enigmatic.

The study by Weyand and colleagues [3] raises more questions than it answers. It serves to redefine the level of our ignorance and whet our appetites for future, more definitive studies. Their findings spark our interest in the role of lymphokines in inflammatory lesions such as giant cell arteritis, but we still need to identify the inciting event, to clarify the pathogenic mechanism of giant cell arteritis, and to define the relation between giant cell arteritis and polymyalgia rheumatica.