T lymphocyte subsets in inflammatory bowel disease: peripheral blood

Peripheral blood T lymphocytes and T lymphocyte subsets have been quantified in 28 patients with ulcerative colitis and 26 with Crohn's disease by an indirect immunofluorescence technique using monoclonal antibodies: OKT3, which detects all peripheral blood T lymphocytes; OKT4 (T cells of helper phenotype); and OKT8 (T cells of supressor-cytotoxic phenotype). Eighteen normal subjects and 16 patients with a variety of non-inflammatory gastrointestinal disorders were studied as controls. No significant differences were found between patient and control groups in the proportions of circulating T lymphocytes or their subsets. When compared with normal subjects, absolute numbers of T lymphocytes were reduced in patients with active ulcerative colitis or Crohn's disease (p less than 0.05). OKT4+ T cell numbers were reduced in ulcerative colitis, whether active (p less than 0.02) or inactive (p less than 0.05) and in active Crohn's disease (p less than 0.05) Numbers of OKT8+ T cells were reduced in active Crohn's disease (p less than 0.01). There were no differences in T lymphocyte numbers between the patient groups and the disease control subjects. The OKT4+:OKT8+ ratio in patients with inflammatory bowel disease did not differ from that in controls. No relation was found between any of the parameters studied and disease activity, site, or extent of disease, or treatment with sulphasalazine or corticosteroids. The presence of Ia-like, HLA-DR antigens on T cells was detected using a double marker immunofluorescence technique. In control subjects up to 7% of OKT3+ cells were HLA-DR+. In only three patients was the proportion of HLA-DR+ cells greater than in controls. These results indicate that the pathogenesis of ulcerative colitis or Crohn's disease does not depend upon an alteration in the proportion of circulating T lymphocytes nor upon an imbalance of T lymphocyte subsets as defined by monoclonal antibodies. The reduction in T lymphocyte numbers may result from mucosal infiltration. The findings also suggest that circulating T lymphocytes are not activated.     

Decreased level of suppressor/cytotoxic T cells (OKT8+) in polymyalgia rheumatica and temporal arteritis: relation to disease activity

Separated peripheral blood mononuclear cells were analyzed by fluorescent microscopy with monoclonal antisera for T cells (OKT3+), helper/inducer T cells (OKT4+) and suppressor/cytotoxic T cells (OKT8+). Thirty-seven patients with polymyalgia rheumatica (PMR), 13 of whom had positive biopsies for arteritis, were studied and compared with 25 age and sex matched normal subjects. The percentages of OKT3+ and OKT4+ T cells were similar in the PMR group and controls, but percentage of OKT8+ T cells was significantly reduced in patients (14.8 +/- 6.8) compared with controls (22.1 +/- 6.3). Values of OKT8+ T cells were extremely low in untreated patients with active, acute disease (7.8 +/- 4.4%) and significantly lower than in prednisone treated patients with inactive disease (17.3 +/- 5.9). These findings indicate that low values of circulating OKT8+ T cells is a feature of PMR and is related to disease activity.     

Phenotype study of blood T lymphocytes in alcoholic hepatopathies

Peripheral blood T lymphocytes and T lymphocytes subsets have been quantified by an indirect immunofluorescence technique using monoclonal antibodies, in 10 patients with fatty liver, 8 with acute alcoholic hepatitis (AAH), 10 with inactive cirrhosis and 7 with cirrhosis and AAH. Twenty normal subjects were studied as controls. As compared to controls (1.81 +/- 0.56 10(9)/l), we found a reduced number of peripheral T lymphocytes (OKT3+) in patients with inactive cirrhosis (0.98 +/- 0.45, p less than 0.001) and in patients with cirrhosis and AAH (1.22 +/- 0.51, p less than 0.02). The OKT4 to OKT8 ratio was normal in patients with fatty liver or inactive cirrhosis, but it was significantly higher in patients with AAH with or without cirrhosis (2.83 +/- 0.79, p less than 0.01, and 2.10 +/- 0.56, p less than 0,02, respectively) than in controls (1.68 +/- 0.24). In both groups, this increased ratio was due to a decreased proportion of OKT8+ circulating lymphocytes (19.2 +/- 6.7 p. 100, p less than 0.01, and 21.8 +/- 4.6 p. 100, p less than 0.02, respectively) when compared to controls (27.1 +/- 4.1 p. 100). The T-cell imbalance observed in patients with liver cell necrosis may be of importance in the pathogenesis of alcoholic liver disease.     

T-lymphocyte subpopulations in rheumatoid arthritis. II. Definition by monoclonal antibodies

Samples of peripheral blood of 27 patients with seropositive rheumatoid arthritis (RA) and 27 healthy age- and sex-matched controls were coded and studied for lymphocyte subpopulations. Monoclonal antibodies and indirect immunofluorescence were used to analyse subpopulations of purified T cells: OKT3 reacts selectively with all peripheral human T cells, OKT4 defines the helper/inducer subpopulation and OKT8 the suppressor/cytotoxic T cells. RA patients had a significantly lower relative lymphocyte count (p less than 0.001). whereas the percentage of all T cells was similar in patients and controls, RA patients had a significantly higher proportion of helper/inducer cells (62 +/- 1.3 vs. 56 +/- 1.0, p less than 0.005) and significantly decreased suppressors/cytotoxic cells (25 +/- 1.5 vs. 32 +/- 1.0; p less than 0.001). This resulted in an increased "immunoregulatory ratio" of helper to suppressor cells in RA patients (2.68 +/- 0.17) compared to their normal controls (1.83 +/- 0.10; p less than 0.001). The proportions of T cells expressing HLA-D-like antigens (defined by a monoclonal antibody to human Ia) was not different in patients and controls. These findings confirm conclusions derived from our previous study of T cell subsets defined by the expression of Fe-receptors (Tm and Tg), that in the peripheral blood of patients with RA, helper mechanisms predominate over suppressor mechanisms. This derangement of the immunoregulatory balance may have an important role in the pathogenesis of seropositive RA.     

Abnormality of T cell subsets in patients with idiopathic thrombocytopenic purpura

Fifty patients with idiopathic thrombocytopenic purpura (ITP) and 50 normal controls were studied for T cell subsets using OKT monoclonal antibodies. These patients were found to have lower proportions of OKT4+ subset and higher proportions of OKT8+ ratio than the healthy controls. The perturbation was more severe in patients with active disease. A significant increase of OKT 4+8+ double-labelled cells deduced by a subtraction method was also observed in the peripheral blood of patients and an inverse correlation was demonstrated between the proportions of OKT4+8+ subset and the platelet counts. Furthermore, a positive correlation between the proportions of OKT4-8+ subset and the platelet counts was also found in patients with active ITP. These data suggest that abnormalities of T cell subsets in the peripheral blood may play an important role in the pathogenesis of ITP.     

Peripheral T-lymphocyte subsets changes in patients with endocrine and metabolic diseases

In this study the authors utilized the OKT monoclonal anti-human T lymphocyte antibodies to determine the normal level of T lymphocyte subsets in the peripheral blood of 30 healthy volunteers, 36 patients with Graves' disease (GD), 4 with hypothyroidism and 30 with diabetes mellitus (DM). It is shown that OKT+3, OKT+4 OKT+8 and OKT+4/OKT+8 ratio of T-lymphocyte subsets in the peripheral blood of 30 healthy volunteers were 72.4 +/- 5.1%, 38.9 +/- 5.2%, 26.8 +/- 4.3% and 1.5 +/- 0.3. In 36 patients with GD OKT+3 was 67.6 +/- 5.8%, OKT+4 was 33.8 +/- 6.6% and OKT+8 was 17.9 +/- 6.1%. These were all significantly lower than those of normal controls (P less than 0.001). OKT+4/OKT+8 ratio was 2.1 +/- 0.6, which was significantly higher (P less than 0.001). The value of OKT+4 and OKT+8 in 10 patients with GD after treatment were markedly elevated. However, the ratio of OKT+4/OKT+8 was significantly decreased (P less than 0.05). OKT+3 of 4 cases of hypothyroidism was 69.7 +/- 3.35%, being similar to that of normal controls (P greater than 0.2). OKT+4 and OKT+8 were 31.73 +/- 4.99% and 18 +/- 2.94% respectively, both of which being markedly decreased (P less than 0.01). The ratio of OKT+4/OKT+8 was 1.97 +/- 0.22, being not significantly elevated (P greater than 0.05). OKT+3, OKT+4 and OKT+8 of 23 cases with DM were 68.8 +/- 7.2% (P less than 0.05), 33 +/- 8.1% (P less than 0.005), and 21.2 +/- 6.5% (P less than 0.001) respectively. All of these were significantly decreased. The ratio of OKT+4/OKT+8 was 1.6 +/- 0.5, being not significantly changed.     

Cytochemistry of normal lymphocyte subsets defined by monoclonal antibodies and immunocolloidal gold

The cytochemical reactivities of 3 acid hydrolases, alpha-naphthyl acetate esterase (ANAE), acid phosphatase and beta-glucuronidase were investigated in normal peripheral blood lymphocyte subsets defined by monoclonal antibodies OKT3, 4, 8 and FMC4 (anti-Ia). A combined monoclonal antibody-immunocolloidal gold/cytochemical staining procedure was used to determine enzyme activities and distributions of reaction product in each subset. Cytochemical profiles for each lymphocyte subset were defined. The majority (greater than 85%) of T cells (OKT3+) were positive for all 3 enzymes whereas a minority (less than 40%) of B cells (FMC4+) displayed reactivity. The cytochemical profiles of T helper/inducer (OKT4+) and T suppressor/cytotoxic (OKT8+) cells were not significantly different and corresponded to that observed for OKT3+ cells; thus none of these enzymes can be used to distinguish normal lymphocyte subsets cytochemically. ANAE reactions were further analysed, in the respective subsets, on the basis of dot-like or scattered/diffuse reactivity. The ratios of cells displaying dot-like: scattered/diffuse reactivity, in the respective subsets, were OKT3+, 5.4:1; OKT4+, 8.1:1; OKT8+, 2.4:1; FMC4+, 0.4:1. The cytochemical profiles and ANAE reactivities of T cell subsets identified by monoclonal antibodies differ from those displayed by T cell subsets defined by Fc receptors and confirms that there is little correlation between subsets defined by these two methods.     

Peripheral blood lymphocyte subsets in polymyalgia rheumatica

Summary Peripheral blood lymphocytes from 23 patients with polymyalgia rheumatica (PMR) were characterized using monoclonal antibodies and flow cytometry in a two-year prospective study. There were no significant differences in absolute numbers or relative percentages of lymphocytes or CD3+, CD4+, CD8+ T cells or the CD4+T cell functional subsets, virgin (CD4+CD45RA+) and memory (CD4+CD29+) T cells, in patients before or during corticosteroid treatment compared to controls. Previous reports on decreased levels of CD8+T cells as a characteristic of PMR/giant cell arteritis was not confirmed. The absolute number and relative percentage of lymphocytes with natural killer cell activity, CD16+ CD56+ cells, were significantly lower in patients with active untreated PMR as well as during corticosteriod treatment compared to controls, but at the two-year follow-up the difference was less marked.     

Immune complexes and autoantibodies in patients with giant cell arteritis and their relationship with autologous rosette-forming cells

Serologic studies and lymphocyte analysis were carried out in 29 patients with giant cell arteritis (GCA). IgA-containing circulating immune complexes (CIC) were detected in 16 GCA patients with or without polymyalgia rheumatica (55%). A significant difference was demonstrated in autologous rosette-forming cells between the patients as a whole, and matched controls (8.6 +/- 2.0 vs 11.6 +/- 2.4, p less than 0.001), and also between patients with and patients without CIC (7.9 +/- 1.6 vs 9.4 +/- 2.0, p less than 0.001).     

T lymphocyte subpopulations in chronic lymphocytic leukemia of B cell type in relation to immunoglobulin isotype(s) on the leukemic clone and to clinical features

Total blood T lymphocytes and subpopulations (OKT4+ and OKT8+ cells) were studied in 59 patients with B cell chronic lymphocytic leukemia (B-CLL). In 48 previously untreated patients, total T lymphocytes were higher as compared to healthy controls (p less than 0.001). T-cells and OKT8+ cells were significantly increased in patients in advanced clinical stage and with progressive disease in comparison to patients with low stage and indolent disease. High numbers of OKT8+ lymphocytes were also seen in patients with a dominance of mu heavy chains on the leukemic clone. Moreover, in this patient group the OKT8+ subpopulation correlated with total B cells (r = 0.68, p less than 0.001) while in patients with a mu delta phenotype no such correlation was seen. After successful cytostatic therapy there was a reduction in total numbers of both OKT4+ and OKT8+ cells, in particular, with a concomitant increase in OKT4/OKT8 ratios.     

Circulating T cell subtypes in polymyalgia rheumatica and giant cell arteritis: variation in the percentage of CD8+ cells with prednisolone treatment.

OBJECTIVES--Some reports have described a decreased percentage of circulating CD8+ cells in patients with polymyalgia rheumatica and giant cell arteritis (PMR/GCA) before treatment and persisting for some months during treatment with corticosteroids. Other studies have found no such changes. There are overt methodological variations between these studies and there may also hidden differences, such as the timing of blood samples. The purpose of this study was to investigate T cell subtypes in patients with PMR/GCA while controlling for variables known to affect T cells. METHODS--Circulating T cell subsets were measured in 36 patients with PMR/GCA before and during treatment with prednisolone. Blood samples during treatment were taken before the daily dose of prednisolone. The whole blood lysis method was used followed by flow cytometry. RESULTS--Compared with controls, CD8+ cells were not reduced before treatment in patients with PMR/GCA (0.44 x 10(9)/l; 28% of lymphocytes). CD4+ cells were also normal (0.78 x 10(9)/l; 48% of lymphocytes). During treatment with prednisolone total T cells increased from 1.18 to 1.59 x 10(9)/l and CD4+ cells increased from 0.78 to 1.05 x 10(9)/l. The percentage of CD8+ cells decreased on treatment from 28 to 25%. CONCLUSIONS--This study does not confirm the finding of some groups that the percentage of circulating CD8+ cells is reduced in patients with PMR/GCA before treatment. It does show that the percentage of CD8+ cells decreases during treatment with corticosteroids. This needs to be considered when designing studies of lymphocyte subsets in diseases treated with corticosteroids.     

Peripheral blood lymphocyte subpopulations in polycythaemia and thrombocythaemia

Lymphocyte subpopulations from peripheral blood of normal subjects and patients with primary proliferative polycythaemia (PPP), idiopathic erythrocytosis (IE) and essential thrombocythaemia (ET) were separated using antihuman immunoglobulin antiserum for B lymphocytes and the following monoclonal antibodies: OKT3, directed against the general T-lymphocyte subpopulation, OKT4 and OKT8, detecting respectively T-helper and T-suppressor lymphocyte subpopulations, OKM1 reacting mainly with monocytes. A decrease in the number of OKT3+ cells was observed both in PPP and IE, with a particular fall of the OKT8+ (suppressor) cells, so that the T4/T8 ratio was significantly increased (P less than 0.03 in PPP and P less than 0.0005 in IE). The ratio remained normal in samples from ET. OKM1+ cells were significantly increased in PPP (P less than 0.04), but not in IE, while in ET there was a rise in a few cases only. The present data point out some definite changes in the circulating lymphomonocytic cell subsets, which may be of interest in the study of this group of myeloproliferative disorders.     

Selective T cell receptor decrease in peripheral blood T lymphocytes of patients with polymyalgia rheumatica and giant cell arteritis

OBJECTIVES: To investigate the phenotype and T cell receptor (TCR) use in peripheral blood T cells in patients with polymyalgia rheumatica (PMR) and giant cell arteritis (GCA). METHODS: Circulating T lymphocyte phenotype and TCR repertoire were studied by flow cytometry using specific monoclonal antibodies in 23 healthy controls and 37 patients with PMR/GCA. RESULTS: Patients with active PMR/GCA showed an inverse relation between naive and memory CD4+ T cells and unchanged expression of activation surface markers compared with controls. CD4+ TCR BV expansions were seen in 12 (52%) controls and in 8 (22%) patients with active disease (p = 0.03). Within the CD8+ subset, the frequency of expansions was similar between groups. Most T cell expansions remained stable over time. Seventeen of the 23 patients with active PMR/GCA disclosed a simultaneous CD4+ and CD8+ T cell depletion for at least one particular BV family with a clear predominance of BV5S2/S3. CONCLUSIONS: The phenotype of circulating T cells in patients with PMR/GCA is similar to that found in aged healthy subjects, except for the surface markers of naive and memory cells and a striking non-activated phenotype. Specific BV expansions in CD4+ and CD8+ T cells, which remain stable over time, are frequent in aged subjects, including patients with PMR/GCA. TCR BV changes in patients with active disease seem to be also age related, except for the significant decrease in certain BV families in both CD4+ and CD8+ T cell subsets, which may favour the participation of a superantigen stimulation in PMR/GCA.     

Peripheral blood T lymphocyte subpopulations defined by monoclonal antibodies in rheumatoid arthritis: distribution in patients untreated and treated by oral gold therapy

The distribution of T lymphocyte subsets was determined in peripheral blood (PB) of two groups of patients with rheumatoid arthritis by using monoclonal antibodies (OKT). In untreated patients the percentage of OKT4+ cells (helper/inducer) was found to be significantly increased as compared to healthy controls. In patients receiving oral gold therapy a similar increase in OKT4+ cells was confirmed; furthermore, these patients showed a significant decrease in OKT8+ cell population (cytotoxic/suppressor) compared to untreated patients and to normal controls. A small numerical superimposition of values of OKT4+ and OKT8+ lymphocytes was observed in untreated but not in treated patients.     

Determination of sub-populations of circulating T lymphocytes in alcoholic cirrhosis using monoclonal antibodies OKT3, 4, 5, Leu 2 and Leu 15. Effect of hepatitis B virus infection

Peripheral T lymphocyte subpopulations were quantified in 24 alcoholic cirrhotic patients, 11 of them having anti-HBs and/or anti-HBc antibodies, and were compared with 35 healthy control subjects, 10 of them having anti-HBs and/or anti-HBc antibodies. The monoclonal antibodies utilized (OKT3, OKT4, OKT8 in simple staining, Leu 2 and Leu 15 in double staining) are considered as markers of mature (CD3), helper (CD4), cytotoxic/suppressor (CD8, Leu 2), suppressor (Leu [2+ 15+), and cytotoxic (Leu 2+ 15-) T cells. In cirrhotics, when compared to controls, the number of CD3 cells was reduced (p less than 0.01); the proportion of CD4 cells was within normal range, and that of CD8 cells diminished (p less than 0.001), contrasting with an increased proportion of Leu 2+ cells (p less than 0.01), related to an increased proportion of Leu 2+ 15+ cells. Leu 2+ 15- lymphocytes were within normal range. In control subjects, a decreased proportion of Leu 2+ 15+ cells was found (p less than 0.05) when Ac HBs and/or Ac HBc were present. In cirrhotics having at least one serologic marker of hepatitis B virus infection, when compared with negative ones, increased proportions of Leu 2+ (p less than 0.05) and Leu 2+ 15+ (p less than 0.05) cells were found. These results show that data concerning T lymphocyte subpopulations are conflicting when various types of antibodies are used. However, they suggest abnormalities of immune regulation, possibly a defect of T suppressor cell function. Hepatitis B virus infection probably modifies immune regulation in alcoholic cirrhosis, and perhaps in normal subjects.     

Analysis of lymphocyte subsets of bone marrow in patients with rheumatoid arthritis by two colour immunofluorescence and flow cytometry.

Lymphocyte subsets defined by monoclonal antibodies were investigated in bone marrow and peripheral blood of 17 patients with rheumatoid arthritis (RA); 13 patients with osteoarthritis or aseptic necrosis served as controls. Patients with RA were found to have a raised OKT 4/8 ratio both in bone marrow and peripheral blood in comparison with the controls. Furthermore, bone marrow of RA showed a lower percentage of OKT 8+ T cells than that of controls. The percentage of HLA-DR+ T cells was higher in bone marrow than in peripheral blood of RA, though a slightly lower percentage was detected in bone marrow than in peripheral blood of controls. Thus T cell subsets in bone marrow of RA differ significantly from those of controls. Patients with RA had a higher OKT 4/8 ratio and a higher percentage of HLA-DR+ T cells in bone marrow than controls, suggesting that T cell subsets in bone marrow of RA are in an immunologically activated state and that T cell subsets are affected by rheumatoid inflammation in bone marrow of RA.     

Circulating lymphocyte subpopulations in Crohn's disease

Circulating lymphocytes were enumerated in 28 patients with Crohn's disease and in 12 patients with other diseases by rosetting and by immunofluorescent staining using monoclonal antibodies for T-cell surface phenotypic markers [OKT3 (mature), OKT4 (helper), and OKT8 (suppressor/cytotoxic)] or polyvalent antisera for surface immunoglobulins (B cells). Total lymphocyte counts were reduced only in those with non-steroid-treated active Crohn's disease. Circulating monocyte counts, proportions of peripheral T and B cells, and percentages and absolute numbers of mature, helper, and suppressor T-cell subclasses in Crohn's disease were not significantly different than in the controls. Helper to suppressor T-cell ratios were comparable in all subjects, varying directly with numbers of helper T cells (p less than 0.05). Individual ratios of helper to suppressor T cells did not correlate with disease activity or location, the use of steroids, serum albumin, or total lymphocyte or monocyte counts. This study provides no evidence for underlying abnormalities of circulating lymphocyte subpopulations in Crohn's disease when compared to subjects with other illnesses. The characterization of lymphocyte subclasses in affected tissues is an important area of continuing investigation.     

Analysis of T lymphocytes after bone marrow transplantation using monoclonal antibodies

Using monoclonal antibodies to cell surface antigens and fluorescent cell sorter analysis, we studied peripheral blood lymphocyte subsets after bone marrow transplantation (BMT). In 13 patients studied 3 mo or more after BMT, the ratio of T-cell subsets defined by antibodies OKT4 and OKT8 was reversed (OKT4/OK%8 = 0.7 +/- 0.3) in comparison to normal volunteers or bone marrow donors (ratio OKT4/OKT8 = 1.7 +/- 0.4) (p less than 0.001). This reversed ratio persisted for up to 3 yr after BMT. In contrast to a previous report, presence of an abnormal ratio of T-cell subsets did not correlate with clinically significant graft- versus-host disease (GVHD). In agreement with a previous study, (26% +/- 8%; less than 4% in normals (p less than 0.001) and antibody OKT10 reactive cells (39% +/- 20% versus 10% +/- 4%) (p less than 0.01), suggesting in vivo activation. However, their PBL did not react with antibody B3/25 (antitransferrin receptor), a marker found on normal PBL after in vitro activation by mitogens (BMT patients less than 5%; normal PBL T cells plus PHA 45% +/- 11%). These results demonstrate that BMT patients have: (A) an abnormal ratio of T-cell subsets in the presence or absence of clinically significant GVDH disease so that these measurements were not useful in monitoring patients; (B) an increased number of T cells with cell surface phenotype (OKT8+, Ia+, OKT10+, B3/25-) that is distinct from normals but similar to patients with infectious mononucleosis or acquired hypogammaglobulinemia.     

T cells in monoclonal gammopathies

Subpopulations of human T lymphocytes were analysed by monoclonal antibodies (OKT3, OKT4, OKT8) in healthy controls and in patients with multiple myeloma (MM), Waldenstr?m's macroglobulinemia (WM) and benign monoclonal gammopathy (BMG). Lymphocytes forming rosettes with SRBC correlated well with T cells staining in indirect immunofluorescence by OKT3 monoclonal antibodies. The relative and absolute numbers of OKT4+ T cells were significantly lower in patients than in controls. Though, the percentage of OKT8+ T cells was increased in patients, the total OKT8+ cell counts were normal in all patient groups. The ratio between OKT4+ and OKT8+ lymphocytes was low in the groups of treated MM and of WM patients compared to controls (P less than 0.001). Moreover, the ratio was lower than the normal range in 27% of BMG and 38% of untreated MM patients. The imbalance between OKT4+ and OKT8+ T cells in untreated MM was more pronounced in clinical stage III patients than in stage I and II patients. The most pronounced changes were noted in treated MM patients. The significance of the altered T cell subsets in monoclonal gammopathies with regard to polyclonal and tumor B cell regulation remains to be established.     

Selective depletion and activation of CD8+ lymphocytes from peripheral blood of patients with polymyalgia rheumatica and giant cell arteritis.

A prospective study of 33 patients with polymyalgia rheumatica/giant cell arteritis (PMR/GCA) was undertaken, firstly, to monitor sequentially peripheral blood CD8+ lymphocyte levels and, secondly, to assess the expression of activation markers on T lymphocyte subsets. The results indicated that there was a significant decrease in absolute numbers and relative percentages of CD8+ T lymphocytes, which returned to normal ranges after approximately 24 months' treatment, and that there was an increased percentage of CD8+ lymphocytes in PMR/GCA which express HLA class II antigens.