Abbreviations used FEV 1 : Forced expiratory volume in I second PCo: Provocative concentration causing a 20% fall in FEV,

Transcription

1 Pulmonary function, activated T cells, peripheral blood eosinophilia, and serum activity for eosinophil survival in vitro: longitudinal study in bronchial asthma Johann-Christian Virchow, Jr., MD,' lbert Oehling,b Lisbeth Boer,c Trevor T. Hansel, MD,' Peter Werner, MD, FCCP,6 Heinrich Matthys, MD, 8 Kurt Blaser, PhD,c and Christoph Walker' Freiburg, Germany, and Davos- Wolfgang and Davos, Switzerland close correlation among the number of activated, peripheral blood T helper cells, eosinophilia, and airflow obstruction has been reported in patients with asthma. To test these cross-sectional data we performed a prospective, longitudinal study investigating the relationships among T-cell activation in peripheral blood, eosinophilia, forced expiratory volume in 1 second (FEV), and serum activity toward eosinophil survival in 20 individuals with asthma over a period of 21 days after admission to a clinic located 1560 m above sea level. During the study, maintenance treatment with inhaled P1 2 -agonists and theophylline was unchanged. Five patients had also been taking inhaled corticosteroids, and this was also not changed during the study period. ccording to the changes in pulmonary function observed at the end of the study, patients were divided into three groups: twelve patients whose pulmonary function improved by more than 10% from baseline (group I), four patients whose pulmonary function deteriorated by more than 10% from baseline necessitating therapeutic intervention (group II), and four patients with no change in FEV, ( < 10% from baseline) (group III). ctivation of T cells as determined by interleukin-2 receptor expression, number of eosinophils, and serum activity toward eosinophil survival in vitro declined in group I; whereas FEV 1 increased. The decrease in FEV in group II was associated with an increase in interleukin-2 receptor-positive T cells, number of eosinophils, and serum activity. In group III FEV, interleukin-2 receptor expression on T cells, the number of peripheral blood eosinophils, and the measured serum activity for eosinophil survival in vitro did not change significantly from baseline. With the exception of FEV, and eosinophil numbers, which came close to reaching statistically significant correlations with the measured serum activity for eosinophils, the percent changes from baseline to the end of the study of all other parameters were closely correlated with correlation coefficients between 0.76 and We conclude that a close correlation does exist between changes in T-cell activation, numbers of eosinophils, serum activity toward eosinophil survival in vitro, and the degree of airflow obstruction in patients with asthma, observed longitudinally, suggesting that these parameters are interrelated. (J LLERGY CLIN IMMUNOL 1994;94:240-9.) Key words: sthma, T-cell activation, eosinophils, FEV1, longitudinal correlation From the Department of Pneumology, University Hospital, Freiburg, Germany; bthe Hochgebirgsklinik Davos, Davos- Wolfgang, Switzerland; and cswiss Institute of llergy and sthma Research, Davos, Switzerland. Received for publication Mar. 25, 1993; revised Dec. 12, 1993; accepted for publication Feb. 7, Reprint requests: Johann-Christian Virchow, Jr., MD, Depart- bbreviations used FEV 1 : Forced expiratory volume in I second PCo: Provocative concentration causing a 20% fall in FEV, Eosinophilia of the peripheral blood, ment of Pneumology, University Hospital, Hugstetterstrasse Es a of the peripheral blood, 55, D Freiburg, Germany. mucosa, 2 and sputum, 3 as well as increased bron- Copyright 1994 by Mosby-Year Book, Inc. chial responsiveness, 4 are characteristic features /94 $ /1/55250 of bronchial asthma. Peripheral blood eosino- 240

2 J LLERGY CLIN IMMLINOL VOLUME 94, NUMBER 2, PRT 1 philia has been shown to directly correlate with impairment of airflow' and degree of bronchial hyperreactivity to inhaled histamine. 5 The formation of eosinophils in the bone marrow and the function of mature eosinophils are regulated by T-cell-derived cytokines. 6 ' 7 Mainly interleukin (IL)-5, IL-3, and granulocyte-macrophage colonystimulating factor have been described as acting on eosinophils."" 4 ctivated T cells have been observed in the peripheral blood of patients with mild to moderate asthma, ' 5 and the number of activated T cells correlates with the airflow obstruction assessed by peak flow measurements. '6 In a previous study we have demonstrated a direct correlation between peripheral blood eosinophilia and the number of activated, IL-2 receptor (CD25)-positive T cells in patients with asthma.'5 These spontaneously release IL-5, IL-3, and granulocyte-macrophage colony-stimulating factor, which enhance eosinophil survival in vitro. 5 The same cytokines have been recovered from bronchoalveolar lavage from patients with asthma. 7 Expression of mrn for IL-5 has been described in bronchial biopsy specimens from patients with asthma. 8 Investigators have related the number of circulating activated T helper cells with airflow obstruction in the larger airways as measured by peak expiratory flow.l6 Both activated T lymphocytes and eosinophils have been identified in bronchial biopsy specimens from patients with stable atopic asthma. 2 ' 19 The number of eosinophils in the peripheral blood count has also been correlated inversely with the severity of asthma, as measured by forced expiratory volume in 1 second (FEV 1 ).' In addition, studies investigating bronchoalveolar lavage have demonstrated an inverse relationship between the number of eosinophils and FEV 1.20 In children with asthma a seasonal increase in exercise-induced bronchospasm was associated with an increase in the number of eosinophils in the peripheral blood count. 2 ' Furthermore, eosinophil granular products released from eosinophils, such as cationic proteins, 2 2 have been correlated with disease activity. 23 The treatment of patients with asthma resulted in reduced levels of cationic proteins with an associated decrease in airflow obstruction. 2 However, to date no study has been published, which attempted to longitudinally (i.e., over a longer time interval) determine in patients with asthma the apparent relationships among activated, IL-5- producing T (helper/inducer) cells, eosinophils, Virchow et al. 241 serum activity of cytokines, and change in pulmonary function. We therefore studied, in patients with asthma, the correlation between the number of activated T cells and eosinophils in peripheral blood, as well as the changes in FEV, and the effect of serum on eosinophil survival in vitro over a period of 21 days. METHODS Patients Twenty patients, 11 women and nine men (mean age, L5.3 years; range, 18 to 71 years), with a diagnosis of asthma as defined by the International Consensus Report on the Diagnosis and Management of sthma, 4 were entered into the study. ll patients had a history of variable degrees of wheezing and coughing associated with chest tightness. Reversibility of airflow obstruction (FEV, and/or peak expiratory flow rate >20%) had been documented previously. The peripheral blood count for eosinophils at the beginning of the study ranged from 360 to 2440/mm 3 ( ; mean ± SD). Sixteen patients had allergic asthma of childhood onset with a positive history for allergen-induced bronchospasm after contact with one or more of the following sources of allergens: house dust, cats, grass pollen, birch pollen. Skin prick tests with allergen extracts of house dust mite (Derrnmatophagoides pteronyssinus), cat dander, grass pollen, and/or birch pollen (Bencard, SmithKlineBeecham, Neuss, Germany) produced a positive wheal (diameter >5 mm) and flare (diameter > 15 mm) reaction in the presence of a positive histamine control and a negative vehicle control. The diameter of the wheal and flare reaction was calculated by dividing the sum of the largest diameters perpendicular to each other by 2. Furthermore, the patients with allergic asthma had elevated total IgE levels between 360 and 2000 ku/l and specific IgE levels of greater than 0.70 ku/l in response to at least one of the above mentioned allergens. Four patients had intrinsic (nonallergic) asthma of adulthood onset with no history of allergen-induced bronchospasm; negative skin prick test results to a large panel of ubiquitous airborne, occupational, and nutritional allergens; normal IgE levels; and a history of onset of asthma after a severe respiratory tract infection. For at least 4 weeks before entry into the study, all patients had been receiving 3 2 -sympathomimetic treatment as baseline medication; 11 patients had also been receiving theophylline orally in a dose that achieved therapeutic serum levels between 10 and 20 mg/l. Five patients had been taking inhaled corticosteroids (beclomethasone dipropionate, 0.5 mg two times a day) regularly for at least 2 months before the study. During the study period neither dose nor type of medication, including inhaled corticosteroids, were changed. If the

3 242 Virchow et al. J LLERGY CLIN IMMUNOL UGUST 1994 current medication had to be changed because of increasing symptoms, patients were withdrawn from further participation in the study. Patients were allowed, however, to use additional 2-sympathomimetics if needed. Before blood sampling and pulmonary function tests, bronchodilator medications were withheld for at least 12 hours. On entry into the study, bronchial responsiveness was measured by inhaled bronchoprovocation with histamine as described elsewhere. 25 Provocative concentration causing a 2050 fall in FEV, (PC 20 ) was established as the concentration of histamine at which a drop in FEV, of 20% compared with baseline occurred. Bronchial provocation with histamine was not performed in patients with baseline FEV, of less than 70% of predicted value. Two patients declined the bronchoprovocation test with histamine. FEV, in all patients at the beginning of the study was 78.8% % of predicted value, with a range of 44.3% to 113.3%. Patients' symptoms were evaluated daily by an independent clinician. Based on this attending physician's assessment, a burst of systemic corticosteroids was started if a patient's status and symptoms deteriorated clinically. Only data collected before initiation of rescue corticosteroid therapy were included in the analysis. ll patients had a normal chest roentgenogram in posterior-anterior and left-lateral projections and remained afebrile throughout the study. ll patients were lifelong nonsmokers and gave their informed consent. The study protocol was approved by the Human Subjects Committee of the Hochgebirgsklinik, Davos. Collection of blood Ten milliliters of venous blood was collected into ethylenediaminetetraacetic acid-containing plastic tubes (Sarstedt, Niimbrecht, Germany) under sterile conditions between 7:15 and 7:30 M on days 1, 4, 7, 11, 14, and 21. Pulmonary function tests On the days of blood sampling, FEV, was measured between 7:30 and 8:00 M with the Jaeger Bodytest (Jaeger, Wfirzburg, Germany). The FEV, measurement was repeated until a reproducibility of greater than 95% was achieved. Pulmonary function tests were obtained no later than 30 minutes after blood was collected. The values for the predicted FEV, were calculated from an algorithm published elsewhere. 26 nalysis of leukocytes and lymphocyte subsets Blood leukocyte counts and differential analysis were measured by automated blood count analysis (Technicon HI; Technicon, Tarrytown, N.Y.). Specific binding of monoclonal antibodies against CD3 (T cells), CD4 (T helper/inducer), CD8 (T suppressor/cytotoxic), CD25 (L-2 receptor), and HL-DR was analyzed as described previously." Determination of serum activity toward eosinophils Purification of eosinophils from patients with asthma was performed by negative selection with anti-cd16- coated magnetic microspheres, as described previously with anti-cd16-coated immunomagnetic particles (anti-cd16 [CLB Fc Gran 11), which were a kind gift of Dr. T. Huizinga (Central Laboratory of the Netherlands Red Cross Blood Transfusion Service, msterdam). 27 These highly purified eosinophils were cultured in tissue microculture plates (Falcon Plastics, Oxnard, Calif.) (105 cells/well) containing l of either RPMI 1640 (Gibco, Grand Island, N.Y.)/10% fetal calf serum or 50% patient or control sera at 37 C and 5% CO 2 for 3 days. Then a 100 ld cell suspension was mixed with 200 p.l phosphate-buffered saline containing 2% fetal calf serum, 0.1% NaN 3, 1 pimol/l ethidium bromide and incubated for 5 minutes at 40 C and analyzed in a cytofluorograph (Epics Profile, Coulter Corp., Hialeah, Fla.). The fraction of viable and dead cells in each sample was determined from 5000 counted eosinophils. Statistical analysis In the statistical analysis the arithmetic mean + the standard deviation was used unless indicated otherwise. For comparison within group I Wilcoxon's test was used. Correlation coefficients and their statistical significance were determined with Spearman's rank correlation analysis. RESULTS Peripheral blood leukocyte and lymphocyte subpopulations were analyzed in conjunction with measurements of pulmonary function parameters in 20 individuals with asthma over a period of 21 days. In four patients the study had to be terminated prematurely because their asthmatic symptoms increased, necessitating change of their baseline medication by addition of a burst of systemic corticosteroids. ccording to the changes in pulmonary function observed on the day of termination of the study, patients were divided into three groups. Group I consisted of patients whose FEV, improved by more than 10% from baseline, whereas patients whose FEVI decreased by more than 10% from baseline were assigned to group II. Group III consisted of patients whose FEV remained unchanged (< 10%) compared with baseline. Twelve of 20 patients showed an improvement in FEV 1 from 71.4% % of predicted value to 96.5% % (p < 0.001) during the 21-day observation period. These patients (group I) had a mean age of ± 16.6 years, and the mean PC 2 0 to histamine, measured in six patients of this group, was mg/ml. In the remaining six patients inhaled histamine provocation was not

4 J LLERGY CLIN IMMUNOL VOLUME 94, NUMBER 2, PRT 1 Virchow et al. 243 I So E P 100 To o--o = E:osinophils *--* = IL-2R+ T ce s -a = C0D4+ IL-2R+ T cells o-e = Serum ctivity for Eosinophils * -* = FEV1 T'.*.Baseline 0 Io E o-o = Eosinophils -- * = IL-2R+ T cells n- = CD4+ IL-2R+ T cells D-o = Serum ctivity for Eosinophils *- = FEV1 o a 0 1W 50 - W Z W* = C 5 & ~ -Baseline c [days] C [days] Group I Group II FIG. 1. Patients of group.the increase in FEV, (ki-) of 12 patients compared with baseline (100%) is reflected by a decrease in eosinophil numbers (o-o), IL-2 receptorpositive T cells (--), and IL-2 receptor-cd4* T cells (ti), as well as serum activity (---) to enhance eosinophil survival in vitro. Error bars: standard error of the mean. performed because the baseline FEV, was less than 70% of predicted value. The observed increase in FEV, in this group was accompanied by a decline in the numbers of circulating eosinophils, activated T cells, activated CD4+ lymphocytes, and in serum activity toward in vitro survival of eosinophils (Fig. 1). Most of the changes, especially in improvement in FEV1, were observed within the first 3 days after admission (Fig. 1). ll five patients who were receiving inhaled corticosteroids on a regular basis were in this group. In group II four patients had to terminate the study prematurely because their FEV 1 declined from 88.1% +± 2.8% of predicted value to 71.9% ± 19.9% and their clinical course deteriorated, necessitating a burst of systemic corticosteroids. These patients' participation in the study was terminated at the time systemic corticosteroid therapy was initiated. Their mean age was years, and the mean PC 20 to histamine measured before the study period in three of these patients was mg/ml; one patient refused the bronchoprovocation test with histamine. The deterioration in FEV 1 observed in group II was accompanied by an increase in eosinophils, activated T cells, activated T helper/inducer cells, and serum activity toward in vitro survival of eosinophils (Fig. 2). In the four patients in group III, FEV, changed less than 10% compared with the baseline values at the beginning of the study (88.2% + 5.8% of predicted value vs 88.2% + 5.9%). Their mean FIG. 2. Patients of group II. The participation of these patients in the study was terminated prematurely because they required a burst of systemic steroids. The decrease in FEV 1 (--) of four patients compared with baseline (100%) is accompanied by an increase in eosinophil numbers (--o), IL-2 receptor-positive T cells (e-), and IL-2 receptor-cd4 + T cells (-), as well as an increase in serum activity (-o) to enhance eosinophil survival in vitro. Error bars: standard error of the mean. _0 E c =o a o--o = Eosinophils -- * = IL-2R+ T cells -- = CD4+ IL-2R+ T cells u-u = Serum ctivity for Eosinophils -- = FEV1 01: ~ ~ - Baseline [days] Group III FIG. 3. Patients of group IIIl. No change from baseline (100%) was observed for any of the measured parameters in four patients. Error bars: standard error of the mean. age was 21.8 ± 3.0 years, and the mean PC 20 to histamine in three of these patients, measured before the study period, was mg/ml; one patient refused the bronchoprovocation test with histamine. The total eosinophil count in the peripheral blood, the CD25+, activated T cells, activated CD25 + CD4 + T helper cells, and serum activity toward eosinophil survival in vitro did not change significantly in this group (Fig. 3). Cellular changes in peripheral blood for the three groups during the study period are summarized in Table I.

5 244 Virchow et al. J LLERGY CLIN IMMUNOL UGUST 1994 TBLE I. Change of FEV, and peripheral blood leukocyte and lymphocyte subpopulations during the study period in groups I to III Group I Group II Group III Day of Day 1 Day 21 Day 1 termination Day I Day 21 FEV 1 (L) ± 20.6* 88.1 ± ± ± ± 5.9 Leukocytes (x 10"/L) 7.0 ± l.lt 6.9 ± ± ± ± 1.1 Neutrophils (x 10 6 /L) 3.7 ± ± ± ± ± 1.1 Eosinophils (x 10 6 /L) 1.2 ± * 0.5 ± ± ± ± 0.2 Basophils (x 10 6 /L) 0.1 ± ± ± ± ± 0.01 Monocytes (x 106/L) 0.5 ± ± ± ± ± 0.03 Lymphocytes (x 10 6 /L) ± ± ± ± CD3 (%) ± ± ± CD4 (%) ± ± ± ± CD8 (%) ± ± _ 1.9 B cells (%) ± ± ± ± 3.9 NK cells (%) ± ± CD4/CD8 2.4 ± ± CD4/IL-2R (%) * ± ± CD4/HL-DR (%) ± ± ± 3.1 CD8/IL-2R (%) 1.8 ± ± ± ± ± 0.4 CD8/HL-DR (%) NK, Natural killer. *p < tp < 0.05 (Wilcoxon matched pairs test). Correlations of the parameters observed were obtained by comparing the percent change of each parameter in each patient from baseline to the end of the study. In the four patients who had to terminate the study because of deterioration of pulmonary function, the last valid data pair, which was obtained before corticosteroids were added to the therapeutic regimen, was used. This comparison of the change in relative numbers of eosinophils and IL-2 receptor-positive T cells between baseline and the time of termination of the study for all patients demonstrated a significant, positive correlation (n = 20, r = 0.75, p < ) (Fig. 4, ). Similar correlations were found for the changes in relative numbers of eosinophils, the change in numbers of IL-2 receptor-positive T helper cells (n = 20, r = 0.70, p < ) (Fig. 4, B), and the change in FEV1 in percent of predicted value (n = 20, r = -0.70, p = ). Correlation of the change in serum activity toward eosinophil survival in vitro and eosinophil numbers almost reached statistical significance (n = 18, r = 0.38, p < 0.12) (Fig. 4, D). The change in relative numbers of activated, IL-2 receptor-positive T lymphocytes strongly correlated with the number of IL-2 receptor- CD4 + T cells (n = 20, r = 0.94, p < ), suggesting that the majority of IL-2 receptorpositive T cells were of the CD4 T helper/inducer cell phenotype (data not shown). n inverse correlation was observed between the change in FEV, in percent from baseline and the change in the numbers of activated, IL-2 receptor-expressing T lymphocytes (n = 20, r = -0.76, p < ) (Fig. 5, ). The change in activated, IL-2 receptor-positive T helper lymphocytes inversely correlated with the change in FEV1 in percent of predicted value (n = 20, r = -0.70, p = ) (Fig. 5, B) and with the change in serum activity toward eosinophil survival in vitro (n = 18, r = 0.54, p = 0.02) (Fig. 5, C). The correlation between FEV 1 and serum activity for eosinophils almost reached statistical significance (n = 18, r = 0.39, p = 0.11) (Fig. 5, D). There was no correlation, however, between the number of CD8 + T cells or IL-2 receptor-positive CD8+ T cells and any of the parameters investigated (data not shown). Furthermore, no correlation was observed between the expression of HL-DR on CD4 + or CD8 + T cells or total numbers of T cells with the investigated parameters in any of the three patient groups, individually or combined.

6 J LLERGY CLIN IMMUNOL VOLUME 94, NUMBER 2, PRT 1 Virchow et al. 245 E -e + U Ir- a B i 20- -,t r = zrn~~~~ C on 1 40! Eosinophils [%] FIG. 4., Correlation between the percent change in peripheral blood eosinophils ( % eosinophils) and the percentage of IL-2 receptor-positive T cells ( % CD25+ T cells) from baseline (baseline = 0%) to the end of the study (n = 20 data pairs; p < ). B, Correlation between the change in absolute numbers of peripheral blood eosinophils ( % eosinophils) and the percentage of IL-2 receptor-positive CD4 + T lymphocytes ( % CD25+ CD4+ T cells) from baseline (baseline = 0%) to the end of the study (n = 20 data pairs; p < ). C, Correlation between the change in absolute numbers of peripheral blood eosinophils ( % eosinophils) and change in FEV, ( % FEV, in percent of predicted value) from baseline (baseline = 0%) to the end of the study (n = 20 data pairs; p < ). D, Correlation between the change in absolute numbers of peripheral blood eosinophils ( % eosinophils) and the change in serum activity to enhance eosinophil survival ( % of serum activity for eosinophils) in vitro from baseline (baseline = 0%) to the end of the study (n = 18 data pairs; p < 0.12) n I - I I - --Jb -.. ~I - -- I r = 0.38 D The number of HL-DR CD8+ T cells tended to be higher in the patients whose pulmonary function deteriorated (group II) but did not reach statistical significance. DISCUSSION In this study we analyzed the relationship of activated T cells, T-cell subpopulations, and their products (eosinophil survival-promoting activity in serum) in relation to the total eosinophil count in the peripheral blood and the degree of airflow obstruction in bronchial asthma. For this purpose, patients being hospitalized with mild to moderate asthma were included in a prospective, longitudinal study. In asthma a number of abnormal immunologic parameters such as peripheral blood eosinophilia, 1 sputum eosinophilia, 3 and the presence of mast cells 4 have been reported including increased numbers of activated T cells, 6 which have been implicated as playing an important regulatory role 5 in inflammatory alterations observed in the bronchial mucosa in asthma. 2 Their presence has been related to the clinical severity of the disease,' 6 which has also been described for peripheral blood eosinophilia' and eosinophil products. 2 3 In only one longitudinal study was a decrease in peripheral blood activated T cell numbers with improvement in peak expiratory flow observed in patients with acute, severe asthma. 6 The authors, however, did not relate their findings to the peripheral blood eosinophilia in their patients. In our study, changes in T-cell activation were inversely correlated with the change in airflow obstruction observed longitudinally, confirming

7 246 Virchow et al. J LLERGY CLIN IMMUNOL UGUST 1994 I- 0 3 I r = "-" FEVI %] 80 C a() a\) V~J - 4(0 o.1-20 o () 0 Cr fj n, L i r = , -4( FEV1 [%].., B D CI _. --. r = 0.54 I Serum ctivity for Eosinophils [%] ( r = FEV1 [%] FIG. 5., Correlation between the change in FEV, ( % FEV, in percent of predicted value) from baseline and change in the percentage of IL-2 receptor-positive T lymphocytes ( % CD25+ T cells) from baseline (baseline = 0%) to the end of the study (n = 20 data pairs; p < ). B, Correlation between the change in activated, IL-2 receptor-positive T helper lymphocytes ( % CD25+ CD4+ T cells) and FEV, ( % FEV, in percent of predicted value) from baseline (baseline = 0%) to the end of the study (n = 20 data pairs; p < ). C, Correlation between the change in serum activity to enhance eosinophil survival ( %/6 of serum activity for eosinophils) in vitro and change in the percentage of IL-2 receptor-positive T helper cells ( % CD25+ CD4+ T cells) from baseline (baseline = 0%) to the end of the study (n = 18 data pairs; p < 0.02). D, Correlation between the change in FEV, (in percent of predicted FEV,) from baseline and the change in serum activity to enhance eosinophil survival in vitro from baseline (baseline = 0%) to the end of the study) (n = 18 data pairs; p < 0.11). a 1Q _I m In L E60 the observations of the study by Corrigan and Kay.' 6 In contrast to their study population, 6 we investigated patients with asthma whose antiasthmatic therapy had not been changed for at least 4 weeks before being referred to an asthma clinic located at 1560 m above sea level on account of their mild to moderate asthmatic symptoms at the time of admission. In a previous, cross-sectional study we reported a close correlation between the number of activated total peripheral blood T cells and the CD4 + T cells with the number of circulating eosinophils in patients with mild to moderate asthma.' 5 This finding was attributed to soluble factors present in serum and in supernatants of unstimulated peripheral blood T cells, which had the capability of enhancing eosinophil survival in vitro and were characterized as containing IL-3, IL-5, and granulocyte-macrophage colony-stimulating factor. These latter cytokines, particularly IL-5, have profound effects on the differentiation of eosinophils and activation of their functions Consequently, because eosinophils" 2 and their prod- UCts23, 24 have been implicated in the pathogenesis of bronchial asthma, we also analyzed the change in the numbers of peripheral blood eosinophils in relation to the numbers of activated T cells (and CD4+ T cells) and pulmonary function. n inverse relationship between peripheral blood eosinophilia and pulmonary function and a direct correlation between the number of activated T cells (and CD4' T cells) in peripheral blood and peripheral blood eosinophils were found when the change measured in percent between baseline and the termination point of the study were correlated. Both of these correla-

8 J LLERGY CLIN IMMUNOL VOLUME 94, NUMBER 2, PRT 1 Virchow et al. 247 tions thus confirm earlier reports 1' ' on bronchial asthma. However, our prospective study extends these observations with longitudinal data, thus allowing each patient to serve as his or her own control. Despite the :Fact that the pulmonary function improved in some patients, whereas in others a decline in FEV:, was observed over the study period, relatively symmetric changes in the number of activated T cells, the number of eosinophils, and the serum level of eosinophil survival-prolonging activity could be demonstrated (Table I, Figs. 1 to 3). In the four patients in whom pulmonary function changed less than 10% from baseline T-cell activation, eosinophil numbers in peripheral blood and serum activity remained relatively constant throughout the observation period. In their group of subjects with asthma Corrigan and Kay'6 were able to demonstrate a decline in numbers of activated peripheral blood T helper cells as the airway obstruction decreased. Our study significantly expands on these observations by showing that not only the numbers of activated T cells and eosinophils and serum levels of eosinophil survival-prolonging activity declined with improving pulmonary function (Fig. 1), but that the opposite was true as well (i.e., declining lung function was associated with a rise in activated T cells, eosinophils, and serum levels of eosinophil survival-prolonging activity) (Fig. 2). These observations and the additional findings in four of our patients with stable pulmonary function and relatively constant numbers of activated peripheral blood T cells, eosinophils, and serum levels of eosinophil survival-prolonging activity support the concept of a causal relationship between these parameters and lung function in patients with asthma. Clear differences in the expression of activation markers on CD4 and CD8 peripheral blood T lymphocytes have been demonstrated in patients with asthma compared with normal control subjects. 5 ' 7 nother study that analyzed the expression of activated T cells in atopic dermatitis showed stable expression of CD25 on T helper cells in a normal control group over a period of 3 weeks, and as in our study, marked changes were observed in patients with atopic dermatitis. 28 The combined results of our study thus link the results of several previous studies,' 2 15, 16 suggesting that activated T cells and eosinophils are interrelated in the pathogenesis of bronchial asthma. number of confounding variables might interfere with the statistical interdependence of the parameters measured. The individual degree of bronchial reactivity in the patients studied may have markedly influenced the pulmonary function because controls were not applied for exposure to nonspecific airway irritants and physical exercise, which are both known to influence the degree of airflow obstruction rapidly. Furthermore, it cannot be excluded that additional baseline medication, especially inhaled bronchodilators, were used without the knowledge of the investigators. Yet, the changes observed in T-cell activation and serum factors together with the peripheral blood eosinophilia were rather closely related to pulmonary function parameters throughout the study period. The reason for the observed improvement in pulmonary function in nine patients with allergic asthma, despite an unchanged medication schedule, could be due to reduced exposure to house dust mite allergen, which is not present at an altitude of 1560 m above sea level. 2 9 ' 3 Because the study was performed during the fall and winter months, reduced exposure to pollen may have contributed to the improvement in pulmonary function experienced by these patients. However, the focus of our study was directed toward the relationship between cellular and cell-associated parameters of asthmatic inflammation and pulmonary function. Thus the assumption that a decrease in allergen exposure was responsible for the clinical improvement in the patients with allergic asthma must remain speculative. Four of the 20 patients included in this study had intrinsic asthma, for which no allergic background could be demonstrated. Respiratory tract infections are frequent causes of exacerbations of intrinsic asthma. However, during the study period no signs or symptoms suggestive of respiratory tract infections could be observed in these four patients. The reason for the improvement in three of four of the patients with intrinsic asthma remains unclear. Two of these patients who experienced improvement and who had a baseline FEV, of 44% and 48% of predicted value, received inhaled corticosteroid therapy at regular intervals, whereas one patient with intrinsic asthma who did not receive topical corticosteroids showed a decline in FEVI and a worsening of clinical symptoms, necessitating treatment with systemic steroids. Despite the fact that the baseline medications were not changed 4 weeks before the study and during the study period, these findings underscore the relative

9 248 Virchow et al. potency of topical steroid therapy in the treatment of chronic asthma. The reasons for the decline in FEV, observed in three patients with allergic asthma remain unknown. detailed search for precipitating factors did not provide any etiologic clues. This is, however, compatible with the clinical observation that not all asthmatic exacerbations can be clearly attributed to exposure to allergen or respiratory tract infections. The reason that the FEV, of one patient with intrinsic asthma and four patients with allergic asthma who received only an inhaled 0 2 -agonist on a regular basis improved during the observation period remains unclear, and it remains speculative whether allergen avoidance at high altitude and/or the balanced use of P 2 -agonists were the sole responsible factors for the observed improvement. lthough the starting (percent predicted) mean FEVt measurements in the patients in groups I to III were similar, the mean histamine PC 20 measurements were different, ranging from 0.06 mg/ml in group II to 1.12 mg/ml in group I. Despite this observation, which suggests that patients with lower FEV, were more likely to experience deterioration in pulmonary function, it must be stressed that PC 20 measurements were not available for all patients. It was not the aim of the study to analyze any association of PC 20 with the spontaneous course of the patients' asthma, and therefore this observation should be interpreted with appropriate care. ctivated CD8 + T cells have been described in the peripheral blood and bronchoalveolar lavage in a previous study of patients with intrinsic asthma.' 7 Our patients who experienced subsequent deterioration in their FEV, tended to have higher percentages of HL-DR-positive, activated CD8 + T cells in peripheral blood at the start of the study; however, this was not statistically significant. During the study, their numbers did not correlate with any of the parameters under investigation. No evidence about their functional role in asthma is available. In our study no change was observed in the percentage of HL-DR expression on CD4 and CD8 + T cells. In conclusion, our data demonstrate an inverse correlation between T-cell activation, serum activity toward eosinophil survival in vitro attributed to circulating cytokines, eosinophilia, and parameters of airflow obstruction over a period of 3 weeks. These observations provide further evidence that there is an interrelationship of these J LLERGY CLIN IMMUNOL UGUST 1994 parameters, suggesting that they may play a role in the development of the syndrome of bronchial asthma. However, the exact interplay that links these parameters requires further elucidation. REFERENCES 1. Horn BR, Robin ED, Theodore J, Van Kessel. Total eosinophil counts in the management of bronchial asthma. N Engl J Med 1975;292: Bousquet J, Chanez P, Lacoste JY, et al. Eosinophilic inflammation in asthma. N Engl J Med 1990;323: Gollasch H. Zur Kenntnis des asthmatischen Sputums. Fortschr Med 1889;7: International Consensus Report on Diagnosis and Treatment of sthma. National Heart, Lung, and Blood Institute, Publication No Eur Respir J 1992;5: Taylor KJ, Luksza R. Peripheral blood eosinophil counts and bronchial responsiveness. Thorax 1987;42: Miyajima, Miyatake J, Schreurs J, et al. Coordinate regulation of immune and inflammatory responses by T cell derived cytokines. FSEB J 1988;2: Silberstein DS, David JR. The regulation of human eosinophil function by cytokines. Immunol Today 1987;8: Sanderson CJ, Campbell HD, Young IG. Molecular and cellular biology of eosinophil differentiation factor (IL-5) and its effect on human and mouse B cells. Immunol Rev 1988;102: Lopez F, Sanderson CJ, Gamble JR, Campbell HD, Young IG, Vadas M. Recombinant human interleukin 5 is a selective activator of human eosinophil function. J Exp Med 1988;167: Raghavachar, Fleischer S, Frickhofen N, Heimpel H, Fleischer B. T lymphocyte control of human eosinophilic granulopoiesis of an idiopathic hypereosinophilic syndrome. J Immunol 1987;139: Rothenberg ME, Owen WF, Silberstein DS, Sobermann RJ, usten KF, Stevens RL. Human eosinophils have prolonged survival, enhanced functional properties, and become hypodense when exposed to human interleukin 3. J Clin Invest 1988;81: Lopez F, Williamson DJ, Gamble JR, et al. Recombinant human granulocyte-macrophage colony stimulating factor stimulates in vitro mature human neutrophil and eosinophil function, surface receptor expression and survival. J Clin Invest 1986;78: Silberstein DS, Owen WF, Gasson JC, et al. Enhancement of human eosinophil cytotoxicity and leukotriene synthesis by biosynthetic (recombinant) granulocyte-macrophage colony stimulating factor. J Immunol 1986;137: Owen WF, Rothenberg ME, Silberstein DS, et al. Regulation of human eosinophil viability, density, and function by granulocyte/macrophage colony stimulating factor in the presence of 3T3 fibroblasts. J Exp Med 1987;166: Walker C, Virchow J-C Jr, Bruijnzeel PLB, Blaser K. T cell subsets and their soluble products regulate eosinophilia in allergic and nonallergic asthma. J Immunol 1991;146: Corrigan CJ, Kay B. CD4 T-lymphocyte activation in acute severe asthma. Relationship to disease severity and atopic status. m Rev Respir Dis 1990;141: Walker C, Bode E, Boer L, Hansel TT, Blaser K, Virchow J-C Jr. llergic and nonallergic asthmatics have distinct

10 J LLERGY CLIN IMMJNOL VOLUME 94, NUMBER 2, PRT 1 Virchow et al. 249 patterns of T cell activation and cytokine production in peripheral blood and bronchoalveolar lavage. m Rev Respir Dis 1992;145: Hamid Q, zzawi M, Ying S, et al. Expression of mrn for interleukin-5 in mucosal biopsies from asthma. J Clin Invest 1991;87: zzawi M, Bradley B, Jeffrey PK, et al. Identification of activated T lymphocytes and eosinophils in bronchial biopsies in stable atopic asthma. m Rev Respir Dis 1990; 142: De Monchy JG, Kaufmann HF, Venge P, et al. Bronchoalveolar eosinophilia during allergen-induced late asthmatic reactions. m Rev Respir Dis 1985;131: Henriksen JM. Exercise-induced bronchoconstriction: seasonal variation in children with asthma and those in rhinitis. llergy 1986;41: Gleich GJ. The eosinophil and bronchial asthma: current understanding. J LLERGY CLIN IMMUNOL 1990;85: Virchow J-C Jr, Holscher U, Virchow C Sr. Sputum-ECP levels correlate with parameters of airflow obstruction in patients with asthma. m Rev Respir Dis 1992;145: Frigas E, Loegering D, Solley GO, Farrow GM, Gleich GJ. Elevated levels of eosinophil granule major basic protein in the sputum of patients with bronchial asthma. Mayo Clin Proc 1981;56: Cockcroft DW, Kilian DN, Mellon JJ, Hargreave FE. Bronchial reactivity to inhaled histamine: a method and clinical survey. Clin llergy 1977;7: Ruihle K-H, Matthys H. Kritische uswahl von Sollwerten fair ein Computerprogramm zur Routinelungenfunktionsdiagnostik. Pneumologie 1976;153: Hansel T, de Vries IJM, Iff T, et al. n improved immunomagnetic procedure for the isolation of highly purified human blood eosinophils. J Immunol Methods 1991 ;145: Walker C, Kaegi MK, Ingold P, et al. topic dermatitis: correlation of peripheral blood T cell activation, eosinophilia, and serum factors with clinical severity. Clin Exp llergy 1993;23: Menz G, Petri E, Lind P, Virchow C. House dust mite in different altitudes of Grisons. EXS 1987;51: Platts-Mills TE. Environmental factors in the inflammation of asthma. In: Godard P, Bousquet J, Michel FB, eds. dvances in allergology and clincial immunology. Park Ridge: Parthenon Publishing, 1992: vailability of JOURNL Back Issues s a service to our subscribers, copies of back issues of the The Journal of llergy and Clinical Immunology for the preceding 5 years are maintained and are available for purchase from the publisher, Mosby-Year Book, Inc., at a cost of $9.50 per issue. The following quantity discounts are available: 25% off on quantities of 12 to 23, and one third off on quantities of 24 or more. Please write to Mosby-Year Book, Inc., Subscription Services, Westline Industrial Drive, St. Louis, MO , or call (800) or (314) for information on availability of particular issues. If unavailable from the publisher, photocopies of complete issues are available from University Microfilms International, 300 N. Zeeb Rd., nn rbor, MI (313) '