Serum interleukin-1 beta is increased in cluster headache Paolo Martelletti, Massimo Granata, Mario Giacovazzo Headache Centre, Institute of Internal Medicine VI, University "La Sapienza", Rome, Italy Cephalalgia Martelletti P, Granata M, Giacovazzo M. Serum interleukin-1 beta is increased in cluster headache. Cephalalgia 1993;13:343-5. Oslo. ISSN 0333-1024 We measured serum interleukin-l beta (IL-1 beta) in 24 episodic cluster headache (CH) patients and 45 normal controls using a specific ELISA method. There was an increase in IL-1 beta in all CH patients compared to controls. IL-1 beta was further inc reased during the ictal phase of CH compared to patients between attacks and normal individuals. Between attacks, IL-1 beta was also significantly increased compared to controls. We suggest that these results represent an activation of the immune sys tem in CH. Paolo Martelletti, Centro Cefalee, Istituto di Clinica Medica VI, Università "La Sapienza", Viale del Policlinico, 00161 Rome, Italy In recent years there has been increased interest in the neuroimmunomodulatory system in cluster headache (CH) (1-5). Our laboratory has reported impaired function of certain cytokines, peptides actively secreted during the immune response. With the use of immunopeptide mapping we have shown decreased expression of interleukin-2 (IL-2) receptors on the surface of lymphocytes (3), and, in contrast, recombinant IL-2 induced upregulation of cytotoxicity (4). Beta-interferon caused a similar increas e on cytotoxicity as IL-2, but together these compounds had no synergistic effect (5). Here we report an increase of the interleukin-1 beta (IL-1 beta) in CH patients and discuss the importance of this finding. Material and methods Patients The study was conducted from March to November 1989 in 24 informed and consenting outpatients suffering from CH (21M, 3F, mean age 46.4 ± 7.9 SD). The approval of our institution's ethics committee was obtained prior to the study. The diagnosis of CH was made according to the criteria of the IHS Classification (6). Patients consecutively diagnosed in our outpatient Headache Centre were entered into the study in a manner that provided equal numbers in cluster crises and between cluster crises. Fo rty-five healthy blood donors (39M, 6F) matched for age (mean age 43.4 ± 9.1 SD) were used as the control group. Subjects received no medication in the 72 h prior to morning blood sampling (08.30). The sera obtained were stored at -70°C until assay ( May 1990 for controls and November 1990 for patients). IL-1 beta assay Serum IL-1 beta was measured by specifically directed enzyme-linked immunoassays (ELISA), trapping the IL-1 beta between a specific mouse mono-clonal anti-human IL-1 beta (antibody 2D8) antibody and the corresponding rabbit polyclonal anti-human IL-1 beta antibody, as previously described (7). The assay was further developed with a peroxidase conjugated donkey anti-rabbit immunoglobulin anti-serum (Amersham) and the peroxidase substrate o-phenylendiamine. The sensitivity of the assay for IL-1 be ta was 10 pg/ml. The IL-1 beta concentration was determined from a standard curve obtained with serial dilutions of known standards (recombinant human IL-1 beta from Biogen, Geneva) with a linearity from 100 to 2000 pg/ml. Values less than 16 pg/ ml were regarded as undetectable, values up to 50 pg/ml as minimally detectable and values of 50-100 pg/ml as low level detection based on the curve linearity. Immunoprecipitation of IL-1 from 35-S-methionine-labelled monocytes with the various antibodi es used in the assays has shown that these antibodies are able to recognize the mature 17 KD IL-1 as well as the 31 KD precursor molecule. Statistics One-way ANOVA and Student's t -test for paired data were used to test differences between basal values (CH between crises) and ictal values (CH during cluster crisis) and between basal or ictal values and control group values. Data were expressed as m ean ± SEM. Values of p < 0.05 were regarded as significant. Results Individual values of IL-1 beta in controls and CH patients are reported in Tables 1 and 2 respectively. The CH population as a whole had increased mean values of IL-1 beta when compared to normal individuals (Table 3). There was a marked difference i n mean values between the patients tested during the