and V), a molecular mass of about 10000 Da would be needed, because ... all repeatability was very good except in Assay ... The best agreement among assays was found between .... Berg S, Brodin B, Heaselvik F, Laurent TC, Mailer R. Ele-.
CLIN.CHEM.38/1, 127-132 (1992)
Seven Different Assays of Hyaluronan Compared for Clinical Utility Ulla Lindqvist,”8 Kenji Chichibu,3 Bertrand Poole,7 and Torvard C. Laurent2
Detpech,4
To compare six assays of hyaluronan (hyaluronic acid; HYA) in serum, developed in different laboratories, we analyzed 10 samples from each of three groups: healthy persons, patients with primary biliary cirrhosis, and patients with rheumatoid arthritis. All the assays are based on the use of affinity proteins specific for HYA, prepared from cartilage or brain tissue, and are analogous to RIA or enzyme immunoassay techniques. The assay results were of the same magnitude. Although statistical analysis indicated that the methods in some cases deviated significantly from one another, this variation was less than the physiological variation in the healthy population. Therefore, the results of clinical investigations in which the various methods have been used are comparable. The analyses have high specificity and sensitivity for primary biliary cirrhosis but are somewhat less suitable for detecting rheumatoid arthritis. A seventh laboratory, which obtained antibodies to HYA, used these in an AlA to analyze a separate series of serum specimens. Results were in agreement with those obtained by one of the other assays. Additional K.yphra..e: ronec4in intermethod rheumatoid arthritis
hyaluronan..binding proteins hyalucompanson pnmaty biliary cirrhosis chromatography,immunoaffinhty enzy.
‘
maticmethods (hyaluronate; hyaluromc acid; HYA) is a found in connective tissues. In the last decade, the development of sensitive assays (1-8) for HYA has allowed researchers to measure its concentration in serum and other body fluids. Investigators are increasingly interested in determining the serum concentration of HYA in (e.g.) people with inflammatory diseases (8-19), liver diseases (15,20-25), malignancies (2,26-30), and septic or other acute conditions (31-33). For example, the HYA concentration can help determine the extent of an inflammatory condition (e.g., rheumatoid arthritis); can signal a decrease in the Hyaluronan polysaccharide
Departments of’ Clinical Chemistry and2 Medical and Physiological Chemistry, University of Uppsala, Uppsala, Sweden. Chugai Pharmaceutical Co. Ltd., Tokyo,Japan. 4Laboratoire d’Oncologie Moleculaire, Centre Henri-Becquerel, Rouen, France. 5Ciba-Geigy Corp.,Pharmaceuticals Division, Research Department, Summit, NJ. 6Department of Biochemistry, Rush-Presbyterian-St. Luke’s Medical Center,Chicago, IL. 7Joint Diseases Laboratory, Shriners Hospital for Crippled Children, and Division of Surgical Research, Department of Surgery, McGill University, Montreal, Canada. 8Address for correspondence: Department of Clinical Chemistry, University Hospital, 8-75185 Uppsala, Sweden. Received September 7, 1990; accepted November 4, 1991.
Ronald L. Goldberg,5 Warren Knudson,6
A. Robin
activity of hepatic sinusoidal cells, where HYA is catabolized; or can indicate the presence of an HYA-producing tumor. The new assays, which are based on the specific affinity of certain proteins for HYA, can be regarded as the third generation of techniques for its analysis. The first generation was based on colorimetric assays of glucuronic acid or glucosamine in purified HYA; the second generation involved the degradation of HYA by specific hyaluronidases, with quantification of the degradation products. The third-generation assays are 100fold more sensitive than the second-generation assays and can determine nanogram amounts of HYA. The specific-affinity proteins used thus far in the assay have been the cartilage proteoglycan or its HYAbinding region and the cartilage link protein (34, 35), the hyaluronectin from brain tissues (36), or HYAspecific antibodies (37). The increasing clinical use of these methods has made it necessary to compare the results obtained by the various techniques. We have investigated whether the various methods give comparable results and if they can be used with the same decision limits in clinical diagnosis. Here we describe the analysis of 30 unknown serum samples by six different assays used in our laboratories. We have also included some data obtained by a seventh technique (8). The results show that the assays give slightly different values, but the deviation is no larger than the variation of the HYA concentration in a control population measured by one technique. MaterIals and Methods
Samples Serum was collected in Sweden from 10 healthy individuals, 10 patients with primary biliary cirrhosis, and 10 patients with rheumatoid arthritis. The diagnosis of primary biliary cirrhosis was based on the presence of anti-mitochondrial antibodies, normal extrahepatic bile ducts on cholangiography, and liver-function tests that indicated cholestasis. Rheumatoid arthritis was diagnosed according to the criteria of the American Rheumatism Association. Each specimen was split into six samples, sodium azide (0.2 g/L, final concentration) was added as a preservative, and all samples were stored at -20 #{176}C until used. One sample from each specimen was sent frozen via express freight to laboratories in France (Assay ifi), Japan (Assay V), Sweden (Assays I and II), and the U.S. (Assays IV and VI) for blind analyses. The determinations were performed in duplicate in each laboratory. Serum samples that had been analyzed in Canada by a radioimmunoassay involving rabbit antibodies reacCLINICALCHEMISTRY, Vol.38, No. 1, 1992 127
tive with unsaturated oligosaccharides of HYA (Assay VII) had been stored frozen at -20 #{176}C for four years. They were shipped to Uppsala and analyzed by Assay II. HYA Assays
For further details of the following assays, see the specific references. Some characteristics of the assays are compared in Table 1. Assay 1(3). This assay, originally described by Tengblad (1), has been modified (38) for use in serum analyses (3). It is based on use of a mixture of the HYAbinding region of the cartilage proteoglycan and the cartilage link protein. The proteins are labeled with 1251 and partitioned between free HYA in solution and HYA-substituted Sepharose#{174} gel (Pharmacia, Uppsala, Sweden) during a 20-h equilibration period. The gel is then pelleted and washed, and its radioactivity is measured in a gamma-counter. Increasing the amount of the free HYA in the incubation mixture decreases the radioactivity in the pellet. Assay 11(4). This assay is developed from Assay I but uses only the purified HYA-binding region of the cartilage proteoglycan. The HYA-substituted Sepharose gel with a particle diameter of 100 m is replaced by an HYA-substituted agarose gel with a particle diameter
