Measurement of cerebrospinal fluid orexin‐A (hypocretin‐1) by enzyme‐linked immunosorbent assay: A comparison with radioimmunoassay

Orexin (hypocretin) is a neuropeptide that plays an important role in arousal state and food intake, and was discovered in 1998.1, 2 The cerebrospinal fluid (CSF) orexin-A (hypocretin-1) level is a biomarker used to diagnose narcolepsy type 1 (NT1).3 Radioimmunoassay (RIA) using a commercial kit (Phoenix Pharmaceuticals, Burlingame, CA, USA) is currently the standard technique for measuring CSF orexin-A. However, RIA requires expensive measuring equipment. In addition, RIA involves the use of radioactive materials and requires specific precautions for the use and disposal of them.4 Therefore, development of a simpler and less expensive measurement method is needed. In this study, we measured CSF orexin-A using an enzyme-linked immunosorbent assay (ELISA) kit (Wako Pure Chemical Industries, Osaka, Japan, under development) and compared it with the measurements obtained by RIA. We selected 80 CSF samples from patients with various hypersomnolence conditions from samples in which orexin-A levels had been previously measured by RIA, and we re-measured them by ELISA (Table S1). CSF orexin-A was detectable by ELISA as well as RIA. We found a significant positive correlation between the ELISA and RIA results (Spearman correlation: ρ = 0.734, P < 0.01, n = 80; Fig. S3). Although the orexin-A levels were more widely distributed with ELISA, those of NT1 patients were lower than those of hypersomnolence patients with normal orexin-A levels (>200 pg/mL) according to RIA. On average, the ELISA results were about fourfold lower than the RIA results. Thirteen of the 56 cases with normal orexin-A levels (>200 pg/mL) according to RIA were below the normal range (<51 pg/mL) with ELISA (>51 pg/mL is normal range for ELISA; Fig. S4); thus, there were discrepancies between RIA and ELISA results. There was no specific hypersomnolence condition that had a significantly different occurrence of these discrepancies. In NT1, there were no discrepancies between the RIA results and the ELISA results (Table S2). Although RIA likely detects only a single epitope of orexin-A, the sandwich ELISA we used needs two different epitopes of orexin-A. Therefore, ELISA likely detects a specific length of the orexin peptide/fragment. This difference may be the reason why the orexin-A measured by ELISA is fourfold lower than that by RIA. Further evaluations of ELISA, especially for the discrepant cases, are warranted to understand the applications of this new method and orexin dynamics in patients with hypersomnia. We attach the details of this study as supporting information (Document S1, Figs S1–S4, Tables S1–S2). Informed consent was obtained from all participants and participants' anonymity has been preserved. This study was approved by the ethics committee of Akita University Graduate School of Medicine (No. 1801), and conforms to the provisions of the Declaration of Helsinki (as revised in Brazil, 2013). The authors declare no conflicts of interest. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.