Background In recurrent pregnancy loss (RPL), the pathogenesis of the majority of cases remains to be explained. Antiphospholipid, syndrome (APS) is one of the disorder responsible for causing RPL and its overwhelmed complement activation recognized as a major pathogenic mechanism. Autoantibodies against complement component 1q subcomponent (aC1q) have been shown to associate with complement activation in primary APS, but the relevance of aC1q in RPL is still unclear. We hypothesized that aC1q would be associated with the pathogenesis of RPL in patients with or without APS, especially in RPL of unknown etiology.
Objectives The aim of this study was to explore the significance of aC1q in RPL.
Methods As a clinical study, we conducted a retrospective cross-sectional study comprising a total of 134 patients with RPL of unknown etiology, 27 with obstetric APS (OAPS), 14 parous patients with connective tissue disease (CTD) without historical obstetric/thrombotic complications and 17 parous healthy controls (HC). Serum levels of aC1q were measured using a solid-phase ELISA (Buhlmann Laboratories AG, Switzerland) and defined as positive using cut-off value of more than 15 U/mL according to the manufacturer. In murine model, 8–12 week-old female BALB/c mice were mated with isolated males and the presence of vaginal plug was defined as day 1 of pregnancy. Mice were treated with intravenous injections of anti-mouse C1q monoclonal antibody (JL-1), isotype control IgG2b or PBS. To block C5a receptor (C5aR), mice were intravenously pre-treated with anti-C5aR antibody, 30 minutes before the injection of JL-1 on day 8. Mice were sacrificed on day 16 of pregnancy and fetal resorption ratios, weight of fetuses and placentas, serum levels of C3a and immunohistochemical staining of complement components on placental tissue were compared among each group.
Results Among RPL, OAPS, CTD and HC, 47 (35%), 8 (30%), 3 (21%) and 2 (12%) were positive for aC1q, respectively. In RPL patients, aC1q was more prevalent (p<0.05) and its titer was significantly higher than in HC (median and interquartile range [IQR] 12 [8–21] vs. 0 [0–4.3], p<0.0001) (Figure 1). In murine model, fetal resorption ratio was higher (p<0.01), weight of fetuses and placentas lower (p<0.05), and serum levels of C3a higher (p<0.01) in mice treated with JL-1 than in control mice. Immunohistological findings showed that complement components were more deposited on placenta in JL-1 treated mice than in control mice. Furthermore, the additional blockade of C5aR cancelled the pathogenic changes in JL-1 treated mice.
Conclusions Clinical findings showed that aC1q could be relevant to RPL. Moreover, we have established aC1q induced pregnancy loss model mice. Our study indicates that aC1q has a pathophysiologic role in RPL and that anticomplement therapy might be effective for at least some groups of patients with RPL for whom specific treatment remains to be established.
Disclosure of Interest None declared