A case of systemic lupus erythematosus with C1q deficiency, increased serum interferon-α levels and high serum interferogenic activity

C1q deficiency associates with high serum IFN-α levels and high serum interferogenic activity. Sir, In this report we present a case of severe SLE on the basis of inherited C1q deficiency, and present additional support for the role of IFN-α in the pathogenesis of the disease. Our patient was diagnosed with SLE at 3 years of age with a history of butterfly rash, lupus nephritis World Health Organization (WHO) class II and pleurisy. He was treated with prednisone and HCQ, and the disease went into remission. At the age of 24 years, after a Salmonella abony-caused enteritis and meningitis, a disease flare was triggered with rapid weight loss, leukopaenia and neurological symptoms with a paretic bladder, hyperreflexia in lower extremities, positive Babinski’s sign and decreased sensation from the hips and below. He was referred to Uppsala University Hospital. Laboratories showed speckled ANA with positive SS-A/Ro52, SS-A/Ro60, SS-B, RNP and Sm antibodies. Anti-dsDNA antibodies were negative. Classical complement function was 8% of normal function and a western blot on serum from the patient showed complete absence of C1q protein, confirming the diagnosis of C1q deficiency (supplementary Fig. S1, available at Rheumatology online). A list of baseline laboratory and imaging results is provided in supplementary Table S1, available at Rheumatology online. Because of progressing neurological symptoms and emergence of radiological signs of myelitis from level Th1 and below, treatment was given with high-dose prednisone, IVIg, CYC, plasmapheresis, MMF and fresh frozen plasma (supplementary Table S2, available at Rheumatology online) [1]. The efficacy of plasma infusions remained for almost 2 years, but when anti-C1q antibodies occurred, plasma infusions was discontinued. To identify the underlying mutation, the genes encoding the subunits C1qA, C1qB and C1qC of the C1q molecule were sequenced by Sanger sequencing, revealing a premature termination codon mutation rs377549148, R69X, located in the third exon of the C1qC gene (supplementary Fig. S2, available at Rheumatology online). The patient was homozygous for this mutation resulting in non-functional C1qC subunit, while both parents (with no known consanguinity) and a healthy brother were heterozygous (supplementary Fig. S3, available at Rheumatology online). The R69X mutation is previously known in two families originating from the same central European region. The phenotype described in association with this mutation was similar to that in our patient with malar rash, discoid rash and oral ulceration, and positivity for ANA, anti-RNP and anti-Sm [2]. To assess whether other previously known SLE-associated gene variants might contribute to the disease pathogenesis beyond the C1q deficiency, we genotyped the 15 most robustly SLE-associated variants in 13 genes (supplementary Table S3, available at Rheumatology online). We found that 10 variants were either homozygous or heterozygous for the risk alleles and among those we noted homozygosity in the IRF5 risk gene variant that correlates to high serum IFN-α activity [3]. Next, serum samples collected at 21 occasions during a period of 2 years were analysed for IFN-α by a dissociation-enhanced lanthanide fluoroimmunoassay [4]. High levels of IFN-α were detected in 18/21 serum samples, and correlated well with disease activity measured by the SLEDAI score (Fig. 1). Correlation between IFNα levels and disease activity High IFN-α levels were found in patient’s serum, and correlated well with disease activity measured by SLEDAI score. *Fever, pulmonary infiltrates, rash, arthritis, mucosal ulcers. **Fever, rash, cutaneous vasculitis, arthritis, leukopaenia. Arrow: fresh frozen plasma infusion. Dashed arrow: Time point of appearance of C1q antibodies (19 U/ml, reference <14). The C1q antibody titre subsequently increased to >100 U/ml. In order to clarify the capacity of patients’ sera to induce IFN-α production in peripheral blood mononuclear cells (PBMCs), such cells were isolated from two healthy blood donor buffy coats by Ficoll-Paque density gradient centrifugation. PBMCs were monocyte-depleted in order to enhance the IFN response. High levels of IFN-α (mean 1958 U IFN-α/ml) were generated when these cells were stimulated with 10% patient serum in combination with necrotic cell material as previously described [5]. When the patient’s own PBMCs were stimulated in the presence of 10% autologous serum and necrotic cell material, 14 U IFN-α/ml was produced, which was markedly lower than the level produced by PBMCs from a healthy control (177 U IFNα/ml), also stimulated with patient sera. There are several possible causes of this discrepancy, for instance low frequency of plasmacytoid dendritic cells among the patient’s PBMCs (0.1%) compared with the healthy donor’s PBMCs (0.6%), reduced IFN-producing capacity of patient’s plasmacytoid dendritic cells due to treatment or concurrent infections, or a refractory state due to in vivo activation. C1q has been shown to inhibit IFN-α production in plasmacytoid dendritic cells, supporting the theory of a regulatory function of C1q [6, 7]. In our case, high levels of serum IFN-α during an extended period of time were detected and correlated well with disease activity, further supporting the role of IFN-α in disease pathogenesis. We therefore propose that a genetic C1q deficiency is classified as a Mendelian type I interferonopathy, which has been discussed previously [8]. Furthermore, a severe SLE phenotype in a patient with prepubertal onset of the disease should raise the suspicion of monogenic cause of the disease, even if the initial treatment is successful. Funding: This work was supported by grants from the Swedish Research Council for Medicine and Health (D0283001), the Swedish Rheumatism Association, King Gustav V’s 80-year Foundation, the Swedish Society of Medicine and the Ingegerd Johansson donation. Disclosure statement: The authors have declared no conflicts of interest.