Whole Exome Sequencing as a Diagnostic Tool of Primary Complement Component Deciencies: A Multicenter Experience of Three Novel Mutations

Diagnosis of primary complement deciencies requires a high index of suspicion. Thus, susceptible patients are often underdiagnosed and untreated. Here, we present a multi-center experience with three novel inborn errors of the classical complement system. This is a retrospective multicenter analysis of computerized medical records of children (> 18 years) admitted in the period between 2003 and 2018 at and and Patients were genetically diagnosed by a complimentary immune work-up. We identied 5 patients (3 males) from four different consanguineous families harboring three novel mutations in the complement components C6-C8. Genetic mutations were identied by whole exome sequencing. Clinical manifestations consisted of meningitis with or without meningococcemia. The immune work-up demonstrated nearly absent levels of CH50, compatible with a complement pathway defect. The mean diagnosis delay was 10.56 (0–30) years. Conclusion: Invasive meningococcal infections may be life-threatening and cause severe neurological sequela. Awareness of risk factors for primary complement deciencies, even at the rs infectious episode, should facilitate prompt immune and genetic investigations, commencing diagnosis and proper treatment.


Introduction
The prevalence of primary complement component (C) de ciencies among patients with primary immune de ciencies (PIDs) is estimated to be 1-6%, though higher rates have also been reported [1]. Decreased levels of terminal components C5-C9 are known to increase susceptibility to infection with Neisseria spp., including Neisseria meningitides [1]. Moreover, inherited disorders of regulators of the alternative pathway, such as properdin de ciency, are found in patients with recurrent meningococcal infections [1; 2].
Clinical investigation of patients with recurrent meningococcal infections usually warrants screening complement activity by measuring the 50% hemolytic complement activity of serum (CH50) and the alternative pathway hemolytic assay (AH50), followed by quanti cation of C1-C9. This facilitates the treatment of susceptible patients with prophylactic antibiotics and meningococcal vaccination, reducing the risk of future invasive meningococcal infections. However, the diagnosis of complement de ciencies requires a high index of suspicion, especially when treating patients with a single infectious episode [3]. Such patients are often underdiagnosed and untreated until the next episode of meningitis or meningococcemia [4].
The introduction of next-generation sequencing has been a powerful tool in the diagnosis of inborn errors of immunity [5]. In addition, accumulating evidence suggests that whole exome sequencing (WES) is effective in promoting the diagnosis of complement de ciencies, such as C1Q de ciency, which induces the development of systemic lupus erythematosus (SLE) [6; 7].
Here, we present a multi-center experience of ve patients with three novel inborn errors of the classical terminal complement system. We discuss the genetic and immune characteristics of these patients and review the corresponding literature.

Immune work-up
The immune work-up consisted of screening the complement system by measuring CH50, followed by quantifying C1-C9. Asplenia was ruled out by abdominal ultrasonography.

Genetic work-up
High throughput sequencing for whole exome sequencing was performed on genomic DNA samples from the patient. Coding regions were enriched with a SureSelect Human All Exon V5 Kit (Agilent) and then sequenced as 100-bp paired-end runs on an Illumina HiSeq 2500 (Illumina Inc). We used the BWA mem algorithm (version 0.7.15)[8] for alignment of the sequence reads to the human reference genome (hg19 or Hg38). The HaplotypeCaller algorithm of GATK version 3.4 was applied for variant calling, as recommended in the best practice pipeline [9]. KGG-seq v.1.0 was used for annotation of identi ed variants [10] and in house scripts were applied for ltering, based on family pedigree and local dataset of variants detected in previous sequencing projects. All mutations of the complement genes were validated by dideoxy Sanger sequencing in the carriers and healthy control. Data were evaluated using Sequencer v5.0 software (Gene Codes Corporation).
Ethical review of the study This study was approved by the institutional review board committees of Shaare Zedek and Tel-Hashomer Medical Centers.

Clinical characteristics
Patient characteristics are detailed in Table 1. We identi ed 5 patients (3 males and 2 females) from four different families (A-D). Consanguinity was noted in three of the families. Excluding P3, all patients were from a Jewish origin. Family history of recurrent meningococcal infections was notable in three patients (P1, P2, and P4). Representative pedigrees of families A and D are shown in Fig. 1A and B, respectively.  Genetic work-up WES revealed a novel homozygous mutation in C6 (c.1786C > T; p.R596X,) in P1 and P2 ( Fig. 2A). A mutation in C7 (c.1135G > C; p.G379R) was identi ed in P3 and P4 (Fig. 2B). P5 had a novel homozygous mutation in C8b (c.361C > T; p.R121X, Fig. 2C). Interestingly, P5 also had a homozygous mutation in CECAM16 (c.703C > T; p.R235C), which is related to hearing loss [11].These variants were considered rare and pathogenic by different predictive software. Sanger sequencing con rmed the genetic ndings with Mendelian inheritance yielding heterozygous carriers among rst-degree relatives, including parents and siblings.

Treatment and outcome
All patients were prescribed extended meningococcal vaccines and prophylactic amoxicillin. Interestingly, prior to the diagnosis of her C6 de ciency, P2 received prophylactic amoxicillin due to rheumatic fever. This may account for her 18-year delay in diagnosis.
The mean current patient age is 24 (16-37) years. Only one patient (P5) has demonstrated residual neurological impairment in the form of neurosensory hearing loss, which may have complicated his primary genetic hearing loss defect.

Discussion
In this study, we described our experience with primary complement component de ciencies. Using WES, we identi ed three novel mutations in C6, C7 and C8b and con rmed the diagnosis by quantifying CH50 and different components of the classical pathway. Two patients (P1 and P2) with meningococcemia were found to have a novel genetic mutation causing C6 de ciency. C6 interacts with C5b, C7, C8, and C9 to form the membrane attack complex, which catalyzes cell lysis [12]. Terminal complement component de ciencies (C5-C9) are associated with an up to 10,000-fold increased risk of invasive meningococcal infection [4]. High prevalence of C6 de ciency has mainly been reported among the African-American population [12], though it is also seen in other ethnicities, such as Spanish [13], Irish [14], and South-African [14].
Our study found a signi cant delay in diagnosis from the initial meningococcal infection. Diagnosis of C6 and C8b de ciencies in P2 and P5 was delayed by 18 and 30 years, respectively. Interestingly, prophylactic amoxicillin given to P2 due to rheumatic fever may also have a role in the delayed diagnosis due to lack of meningococcal infections while P2 was on prophylactic treatment. This further demonstrates the need for a high index of suspicion in patients with a single meningococcal infection episode.
Our report corresponds with other studies. In one study of 22 adult patients who previously experienced a single event of meningococcal meningitis, 2 were diagnosed with C6 and C7 de ciencies [3]. In South Africa, 12.8% of patients with invasive meningococcal infections have been found to have C5 and C6 de ciencies [15].
However, screening all children presenting with a single invasive meningococcal disease for complement de ciencies is costly and has not been routinely implemented, therefore investigation is directed by diagnostic clues. The presence of consanguinity, a family history of meningococcal infections, and recurrent episodes of invasive meningococcal infection are obvious "red ags" that indicate a need for investigation. Grumach et al. detailed other signs of complement de ciency, including meningococcal meningitis in a patient > 5 years old; meningococcal infections with less common serotypes; the presence of autoimmune disease, such as SLE; angioedema without urticaria; other recurrent bacterial infections; and renal and ophthalmic involvement in complement regulatory protein de ciencies, including factor H de ciency in hemolytic uremic syndrome [1].
The patients in our cohort presented with a single meningococcal meningitis event with or without meningococcemia. The family history and consanguinity in families A, C, and D were indicators to start immune and genetic work-up. Furthermore, identi cation of meningococcal serotype Y in P1 and P2 was also a clue, as the most common serotypes reported in Israel are B and C (76.9% and 10.9%, respectively)[16].
Use of CH50 screening complemented by WES enhanced the diagnosis of some of our patients. We suggest implementing this methodology in patients with a single invasive meningococcal infection as a "red ag" for primary inborn errors of immunity to reduce the diagnostic delay. This will facilitate better patient care, decrease rates of neurological complications and promote early genetic counselling of siblings.
Our study has several limitations, particularly its size and retrospective design. Prospective large cohort studies investigating complement de ciencies using WES and CH50 screening in patients with a single meningococcal infection are needed.
In conclusion, management of a patient with a rst invasive meningococcal infection requires awareness of complement component de ciencies. Upon clinical signs, prompt immune and genetic investigations should be recommended, commencing diagnosis and proper treatment.