Background Milroy and Milroy-like disease are rare disorders characterised by congenital lymphoedema caused by dysfunctional lymphatic vessel formation. Loss of extracellular response mediated by vascular endothelial growth factor receptor 3 (VEGFR-3) is associated with Milroy disease, and VEGFR-3 gene is mutated in around 70% of the cases diagnosed. The only genetic alteration known to be associated with Milroy-like disease was recently identified in a family with a frameshift mutation in vascular endothelial growth factor C (VEGFC) gene, which encodes a VEGFR3 ligand.
Methods and results We report a newborn patient with an external phenotype consistent with Milroy disease and a truncating mutation (p.R210X) in the VEGFC gene detected by exome sequence analysis. Subsequent analysis, by lymphoscintigraphic scan, performed for research purposes, allowed us to correct the diagnosis, confirming patient’s disease as Milroy-like. The mutation segregates with the phenotype in the family according to a dominant model with full penetrance.
Conclusions The clinical presentation, similar to Milroy disease, indicates an overlapping of the external phenotype of both diseases, suggesting that genetic analysis of VEGFC would be useful in diagnosing patients that present with Milroy features but have no mutation in VEGFR-3. Establishing a well-defined genetic pattern would help with differential diagnosis.
- Clinical Genetics
- Genetic Screening/Counselling
Statistics from Altmetric.com
Milroy disease (hereditary lymphoedema type IA; MIM 153100) is a rare autosomal dominant disorder characterised by congenital lymphoedema, caused by a dysfunctional lymphatic vessel formation due to a loss of extracellular response mediated by vascular endothelial growth factor receptor 3 (VEGFR-3, MIM 136352). VEGFR-3 is the only gene associated with this disease, and approximately 70% of the patients diagnosed with this condition carry mutations in this gene.1 Mutations in VEGFR-3 show incomplete penetrance and variable expression, making genotype-phenotype correlations difficult.1
Milroy-like disease is characterised by symptoms similar to those in Milroy disease; however the former exhibits a distinct aetiology. Its differential diagnosis, according to Connell et al,2 is primarily based on presence/absence of VEGFR-3 mutations. The lymphoscintigram in milroy disease is consistent with a functional aplasia.3
VEGFR-3 encodes a transmembrane receptor involved in the lymphangiogenesis. This receptor mediates the transduction of extracellular signals induced by vascular endothelial growth factor C (VEGFC, MIM 601528) and vascular endothelial growth factor D (VEGFD, MIM 300091), activating intracellular pathways that trigger lymphangiogenesis. While both ligands bind the receptor, VEGFC but not VEGFD is indispensable for the development of the lymphatic system.4 ,5
The association of VEGFC with Milroy disease was previously studied in a cohort of patients with typical Milroy disease; however, no mutations in the VEGFC gene were identified.6 Recently, a frameshift mutation in VEGFC was identified by exome sequencing in a patient with Milroy-like disease.7 In vitro and in vivo assays for the biological function of VEGFC activity indicated that a frameshift mutation significantly reduces or completely abolishes the activity of the protein. The authors concluded that the frameshift mutation in VEGFC underlies the disease mechanism of action in this subset of patients and suggested additional study of VEGFC in Milroy-like disease patients lacking VEGFR3 mutations.7
We report a family with congenital lymphoedema, where the proband was a macrosomic female baby born after a 41-week gestation period with marked oedema on the dorsal and external regions of the lower limbs, which was more intense on the left side. This asymmetric pattern persisted over the course of the study. An echocardiogram did not reveal any cardiac aberrations, and proteinuria test was negative. One month after birth, the patient presented with lymphoedema with a positive pitting test. Two months after birth, she had a harmonic macrosomia and presented with lymphoedema with pitting in the left foot with minimal fibrosis in the leg and minimal oedema in the right foot (figure 1A). Three months post-birth, after 1 month of rehabilitation, the oedema had improved and the patient presented with soft oedema without fibrosis predominantly on the left foot and the back and lateral face of the left leg. The cranial ultrasound was normal. In the following months the oedema persisted without marked changes. At 20 months old, oedema, without pitting, was observed in the lower left leg in the ankle and dorsum of the foot (figure 1B). The toenails appeared upslanting and deep toe creases were observed. The patient also presented with papilloma and cellulitis, though itching and pain were not present.
The proband is the second of two daughters born to unrelated Caucasian parents. The father (figure 1C) and grandmother exhibit similar symptoms. Family history of the disease dates back to the patient's great-grandmother and her three sisters, as relayed by the grandmother; but no medical records exist. The disease is not present in the mother's side of the family (figure 2C). Clinical findings are summarised in the online supplementary table S1.
The proband's father (Patient III:3 in figure 2C) is 38 years old. He is the first son of a two children born to unrelated Caucasian parents. He presented at birth with a hydrocoele and an oedema in the lower right limb from the knee to the foot, and presents with deep toe creases (figure 1C) that have been treated with physiotherapy. He has no handicap, but his oedema is aggravated when running or walking. Lymphoscintigraphic assessment of uptake and transit through the lymphatic system revealed no main-tract filling in the lower right limb (figure 1D).
The proband's grandmother (Patient II:2, figure 2C) developed oedema of her left leg in her 30 s. She suffers of oedema in the left leg and presents with cellulitis, upslanting toenails and deep toe creases. Cellulitis is also presented in her leg. In addition to her one son, two other pregnancies resulted in a miscarriage and a child that died at birth, but there was no clinical report noting oedema in the latter. However, history of recurrent miscarriages has been reported1 related to Milroy disease.
DNA was extracted from blood samples using Promega reagents according to the manufacturer's protocol. Exome capture was performed on the proband DNA using the Agilent SureSelect Human All Exome 50 Mb kit. Exome sequencing was performed using the SOLID 5 XL instrument, resulting in ∼94% recovery at >30×coverage. LifeScope (Life Technologies) and Genome Analysis Toolkit (GATK) were used for SNP and indel variant identification,8 and ANNOVAR tools were used for annotation.9 Variants that were represented with an allele frequency of more than 1% in dbSNP, 1000Genomes or our internal exome data set were filtered out.
Exome analysis of the sample revealed a heterozygous nonsense mutation that caused a stop codon in the exon 4 of the VEGFC gene (c.C628T (p.R210X); RefSeq accession number NM_005429) (figure 2A). This mutation is predicted to cause a premature termination of the protein and to be disease-causing by the Mutation Taster prediction algorithm (http://www.mutationtaster.org). The mutation was confirmed by Sanger sequencing using an ABI 3730XL capillary sequencer (Applied Biosystems) (figure 2B). Genetic analysis of the family members confirmed that the mutation is present in symptomatic relatives and absent in asymptomatic relatives (figure 2C): the proband's sister (Patient IV:3) and uncle (Patient III:2) do not present any of the indicated clinical characteristic compatible with Milroy disease, and they do not carry the mutation.
We present a case of a three-generation family with an external phenotype of typical Milroy disease without mutations in VEGFR-3, where a stop codon mutation in VEGFC was detected. All carriers of the novel VEGFC mutation exhibit the clinical diagnostic criteria for Milroy disease, including lower-limb swelling at birth or soon thereafter and upslanting toenails.
To date, only one family with Milroy-like disease and carrying a frameshift mutation in the VEGFC gene has been reported.7 Until that report, patients with lymphoedema symptoms resembling the Milroy phenotype, but who lack a family history of Milroy disease and do not carry a VEGFR-3 mutation, were characterised as having Milroy-like disease due to the unclear inheritance pattern of the phenotype.2
In the previous7 and the present studies, the patients have family history, an early age of onset and fulfil the clinical diagnostic criteria for primary lymphoedema. As Gordon et al7 states, scan appearances of the lymphoscintigraphy examination were similar but not typical of those seen in patients with Milroy disease, so they described their patients as having Milroy-like disease. Our family showed reduced uptake with evidence of rerouting similar to the previous description. At this point the possibility exists of a mild form of Milroy or Milroy-like disease phenotypically distinct from the VEGFR-3 related Milroy disease, but at the moment there are only two described families with mutation in VEGFC gene. Even though all carriers of the novel VEGFC mutation fulfil the clinical diagnostic criteria for Milroy disease, they all show a non-typical lymphoscintigraphic scan.
These observations coupled with the overlapping phenotypes of Milroy and Milroy-like disease emphasise the need for scan test or the establishment of clear genetic patterns to distinguish these diseases to enable accurate diagnoses.
The overlapping phenotype can be explained by the degree to which VEGFR-3 signalling is perturbed due to mutations in VEGFR-3 or VEGFC, and other factors that have been shown to regulate lymphangiogenesis, such as the activity of Notch signalling.10
We report a new clinical case of a Milroy-like disease with a stop mutation in the VEGFC gene that is located in the same exon and causes the same protein alteration reported by Gordon et al.7 The previously described frameshift mutation leads to a predicted premature stop codon 10 amino acids further downstream from codon 191, while patients in this study harbour a stop codon at position 210. The predicted truncated proteins are 201 and 210 amino acids long, respectively. Both mutations are localised in the VEGF-homology domain of the VEGFC protein, which is essential for the binding to VEGFR-3 and VEGFR-2.11 Furthermore, ‘in vitro’ assays suggest that the secretion of truncated protein is strongly impaired compared to wild-type VEGFC.7 Accordingly, stop mutations in this region are presumed to cause haploinsufficiency of VEGFC as Gordon et al7 indicated.
The external clinical presentation of patients with known VEGFC mutations is similar to that of typical Milroy disease, but differs in the incomplete functional aplasia in the lymphoscintigraphic scan. The data suggest that mutations in this ligand of VEGFR3 cause a distinctly, possibly milder, phenotype than mutations in the VEGFR3 receptor itself. As Connell et al indicate, VEGFC screening should be performed in patients with ‘Milroy-like’ lymphoedema. VEGFC screening in patients that exhibit congenital lymphoedema but do not carry a known VEGFR-3 mutation will allow further characterisation of this phenotype and should contribute to the genotype-phenotype discussion.
The similarity of phenotypes among VEGFC mutation-positive patients with Milroy-like disease and patients with Milroy disease complicates differential diagnosis of the diseases and necessitates a lymphoscintigraphic scan or a genetic test to confirm the diagnosis. Although a previous study did not detect VEGFC mutations in patients with Milroy disease, this case study supports VEGFC mutations as a potential hallmark of Milroy-like disease. As a result, VEGFC screening may be useful in differential diagnosis of patients with Milroy symptoms, and could improve upon existing diagnostic criteria that are based on age of onset and family history of the disease.
We thank the patient and his family for agreeing to participate in this study.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online supplement
Contributors The study was conceived and designed by EB-B and FB. MJF-S, AP-M, CG-M and MLC reviewed the clinical findings, submitted photographs on patients, reviewed the draft and revised the paper. The bioinformatics pipeline was performed by JA and the NGS analysis by FB. Sanger sequence confirmation was performed by RL. The manuscript was written by EB and FB. All aspects of the study were supervised by AC and FB.
Competing interests None.
Patient consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.