Genotype-Phenotype Correlations in Phelan-Mcdermid Syndrome

Background: Phelan-McDermid syndrome (PMS, OMIM# 606232) is characterized by intellectual disability, muscle hypotonia, delayed or absent speech and autism spectrum disorder (ASD). PMS is due to heterozygous chr. 22q13.3 terminal deletions or, more rarely, point mutations involving the SHANK3 gene, crucial to the formation and plasticity of excitatory synapses. Disruption of SHANK3 causes approximately 0.5% of ASD cases, with higher rates reported in ASD co-morbid with ID. Patients with PMS may undergo behavioral regression, with two incidence peaks around 6 years of age and in adolescence.

Objectives: This study aims to assess genotype-phenotype correlations in PMS.

Methods: Fifty PMS patients (2-42 y.o., M:F=1:1) were recruited at the Interdepartmental Program “Autism 0-90” of the “G. Martino” University Hospital (Messina, Italy). All patients underwent medical history collection, neurological examination, behavioral observation, medical work-up and psychodiagnostic testing, including ADOS-2, ADI-R, Leiter-3 or GDMS, VABS-II, VAS, CGI, CBCL, SCQ, SSP, WHOQOL, QOL-A, ABC and RBS-R. Genetic diagnosis was confirmed by array-CGH (400K Kit, Agilent) or by targeted Sanger sequencing in 46 patients, by FISH in 1 case and by karyotype in 3 cases.

Results: All patients received a DSM-5 diagnosis of Intellectual Disability and/or global development delay; 15/50 (30.0%) also satisfied DSM-5 and ADOS criteria for ASD. Moreover, all patients share deficits in gross and fine motor skills, as well as in alternating movement, visuo-motor and bimanual coordination. Deletion size ranges from 25 Kb to 9 Mb, spanning up to 7 autosomal dominant disease genes (SHANK3, SCO2, PLXNB2, UPK3A, SAMM50, PNPLA3, TCF20). Large deletions (>5 Mb), spanning from SHANK3 to UPK3A and beyond, are significantly associated with: (a) female sex (P<0.05), (b) more severe clinical phenotypes and greater developmental delay (P<0.05 for delay in walking and sphincter control, P<0.01 for deficits in receptive language); (c) presence of multiple brain MRI abnormalities (P<0.01). The largest deletions involving also TCF20 always yield a very severe phenotype with extreme muscle weakness, hypotonia, difficult weaning, reduced spontaneous activity and lack of autonomous walking. On the other hand, deletion sizes >0.42 Mb, spanning from SHANK3 to PLXNB2 and beyond, tend to be associated more frequently with renal malformations, but these are occasionally observed also with smaller deletions. Clinical regression occurred in 12/50 (24.0%) patients and may be more frequent in carriers of small deletions encompassing only SHANK3 (P=0.14). In general, large phenotypic variability is observed among carriers of identical or similar deletions.

Conclusions: Interindividual differences in PMS severity may stem from at least three sources: (a) deletion size involving other functional genes in addition to SHANK3, especially PLXNB2 and TCF20; (b) additional mutations, microdeletions or epigenetic influences in the non-deleted allele; (c) greater penetrance of familial genetic loading for neuro-behavioral disturbances in the presence of a SHANK3 synaptopathy. Larger deletions may be rarely compatible with life in male offspring. The medical work-up for PMS should be broad, structured, and consistent, and should not be driven by deletion size, although dominant and recessive genes located in the deleted segment can point toward specific vulnerabilities.