Genotype/Phenotype Correlations in 103 Children from 87 Families with Hereditary Spherocytosis

Hereditary spherocytosis (HS) is the most common congenital hemolytic anemia in Caucasians. Patients with HS show a high degree of phenotypic variability from asymptomatic to transfusion dependence. Much of this variability stems from HS not being a single uniform disorder but instead being a collection of disorders involving mutations in 5 different genes (ANK1, SPTB, SPTA1, SLC4A1, and EPB42 ) encoding for the 5 major cytoskeletal proteins in red blood cells (ankyrin, β and α spectrin, band 3, and protein 4.2 respectively). These proteins are responsible for maintaining the biconcave disk morphology of red blood cells.

Traditionally the diagnosis of HS has been made without genetic testing. We believe that knowledge of the HS subtype may impact future clinical management decisions, e.g. what type of splenectomy to perform-partial vs total and impact on genetic counseling needs. As a result we are now pursuing genetic testing on all our patients with HS.

Over the past 16 years at Sick Kids Hospital (Toronto, Canada), we have followed 257 children with HS. In the past year we have also been offering genetic testing (following informed consent) to all children (<18 years old) still being followed in our center. DNA is sent to Prevention Genetics (Wisconsin, USA) where Next Generation sequencing is performed. Mutations are verified by Sanger sequencing.

In total, we have genetic test results on 103 children from 87 families to date. Mutations in ANK1 were most common (50 children/43 families); followed by mutations in SPTB (32 children/26 families); SLC4A1 (7 children/7 families) and finally SPTA1 (6 children/3 families) EPB42 mutations were not found to be causative for HS in our cohort. In 8 children (8 families) no causative mutation for HS was found. In 23% of children multiple mutations were found -particularly heterozygous mutations of SPTA1 in combination with ANK1 or SPTB mutations.

The majority of mutations in our patients were nonsense, frameshift, or splice site mutations. Most were novel - not previously described mutations, and unique to families. There were 5 exceptions to this; 1) c.4339-99C>T in SPTA1 (referred to as the α^LEPRA mutation) in 4 children/2 families with an α spectrin form of HS, and in another 2 children (1 family) with an ankyrin form of HS; 2) an ANK1 c.5097-33G>A (8 children/6 families); 3) an ANK1 c.1405-9G>A (4 children/3 families); 4) SPTB c.6037C>T (3 children/2 families); and 5) SPTB c.5266C>T (3 children/2 families). The ANK1 c.5097-33G>A mutation had not been previously identified and yet was the most frequently detected mutation in our HS population.

Most children (68/103) were found to have autosomal dominant (AD) HS: 33 children with the ankyrin subtype, 26 with the β spectrin subtype, 4 with the band 3 subtype and 5 in whom no mutation could be found but where there was a clear history of AD inheritance. Autosomal recessive (AR) inheritance was confirmed in 7 children - all 6 with α spectrin form of HS and 1 with a β spectrin form of HS. Twenty-eight children were spontaneous new mutations for HS: ANK (n=17), β spectrin (n=5), band 3 (n=3) and 3 in whom no mutation could be found. In all AR forms of HS, the index case had initially been thought to be a spontaneous new mutation for an AD form of HS; genetic testing resolved the inheritance pattern.

Patients were categorized according to disease severity, primarily on the basis of their need for transfusions (a reflection of baseline hemoglobin) and splenectomy. The proportion of children that have required transfusions and needed splenectomies was: 83%/83% (α spectrin); 48%/18% (ankyrin), 23%/10% (β spectrin) and 29%/0% (band 3).

Most children in our center undergoing splenectomy have undergone partial splenectomy; few have required subsequent total splenectomy. However, of the 5/6 children with α spectrin form of HS that underwent partial splenectomy 3 have subsequently needed, or are being considered for, total splenectomy. This suggests that due to the extreme severity of the α spectrin subtype of HS, children with this form of HS may not do well long-term with partial splenectomy and are likely to eventually require a total splenectomy.

Our study to date represents one of the largest and most comprehensive genetic analyses of a cohort of HS patients. Our findings will add to the growing understanding of the disease, and will be important to provide comprehensive genetic counseling and possibly in the future to guide management of selected cases.