This is the first study that describes phenotypical commonalities and variation in a large cohort of families representing 50 patients with non-classic Pompe disease. We confirmed the presence of phenotypical commonalities within sibships, including: symptom onset either in childhood or in adulthood, similar presenting symptoms (in 70% of families), and the occurrence of particular symptoms such as ptosis and bulbar weakness. However, the rate of disease progression varied substantially within some families and also between families with the same GAA genotype.
Infants with classic infantile Pompe disease show a consistent phenotype, also within sibships, while the clinical spectrum of non-classic Pompe disease is much more variable regarding onset and progression of the disease
[3, 6–8, 12]. Since siblings share the same set of GAA mutations and are genetically more related than non-relatives, we expected to find less variation in phenotype within sibships compared to the overall Pompe population. In our 22 Dutch families the presenting symptoms were often the same. As limb girdle weakness is known to be the most frequently occuring initial symptom in Pompe disease, it is not surprising this was found frequently in all siblings of a family
[3, 5–7, 12]. Other symptoms also seemed to cluster within families. Remarkably, bulbar weakness was a presenting symptom in all three siblings of one family. Bulbar weakness has been reported in approximately 25% of adult Pompe patients, but rarely as first symptom (1-2%)
[6, 12, 25]. It was also found in all siblings of a second family but later on in the course of the disease. The same applies to ptosis
[26–28]. Ptosis has been reported in 23% of the Dutch adults with Pompe disease
. In our cohort of families we found the same percentage. In four families ptosis was present in all siblings. All carried the c-32-13 T > G mutation, but the second GAA mutation differed between the families. The same applied to the two families with bulbar weakness, suggesting a role for (epi)genetic factors in the clustering of symptoms within families.
We also found differences in phenotype within the same family. The median difference in symptom onset between siblings was nine years, with extremes of 20 to 31 years in three families. Since time of symptom onset was based on patients reporting their own history, there may be a recall bias leaving some uncertainty in the differences between time of symptom onset between siblings. However, it is inevitable that large differences will occur within some families. The majority of wheelchair and/or ventilator-dependent patients had an ambulant or non-ventilated sibling, while they had already been using these supportive measures when they were the same age. The duration of the disease (time from symptom onset) could not always explain this difference, since a third of the patients were wheelchair or ventilator-dependent and needed these resources at an earlier stage of their disease than their ambulant or non-ventilated sibling.
We looked for possible factors explaining the observed phenotypical variation between siblings. Siblings who had the disease for longer were often more severely affected. However, this does not explain why these siblings developed symptoms at a younger age and why their disease progressed more rapidly. Gender could play a role since twice as many males as females were more severely affected. Earlier studies on the natural course of Pompe disease also show male gender to be a predictive factor for a more severe respiratory status
[12, 29]. GAA activity did not explain differences in phenotype, since in only three families did the most severely affected patient have the lowest amount of enzyme activity in fibroblasts. This is in accordance with previously published data on Pompe patients, describing the absence of a correlation between the clinical course and residual activity in patients with the c-32-13 T > G/null genotype
. Other co-morbidities contributed to phenotypical differences in only one family.
The substantial differences in age of symptom onset observed in the 12 families with an identical GAA genotype (c.-32-13 T > G/c.525delT (r.0)) suggest that other factors such as variability in genetic background, modifying genes or environmental factors are likely to play a role
[5, 8, 14, 16, 20, 30]. An example of a potential modifying gene is the angiotensin-converting enzyme gene, which has been described as playing a role in modulating phenotype and prognosis in Pompe disease
[31, 32]. This and other modifying genes might also explain the clear differences between families with the same GAA genotype, and the clustering of symptoms such as ptosis and bulbar weakness in certain families.