Response to Nogales-Gadea et al.

To the Editor: In their letter, Nogales-Gadea et al.¹ present a thoughtful analysis of several points from our recently published paper entitled “Determining the Prevalence of McArdle Disease From Gene Frequency by Analysis of Next-Generation Sequencing Data.”² The main thrust of our paper is that McArdle disease is more common than currently appreciated. Overall, Nogales-Gadea et al.¹ agreed with our conclusion, if not taking it even further, but they differed on the weighting of some of our explanations for why it may be underrecognized. These interesting and important questions are worthy of further discussion.

Nogales-Gadea et al.¹ agree that one explanation for our finding is that an unknown, but significant, proportion of patients are undiagnosed by virtue of having mild or no symptoms, which could be due to the intrinsic biology of the disorder or because the level of physical activity of many individuals with biallelic mutations in PYGM is not sufficient to trigger development of recognizable manifestations. We fully agree with this and believe that the penetrance and expressivity of McArdle disease are complex questions that deserve further study.

In our paper, we showed several alternative calculations based on carrier frequency that led us to estimate that prevalence of McArdle disease was 1/7,650 to 1/80,478, which is lower than previously estimated (1/100,000). We thought that the high end of this range (1/7,650) was unlikely and suggested as one possible explanation that some variants reported as pathogenic may actually be benign, consistent with the findings of Bell et al.³ Nogales-Gadea et al.¹ questioned this hypothesis. They have very recently reviewed PYGM mutations in McArdle disease, and they generously shared a draft manuscript with us to support our writing of this response.⁴ We repeated our calculations, including all variants reported as pathogenic in their manuscript, using the European-American subset of the National Heart, Lung, and Blood Institute’s Exome Sequencing Project data set. The result was a prevalence estimate for McArdle disease of about 1/6,000, more common than even the highest estimate in our paper. More accurately, we should say that this is the predicted prevalence of individuals who harbor biallelic pathogenic mutations in PYGM. Assuming this to be true, one must then explain why the prevalence of the McArdle phenotype appears to be so much lower than this. We and Nogales-Gadea et al.¹ invoke various explanations that relate to the penetrance and expressivity of this trait and believe that it can be considered a gene–environment interaction, for which sustained energetic exercise is the environmental exposure or trigger. The authors are surely correct when they suggest that sedentary individuals have a lower probability of manifesting symptoms that might lead to a diagnosis of McArdle disease. Another potential explanation is the existence of modifiers. These could be acting in cis, analogous to the c.350G>A p.Arg117His and poly-T tract (ref. 5) or in trans, analogous to variants in DCTN4 (ref. 6), both of which modify the phenotype associated with mutations in CFTR.

We also agree with Nogales-Gadea et al.¹ that next-generation sequencing highlights other potentially interesting issues. We did not address in our manuscript the observation that European–American data sets contain a number of apparently null PYGM variants that have not been observed in patients with disease. We have demonstrated that—at least in autosomal dominant disorders—about 50% of apparently null variants are associated with an abnormal phenotype.⁷ In addition, there are dozens of other rare variants in this gene, all of which are unlikely to be benign. Were many of these PYGM variants to be pathogenic, it would raise the estimate further, which seems incredible to us.

Clearly, we have a great deal to learn about the genetics, biology, and phenotypic consequences of mutations in PYGM. The letter by Nogales-Gadea et al.¹ amplifies and extends our work and underscores the utility of genomics in improving our understanding of the full spectrum of variation. We are grateful for their thoughtful critique and hope that we will have the opportunity to further benefit from their leadership and experience in elucidating this important and fascinating disorder.

Disclosure

M.D.C. declares no conflict of interest. L.G.B. has a potential conflict of interest to declare, as he is an unpaid consultant to the Illumina Corporation. He also receives royalties from Genentech and Amgen, although those are not likely relevant to the topic of this paper.