Background Acute Intermittent Porphyria (AIP) is characterized by neurovisceral attacks including abdominal pain and neuropathy, resulting from heterozygous pathogenic variants in hydroxymethylbilane synthase (HMBS). Three studies found the prevalence of pathogenic HMBS variants in genetic datasets to be 1:1299, 1:1675, or 1:1782, but did not include clinical information for the individuals in those large general population genetic datasets. Based on these studies and the prevalence of diagnosed AIP, the penetrance of AIP was described to be low (<1%); however, the prevalence of diagnosed or undiagnosed patients in those population datasets was unknown. A higher prevalence of symptomatic individuals might be expected in a patient population sample. The Mass General Brigham (MGB) Biobank is a database containing clinical data and biologic samples of over 135,000 patients, including annotated exome data for 13,000, collected through the MGB hospital system. The aim of this study was to evaluate the genetic prevalence and disease penetrance of AIP using the MGB Biobank exome dataset in parallel with clinical information obtained from patient medical records.

Methods Patients evaluated at MGB were eligible to enroll in the MGB Biobank research protocol. We identified individuals with disease-associated HMBS pathogenic variants in the dataset, and the medical charts of these individuals were reviewed.

Results Among the 13,000 annotated exomes, 8 patients were identified with pathogenic HMBS variants, corresponding with a genetic prevalence of 1:1625. These 8 patients had 3 well-described pathogenic variants in HMBS (NM_000190.3): p.Arg167Gln, p.Arg173Trp, and p.Arg225X. Two of the 8 patients had been previously diagnosed with AIP. However, detailed review of the 6 remaining undiagnosed patients revealed that 4 had a history of at least one episode of unexplained abdominal pain, 3 had unexplained neuropathy, 3 had hyponatremia, and 3 had hypertension. Further, at least 4 patients were exposed to common AIP triggers such as porphyrinogenic medications, anesthesia, or tobacco.

Conclusion Using the MGB Biobank, we found the prevalence of pathogenic AIP variants to be 1:1625, which is consistent with previous studies in general population databases. AIP disease penetrance in our patient database was at least 25% but may be higher due underdiagnosis and failure to recognize the nonspecific disease symptoms. After recent approval to recontact patients, we will pursue biochemical testing of the 6 patients with pathogenic HMBS variants but no AIP diagnosis. Similar studies in other datasets are essential to better characterize the prevalence, penetrance, and underdiagnosis of the acute porphyrias and achieve earlier diagnosis. At ICPP, we plan to present additional data for the 50,000 exomes now available in the MGB Biobank and results of outreach to the undiagnosed patients for clinical evaluation and biochemical testing for AIP.