Background/Objectives: Autosomal recessive polycystic kidney disease (ARPKD) is a severe ciliopathy caused by biallelic pathogenic variants in PKHD1, characterized by variable renal and hepatobiliary involvement. The widespread use of next-generation sequencing (NGS) has revealed a large number of rare PKHD1 variants, creating major challenges in distinguishing molecularly confirmed ARPKD from a broader spectrum of PKHD1-associated disease. Methods: We performed an integrated clinical and molecular analysis of 68 individuals referred for suspected ARPKD. Using phase-aware and family-informed ACMG classification, patients were stratified into three genetically defined groups: 40 with molecularly confirmed ARPKD (biallelic pathogenic, likely pathogenic or segregation-supported VUS-LP variants in trans), 10 with biallelic PKHD1 variants of uncertain pathogenicity, and 18 monoallelic carriers. Genotype–phenotype correlations were restricted to the molecularly confirmed ARPKD group. Results: Among the 40 molecularly confirmed ARPKD patients, 17 (42.5%) carried two loss-of-function (LoF) alleles, 16 (40%) carried one LoF allele, and 7 (17.5%) carried only non-LoF alleles. A strong allele-dose effect was observed. Neonatal or infantile onset occurred in 88% of LoF/LoF patients, compared with 56% of LoF/non-LoF and 29% of non-LoF/non-LoF individuals (p < 0.001). Progression to renal replacement therapy occurred in 65%, 31%, and 0% of patients (p = 0.002). In contrast, hepatobiliary disease was highly prevalent across all genotype classes and showed no significant association with LoF burden. Conclusions: Phase-aware and family-informed interpretation of PKHD1 variants distinguishes a molecularly confirmed ARPKD core from a broader PKHD1 variant spectrum. Within confirmed ARPKD, loss-of-function allele burden is the primary determinant of renal and perinatal severity, whereas hepatic disease is largely independent of truncating allele burden. These findings refine diagnosis, prognosis, and genetic counseling in the genomic era.
Lapunzina-Soler et al. (Wed,) studied this question.