An innovative and effective NGS-based protocol for GBA1 gene analysis in Parkinson disease.

Variants in the GBA1 gene represent the most common genetic risk factor for Parkinson’s disease (PD). Overall, 8–12% of PD patients carry a heterozygous GBA1 variant, although prevalence varies considerably across populations. Large-scale sequencing of GBA1 remains challenging with conventional diagnostic methods—namely Sanger sequencing and standard short-read next-generation sequencing (sr-NGS)—due to the high sequence homology between GBA1 and its pseudogene GBA1LP, which often results in false-positive and false-negative findings. This complexity underscores the need for a reliable, accurate, and rapid alternative for GBA1 analysis, a requirement that is particularly relevant for the implementation of genotype-based personalized therapies, such as deep brain stimulation (DBS) in GBA1-PD patients. In this work, we designed, optimized, and validated LONG-NEXT, an innovative NGS-based protocol aimed at streamlining large-scale GBA1 sequencing. The approach employs a single long-range PCR amplicon (6.5 kb) encompassing the entire GBA1 gene as a template for short-read sequencing library preparation, combined with a dedicated bioinformatic pipeline that masks the GBA1LP sequence in the reference genome. The protocol was first tested on selected cases with suspected genotyping errors from conventional methods (n = 13) and subsequently validated on two cohorts previously screened either by Sanger sequencing (n = 101) or standard sr-NGS (n = 294). In the optimization phase, reanalysis with LONG-NEXT revealed three sr-NGS false positives caused by pseudogene read mismapping, four Sanger false homozygotes due to PCR-related allele dropout, and six false negatives shared by both techniques, attributable to pseudogene misalignment or allele dropout. Validation identified one additional Sanger false homozygote and one sr-NGS false negative. Within the framework of a study involving a large multicentric Italian PD cohort (PARKNET Study Group), investigating the prognostic impact of GBA1 variants on long-term motor and non-motor outcomes in PD patients who had undergone DBS surgery, I employed LONG-NEXT as genetic screening method for PD patients recruited at the IRCCS Mondino Foundation (n=137). We identified 41 GBA1 variant carriers and 96 wild-type PD patients. Among the positive patients, 17 are carriers of severe variants (41%), 2 of complex allele (5%), 8 of mild variants (20%), 8 of risk variants (20%) and 6 of unknown variants (15%). In addition, glucocerebrosidase (GCase) enzymatic activity was assessed in a subgroup of 30 PD patients (15 GBA1-PD, 15 nonGBA1-PD), revealing significantly reduced activity in GBA1-PD, consistent with previous literature. The LONG-NEXT protocol proved to be a robust, rapid, cost-effective, and accessible strategy for precise GBA1 genotyping, enabling targeted therapeutic approaches. Our findings confirm that GBA1-PD patients derive motor benefit from DBS, while their accelerated cognitive decline appears to reflect the natural history of the disease rather than an effect of the surgical intervention. Future work will expand GCase activity profiling and investigate additional blood-based biomarkers—such as substrate accumulation (e.g., glucosylceramide and glucosylsphingosine) and other lipid metabolism–related markers (e.g., ceramide, sphingomyelin, sphingosine)—using advanced mass spectrometry, with the aim of defining a distinct biochemical signature for GBA1-PD.