The Gaucher earlier diagnosis consensus point-scoring system (GED-C PSS): Evaluation of a prototype in Finnish Gaucher disease patients and feasibility of screening retrospective electronic health record data for the recognition of potential undiagnosed patients in Finland

Abstract

Background Gaucher disease (GD) is a rare inherited multiorgan disorder, yet a diagnosis can be significantly delayed due to a broad spectrum of symptoms and lack of disease awareness. Recently, the prototype of a GD point-scoring system (PSS) was established by the Gaucher Earlier Diagnosis Consensus (GED-C) initiative, and more recently, validated in Gaucher patients in UK. In our study, the original GED-C PSS was tested in Finnish GD patients. Furthermore, the feasibility of point scoring large electronic health record (EHR) data set by data mining to identify potential undiagnosed GD cases was evaluated. Methods This biobank study was conducted in collaboration with two Finnish biobanks. Five previously diagnosed Finnish GD patients and ~ 170,000 adult biobank subjects were included in the study. The original PSS was locally adjusted due to data availability issues and applied to the Finnish EHR data representing special health care recordings. Results All GD patients had high levels of the biomarker lyso-Gb1 and deleterious GBA mutations. One patient was a compound heterozygote with a novel variant, potentially pathogenic mutation. Finnish EHR data allowed the retrospective assessment of 27–30 of the 32 original GED-C signs/co-variables. Total point scores of GD patients were high but variable, 6–18.5 points per patient (based on the available data on 28–29 signs/co-variables per patient). All GD patients had been recorded with anaemia while only three patients had a record of splenomegaly. 0.72% of biobank subjects were assigned at least 6 points but none of these potential “GD suspects” had a point score as high as 18.5. Splenomegaly had been recorded for 0.25% of biobank subjects and was associated with variable point score distribution and co-occurring ICD-10 diagnoses. Discussion This study provides an indicative GED-C PSS score range for confirmed GD patients, also representing potential mild cases, and demonstrates the feasibility of scoring Finnish EHR data by data mining in order to screen for undiagnosed GD patients. Further prioritisation of the “GD suspects” with more developed algorithms and data-mining approaches is needed. Funding This study was funded by Shire (now part of Takeda).

Screening for patients with Gaucher's disease using routine pathology results: PATHFINDER (ferritin, alkaline phosphatase, platelets) study

“…One patient was diagnosed with GD clinically and confirmed on genetic testing and one extra heterozygote was identified. Using the more comprehensive Gaucher earlier diagnosis consensus point‐scoring system which includes clinical as well as laboratory data, one study identified 42 individuals scoring >7.5 point threshold with reduced lysoGB1 (<1.2 ng/mL) but failed to identify any new cases of GD from 170 000 individuals in a Finnish biobank 27 …”

“…Using the more comprehensive Gaucher earlier diagnosis consensus pointscoring system which includes clinical as well as laboratory data, one study identified 42 individuals scoring >7.5 point threshold with reduced lysoGB1 (<1.2 ng/mL) but failed to identify any new cases of GD from 170 000 individuals in a Finnish biobank. 27 The strength of the PATHFINDER studies is the use of primary care samples to ensure a lack of selection bias compared to cohorts identified from secondary care clinics. The study is limited by the assumed prevalence of asymptomatic GD being lower than suggested in the literature and by the lack of clinical phenotyping data to pair with the laboratory analyses so that a more specific clinical scoring system for GD could not be used.…”

Screening for potential undiagnosed Gaucher disease patients: Utilisation of the Gaucher earlier diagnosis consensus point-scoring system (GED-C PSS) in conjunction with electronic health record data, tissue specimens, and small nucleotide polymorphism (SNP) genotype data available in Finnish biobanks

“…We have recently introduced a novel approach for the screening of undiagnosed GD patients in Finland [ 15 ]. If successful, the approach that utilises Gaucher Earlier Diagnosis Consensus point-scoring system (GED-C PSS) [ 16 ], automated point scoring of longitudinal electronic health record (EHR) data, and Finnish biobank data and specimens allows the screening of large populations.…”

“…The applicability of the Finnish EHR data of the secondary/tertiary health care and the GED-C PSS in the automated point scoring was demonstrated in our previous study assessing ca 160,000 sample donors in Auria biobank (Turku, Finland) [ 15 ]. The GED-C PSS of 32 signs/covariables, originally developed for patient chart review-based point scoring of type 1 and type 3 GD, independently from the current study, represents a prototype PSS that hasn't been assigned for a commonly accepted cut-off value but which has so far been validated in 25 UK patients as well as tested in five Finnish GD patients [ [15] , [16] , [17] ]. In the five Finnish GD type 1 patients, the GED-C point score was 6–18.5 per 28–29 manually assessed signs/covariables per patient [ 15 ].…”

“…The GED-C PSS of 32 signs/covariables, originally developed for patient chart review-based point scoring of type 1 and type 3 GD, independently from the current study, represents a prototype PSS that hasn't been assigned for a commonly accepted cut-off value but which has so far been validated in 25 UK patients as well as tested in five Finnish GD patients [ [15] , [16] , [17] ]. In the five Finnish GD type 1 patients, the GED-C point score was 6–18.5 per 28–29 manually assessed signs/covariables per patient [ 15 ]. In the screened Auria biobank population, the score range was 0–13.5 in the automated assessment of 27 GED-C signs/covariables with 0.72% (n ≈ 1,160) of the subjects having a score of six or more points [ 15 ].…”