Normative Data of the VARIETE Cohort Influences Clinical Interpretation of Total IGF-I Levels Measured By the IDS-iSYS Assay
Presentation Number: OR21-1
Date of Presentation: April 2nd, 2017
Aimee J Varewijck1, Aart J. van der Lely2, Sebastian J.C.M.M. Neggers1, Leo J. Hofland3 and Joseph A M J L Janssen*3
1Erasmus MC, Rotterdam, Netherlands, 2Erasmus University Medical Center, Rotterdam, Netherlands, 3Erasmus Medical Center, Rotterdam, Netherlands
Introduction and Background: We previously reported that introduction of the IDS-iSYS IGF-I assay may have considerable consequences for the clinical diagnosis and interpretation of IGF-I levels during therapy of GH deficient (GHD) subjects1. Normative data for the iSYS IGF-I assay have been published by Bidlingmaier et al2. Recently Chanson et al. published normative data for the iSYS IGF-I assay based on a large random sample of the French general adult population (VARIETE cohort)3.
Aim of the study:To investigate whether normative data from two different studies (Bidlingmaier et al. vs. Chanson et al. in the VARIETE cohort) lead to differences in clinical interpretation (Z-scores) of total IGF-I levels measured by the IDS-iSYS assay.
Methods: Total IGF-I values were measured in 102 GH deficient subjects by the IDS-iSYS assay: Measurements were performed before starting GH treatment and after 12 months of GH treatment. Individual Z-scores were calculated twice: using normative data published by Bidlingmaier et al. and published by Chanson et al..
Results:After 12 months of GH therapy, total IGF-I increased from baseline 9.6 ± 4.2 nmol/L (mean ± SD) to 16.5 ± 6.8 nmol/L (P<0.001). When using normative data of Bidlingmaier et al., Z-scores increased from baseline -1.41 (-3.14 to +1.76) (mean (range)) to 0.21 (-3.00 to +3.28). When using normative data of Chanson et al., Z-scores increased from-2.40 (-4.52 to +1.31) to -0.65 (-4.32 to +2.79).At baseline Z-scores calculated from normative data of Bidlingmaier et al. differed significantly from those calculated from normative data of Chanson et al. (p<0.001). Z-scores calculated from both studies also differed significantly after 12 months of GH treatment (p<0.001).At baseline, 71.6% of all subjects had IGF-I Z-scores ≥-2, when using normative data of Bidlingmaier et al., while this was 31.4% when normative data of Chanson et al were used to calculate Z-scores.After 12 months of GH therapy, 94.7% of all subjects had IGF-I Z-scores ≥-2SD when using normative data of Bidlingmaier et al. while this was 84.0% when normative data of Chanson et al. were used to calculate Z-scores.
Conclusions: In adults with GHD, Z-scores of total IGF-I IDS-iSYS results based on normative data from two different sources differed significantly and would most likely have led to different clinical conclusions and GH dose adjustments. Therefore, in the next Consensus Statement on the Standardization and Evaluation of IGF-I Assays, criteria should be formulated on how to validate that normative data of an IGF-I assay are representative for the healthy population.
Nothing to Disclose: AJV, AJV, SJCMMN, LJH, JAMJLJ