Key points are not available for this paper at this time.
Abstract Disclosure: C.L. Chik: Advisory Board Member; Self; Ipsen, Novo Nordisk, Novartis Pharmaceuticals. Grant Recipient; Self; Pfizer, Inc.. C. Gandhi: None. J.A. Rivera: None. M. Denis: None. S. Van Uum: Advisory Board Member; Self; Ipsen, Pfizer, Inc., Novo Nordisk, Spruce. S.Z. Ezzat: Advisory Board Member; Self; Bayer, Inc., Eli Lilly Self; Roche Diagnostics. Objective: IGF-1 measurements were transitioned from the IDS-iSYS to the Roche Elecsys® immunoassays by the central laboratory during our clinical study that assessed strict IGF-1 control (I-Con) and quality-of-life scores in patients with acromegaly. Given the known variabilities of the IGF-1 measurements and the critical nature of IGF-1 results to the validity of the study, a retrospective analysis of the stored samples was completed using both methods. Methods: IGF-1 was measured in available stored samples (n=38) using IDS-iSYS at one of the hospital laboratories and the Roche Elecsys® at the central laboratory. In addition, the end-of-study (EOS) IGF-1 results were compared with IGF-1 measurements during the study (DTS) at the central and the local laboratory of the two sites with more than 10 stored samples, with one site using IDS-iSYS and the second site using IGF-1-RIACT. Bland-Altman plot, regression analysis and intra-class correlation were used to assess the reliability of the IGF-1 measurements. Results: Ten acromegaly patients from four sites in Canada participated in the clinical study. After six months of addition of pegvisomant to a somatostatin analog (n=6) or dose escalation of pegvisomant (n=4), IGF-1 decreased from 258.5 ± 59.2 µg/L 121.8 ± 14.4 % upper limit of normal (%ULN) to 184.1 ± 41.1 µg/L (86.7 ± 20.0 %ULN) (p=0.001). Two to five stored samples were available from the 10 patients (n=38) and the mean difference of IGF-1 between the IDS-iSYS and Roche Elecsys® was 10.8 µg/L or 4.6% when expressed as %ULN, with a trend to a higher value at the central laboratory. Linear regression of the difference between IGF-1 measurements against their average showed a slope of 0.1 (p0.001) or 0.15 (p=0.01) when expressed as %ULN. Intra-class correlations between the central laboratory DTS and EOS measurements were compared with those by the local laboratory. In the site that used IDS-iSYS for IGF-1 measurements, the intra-class correlations with the Roche Elecsys® was 0.99 (n=11) for DTS measurements and 0.98 (n=13) for EOS measurements. In the second site that used IGF-1-RIACT, the intra-class correlations with the IDS-iSYS was 0.38 (n=13) for DTS results, and with the Roche Elecsys® was 0.64 (n=15) for EOS results. Conclusion: Although there were variabilities between the IDS-iSYS and the Roche Elecsys® IGF-1 measurements, the small difference between the two IGF-1 assays confirmed that the transition from the IDS-iSYS to the Roche Elecsys® by the central laboratory had little impact on the IGF-1 results of our study. Using intra-class correlations to assess reliability of the IGF-1 measurements, the reliability of the IGF-1 results appeared dependent on the IGF-1 method used in the individual site. Presentation: 6/2/2024
Chik et al. (Tue,) studied this question.