Skip to main content
Search
Search
Advanced Search
Skip main navigation
Open access
Research Article
8 May 2026

Evaluation of steroid toxicity using the Glucocorticoid Toxicity Index-Metabolic Domains instrument in adults with myasthenia gravis: a real-world study in the USA

Authors: Glenn Phillips https://orcid.org/0009-0001-0146-6916 [email protected], Cynthia Qi, Deborah Gelinas, Martha Stone, Anthony Chamberas, Dakshinamoorthy Amirthaganesan, Rucha Kulkarni, Albert Whangbo, and John H StoneAuthor Info & Affiliations
Volume 15, Number 6
https://doi.org/10.57264/cer-2025-0141
PDFFormats

Abstract

Aim: The Glucocorticoid Toxicity Index-Metabolic Domains (GTI-MD) has been utilized to assess steroid toxicity in clinical trials and practice but has not been applied previously to large datasets. The GTI-MD was used to investigate steroid toxicity in patients with myasthenia gravis (MG) using a large real-world dataset. Materials & methods: Steroid toxicity was assessed in adults with MG (≥2 diagnoses occurring between ≥30 and ≤730 days apart) using Optum® electronic health records. Patients were categorized into steroid-initiator (MG-SI) or steroid-naive (MG-SN) cohorts. The GTI-MD algorithm assessed changes in steroid toxicity during follow-up using vital signs and simple laboratory tests pertaining to toxicity in the body mass index, blood pressure, glucose tolerance and lipid metabolism domains. Changes in steroid toxicity were reported as cumulative worsening scores (CWS) and aggregate improvement scores (AIS). Results: A total of 682 (MG-SI:377; MG-SN: 305) patients with MG were included. The potential for steroid toxicity at index date, as assessed by the GT-SNAPSHOT scores, was comparable between the cohorts at baseline (92.0 ± 31.1 vs 88.8 ± 32.8; p = 0.117). At follow-up, the MG-SI cohort had higher CWS scores than the MG-SN cohort (22.6 ± 22.8 vs 18.7 ± 21.2; p = 0.023) and were more likely to exceed the 10-point minimum clinically important difference (68% vs 59%; p = 0.016) and the 20-point threshold (44% vs 36%; p = 0.030). Patients with multiple steroid exposures had higher CWS (25.7 ± 24.8 vs 19.2 ± 19.9; p = 0.006) and AIS (9.6 ± 34.7 vs -0.3 ± 33.7; p = 0.005) values than one-time steroid users. Patients who received ≥20 mg/day of steroids (prednisone-equivalent) at index had higher CWS values (25.0 ± 23.0 vs 19.9 ± 22.3; p = 0.033) and AIS values (8.3 ± 32.9 vs 1.1 ± 35.9; p = 0.044) compared with those who received <20 mg/day of steroids. Conclusion: The GTI-MD is a valuable and sensitive method for unmasking steroid toxicity in MG using large datasets to gather real-world evidence.

Plain language summary: Using routine medical data to understand the side effects of steroids in people with myasthenia gravis

What is this article about?

This study looked at how steroids, commonly used medications, can cause side effects in people with myasthenia gravis, a rare muscle-weakening condition. The researchers used a simplified tool called the Glucocorticoid Toxicity Index-Metabolic Domains (GTI-MD), which focuses on common health measures like body weight, blood pressure, blood sugar and cholesterol, to track these side effects using real-world health records.

What were the results?

The study included over 600 people with myasthenia gravis, comparing those who began steroid treatment with those who did not. People who used steroids showed greater increases in negative health effects over time, especially in body weight. Those who took higher doses or had multiple courses of steroids experienced more side effects. The simplified GTI-MD tool was sensitive enough to detect these changes using routine health data.

What do the results mean? Why is this important?

Steroid side effects can be serious and may go unnoticed without proper monitoring. This study shows that it is possible to track steroid toxicity using everyday medical data, helping doctors spot problems early and adjust treatments to reduce harm. The findings also suggest that people starting strong or repeated steroid treatments may need closer follow-up to manage risks. This approach could be used in other diseases where steroids are commonly prescribed.

Supplementary Material

File (supplementary data.docx)

References

Papers of special note have been highlighted as: • of interest; •• of considerable interest
1.
DeHart-McCoyle M, Patel S, Du X. New and emerging treatments for myasthenia gravis. BMJ Med. 2(1), e000241 (2023).
PubMed
Google Scholar
2.
Ye Y, Murdock DJ, Chen C, Liedtke W, Knox CA. Epidemiology of myasthenia gravis in the United States. Front. Neurol. 15, 1339167 (2024).
PubMed
Google Scholar
3.
Salari N, Fatahi B, Bartina Y et al. Global prevalence of myasthenia gravis and the effectiveness of common drugs in its treatment: a systematic review and meta-analysis. J. Transl. Med. 19(1), 516 (2021).
PubMed
Google Scholar
4.
Antonini G, Habetswallner F, Inghilleri M et al. Real world study on prevalence, treatment and economic burden of myasthenia gravis in Italy. Heliyon 9(6), e16367 (2023).
PubMed
Google Scholar
5.
Santos E, Coutinho E, Moreira I et al. Epidemiology of myasthenia gravis in northern Portugal: frequency estimates and clinical epidemiological distribution of cases. Muscle Nerve 54(3), 413–421 (2016).
PubMed
Google Scholar
6.
Breiner A, Widdifield J, Katzberg HD, Barnett C, Bril V, Tu K. Epidemiology of myasthenia gravis in Ontario, Canada. Neuromusc. Disord. 26(1), 41–46 (2016).
PubMed
Google Scholar
7.
Rodrigues E, Umeh E, Aishwarya, Navaratnarajah N, Cole A, Moy K. Incidence and prevalence of myasthenia gravis in the united states: a claims-based analysis. Muscle Nerve 69(2), 166–171 (2024).
PubMed
Google Scholar
8.
Alhaidar MK, Abumurad S, Soliven B, Rezania K. Current treatment of myasthenia gravis. J. Clin. Med. 11(6), 1597 (2022).
PubMed
Google Scholar
9.
Patel NJ, Jayne DR, Merkel PA et al. The glucocorticoid toxicity index-metabolic domains, an abridged version of the glucocorticoid toxicity index: post-hoc analysis of data from the advocate trial. Lancet Rheumatol. 5(7), e413–e421 (2023).
PubMed
Google Scholar
• Validates the Glucocorticoid Toxicity Index–Metabolic Domains (GTI-MD) as a robust surrogate for the full GTI, supporting its use in clinical trials and real-world studies to quantify steroid toxicity.
10.
Habib AA, Narayanaswami P. Would glucocorticoids be approved for clinical use if discovered today? Muscle Nerve 70(1), 9–11 (2024).
PubMed
Google Scholar
•• Offers a critical reflection on the modern risk-benefit profile of steroids, aligning with the study’s motivation for evaluating toxicity.
11.
Miloslavsky EM, Naden RP, Bijlsma JW et al. Development of a Glucocorticoid Toxicity Index (GTI) using multicriteria decision analysis. Ann. Rheumatic Dis. 76(3), 543–546 (2017).
PubMed
Google Scholar
• Describes the development of the GTI using a rigorous decision-analysis framework, forming the methodological foundation of this study.
12.
Liang Y, Zeng FA, Sheriff T et al. Evaluation of the toxicity of glucocorticoids in patients with autoimmune blistering disease using the glucocorticoid toxicity index: a cohort study. JAAD Int. 6, 68–76 (2022).
PubMed
Google Scholar
13.
Stone JH, McDowell PJ, Jayne DRW et al. The Glucocorticoid Toxicity Index: measuring change in glucocorticoid toxicity over time. Semin. Arthritis Rheum. 55, 152010 (2022).
PubMed
Google Scholar
•• Provides the definitive framework for the development and application of the GTI, including cumulative worsening scores and aggregate improvement scores metrics used in the present study.
14.
Bahap-Kara M, Sariyildiz E, Zengin HY, Karadag O, Bayraktar-Ekincioglu A. Prospective assessment of glucocorticoid toxicity in rheumatology practice: a focus on the glucocorticoid toxicity index. Rheumatology (Oxford) 64(3), 1010–1018 (2024).
Google Scholar
• Demonstrates practical application of the GTI in routine rheumatology care, highlighting real-world feasibility.
15.
McDowell PJ, Stone JH, Heaney LG. The role of quantification of glucocorticoid-associated toxicity in severe asthma. J. Cell Immunol. 3(1), 31–35 (2021).
PubMed
Google Scholar
16.
Choi YC, Zhang D, Tyczynski JE. Comparison between health insurance claims and electronic health records (EHRs) for metastatic non-small-cell lung cancer (NSCLC) patient characteristics and treatment patterns: a retrospective cohort study. Drugs Real World Outcomes 8(4), 577–587 (2021).
PubMed
Google Scholar
17.
Optum EHR data. Gain visibility across therapeutic areas and the continuum of care. https://www.optum.com/en/business/life-sciences/real-world-data/ehr-data.html
Google Scholar
18.
Jayne DRW, Merkel PA, Schall TJ, Bekker P. Avacopan for the treatment of anca-associated vasculitis. N. Engl. J. Med. 384(7), 599–609 (2021).
PubMed
Google Scholar
19.
Rice JB, White AG, Johnson M et al. Healthcare resource use and cost associated with varying dosages of extended corticosteroid exposure in a us population. J. Med. Econ. 21(9), 846–852 (2018).
PubMed
Google Scholar
• Highlights the substantial healthcare burden and costs linked to prolonged corticosteroid exposure, reinforcing the need for toxicity monitoring tools.
20.
Kabadi S, Yeaw J, Bacani A et al. Healthcare resource utilization and costs associated with long-term corticosteroid exposure in patients with systemic lupus erythematosus. Lupus 27(11), 1799–1809 (2018).
PubMed
Google Scholar
21.
Park H-S, Kim K, Yu MH et al. Risk of fracture in patients with myasthenia gravis: a nationwide cohort study in Korea. J. Bone Mineral Res. 39(6), 688–696 (2024).
PubMed
Google Scholar
|