A new prospective study published in Seminars in Arthritis and Rheumatism confirms what clinicians have long known but have been unable to quantify: steroid-toxicity is universal in patients on long-term glucocorticoid treatment, and six months of exposure marks a critical inflection point in both objective toxicity burden and patient-reported quality of life.
The LONG-TOX cohort, led by Naomi Patel, MD MPH, and John H. Stone, MD MPH of Massachusetts General Hospital and Harvard Medical School, follows patients with rheumatic diseases treated with long-term steroids. Researchers measure steroid-toxicity, quality of life, and healthcare resource utilization in parallel. The baseline findings from the first 90 participants have just been published, and they are striking.
Measuring what matters: a sequenced approach
The LONG-TOX study deploys instruments from the GTI family in a deliberate sequence designed to capture both the burden patients carry at the point of entry and the trajectory of toxicity over time.
At enrollment, the GT-SNAPSHOT establishes a comprehensive baseline score of cumulative steroid-toxicity on a scale of 0 to 1592. This cross-sectional assessment captures the effects of steroid exposure on the patient before longitudinal tracking begins. From that point forward, the GTI takes over, measuring changes in steroid-toxicity between visits at months 0, 6, and 12. Running alongside both is the SF-36, which captures patient-reported quality of life, including energy, pain, emotional well-being, and general health perception.
Dr Patel explained the study goals:
“We’re really trying to delve into the relationships and the societal impact of glucocorticoid toxicity, by measuring the relationships between GTI scores at different time points, healthcare resource utilization, and quality of life.”
The GT-SNAPSHOT answers the question: “How much damage has already been done by steroids?” The GTI answers: “Is the steroid-toxicity getting better or worse over time?” The SF-36 answers: “How is steroid-toxicity affecting the patient’s daily life and experienced health?”
Six months changes everything
Every one of the 90 enrolled patients had a positive GT-SNAPSHOT score at baseline, meaning all had measurable steroid-toxicity at the baseline visit, some within days of starting treatment. Among those with more than six months of prior steroid exposure versus those who had not had prior steroid exposure, GT-SNAPSHOT scores were higher, indicating higher toxicity (205 vs. 160), and two SF-36 domains showed clinically and statistically significant decline: Energy/Fatigue (35 vs. 50, p=0.01) and General Health (30 vs. 60, p<0.001).
A General Health score of 30 against a population norm of 50 is not a statistical artifact. It represents a patient who reports their health is substantially worse than that of the general population. And it emerges within six months.
The most prevalent toxicities at the time of enrollment were neuropsychiatric: insomnia (66%), depression (44%), mania (34%), and cognitive impairment (28%). These are not late-stage complications. These can occur soon after initiation of steroids and are patient-felt. As Dr Patel has noted, they often go unrecorded:
“Some of the things that matter most to patients don’t perfectly line up with what’s perceived to be most important by the clinician. Patients often report weight gain, mood changes, acne, bruising, and other skin changes as being key concerns.”
What this means in the clinic
The LONG-TOX findings carry a direct message for rheumatologists and other prescribers: steroid-toxicity can accumulate in your patients faster than clinical intuition suggests, and it is measurable. The GT-SNAPSHOT can be deployed at the point of care to establish a patient’s toxicity burden at any stage of treatment, providing the kind of objective baseline that informs tapering decisions, guides conversations about steroid-sparing alternatives, and surfaces neuropsychiatric and quality-of-life impacts that routine appointments rarely capture in full.
Dr Patel is direct about the gap between what patients experience and what clinicians record:
“Patients can feel isolated and unheard if concerns about their treatment are overlooked or their clinicians don’t have alternatives they can turn to.”
Alongside the GT-SNAPSHOT, tools like Sam (steroidsandme.com) provide patients with a digital companion for education and real-world side effect tracking between visits, strengthening the doctor-patient connection where it matters most.
Why real-world data changes the equation
Clinical trials systematically underestimate the burden of steroid-toxicity. Trial exclusion criteria may remove the most vulnerable patients; controlled dosing protocols reduce cumulative exposure below real-world levels. LONG-TOX is designed to close that gap.
Dr Patel is direct about the scale of the problem:
“There are probably not too many clinic days that go by when we don’t see patients on steroids.”
Between 1–3% of the general adult population uses steroids long-term, and for many autoimmune conditions, steroid-sparing agents reduce but do not eliminate steroids completely. Her simulation modeling in ANCA-associated vasculitis has validated that in just six months, more than 10% of patients on even moderate steroid regimens can develop new-onset diabetes.
The GT-SNAPSHOT and SF-36 data reported here are the foundation. Together, they establish what good steroid-toxicity evidence looks like: a validated clinician-reported outcome measuring objective toxicity across nine health domains, paired with a patient-reported outcome capturing the lived experience of that burden.
Nina Liu, Director of COA Portfolio Management at IQVIA, is excited about the use of the GTI Family in conjunction with patient-reported outcome measures such as the SF-36 and SF-36v2:
“Pairing objective toxicity measures with the SF-36v2 gives a more complete picture of steroid burden, one that reflects both clinical impact and lived experience. As the FDA continues to prioritize patient-reported outcomes in drug evaluation, measures like the SF-36v2 help ensure that treatment benefits are defined in terms that matter to patients.”
As the cohort matures, the GTI will track whether steroid-toxicity improves, stabilizes, or worsens across disease types and treatment regimens, generating the longitudinal real-world monitoring that trial datasets cannot provide.
The implications for drug development
For pharmaceutical teams developing steroid-sparing therapies, LONG-TOX sets a new benchmark. The GT-SNAPSHOT establishes what a patient has already accumulated at a specified baseline time point, for example, prior to starting a new therapy. The GTI then measures the steroid journey from that baseline. Together, they give development teams a longitudinal steroid-toxicity narrative, one that dose-reduction data alone cannot provide. Health technology assessors and payers will increasingly require contextualized value of steroid-sparing agents.
The LONG-TOX study continues. Disease-specific and exposure-stratified analyses will follow as the cohort grows. The evidence being generated now will define the real-world standard against which the next generation of steroid-sparing therapies and point-of-care tools must demonstrate their value.
Reference
Patel NJ, Wang J, Jiang B, Jha I, McMahon GA, McMahon AE, Chiha T, Choi HK, Stone JH. A novel cohort to assess longitudinal glucocorticoid toxicity in individuals with rheumatic diseases: objectives, design, and initial baseline characteristics of the LONG-TOX cohort. Semin Arthritis Rheum. 2026 Feb;76:152897. doi: 10.1016/j.semarthrit.2025.152897.
Naomi Patel, MD MPH is a rheumatologist and clinical researcher at Massachusetts General Hospital and Instructor of Medicine at Harvard Medical School, specializing in vasculitis and the quantification of steroid-toxicity in rheumatic diseases.
John H. Stone, MD MPH is Professor of Medicine at Harvard Medical School, The Edward A. Fox Chair in Medicine at MGH, and co-founder of Steritas.
Nina Liu is Director of COA Portfolio Management at IQVIA, where she leads strategy and implementation for clinical outcome assessments across pharmaceutical development programs.