Primary biliary cholangitis is a chronic autoimmune condition in which the immune system progressively destroys the small bile ducts inside the liver, causing bile acids and other cholestatic substances to accumulate in the blood and tissues. It predominantly affects women and is typically diagnosed in middle age; its most tormenting symptom is cholestatic pruritus—an itch that researchers describe as serious, debilitating, and under-recognized. Between 55 and 89 percent of patients experience it at some point, and up to 53 percent develop clinically significant itching over the course of their disease. The burden extends far beyond the skin: pruritus in PBC is frequently accompanied by sleep disturbance, depression, and cognitive impairment, and it can erode health-related quality of life as thoroughly as the underlying liver injury itself. Yet treatment options remain thin. Current guideline-recommended therapies rest on insufficient clinical evidence of efficacy and often cause tolerability problems, leaving itch in PBC what the field itself acknowledges as an unmet medical need—one that has only recently begun to attract purpose-built drugs.

Numerical rating scales have long been the workhorse of itch research. In trials, patients rate their worst itching on a scale from 0, meaning no itching, to 10, meaning the worst itching imaginable, and investigators track how those numbers change under treatment. The WI-NRS has been formally validated for measuring itch in PBC, but until now nobody knew how large a score reduction had to be before it reflected a change patients would call meaningful. That gap matters more than it might seem: a therapy can produce a statistically significant shift in a rating scale that is nonetheless too small for patients to notice in their daily lives, or a trial can miss a genuine benefit because its responder definition was drawn arbitrarily. Meaningful change thresholds, sometimes called responder definitions, are therefore essential for interpreting patient-reported outcomes, and the US Food and Drug Administration specifically recommends anchor-based methods—linking scale changes to an external, patient-anchored variable—for establishing them.

The thresholds unveiled in the new paper trace their lineage to linerixibat, a minimally absorbed ileal bile acid transporter inhibitor recently approved in the United States and the United Kingdom for cholestatic pruritus in PBC. The drug works by blocking a transporter in the ileum that normally recycles bile acids back into the bloodstream, thereby shrinking the pool of itching-provoking bile acids that accumulate when bile flow is impaired. Linerixibat’s development program spanned a Phase 2a proof-of-concept study, the Phase 2b GLIMMER dose-response trial, the pivotal Phase 3 GLISTEN study, and a long-term safety study, all built on validated patient-reported measures. GLIMMER, which ran from January 2017 to April 2020, opened with a four-week placebo run-in before twelve weeks of double-blind treatment and a four-week single-blind placebo phase—an architecture built to expose how much of any apparent benefit might stem from expectation rather than pharmacology. GLISTEN, conducted from December 2021 to December 2024, enrolled 238 adults aged 18 to 80 with PBC and moderate-to-severe pruritus, defined as a WI-NRS score of at least 4. Participants took oral linerixibat at 40 milligrams twice daily or matching placebo for 24 weeks, and the drug produced significant, rapid, and sustained reductions in both itch and itch-related sleep interference. The new threshold analysis drew primarily on Part A of GLISTEN, its 24-week primary intervention period, with supportive data from GLIMMER.

Inside GLISTEN, symptom measurement was granular and relentless. Participants carried an electronic diary and recorded their worst itch severity twice daily on the 0-to-10 scale; the higher of the two readings became the day’s worst itch score. Averaging seven daily values produced the Weekly Itch Score, and the worst weekly score within a month defined the Monthly Itch Score—the metric underlying GLISTEN’s primary and key secondary endpoints. Sleep was captured in parallel: every morning, patients rated how much itching had interfered with their sleep, from 0, meaning no interference, to 10, meaning complete interference, generating weekly and monthly sleep scores through the same averaging logic. Alongside these diaries, patients completed two global anchors. The Patient Global Impression of Severity asked them to rate their itch on a five-point scale from absent to very severe. The Patient Global Impression of Change used a seven-point Likert scale from very much improved to very much worse—and critically, it included a follow-up question asking whether the change they had experienced was, in their own judgment, meaningful.

That follow-up question became the study’s secret weapon. Because patients explicitly confirmed whether their change in itch was meaningful, the analysts could restrict their calculations to people who directly endorsed the significance of their own improvement, sharply reducing the risk of misclassification. The team first verified that the anchors were statistically sound, requiring Spearman rank-order correlations of at least 0.30 between anchor responses and changes in the weekly itch and sleep scores—a recognized minimum standard for anchor validity. Since treatment assignment was irrelevant to the question being asked, data from the linerixibat and placebo arms were pooled, and the analyses relied only on blinded observations. The primary anchor-based analysis examined patients whose PGI-S severity rating had improved by one point; a supportive analysis used patients rating themselves moderately improved on the PGI-C. These estimates were then cross-checked against distribution-based benchmarks, including one-half and one-quarter of the standard deviation of change and the standard error of measurement. Because GLISTEN’s headline endpoints were built on monthly averages rather than weekly ones, the team also ran exploratory analyses on monthly itch and sleep scores to test whether the same thresholds would hold at the monthly granularity that sponsors and regulators most often report. Finally, cumulative distribution function curves—plots showing how completely the responder groups separate at every possible score change—provided a visual sweep of the evidence before the findings were triangulated into single thresholds.

The numbers converged with striking consistency. Among the 238 GLISTEN participants—a group that was 95 percent female, with a mean age of 55.8 years, a median PBC duration of 61.2 months, and a median itch duration of 71.6 months—62 percent were white, 30 percent Asian, and 23 percent Hispanic or Latino. Their baseline monthly itch score averaged 7.3 out of 10, close to the ceiling of the scale. Patients who reported a meaningful change in itch at Week 24 had improved by an average of 3.0 points (standard deviation 1.7) on the weekly WI-NRS if their PGI-S severity rating had dropped by one point, and by 2.9 points (standard deviation 2.0) if they rated themselves moderately improved on the PGI-C. The corresponding sleep-interference improvements were 2.6 points (standard deviation 2.1) and 2.4 points (standard deviation 2.1) on the weekly SI-NRS. Distribution-based estimates for both scales ranged from 0.42 to 0.85 standard deviations of change, bracketing the anchor-derived values. When every line of evidence was triangulated, the team settled on reductions of at least 3 points on the WI-NRS and at least 2.5 points on the SI-NRS as the thresholds patients themselves consider meaningful.

Those figures may look modest, but context transforms them. A patient starting near 7 on the itch scale who improves by 3 points has shed nearly half of a severely burdensome symptom—something patients reliably recognize as life-changing rather than statistical noise. The distinction between statistical significance and clinical meaningfulness has haunted patient-reported outcome research for decades, and the cumulative distribution function curves in the new analysis illustrate it vividly: patients reporting meaningful improvement separate cleanly from those who do not only once score changes cross the three-point line on itch and roughly the 2.5-point line on sleep. With validated thresholds in hand, trial sponsors can now define responders and non-responders in a standardized way, regulators can weigh whether a drug’s effect size crosses the bar of human perception, and clinicians can interpret whether an individual patient’s improvement on a rating scale reflects a benefit they can actually feel.

The study also fills a gap that stretched beyond itch itself. Meaningful change thresholds have been characterized for itch intensity in dermatological conditions and in chronic kidney disease, but estimates in PBC were sparse, and no threshold had ever been established for pruritus-related sleep interference measured on a 0-to-10 scale in this population. That omission mattered, because disturbed sleep is one of the most corrosive consequences of cholestatic itch, driving exhaustion, depressed mood, and cognitive fog. By anchoring the SI-NRS to the same global impression measures patients used to judge their itch, the researchers demonstrated that sleep interference can be evaluated with the same rigor as itch intensity. The authors argue the findings support the use of both scales in future PBC trials of new antipruritic treatments, and may inform wider hepatology clinical practice, where itch severity is too often assessed informally or overlooked entirely.

In an era when drug approvals increasingly hinge on demonstrating benefits that patients genuinely value, the analysis is a case study in patient-centered measurement. The magnitude of a treatment effect, the work suggests, should not be defined by statisticians alone but calibrated against the lived experience of the people the medicine is meant to help. For the PBC community—a population that has watched itch go under-recognized and undertreated for decades—the new thresholds offer something concrete: a shared, evidence-based definition of what better actually looks like. If a future therapy trims a patient’s worst itch score by three points, or quiets itch-driven nights by two and a half, the field can now say with confidence that the change is real, it is meaningful, and it is worth pursuing.