Elafibranor in Primary Biliary Cholangitis: What Two-Year ELATIVE Data Add to PBC Care

For a chronic disease such as primary biliary cholangitis, a good result at 52 weeks is reassuring. The harder question is what happens after that.

Does the biochemical response hold? Does treatment remain tolerable with longer exposure? Do symptoms that matter to patients—particularly itch and fatigue—move in the same direction? And, perhaps most importantly, are we seeing something that is likely to matter to the natural history of PBC, or simply a durable improvement in surrogate markers?

Those are the questions behind the latest analysis from the ELATIVE phase III trial, published online in the Journal of Hepatology on August 26, 2026.

The report extends the randomized placebo-controlled experience with elafibranor to 104 weeks and adds longer-term observations from the trial's open-label extension. For hepatologists already familiar with the positive 52-week ELATIVE results, the interest here is less about discovering a new efficacy signal and more about asking whether the original signal survives longer follow-up.

Broadly, it does.

But the distinction between a durable biochemical response and proven prevention of liver-related clinical events remains important.

Why the second year matters

PBC treatment is rarely judged over a few months.

Patients may remain on therapy for years, and the clinical objective extends well beyond lowering an isolated laboratory value. Persistent cholestasis is associated with disease progression, and patients with an inadequate response or intolerance to ursodeoxycholic acid remain a group in whom additional therapeutic options are needed.

Elafibranor is a peroxisome proliferator-activated receptor alpha/delta agonist being evaluated in this setting.

The earlier 52-week ELATIVE analysis had already established a substantial biochemical treatment effect: 51% of patients receiving elafibranor achieved biochemical response compared with 4% receiving placebo.

The new analysis asks a more useful long-term question: does that response persist?

That matters because an impressive week-52 result followed by loss of effect, treatment discontinuation, emerging toxicity, or deterioration in other disease measures would tell a very different clinical story.

The two-year data therefore add an important dimension to the original trial, even though they do not yet answer the question clinicians ultimately care about most—whether treatment reduces decompensation, transplantation, or death.

ELATIVE: what was actually studied?

ELATIVE is an ongoing double-blind, placebo-controlled phase III trial with an open-label extension.

A total of 161 patients with PBC and inadequate response or intolerance to UDCA were randomized in a 2:1 ratio to receive either:

• Elafibranor 80 mg once daily, or

• placebo.

Patients entered with alkaline phosphatase levels of at least 1.67 times the upper limit of normal and total bilirubin no greater than 2 times the upper limit of normal.

The protocol excluded patients with autoimmune hepatitis or PBC–autoimmune hepatitis variant syndrome and those with evidence of hepatic decompensation.

That population definition is worth keeping in mind. These findings apply most directly to the type of compensated PBC population actually enrolled—not automatically to patients with advanced decompensated disease or overlap syndromes.

The blinded study included a common 52-week period followed by a variable second period in which participants could remain on their randomized treatment for as long as 104 weeks. Patients completing the double-blind phase could subsequently enter the open-label extension and receive elafibranor.

The architecture gives us two different types of evidence.

The randomized 104-week comparison is the more robust efficacy signal.

The extension helps answer questions about persistence and longer exposure, but—as with most open-label extensions—the farther we move from randomization, the more cautious interpretation should become.

At 104 weeks, the biochemical effect was still there

Among patients who remained in the placebo-controlled phase through week 104, 64.3% of elafibranor-treated patients—18 of 28—met the study definition of biochemical response.

No placebo-treated patient did.

ALP normalization occurred in 10.7% (3 of 28) of patients receiving elafibranor and in none receiving placebo.

The definition of biochemical response deserves attention because it was not simply “ALP improved.”

Patients had to achieve:

ALP <1.67× the upper limit of normal,
with at least a 15% reduction from baseline,
and total bilirubin at or below the upper limit of normal.

That composite makes the result more clinically interpretable than a modest numerical enzyme change alone.

The direction of the continuous biochemical measures was also consistent with the response analysis.

At week 104, mean ALP had fallen by 46.1% in the elafibranor group, while it had increased by 23.5% in the placebo group.

Mean total bilirubin decreased by 5.5% with elafibranor and increased by 32.6% with placebo.

For the practising hepatologist, that is probably the most important finding in the new report.

The biochemical effect observed during the first year did not appear to disappear during the second.

That strengthens confidence in durability.

It does not, however, convert a biochemical endpoint into a clinical-outcome endpoint.

What happens when treatment continues beyond the blinded phase?

The open-label extension included 138 patients.

Of these, 93 continued elafibranor, while 45 crossed over from placebo to elafibranor.

Mean elafibranor exposure was 115.2 weeks among patients receiving continuous treatment and 59.4 weeks in the crossover group.

Among continuously treated patients, biochemical response was seen in:

• 58.8% (47/80) at week 104, and

• 65.0% (13/20) at week 156.

ALP normalization occurred in:

• 15.0% (12/80) at week 104, and

• 25.0% (5/20) at week 156.

The crossover cohort provides another useful observation. After 52 weeks of elafibranor treatment, 51.2% (21/41) achieved biochemical response and 22.0% (9/41) achieved ALP normalization.

The pattern is reassuring.

Patients remaining on treatment continued to demonstrate biochemical responses, while those switching from placebo developed responses after starting elafibranor.

But week 156 deserves a small asterisk—not because the finding is uninteresting, but because the denominator has become small.

Only 20 continuously treated patients contributed to the reported week-156 biochemical-response analysis.

A response rate of 65% sounds impressive. Thirteen responders, however, remind us how easily percentages can look more definitive than the underlying sample warrants.

The extension therefore supports durability; it should not be read as a new randomized efficacy comparison extending to three years.

PBC is more than an alkaline phosphatase value

Biochemical endpoints dominate PBC trials for understandable reasons, but patients frequently experience the disease through symptoms.

Pruritus can be relentless.

Fatigue can be disabling.

And neither is always neatly predicted by the liver chemistry panel in front of us.

ELATIVE therefore assessed patient-reported symptom outcomes in participants who had moderate-to-severe symptoms at baseline.

At week 104, mean change in PROMIS Fatigue Short Form 7a was:

• −6.3 with elafibranor

• versus −0.4 with placebo.

For pruritus, mean change in the PBC Worst-Itch Numeric Rating Scale was:

• −4.1 with elafibranor

• versus 0.3 with placebo.

During open-label follow-up, numerical improvements among continuously treated symptomatic patients also persisted through week 130, with mean changes of −4.8 for fatigue and −4.0 for itch.

These are clinically interesting findings because both symptoms contribute substantially to the lived burden of PBC.

But they need appropriate framing.

These were analyses in patients with moderate-to-severe symptoms at baseline. Longer-term extension data are also affected by smaller numbers and missing observations over time.

The findings therefore support a symptom signal.

They do not establish that every patient starting elafibranor should expect meaningful improvement in fatigue or pruritus, nor do they make elafibranor primarily an antipruritic or antifatigue treatment.

Fatigue in particular remains biologically and clinically complicated. One laboratory improvement rarely explains the entire symptom.

What did the fibrosis markers show?

The study also followed several non-invasive fibrosis-related measures, including liver stiffness measurement, Enhanced Liver Fibrosis score, and PRO-C3.

At week 104:

• median liver stiffness change was 0.0 kPa with elafibranor and −0.8 kPa with placebo;

• median ELF score changed by 0.1 in both groups;

• median PRO-C3 changed by −2.6 μg/L with elafibranor and +1.7 μg/L with placebo.

The sensible word here is stability.

Not regression.

Not reversal.

And certainly not proof of an antifibrotic effect.

Stable non-invasive fibrosis-related markers during longer treatment are reassuring in a progressive liver disease, particularly when substantial biochemical improvement is occurring simultaneously.

But liver stiffness and circulating fibrosis markers are influenced by multiple biological and technical factors. The available report does not demonstrate histological fibrosis regression or establish that elafibranor prevents progression to cirrhosis.

Those are much larger claims than these measurements can support.

Longer exposure has not produced an obvious new safety story

According to the report, elafibranor was well tolerated through treatment exposure extending to 182 weeks, with the authors describing a favorable longer-term safety profile.

That is important.

A therapy intended for chronic use needs to remain tolerable well beyond the timeframe required to demonstrate regulatory efficacy.

At the same time, no phase III extension—particularly one with diminishing numbers at later time points—can fully characterize uncommon long-term adverse events.

Longer follow-up, post-approval pharmacovigilance, and real-world experience will still be needed to understand uncommon toxicity, treatment persistence, reasons for discontinuation, interactions, and performance in patients whose comorbidity burden differs from that of a clinical-trial population.

So far, though, the longer ELATIVE experience does not appear to undermine the safety signal established during the original blinded study.

What should this change in our interpretation of elafibranor?

The week-52 ELATIVE results established that elafibranor could produce a substantial biochemical response in patients with PBC who had inadequate response or intolerance to UDCA.

The new analysis adds something different: durability.

The treatment effect remains evident through two years of randomized follow-up in the patients contributing data, while the extension provides further evidence that biochemical response can persist with continued therapy.

The symptom findings add another potentially meaningful dimension, particularly for itch, although they deserve less certainty than the principal biochemical outcome.

And the stability of non-invasive fibrosis-related measures is reassuring without being proof of disease modification.

Taken together, the study strengthens the longer-term evidence base for elafibranor as a treatment option in the studied PBC population.

That is useful information for clinical discussions.

It is not the same as demonstrating that treatment improves transplant-free survival.

What ELATIVE still cannot tell us

There are several conclusions that should remain outside the boundaries of this analysis.

First, this report does not demonstrate a reduction in mortality, hepatic decompensation, liver transplantation, or cholangiocarcinoma.

Those outcomes were not established here.

Second, biochemical response should not be casually translated into guaranteed individual clinical benefit. Biochemical markers are highly relevant in PBC and form an important part of risk assessment, but this particular analysis does not directly show that achieving the reported response prevents a future liver-related event.

Third, ELATIVE was not a head-to-head trial of available second-line therapies.

The study therefore cannot tell us whether elafibranor is clinically superior or inferior to fibrates, obeticholic acid, seladelpar, or other evolving PBC treatment strategies.

Finally, extrapolation to populations not represented in the study requires caution. Patients with hepatic decompensation and PBC–autoimmune hepatitis variant syndrome were excluded.

The longer-term data are reassuring, but they remain data from a defined trial population.

The next important question is no longer whether ALP falls

ELATIVE now provides a progressively clearer answer to the early efficacy question.

Elafibranor lowers cholestatic biochemical markers, and for many patients that response appears to persist.

The research question now moves downstream.

Do patients maintaining these responses accumulate fewer liver-related events over time?

Do trajectories in ALP, bilirubin, fibrosis markers, and symptoms translate into measurable differences in progression?

How does treatment perform outside trial eligibility criteria?

And as the PBC treatment landscape becomes more crowded, which patient characteristics should guide selection between therapies?

Those questions will require longer observation, real-world cohorts, and eventually evidence connecting surrogate improvement with outcomes patients actually experience.

For now, the two-year ELATIVE analysis moves the elafibranor story forward without completing it.

Clinical Takeaway

The longer-term ELATIVE data strengthen the evidence that elafibranor 80 mg once daily can produce a durable biochemical response in patients with PBC and inadequate response or intolerance to UDCA.

At week 104 of the placebo-controlled period, 64.3% of elafibranor-treated patients achieved biochemical response versus none receiving placebo, with continued biochemical responses observed in the open-label extension. Patient-reported fatigue and pruritus also showed encouraging numerical improvements in symptomatic subgroups, while non-invasive fibrosis-related markers remained broadly stable and longer treatment exposure was reported as well tolerated.

For clinicians, the important addition is durability—not proof of disease modification.

ELATIVE supports greater confidence that the biochemical effect can persist beyond the first year. It does not yet demonstrate prevention of decompensation, transplantation, or death, and longer-term clinical-outcome and real-world evidence will be necessary before those conclusions can be made.

Source:

Elafibranor in primary biliary cholangitis: two-year placebo-controlled outcomes and long-term open-label data from the ELATIVE® phase III trial. Journal of Hepatology. Published online August 26, 2026. DOI: 10.1016/j.jhep.2026.08.016.