Back Back
PBC: Assessing Response
New Criteria, New Hope: Rethinking When to Start Second-line Therapy for PBC

Released: December 17, 2024

Expiration: December 16, 2025

Sonal Kumar
Sonal Kumar, MD, MPH
Activity Information

Activity

Progress
1
Course Completed
Key Takeaways
  • Studies at AASLD 2024 offer new insights that could help us make more-informed decisions about initiating second-line therapies for PBC.
  • Although not all patients may achieve the specified ALP decrease, this does not rule out potential benefits, and patients could still respond well to second-line therapy.

One of the major challenges in treating primary biliary cholangitis (PBC) has been the lack of consensus regarding the definition of an appropriate response. Traditionally, treatment starts with ursodeoxycholic acid, and the patient’s response is assessed after 1 year. Various criteria have been published for assessment, but there is no unified agreement on which should be used.

Several criteria, such as the Barcelona criteria (>40% decrease or normalization of alkaline phosphatase [ALP] levels at 1 year) and Toronto criteria (ALP level of <1.67 times the upper limit of normal in 2 years), are commonly referenced. In addition, the POISE criteria, derived from the POISE trial of obeticholic acid, offer this approach to assessing treatment response: ALP of less than 1.67 times the upper limit of normal, at least 15% ALP reduction from baseline, and a normal bilirubin level.

Recently, there has been a shift toward aiming for ALP levels to fall within the normal range. The Global PBC Study Group has published updated criteria, recommending that ALP levels be below the upper limit of normal and that total bilirubin level be less than 0.6 times the upper limit of normal.

At the American Association for the Study of Liver Diseases 2024 annual meeting (AASLD 2024), a study of this new approach was studied by Nancy Reau and her colleagues at Rush University Medical Center was reported. In this study, they examined their institution’s experience with initiating second-line therapy. They found that only 34.5% of patients met the new Global PBC Study Group criteria for an adequate response to ursodeoxycholic acid, compared with 76.5% using the Toronto criteria. This is important because it suggests that, according to the new criteria, a much larger group of patients may be eligible for second-line therapy than previously recognized using older criteria.

To me, this highlights the need to reassess our patient population. Applying the Global PBC Study Group criteria can better identify individuals who may benefit from second-line therapy, ensuring that more patients receive appropriate treatment based on the latest data.

All or None, or the Lower the Better?
As mentioned, in the POISE trial the composite endpoint included 3 components:

  1. ALP less than 1.67 times the upper limit of normal
  2. 15% or greater ALP reduction from baseline
  3. Bilirubin normalization

This composite endpoint was used as a dichotomous variable. If all 3 components were met, the patient was considered a responder, but if any component was not met, the patient was considered a nonresponder. These POISE criteria have been used as a composite endpoint for all recent PBC studies, including those with seladelpar, elafibranor, and obeticholic acid.

But when thinking about the first component of this endpoint—reaching a final ALP level of less than 1.67 times the upper limit of normal—the baseline ALP level matters. A patient with an ALP level of 1000 IU/L has further to go compared with someone starting at 300 IU/L, making it harder to meet that first criterion and therefore the composite endpoint.

At AASLD 2024, a study by Cynthia Levy and colleagues evaluating how baseline ALP level affects response to treatment with elafibranor was reported. The study explored whether patients with higher baseline ALP levels were less likely to meet the composite endpoint because of their higher starting values.

The results showed that if the baseline ALP level was less than or equal to 2 times the upper limit of normal, 86% to 87% of patients achieved a biochemical response, meeting all 3 components of the composite endpoint. But as baseline ALP increased, the response rate decreased. Patients with an ALP level of greater than 4 times the upper limit of normal were less likely to achieve the composite endpoint, because it becomes harder to reach the required level of less than 1.67 times the upper limit of normal.

However, this does not mean that there is no effect. The mean percentage reduction in ALP from baseline to Week 52 was consistent across all starting levels. Whether ALP was less than or equal to 2 times the upper limit of normal or greater than 4 times the upper limit of normal, the reduction in ALP was around 30% to 40%. In other words, looking at the percentage change in ALP alone, patients showed similar responses regardless of their starting ALP levels. This is not evaluating the composite endpoint, but it is still valuable. A 30% to 40% reduction is important, especially when starting with extremely high ALP levels, because we know that the higher the ALP level is, the higher the risk to the patient is.

In my opinion, lowering ALP is always beneficial because the lower the ALP level is, the better. Even if the treatment does not bring ALP levels within the normal range or even below 1.67 times the upper limit of normal, a 30% reduction still holds value. Patients who are technically nonresponders by POISE criteria may still benefit from the treatment. To me, it is important to look beyond the composite endpoint and consider the broader picture. The outcome should not be seen as a simple success or failure, because that does not fully reflect whether the patient benefits. Even patients who do not meet the composite endpoint can still be receiving a benefit.

Your Thoughts?
How will these 2 studies presented at AASLD 2024 affect your treatment decisions for PBC? How do you assess response in patients with PBC? Join the discussion by posting a comment below.