analysis-2030262 1920

GENFIT has published the first results of its Elafibranor efficacy and safety clinical trial for patients with Primary Biliary Cholangitis (PBC) and an inadequate response to Ursodesoxycholic acid.

Those results are amazing but before commenting its, it seems important to me to replace them in the context of this rare and difficult to cure disease.


For years the only drug available against PBC was Ursodesoxycholic Acid (UDCA), now sold as a generic, but a number of patients (40% according to Intercept) are intolerant or non-responders to this treatment.

There was therefore significant pressure from the patient associations to find an alternative treatment to the UDCA.

This disease is slowly evolving, so it is difficult to carry out clinical studies that involve a long enough time to see results and constraints of reasonable delays in the development of a highly anticipated drug.

However, a study on histological improvement takes years, in this configuration, regulatory agencies such as the FDA are trying to determine early indicators of the disease that can be identified more quickly and which, in combination, seem to be predictive of histological improvements.

These are therefore alternative clinical study objectives, called surrogate endpoints, that do not recognize the improvement of the disease as such but give indications that suggest that the benefit of the drug could be positive in the future.

If a drug can over a period of time reduce these indicators to the predetermined thresholds, it will be statistically conceivable that it will improve the prognosis of the disease and may be given a marketing authorization.

In the case of PBC, to determine precisely these thresholds and criteria, a very precise statistical analysis had to be carried out beforehand. This complete analysis can be found in this 2014 study published in GastroEnterology. (Levels of Alkaline Phosphatase and Bilirubin Are Surrogate End Points of Outcomes of Patients With Primary Biliary Cirrhosis: An International Follow-up Study ).

This study was based on data from the largest study conducted by the Global PBC Study Group in the Netherlands.

These data cover a total of about 5000 patients with PBC at different stages of the disease, from the least advanced to the most advanced.

The studies concluded that, in this disparate population, acceptable serological surrogates for predicting long-term disease improvement and therefore for marketing authorization were as follows:

it is a mixed endpoint grouping together the following 3 criteria:

  • Alkaline Phosphatase (ALP) <1.67 times the ULN (Upper Limit of Normal) which, unless I am mistaken, is located at about 100 U / L but seems to vary according to the laboratories and the sex of the patients.
  • 15% decrease in Alkaline Phosphatase (ALP) vs baseline
  • Bilirubin level brought back to normal and constant.


the FDA insisted that "ALP as a stand-alone biomarker has not been validated in randomized clinical trials to predict clinical outcomes. "

One could say that the criteria to be obtained in a clinical study required for the market agreement of a drug to treat PBC are clear, argued, supported by an international study on more than 5000 patients and are therefore established. 

But two years ago,  Intercept Pharmaceutical gave the FDA the results of its POISE study (OCA as PBC treatment), and all did not go as planned.

The endpoints of his study were those mentioned above and the OCA was very successful in achieving the results of the composite criterion.

image-17 med

The OCA reached the endpoint in 40 to 50% of the patients where the placebo was 10%.

It might have looked like a total success, but then, the FDA discover that, contrary to what had been agreed between Intercept and the FDA at the beginning of the study, Intercept had not recruited the expected patients.

Indeed, one of the important criteria was to bring total bilirubin back into the normal range, but the FDA found that 90% to 94% of the patients included in the study (regarding arms) were already in the normal range at baseline. Intercept had therefore included only slightly affected patients contrary to FDA requests. The population included in the study therefore had nothing to do with the one used to establish the criteria validating the study.

POISE ULN Bilirubine

The FDA therefore considered that the criteria achieved were not acceptable as they were, because they were inappropriate for the studied population. Since the bilirubin criterion was no longer acceptable in the Intercept study, only the decline in ALP remained valid, but the FDA had pointed out before, “ALP as a stand-alone biomarker has not been validated in randomized clinical trials to predict clinical outcomes ".

Probably under the pressure of patient organizations starving for a new medication, the FDA tried to see if they could extract from the base of the 5000 patients of the Global PBC Study Group, a subgroup with the same characteristics as the population recruited by Intercept, and then removing bilirubin criterion, looked for other threshold criteria based on the decrease in ALP that may be predictive of clinical improvements, and finally checked to see if these criteria were met in the population of the ICPT study.

After a period, the FDA concluded that a criterion requiring a 40% decrease in ALP below a threshold of 2 x UNL appeared to be predictive of improved PBC !

This is a new composite surrogate endpoint, more demanding on the decrease of the ALP in (40% instead of 15%), and curiously less demanding in terms of threshold to reach (ALP under 2 x UNL instead of 1.67 x UNL) but above all, they removed criterion on total bilirubin.

 One can say that these criteria have been established 'post hoc' in an 'ad hoc' way for the Intercept study, and have thus enabled Intercept to obtain a marketing agreement for the OCA in the PBC.


So, we end up with two composite endpoints approved by the FDA to obtain a marketing agreement for a drug in the PBC.

N° 1 - An academic composite endpoint, the initial endpoint, incorporating the following criteria:

  • Alkaline Phosphatase (ALP) <1.67 times ULN (Upper Limit of Normal)
  • 15% decrease in Alkaline Phosphatase (ALP) vs baseline
  • Bilirubin level brought back to normal and constant.


N° 2 - A second endpoint established following the Intercept POISE study, integrating the following criteria: 

  • Alkalin Phosphatase (ALP) <2.00 times ULN (Upper Limit of Normal)
  • 40% decrease in alkaline phosphatase (ALP) vs baseline


In all likelihood, the both endpoints should be now recognized by the FDA, even if it seems logical that the scientific community considers that the former has more legitimacy because it is the result of a long and serious international study.


Pruritus is one of the most serious symptoms of the disease, with fatigue, it is one of the most disabling effects of PBC. 

Payers are very demanding about the adherence to the treatments they support, so they are often suspicious of drugs with unpleasant side effects such as pruritus because they know that it deters patients from taking their treatment.

Beyond improving the life expectancy, considering the improvement of pruritus is essential for any treatment.

In the case of PBC, pruritus is a direct consequence of the disease, so a drug should be able to reduce this inconvenience or at least not aggravate it.

Unfortunately, this is what the OCA does, inducing additional pruritus. 

to date, it seems clear that any effective drug on the PBC that would further reduce pruritus would quickly replace the OCA in the market, for obvious reasons of patient comfort.

Elafibranor has consistently shown that it improves the pruritus of patients, so this is a strong argument for this drug, which, once on the market, could quickly take the lead.


The problem of recruiting patients with high bilirubin encountered by INTERCEPT was not necessarily due to negligence on their part, but may be due to the difficulty of finding such patients. 

It seems that CYMABAY encountered the same problem in their recruitment of Phase 2 for Seladelpar 

CYMABAY Ph2 PBC trial, patients baselines

This problem could perhaps also affect the GENFIT trial. In the absence of data on the total bilirubin level of patients recruited in the GENFIT study, it is difficult to know so far.


As We dont know to date if GENFIT recruited  majority of patients had a total bilirubin plasmatic levels over the normal level  we should rely only on the published data.

If  GENFIT reaches endpoint  Elafibranor will have met the same criteria as the OCA and will therefore be a logical candidate for obtaining a MA, especially as OCA's safety issues are challenging many people.

The ideal would therefore be to reach an ultra-demanding composite criterion comprising a synthesis of the most stringent criteria of the two endpoints in a population with high bilirubin at baseline.

  • Alkaline Phosphatase (ALP) <1.67 times ULN (Upper Limit of Normal)
  • 40% decrease in alkaline phosphatase (ALP) vs baseline
  • Bilirubin level brought back to normal and constant.

If all of these criteria are met, Elafibranor is sure to be considered as the ideal treatment, far ahead of the OCA, especially as Elafibranor achieved improvements in patient's pruritus when the OCA worsened it.


The published results show that 80 mg dose of Elafibranor conduct to an ALP reduction of 48 % vs an increase of 3% for the placebo as Primary Outcome Measure (p<0.001). 120 mg dose of Elafibranor conduct to an ALP reduction of 42 % vs an increase of 3% for the placebo as Primary Outcome Measure (p<0.001). 

Regarding the N°1 Consensual Endpoint,  included in the trial as  first in Secondary Outcome Measures :

  • Alkaline Phosphatase (ALP) <1.67 times ULN (Upper Limit of Normal)
  • 15% decrease in Alkaline Phosphatase (ALP) vs baseline
  • Bilirubin level brought back to normal and constant.

With the 80 mg dose of Elafibranor 67% of patients reached this endpoint  vs 6,7% for placebo (p=0.001).

With 120 mg dose of Elafibranor 79% of patients reached this endpoint  vs 6,7% for placebo (p<0.001).

all that with an improvement in pruritus 

The results are amazing !


ALP decrease at the end of the trial

ALP decrease at week 12

FDA Consensual Criteria at the end of the trial 

FDA Consensual Criteria at week 12

(Cymabay didn’t published their results on FDA consensual criteria at week 12 but only at week 26, Intercept published the data at week 12 so we can only compare Elafibranor and OCA at week 12 on that criteria.)


To conclude, the results of the Elafibranor in the PBC are amazing and far better than those obtained with difficulty by OCA. 

It should be remembered that since its launch on the market,  INTERCEPT has to put a safety warning (Black Box) on the OCA boxes following the death of many patients.

Notice tha GILEAD never published their GS 9674 results in PBC despite their trial has been completed for months!

The results obtained by Elafibranor in 12 weeks are equal or better than Cymabay’s results obtained in 26 weeks.

Elafibranor is also reducing pruritus and is presenting a better safety profile. 

Even if it is about phase 2 results out of 45 patients, the published results of Elafibranor trial are more than encouraging and should attract the attention of the associations of PBC patients who still do not have a serious and safe alternative solution to UDCA.

To my opinion, GENFIT is on the path for a marketing agreement in PBC.

Gery Divry

Notice that I am neither a physician nor a biologist or financial analyst, my point of view is only that of an enlightened amateur, so it must be taken for what it is, a questionable point of vie

WWW.NASHBIOTECHS.COM  -  Copyright G DIVRY 2015-2016  - Contact and TERMS OF USE