fruit-424182 1920

How to confuse readers comparing apples and oranges

Bloomberg is writing in its note :

Leading Candidates

Intercept is testing Ocaliva, already approved to treat a rare liver disease called primary biliary cirrhosis, against NASH. Enrollment of a final-stage trial of 2,000 people is expected to be completed next year, with interim results due in 2019. A potential concern is that in an earlier trial some patients experienced itching when taking the drug, in some cases so severe that it interfered with basic activities. CEO Mark Pruzanski said in an interview that he hopes Ocaliva will become a backbone treatment, used in combination with other drugs.

Genfit is developing elafibranor, a once-a-day-treatment that raises fat metabolism and fights inflammation, to “treat the underlying cause of fibrosis,” Dean Hum, Genfit’s chief scientific officer, said in an interview.

While elafibranor failed to meet a primary goal in a mid-stage trial last year, it showed promise in a subset of patients with moderate and severe forms of the disease. A final-stage trial in 2,000 moderate and severe patients is underway, and the company may have data by late 2018. 

It is funny to read this note because the way it is redacted could let the readers think that INTERCEPT is testing Ocaliva on a wider part of population than GENFIT, who, because of a supposed previous failure is obliged to test Elafibranor only in moderate and severe patients population.

In fact it is not true, 

INTERCEPT Phase 3 population recruitment criteria are, as they presented it

1  NAS score ≥4.

2  Histologic evidence of fibrosis stage 1, stage 2 or stage


GENFIT Phase 3 population recruitment criteria are, as published

1  NAS score ≥4.

2  Fibrosis stage of 1 or greater and below 4


As you can see the populations are similar, on the paper they are identical, in fact INTERCEPT explained previously that to be included, the F1 patients should be obese, T2 diabetics or to have elevated ALT.

So the Bloomberg note is a little bit ambiguous on that point.

The other point is the mention of  GENFIT failure on primary goal last year.

Again the text is facially right, but unbalanced and not completely true, and notice that previous results of Ocaliva are not mentioned.

To be pertinent, and i imagine that Bloomberg readers expect pertinent data's from a worldwide agency like Bloomberg, they should have wrote:

With Ocaliva, INTERCEPT reached previously  its phase 2b primary endpoint, 'a 2pts decrease of NAS score without worsening of Fibrosis’ , but this endpoint is not considered anymore as relevant by regulation agencies to get an agreement on NASH. Conjointly INTERCEPT missed the secondary endpoint of its study, 'NASH reversion without worsening of Fibrosis' and never met in a clinical study the new stringent regulatory end-point required now in phase 3.

Last year, GENFIT failed to reach  its phase 2b primary endpoint, 'NASH reversion without worsening of Fibrosis' but Elafibranor is, at this time, the only drug wich reached  the new stringent regulatory end-point required now in phase 3 on all its ITT population.


It isn't obvious, but it is more pertinent, because you cannot compare different studies with different endpoints on different populations and then let people thinking that INTERCEPT reached its primary endpoint and not GENFIT, with no any more explanation.

It is as comparing a "hole in one » missed in golf with a successful penalty kick in football. 

it is comparing Apples and Oranges.

The problem is that many analysts don’t understand what is special in the NASH clinical trials.

It is an emerging disease, not a well defined one, so the definition of the disease is smoothly modified every year and the regulatory endpoints are also modified to follow the scientific knowledge.

A first example, INTERCEPT in FLINT.

Despite they didn't followed the basic standards rules of clinical studies (26% of patients started medication impacting the disease DURING the study and were not removed from the results), It was announced that they reached their primary endpoint (2 points decrease of NAS score without fibrosis worsening). 

Following basic analysis, it is a success, but in the real world it is not so evident because the target they reached is not anymore the target to reach.

The FDA announced clearly that NASH reversion without fibrosis worsening was now the target to reach and ICPT who had this endpoint as secondary in the FLINT trial was unable to reach it despite they counted patients who did not have NASH at the beginning of the trial as cured patients in their results. (Read:  INTERCEPT'S STUDY CURIOSITIES)

A second example, GENFIT in GOLDEN

The primary endpoint of the GOLDEN study was NASH reversion without worsening of fibrosis, an endpoint harder to reach. The study included patients  with a NAS score of 3 to 8 inclusive while Intercept trial excluded the NAS score <4. 

Indeed the NAS 3 group of patients (15%) who got a reversion under placebo  prevented GENFIT to have a statistically significant result in the entire population. This effect could have been anticipated because the false positives biopsy results are mathematically high in less advanced disease population. ( to see why, please read, BIOPSY IN QUESTION ! FALSE NEGATIVES  - FALSE POSITIVES ! A POPULARIZATION !)

Following basic analysis, it is a failure, but In July 2015, a new and stricter definition of the NASH reversion was proposed by the NASH Key opinion leaders and endorsed by the FDA. (To see the difference between the two definitions, please read, THE NEW DEFINITION OF NASH STRIKE THE NASH PLAYERS)

It is for now the definition to be used in Phases 3 trials.

Logically as the publication of the results was about to be published Genfit was asked by the authors of the publication to let them use this more strict definition on the GOLDEN data in the publication analysis, witch make sense.

Stricto sensu it is a' post hoc’ analysis and some analysts mention it as not relevant but they are wrong because it is not an 'ad hoc’ analysis with criteria chosen post hoc to present positives results.

In this case GENFIT did not interfere in the new definition parameters, the new definition grid promoted by the FDA and unknown at the beginning of the study, was applied by the authors on GOLDEN raw data, including the NAS 3 patients who perturbed the previous study results.

Surprise!  The result of the study with this new consensual endpoint was clear!  The GOLDEN study reached this endpoint on ITT population without any statistical treatment ... 

ICPT was supposed to do the same on FLINT raw data, but they didn’t, to my opinion it would have been a fiasco but no one can know! 

It shows clearly why it is not easy to have a definite position on NASH trial results just reading abstracts.

In a few months, what seemed evident at the beginning?

FLINT = Success and GOLDEN = Failure,

Was totally inversed by the evolution of scientific knowledge of the disease.

   FLINT reached the wrong target and missed the good one 

   GOLDEN missed the wrong target but reached the good one

Many Analysts made their opinion at the results announcement and did not take the time to update their knowledge, so they are still convinced that FLINT was a success and GOLDEN à failure

They are not pertinent in their analysis because they are investing or imitate others to invest on outdated data.

To resume, the consensual endpoint to reach now in a phase 3 trial is:

Reversion of NASH without worsening of fibrosis (with the new definition of reversion).

And the only drug which had positive results under those stringent criterias in a clinical study, is Elafibranor, OCA never reached it.

This demonstration show you how it is premature to take brutal position on NASH trial results without taking the time to analyze the reality and not relying on the black or white picture provided by superficial analysts.

Tobira results on fibrosis is interesting, they used that endpoint at secondary but in fact it was the main target of cenicriviroc.

Targeting NASH without a strong background in preclinical, they took the FLINT endpoint as primary.

But people who analyzed the preclinical studies are not really surprised that they missed the NASH endpoint (read :  TOBIRA )

Post hoc analysis (as done by ICPT and GENFIT) will maybe show subgroups with more efficacy and could help them to attempt a Phase 3 trial, alone or in combo. 

Post hoc studies on an emerging disease not well defined are essential.. it is the only way to progress ..

With a classic point of view, all the NASH studies are a failure; even FLINT trial with their enormous bias should have been rejected. 

In real world all those studies were important to progress in the knowledge of the disease. 

All the studied drugs had impacts on the disease, maybe not exactly as it was expected initially, but it is not a reason to drop it in the trashcan?

Fortunately it seems it is also the opinion of the FDA who need urgently a drug to stop or slow the silent pandemic named NASH .





Share on StockTwits

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