Durect (DRRX): Not sure they have much of a chance with their AH trial
Almost the biotech equivalent of a Hail Mary
Sometimes when looking at a clinical trial you just wonder “why did they go to such a large trial when they have almost no data?” Durect is one of those cases. Based on an open label 19-patient trial in alcohol-associated hepatis (AH) they moved larsucosterol (DUR-928) on to a 300-patient Phase 2b trial.
The original Phase 2a trial enrolled both moderate and severe AH patients, following them for 28 days post-treatment after receiving either 30mg, 90mg or 150mg of drug. The Phase 2b is enrolling only severe patients, administering only the 30mg and 90mg doses and follows the participants for 90 days. So as you can see from the below only 8 of those patients in the Phase 2a trial received the doses (30mg and 90mg) that are to be studied and were severe enough to potentially be eligible for the larger Phase 2b trial. None were followed for 90 days as part of the trial.
The company likes to focus on the fact that none of these 19 patients died during the 28-day period, which obviously is a good thing but is not necessarily a sign of efficacy of the compound. See below the mortality seen across different trials in AH. It is literally all over the place.
Additionally, they enrolled younger patients (all patients were between the ages of 36-45) who have a higher chance of survival (note the below chart is survival during the hospital stay and not 28-day survival).
Also, based on their Model for End-Stage Liver Disease (MELD) scores, which predicts 3-month mortality due to liver disease, it doesn’t look like the drug had much of an impact, especially in severe patients.
Then there were the Lille scores which are used as a prognostic indicator for patients unresponsive to corticosteroids. A score of below 0.45 indicates a 6-month survival rate of 85% while a score above that indicates only a 25% 6-month survival rate. Look what happened at the 150mg dose which seems to have been doing harm compared to the lower doses (though granted the trial was small).
Let’s move on to the Phase 2b trial. It has 3 arms, placebo (standard of care, which means corticosteroids) and then 30mg and 90mg of larsucosterol.
The primary endpoint is 90-day survival or liver transplant. The company has been pretty evasive on the powering with lots of hedging language. Here is what they said on the Q2 2023 earnings call:
Question – Sean Kim: Hi. Thank you for taking my questions. My first question is could you please remind us whether the statistical plan for the AHFIRM trial has been submitted and if you have provided any detail on the statistical powering for the trial?
Answer – James E. Brown: Yeah. It certainly has been submitted but we haven't really provided a great bit of color to that suffice to say we designed it such that 300 patients should -- the trial proceed as we hope and the placebo group perform as history has dictated, and we're in the range of events that would allow for that then that will have a statistically significant trial but it all remains to be seen, as always. The data will dictate at the end.
They hope the placebo arm performs “as history has dictated?” Here is history:
The company likes to talk about a 30% mortality rate at 90-days but that is certainly not set in stone and not necessarily consistent across trials. And in a very recent trial of prednisone (which will likely be used in the control arm) versus anakinra and zinc in 147 severe AH patients, the prednisone arm had a 90% survival rate at 90 days. There literally might not be room for improvement on this by larsucosterol even if the drug actually has activity. It seems that it would have made a lot more sense for the company to have gone the intermediate route and ran a trial with something like 50 patients, while using a 90-day endpoint rather than jumping directly into this large trial, which I am guessing is costing the company between $15-30 million based on standard per-patient clinical trial costs.
Anyway, we’ll see what happens.
Please like, share and subscribe!
Stocks mentioned:
Although you make good points at time, at other times they are borderline off-topic. Ex. This trial is on severe patients or the 2a mild and severe. Yet your comparaison to the historical data mixes all types of AH patients and we do not know the numbers of patients behind it. So quite the meaningless graph. Another example, the company in a quarterly call mentioned that after 90 days they were able to contact 17 of the participants - with one missing (dead or alive?). Albeit it wasn’t in the 2a trial design it is worth mentioning. You also do not talk about the MOA and how it ties nicely with the mechanism behind AH.
So it’s a nice display of scepticism, which is healthy, but lacks depth as it seems very selected in its information and interpretation.