Biotech Values

[iwfal]

FGEN Prediction: when the full data on Non-Dialysis (ND) MACE/MACE+ is made available, I predict there will be a substantial difference in HR for patients above GFR=15 vs the HR for those with baseline GFR below 15 – with high GFR patients having much better MACE HR than low GFR patients. Why do I predict this? Because:


a) Low GFR patients traditionally have a high rate of adverse events with renally active drugs, regardless of drug class. For instance, the classes of drugs most prescribed for slowing renal failure are ACE/ARB, but they are a concern once the GFR is too low. And PHD Inhibitors are definitely renally active.


b) Fibrogen started talking about ‘GFR decline’ benefits specific to GFR>=15 patients in late February, when previously (Jun last year – see 8k transcript) no such differentiation was made with respect to baseline GFR. It is, of course, possible to interpret the Feb disclosure in other ways – e.g. as a hint as nothing more than benefit in GFR decline did not occur in patients with baseline GFR under 15. But even many of these other interpretations imply some adverse event benefits (e.g. MACE) for GFR>15 patients since declining GFR leads to more adverse events (and once under 15, it would generally be a fast decline in terms of adverse events)


c) A late breaking decision to split ND patients at GFR=15 (presumably driven by looking in detail at some of the previously unblinded, but smaller ND trials) would also substantially explain the complete catastrophe of the early May cc since:

C1) If they did only recently decide to split ND at GFR=15, then they potentially wouldn’t have had time to finish discussions with the regulatory agencies before their planned unblinding of the big MACE trials.

C2) Assuming an unresolved set of discussions with a regulatory agency (i.e. FDA – although it will be interesting to learn why no such issue with EMA) the companies would have to be EXTREMELY reluctant to try to discuss it in public, since regulatory agencies generally look unfavorably on litigating in the court of public opinion.

C3) NOTE: I am assuming that the ITT ND MACE data is not great. Probably meets non-inferiority, but maybe not – either way I’d be surprised if was very good. (If it had some redeeming ITT characteristics I assume they would have said something, however reluctantly, as they did with Stable Dialysis.)

Final note – I am reluctant to speculate on how much clinically better (e.g. does GFR>15 have MACE/MACE+ HRs under, say, 0.9, while GFR<15 is over 1.2? I don’t know.)

Previous predictions related to FGEN:

a) I speculated in 2016 that there were would be a GFR benefit for ND patients. This seems to have been largely confirmed, although they still haven’t clarified whether it is a one-time bump or a change in slope. (At this point I suspect primarily just a bump)

b) I speculated that there would be at least a transient issue with hyperkalemia or acute dialysis in the ND trials. We will need to wait to see the FDA analysis to know.

c) I speculated that there would be an imbalance in liver related adverse events. Perhaps as mild as statin-like LFTs, perhaps more. Given that the trials made it to the end, obviously problems, if any, were on the milder end of the spectrum. But, again, we will need to wait for the FDA documents to see whether there were any imbalances of milder issues (note: AKBA recently disclosed a Hy’s Law, and their trials have nonetheless continued well beyond that event).

Quote from FGEN 18Q4 cc at end of Feb:

Preliminary results from a full analysis on patients with baseline eGFR 15 or higher in the Phase III placebo-controlled studies show that 1-year decline in eGFR in roxadustat is significantly less than placebo in CKD Stages 3 and 4 patients. We believe roxadustat treatment could offer significant clinical benefit in the non-dialysis-dependent patients by attenuating renal progression, as seen in slowing down the declines in eGFR over time.