DFV890 (NLRP3 inhibitor) ✔️
Novartis AG
Confirmed correct on May 17, 2024 in Figure 1 of the Clin. Transl. Sci. paper describing the ph1 study.
DFV890 (formerly IFM-2427) is an oral NLRP3 inhibitor acquired by Novartis (NVS 0.00%↑ ) from IFM Tre in 2019. It is currently being developed for a variety of inflammatory diseases, including osteoarthritis, for which it is in phase 2 trials (NCT04886258).
NLRP3 is highly expressed in macrophages and is part of the inflammasome, which is a protein complex that detects products from damaged cells, such as extracellular ATP.
I stumbled across Novartis’ inflammation portfolio while I was perusing my favorite patent database, the EPO. A patent on dosing regimens for an NLRP3 inhibitor for osteoarthritis (OA) caught my eye, so I decided to look into this.
Which one is it?
My approach for this case was different than usual, since I had started from the patent space, rather than from a company’s pipeline. This wouldn’t be an issue if the company only had 1 compound per therapeutic area…but Novartis has 5 active trials with 4 different drugs being trialed for OA.
So which one is the dosing patent (‘399) referring to?
From the title, you already know it’s DVF890. But how did I get there?
Not all roads lead to Rome
The dosing patent (‘399) provides a compound structure & specifies 2 isomers. But we need to connect it to one of the 4 compounds being trialed. So which one is it?
As I dig deeper into the patent, I encounter 3 examples:
Example 1: Clinical study with Compound 1A
Example 2: Clinical first-in-human (FIH) study
Example 3: various in vitro methods on how to study the efficacy of candidate compounds
Nestled in the clinical study examples is a useful detail: Compound 1 is dosed ORALLY.
Important detail! Of the 5 compounds Novartis is currently trialing in OA, only 1 is orally administered: DVF890.
And, as an additional check: when you look up DVF890, you will immediately see that it was assessed in a phase 2 trial for COVID-19 as an oral NLRP3 inhibitor.
Okay, so now we’ve narrowed it down to DFV890 as the likely compound that the dosing patent (‘399) is talking about. But can’t jump the gun now! We’ve got to make sure that Compound 1A in ‘399 is actually DFV890.
Lil’ detour: structure of MCC950 points to oral RoA for Compound 1A
Apart from the dosing patent (‘399) plainly stating that Compound 1A is an oral compound, I came across a tool inhibitor of NLRP3, MCC950, early on in this case. In fact, when I was searching to see if the structure of DFV890 had been published, I almost mistook MCC950 for being DFV890 at first.
Veeeeeery similar! But we can see that Compound 1 has key modifications to improve patentability I mean pharmacology.
Anyway, the reason why I bring up MCC950 is because it is also dosed orally in preclinical studies. This strengthens the case that the structure of Compound 1 in ‘399 is oral. Yes, I know they literally say it’s dosed orally in the patent, but it’s a good check to see that a related structure also has the same RoA.
Final touches
Okay, so we have all this set up suggesting Compound 1 in ‘399 may be DFV890. Let’s close this one out.
Example 1 in ‘399 describes a clinical study design with Compound 1A. Not just any clinical study. It specifically describes, “A randomized, two-arm, placebo-controlled participant and investigator-blinded phase 2 study investigating the efficacy, safety and tolerability of Compound 1A in patients with symptomatic knee osteoarthritis.”
Which, as it so happens, is the exact title of the phase 2 DFV890 trial (NCT04886258)—right down to the lack of Oxford comma. Just switch out “Compound 1A” for “DFV890”.
Excellent start. Now let’s see if the study details in the ‘399 patent and the phase 2 trial match up. As we can see, the DETAILS of the primary & secondary endpoints line up—right down to the measurement tools & measurement time frames.
I know what some of you are thinking—yeah, that’s cool Victoria, but aren’t these just endpoints and measurement tools that ANY OA drug would look at?
Good point!
Keeping in mind the establishing context we have already laid out, I present to you: the DOSING ARMS!
Had to go the EU Clinical Trials Registry (EudraCT Number: 2020-006104-17) because Trials.gov is being cheap with the info. As we can see, the doses for Arm 1 and Arm 2 described in the ‘399 patent perfectly match the description provided in the EU Trials Registry for the phase 2 DFV890 trial. Arm 1: 10 mg BID, Arm 2: 25 mg BID.
Side note: I can bank on the EU Trials Registry to come through with the CAS number for registered trials. But sadly, no dice here.
De-CYPhering the answer!
As I mentioned, I think the linchpin of this case was…CYP isoforms! On the subject of drug-drug interactions, the ‘399 dosing patent mentions:
So Compound 1A is a CYP2C9 substrate. Now, this could just be my inexperience talkin’ here, but I found this interesting because most drugs I’ve looked at that are subject to CYP metabolism are primarily CYP3A4 substrates (examples: ritonavir, remdesivir, etc.). So, this CYP2C9 detail caught my eye.
I checked all registered trials of DFV890. Sure enough, all but 1 of the 4 trials (NCT04868968) include CYP2C9 modulators in their exclusion criteria. This strengthens the link between Compound 1A and DFV890.
The icing on the cake can be found in the study protocol (Clinical Trial Protocol CDFV890D12201) for the DFV890 in COVID-19 phase 2 trial (NCT04382053; Madurka et al Infection 2022 from above). On Trials.gov, this trial says it HAS RESULTS, which almost always means that the study protocol is in included.
Browsing through the protocol, I find this detail:
There you have it: DFV890 is mainly eliminated by hepatic CYP-mediated metabolism, with CYP2C9 being the main isoform involved: 68%, which is exactly the same % as in the ‘399 patent. Okay, sure, there is the 29% vs. 30% CYP3A4 metabolism between the patent and the study protocol. But in the face of all other evidence here, I’m going to give this one up to rounding.
So, in my opinion, Compound 1A = DFV890.
Different detective strategy for this compound. But that’s what makes it interesting! Catch y’all in the next one :)
Have a compound that you want me look into? Suggest a compound here.