IDE397 (MAT2A inhibitor)
Ideaya Biosciences
IDE397 is an oral MAT2A inhibitor by Ideaya Biosciences (IDYA 0.00%↑) that is currently in phase 1/2 trials for solid tumors harboring methylthioadenosine phosphorylase (MTAP)-homozygous deletions (NCT04794699).
Some bio background on why people think this is a good target
Homozygous deletion of CDKN2A/B at the 9p21 chromosome is an early clonal event in tumorigenesis. MTAP is frequently co-deleted with CDK2NA tumor suppressor locus across multiple tumor types, which has spurred much interest in finding therapeutic targets that emerge as a result of MTAP-homozygous deletion. This is because there is a distinct metabolic phenotype associated with MTAP-deleted tumors: accumulation of its substrate, methylthioadenosine (MTA).
Loss of MTAP yields selective dependence on protein arginine methyltransferase 5 (PRMT5) and its binding partner WDR77.
Extreme intracellular accumulation of MTA can inhibit PRMT5, a key regulator of transcription, which can later lead to cell death.
Deletion of MTAP in cancer cells does not, by itself, sufficiently increase levels of intracellular MTA to yield cell death, however. So people are trying to figure out ways to further increase levels of MTA in cancer cells.
How to do this? By inhibiting other enzymes that metabolize MTA!
Two main enzymes metabolize MTA: MAT2A and PRMT5. Ideaya has chosen to develop a MAT2A inhibitor to try and boost intracellular levels of MTA in MTAP-deleted cells which could theoretically kill cancer cells in a selective manner and with a wide therapeutic window.
Very cool if it does work. The whole premise hinges on the ability to therapeutically increase levels of MTA to the extreme.
Coincidentally, the lab I used to work in published a study using MTAP-deleted cell lines & primary glioblastoma tuumors questioning the therapeutic feasibility of this strategy. Worth a read.
Hello darkness my old friend
Anyway, it’s been a while since I’ve posted on an oncology target! I recently saw an old post from Evaluate Vantage on how eeeeverybodyy is entering the MTAP space. So why not check up on one of the OG MTAP pipelines: Ideaya.
As I mentioned, I did brush shoulders with the MTAP space and so the main players have been in my periphery for some time, including IDE397.
To Espacenet I went!
Searched “ideaya biosciences” and sorted the results by descending priority date and well, look what I found.
Two combination patents, what luck! Maybe even better than landing a solid dosage forms patent.
Both patents describe combinations with the same “Compound A”…
Perhaps this is a quirk of oncology patents but the linchpin for the the last 2 oncology compounds I’ve cracked has been the combo therapy data in xenografts. Or maybe it’s just that those are the only ones I can figure out haha. Anyway…
Same story, different compound
This story is short. And thank goodness it is because the last post on SAGE-324 was long as heck. I was SUPER SURPRISED that no one has (publicly) blown this one open yet…because there are fingerprints EVERYWHERE.
As soon as I saw the combo data, I began digging through Ideaya’s investor materials. Thinking this was going to be IK-930 2.0, I searched through their 2023 and 2022 10K filings, hoping that I would come across equivalent graph showing the combo data in mice.
But sadly, Ideaya wasn’t going to make it that simple! No graphs to be found. Instead, I found these gems:
Suuuuper key. Remember the combo patents? No, not the data—forget that for a sec! The inventors! For the type I PRMT inhibitor combo patent.
I was surprised by this co-inventorship when I first saw this too—caught it as I was highlighting filing dates (as is my habit). Pretty unusual to see co-inventorship from 2 biotech/pharma companies (at least from what I’ve seen), so I decided to look further.
Behold these articles from Fierce Biotech. Turns out, Ideaya was about to make a deal with GSK for exclusive licensing rights to IDE397. But GSK backed out in Aug 2022, after axing their MTAP program (PRMT5 & PRMT1).
And just to check, I searched the structure of GSK3368715 (the PRMT1 inhibitor that GSK ditched). Sure enough, the structure on Selleckchem matched ”Compound B” in the type I PRMT inhibitor patent (‘808).
So you can see why I leapt with excitement when I saw that blurbs in the 2023 & 2022 SEC 10Ks! I found my footing for this compound. AKA, “Compound A” in the combo patents may be IDE397.
My spidey senses told me that I might find the graphs I sought in one of the company presentations. So I ditched my efforts scouring the SEC wasteland. There was only 1 company slide deck and it was last updated April 2023. Super fresh. How convenient. Let us begin 🥬 .
Ctrl+F ”397”. Enter, enter, enter, enter…
Aha!
Combination data with a PRMT5MTA-2 inhibitor in an NSCLC (MTAP-/-) CDX! Just need to see which of the graphs in the type 2 PRMT5 patent (‘806) this could be…
How did I know to look in the type II PRMT patent instead of the other one? Well actually I didn’t–at first.
Whenever I open up a patent PDF, I will skim through the whole thing and make mental notes of my impressions. When I saw the graphs in the company slides, I remembered seeing those doses in the type II PRMT patent, so I went there. Just so happened that my memory served me correctly, so I just went with it. Also, I figured that the most up-to-date company slides would probably not include data from a discontinued partnership (GSK).
It’s a match!
So which of the graphs in the type II PRMT inhibitor patent (‘806) could it be? I skimmed carefully...until I found what I was looking for! Figure 4!
I actually looked through these graphs without knowing what CDX model the data corresponded to...because:
Graphs were untitled
Info saying what corresponded to what was buried in the Example descriptions that I did not want to look through at that point because I was too excited to find a match.
In retrospect, I probably should’ve tried to label the graphs first, but alas…
I mention this because looking at untitled graphs can be quite disorienting—even if I was only looking for patterns. This is because, in this case—unlike IK-930—what is presented in the company graph and what is shown in the patent is not exactly the same. The patent graph is more cluttered because it includes way more treatment arms. The company presentation graph is cleaned up to show what they want to show. So I was a bit disoriented but I managed to find my way.
And voilà!
After orienting myself on the vehicle (making sure the shape of the graph is the same, making sure the bounds of the standard deviation bars match), I weaved through the lines to find just what I was looking for.
There it is!
And just to check to make sure they’re both the right CDX model, I look back through the ‘806 patent. From Example 6:
So the data from Figure 4 & 4a were generated in NCI-H838, a NSCLC CDX that’s MTAP-deleted, consistent with the company presentation. Check.
Coupled with the fact that we know GSK was once looking to combine their (discontinued) PRMT1 inhibitor GSK3368715 with IDE397 & we find the structure of GSK3368715 in the type I PRMT inhibitor patent (‘808), I’m ready call it.
Compound A = IDE397
And looking back through the type II PRMT inhibitor patent (‘806), we find the structure of Compound A is also the same Compound A as in the type I PRMT inhibitor patent (‘808; AKA, the one with the GSK compound). Check.
Wooo!
And if you’re paying attention, I’m sure you have some thoughts on what “Compound B” might be. I certainly have some ideas! But need more confirmation first to get to the ≥99% confidence Molecular Sherlock aspires to ;) So until then, peace out & see you in the next one!
Have a compound that you want me look into? Suggest a compound here.
https://www.targetmol.cn/target/methionine_adenosyltransferase_%28mat%29
S-Adenosylmethionine synthetase, also known as methionine adenosyltransferase (MAT), is an enzyme that creates S-adenosylmethionine (also known as AdoMet, SAM or SAMe) by reacting methionine (a non-polar amino acid) and ATP (the basic currency of energy).