Hello all,
There have been few times in my life when I have wanted something so much and been so unable to get it. Let me tell you, I really want navitoclax. I mean, I really want navitoclax. To regular readers of this space, there is no need to elaborate why.
But for now at least, I can't have it. Accepting this idea has been difficult.
The problem is one of cost. Navitoclax is a mighty big molecule. It's unusually large for a for anything normally considered a "drug". Because of this, per gram it is typically available for $1000 to $1200. While I can theoretically afford this, it's not something that would be comfortable for me, especially if I wanted to experiment with it more that a couple times per year.
Logic has already investigated a group buy. (Thank you for that.) This brings the price down a bit, which is good, but not good enough. Unfortunately, the minimum order for such a synthesis purchase seems difficult to accomplish at this price. I am still hopeful that this group buy will happen but I'm not going to hold my breath.
So how to solve? What is the next step? I have spent a great deal of time over the last month trying to figure this out. Below, I am going to enumerate the various choices that I've wrestled with and weigh them here for your entertainment and edification. Note that my goal and the purpose of this list is specifically to kill or silence senescent cells. I realize that there are many other options (TOR, etc) and don't wish to disparage them by act of omission. If you are short on time, please skip to the end to item #8 for my conclusions and a possible group buy proposal that I hope you will consider.
The Options:
1: Wait for the navitoclax group buy to happen (i.e. Do nothing but wait.):
As I mentioned above, I'm still hopeful that Logic can make it happen but it might be quite a while so I am going to say no to this option.
2: Wait for navitoclax to be commercialized for our purposes:
A company named
Unity Biotech has sprung up to do exactly this. They have renamed it UBX0101.
Jeon, O.H., Kim, C., Laberge, R.-M., Demaria, M., Rathod, S., Vasserot, A.P., Chung, J.W., Kim, D.H., Poon, Y., David, N., et al. (2017). Local clearance of senescent cells attenuates the development of post-traumatic osteoarthritis and creates a pro-regenerative environment. Nat. Med.
Unfortunately, they still have to go through all the steps of getting it approved just like any company. In the best of cases it will be years before we have a pill to take. I can imagine that they are going to be setting the price point very, very high.
Question: If you started a company selling a drug that would only need to be taken once a year, what would you price it at? What if that drug was not to cure a disease but to sell to rich people who did not want to get old?
Answer: I would price what the market would bear. $50,000 per treatment? $80,000? The sky is the limit.
For now not an option.
3: Get a group buy together for the FOXO4-DRI drug:
Baar, M.P., Brandt, R.M.C., Putavet, D.A., Klein, J.D.D., Derks, K.W.J., Bourgeois, B.R.M., Stryeck, S., Rijksen, Y., Willigenburg, H. van, Feijtel, D.A., et al. (2017). Targeted Apoptosis of Senescent Cells Restores Tissue Homeostasis in Response to Chemotoxicity and Aging. Cell 169, 132–147.e16.
I have news for all of you waiting for FOXO4-DRI to become available. If you think navitoclax is expensive just wait for how much this one is going to cost:
Here is the sequence:
GSHMLEDPGAVTGPRKGGSRRNAWGNQSYAELISQAIESAPEKRLTLAQIYEWMVRTVPY
61 D-amino acids. Going rate is $55/residue for 100mg @98% purity. That's $3355 for 100mg. And that's a good deal.
Peptide synthesis is a mature, highly competitive business. This price is not likely to come down unless a major pharmaceutical company decides to make it giant reactors. Unless you know someone at the factory with sticky fingers, it's going to be expensive.
Not an option for now.
4: Buy a used peptide synthesis machine and make some FOXO4-DRI myself:
I can see some reasonably prices machines for sale right now at less than $3000. But the machine is not the problem. It's two things: the cost of reagents and the purification. Getting the crude peptide purified at large scale is a huge pain. This is why the markup is so high going from crude peptide ($2/amino acid) to 98%+ ($55/amino acid). In a past life used to purify solid phase synthesis nucleic acids, which are similarly synthesized. It's easy at a few milligrams because you can use HPLC (not that I have one of those right now anyway) but as you go up, the equipment gets expensive fast.
I'm not going to rule this option out but I don't want to try with limited equipment and money.
4: Wait for the FOXO4-DRI drug to be commercialized:
Again, years of waiting. Plus, the economics of the price are unlikely to change magically. See #2 above.
Not an option for me right now.
5: Intravenous quercetin:
Starting to feel desperate here... If I can't have anything else, well then fine. Can I make I.V. quercetin work?
This first thing is getting it to dissolve. It's pretty much insoluble, which is why these researchers in the phase I trial used DMSO. But I know something now that they didn't: Beta cyclodextrin greatly increases the solubility of quercetin:
Jullian, C., Moyano, L., Yañez, C., and Olea-Azar, C. (2007). Complexation of quercetin with three kinds of cyclodextrins: an antioxidant study. Spectrochim Acta A Mol Biomol Spectrosc 67, 230–234.
But what about the dosage? Look here for that info:
Ferry, D.R., Smith, A., Malkhandi, J., Fyfe, D.W., deTakats, P.G., Anderson, D., Baker, J., and Kerr, D.J. (1996). Phase I clinical trial of the flavonoid quercetin: pharmacokinetics and evidence for in vivo tyrosine kinase inhibition. Clin. Cancer Res. 2, 659–668.
From the paper above, it looks like the dose-limiting side-effects are severe nausea and kidney damage. (Yay!)
From page 662, it looks like severe vomiting start at about 630 mg/m^2 which for me is 1209mg. Renal toxicity starts at that point too so I would need to stay as far under 1200mg as possible. (Note that even close to this dosage would involve chemotherapy level nausea.)
Okay, now that I know what dosage I want to avoid, how much do I actually need to be effective?
On page 667 it says the 50% effective concentration is someplace in the range of 1 to 10 micromolar. Of course no one takes drugs at the EC50. You would need to be higher than that, probably quite a bit higher since drug-receptor binding is not linear.
But then here is the bad news:
"At doses above 630mg/m^2 [1200mg for me], quercetin plasma levels where in excess of 1 micromolar for at least 3 hours."
Crap. These numbers don't work. To get to a therapeutic dose, I would have to be in severe vomiting and renal toxicity range for like eight hours. This would meaning involving not just one bolus injection but a drip. Sounds like torture to me and a possible death sentence if I'm not careful.
Nope. Forget that terrible idea. No way I'm doing IV quercetin. (And before you click "Dangerous, Irresponsible" re-read the previous sentence. I am not going use I.V. quercetin.
6: Synthesize navitoclax myself:
Yeah, I'm a chemist, among other things. However, there is no way I want to macro-scale synthesize this entire thing in my "spare" time. It's too daunting to think about while still needing to work at my main job and keep all my other projects alive.
But how about something smaller?
7: Synthesize part of navitoclax to help reduce the price:
Retrosynthetic analysis of navitoclax gives us three main pieces:
Here are my synthetic goals:
No reactions needing chromatography.
No air sensitive reagents or schlenk lines.
As few expensive reagents or solvents as possible.
Extend synthesis all the way back to cheap raw materials.
I started thinking about making just piece III above. I was only half serious about the idea but after a few days reading and sketching ideas, I had something that looked pretty workable. A while later I spent another couple days and made it even better. Astrazeneca has a patent where they do it in eight step at only 6% yield (US2012/0035134 A1). Abbvie does it in 19% overall yield (US2005/272744 A1). I also found a Chinese patent that claims 24%. By combining all these along with some other reactions I changed to meet the goals listed above, I think I can do it in seven steps at someplace around 40-45% yield starting from D-aspartic acid (cheap) if I don't screw up. How can I achieve a better yield than the patent? I can combine reactions from multiple patents and other literature in ways these patent writers legally cannot.
Next I moved on to piece I. This one looks bigger but I think it's actually easier. I think I can do this in seven steps starting from para-cresol (fairly cheap). The fun part is that I think I can get derive the benzyl rings almost completely from para-dicholobenzene (very cheap). There would be some amount air-sensitive steps but they are manageable the way I have it planned.
I'm not going to bother with piece II. It would involve large scale chromatography to separate out various product isomers and I can't envision doing that at home. Too messy, too expensive.
In any case, I'm guessing it would take at least 6 months each of spare time to do these at large quantity. Some of my loved ones over the years seemed to think that chemistry is like waving a magic wand over something and poof! Nope, actually, it's a lot of work, especially when you don't have all the right equipment and space. But, after putting almost two weeks of effort doing synthesis planning, I have to at least try one of them, maybe just for the fun of it, so I've started to slowly assemble the items I need.
Nonetheless, I am back at playing the waiting game, even if it's waiting for myself.
8: Use a senotherapeutic rather than a senolytic:
This is the choice that I have struggled over more than any other and the one I most want to start a discussion about. Here is the story: A few years ago before I ever heard of navitoclax, dasatinib or Longecity I was obsessed with the SASP (Senescence-Associated Secretory Phenotype) and "inflammaging". I spent nearly all my free time researching it for what was about a year. As a reminder, here is the basic sequence:
After cells are damaged or reach their division limit this triggers two parallel pathways:
A) p19ARF is activated -> p53 is activated -> p21 is activated -> CDK2 is blocked -> the cell cycle is stopped.
B) p16INK4a is activated -> CDK4 and CDK6 are blocked -> the cell cycle is stopped.
Along with this change in the cell cycle, the cells start pumping out inflammatory cytokines. What follows is an inflammatory chain reaction with IL-1alpha being the initiator and IL-6 causing the most damage and sustaining a viscous cycle.
Campisi, J., Andersen, J., Kapahi, P., and Melov, S. (2011). Cellular senescence: a link between cancer and age-related degenerative disease? Semin Cancer Biol 21, 354–359.
At the time I realized that if I could somehow block IL-6 then maybe I could interrupt or slow down aging. Unfortunately, there are still no small molecule inhibitors of IL-6 receptor. However, there are inhibitors of its downstream signal transducer, Janus kinase 1 (JAK1). Most notably
Jakafi (Ruxolitinib) and
Xeljanz (Tofacitinib). Sadly, these drugs have dose-limiting side effects due to the fact that they are not selective for JAK1, affecting JAK1/JAK3 and JAK1/JAK2 respectively. At this point I started looking really hard for a selective JAK1 inhibitor. The first one that was publicized was from a European pharmaceutical research company called Galapagos. At the time it was called GLPG0634 but is now is goes by
Filgotinib . At the time there was no structure available so I just kind of watched and waited.
In the meantime, my hypothesis about JAK1 and IL-6 was confirmed. Check out the following articles:
Xu, M., Tchkonia, T., Ding, H., Ogrodnik, M., Lubbers, E.R., Pirtskhalava, T., White, T.A., Johnson, K.O., Stout, M.B., Mezera, V., et al. (2015). JAK inhibition alleviates the cellular senescence-associated secretory phenotype and frailty in old age. PNAS 112, E6301–E6310.
In the above paper, they show, ruxolitinib reduces frailty even in very old mice.
Xu, M. et al. (2016) Perspective: targeting the JAK/STAT pathway to fight age-related dysfunction. Pharmacol. Res. 111, 152–154
Doles, J.D., and Olwin, B.B. (2014). The impact of JAK-STAT signaling on muscle regeneration. Nat Med 20, 1094–1095.
Soto-Gamez, A., and Demaria, M. (2017). Therapeutic interventions for aging: the case of cellular senescence. Drug Discov. Today.
Xu, M., Palmer, A.K., Ding, H., Weivoda, M.M., Pirtskhalava, T., White, T.A., Sepe, A., Johnson, K.O., Stout, M.B., Giorgadze, N., et al. (2015). Targeting senescent cells enhances adipogenesis and metabolic function in old age. eLife e12997.
After reading one of the papers above, I decided to take a crack at synthesizing filgotinib. Due to life events, I put on the back burner. This was when I discovered Dasatinib and starting looking for a source. I stumbled across Longecity and here I am.
What I am proposing is a Group Buy for one of these JAK inhibitors. Since it would have to be taken every day, the following criteria would have to be met:
1: Lack of or limited side effects.
2: Safety established in at least Phase II studies.
3: Small enough molecule to be synthesized at reasonable cost.
4: Potent enough so that small amounts could be taken per day.
Right off the bat, Ruxolitinib and Tofacitinib are off the consideration because of their non-specificity for JAK1 and therefore high amount of side effects. That leaves us three drugs that are currently in development for JAK1:
1: Filgotinib from Galapagos/Gilead:
Dosage: 25-100 BID or 50-200 QD
2: Upadacitinib from Abbvie:
Dosage: 6-18mg BID
3: PF-04965842 from Pfizer:
Dosage: 200mg BID I think? (Not a lot of info about this drug yet. Can't remember where I saw this.)
Off the bat, upadacitinib looks the best. However, I've spent the last three days deciphering the Abbvie patent. In spite of being about as large as filgotinib, it's much more difficult to synthesize. 17 steps for upadacitinib while filgotinib is seven or eight, depending on how far back you go. Plus, upadacitinib is optically active (chiral), which make the synthesis even more difficult. (And as I've said before, cost does not grow linearly with steps, it's exponential.)
Here is the Abbvie patent if you are interested:
Voss, J.W., Camp, H.S., and Padley, R.J. (2015). Jak1 selective inhibitor and uses thereof. US20150118229 A1
I think upadacitinib is the best one without considering price. It's about 5X more potent than filgotinib but unless it's less than 5x more expensive per gram which I doubt, I would maybe recommend filgotinib.
A years supply of filgotinib would be something like 18-36 grams. If it's cheap that seems like a reasonable purchase, considering its about the size and complexity as dasatinib and I bought 20 grams of that stuff for very little.
A years supply of upadacitinib (a better drug in terms of potency and lack of side effects) would be 4.4 to 13.1 grams.
So that is my humble suggestion for a viable next step. It is backed by peer-reviewed research showing positive effects on aging and clinical trials demonstrating safety in humans.
I welcome your comments.
Edited by jmorris, 25 May 2017 - 05:38 AM.