There have been many headlines in recent days about a potential malaria cure. I'm not sure what set these off at this time, since the paper describing the work came out back in the spring, but it's certainly worth a look.
This all came out of the Medicines for Malaria Venture, a nonprofit group that has been working with various industrial and academic groups in many areas of malaria research. This is funded through a wide range of donors (corporations, foundations, international agencies), and work has taken place all over the world. In this case (PDF), things began with a collection of about 36,000 compounds (biased towards kinase inhibitor scaffolds) from BioFocus in the UK. These were screened (high-throughput phenotypic readout) at the Eskitis Institute in Australia, and a series of compounds was identified for structure-activity studies. This phase of the work was a three-way collaboration between a chemistry team at the University of Cape Town (led by Prof. Kelly Chibale), biology assay teams at the Swiss Tropical and Public Health Institute, and pharmacokinetics at the Center for Drug Candidate Optimization at Monash University in Australia.
An extensive SAR workup on the lead series identified some metabolically labile parts of the molecule over on that left-hand side pyridine. These could fortunately be changed without impairing the efficacy against the malaria parasites. The sulfonyl group seems to be required, as does the aminopyridine. These efforts led to the compound shown, MMV390048, which has good blood levels, passes in vitro safety tests, and is curative in a Plasmodium berghei mouse model at a single dose of 30 mg/kg. That's a very promising compound, from the looks of it, since that's better than the existing antimalarials can do. It's also active against drug-resistant strains, as well it might be (see below). Last month the MMV selected it for clinical development.
So how does this compound work? The medicinal chemists in the audience will have looked at that structure and said "kinase inhibitor", and that has to be where to put your money. That, in fact, appears to have been the entire motivation to screen the BioFocus collection. Kinase targets in Plasmodium have been getting attention for several years now; the parasite has a number of enzymes in this class, and they're different enough from human kinases to make attractive targets. (To that point, I have not been able to find results of this latest compound's profile when run against a panel of human kinases, although you'd think that this has surely been done by now). Importantly, none of the existing antimalarials work through such mechanisms, so the parasites have not had a chance to work up any resistance.
But resistance will come. It always does. The best hope for the kinase-based inhibitors is that they'll hit several malaria enzymes at once, which gives the organisms a bigger evolutionary barrier to jump over. The question is whether you can do that without hitting anything bad in the human kinome, but for the relatively short duration of acute malaria treatment, you should be able to get away with quite a bit. Throwing this compound and the existing antimalarials at the parasites simultaneously will really give them something to occupy themselves.
I'll follow the development of this compound with interest. It's just about to hit the really hard part of drug research - human beings in the clinic. This is where we have our wonderful 90% or so failure rates, although those figures are generally better for anti-infectives, as far as I can tell. Best of luck to everyone involved. I hope it works.
1. milkshake on September 4, 2012 9:55 AM writes...
3,5-diaryl-2-aminopyridines are pretty promiscuous class of inhibitors. I was on c-Met project in SUGEN when Crizotinib class of compounds was optimized from a HTS 3,5-diaryl-2-aminopyridine hit just like this one (I worked on another series) and these 3,5-disubst aminopyridines were non-selective like hell even after a year and half of optimization. Management liked it because the compounds covered all sorts of mutant c-Met kinases that drive malignancies and can be quite distinct from the wild form... The irony is, Pfizer re-suscitated the project for LKA kinase - a target bearing some active site/activation loop similarities with Met. So crizotinib was never SAR otimized for its current indication target.
Permalink to Comment2. John Wayne on September 4, 2012 10:35 AM writes...
The probable promiscuous nature of this molecule may be it's key to success against malaria. It has been my experience and observation that molecules that target a specific biochemical pathway are subject to rapid resistance, while 'dirtier' compounds have a greater potential lifetime in clinical practice.
Let's hope that this molecule is promiscuous enough to remain an effective therapy against malaria while maintaining a reasonable safety profile in humans. Nice work to everybody on the team.
Permalink to Comment3. Pete on September 4, 2012 10:40 AM writes...
Looks interesting. Here's the spin from S'frica:
allafrica.com/stories/201208300353(dot)html
Permalink to Comment4. Dr. L on September 4, 2012 11:24 AM writes...
In my opinion, any single small molecule compound against malaria is doomed. P falciparum is simply too intelligent of an organism for such attempts to be successful. Biologics (e.g., bioengineered falciparum enzymes which act as Troyan horses) are probably the next good choice.
Permalink to Comment5. cynical1 on September 4, 2012 11:56 AM writes...
A couple questions - I don't know the animal models for malaria at all but do the existing agents require more than one dose to eliminate the parasite?
Also, is the resistance that is observed from existing agents caused by elimination of the unmetabolized drugs in feces and/or urine winding up in the water where mosquitoes breed and presumably drink thus exposing the parasite to subclinical doses of the drug? Or is it from mosquitoes biting infected people who have developed the resistant parasite from treatment and then passing it on to others? Or neither?
Permalink to Comment6. barry on September 4, 2012 12:02 PM writes...
part of the joy of a phenotypic screen in infected mice is that we already know that the mice survive treatment, even if the cmpd proves to be an indiscriminate kinase poison (does staurosporine work in this model too?)
Permalink to CommentOf course if the side effects are grim enough, compliance will be an issue and this will remain a last-line treatment. Still, we're talking about a disease that claims a million lives a year.
7. Chemjobber on September 4, 2012 12:09 PM writes...
Am I correct in understanding that artemisinin is the drug of choice? If so, how does this compound compare?
Permalink to Comment8. petros on September 4, 2012 12:43 PM writes...
Dr L's point ignores the problems of treating malaria patients, mostly children, in hot, humid climates.
Therapeutics need to have a very low cost of goods, a very short course of treatment, ideally a single dose, and to have a high degree of chemical stability because refrigeration is rarely available at the point of treatment.
Permalink to Comment9. Anatoly on September 4, 2012 12:54 PM writes...
@4
Biologics? Into erythrocytes? Really?
Come on, biologics are not a cure-all.
Permalink to Comment10. Pete on September 4, 2012 1:21 PM writes...
Although the original hits were identified from a screen of a kinase-focussed library, it's not clear (to me at least from a cursory read of the paper of my computer screen) that this compound is hitting a kinase. Adenine mimics have the potential to interfere with a cell's ability to both use and make ATP.
I did wonder about whether the amino group was essential. Could the electron withdrawing groups be preventing the pyridine from protonating in an acidic intracellular compartment?
Permalink to Comment11. Dr. L on September 4, 2012 2:08 PM writes...
@9 It is long well known that Plasmodium infected RBCs are capable of efficient endocytosis so a transport of structurally robust enzymes into these cells is feasible.
@9 I meant to say that for anything close to malaria eradication we have to go beyond small molecules or vaccines made from a handful of random antigens. I agree with you that the next stage of the battle against malaria will be difficult, and the efficient delivery of medication to those patients remains a big hurdle.
Permalink to Comment12. g0pr on September 4, 2012 2:19 PM writes...
@1 I think you meant ALK (Anaplastic lymphoma kinase)... It's interesting that you mention "Crizotinib class of compounds was optimized from a HTS 3,5-diaryl-2-aminopyridine hit"... However, it was supposed to be an outcome of rational structure based design based on crystal structure of their (Pfizer) indolinone compound (with c-Met)[J. Med. Chem paper from Pfizer DOI: 10.1021/jm2007613]!! So it's case of make a nice story for the sake of publication??!! Just sayin..
Permalink to Comment13. milkshake on September 4, 2012 2:34 PM writes...
@12: Sorry should have been ALK (I made a typo). The project story goes like this: SUGEN had a series of c-Met oxindole compounds that looked like Sutent with the 5-fluoro replaced with halogenated benzylsulfone. These were truly terrible compounds from PK point of view with unpredictable multiple metabolism modes, hard to formulate (they tend to precipitate in the injection site) and so on. I too wasted my time on these. The breaktrough came with Pharmacia HTS screen of Kalamazoo libraries that provided two unrelated lead series. Solving a crystal structure of c-Met cocrystal both with the old oxindole-sulfone lead and the new aminopyridine series (and also for the third lead) showed a decent overlap of the two so the oxindole compound binding mode did in the end inform the design of Crizotinib series
Permalink to Comment14. PharmaHeretic on September 4, 2012 6:19 PM writes...
So the single biggest advancement over the last 30-40 years have been all the drugs and procedures to treat hypertension and other cardio-vascular diseases. Ironically advances in the treatment of cancers has had a much smaller impact on its share of mortality.
http://www.scientificamerican.com/article.cfm?id=longevity-why-we-die-global-life-expectancy#
Permalink to Comment15. Pali on September 4, 2012 9:52 PM writes...
It is funny that somebody would think that a kinase inhibitor has the ability to cure malaria. Same is true for cancer.
Permalink to Comment16. Matthew K on September 4, 2012 10:03 PM writes...
If the kinase is essential for continued survival, specific to the parasite and well targeted by the drug, where's the problem?
Permalink to CommentClearing a parasite is a TOTALLY different problem to eliminating mutant human cells. Why would you compare it to cancer at all?
17. SMILES on September 5, 2012 12:25 AM writes...
Watch out the genotoxcity issues for this kind of aromatic amine esp as regulatory recommended at much high doses in GLP tests will be used
Permalink to Comment18. Kate H on September 5, 2012 2:11 AM writes...
The headlines have probably been triggered by the press release from H3-D at the end of August: http://www.uct.ac.za/dailynews/?id=8220
Permalink to Comment19. Jess on September 5, 2012 4:46 AM writes...
This really isn't as big a deal as everyone is making it, the parsite species infecting and killing over 1 million people a year is plasmodium falciparum, not plasmodium berghei. the mechanisms of infection are quite different between humans and animals. Even if by some small chance it works in humans it will take many years of clincal trails before it can be put on the market.
Permalink to Comment20. malaria madness on September 5, 2012 6:49 AM writes...
Jess -
There are (not many) no good in vivo animal models of P. falciparum which is why the P. berghei model is used. It is one the standard in vivo models used for antimalarial discovery and a good benchmark.
See NRDD 2004 509-520
"So, in vivo evaluation of antimalarial compounds typically begins with the use of rodent malaria parasites. Of these, P. berghei, P. yoelii, P.chabaudi and P. vinckei(see Further Information, Antimalarial drug discovery: efficacy models for compound screening) have been used
Permalink to Commentextensively in drug discovery and early development. Rodent models have been validated through the identification of several antimalarials — for example, mefloquine, halofantrine and more recently artemisinin derivatives. In view of their proven use in the prediction of treatment outcomes for human infections, these models remain a standard part of the drug discovery and development pathway."
21. milkshake on September 5, 2012 12:37 PM writes...
@17 EWG-substituted anilines have much lesser genotoxicity potential (see Me-anthranilate, PABA, p-aminosalycilic acid are perfectly acceptable in massive doses) because the metabolic activation by N-hydroxylation is not as facile as with el rich anilines. And I have yet to see any reference to genotoxicity of aminopyridines, aminopyrimidines and the like
Permalink to Comment22. A Nonny Mouse on September 5, 2012 1:49 PM writes...
Main problem with this compound will be the cost of synthesis for third world use; both the boronic acids used in the synthesis are very expensive. It will need a very different approach if it is to be used for anything but Western visitors.
Permalink to Comment23. MoMo on September 5, 2012 2:41 PM writes...
Right on Nonny Mouse! The catalysts alone will send the finances running. Many a malaria program has suffered from the following words-
Is it cheap to make-less than 14 cents per dose?
Is it safe for children?
Not fair questions when people are dying, and you would think Gates Foundation wouldnt care-
But these are the words even they speak.
Permalink to Comment24. metaphysician on September 5, 2012 4:06 PM writes...
23-
Why wouldn't they ask exactly those questions? The Gates Foundation isn't in the business of funding malaria research, its in the business of producing malaria treatments. There is no point funding a treatment that will be useless for the intended purpose: treating third world poor.
Permalink to Comment25. barry on September 5, 2012 5:08 PM writes...
plasmodium and homo parted evolutionary ways a long time ago. If this is a promiscuous kinase poison (as seems likely), still--if one were to invest the effort to identify the molecular target and solve a co-crystal structure by x-ray diffraction--there's a better chance to get to a clean malaria drug here than there is to get a clean cancer drug from a kinase program
Permalink to Comment26. Pali on September 6, 2012 4:03 AM writes...
#16/Matthew K: Kinase inhibitors are grossly overrated as drug targets. See for instance comments #1 and #2 - already indicating a different mechanism for this class of compounds. And please explain how kinase inhibitors get into cells? All of the above is true for cancer as well and suggest the mechanism of action is different from the claimed one.
Permalink to Comment27. MoMo on September 6, 2012 2:32 PM writes...
MetaPhysician- Why ask those questions?
Because low-cost cheap drugs for 3rd world countries are already mined and being used.
face it, the days of cheap and easy drugs are gone.
But many compounds have failed the MMV litmus test and may have worked fine- except for being cheap and "safe" for dying children. (those with cerebral infections)
I know the MMV methods firsthand, and while commendable for their bringing the "business" mindset to the malaria problem, so far the program has not been the savior it had hoped to be.
But keep trying, and maybe one of the favored groups will be "successful"
Permalink to Comment