Fragments in the Clinic: AT9283

Returning to the discussion of fragment-derived compounds that have made it into the clinic, researchers at Astex have published a nice account of the discovery of AT9283, an Aurora kinase inhibitor, in the latest issue of J. Med. Chem. The compound is in phase I testing for the treatment of solid tumors and in phase I/II testing for hematological malignancies.

While pursuing CDK inhibitors, an effort that yielded the clinical compound AT7519 (as highlighted in August last year), Astex researchers discovered that some of their pyrazole-benzimidazoles were potent and highly ligand-efficient Aurora A inhibitors. This illustrates that Nobel laureate James Black’s famous dictum, “the most fruitful basis of the discovery of a new drug is to start with an old drug,” applies to fragments as well – especially when going after kinases.



Crystallography revealed the binding modes of Compounds 5 and 7 (above) to Aurora A, and structure based design suggested adding a morpholine group to improve potency (as well as solubility). This did improve cell potency, but the resulting molecules exhibited very high plasma protein binding. A further series of structure-guided SAR studies succeeded in replacing the phenyl amide with a cyclopropyl urea, resulting in the highly potent and less lipophilic AT9283. This molecule inhibits both Aurora A and B and shows low nanomolar cellular activity consistent with inhibition of Aurora B. It also shows a clean CYP profile, good solubility, and exhibits significant tumor growth inhibition in mouse xenograft models.

Perhaps unsurprisingly given the molecule’s origins, AT9283 hits a number of other kinases too. Some of these, such as JAK2, Flt-3, and the Abl T315I mutant, are attractive cancer targets in their own right. Indeed, the fact that the molecule binds exclusively in the ATP-binding pocket allows it to inhibit kinases, such as Abl(T315I), that are resistant to many adaptive-pocket binding inhibitors such as imatinib. However, the molecule also hits more than 20 other kinases with similar potency, which could lead to off-target side effects. That said, specificity may not be everything it was once thought to be: sales of sunitinib, probably the most non-selective of approved kinase inhibitors, were $627 million for the first nine months of 2008. There is a raging debate in the kinase field over the importance of specificity. It is a debate that only more data will resolve, and AT9283 represents an attractive data point.

The Big Get Bigger

In case you missed it, Pfizer, the 800 pound Gorilla of Big Pharma, just ate Wyeth. I can tell you in the little town of Collegeville, where Wyeth has a site, there is immediate concern, not just about friends and neighbors, but about the town in general. (Although, GSK also has a site here, but still 8000 jobs are probably gone.)

What does it mean for FBDD? Wyeth has an active group in Boston and Pfizer has one in San Diego. I think there are also groups scattered at other sites (Groton ?).

Here are the key 'grafs:

The deal came as New York-based Pfizer set out a full house of issues: a 90 percent drop in income, a hefty charge to end an investigation, a severe cut in its dividend, a shockingly low profit forecast for 2009 and 8,000 job cuts starting immediately.

That's all on top of the colossal problem triggering this deal: the expected loss of $13 billion a year in revenue for cholesterol fighter Lipitor starting in November 2011, when it gets generic competition.

Pfizer also plans by 2011 to cut about 8,190 jobs, 10 percent of its workforce, as part of what it expects will be a staff reduction totaling 15 percent of the combined companies' workers -- implying a total job loss of almost 20,000.

That's a lot of jobs lost, a lot of them will be R&D. I can't imagine that Pfizer will want more than one FBDD group when many companies have zero. I think we will see many colleagues out of a job, fragment and non-fragment.


There are >48,000 hits for Pfizer wyeth merger as of 9pm EST. There is lots of excellent analysis. I wpn't try to duplicate the efforts of others much better at that than I.

Golden discoveries or numerology?

Masaya Orita and colleagues from Astellas Pharmaceuticals have published a thought-provoking paper in Drug Discovery Today (in press). In it, they describe two new measurements based on the golden ratio.

As mathematicians, art historians, and readers of The Da Vinci Code know, the golden ratio, or phi, is an irrational number whose first ten digits are 1.618033988. Phi describes the relationship between two numbers, such as 6765 and 4181, in which the ratio of the sum of the numbers to the larger number is equal to the ratio of the larger number to the smaller number. It pops up in many unexpected places, though, like Elvis, many of these sightings are disputed. Now it may have made (two!) appearances in the world of fragment-based drug discovery.

The authors examined 30 examples of fragment-based ligand discovery in which the final compound had an affinity better than 100 nM and a MW less than 600 while the starting fragment had an affinity greater than 1 micromolar. They found that the average number of heavy (non-hydrogen) atoms of the final compound was 28.933, the average number of non-hydrogen atoms of the fragment was 17.833, and thus 11.1 heavy atoms were grown or added to the fragment during optimization. 28.933 / 17.833 is approximately equal to 17.833 / 11.1, which is approximately equal to phi, the golden ratio.

The authors suggest that, if a protein target has known inhibitors with N heavy atoms, a fragment library might be more likely to produce hits if it contains compounds that have N/phi heavy atoms. I’m not sure this is the best strategy. It seems that, regardless of the target, one will want to keep the final molecular weight low, and thus a “Rule of 3” approach is probably the best bet (which, as the authors note, is related by phi to the “Rule of 5”). That said, perhaps it is worth screening larger fragments for particularly intractable targets such as protein-protein interactions, which seem to require larger ligands.

The second observation of phi is based on a reanalysis of Kuntz’s seminal “Maximal affinity of ligands”, which includes binding data for more than 150 ligand-receptor interactions. After removing heavy metals and other non-drug like ligands, and plotting ligand efficiency vs heavy atoms for the strongest-binding ligands, Orita and colleagues found that, as the number of heavy atoms doubled, the maximal ligand efficiency decreased by a factor of phi. From this they derived a new measurement:

%LE = (LE / maxLE)*100
Where maxLE = phi^log2(10/HA)

This measurement is intended to give a sense of how closely any ligand with a certain number of heavy atoms approaches the maximum ligand efficiency achievable for a ligand with the same number of heavy atoms.

The paper is a fun read (don’t be put off by the equations!), but will the observations of phi hold up to further scrutiny? And will the new indices be useful? The authors are appropriately circumspect:

Why does the Golden Ratio appear in FBDD? This might be an artefact caused by human minds (medicinal chemists), to whom such a ratio is attractive. It is expected that arguments about the existence and usefulness of the Golden Ratio in the field of drug discovery will be advanced in future.

What do you think? Are these demonstrations of patterns in medicinal chemistry, or of pattern-finding instincts in medicinal chemists?

Ligand efficiency for antibiotics

Back in October of last year we highlighted a paper in Science that disclosed a new antibiotic targeting the bacterial protein FtsZ. The compound was derived through fragment-based techniques, though at the time no details were provided. A new paper in BMCL now provides some of the early medicinal chemistry, and also introduces an interesting new tool for evaluating antibiotics.

As mentioned in the Science paper, the researchers (led by Prolysis but with a number of contributors from Evotec and Key Organics) started with the fragment-like (MW = 151, 11 heavy atoms) 3-methoxybenzamide. An initial survey of “SAR by catalog” soon moved to the synthesis of analogs that could be assembled in up to four steps from commercially available compounds. This study found that the amide was essential, and only limited substitutions around the aromatic ring were tolerated. Turning to the alkoxy group, the authors took the classic “methyl, ethyl, butyl” approach, but kept going all the way to dodecyl. Intriguingly, a nonyloxy substituent proved to be optimal, better than either 8 or 10 carbon chains. Adding two fluorine atoms to the aromatic ring improved the potency further. Although the paper does not describe the final push to PC190723, the authors do describe the desire to replace the long alkyl chain and its likely attendant problems.



The paper also defines an interesting variation of ligand efficiency:

Antibacterial efficiency = -ln (MIC) / N, where
MIC = minimum inhibitory concentration (mg/ml) and
N = non-hydrogen atoms

Although the metric has a few quirks (for example, low-efficiency compounds can actually have negative numbers), “good” values correspond roughly to good LE values; clinically approved low molecular weight antibiotics have antibacterial efficiencies in the 0.26-0.32 mg/ml/atom range.

So for all you folks working on antibiotics, not only are fragments a viable starting point, you now have a new way to evaluate progress.

Fragments in the Clinic: Indeglitazar

Following up on the discussion of fragment-derived compounds that have made it into the clinic, the first 2009 issue of PNAS describes the discovery of indeglitazar, which I believe is the USAN name of Plexxikon’s PLX-204/PPM-204. Indeglitazar is a pan-agonist of the peroxisome proliferator-activated receptors (PPARs), and has been in clinical trials for treatment of type 2 diabetes.

Plexxikon’s version of fragment-based screening, “scaffold-based discovery,” entails screening several thousand small to medium sized fragments (150-350 Da) in a biochemical assay, followed by crystallographic analysis of active molecules. In the current case, the researchers screened their collection against PPAR alpha, gamma, and delta, looking for molecules that activated two or more. After the primary screen, 170 molecules were characterized crystallographically, and about a quarter produced at least one structure. The substituted indole fragment (below) showed very weak activity, but bound snugly in a large pocket with its NH positioned toward a second pocket. Structure-guided design led to the more potent phenyl sulfonamide shown in the middle of the figure, and synthesis of just 20 additional compounds resulted in indeglitazar, which activates PPARs alpha, gamma, and delta. Ligand efficiency remained fairly constant throughout optimization.



Indeglitazar is a full agonist of PPAR alpha but only a partial agonist of PPAR gamma and delta; this may provide a better side effect profile than full activators. The molecule shows impressive pharmaceutical properties (high oral bioavailability, long half-life, etc.) as well as promising activity in mouse and rat models of diabetes (lower blood glucose, insulin, total cholesterol, triglycerides, free fatty acids, etc.). In contrast to other PPAR agonists, which sometimes cause weight gain, indeglitazar also caused weight loss in rodent and primate models; the authors suggest this could be because it affects all three PPARs.

Although indeglitazar was advanced to phase 2 trials in collaboration with Wyeth, increasing concerns over the potential side effects of PPAR agonists have caused Wyeth to discontinue development of this compound for diabetes, and as of November of 2008 the molecule was available for licensing.

Nonetheless, this is an impressive story, and appears to be the first example of using fragment-based methods to discover an agonist, as opposed to an inhibitor.

Fragments in the Clinic

Jeff Albert posted an interesting topic in the LinkedIn FBDD group's discussion section.
What is the current status of FBDD based drugs in the clinic.

Justin Bower from AZ posted:

ABT 263 (Phase II Bcl-2/Bcl-xl inhibitor) Abbott
ABT 869 (VEGF & PDGFR Phase I) Abbott
SGX-523 (Met Phase I) SGX (No Longer in Clinic, per comments)
SGX-393 (Bcr-Abl) SGX
AT-7519 (CDK1, CDK2 Phase I) Astex
VER-52296 or NVP-AUY-922 (hsp90, Phase I/II) Vernalis plus another in Phase I

Gianni Chessari, from Astex, noted this list (from a review coming out soon-ish?), with repeats removed:

AT9283 Astex Aurora Phase 2
LY-517717 Lilly/Protherics FXa Phase 2
PLX-204 Plexxikon PPAR agonist Phase 2

ABT-518 Abbott MMP-2 & 9 Phase 1
AT13387 Astex HSP90 Phase 1
IC-776 Lilly/ICOS LFA-1 Phase 1
PLX-4032 Plexxikon B-RafV600E Phase 1
PLX-5568 Plexxikon Kinase Inhibitor Phase 1
SNS-314 Sunesis Aurora Phase 1
LP-261 Locus Tubulin Phase 1
DG051 deCODE LFA4H Phase 1

So, what's missing? I would also be curious as to how these projects progressed.

Update: I have removed duplicates from the two lists. One (SGX-523) has fallen out of the clinic (per comments).