Drug Candidates Unveiled

In accordance with what has rapidly become a tradition at the American Chemical Society's spring national meeting, pharmaceutical researchers gathered in New Orleans this month to disclose the chemical structures of potential new medicines for the first time.

The five speakers at the Division of Medicinal Chemistry's "First-Time Disclosure of Clinical Candidates" symposium, which was supported by Wyeth Pharmaceuticals, described drug candidates for prostate cancer, diabetes, cardiovascular disease, and hepatitis C. Symposium organizer Andrew W. Stamford, director of medicinal chemical research at Schering-Plough Research Institute, in Kenilworth, N.J., characterized this year's candidates as a strong group targeting diverse therapy areas. "These are compounds that have made it into the clinic" and continue to wend their way through clinical trials, he told C&EN. "That's not easy to do."

Leading off the symposium was John A. McCauley, senior research fellow in medicinal chemistry at Merck Research Laboratories, in West Point, Pa. He told attendees about Merck's hepatitis C drug candidate, MK-7009.

More liver transplantations are performed due to hepatitis C, an infection caused by a single-stranded RNA virus called the hepatitis C virus (HCV), than any other liver malady. The rate of disease progression is highly variable; most patients remain symptom-free for an extended period. Many patients eventually progress to symptomatic hepatitis and from there to fibrosis, cirrhosis, liver cancer, and death.

Today, doctors treat hepatitis C with a combination of ribavirin and polyethylene glycol-derivatized (PEGylated) interferon-α. However, this therapy has side effects, such as anemia and depression. In addition, it's effective only in about half of patients with the most common form of the virus. Such shortfalls underscore the need for additional treatment options.

At last year's symposium, GlaxoSmithKline (GSK) researchers discussed their HCV polymerase inhibitor, GSK625433 (C&EN, May 7, 2007, page 56), which is now in Phase I clinical trials. The Merck team focused on a different target—the HCV NS3/4A serine protease. This enzyme is essential for viral replication because it helps to cleave the long peptide precursor that is translated from the viral genome. This process generates proteins that are required to make more viruses. Because combination therapies that include a protease inhibitor have been successful in treating the human immunodeficiency virus, researchers are hopeful that a similar approach could derail HCV as well.

The Merck team recently published their MK-7009 discovery strategy (J. Am. Chem. Soc. 2008, 130, 4607). They started by examining published structures showing an inhibitor bound to a portion of the HCV NS3/4A protease. To learn more about how the full-length protein may affect inhibitor binding, the researchers used molecular modeling to develop a picture of this inhibitor bound to a crystal structure of the full-length protease. This picture proved to be a valuable tool in the design of a new series of protease inhibitors.

To make such inhibitors, the team developed an efficient synthesis that featured a ring-closing metathesis step. The clinical candidate's structure also is amenable to other cyclization strategies, which may prove useful for large-scale production, McCauley said. "You could envision using reactions such as a Suzuki coupling or a Sonogashira reaction," he added. The result of the team's drug design strategy was MK-7009, a potent and selective inhibitor of HCV NS3/4A protease. The team determined that the compound efficiently reached its site of action, liver tissue, in preclinical animal studies. MK-7009 is administered orally and is currently in Phase I clinical trials.

Following McCauley's talk, Mark E. Salvati, associate director of discovery oncology chemistry at Bristol-Myers Squibb (BMS), in Princeton, N.J., disclosed the structure of BMS-641988, a drug candidate for the treatment of advanced prostate cancer. BMS-641988 is an antagonist (blocking agent) associated with the androgen receptor (AR), which modulates gene expression in response to those male hormones that promote the growth of prostate tumors. Salvati suggested that BMS-641988 may give new hope to patients with advanced prostate cancer because it emerged from research that answered some questions about why established prostate cancer treatments ultimately fail.

Currently, patients with prostate cancer may be given chemical or surgical castration treatment along with an AR antagonist, a therapy referred to as androgen ablation. This therapy seeks to deprive the body of androgens and yields about 18 months of disease remission. However, nearly all of these patients experience a relapse where androgen ablation therapy no longer staves off the cancer.

The latest research in this area has found that such relapses are directly related to reactivation of the AR signaling pathway in tumor cells. Through pathways that reactivate AR, prostate cancer cells find a way to grow despite the reduced androgen levels created by androgen ablation therapy. Therefore, Salvati noted, using highly potent androgen receptor antagonists that more completely block the AR signaling pathway might be a better treatment option for advanced prostate cancer than the currently available therapies.

The BMS team applied structure-based drug design and medicinal chemistry approaches to develop BMS-641988. They began by analyzing how a known AR blocker interacted with its target and made modifications to increase potency and generate new antagonists that bind differently to the AR compared with existing antagonists. Such antagonists may circumvent many of the pathways leading to androgen ablation relapse, Salvati told C&EN. The efforts of his team culminated in BMS-641988, a highly potent and selective AR antagonist. In preclinical tests, BMS-641988 effectively blocked the growth of an in vivo human prostate tumor model made resistant to continued bicalutamide treatment. Bicalutamide is the gold standard AR antagonist. In addition, the team showed at the genomic and proteomic levels that BMS-641988 better blocked the AR signaling pathway compared with bicalutamide. BMS-641988 is given to patients orally and is currently in Phase I clinical trials for the treatment of prostate cancer.

The third speaker was Bradley R. Teegarden, associate director of medicinal chemistry at Arena Pharmaceuticals in San Diego. He described his team's discovery of APD791, a drug candidate that interferes with blood clot formation.

Clotting is essential to prevent excessive blood loss after injury. But when blood vessels are obstructed and blood flow is restricted, clotting can also lead to heart attack or stroke. Many patients receive clot-reducing therapy to lower the risk of strokes or heart attacks. Such treatments include aspirin and clopidogrel, which is marketed under the brand name Plavix by BMS and Sanofi-Aventis. But these drugs prevent clotting so well that they can lead to excessive bleeding after an injury, Teegarden said.

Clots are an aggregate of specialized cells called platelets that circulate in the blood. APD791 aims to block clots by impeding the signaling of ₅-HT_2A, a G-protein-coupled receptor on platelets that binds 5-hydroxytryptamine, also known as serotonin.

The Arena team chose to target this serotonin receptor because it is more of a peripheral player in clotting. "Serotonin doesn't directly work on platelet aggregation," Teegarden said. Rather, the neurotransmitter amplifies other factors that activate platelet aggregation. So lowering the serotonin receptor's signal might leave behind enough baseline clotting to avoid a bleeding risk, he noted.

Arena sought to develop a drug that was selective for ₅-HT_2A over other serotonin receptors and that was soluble enough to be efficiently carried in blood plasma. In a key advance, the team incorporated cyclic nitrogen-containing moieties into a lipophilic ₅-HT_2A inhibitor to improve its solubility. Further tweaks to this scaffold yielded APD791, which is selective for ₅-HT_2A and showed good clot-reducing activity in preclinical tests in dogs and monkeys. APD791 is given to patients orally and is currently in Phase Ib clinical trials, which are part of a series of trials to assess safety.

The final two speakers at the symposium discussed approaches to tackling diabetes. One team is targeting glucose production, whereas the other hopes to control receptors involved in carbohydrate and lipid metabolism.

At Metabasis Therapeutics in San Diego, director of medicinal chemistry Qun (Max) Dang and his colleagues developed MB07803, a second-generation fructose-1,6-bisphosphatase (FBPase) inhibitor for the treatment of type 2 diabetes.
FBPase is an enzyme in the liver and kidney that controls glucose biosynthesis in the body.

The liver tissue in type 2 diabetics responds poorly to the hormone insulin. As a result, the liver makes excess glucose, leading to high blood glucose levels that in the long term can cause blindness, kidney disease, and cardiovascular disease.

Inhibiting FBPase decreases glucose production and would thereby lower blood glucose levels in type 2 diabetics. Metabasis focused on developing mimics of adenosine 5′-monophosphate (AMP), which is a natural allosteric inhibitor of FBPase.

At the 2005 ACS spring national meeting, Metabasis described its first-generation candidate, MB06322 (CS-917), which the company had advanced into clinical trials with Tokyo-based corporate partner Daiichi Sankyo. But development of CS-917 has since been discontinued in favor of the second-generation candidate that Dang discussed in New Orleans.

The Metabasis team had two priorities when developing MB07803—improving upon CS-917's oral bioavailability (ability to be absorbed) and eliminating the formation of a CS-917 metabolite that is generated by acetylation of the amino group. This metabolite is inactive because it cannot bind to FBPase. Moreover, the metabolite's formation may contribute to some other limitations the team observed with CS-917, Dang said.

The team's medicinal chemistry breakthrough came when it installed a bulky electron-withdrawing group on the heterocycle where the amino group is attached. This modification reduced N-acetylation by conjugation effects. Then, to improve bioavailability, the team fine-tuned the moiety that masked CS-917's polar phosphonic acid group. The mask is only temporary; the compound rapidly converts in vivo to the phosphonic acid, which is the active agent. These modifications led to MB07803.

MB07803 lowered glucose levels in animal tests, and in Phase I clinical trials orally administered MB07803 was found to be safe in healthy patients, with no detectable N-acetylation in the projected therapeutic dose range. The drug is now in Phase IIa trials, which are part of a series of trials to test efficacy.

Novartis Pharmaceuticals is also working on a type 2 diabetes therapy. At the meeting, T. R. Vedananda, a program head in the department of diabetes and metabolism at Novartis, in Cambridge, Mass., talked about how his team developed cevoglitazar, which activates receptors that orchestrate food breakdown and energy production in cells.

Cevoglitazar targets peroxisome proliferator-activated receptors (PPARs), which regulate many genes involved in fat and sugar metabolism. One particular PPAR subtype, PPAR-γ, is already the target of established diabetes medications, such as GSK's rosiglitazone, marketed under the trade name Avandia. Avandia lowers blood glucose levels by increasing cells' responsiveness to insulin. Last year, FDA issued a safety alert for Avandia due to data indicating the drug increases risk of sometimes-lethal heart-related side effects (C&EN, May 28, 2007, page 8, and June 11, 2007, page 24). PPAR activators like cevoglitazar appear to remain promising enough as a class that Novartis has continued to pursue these targets.

Cevoglitazar activates PPAR-γ, as well as another subtype, PPAR-α. PPAR-α is thought to be the target of lipid-lowering drugs called fibrates. According to Vedananda, activating both PPAR family members with one drug could lower glucose and lipid levels at the same time—a strategy that might benefit an overweight type 2 diabetic patient.

The Novartis team's discovery approach combined database screens with structure-based drug design. They began by examining published X-ray structures of PPAR-γ with known ligands already bound to it. Next, they pared that extensive binding data down to three key hydrogen bonds and a phenyl ring that they deemed necessary for good receptor activity and binding. Using a molecular modeling program, the team searched three-dimensional databases containing nearly 1.3 million structures for molecules that could make those key hydrogen bonds. Transforming virtual hits from that initial computer search into functioning drugs required medicinal chemistry expertise, Vedananda emphasized. The Novartis team used structure-based drug design to optimize their best hits and ultimately made cevoglitazar, which was effective at lowering lipid and glucose levels in several animal models. Cevoglitazar is delivered orally and has completed Phase I clinical trials.

Asked to comment on the importance of the annual first-disclosures sessions, Metabasis' Dang noted that it provides a useful opportunity for pharmaceutical researchers to share how they have solved specific medicinal chemistry problems.

The first-disclosures symposia highlight the central role that medicinal and synthetic organic chemists play in the drug development process. "There aren't many forums where chemists can showcase their accomplishments," BMS's Salvati said.

Merck's McCauley agreed. As a drug becomes more successful, people typically hear more about the drug itself and less about the arduous process that went into its discovery, he said. "This symposium is the top place for medicinal chemists to tell their story."