Back In The Game
Investors are a fickle bunch. it can take as little as one failure for Wall Street to write off a technology that only a few years earlier had been hyped as "the next big thing." Stock prices plummet, and funding becomes scarce for companies on the cutting edge of science. Such setbacks only serve to further delay a technology from fulfilling its promise.
But history has shown that it takes only one success in the biotechnology sector for the money to come back. After a lengthy period in the doghouse, oligonucleotide-based antisense drugs appear to be making strides toward getting back into the good graces of both investors and potential development partners.
It took several glancing blows and one heck of a sucker punch to fully shake the confidence of the investment community in oligonucleotide-based antisense therapeutics. A major hit came as early as 1998, when Gilead Sciences, a pioneer in the antisense space, abandoned the technology after its gene-code-blocking development pact with Glaxo Wellcome fell apart. Gilead subsequently sold its antisense patent estate, which had been more than a decade in the making, to Isis Pharmaceuticals.
That same year, antisense won its first and only approval to date: Isis' Vitravene, an oligonucleotide-based drug for the treatment of cytomegalovirus retinitis, an eye disease. The drug provided some validation of the technology, but because it is locally administered it didn't prove that oligonucleotides could work in indications where drugs need to have a systemic effect, such as cancer.
Despite that minor victory, wariness mounted as late-stage trials of systemically administered antisense drugs produced disappointing results. In 2003, Affinitak, developed by Isis and Eli Lilly & Co., did not meet its primary end point in a Phase III trial for lung cancer. Isis suffered another Phase III failure the following year with its Crohn's disease drug, Alicaforsen.
The coup de grace was what some consider the colossal failure of Genta's cancer drug, Genasense. In a highly publicized deal, Aventis agreed in 2002 to pay up to $480 million to codevelop Genasense, which was in late-stage studies for skin cancer. When a Food & Drug Administration advisory committee did not recommend approving the drug for melanoma in 2004, Aventis pulled out of the deal.
Yet those failures can largely be chalked up to a steep technology learning curve, industry watchers say. When it comes to oligonucleotide-based therapy, "it is not your father's Oldsmobile," explains Needham & Co. stock analyst Mark Monane, who follows most of the antisense companies. "These are new companies with new strategies and new targets. There have been huge advances, but there's a large myth that antisense doesn't work."
The initial products pushed through the clinic were based on first-generation antisense technology and had serious limitations, most notably in potency, delivery, and duration of action. But while companies were testing those early drugs in patients, they were making a parallel effort to improve the technology through chemical modification.
Drugs based on the results of that work, so-called second-generation antisense, are currently winding their way through the pipeline with what appear to be more compelling results.
The science behind antisense is both straightforward and complex. Many diseases are caused when proteins go awry, as in the abnormal production of enzymes, antibodies, or hormones. Traditional small-molecule drugs affect disease progression by attacking those aberrant proteins. In contrast, antisense technology targets the RNA that controls the production of the aberrant proteins.
During the translation phase of protein production, messenger RNA (mRNA) carries a message with the blueprints for building proteins from the DNA to the ribosome, the protein factory of the cell. An antisense drug tries to intercept that message by binding to and then degrading the mRNA, thus preventing those proteins from ever being made.
"Everyone knows the DNA goes to the RNA goes to the protein," Monane says. However, RNA has turned out to be quite an elusive target. Antisense technology has been helpful in terms of gene validation, but, he adds, "the question is, Can you make drugs with it?"
A range of technologies fall under the umbrella term "antisense," differentiated only by the approaches they use to break down the mRNA. Approaches generating interest include RNA interference (RNAi), micro-RNA, and the use of oligonucleotides to inhibit toll-like receptors.
Much of the enthusiasm among investors and big pharmaceutical companies has shifted to RNAi technology, which uses double-stranded oligonucleotides to intercept mRNA rather than single-stranded oligonucleotides, which make up first- and second-generation antisense drugs.
The technologies are more similar than they are different, notes C. Frank Bennett, senior vice president of research at Isis Pharmaceuticals. "At the end of the day, they're both catalytic mechanisms and antisense mechanisms, in that the antisense strand is what binds to the target and causes cleavage of the message," he says.
The main difference between the two approaches is that the mechanism of first- and second-generation oligonucleotide-based drugs is fairly well-characterized, whereas RNAi is still in its early days. In fact, Alnylam Pharmaceuticals, a leader in RNAi, only recently showed that short interfering RNAs could have a systemic effect, a feat achieved long ago by earlier generations of antisense molecules.
When trying to see through the fact and fiction of antisense, it is important to note that companies developing RNAi drugs "haven't had a chance to fail yet," cautions Kate Winkler, senior executive director for health care and life sciences research at Gold Crown Capital and a longtime antisense watcher.
Although RNAi may be considered the new heavyweight in the antisense arena, there is growing evidence that there is some fight left in second-generation oligonucleotides. A year ago, shares of Isis, considered to be the leader in the technology, hovered around $3.00. But in the past three months, Isis stock has almost tripled in value.
Similarly, AVI BioPharma, which boasts competing antisense technology, has seen its stock price double in that same time. And on a much smaller scale, even Genta's stock price has recovered modestly this year after languishing at its nadir following the Genasense disappointment.
Several factors are contributing to mounting interest in antisense companies: Late-stage data are starting to emerge for several second-generation oligonucleotide-based drugs, and some antisense companies find themselves working on "hot" products, such as treatments for bird flu or innovative cholesterol-lowering drugs.
Isis is by far the front-runner in the antisense space and could be the company to convince both investors and big pharma that the technology is worth another serious look.
"The bottom line is the evidence is very clear that the antisense approach is working for its target reduction both at an RNA and at a protein level and that those are associated with pharmacology outcomes," Winkler says.
Underpinning that success has been the steady evolution of antisense technology. When scientists started thinking about using oligonucleotides to intercept mRNA, they first created drugs with phosphodiester backbones, and they looked like DNA found in nature. Those drugs never made it to human trials because the structures proved to be too similar to ones found in the body, causing enzymes to recognize them swiftly and break them down.
The next step from that "zero-generation" phosphodiester-based technology came when scientists tweaked the molecular backbone by exchanging an oxygen for a sulfur. The resulting phosphorothioate backbone became the foundation for a handful of drugs that made it as far as Phase III trials, most notably Genasense from Genta and Affinitak from Isis and Eli Lilly. However, those first-generation antisense drugs suffered from fleeting half-lives and were not highly potent during their short time in the body.
To compensate for these deficits, patients tested with those early antisense drugs were subjected to the unsavory and less marketable dosing regimen of intravenous infusion once every few days. Easier to administer subcutaneous injections were found to cause an inflammatory reaction in the skin or lymph nodes.
The real breakthrough came in the mid-1990s, when Isis scientists, working in collaboration with Ciba-Geigy, uncovered one key to the limitations of first-generation antisense drugs: Phosphorothioate decreases the oligonucleotide's binding affinity for RNA, thereby compromising the potency of the drug as well as its duration of action.
Isis overcame the hurdles by modifying the ribose ring of the individual nucleotides. When the ring is ornamented with 2′-O-methoxyethyl (MOE) groups, second-generation antisense drugs become both DNA-like and RNA-like, thereby increasing their binding affinity.
Because these second-generation antisense drugs can be dosed weekly or monthly via infusion or subcutaneously, companies are able to explore their potential applications in a much broader range of diseases. They can even consider the possibility of introducing oral antisense formulations. "We can now go off into chronic diseases and diseases for which patients will tolerate once-a-month injections of the drug," Bennett says.
The evolution of the technology has shaped Isis' thinking about where antisense will prove most effective. For example, when first-generation drugs ruled, the Isis pipeline was focused on oncology, exploiting the ability of oligonucleotides to sensitize cancer cells to chemotherapy. Now, the company's lead second-generation products are aimed at cardiovascular disease and metabolic diseases, although with some attention still on cancer, Bennett says.
Two drug candidates are generating the most interest: the cholesterol-lowering compound ISIS 301012 and the diabetes drug ISIS 113715. Multiple Phase II trials are under way for both products, and early results from each show promise.
ISIS 301012 inhibits production of apoB-100, a protein involved in making and transporting low-density lipoprotein, or "bad" cholesterol. Isis is initially studying ISIS 301012 in patients with familial hypercholesterolemia, a small group of people who develop very high cholesterol in their late teens and early 20s. Isis is trying to position the drug as an add-on or alternative therapy for patients who are not adequately treated with statins, Bennett says.
The market for the product could be enormous because it will not compete with statins. Rather, the drug will likely be used in combination with statins, a therapeutic niche that has been underserved in the cholesterol market, says Needham & Co.'s Monane. Though combinations of drugs are routinely used to treat diseases like high blood pressure and cancer combination cholesterol therapies, he notes, have been "problematic."
Isis is currently developing the drug on its own but does not plan to bring it to market itself, Bennett notes. "I think they are probably going to partner 301012 after good Phase II data, when the drug will have good value built into it," Winkler says.
She believes there could also be intense partnering interest in the diabetes drug, should it fare well in Phase II trials. ISIS 113715 inhibits production of protein-tyrosine phosphate 1B, an enzyme that affects insulin's ability to regulate blood sugar levels. "It is an interesting target because no one can touch it with small molecules because it's a phosphate binding site," Winkler notes.
ISIS 113715 sensitizes cells to allow less insulin to be used to keep glucose at the right levels, Bennett notes. This means the drug has the potential to be used alone in patients in early stages of type 2 diabetes who have yet to start taking insulin. It could also help lower the dosage of insulin required by patients in more advanced stages.
Because Isis has a fairly tight grip on the intellectual property surrounding both first- and second-generation antisense technology, it also has a stake in several of the other later-stage antisense drugs in development elsewhere. The most advanced product, with the potential to beat both ISIS 301012 and ISIS 113715 to market, is a cancer treatment being developed by OncoGenex Technologies.
The drug, OGX-011, targets the protein clusterin, which diminishes the efficacy of chemotherapy. It is in four Phase II trials for a range of cancers, including breast, prostate, and lung.
Eli Lilly, which stuck with the technology even after the failure of Affinitak, is also pursuing several antisense drugs, both in conjunction with Isis and through a license for the Isis technology. Two second-generation cancer drugs are currently in Phase I trials.
Australia's Antisense Therapeutics has a broad licensing arrangement with Isis that is the basis for the three antisense drugs in its pipeline. The company's lead candidate, ATL 1102, inhibits production of CD49d, a subunit of an antigen that is believed to be involved in enabling white blood cells to enter sites of inflammation. The drug is currently in Phase IIa trials for multiple sclerosis.
With a long list of companies pursuing antisense drugs through technology from Isis, AVI Biopharma stands out as one of the few firms developing drugs with a different backbone structure. But because it developed its antisense technology, called Neugene, in-house, AVI has lagged behind some of its competitors in producing clinical candidates.
"AVI felt that the technology needed to improve on its pharmacology and applicability to human trials before it was going to have an impact in the clinic," says AVI's chief executive officer, Denis R. Burger. While other companies were pushing first-generation products through the clinic, "we went back to the chemistry lab," he adds.
The company believed that the next-generation technology needed to address several critical issues: stability, delivery, ease of manufacture, and safety. To tackle the stability problem, AVI added a nitrogen to the ribose ring to generate a six-membered morpholine ring. This modification, Burger says, enables the molecule to behave more as a peptide than as a classic nucleotide, which improves stability and facilitates polymerization.
The second modification AVI made was to neutralize the backbone of the oligonucleotide. First-generation antisense drugs consist of a string of 18 to 24 phosphorothioate subunits, each carrying a negative charge. "That's a huge load to drag across the cell membrane to get to a target," Burger notes. "Further, things that are that charged tend to stick to everything and therefore disappear from circulation rather quickly."
The end result is a phosphorodiamidate morpholino oligomer structure that AVI says is stable and highly specific. Like Isis' second-generation antisense technology, AVI's backbone is modified so that it is not found in nature. As a result, the body does not make enzymes that can recognize and degrade the drug.
AVI completed a Phase I trial of a hepatitis C drug based on the technology in late 2005 and expects to report results from a Phase II trial for the drug at the end of this month.
AVI's concept is compelling, but analysts caution that it has yet to be proven in the clinic; the results from that late-stage trial will thus be an important test. The company is also developing drugs to address a range of viruses, including ebola, Marburg, and avian flu, which are high-profile diseases. These targets could explain some of the investor interest in the company.
Though most antisense companies have refocused their pipelines on next-generation technology-be it second-generation antisense, RNAi, or another antisense approach-several players maintain that the early technology still holds therapeutic potential.
"I believe the overwhelming issue is not the chemistry or the backbone, but the selection of the target," says Raymond P. Warrell, president and CEO of Genta. "Antisense is nothing more than protein knockout strategy, and if you're going to be successful, the most important thing is what proteins you choose to knock out."
Genasense targets Bcl-2, a protein that keeps chemotherapy from killing cancer cells. Genta continues to pursue approval for the drug, despite the 2004 thumbs-down from FDA for the melanoma indication. In December, the company filed a New Drug Application seeking approval to use Genasense in combination with two chemotherapy drugs in patients with chronic lymphocytic leukemia. The drug is also being studied for lung cancer, prostate cancer, and acute myeloid leukemia.
Warrell says Genta will put a second first-generation antisense drug that targets the oncogene c-myb into the clinic in the first half of this year.
Lorus Therapeutics is also developing antisense drugs based on first-generation technology. The company's lead antisense candidate, G2040, targets a subunit of ribonucleotide reductase (RNR), an enzyme that is essential for DNA synthesis and cell proliferation. RNR is overexpressed, or highly active, in cancer cell lines, a state that enables cancer cells to resist chemotherapy agents, says Yoon Lee, director of research at Lorus.
By down-regulating the activity of RNR using G2040, cancer cells will theoretically become more sensitized to chemotherapy, Yoon explains. Lorus is looking at the efficacy of G2040 across a variety of tumor types in seven different Phase I or Phase II trials. It is hoping to partner the drug to help move it more swiftly into Phase III studies.
With so many oligonucleotide-based antisense drugs, both first- and second-generation, pushing to the mouth of the pipeline, companies and analysts agree that the individual winner of the race is not as important as simply getting a broad-indication drug across the finish line.
The pivotal point will come when the first drug is successful in the market, Warrell says. Once such a drug reaches the market, he adds, "the entire field will be galvanized."
Yet Gold Crown Capital's Winkler warns that despite the signs of revival, investors and big pharma partners are still reluctant to embrace antisense technology. She maintains that much of the recent enthusiasm over Isis has been generated because its lead candidate is a cardiovascular drug, not because of the underlying technology. "People still don't want to believe in antisense," she says.