Sylvia Davidson

Research suggests importance of haplotypes over SNPs

Orchid BioSciences (Princeton, NJ) points out that “in some cases a single SNP will be responsible for a phenotype, whilst in other cases it will be multiple SNPs, either linked [i.e., grouped in a haplotype] or independent [perhaps on different chromosomes].” Mark Edwards chief scientific officer of Oxagen (Abingdon, UK) agrees. “It is dangerous to generalize [about SNPs and haplotypes],” particularly in the area of diseaseassociated genes because just not enough is known yet about these genes. Oxagen is working on the identification of genes associated with type 2 diabetes, inflammatory bowel disease, psoriasis, and asthma. However, the commercial reality is clear: if a single SNP were found to be the basis for variation in drug response or disease risk, the development of a diagnostic product would be comparatively straightforward because it would require a single assay only. Whereas if a combination of several SNPs was found to correlate with phenotype, development of a diagnostic would be more complex and costly, potentially requiring an assay for each SNP involved. As a result, diagnostics based on single SNP assays are expected to reach the market within the next year or two: One of these comes from Gemini Genomics (Cambridge, UK), a specialist in gene discovery through twin and population genomics. It licensed a SNP in the collagen alpha1 gene to the BritishNorwegian diagnostics company AxisShield (Dundee, UK) in April 1999 for production of an osteoporosis predisposition test. Gemini’s CSO Patrick Kleyn says the SNP is predictive of a much higher chance of fractures in women with osteoporosis. However, cheap efficient haplotype tests are some way off. “Instruments and kits are not yet at the right level to make it cheap and quick to haplotype people,” says Ruano. “The commercial pressure is now on [the instrumentation companies] to develop a test that reduces the cost and increases the specificity,” of haplotype tests. Difficulties lie in haplotype determination in the first place. Direct determination of haplotype (molecular haplotyping) is only possible by sequencing the two homologous chromosomes independently to reveal which SNPs appear on the same piece of DNA—a laborious process that would be very expensive to perform on a large group of people. Alternatively, algorithms can be used to infer haplotypes based on analysis of SNPs in patients and information on haplotype frewere identified by sequencing and classified into two functionally related groups. Five haplotypes were found significantly more frequently in those people with substance dependence than in controls. Genaissance CEO Gualberto Ruano believes that the haplotype approach may be essential in recognizing genetic factors that create a tendency towards a particular disease or pharmacogenetic phenotype. For example, there may be three SNPs in a gene (including regulatory region, and all coding

Biogen, Serono squabble over MS market.

and that Biogen’s clear intent is to “damage Serono through harassment, defamation and manipulation of the media.” The civil case began on July 11. EVIDENCE data shows that at 24 weeks, only 25.1% of patients treated with Rebif had experienced a relapse during six months of treatment compared with 36.7% of Avonex patients. Serono expressed this as an adjusted odds ratio of 1.9, or that patients treated with Rebif had 90% greater odds of remaining-relapse free relative to patients treated with Avonex. Rebif also achieved high statistical significance for the secondary endpoints of the study, including average number of relapses during the six months of treatment, relative risk of experiencing a relapse, and need for steroids to treat relapses. “The company set themselves a very high hurdle to show efficacy at six months...but these guys did not just climb the hurdle, they pole-vaulted over it,” says Steve McGarry, analyst at Goldman Sachs (London). Biogen, whose stock dropped 9% the first day of trading after the presentation, has criticized the trial on several points: the short duration of the trial, use of time to relapse rather than disability progression as the primary endpoint, and that the patients and treating physicians were not blinded. Biogen’s Nancy Simonian, senior medical director of research, says that as MS is a chronic disorder, trials should last at least two years as “short-term changes do not necessarily reflect what is happening in the long-term.” Simonian also objects to the “unconventional way” Serono presented its data, referring in part to use of the 90% odds ratio which she claims makes the difference between Rebif and Avonex look greater than it actually is. Avonex accounted for 82% (

British biotech redefines itself.

761 million) of Biogen’s revenue in 2000; clearly, keeping hold of orphan status is important for the company (see box). Samir Shah, vice president corporate marketing neurology at Serono, points out that the trial design and endpoints had been “preagreed with the FDA.” He adds that the nature of the different treatment regimens for the two products made it impossible to blind the patients and the treating physicians, however “precautions were taken to avoid bias” as the evaluating physicians were blinded, for example. And as for the duration of the trial, Shah says that four year data is also now available from the PRISMS study (Neurology 56, 1628, 2001), which shows that patients receiving high dose Rebif for four years had less impairment and disability than those receiving low-dose Rebif or placebo. Two-

Financing bonanza set to continue.

tion often has the upper hand. “If NIH continues to do business as usual, it runs the risk of losing a major opportunity,” he says. This warning is not exaggerated; it’s very real. We’re talking of how to support research to make new tools.” But top NIH officials say that traditional review regimes are already being revamped to ensure that nanotechnology projects are funded. NIH Deputy Director for Extramural Research Wendy Baldwin says that some of these changes already are being put in place. “We’ve explicitly changed the review criteria so that it’s not only hypothesis-driven but also design-driven,” she says. Moreover, the entire peer review system at NIH is in the throes of a multi-year retooling effort, one that entails changes in staff, reviewer attitudes, and in the overall “culture.” That kind of “social change takes a long time,” she says. Jeffrey L. Fox

Incyte SNPs up Hexagen for new firm

The number of biotechnology initial public offerings (IPOs) actually grew during the traditionally quiet months of July and August. The post-IPO performance of stock that floated earlier in the year has been good both in the US and in Europe, and the trend looks set to continue for the recent spate of offerings. Analysts agree that this is likely to fuel further IPO activity, and don’t expect enthusiasm in the US to diminish before 2001, despite the high number of financings. However, they predict the relative immaturity of the markets to result in less stable post-IPO performance of companies listing in Europe. Between late June and early August, 20 biotech companies floated on Nasdaq (including dual listings), compared with 20 in the previous five months; meanwhile, European exchanges saw 11 IPOs, compared with the first 5 months of the year. The share prices of most of the companies that have floated recently in the US have held up well, with prices in early August still significantly above initial offers (see table). Many stocks rose over 20% in their first week of trading while some performed even better. For instance, Applied Molecular Evolution (San Diego, CA) and Ilumina (San Diego, CA) rose almost 40% and over 170% respectively in their first week. Although stocks of companies that floated earlier in the year plummeted in the aftermath of the Clinton/Blair statement in March (Nat. Biotechnol, 8, 365), they had in the main recovered by August. For instance, share price of Diversa (San Diego, CA) closed at

Hidden biotechnology worth over

34.00 on August 8, up 41% from its initial offer price on February 2, but almost 200% below the highs seen for the stock in early March. “In general, the IPOs [in the US] are meeting expectations,” says Steven Burrill, CEO of investment bank Burrill & Co. Traditionally, however, biotechnology booms have come to an end as supply outstrips demand. But “investor appetite hasn’t been filled yet,” says Burrill. As a result, “The queue that’s lining up [to join Nasdaq] for September and October is awesome.” There are at least 14 companies planning to go public in late summer, including Ciphergen Biosystems (Fremont, CA), Introgen Therapeutics (Austin, TX), and DoubleTwist.com (Oakland, CA). Amersham Pharmacia Biotech (Piscatawy, NJ) is among the 21 companies currently planning to float on Nasdaq before the end of the year. Although Burrill does note that stock of companies that went public in early 2000 might start to feel some “downward pressure” in August as people who bought into IPOs are allowed to sell their shares, he thinks it will be 2001 before companies start performing below investors expectations. As for Europe, says Steve McGarry, analyst at Goldman Sachs, “in general the stock of new IPOs is doing well.” For instance, TranXenoGen’s (Worcestor, MA) stock had risen 26.5% to £2.62 by August 8 after going public on July 18. And this is the trend for all the IPOs so far in 2000, according to Genghis Lloyd-Harris, analyst at Credit Suisse First Boston. For instance, a week after Weston Medical’s IPO on May 3, its stock price was up 8.8% to £1.85, and by August 8 it had jumped a further 39.7% to £2.52. However, analysts are more negative about the prospects of post-IPO performance in Europe than in the US over the coming months. One distinguishing factor is that several European markets have seen the first listing of a biotech company: For instance, BioTie Therapies (Turku, Finland) listed on the Helsinki exchange in Finland in mid-July and Biosearch Italia (Gerenzano, Italy) went public on the Nuovo Mercato in late July. Although Biosearch Italia’s share price had only risen 23.2% on the offer price by August 8 and BioTie Therapies hovered around its original offer, in the Swiss market, where Actelion (Allschwil) and Modex Therapeutics (Lausanne) are among the first biotechnology companies listed, both had seen spectacular rises in stock price since their IPOs, jumping 179% to CHF725 and 79% to CHF300 since their IPOs in May and June, respectively. McGarry explains that when a sector first starts to appear on an exchange, the novelty tends to result in overvaluations of those companies, which subsequently fall as more companies in that sector list. He cites the behavior of biotechnology stocks on Germany’s Neuer Markt, which fell more than 20% in the first week of August. “Now that there are six to seven companies on the German market, valuations start to go down,” says McGarry. From now on, he adds, Financing bonanza set to continue