Corrections or additions?
This article was published in U.S. 1 Newspaper on December 8,
1999. All rights reserved.
Balancing the Genetic Code
by Christopher Mario
Digits rule our lives. Transpose two numbers in a phone
number: sorry, wrong number. Enter $101 instead of $110 in your
rubber checks. Switch two letters in the HTML of your website: crash
Messing up your digits can lead to trivial and easily fixable problems
like those above, or they can be more serious, like the pesky "19"
lurking in the software of the world’s computers, two little digits
that have some people stocking up on creamed corn and diesel fuel
for a disaster that may or may not happen on January 1, 2000.
But there are another set of digits, far more important than any
You can’t see them. You know absolutely nothing about them. Yet you
have about three billion of them with you right now. Rather than
these digits are expressed as letters: A, C, T, and G, for the
adenine, cytosine, thymine, and guanine, known collectively as the
bases of DNA. These digits or bases are the building blocks of the
genes that make you you.
Now for the bad news: your digits may be out of whack. Switch a few
digits around in your genes, and you could have a much more serious
problem than a bounced check or a crashed computer: you could be
at the end. Digits rule our lives. They can rule our deaths too.
Or maybe not. Work now under way at Forrestal Center-based Orchid
Biocomputer may in the future provide medical science with a way not
only to find out if your genetic checkbook is in the red, but also
to bring it back into balance.
At the heart of Orchid’s efforts is something called SNPstream, an
automated, computer chip-like device that can identify the transposed
digits in genes — which scientists call single nucleotide
or SNPs, pronounced "snips" — quickly, accurately, and
cheaply. If all goes according to plan, these little snip chips may
hold the key to a cornucopia of new medical diagnoses and treatments
— from predicting predisposition to genetically based diseases,
to targeting existing drugs to the individuals for whom they will
be most effective, to identifying the genetic bases of diseases and
developing drugs to treat them.
SNPs are the most common form of variation in genes from individual
to individual. And that’s what makes them so important, according
to Dale Pfost (pronounced "Post"), chairman and CEO of Orchid
"Before about 1990, the notion would have been that there were
all sorts of diversities expressed on the genome, but nothing special
about single bases," Pfost says. "And it’s true that there
are other ways differences are expressed, such as variable repeat
lengths of individual bases, or insertion of deletions of whole hunks
of DNA. But the most common form of variation is in individual bases
flipping from one to another" — an A for a T, a T for a C
— "and any individual human has hundreds of thousands to the
low millions of these variations. And it is now believed quite broadly
that by setting SNPs very well very quickly, one will find many
to improve healthcare."
Pfost is not alone in that belief. In April of this
year, 10 large international pharmaceutical companies and the Wellcome
Trust, a British medical research foundation, in April of this year
established the SNP Consortium, a $45-million public-private
that in collaboration with the Human Genome Project has set a goal
of finding and mapping 300,000 common SNPs.
This blue-ribbon group thinks SNPs matter. A lot. According to the
consortium’s website, "variations in DNA sequence can have a major
impact on how humans respond to disease; environmental insults such
as bacteria, viruses, toxins, and chemicals; and drugs and other
This makes SNPs of great value for biomedical research and for
pharmaceutical products or medical diagnostics. Scientists believe
SNP maps will help them identify the multiple genes associated with
such complex diseases as cancer, diabetes, vascular disease, and some
forms of mental illness."
In other words, finding and understanding the function of SNPs may
very well lead to a revolution in healthcare. They may lead to a new
way of diagnosing and treating diseases, like diabetes, against which
current therapies are inadequate. And perhaps most intriguingly of
all, SNP technology may enable your doctor to predict conditions to
which your personal genetic makeup makes you susceptible, and to
treatments that have been shown to be most effective in patients who
share with you a particular SNP.
But there’s still a lot of work to be done. Nobody knows for sure
how valuable individual SNPs will be in treating and diagnosing
— the science of correlating SNPs and diseases is still in its
infancy, and it may be that most SNPs work in groups, rather than
individually, making the determination of their functions
more difficult. Nobody has yet identified all the 300,000 or so most
common SNPs that scientists figure to be the most likely suspects
in causing disease, rather than, say, determining the shape of your
earlobe. And crucially for Orchid Biocomputer and Dale Pfost, the
science and the market have yet to determine who among the handful
of competitors — including not just Orchid but also such Silicon
Valley companies as Incyte and Affymetrix and Maryland-based Celera
— will have the best SNP mapping technology.
Pfost thinks it will be his. Here’s how the technology developed.
In September, 1998, Orchid acquired a Baltimore company called
Tool and along with it Molecular Tool’s proprietary SNP technology,
known as genetic bit analysis (GBA).
"Molecular Tool developed their technology to look at SNPs in
1990," Pfost says of GBA. "The first application was to do
identity and paternity testing, and one of the first commercial
was to do paternity testing for the Jockey Club of America. Because
of the value of stud service and the importance of verification of
a foal’s parentage, it was a valuable technology. They then applied
it to human paternity and other forms of forensic testing."
But as Molecular Tool did it, the technology was slow and expensive
— far too much of both to be useful in broader medical efforts
using SNPs. And in that problem Pfost and his colleagues at Orchid
— one of whom knew two Molecular Tool board members and brought
the companies together — saw an opportunity.
Founded in 1995 as one of the first Sarnoff spinoffs, Orchid was an
early developer of automated microfluidics processors — miniature
chemistry labs that can manage and perform hundreds of chemical
simultaneously on a computer chip-like object made of glass, silicon,
and polymers that’s about the size of a credit card.
These so-called "ultra-high throughput" research
tools, which use pressure and electrical charges to guide tiny amounts
of often extremely expensive chemicals through channels on the chips,
were Orchid’s entire reason for being when first spun off from Sarnoff
with the additional backing of Anglo-American pharmaceuticals giant
Now, with the addition of Molecular Tool’s GBA technology, Orchid
has embarked on a related but entirely new voyage. While the company
will continue to refine its microfluidics technologies — in
right now is a multi-level chip that will accommodate 10,000 chemical
chambers — Orchid’s new additional focus combines the speed and
economy of the firm’s chips with the possibly revolutionary potential
of Molecular Tool’s SNP mapping technology.
Or, as Pfost puts it, the marriage of the two companies has united
hardware and software to create what Pfost believes will become
like what "Wintel" — Windows plus Intel — is to the
computer industry: an industry-dominating powerhouse. Just as Wintel
is to desktop computing, Orchid will be in the use of SNPs to develop
and improve diagnoses and treatments for unmet medical needs, Pfost
"It’s not as if our chip doesn’t have an intrinsic value in its
own right," Pfost says of his Wintel metaphor. "But what we
needed to have under our roof was proprietary chemistry to run in
it. We already did chemistry and biology on chips. We had that
in place. What we were looking for was something to go into the chip.
We found Molecular Tool. And now we are in the middle of what will
be one of the most exciting areas of biotechnology for the next 15
It’s probably not where Pfost would have expected to find himself
on the day in the early 1980s when, while a Ph.D. candidate in physics
at Brown, he visited a friend at Harvard Medical School. Walking
a Harvard lab that day, he happened to notice a 96-well plate being
used in a chemistry experiment. Pfost, who had been working in his
own lab on automated systems to inspect microchips using
to move the chips in and out of holders, thought that the two
— moving chips in and out of holders, moving chemicals in and
out of wells — were kind of similar.
Which led to a product called the Biomek, 96 tiny test tubes made
of injection-molded plastic in an object the size of a credit card,
and a company called Infinitek, which he founded in 1982 while still
in graduate school. The Biomek "became a galvanizing industry
standard for automation and experimentation," Pfost says today.
He sold the company to SmithKline in 1985, and stayed with it as a
SmithKline employee, building it into a $25 million business.
From there Pfost went to England, where he had been recruited to start
a business called Oxford Glycoscience, now a publicly traded company
in the UK that does automated drug discovery in complex carbohydrate
research. In 1996, he came to Princeton to help establish Orchid (U.S.
1, January 29, 1997).
And that he has done, shepherding the company through a $27.5 million
private placement, a $16 million lease financing package with a
capital group called Oxford Venture Finance, and a $15 million
with Motorola to enhance the functionality and manufacturability of
Orchid’s microfluidics chips. The company is now raising at least
$40 million in additional financing, and recently opened a new 32,000
square foot facility at 303 College Road that includes the company’s
"Genotyping Center of Excellence," a facility that will be
able to map or "score" millions of SNPs a day by the end of
2000. With nearly 80 employees, Pfost expects the company’s revenues
to be in the "teens of millions" this year, and anticipates
an IPO sometime next year, markets permitting.
But the crown jewel of Pfost’s tenure at Orchid is certainly
the product of the combined technologies of Orchid and Molecular Tool,
which right now can scan over 25,000 SNP genotypes per day. Yet that
too is just a beginning, because discovering SNP variations quickly
and accurately is merely the first step. Next for Orchid will be the
difficult and still uncertain task of figuring out what those
"Discovering is the beginning, but the fact is, most SNPs won’t
matter," Pfost explains. "Our task is to find the ones that
do, the ones that may account for the variable responses to drugs,
ranging from severe adverse drug response causing hospitalization
or death to complete lack of effect, which can be equally lethal.
Range of variability in response to medications is amongst the largest
unmet medical needs facing healthcare. And I believe we are in a
to help find the telltale signs in the form of SNPs that will enable
us to identify individuals susceptible to variable responses to
prior to their use."
The SNPstream technology may also provide opportunities to develop
new drugs targeted to specific conditions caused by SNPs, and to that
end, Orchid will provide its technology to pharmaceutical research
firms for use in their work, either by licensing the technology or
by providing the services in Orchid’s facilities. But for now, the
company’s primary commercialization strategy is focused on variable
drug response for a number of reasons.
First, because early research appears to indicate that many such
may indeed be traceable to SNPs. Second, because finding these SNPs
and identifying their roles in specific responses will provide a means
for the company to protect their findings with patents and thus lead
to revenues, most likely in the form of royalties from drug
But most important, variable drug response is a serious problem for
which SNP technology, if effective, would provide the only alternative
to what we have today — which is to take the drug and see what
"We are prioritizing our efforts to be able to make the most
benefits available," Pfost explains. "Adverse drug reactions
lead to 1 million hospitalizations and at least 120,000 deaths
And in terms of lack of efficacy of drugs, well, there’s really no
way to know, but the numbers are much larger."
A number of preliminary studies have indicated a direct link between
individual SNPs and drug response, Pfost says. In one, the
of a family of cardiovascular drugs known as beta blockers in patients
with severe congestive heart failure seemed to be linked directly
to one SNP in the patients’ genes. In another, SNPs appeared to be
important in the failure of about 30 percent of childhood leukemia
patients to respond to new treatments — treatments that are highly
effective for the other 70 percent of patients.
More studies will be needed, on these aand a whole host of other
But if Pfost is even close to right about the impact of SNPs on the
future of healthcare, Orchid’s future will be very bright indeed.
And for Pfost himself. His upcoming IPO, if successful, will quite
possibly make Pfost, his colleagues, and Orchid’s investors very,
very rich. For comparison’s sake, consider that two of Orchid’s
traded competitors, Incyte and Affymetrix, have market valuations
of $1 billion and $3 billion respectively. Do the math: you don’t
need to have all that large an ownership stake in a company of such
value to realize a very big payoff.
Which is all very nice, Pfost says, but it’s not what drives him.
"My primary motive is to see technology successfully employed
and improving lives," says the California native, who is 42,
and has a nine-year-old son. "The reason we all work 80 or 90
hour weeks is that what we’re doing does matter. It’s like in school,
you study because you enjoy learning, and the report card is just
a form of `attaboy’ and positive reinforcement. For me, money falls
into that category."
Box 2197, Princeton 08540-2197. Dale R. Pfost Ph.D, chairman and CEO.
609-750-2200; fax, 609-750-2250. Home page:
It seems like every self-respecting company —
lowtech, and in-between — has a web strategy these days. Orchid
Biocomputer is no exception. But Orchid won’t be selling books or
toys or auctioning off Pokemon cards. No, Orchid will be selling
you really can’t do without: your genes.
Well, not exactly. But at the soon-to-be-launched GeneShield.com,
you’ll find something that no other website can offer, we are sure:
a map of your very own SNPs.
GeneShield will provide direct-to-consumer genotyping on a
basis. With your subscription, you will get a quarterly advisory on
what has been learned lately about your SNPs and how they may
you to certain drug reactions.
Squeamish? Rest easy. No blood will be required. Just purchase your
sample prep box on the website or at the pharmacy. Inside you will
find a few forms to fill out, a Q-tip, and some very simple
swab the inside of your cheek, put the swab in the handy tube
and drop the whole shebang in the mailbox for the very short trip
to College Road.
A few weeks later you will receive your results, along with the
disk and password you will need to access your quarterly updates.
But don’t expect any medical advice: that’s between you and your
who can visit GeneShield’s physician support website for answers to
The cost? "Not expensive," says Dale Pfost. "Definitely
within the realm of an impulse purchase."
As for information about SNPs that may predispose you to a certain
diseases, well, GeneShield will be able to provide that too. But Pfost
has a question for you: do you really want it?
"We believe that SNP scoring and testing will become a standard
of medical practice, and we feel the best gatekeepers of that
would be the patients themselves," Pfost says. "But
People are uncomfortable with learning about that."
And anyway, Pfost questions how useful any knowledge of the actions
of individual SNPs will be in predicting disease predisposition, at
least in the near term.
"Not many diseases attributable to single SNPs are known, and
diseases tend to be complex: what one calls a disease is often a very
large family of forms of a disease," he notes. "The telltale
signs may be thousands in number, and some complex combination of
SNPs will probably determine relative risk. Whereas in case of a lack
of efficacy of drug, there are situations where a single solitary
SNP can make a decisive determination."
— Christopher Mario
Corrections or additions?
This page is published by PrincetonInfo.com
— the web site for U.S. 1 Newspaper in Princeton, New Jersey.