79
that market vast databases of genes, searchable
by utility, all of which are economical. Huon de
Kermadec picked two sequences of genes: one
that produces a protein that can be cut to make
insulin, and one that makes the yeast resistant
to a specific antibiotic.
Using these genes, he and the team created
a formula for a plasmid, a tiny circular piece of
DNA. Then he hired a company to manufacture
a small quantity of these plasmids, which they
sent to Counter Culture Labs in a tiny plastic
vial. Dozens of companies offer this sort of ser-
vice in the U.S., at prices as low as a few hun-
dred dollars. Next, the team jammed the plas-
mids into yeast cells, and added an antibiotic
solution to the cell cultures. That’s where the
antibiotic-resistance gene comes in handy—
the ones that successfully adopt the plasmid
into their own DNA will survive, and those that
don’t will die out. The next step would be to
grow the surviving cells, which should contain
the genes to produce both antibiotic resistance
and the precursor protein for insulin.
The team thinks they’ve gotten this far.
They know the yeast produces a molecule the
size of insulin—which is a pretty good sign it is,
in fact, insulin. But they’re scientists, so they’re
not popping the champagne yet. They want to
confirm that the molecule is indeed insulin by
using mass spectrometry, a precise technique
that allows scientists to identify specific pro-
teins, before they say for sure.
Then they will have to prove their insulin
is pure enough to inject into a person. They’ll
also have to demonstrate they can make
medical-grade insulin every time with their
process. Then, to get it to the people, they’ll
need to standardize the equipment so other
people can manufacture or buy it.
When asked if he thought his team would
eventually create a yeast that could produce insu-
lin, Huon de Kermadec responded confidently.
“Yes, of course,” he says, “it isn’t rocket sci-
ence.” But then there are the regulatory hurdles.
Di Franco has been reading up on the history
of democracy in ancient Athens and is trying to
craft his organization’s bylaws in the spirit of the
world’s first democracy. He also wants its prod-
uct to be democratically affordable : Di Franco
thinks roughly $10,000 should be enough to
get a group started with the equipment needed
to produce enough insulin for 10,000 people.
Each of these $10,000 setups would be some-
where between a middle- school science experi-
ment and an industrial laboratory, requiring
rooms of equipment; think something closer
to a medical-grade brewery than to a counter-
top bread machine. The resulting product, he
says, would cost someone with diabetes doz-
ens of dollars a month instead of hundreds.
That’s noble, in theory, but there’s a reason
why the FDA puts a lot of effort into certify-
ing the labs that make our medicines: mistakes
can be fatal. The U.S. drug-development system
may be expensive, but it does guarantee quality.
The Open Insulin Project falls into a black
hole outside of FDA regulation, according to
Peccoud, the Colorado State professor. For one
thing, the project may skirt some regulations
by being a nonprofit. Also, the FDA allows in-
dividuals to largely do whatever they want to
themselves. “If you want to inject yourself with
home-brew beer, there’s no law to stop that,”
says Peccoud.
If it does reach a production phase, Open
Insulin would have to conform to Good Manu-
facturing Practice, the FDA rules for factories
that make medicine, food, cosmetics and med-
ical devices. And because the group plans to
share its insulin-production framework online,
crossing state lines, there may be other legal is-
sues on the horizon. One solution might be to
partner with other players in the health care
system, like hospitals and pharmacies, which
create custom versions of everything from acet-
aminophen to opioids in a process called com-
pounding and navigate the demands of the U.S.
system already. However, that’s likely to make
the final product more expensive.
Ultimately, it’s not clear that the Open Insu-
lin Project’s real goal is to facilitate insulin mini-
labs across the U.S. The group intends to put
the plan for their designer insulin- producing
yeast online as soon as it’s done, but only for
“research purposes,” says Di Franco. And with-
out brewing facilities or the ability to check and
purify the hormone, the plans themselves are a
long way—scientifically and legally—from the
point where anyone will be injecting home-
grown insulin. Di Franco has offered up his own
body as a proving ground once the lawyers sign
off: “I’d be thrilled to be the first person to take
the insulin,” he says.
There’s ample evidence that insulin doesn’t
need to be as expensive as it is in the U.S., even
without DIY labs. For one thing, just across the
border with Canada, a vial of insulin costs $30.
In January 2020, Colorado will become the first
U.S. state to put a $100 cap on the co-pays in-
sured patients pay for insulin. Minnesota is
considering a similar law.
“It’s an old drug,” says Peccoud. “It’s not
hard to produce. It should not be more expen-
sive than Tylenol. Insulin is just pure greed.
And a failure of government.” □HEALTH CARE
I N N OVAT O R S
Christine
Lemke
The biggest
Big Data
There are 7.5 billion
humans, and tens of
millions of us track our
health with wearables like
smart watches, as well
as with more traditional
devices like blood-pressure
monitors. If there were
a way to aggregate all
that data from even a few
million of us and make it all
anonymous but searchable,
medical researchers
would have a powerful
tool for drug development,
lifestyle studies and
more. California-based
Big Data firm Evidation
has developed just such
a tool, with information
from 3 million volunteers
providing trillions of data
points. Evidation partners
with drug manufacturers
like Sanofi and Eli Lilly
to parse that data; that
work has led to dozens
of peer-reviewed studies
already, on subjects ranging
from sleep and diet to
cognitive-health patterns.
For founder Christine
Lemke, one of Evidation’s
ongoing projects, to see
if new technologies can
effectively measure chronic
pain, is personal: Lemke
has a rare genetic disease
that causes frequent back
pain. Evidation is partnering
with Brigham and Women’s
Hospital on the project.
ÑJeffrey Kluger