“2017 was a good year for Medical startups” was
a common phrase in the media. If you Google “Medical startups”, you will find
“Artificial Intelligence diagnoses 80 times better than doctors.”, “Farmington
medical startup targets hearing loss with new drugs”, Online Counselling Apps,
Cloud-based Medical Records, … What you will not find, however, is a startup
that develop drugs to treat diseases. So, what makes this field particularly
harder to get in as a startup, compared to biomedical engineering, digital
solutions to healthcare, or even developing drugs for disabilities? Why hasn’t
there been a Bill Gates or Steve Jobs for Drug Research and Development
(R&D), only big Pharma names like Pfizer and Roche? It’s not that there
aren’t tens of thousands of brilliant inventive scientists working to save
people, but it’s the nature of the drug R&D, itself, as well as today’s
economy that does not allow scientist-entrepreneurs to develop new drugs themselves.
First of all, one must define what a startup is.
According to Investopedia, “A startup is a young company that is just beginning
to develop. Startups are usually small and initially financed and operated by a
handful of founders or one individual”. To clearly examine the nature of Drug
R&D, I would like to do a quick overview of the steps of the process, in
this example, a hypothetical case study of the development of the drug Rumotak
for inflammation diseases, shown in the book Science Business by Pisano, 2006.
1.
Target Identification and Validation: The company finds the specific
biochemical pathway, receptors, protein, and gene that serve as a suitable
point for the intervention of a disease. For the Rumotak team, it took them
four years to yield a set of potential drug targets and to put up a hypothesis.
It took another six months of experimentation, gathering studies and data for
validation, and making sure it was an “druggable” reasonable target for their
effort.
2.
Lead identification and Optimization: The company finds or engineers a molecule
that inhibits this target. The Rumotak team had to discover a molecule that
inhibits this particular enzyme. They tested tens of thousands of molecules
against the target, narrowed down the list of possible compounds, and found
“analogs” of these compounds. This was seven years into the project.
3.
Preclinical Development: Experiment in vitro and in vivo to generate data on
the safety and potential effectiveness of the compound. This process usually takes
one year. The result is a compound that will go into the next phase. For
Rumotak, it was CC-30.
4.
Human Clinical Trials: The company conducts trials on human participants to
evaluate the safety and effectiveness of a drug on a defined population. There
are three main phases for this process
Phase
1: Experiment on a small sample of the population to study safety, usually from
1 to about on 100. The research team could use patients to experiment for
diseases like cancer or HIV/AIDS. Rumotak spent 1 year and 10 million dollars
on this phase.
Phase
2: Experiment on a targeted population, parallel a control group, to
study both safety and effectiveness in different doses. Each of 50 to 500
people. This phase usually takes 1 to 2 years.
Phase
3: Experiment on a large group of patients, ranging from 100 to 10000. Cost of
such trials can range from 50 million to 500 million dollars for about four
years. This is to confirm the data on the drug.
5.
Regulatory Approval: The company has to compile all the data about the drug and
submit them to the FDA. The FDA, in a year or more, will review and follow up
on the company, making sure the drug is safe and establishing the content the
company can market for the drug.
There are a few challenges for a startup to
develop a successful drug:
I.
Money
It is obvious to see that no operation can
conduct the entire aforementioned process without spending big money. A study
in the “Nature reviews Drug discovery” reveals that both capitalized and
out-of-pocket costs for bringing a single new drug to market is estimated to be
1.8 billion and 870 million dollars, respectively. In real life, the
research is usually conducted in the first few phases in the public sector, by
labs receiving funding from the Universities, Federal and States’ Institutions
and other funds with the same interest; or even in small private labs. Then
they will simply abandon, cooperate with, or sell the project to the heavy
hitters in the industry to bear the huge cost. Eli Lilly and Co poured
significant resources into their antibiotic. Until it was abandoned, Cubist
Pharmaceuticals picked it up and give to the market Cubicin after spending $220
million on R&D.
The Drug R&D process is a rigorous method of
discovery and problem solving, resulted from 150 years of development since the
early days when Paul Ehrlich hypothesized the concept of “chemoreceptor”. It
was refined and supervised to produce to safest and most effective drug. So if
the company perform any cost cutting that was not justified, it will definitely
get a word from the FDA.
Yet, even when the cost is at nine or ten
figures per drug, this is still not the deal breaker for venture funds.
Startups companies valued in tens of billions are common and extremely
attractive in the tech world. The problems start when you combine great cost
with great risk and a arduous amount of time before earning any revenue.
II.
Time
The simple anecdote above has already shown that
it takes about 15 years to get Rumotak to the market. According to a research
from The Pharmaceutical Research and Manufacturers of America (PhRMA), it takes
approximately 10 years to get any drug to the market.
Uber, a company valued at 48 billion dollar,
after 8 years operating has yet to turn profit. During its run, it is
constantly expanding, hiring, developing the tech and management; at the same
time it is providing 5 billion drives to customers around the world as of this
very moment. Uber is slowly refining their services to finally become the best
version of itself, standing side by side with Google, Apple, and Tesla. And
even without reports of profit, investors are still excited and betting on
Uber. It’s because they’ve seen this before: Amazon was without profit for the
first decade, losing 284 million dollar before blowing up and making everyone
richer and Jeff Bezos the richest man on Earth. They know good things take
time.
However, not only a hypothetical drug startup
wouldn’t generate profit for the first 10 years, it also wouldn’t even earn
revenue. It would have exponentially growing expense for years, before the drug
went through and the company tasted that sweet stream of revenue, at least
until the term of patent ran out. While investors for digital tech startups
would pour out champaign for every quarter the company’s projection of value
came out great, the investors for medical startups would be holding their
breath for a decade, losing sleep because of the risk. The equivalent of such
imagined drug startup would be like if Uber developed the app for 15 years
without releasing, and when it did it got struck down by FDA. Why would anyone
with a sane mind even consider investing in a drug startup?
III.
Risk
According to Matthew Herper, from Forbes, “A
company hoping to get a single drug to market can expect to have spent $350
million before the medicine is available for sale. In part because so many
drugs fail, large pharmaceutical companies that are working on dozens of drug
projects at once spend $5 billion per new medicine.”(“Drug” here means a new
molecular entity). Indeed, failure costs and risk profits.
And make no mistake, the risk
of drug R&D is high, growing at every step of the process. Nature produced
countless forms of existence for a molecule, infinite permutations of particles
numerous structures. In the Target Identification phase, a company would be
testing about just about every chemoreceptors, genes, cells and proteins
remotely related to inflammatory diseases. Afterwards, they will have to
validate about 100 to 300 targets, but as with all scientific research, the
answer is never 100% clear-cut, but a probability distribution. In the next
phase, to find a molecule that binds, company would be screening 1 million
compounds per week, narrowing it down to a compound that has on average 1 in 5000 chances this
would work and can become commercially viable. Preclinical development will
help confirm if that drug is the 1 in 5000, testing it on mice, a method that
has fundamental problems due to biological differences between mice and humans.
The overall probability of Clinical trial success (the likelihood that a drug
entering clinical testing will eventually be approved) is estimated to be less
than 12%. And only 8% of drugs that got this far actually got approved by the
FDA and went to the market. This still doesn’t guarantee that the drug will pay
off its investment. The cost of success/failure for discovery is so large that
most of the 95 biggest pharmaceutical companies in the US in the study sited
from Bernard Munos, produced only 1 new drug in the last 10 years.
All that, to produce a pill- a deceptively
simple product. No moving parts, no warranty or maintenance plan, drugs are
visibly less intimidating than a microprocessor or VR glasses. However, while a
camera might delight you, a drug can kill you. Side effects or misuse can cause
serious harm to a patient, which means lawsuit for the firm. Patent challenges
are also commonplace. Due to either inadequate research or a poor understanding
of the newly patented product or process, a patent may be granted that
infringes on older patents, or may be granted for a non-patentable concept. The number of patent challenges has increased
rapidly since the late 1990s. Over 80 percent of the new molecular entities
(NMEs) experiencing first generic entry in 2011–12 experienced a patent
challenge, compared with an average of less than 20 percent prior to 1998
(Grabowski, Long, and Mortimer 2014). If a drug startup had somehow magically
survived through 15 years of development, a class action can bring the company
without a strong legal team down.
IV.
Human resources
Finally, this is another problem for every
startup, including our hypothetical one. Every project as complex as this must
acquire the best people, not only in a field, but in every field. For a digital
tech startup, it will need software engineers, data analysts, computer
scientists, … The same goes a medical startup; it needed people from all
backgrounds. In just the Target Identification phase alone, Rumotak needed
molecular geneticists to identify specific genes that might be involved in the
disease. A bioinformatic group will do the same thing, but by gathering genetic
sequence data from public and private database. Cell biologist and biochemist
must validate the target worthy of research effort, and so on,... At the same
time, it must cooperate with labs in Universities and hospitals to
increase productivity.
And while it only takes a team of 20-year-olds
in a garage to develop a lean product ( a simplified version of the final
product for showcase and pitching to investors), it’ll take several teams of
veteran professors and doctors just to prove the problem can be solved. So the
cost of poaching talents are high during the very first phase of the process.
There is also a certain mindset that might
impede the integration of these experts. First, there has long been a feud
between the private sector and public sector. As it has been examined in the
book Genome War by James Shreeve, there is prejudice from scientists from the
public sector, equating the transition to the private sector to “selling your
soul to the devil”. And when you actually get these scientists into the
company, you will have to build a culture and structure around them that will
help these “isolated islands of expertise” to cooperate and egos to compromise.
Overall, it would be a nightmare for the HR department.
IV.
Conclusion
In conclusion, drug R&D is a complex process
that is time-consuming, expensive and fraught with risk. A new company with
little fund, dividend-focused investors and lack of talent will not survive in
an industry where even big pharmas struggle. For a startup like such to exist,
it will need a perfect storm. The entrepreneur will need a deep pocket,
management skills, a tolerance for huge risk and connections to both the
scientific community and the private sector. So only if both Craig Venter and
Jeff Bezos partnered up, I might be willing to bet on them.
Bibliography:
-
https://www.investopedia.com/ask/answers/12/what-is-a-startup.asp
“What
exactly is a startup?”
By
Amy Fontinelle, Investopedia, December 15, 2017.
-
“Science Business: The Promise, The Reality, And The Future of Biotech”
By
Gary P. Pisano, Harvard Business School Press, 2006.
·
https://www.forbes.com/sites/matthewherper/2013/08/11/the-cost-of-inventing-a-new-drug-98-companies-ranked/#5999f2432f08
“The
cost of inventing a new drug: 98 companies ranked”
·
https://www.forbes.com/sites/matthewherper/2013/08/11/how-the-staggering-cost-of-inventing-new-drugs-is-shaping-the-future-of-medicine/#37db10df13c3
“How
the staggering cost of inventing drug is shaping the future of medicine.”
By
Mathew Herper, Forbes, 2016.
-
“The Price of Innovation: New Estimates of Drug Development Costs.
DiMasi,
Joseph A., Ronald W. Hansen, and Henry G. Grabowski, Journal of Health
Economics, 2003.
No comments:
Post a Comment