Failure, Innovation and
Now there are three words that are not often in the same
The front cover of the NY Times Magazine on Sunday, November
16, is called “In Praise of Failure” and trumpets the Innovations issue containing
a host of articles about inventiveness and the critical ingredient that failure
is to success. But it was an article by Dan Hurley on page 61 that caught my
eye. It had actually appeared on my kindle earlier in the week, but I waited to
write about it so that I could send you the link:
The article is called, “Why Are There So Few Blockbuster
Drugs Invented Today?”
The first half of the story focuses on the work of inventor
Todd Zion to create a self-regulating form of insulin. It’s fascinating
chemistry and biology and worth a read. However, it is the second half of the
article that was most interesting to me because, (SURPRISE!) it promotes a
theory that I have been pushing for about 10 years. That is that the genome
project as a source of medical innovation was overhyped—especially as it
applies to cancer.
Here’s a quote about genomics from the article: “But this
golden road to pharmaceutical riches, known as target-based drug discovery, has
proved to be more of a garden path.”
Chronic diseases that kill Americans including cancer are
not monogenic etiologically or even defined by multigenic aberrancies. They are
more complicated than that. Even when a target is identified, hitting it “right
between the eyes still often fail(s).”
“If you read them now, the claims made for genomics in the
1990s sound a bit like predictions made in the 1950s for flying cars and
A study quoted in the article showed that investment in genomic
drug development has had a smaller and smaller return on investment over the
years. Another study showed that only 17 of 50 novel drugs approved by the FDA
between 1999 and 2008 came from targeted research. More came the old-fashioned
way—trial, error and luck. Sort of like that drug for angina turned into Viagra.
It has been clear to me for a while that certain known
aspects of cancer make this outcome the predictable one in oncologic
also be a generalized homeostatic failure of the systems in the human body that
prevent carcinogenesis and the genetic abnormalities could be a symptom rather
than a cause of the disease. The success of some of the latest immunotherapies
suggests that targeting the non-cancer may be as useful as targeting the malignant
cells or their genes. Furthermore, renal failure can be caused by lots of
things (including, but not limited to, ethylene glycol), but when a patient’s
GFR drops below 5, my guess is that a biopsy of that failing kidney will yield
only the information sufficient to discern the tubular vs. glomerular origin of
the renal insult, not the agent. The same may be true of cancer. What we see
clinically may be the end-stage view of the failure of multiple potential paths
in the homeostatic anti-carcinogenic surveillance systems.
of malignant and non-malignant components that work together to promote the
cancer’s growth and spread. The genes of the cancer cells may or may not be the
sole cause of the problem and by the time the problem is diagnosed there are
multiple genes in play.
make room for the next generation of humans. After all, mostly old people get
cancer and their utility to the universe (procreation) has been used up like
collegiate eligibility for varsity sports. (My utility to the universe and to Duke basketball have both been up for years.) Medicine is the most anti-Darwinian intellectual
discipline of all as we doctors push back on God’s plans with the usual results
found serendipitously. What the heck is wrong with that?
test development, we would go a lot faster. Who elected them arbiters of
anything? Surely not Congress for the federal budget doesn’t pay for the entire
FDA thus the need to extract users’ fees from pharma to support the very agency
regulating pharma. Talk about conflict of interest!
all of that, I suggest you go back and now read the first half of the article
about how Dr. Zion is developing the self-regulating form of insulin without
the use of genomics.
in cancer research have to make up our minds. Either we are going to go about
this in the sloppy way we always have, tripping over the answers or we will try
to strategically plan the cure. People like Dr. DePinho and the drug companies
and Stand Up 2 Cancer are betting on door number 2. With enough thinking and
planning we can cure cancer—you know, make it history.
those of us who have touched the patients, done the research, written the
protocols and analyzed the data know better. The real motto of success is
“better to be lucky than good.” That also means people with too few failures
need to try harder. Personally, during my research career, I was never at a
shortage for failure. And if failure is defined as getting unexpected and
initially inexplicable experimental results, some of my best successes, limited
though they were, came from failures that, once explained, allowed the design
of better experiments and the finding of more important, unanticipated molecular
mechanisms of antineoplastic drug action.
you wish to be innovative, try something. If you wish to be doing the same old
thing all the time, plan a bit more when it comes to science. Play it safe.
Safe for you, not for the patients.
my favorite scientific axiom: “Man plans. God laughs.”