A few weeks ago, I traveled to Washington, D.C., to attend a National
Cancer Policy Forum-sponsored workshop, titled “Informatics Needs and
Challenges in Cancer Research.”
I was interested in the content — how can we integrate and make
sense of the vast amount of continuously generated cancer-related data —
as well as the individuals involved. The vice-chairwoman of the conference was
Amy P. Abernethy, MD, of Duke University School of Medicine, one of my
mentors and a visionary in the area of rapid-learning cancer care. All of the
other speakers — from academic thought leaders to authors to industry
executives — had extraordinary insights, as well.
William Wood
One of the presenters — Atul J. Butte, MD, of Stanford
University School of Medicine — gave a particularly provocative talk. In
our current environment, he said, tremendous amounts of data already are stored
and available in the public domain, and virtually all parts of scientific
research can be commoditized. With a few mouse clicks, Butte showed how
interested individuals easily could access public-domain gene expression
profile datasets and — funds permitting — could outsource laboratory
research, including animal models, to test hypotheses.
Pointedly, he paused before declaring the only thing that could
not be commoditized is the ability to ask good translational questions.
Half-jokingly, he wondered, if computer, IT and Internet companies could be
started in garages, are we really that far away from a garage biotech?
Many of Butte’s comments clearly were meant to be thought-provoking
rather than taken at face value, but they raised important questions. If the
core asset in scientific inquiry is intellectual capital, how much untapped
potential exists around us every day?
Some possibilities:
- Clinicians who derive biological insights from patient care but lack
a laboratory (or even “dry lab”) presence or partnership to drive
insight to discovery.
- Scientists outside of medicine who can think of new ways to approach
old problems.
- Patients, families or friends — think of PatientsLikeMe (www.patientslikeme.com) — who want to contribute to
solving the diseases that afflict them.
Indeed, lots of the data to get us started on fundamental questions, or
even real-world clinical problems, already are available. They just need to be
put together in the right way.
Last year, Butte published two articles in Science Translational
Medicine in which he matched disease gene expression signatures and drug
gene expression profiles and came to some surprising conclusions. Could an
anti-seizure drug work in inflammatory bowel disease? An anti-ulcer drug for
lung adenocarcinoma? Additional studies are under way to validate these
possibilities.
Recently, I attended another lecture — this time by one of
Butte’s mentors, Isaac Kohane, MD, PhD, of Children’s Hospital
Boston — at a personalized medicine symposium.
In a more immediately clinically relevant example, Kohane showed how
aggregating the entirety of clinical observational data from Boston-area
hospitals during the past several years demonstrated a surprising spike in
heart attack incidence that then went away nearly as quickly as it came on.
The clear correlates, in retrospect, were the introduction and
withdrawal of rofecoxib (Vioxx, Merck) from the market. Again, the common
denominator of this example and others was the initiative and ability of a
creative individual to translate insight to discovery, using data and tools
already available.
I’ve been mulling over these talks as I’ve been scanning my
Twitter feed today. This is the weekend of the American Association for Cancer
Research conference (March 31-April 4), at which many of the “big
ideas” in cancer are being presented and discussed.
I wrote about Twitter in my last column. This and other social media
outlets have become, in a way, a modern-day equivalent of the European
coffeehouses that ushered in the Age of Enlightenment.
Traditional institutional barriers dissolve as ideas and insights from
thinkers everywhere, within and outside academic medicine, flow freely with
astonishingly rapid speed on virtually any topic, cancer certainly included. In
some ways, Twitter is a virtual and continuous scientific meeting on any topic
of the reader’s choice. Sometimes, the virtual and the on-the-ground
meetings coincide, as they have today when many of the tweets and posts that I
see are largely revolving around the data and lectures being unveiled, in real
time, at AACR.
Earlier today, one of the AACR attendees observed the enormity of the
activity ongoing in the convention center and tweeted: “You’d think
… we would’ve already had a cure.”
On the one hand, cancer-related scientific discovery — in the lab,
translationally, and in the clinic — is on the threshold of a potentially
very exciting era.
The book The Emperor of All Maladies by Siddhartha
Mukherjee, MD, PhD, is emblematic of a willingness of the cancer community
to engage with big, paradigm-shifting ideas. As he documents well, the greatest
progress in the past has come from risk-taking rather than incrementalism.
As tangible evidence of this approach, the NCI has launched its
Provocative Questions Project — available at
http://provocativequestions.nci.nih.gov — that promises
to “identify perplexing problems to drive progress against cancer.”
Among these questions are:
- How does the lifespan of an organism affect the molecular mechanisms
of cancer development?
- Can we use our deepening knowledge of aging to enhance prevention or
treatment of cancer?
- How does obesity contribute to cancer risk?
One of my colleagues recently submitted a proposal for grant funding to
address one of these questions, and many, many others are presenting within the
convention halls in Chicago at AACR.
On the other hand, it seems that despite our willingness to tackle
larger questions, we remain constrained by many of the usual barriers within
our traditional infrastructure. These include challenges in sharing data or
approaches with others because of competitive concerns around funding or
publications; a desire to achieve personal or institutional credit; incentives
to pursue incremental rather than transformative research, and to pursue the
same themes rather than new directions, as the safest route to continued
funding; external pressures to study specific pathways or compounds; and an
inability to access the large untapped reservoir of intellectual capital that
resides within clinicians, non-medical scientists, patients and other parties,
as well as an inability to properly make use of large amounts of already
available biological and clinical data.
So here, then, is another provocative question: What if we imagined a
new, complementary approach to cancer-related scientific discovery, a utopia
that could mirror the scope and pace of Twitter-like social media or community
“wikis” with “crowd-sourced” questions and
“crowd-sourced” answers?
The rules of the game would need to be very different. Everyone would
need to share everything with everyone else in real time. In this space, the
usual individual goals of prestige, funding and publications would need to
become secondary or maybe not even relevant. Anyone could contribute, and by
significantly expanding the breadth of involved individuals for any given
problem, specific skills could quickly be brought to bear on particular aspects
of the issues.
To do this, individuals with specific resources — such as
laboratory experiments, bioinformatics computational tools or clinical
observational data — would need to contribute when required to help the
cause. Versatile thinkers would need to be able to shift focus quickly and
frequently as the important questions change and evolve. To drive this
effectively, those involved would need to be motivated by a deeply aspirational
goal — a community-held mission to dramatically improve the care of cancer
patients within our lifetime — akin to eliminating HIV in sub-Saharan
Africa, or, in an earlier era, putting a man on the moon.
This vision raises many questions that are difficult to answer. What are
the limits of altruism? Who would pay for the research, and how? What entity
would be credited if any of the efforts are successful? What format would the
endeavor take, and who would oversee and manage it? What strategy could be used
to tap the reservoir of intellectual capital that needs to be tapped, to engage
the minds that need to be engaged?
At this point, I don’t have any answers to these questions. But
it’s a provocative idea, isn’t it? Who wants to help?
References:
- Dudley JT. Sci Transl Med. 2011;3:96ra76.
- Sirota M. Sci Transl Med. 2011;3:96ra77.
For more information:
- William Wood, MD, is assistant professor of medicine in the
division of hematology/oncology at the University of North Carolina in Chapel
Hill. He may be reached at william_wood@med.unc.edu. Dr. Wood
reports no relevant financial disclosures.