Biomedical Engineering Reference
In-Depth Information
most signifi cant with respect to the likelihood of the cancer actually leading to
death, and more importantly,
for the patient, offering a different set of prospects
for treatment
. . . . [T]he genetic testing that ' s done on a cancer [involves actually
comparing] the cancer to normal tissue from the same individual to look for
mutations that occurred during life that gave rise to that cancer.
Because an individual patient is being analyzed, and the genomic profi le is
created from that particular person's tumor, this process is sometimes referred
to as “ personalized genomic medicine. ”
Cancer Commons is Web-based paradigm whereby patients with advanced
cancers (along with their physicians*) can enter the relevant parts of their
clinical and genomic profi les, to the extent these are known, and receive infor-
mation regarding potentially useful treatments based upon the latest science.
This information comes from the Cancer Commons
molecular disease model
which is, in effect, a
living review paper
regarding genomic subtypes of cancer
and their management. The molecular disease model is created by experts in
the fi eld of genomic cancer treatment, usually in the specifi c type of cancer
with which the patient has been diagnosed. It encodes mappings among these
aspects of genomic medicine: genomic profi les, biochemical mechanisms of the
disease (commonly called “pathways”), and treatment options, including drugs
and combinations of drugs, clinical trials, and other treatment options. Together
these constitute a set of “actionable subtypes” of cancer, each representing a
functionally different molecular etiology driving the disease, and links it to
distinct therapeutic approaches. The molecular disease model also contains
citations to the relevant literature, including case studies that support the
mappings.
The molecular disease model is a “living” document in at least two senses.
First, it is represented in both human-readable and computationally accessible
formats and can thus be used directly by the algorithms that enable the Web-
based Cancer Commons platform to produce information regarding potential
treatments or trials appropriate for particular genomic and clinical profi les.
Second, it is constantly updated by expert curators based on the latest fi ndings
and opinions in the fi eld, as well as by information resulting from the case fl ow
that takes place within Cancer Commons itself. An initial disease model may
be created top down, for example, by a panel of recognized disease experts,
or bottom up, for example, via a statistical analysis of clinically annotated
genomic data. Regardless of how it is created, the initial model is likely to be
incomplete and contain errors and so will need to be refi ned, as described in
the next section.
*We speak in a number of places of what patients do, as though there are no physicians involved.
In reality, the target users of Cancer Commons are physicians (usually oncologists), physician/
patient dyads, or what we call “superpatients” who can operate nearly at the level of oncologists
but with a focus on their own disease.
Search WWH ::
Custom Search