226
on patients with breast cancer and lung cancer. This technology will allow physicians
to tailor the treatment and prognosis for an individual patient, using a small panel of
genes selected from thousands of genes.
Activating mutations of KIT or kinase platelet-derived growth factor receptor
alpha (PDGFRA) are found in the vast majority of gastrointestinal stromal tumors
(GISTs), and the mutational status of these oncoproteins is predictive of clinical
response to imatinib. PDGFRA mutations can explain response and sensitivity to
imatinib in some GISTs lacking KIT mutations.
Genotype-Dependent Effi cacy of Pathway Inhibition in Cancer
Therapeutic inhibition of genetically activated oncoproteins can induce massive
apoptosis of tumor cells, which may lead to dramatic regression of cancer. The
PI3K and MAPK signaling pathways are central regulators of oncogenic transfor-
mation and tumor maintenance. Systematic linking of drug response to genomic
aberrations in NSCLC, as well as in cell lines of other tumor types and in a series of
in vivo cancer models, has shown that tumors with genetically activated receptor
tyrosine kinases depend on PI3K signaling, whereas tumors with mutations in the
RAS/RAF axis depend on MAPK signaling (Sos et al. 2009 ). However, effi cacy of
downstream pathway inhibition is limited by release of negative feedback loops on
the reciprocal pathway. By contrast, combined blockade of both pathways can over-
come the reciprocal pathway activation induced by inhibitor-mediated release of
negative feedback loops and results in a signifi cant increase in tumor apoptosis.
Thus, by using a systematic chemogenomics approach, genetic lesions connected to
PI3K and MAPK pathway activation can be identifi ed and provide a rationale for
combined inhibition of both pathways. These fi ndings may have implications for
patient stratifi cation in clinical trials.
Mutation Detection at Molecular Level
It is known that genetic mutations are responsible for sensitizing some tumor cells
to chemotherapy, while other mutations render tumor cells completely resistant to
drug treatments. Research progress in this area has been slow because analysis of
DNA from tumors is complicated by varying amounts of tumor cells in patient
samples. Furthermore, the heterogeneous nature of many tumors makes it diffi cult
to accurately sequence the tumor DNA, which is required in order to personalize
treatment. This is compounded by cost-prohibitive, conventional low-resolution
sequencing methods that lack suffi cient accuracy to characterize the DNA in can-
cerous cells. Next generation sequencing can be used for the detection of cancer
gene mutations present at extremely low levels. Microreactor-based 454
Sequencing™ (Roche) advanced sequencing technology can generate hundreds of
thousands of DNA sequences in one run, rapidly and comprehensively conducting
high-throughput nucleotide sequencing, with specifi c application to sequencing of
10 Personalized Therapy of Cancer