Personalized Health Care (PHC) in Cancer - Genomic Biomarkers for Pharmaceutical Development - page 41

treatments in the second half of the study (200 patients). In addition to the BATTLE-2 study,
the Lung Cancer Mutation Consortium (LCMC), an initiative of the National Cancer Institute
comprised of 14 leading cancer centers across the US, has formed to evaluate the frequency
of key mutations in NSCLC, including KRAS, EGFR, HER2, BRAF, PIK3CA, AKT1, MEK1,
and NRAS using standard multiplexed assays and ALK rearrangements and MET amplifica-
tions using FISH [129] . In 2011, 830 patients had been registered and 60% of these tumors had
a driver mutation detected, 95% of which were mutually exclusive [129] . Patients with spe-
cific mutations are offered participation in LCMC-linked trials where their mutational status
would suggest a higher probability of success. Creative trials and multi-center initiatives with
the goal of evaluating key biomarker-treatment relationships will continue to advance the
field and identify new markers that could become diagnostic tests.
The application of molecular diagnostics will continue to revolutionize the drug discovery
and development process, with gene-based and molecular diagnostics testing growing at a
30-50% rate [120,123,124] . As many as 1500 genes and 5000 proteins are currently candidates
for new molecular tests, and diagnostic divisions already exist in many companies to com-
bine drug development with the production of diagnostic tests. Currently, the use of com-
panion diagnostics for patient screening in clinical trials is surprisingly infrequent, often due
to the challenges in identifying the exact genetic cause for most diseases where this remains
unknown, and the likelihood that multiple factors are important for disease development / pro-
gression, making it difficult to identify a test that would yield a definitive response. However,
the molecular diagnostics industry is constantly evolving. New technologies such as next gen-
eration sequencing (NGS) and new disease associations / clinical opportunities for predicting
efficacy and monitoring disease outcome are constantly emerging.
Standardization of NGS workflows could substantially reduce the cost of companion diag-
nostic assay development, as most of the validation around instrumentation and sequencing
protocols would already be in place; therefore, a new assay would only need to be validated
with respect to the genomic value in a particular cancer type or patient subset. Furthermore,
NGS could have a central role in the discovery of new genomic biomarkers since many dif-
ferent types of experiments can be performed on a single machine. For example, the FDA
recognizes 41 genes or chromosomal rearrangements as pharmacogenomic biomarkers,
most of which are evaluated by laboratory tests [130] , but NGS could allow all biomarkers
to be combined (plus some others) in a single test. In support of this concept, a panel that
sequences 92% of all genomic rearrangements relevant to TKIs has been developed recently
[131] . For known alterations, it would be more cost-effective to use these types of NGS gene
panels instead of multiple individual diagnostic tests, so as we increase our understanding
of various cancers, and as new drugs become available, these types of panels may make the
development of companion diagnostics more feasible.
2.4 CONCLUSIONS
Clinical innovation always seems to take longer than expected. A decade after comple-
tion of the human genome sequence we still only have a handful of FDA approved com-
panion diagnostics that predict response to therapy. These successes point to a promising