Approximately 320,000 patients are diagnosed with some form cancer in the UK each year.
Despite improved therapies, half of these patients still die of their disease, usually
after the failure of anticancer drug treatment. To date, mechanisms of drug resistance
and also determinants of treatment toxicity which varies significantly between individual
patients are poorly understood. Recent advances in molecular profiling techniques allow
the comprehensive interrogation of large numbers of genes, transcripts and proteins which
may be altered and contribute to progression and drug resistance in tumour tissues.
These technologies, including genome wide RNA expression and methylation profiling and
DNA sequencing can be used to analyse the causes of cancer drug resistance in detail and
this information can subsequently be used to develop better cancer drugs which prevent or
overcome resistance and prolong patient survival. Furthermore, molecular studies can help
to understand why some patients develop excessive toxicities during treatment with drugs
which are well tolerated by most patients. This could help to develop tests that predict
which individual patients will not tolerate a specific drug.
Importantly, we recently demonstrated that individual human tumours can contain highly
heterogeneous cancer cell populations, for example good and poor prognosis cancer cells
within the same tumour. This intratumour heterogeneity may have hampered the
identification of markers of drug sensitivity or tumour aggressiveness in the past as
single tumour biopsies are unlikely to reveal this complexity. Thus, our project uses our
established sample collection protocols to comprehensively sample surgical specimens.
This will allow the analysis of intratumour heterogeneity and its impact on outcome which
is an urgent clinical need.
Tumour tissues can also change their molecular characteristics over time, for example
during treatment, which mandates the longitudinal collection of tissue and blood
specimens in order to understand how treatment resistance develops. We and others have
recently shown that blood samples can be used to obtain information about tumour
progression without the need for rebiopsies (Gerlinger, unpublished results and Forshew
et al. Sci Transl Med 2012, 4:136ra68). Our aim is to regularly collect blood specimens
from which tumour DNA and tumour cells can be extracted and studies as a surrogate of the
tumour lesions. Taken together, multiregion sampling of tumours at surgery and
longitudinal monitoring of molecular alterations over time should provide crucial
insights into tumour heterogeneity and tumour evolution critical for progression and drug
resistance.
The understanding of mechanisms of tumour progression also requires the use of cancer
model systems in the laboratory and established cancer cell lines are generally used for
this purpose. Such cell lines have often adapted to the laboratory tissue culture
environment which changed their characteristics over time. Thus, they frequently behave
differently from cells isolated directly from fresh tumour specimens, hindering their use
for the identification of drug targets and resistance mechanisms. To circumvent this
limitation, we will also use the collected tissues to isolate primary tumour and stromal
cells and to maintain them in the laboratory for a limited period of time. This will
provide the opportunity to study tumour characteristics in greater detail and will
prevent false conclusions which can arise from artifacts which cannot be avoided in long
term tissue culture.
Together, these efforts should improve our understanding of tumour evolution over time
and reveal some of the mechanisms whih allow tumours to spread through the body and to
develop cancer drug resistance. This will be a major improvement over traditional
approaches which have most likely failed to identify mechanisms of cancer progression and
treatment failure because they relied on single biopsies, missing heterogeneous changes
in individual tumours and changes occurring over time (for further information: see
Gerlinger et al (NEJM, March 2012) and Yap et al (Sci Transl Med, March 2012).
