chapter  10
Predictive Markers for Targeted Breast Cancer Treatment
ByHans Christian B. Pedersen and John M. S. Bartlett
Pages 15

Currently selection of breast cancer treatment is guided predominantly by patient

prognosis using classical pathological assessment of tumors, with higher risk

patient groups being offered more aggressive therapy. The choice of treatment

regimen is guided by a very small number of predictive biomarkers (Table 1) (1).

However, it is now clearly recognized that not only may the risks of treatment

outweigh the benefits in some patient groups but that not all patients are equal

with respect to their response to and benefit gained from exposure to novel and

established therapies. There is increasing interest in the identification of bio-

logical markers to improve the targeting of established therapies to those cancers

likely to respond. In addition, development of new molecular targeted drugs has

led to an urgent need for research into predictive markers that can distinguish

patients who benefit from treatment. New proteomic technologies are leading to

markers being discovered at an increasing rate, with protein expression, cellular

localization, and posttranslational modification being recognized as those of

potential importance. Many new immunohistochemical (IHC) markers are

currently being tested for predictive potential for various therapies in breast

cancer. Through large-scale gene expression studies using microarrays, it has

been realized that tumors have molecular signatures that may determine response

to treatment and patient prognosis. Many candidate-predictive and prognostic

signatures in breast cancer are being explored, and some are already in use to aid

patient management (e.g., OncoType Dx1, Genomic Health, Redwood City,

California, U.S.). The challenge for the future is to identify both the biological

and diagnostic determinates of treatment response in order to allow matching of

appropriate treatments to specific patients.