For metastatic colorectal cancer, there has been increasing recognition that certain molecular changes in the genes of the tumor (often called biomarkers) can help predict the benefit from certain chemotherapies like cetuximab.
One of the most frequently used biomarkers is the KRAS mutation, which predicts lack of efficacy to cetuximab and panitumumab. This has been replicated in at least a dozen large randomized studies. Several recent developments have added complexities to what had appeared to be a clear biomarker and a consistent treatment algorithm.
Recently, investigators from Europe reported that this relationship may not be straightforward for one of the more common mutations in KRAS (Tejpar, Bardelli et al., JAMA '10). Related work has identified other less common, mutations in additional regions of the KRAS gene, but the clinical implications of this are also not clear for these mutations. Adding to this complexity is the recognition that additional genes, including BRAF and NRAS, may also modulate this sensitivity.
Recently BRAF was included in the treatment recommendations from the National Comprehensive Cancer Network (NCCN) which are commonly used as a guideline for best practices in the community. However, as additional data was acquired, this relationship became less clear and there are calls for this to be removed from the guidelines until further clarity is provided.
These findings collectively reiterate several broad themes in clinical oncology:
- Biomarkers are a constantly evolving field for colorectal cancer and many other tumor types.
- Dissemination of the best practice patterns for biomarkers will require a new model of physician education.
An effort to include broader interpretation of biomarkers in test results to physicians is one necessary step to improve care. As the biomarker field continues to add complexity to clinical care, it provides opportunities for integration of electronic decision models into patient care. Ultimately, this added complexity will improve the outcomes of patients, and should be embraced and included with recognition of the evolution of this field.
For biomarkers, nothing is constant but change.