Biology

This is an excellent question. Caris is a member of a group of molecular profiling companies. These groups extract protein, DNA or RNA and measure the genetic information present in fixed tissue. Their intent is to match patients to therapy based on this static data. That is, the presence of a gene should, according to their reasoning, confer sensitivity to a drug. The proof of this concept, however, is sorely lacking. If we examine one publication in the literature (VonHoff D et al, J Clin Oncol Nov 2010), we find an objective response rate (measurable benefit in 66 patients treated using molecularly selected drugs) of 10%. Indeed, the positive results reported in this study reflected not a response rate but instead a 30% improvement in the time to disease progression, associated with molecular drug selection, compared with the patients prior, most recent (unsuccessful) therapies. To put this in context, a patient could arguably fail a physician-selected treatment after 1 month, and then receive a Caris selected therapy. If they responded for 1 month plus 10 days (a 30% improvement), they were counted as a success. The clinical relevance of that degree of improvement seems questionable.
While we fully understand the excitement surrounding genetic analyses, the more sophisticated researchers in the field are beginning to appreciate that the complexity of human biology demands more global (functional) analytic platforms that encompass all of the mechanisms of response and resistance. In this light, the presence of a gene cannot predict whether that gene will be expressed, active, counter-acted by a complementary gene, or functional.
We believe that human biology must be taken at face value in its most complex state. This is known as the phenotype and must be examined for the biological features that these interconnected cellular systems create. This field now known as biosystematics, or systems biology, recognizes the redundancies and uncertainties that separate the genotype (molecular profiling) from the phenotype (functional analyses) are not trivial. Our laboratory conducts phenotype analyses. Caris conducts genotype analyses.

One might imagine three platforms for the study of human tumor biology. Genomics studies information at the level of DNA, while proteomics examines protein expression. Functional analyses approximate the living organism by examining the behavior of cancer cells in their own environment. This allows heretofore unrecognized complexities to be examined in real time. A growing number of investigators around the world are beginning to recognize the important and clinical application of these techniques.

Bioinformatics is computer science applied to biological questions. Since all genomics-type sciences such as next-gen sequencing, transcriptomics and proteomics generate large amount of data, computational and statistical approaches are required to evaluate and analyze the generated data to guarantee high data quality. Additionally, since more and more proteomics and genomics data are generated and deposited into the public domain data mining and integration strategies, based on bioinformatics, are becoming more common. In the context of cancer research integration of multiple datasets can be used to validate individual results and prioritize biological pathways or targets for validation. Bioinformatics is therefore highly integrated with proteomics for the interpretation of results. Ultimately these types of “systems” strategies will shed new light on cancer biology and ultimately provide better understanding human cancer biology. This in turn will lead to more rationally designed drugs, more individualized treatments and better biomarkers.

There are numerous additional biomarkers under investigation. It is important to note that different biomarkers are likely to be useful for different reasons. Some biomarkers may help predict or determine responses to particular treatments and others may be useful in predicting how aggressive a particular cancer may be. Still other biomarkers are being tested as new diagnostic tools, for early detection of breast cancer.

Biomarkers can be found in the tumors themselves or in blood, urine, feces or even in breath samples.
They can be made of protein, RNA, or smaller biochemicals. Some biomarkers are actually measures of larger 'states' of the cell, including the stability of the chromosomes. There are too many under investigation to list them all here, but they include: Ki67 (a protein marker of cell division), chromosome length, and some microRNAs. For more details, see http://www.ncbi.nlm.nih.gov/pubmed?term=breast%20cancer%20biomarkers .

A biomarker is something that provides information (often indirectly) about a biological process or event. An example is elevated cholesterol. High cholesterol is frequently associated with coronary heart disease and so is a biomarker for heart disease.

With respect to cancer, there are different types of biomarkers that are used to guide treatment. In breast cancer, there are several that are routinely checked by pathologists.

An example using the estrogen receptor (ER): The estrogen receptor is a protein the binds to estrogen (when it is present). Once they are bound together, the receptor:estrogen duo can cause the activation/repression of a variety of genes, altering cellular activity. In general, estrogen acts as a growth factor for breast cells. When breast cancer is diagnosed, the cancer is examined for the presence of the the estrogen receptor. If it is present, that is an indication that the cancer cells are still 'listening' to the signals sent via estrogen. If the ER is NOT there (but the cells are still dividing - causing the cancer), it is an indication that the cells are more abnormal. They no longer require the signals provided by estrogen.

What does this mean for the patient?
Breast cancers that express the estrogen receptor (ER positive) may respond to treatments that block estrogen production/activity (i.e. tamoxifen).
ER negative tumors may be more aggressive and are not nearly as likely to respond to ER blockers (antagonists).

Other biomarkers in breast cancer include the HER2 protein (a growth factor receptor) and the progesterone receptor (PR).

The identification of biomarkers allows clinicians to personalize the treatment of individual patients, based on their molecular profiles.

The term 'ligand' is a generic one. It is not specific to cancer. A ligand is something that binds to something else. Typically, the term is applied to the things that bind to cellular 'receptors' to turn the receptors on. As an example, estrogen (a lipid hormone) is the ligand for the estrogen receptor (a protein). Once the receptor has bound estrogen, it can do its job. In this case that would be to turn on and off a set of genes. Other receptor/ligand pairs do different things. They can make cells divide, make cells die, or do things less dramatic, like cause cells to take up nutrients from their environment.

As an aside, some cancer drugs work by blocking the interactions between cellular receptors and their ligands. Examples include tamoxifen, which blocks the estrogen receptor:estrogen interaction and Avastin® which blocks the VEGF:VEGF receptor interaction by binding to free VEGF.