Diamonds in the rough
The many facets of biomarkers

Definition and discovery

Although the generic term of a “biomarker” can be broadly applied to a very wide range of fields including astrobiology and ecology, one of its most common usages is within the field of medicine. In this context, a biomarker refers to the presence of some biological molecule that is correlated with a disease. An inherent property of a biomarker is therefore that its presence or abundance can be used to distinguish healthy individuals from those affected by the specific disease. Within the field of oncology, the importance of biomarkers has long been recognized since the link between early detection (and treatment) of a tumor and long-term survival of patients is well established.

With the development and growth in the last two decades of novel technologies capable of surveying for global changes in gene expression and protein abundance, it is unsurprising that the application of these technologies to cancer samples has revealed a wide range of cancer-specific alterations. Even before these latest developments, pioneering discoveries such as inherited mutations in BRCA1/2 and the expression of the prostate-specific antigen (PSA)had already laid the groundwork for the use of cancer biomarkers for diagnosis and risk assessment. At the same time, although the massive increase in data has highlighted many potential new biomarkers, their development is a long process. After identification and validation in multiple disease cohorts, a suitable assay needs to be developed and clinical utility of the marker needs to be established. The last steps involve obtaining regulatory approval and commercial development of the assay and reagents to allow its clinical integration.

Therapeutic application of biomarkers

Although part of the focus of cancer biomarkers has been for improving diagnostics, because of their selective/biased presence on cancer cells, some also represent potential therapeutic targets. For this recognition to happen successfully, another key biotechnology is typically required, and that is monoclonal antibodies (mAb). The mAb developed against a biomarker from any single clone are identical (simplifying regulatory approval) and provide the exquisite sensitivity required to recognize only the biomarker of interest. Once the specificity of the mAb has been verified, there are a wide range of therapeutic approaches available. First, a cytotoxic drug can be conjugated to the mAb, where it acts as a “toxic payload”. When such antibody-drug-conjugates (ADC) are administered, the antibody acts as the delivery mechanism, recognizing only the biomarker on the surface of cancer cells.

Once attached, the antibody and drug will enter the cell, allowing the drug to induce lethal damage to the cancer cell. Alternatively, the antibody alone can be administered, where it will circulate and bind to the surface of the cancer cell. Like any antibody produced by the host, other immune cells can recognize the constant (Fc) portion of the antibody, leading to a cytotoxic response (antibody-dependent cellular cytotoxicity; ADCC) that will kill the cancer cell. While highly effective, both ADC and ADCC rely on injection of foreign antibodies into the patient whose own immune system may mount a response against these antibodies, reducing their effectiveness.

While extensive research has already been published on methods to avoid these limitations, another approach using chimeric antigen receptors (CAR) seeks to avoid the problem altogether. Normally, an individual’s cytotoxic T or Natural Killer (NK) cells are educated during their development such that they do not recognize the individual’s own cells, thus preventing auto-immune diseases. CARs provide a way to bypass this self-tolerance in order to direct the immune system to attack cancer cells. To do this, the portion of the mAb that recognizes the cancer biomarker is fused to the end of a receptor that normally activates T/NK cells. The immune cells engineered to express this CAR are expanded ex vivo and then re-injected into patients function as a sort of “living drug”: The cells become activated when they recognize the cancer biomarker but lack the normal inhibitory signals to self cells, allowing them to kill tumor cells. CAR-T and CAR-NK cells have shown incredible promise against a wide range of cancers, however some solid tumors are able to maintain immunosuppressive microenvironments for reasons that are not entirely clear, limiting the effectiveness of this approach.


While all of these therapeutic approaches have shown great promise, they still ultimately depend on the accurate identification of potential candidate genes based on their expression or specific mutations. These constraints underscore the importance of robust tools for analyzing NGS data in order to maximize the chances of picking the right target to push through to clinical trials. In the end, harnessing the human immune system may hold the ultimate solution for specific and effective cancer treatments that avoid all of the pitfalls of current chemotherapeutic regimes.

Seeq, the search engine for human genomics, is available at