Biomarkers: key players in personalised radiotherapy

> Welcome to community Information > Technology in Surgery and Therapy > Index: All Articles > Home >

Personalised medicine or individualised therapy is an approach that was greatly advocated over the last decade and remains the leitmotif in the forthcoming years. The “one size fits all” system that used to dominate the world of radiotherapy and oncology is no more an accepted way of treatment, as patients are individuals, with particular characteristics and responses to treatment, therefore they require personalised therapy. As Sir William Osler – the father of modern medicine stated: “Variability is the law of life, and as no two faces are the same, so no two bodies are alike, and no two individuals react alike and behave alike under abnormal conditions, which we know as disease”. A patient-tailored treatment must therefore, consider patient-specific characteristics, whether associated with normal tissue response or tumour response.  

Tumour Variability

The latest technological and radiobiological advances allow a highly accurate delivery of radiation, leading to an increased therapeutic ratio (i.e. high tumour control with minimal damage to the normal tissue). This is also owing to the identification of numerous tumour markers and to the possibility for their specific in vivo targeting. We know from radiobiological studies that certain properties exhibited by tumours such as: hypoxia, angiogenesis, cellular proliferation, and the fraction of cancer stem cells present in a tumour, are probably the main culprits for treatment resistance and failure in radiotherapy. Consequently, their identification and targeting plays a crucial role for an optimum outcome.

Biomarkers

According to the National Cancer Institute, a biomarker is a “biological molecule found in blood, other body fluids, or tissues that is a sign of a normal or abnormal process, or of a condition or disease”. Biomarkers come in different forms: proteins, antibodies, nucleic acids, collection of alterations (gene expression). For a biomarker to become clinically noteworthy it has to meet several requirements: high sensitivity and specificity, biochemical stability, high predictive value while also being non-invasive. Biomarkers that meet the above criteria can have valuable applications throughout the entire sequence of medical appraisal, starting from diagnosis, tumour staging, patient stratification, treatment monitoring and concluding with outcome prediction and risk assessment.

Cancer Stem Cells

Among all tumour properties, the presence and the role of cancer stem cells in cancer growth and development is probably the hottest current topic in radiobiology. Although cancer stem cells represent only a small subpopulation of cells (usually below 1% of all cells constituting a tumour), they are powerful entities with unique properties. There are several specific characteristics that define cancer stem cells: higher radioresistance compared to non-stem cancer cells, “immortality” (i.e. the ability of self-renewal), the capability to create all heterogeneous lineages of the original tumour (including differentiated cells that have organ-specific properties) and their preferential residence in special microenvironmental niches within the tumour in order to preserve and protect themselves. Given the fact that a single cancer stem cell is capable of growing (or regrowing) a whole tumour, the eradication of these subpopulation of cells is critical.

Identifying Biomarkers for Cancer Stem Cells

However, in order to be eliminated, they must be first identified. Biomarkers play an important role in the identification of cancer stem cells, owing to the unique properties exhibited by these cells. So far, three classes of biomarkers for cancer stem cells have been researched:

• cell surface markers (for proteins that are outside the cell membrane),
• internal markers (proteins from inside the cell),
• functional markers (i.e. identify cellular processes).

The most commonly used markers for the identification of cancer stem cells are cell surface markers such as CD (cluster of differentiation) molecules, which are surface proteins that allow the analysis of cell differentiation, thus distinguishing between stem-like cells and differentiated cells. CD markers comprise of a vast group of surface markers, which are often tumour specific. To increase their specificity, it is not uncommon to employ two or more such markers for labelling cancer stem cells in laboratory settings. 

Imaging Biomarkers

Another technique to identify stem-like cells in tumours is the use of functional imaging, such as PET (Positron Emission Tomography) and MRI (Magnetic Resonance Imaging), and is growing in popularity. In this regard, ⁶⁴Cu-labelled stemness-specific agents have been evaluated for various cancer types to allow the detection of cancer stem cell fraction via PET imaging. Furthermore, superparamagnetic iron oxide nanoparticles for MRI have been used to image cancer stem cells, research that is still in investigative stages.

While biomarkers, in general, show great promise in personalised oncology, no gold standards have been yet established to reflect the characteristics of the tumour as a whole, and there are still several challenges that need to be overcome in this field.

Biomarker Analysis

One of the challenges is the quantitative imaging of biomarkers which is a precondition for precise treatment response assessment. Another challenge facing diagnostic imaging is the standardisation of imaging protocols to enable the generation of imaging databases. Databases comprising of large amounts of data are valuable sources for data mining, i.e. for harvesting information based on various image patterns. Data mining is undertaken via machine learning and statistical methods for data analysis. In this respect, an emerging machine learning technique is radiomics, which supports the extraction of quantitative data from medical images that reflect diverse cellular characteristics and establishes correlations between image features and patient outcome. 

Using Radiomics

Developments in radiomics already allow for the analysis of quantitative landscapes of various tumour types, thus serving as digital biomarkers. Image patterns and features that are identified through radiomics are typically beyond the grasp of the human eye, which reflects the omission of potentially important information concerning tumour assessment and treatment planning.

The current scientific literature shows that big data and radiomics are likely to assist radiation oncologists in patient stratification, therefore treatment personalisation. Nevertheless, to become a widely used practice, the field of radiomics requires more robust data, validation across clinics and standardisation of data reporting and analysis. Radiation oncology was built decades ago on strong interdisciplinary pillars that today support new levels of scientific disciplines allowing a more personal approach towards patient care. We have to take advantage of these new opportunities to further improve our knowledge and to better exploit the field of data science for the benefit of our patients.

Loredana Marcu PhD, 1 February 2020

===============================================

> Welcome to community Information > Technology in Surgery and Therapy > Index: All Articles > Home >