The Role of Metabolomics in Diagnostics

numares AG has recently entered into a collaboration with Oxford University to develop a metabolomics-based test which could help improve the diagnosis and therapeutic monitoring of patients with multiple sclerosis (MS). 

To learn more about the role that metabolomics could play in the diagnosis of diseases such as MS, and the advantages it could offer over current methods, I spoke to Volker Pfahlert, Chairman of the Executive Board of numares AG.

AM:  What are some of the limitations of current methods of diagnostics?

VP: The current industry standard is chemistry-based diagnosis, for which tests often evaluate a single biomarker and can be expensive. These chemistry-based tests evaluating single biomarkers often require complex sample preparation, test-specific reagents, and work by sequential measurement of biomarkers, leading to time- and cost-consuming processes.

numares’ differentiated AXINON® software-based in-vitro diagnostic (IVD) system evaluates nuclear magnetic resonance (NMR) spectroscopy creating a standardized 1H-spectrum to evaluate metabolic biomarker networks. The output from these NMR analyses provide physicians valuable information on the disease status of patients. An NMR spectrum reflects hundreds of metabolites in a sample, however biological samples evaluated on different NMR machines can produce different results. We developed Magnetic Group Signaling® (MGS®) to enable NMR for highly standardized sample processing and rapid throughput testing to ensure reproducible results for higher sensitivity and specificity. More accurate test results translate into earlier and improved medical decisions.

numares’ tests address unmet medical needs in the indication areas of cardiovascular diseases, nephrology, oncology and neurology, shaping another important pillar in precision medicine. Beyond clinical relevance, numares’ AXINON system also addresses the challenges of affordability because it has a software-like cost structure, does not require manufacturing chemical plants, and does not require analyte-specific reagents (consumables). It offers straightforward new test development, unlimited re-processing of spectra that enables further testing under new aspects and importantly, implementing a new test on the system is as easy as downloading an app on your smart-phone. Our focus is to develop a broad range of medically-needed tests that address these concerns. 

AM: What role can metabolomics play in diagnosing human diseases, and what advantages can this offer?

VP: numares’ approach utilizes the effects of a disease on the dynamics of human metabolism. These effects can be recognized as specific changes in a biomarker network, generated by changes in the metabolic “machinery” caused by the disease. The concentration of metabolites in various body fluids depends on the genetic background and living conditions of an individual. This metabolic pattern is subject to normal physiological variation, but may also reflect pathological processes in the human body. Nuclear magnetic resonance (NMR) spectroscopy allows for the simultaneous quantification of metabolites in human specimens such as serum or urine. This metabolic ”snapshot” or system-wide data can be used in diagnostics for extraction of a subset of data – specific metabolic biomarker networks for risk assessment, disease diagnosis, extent/severity of disease progression, therapy monitoring and deeper understanding of physiological processes in the patient.

For example, our renalTX-Score test, recently launched in Europe, evaluates the status or even better the disease-related changes of a metabolic biomarker network rather than quantifying just one biomarker, as with traditional diagnostics, in order to diagnose rejection of a transplanted kidney. One important differentiator is that in our world the metabolic biomarker network is described by a mathematical equation as a narrative to the changes in the metabolic “machinery”. The current diagnostic standard for acute rejection is biopsy of the kidney, however this invasive intervention can damage the transplant, cause discomfort and stress in the patient and is relatively expensive. renalTX-SCORE, for the first time, enables a regular and close monitoring of the transplanted kidney without additional strain on the patient and the organ.

AM: Can you tell us about some of the hurdles faced when applying metabolomics to human diagnostics?

VP: I think there are a variety of challenges. First, the multi-metabolite aspect described before has some severe consequences. For example, when developing single-marker tests only one marker has to be found; we have to find several markers to set up the metabolic biomarker network. This poses extra technical questions on the machinery applied to find those markers. By applying techniques from artificial intelligence this can be managed. Another aspect is the sample size; it is a simple result from Big Data that such explorations require 5 to 10 times more samples as a basis than when looking for single-markers. For example, the renalTX-SCORE mentioned above, was developed using several thousands of samples, which is an unusually large number for a diagnostic test development. Collecting this large number of samples is a challenge. Personally, I think that the greatest challenge is the mindset needed: to handle the described issues the standard approaches applied in the diagnostic industry cannot work. One has to be willing to approach the named challenges with completely different methods, as the mentioned artificial intelligence techniques. 

AM: How important are academic collaborations in the development and clinical translation of these technologies?

VP: We believe that academic collaborations are mutually important for both numares and researchers for the advanced development of tests that can be translated into clinical use. Collaborations with researchers at academic centers enable us to deploy our system to assist them in their work to both a) understand human disease and b) further expand our product pipeline through external review, insight, and validation of our technology to develop diagnostics based on that preliminary scientific work.  

Specifically, we recently announced a collaboration with Oxford University to develop an in-vitro diagnostic test to improve therapeutic decision making for patients with multiple sclerosis (MS). Oxford is a particularly valuable partner for this development, as it published the first scientific evidence for a potential diagnostic test based on metabolomics analysis of MS patient samples by using NMR technology and has long-standing expertise in MS research. Together, the goal is to develop a diagnostic test that reliably identifies the disease transition from RRMS (relapsing remitting MS) to SPMS (Secondary progressing MS), which could replace the current retrospective approach based on evaluating progress of neurological symptoms or incomplete remission after a relapsing episode. As secondary progressive MS patients do not benefit any more from standard RRMS therapy based on ß-interferon, timely onset of mitoxantron medication is highly warranted to decelerate further disease progress.

AM:  What do you think the future holds for human diagnostics?

VP:  As technology continues to improve and become more widely-adopted, we believe diagnostics will shift towards real-time precision medicine to be:

• more accurate through evaluation of metabolic biomarker networks compared to current approaches – or fingerprints of disease;
• more comprehensive through parallel measurement of biomarkers (vs. sequential measurement); and
• more affordable through simple sample preparation and software-based analytics for a fast, easy and cost-effective process. 
  
  

It is our vision that it is extremely important to develop more accurate and cost-effective diagnostics, as this will lead to actionable clinical decision support that can significantly impact patient outcomes. Improved affordability will also alleviate increasing global healthcare costs. 

Volker Pfahlert was speaking to Anna MacDonald, Editor for Technology Networks.