Oliver Schacht, CEO of OpGen, Inc., examines the potential of bioinformatics in helping to fight antimicrobial resistance.
Bioinformatics
Infections pose numerous challenges to healthcare systems and contribute to higher morbidity, mortality and significantly increasing costs. While clinicians need to make rapid treatment decisions, often they rely on empirical use of antibiotics without knowing the causative pathogen or whether it is characterised as drug-susceptible or drug-resistant; consequently, leading to increasing antimicrobial resistance (AMR) and multi-drug resistance (MDR). Solutions to support the fight against antibiotic-resistant bacterial and fungal infections must be at the forefront of healthcare professional’s minds as we begin a new year. Currently, these so-called “superbugs” are responsible for 35,000 deaths and 2.8 million infections annually in the US alone. In October 2020, the WHO placed antimicrobial resistance (AMR) on its list of the top 10 global public health threats. What’s more, the emergence of Covid-19 has heightened and accelerated additional AMR risks among hospitalised patients; it has been reported that 72% of Covid-19 patients received antibiotics even when not clinically indicated in most cases.
In 2021, technology platforms and solutions that simplify and facilitate detection of drug-resistant infections will continue to evolve and provide value in the fight against AMR. With more and more healthcare systems adopting platforms that deliver treatment insights to enable smarter use of antibiotics, we can expect to see an uptick in cloud-based solutions that encourage collaboration between healthcare facilities and providers. Additionally, as the world continues to grapple with the Covid-19 pandemic, data-sharing platforms capable of delivering broader insights on infectious disease outbreaks will become valuable solutions for epidemiologists and frontline workers alike. Here are a few ways these platforms can impact outcomes and inform decision-making at the patient level and more broadly:
Discovering Threats & Tracking Outbreaks on a Larger Scale
Available cloud-based software makes it quick and easy for healthcare providers to analyse genetic information and discover novel AMR threats and markers. These databases contain genotypic and phenotypic profiles for tens of thousands of clinical isolates compiled over many years or even decades and are continuously updated with new next generation sequencing (NGS) data, pathogen strains and AST information on hundreds of antibiotics.
Machine learning algorithms make it possible to draw on these databases not only to identify which drugs would be effective or ineffective against a particular pathogen, but they also enable healthcare workers to be alerted of AMR threats among their patient populations – both from known and novel pathogens and resistance markers. The ability to detect these threats quickly and access data on treatment outcomes can empower healthcare workers to take action against potential outbreak events more decisively.
Data-sharing technology that brings together individual diagnostic test results, genetic analysis for pathogen identification, treatment outcomes and other related data is essential for advancing the understanding of infectious diseases, and it can serve as a crucial tool for collaboration in the broader fight against infectious diseases. Across states and regions, healthcare facilities can exchange patient and AMR data through these digital surveillance infrastructures, effectively breaking down the ‘records gap’ to share insights on pathogen behavior, treatment outcomes and infection patterns in real time.
- Predicting Genetic Resistance & Antibiotic Susceptibility
The genetic prediction of phenotypic antibiotic resistance based on analysis of whole genome sequencing (WGS) data is becoming increasingly feasible, and it is already accessible to clinical microbiology laboratories. Available databases combine broad resistance profiles with high-quality genetic information to enable accurate pathogen identification and antibiotic resistance detection. Automated interpretive tools which had been lacking or were limited in analysis capability in the past are becoming more refined. State-of-the-art bioinformatics tools and artificial intelligence (AI)-powered algorithms and prediction models are now available that can identify resistance markers and predict antimicrobial susceptibility (AST). Recent studies have showcased NGS workflows capable of identifying pathogens correctly with 100% sensitivity and specificity and antibiotic resistance markers with >95% sensitivity and >99% specificity.
Advancements in rapid molecular diagnostic testing is also expediting diagnostic capabilities. With these solutions, native patient samples can be tested immediately and deliver results in just a few hours, compared to days for conventional culture methods. Once the results are available, they can be added to and analysed against existing databases, providing healthcare workers with access to up-to-date information on genetic resistance markers and diagnostics insights based on data local to their facility as well as cloud-based data.
Supporting Decision-Making
Digital health platforms that rely on NGS to characterise pathogens and resistance markers can also be combined with reporting systems that empower decision-making using artificial intelligence. In these cloud-based clinical decision systems, AMR markers can be linked to treatment responses, guiding healthcare workers as they review individual patient diagnostic test results and look to make responsible treatment decisions.
Cloud-based reporting systems can not only improve individual patient outcomes through broader accessibility to pathogen and resistance marker data, but they can also encourage antibiotic stewardship in healthcare facilities. With more accurate, comprehensive data available on a patient’s condition and the presence of any AMR markers, doctors can administer antibiotics more responsibility, taking measures to cut down on overuse and misuse of invaluable first line antibiotics and preserve newly developed ones for those cases where they are truly indicated.
As the WHO and countless others have noted, AMR is one of the greatest threats to public health and humanity that must be taken seriously. Tackling the challenge of drug-resistant superbugs requires a multi-pronged approach. Developing new antibiotics and maintaining local databases of pathogen information and AMR markers will not be enough. Bioinformatics platforms are evolving to offer data analysis and cloud-based sharing capabilities that can support healthcare professionals working rapidly to identify individual cases of AMR and broader threats to public health.