Leveraging new and upcoming technologies against AMR
Healthcare has evolved drastically in the past few decades owing to the scientific and technological developments. Today, technology has become an integral part in identifying and developing solutions against various healthcare challenges, including the silent epidemic of antimicrobial resistance (AMR).
AMR is a multi-faceted global problem, one that requires multi-pronged measures. In recent years, advancements in therapeutics, diagnostics, prevention, and surveillance technologies have catapulted manifold, this aspect needs to be integrated and leveraged in our efforts to effectively achieve the mandates of the Global Action Plan [GAP] and the National Action Plan [NAP] against AMR. The scientific fraternity and public health experts agree that incorporating these technologies and integrating associated scientific understanding is going to be crucial in combating AMR going forward.
Current antimicrobials have been derived primarily from bacteria and plants. However, since the 1970s the discovery of newer antibiotics has stagnated. Emerging technologies are now allowing us to tap newer sources, like marine microbiome, that potentially enable exploration for newer antimicrobial molecules. Genomic and metagenomic analysis of microbes, along with synthetic biology approaches, open broader avenues for antimicrobial discovery in a shorter timeframe. For example, genomic mining tools, like antiSMASH and PRISM, have helped in identifying candidate antibiotic biosynthetic gene clusters [BGC] that may reveal new antimicrobial leads.
Microfluidics and microfabrication technology have replicated complicated growth medium compositions and conditions required for the bacteria to be maintained outside its natural environment. This is important to grow, scale-up, investigate and sustain these bacterial cultures in the laboratories which can potentially speed up the antimicrobial discovery process. The novel antibiotic, teixobactin, was discovered in Eleftheria terrae through this micro cultivation method.
Popular technologies like CRISPR/ Cas9 has be used in lab settings to facilitate targeted delivery of antimicrobials. To illustrate this further, a 2019 study at the Mississippi State University demonstrated the use of bacteriophage gene-editing that enhanced the targeted action and bactericidal activity of the antibiotic Fosfomycin against bacteria causing bone infections. However, further investigations are needed to check for safety of this approach to assess its potential integration in the broader drug delivery pipeline.
In last couple of years Artificial Intelligence has found tremendous applications in development of newer drugs to being incorporated in the tests and devices. A group of MIT Researchers have identified a new compound called Halicin using machine learning algorithm. The molecule was effective against C. difficile, A.baumanii and M.tuberculosis. AI & ML based algorithms can potentially reduce the cost and time involved in antimicrobial discovery and development.
Antimicrobial peptides are being explored as new therapeutic molecules to treat infections. AI models have also helped in modelling interactions between the amino acids via neural networks and have increased the probability of selecting the most promising drug candidates.
In addition to use of technology in new antimicrobial discovery and better delivery systems, another aspect that needs to be focused is evidence-based use of the antimicrobials. Presently, the indiscriminate use of antibiotics has been repeatedly attributed to the lack of appropriate diagnostics at point-of-care, to assess the pathogen type and the resistance pattern, to guide clinical decision. Diagnostic intervention in the clinics will be invaluable in prescription of appropriate antibiotics, thereby potentially decreasing the treatment time and in controlling further spread in case of an infectious pathogen.
Culture-based antibiotic susceptibility test takes somewhere between 24–72 hrs to deliver the results. Technological advancement in culture-free assays has enabled development of rapid antibiotic sensitivity testing (AST) that provides results under 30 minutes to few hours. Next Generation Sequencing (NGS) based tests profile markers across the entire genome, thereby providing highly accurate results in a shorter time period. In a German clinical study conducted at Heidelberg University Hospital diagnosis was carried out for identifying infectious microbes in septicemia patients. The turnaround time for this test was around 16h for sequencing which gives a competitive edge over blood culture tests which typically take 24 hrs for results and a further 5 days of incubation.[i]
Integration of technologies like biosensor, microfluidics and chip-based technologies can address challenges of low initial pathogen load in clinical sample for testing and the presence of contaminants that may hamper testing efficiencies. Rapid results from low volume and low-quality samples can direct prescription decisions in the clinics, especially in low-resource settings, and thereby contribute to minimizing antibiotic consumption in these countries.
AI based apps are now capable of analysing disk-diffusion antibiograms. This simplifies the measurement task and is easily adaptable in resource-limited settings. Using antimicrobial susceptibility datasets, AI models and machine learning techniques can predict resistance to specific antibiotics. In a 2018 study, An AI tool could accurately predict point mutation concerning resistance to first-line drugs from data obtained from 10,209 of M. tuberculosis collected from 16 countries. [iii] Such methods can effectively drive clinical management of infections and go a long way in unnecessary prescription of antibiotics.
Digital Chest X-Ray Radiography [CXR] and computer aided solutions for Tuberculosis Diagnosis which utilises AI/ML algorithms is now being scaled by various public and private agencies to scan large numbers of Chest X-Rays. These algorithms can pick up subtle information from CXRs which might have been missed by the technician. In case of analyzing cultures, standard microscopy may not adequately stain microbes in a low volume or a low-bacterial load samples, especially in immune-compromised patients and pediatric patients. However, AI/ML algorithms can pick these minute details from the culture plates or slides and identify the infection. This reduces the number of false negatives that might be reported from microbiology labs, in turn bringing more patients into formal care who might have otherwise been untreated. The ultra-portable models and offline CAD solution has enhanced the screening numbers and made it available in remote areas. Additionally, they are also in the works of building software and mobile application to leverage the data systems from various local and global sources for AMR surveillance. AI-driven healthcare applications are immense that can improve health-outcomes but are highly dependent on the quality of data available.
Preventive approaches and interventions are the first line of action required in curtailing disease and its spread. Investing in preventives can reduce the disease burden in a population, and thus reduce the demand on antimicrobials and its indiscriminate use. This is probably the most important AMR intervention that can save lives and resources downstream. Ranging from simple hygiene and sanitation practices to advanced developments in vaccines and biotherapeutics, the strategies are now available to address different facets of prevention and control.
Hygiene, Sanitation & Infection Control: Technology is aiding in prevention of Hospital-acquired infections (HAI) such as GloGerm which uses an invisible fluorescent marker to track high-touch surfaces. This helps in tracing and maintaining sterile environment. Wireless Sensor Network has also been a gamechanger in establishing hand-hygiene protocols in hospital environment. For example, ZigBee technology has a proprietary system that automatically tracks hand hygiene dispenser usage where workers wear pager-sized badges. Each use is sensed through badge at the station and the data is monitored for healthcare provider compliance. C-CAMP supported start-up, Biomoneta, has developed Zebox to lower the spread of infection by trapping and killing airborne microbes within 20 mins. Coeo Labs, on the other hand, are building an AI-based device ‘VAPCare’. This sensory device sucks out secretions from the oral pathway and reduces the chances of acquiring Ventilator-Associated Pneumonia.
Vaccines: Every year 2–3 million deaths are averted through vaccine inventions, especially in low- and middle-income countries (LMICs) where the healthcare infrastructure is sparse and fragmented. By preventing infections and disease progression, vaccines reduce the chances of microbes from mutation and thereby developing resistance. Also, disease prevention leads to reduced antibiotics pressure in the community, inadvertently lowering the risk of AMR as well.
Traditionally viewed as a long and arduous process, involving the use of attenuated or killed microbes, or alternately culturing microbes and subsequently purifying the desired subunits, recent advancements in vaccine development have made it easier to fast-track identification of potential candidates and conduct the necessary clinical validations.
· It is now possible to design immunogenic microbial subunits separately instead of first growing the microbes and then purifying the immunogenic components. This has brought down the cost and the time required for producing these at scale.
· Through Reverse vaccinology, potential antigen encoding genes can be selected and tested for vaccines in invitro and in vivo preclinical trials. This approach has successfully produced Meningococcus B vaccine and that is in trial for vaccines against E. coli and P. aeruginosa.
· Traditionally, glycoconjugate vaccine manufacture involved extensive multi-step bio-chemical processes. Bioconjugation has now allowed glycoconjugate vaccine production through a single fermentation step using enzymes like oligosaccharyltransferases.
Biotherapeutics: It is an emerging preventive strategy. Monoclonal antibodies [mAbs], bacteriophage, and microbiota-based strategies are promising approaches.
· Monoclonal antibodies can be given as a prophylactic treatment or combined with antibiotics as adjuvants.
· On the other hand, Phage therapy uses lytic phages which selectively target pathogen and eliminate them. These are target-specific thereby minimising the disruption of gut microflora. Also, phage cocktails when applied on dwelling devices like catheters can inhibit bacterial growth, thus preventing chances of infection.
· Microbiota-based interventions have been picking up in recent years, with an objective to reduce recurrence of infections particularly for C.difficile.
Stewardship and Surveillance
As emphasized in GAP, to tackle antimicrobial resistance, optimising the use of antimicrobial medicines is paramount. Through synchronized stewardship programmes and surveillance, the goal is achievable. Stewardship program assists in regulating the use of antimicrobials and surveillance provides insights into the current trends of antimicrobial consumption.
Digital tools like Clinical Decision Support Systems [CDSS], and Electronic Medical Record [EMR] can doubly aid in implementing seamless stewardship initiatives. CDSS can provide alerts and reports on several parameters like time period of patients on antimicrobials, patients on inappropriate therapy which makes potential interventions easier. In a three-month trial of a Surveillance software [Sentri7] conducted at UnityPoint Health-St. Luke’s, a 151-bed community hospital in Sioux City, IA, USA, it was observed that there was an 87% improvement post-implementation in antimicrobial management like antibiotic dose adjustments, tracking duration of antimicrobial therapy, etc. The software provides information on lab culture, susceptibility results. A total of 572 patients and 774 alerts were assessed.[iv]
ICMR is developing i-AMRSS software for management and analysis of AST and monitoring antimicrobial consumption data with hybrid model data entry through online and manual entry with distribution and accessibility permission at multiple levels. The real-time analysis of data would provide warnings that can be utilised to monitor the infection control practices. They also aim to identify local geographical hot-spots and provide alerts, sample wise comparison, etc.
Whole genome sequencing allows detection and tracking of specific resistant markers that can be used to track emerging AMR patterns. Advancing bioinformatic tools now allow for early mapping of resistance and prediction of emerging trends. Other than this, genomic surveillance can identify gene variants that are attributed to high microbial transmissibility, infectivity or severe clinical manifestations. This understanding can enable local public health authorities to line up appropriate upstream and downstream interventions to control the emerging healthcare situation in time.
Enabling emerging technologies against AMR
For technologies to thrive, they must be supported and scaled appropriately for on-ground impact. Many organisations are providing a platform for enhanced discovery and access to these new technologies.
CARB-X, a Boston-based global non-profit partnership is leveraging multiple newer technologies to accelerate the development pipeline of new antibiotics, vaccines, rapid diagnostics and other products to diagnose, treat and prevent infections caused by antibiotic-resistant bacteria. Their funded portfolio includes rapid point of care diagnostic tests like the ones developed by Novel Microdevices that aspire to provide results in less than 20–25 minutes, in time to influence critical treatment decisions. CARB-X has also invested in high-value innovations of non-traditional treatment modalities under development by Phico Therapeutics. Phico has engineered bacteriophages to deliver directly to the lung antibacterial small acid-soluble spore proteins (SASPs) for treating ventilator-associated bacterial pneumonia. Affinivax under CARB-X is developing a multivalent vaccine for infections caused by Staphylococcus aureus, using their proprietary Multiple Antigen Presenting System (MAPS) technology.
In India, Centre for Cellular and Molecular Platforms (C-CAMP) acts as an enabler for cutting-edge life-science research and innovations. Through the state-of-the-art technology platforms, C-CAMP is fostering and amplifying innovations under three verticals of Preventatives, Diagnostics and New Drugs. C-CAMP supported SpotSense has built a diagnostic pacifier platform for saliva-based screening of Neonatal Sepsis. Module Innovations through their platform technology aim to make rapid point of care diagnostics for microbial infection. Their phenotype test, ASTSense, is world’s first 2-hour antibiotic susceptibility test of UTI causing uropathogens. Adiuvo Diagnostics is a deep tech company focused on realizing rapid, diagnostic solutions leveraging multi-spectral imaging and AI. Their first proprietary platform “Illuminate” is the world’s first AI integrated device for rapid pathogen assessment in less than 2 min using a novel “Optical biopsy” technique which is currently commercialized in India. RapidDx Technologies, a young company working on accelerating antibacterial susceptibility testing have developed patented FRET-based sensors to test AST in less than 5 hours.
Last mile access
In a 2019 systematic analysis of global burden of antimicrobial resistance, the highest rate of death due to infections were observed in sub-Saharan Africa and South Asia. Scarcity of lab-infrastructure, and unavailable microbial testing for informed treatment decisions remain one of the key drivers. Technology enabled interventions for AMR need to cater to these vulnerable populations in resource-limited settings.
For the technological advancements in AMR to be impactful in reducing the disease burden in these low- and middle-income countries [LMICs], they need to be accessible, cost-efficient and easy-to-use as well as need to be supported by enabling polices and regulatory environment.
[ii] Pascucci M, Royer G, Adamek J, et al. AI-based mobile application to fight antibiotic resistance. Nat Commun. 2021;12(1):1173. Published 2021 Feb 19. doi:10.1038/s41467–021–21187–3
[iii] CRyPTIC Consortium and the 100,000 Genomes Project, Allix-Béguec C, Arandjelovic I, et al. Prediction of Susceptibility to First-Line Tuberculosis Drugs by DNA Sequencing. N Engl J Med. 2018;379(15):1403–1415. doi:10.1056/NEJMoa1800474
[iv] Bremmer DN, Trienski TL, Walsh TL, Moffa MA. Role of Technology in Antimicrobial Stewardship. Med Clin North Am. 2018 Sep;102(5):955–963. doi: 10.1016/j.mcna.2018.05.007. Epub 2018 Jul 14. PMID: 30126584.
Disclaimer: The blog is a compilation of information on a given topic that is drawn from credible sources; however this does not claim to be an exhaustive document on the subject. It is not intended to be prescriptive, nor does it represent the opinion of C-CAMP or its partners. The blog is intended to encourage discussion on an important topic that may be of interest to the larger community and stakeholders in associated domains.