Precision medicine has become a defining paradigm in modern healthcare, with companion diagnostics (CDx) playing a central role in tailoring therapies to individual patient characteristics 1, 2. While the utility of CDx has been well established in oncology, its role in infectious diseases is only now beginning to be realized. The unique intersection between host genetics, pathogen diversity, and therapeutic variability offers fertile ground for applying CDx approaches in infectious disease management. One of the most striking early examples of CDx in infectious diseases is the use of HLA-B*57:01 genotyping to prevent hypersensitivity reactions to abacavir, an antiretroviral medication used in HIV treatment. This case not only highlights the power of pharmacogenetics in improving safety and efficacy but also serves as a model for the broader application of companion diagnostics in infectious diseases. In this commentary, we explore the current state, challenges, and future potential of CDx in the infectious disease landscape, using abacavir as a case study and reflecting on broader implications for clinical practice and research. A companion diagnostic is a medical test that provides essential information for the safe and effective use of a corresponding drug or therapy. While the term originated in oncology, its broader definition encompasses tests that identify biomarkers—either from the host or the pathogen—that can influence drug response, toxicity, or efficacy. In infectious diseases, this includes pharmacogenetic tests (e.g., host genes affecting drug metabolism or immune response), pathogen genotyping (e.g., identifying resistance mutations), and immunologic profiling. Companion diagnostics in infectious diseases can serve several functions: (1) Predict adverse drug reactions (e.g., hypersensitivity, hepatotoxicity), (2) Guide antimicrobial selection based on pathogen genotype or resistance profile, (3) Stratify patients for immune-modulating therapies (e.g., corticosteroids, monoclonal antibodies), and (4) Predict disease progression or response to vaccines. Despite these broad applications, the routine clinical use of CDx in infectious diseases has lagged behind other specialties, in part due to the complexity of host-pathogen interactions and the logistical challenges of implementing molecular testing in real-world settings. However, advances in genomic medicine, decreasing costs of sequencing technologies, and increased awareness of antimicrobial stewardship are creating new momentum for CDx adoption. Abacavir is a guanosine analog nucleoside reverse transcriptase inhibitor (NRTI) that has been a component of combination antiretroviral therapy (ART) for HIV (human immunodeficiency virus) infection since its approval in the late 1990s. Although abacavir is generally effective and well-tolerated, approximately 5%–8% of patients were found to develop a severe hypersensitivity reaction (HSR) within the first 6 weeks of treatment. This multi-systemic reaction, characterized by fever, rash, gastrointestinal, and respiratory symptoms, can be life-threatening and is difficult to diagnose clinically due to its non-specific presentation. In the early 2000s, research identified a strong genetic association between abacavir HSR and the HLA-B*57:01 allele. Mallal et al. first reported in 2002 that individuals carrying HLA-B*57:01 had a markedly increased risk of abacavir-related HSR, with a positive predictive value of 50% and a negative predictive value approaching 100% 3. This finding was subsequently validated in the pivotal PREDICT-1 trial 4, a prospective, randomized, double-blind study published in 2008. The study demonstrated that prospective screening for HLA-B*57:01 before abacavir initiation reduced the incidence of HSR from 7.8% to 0% 4. Following this breakthrough, HLA-B*57:01 testing became the first FDA-approved pharmacogenetic test in the field of infectious diseases and is now a standard component of HIV treatment guidelines globally. This case has become a canonical example of how CDx can dramatically enhance drug safety and exemplifies the value of integrating host genomic information into infectious disease management. While the abacavir case is the most widely cited, the growing relevance of companion diagnostics in the management of several other infectious diseases is equally illustrative. Hepatitis C virus (HCV) and IL28B: Before the advent of direct-acting antivirals (DAAs), the standard of care for HCV involved pegylated interferon and ribavirin. However, response rates varied widely. Genome-wide association studies revealed that polymorphisms near the IL28B gene (encoding interferon lambda-3) could predict treatment response, particularly in HCV genotype 1 infection. Patients with the CC genotype had significantly higher rates of sustained virologic response (SVR) than those with CT or TT genotypes 5. Although now largely obsolete due to DAAs, this example underscores the utility of host genomic studies in predicting treatment efficacy. Tuberculosis and NAT2 (N-acetyltransferase 2): The NAT2 gene encodes N-acetyltransferase 2, which metabolizes isoniazid, a cornerstone drug in tuberculosis treatment. Genetic polymorphisms in NAT2 affect acetylation status, classifying individuals as slow, intermediate, or fast acetylators. Slow acetylators are at increased risk of drug-induced hepatotoxicity, while fast acetylators may have subtherapeutic drug levels 6. Although NAT2 testing is not yet standard in tuberculosis programs, it represents a potential CDx application that could improve safety and effectiveness. COVID-19 and Host Genetics: The COVID-19 pandemic accelerated interest in host genetic susceptibility and immune response variation. Genome-wide association studies identified several loci associated with disease severity, such as variants on chromosome 3 and in genes related to type I interferon signaling 7. Although no routine CDx tests were adopted during the pandemic, these findings open the door for future use of host biomarkers to guide antiviral or immunomodulatory therapy. The sophistication and ease of use of artificial intelligence (AI) points to its increasing role in the development and implementation of companion diagnostics. The integration of data sets from genomic, imaging and epidemiological sources requires the ability of AI to identify specific signatures relevant to biomarker discovery 8, 9. Further, more accurate pathogen identification as well as characterization of drug-resistance profiles is expected to help clinicians provide more patient-specific therapies. Combining machine learning with an AI model has been shown to improve the accuracy and sensitivity of resistance prediction in Mycobacterium tuberculosis isolates, with clear benefits in making treatment decisions 10. In the management of urinary tract infections, machine learning algorithms using factors such as patient history and resistance profiles cut second-line antibiotic usage by 67% and reduced incorrect prescriptions from over 8% to below 6% 11. The clear benefits of CDx in the management of infectious diseases have spurred efforts to address some of the inherent challenges to their widespread adoption. (1) Cost and Access: Point-of-care technologies such as portable nucleic acid testing devices are increasingly available and affordable, facilitating their adoption in rural and underserved populations; (2) Turnaround Time: Point-of-care microfluidic devices have shown promise in reducing turnaround time to identify bacteria involved in sepsis and biomarkers 12. Further, automation and workflow integration in diagnostic labs also will help reduce the time required for CDx tests; (3) Clinical Integration: This requires a multidisciplinary approach to establish a cross-functional team of clinicians, pathologists and IT specialists to manage clinical workflows, electronic health records, and practitioner training; (4) Ethnic Diversity: Awareness of ethnic as well as gender and age differences has prompted regulatory agencies like the FDA to require diversity plans as part of CDx submissions. Over time this will facilitate the establishment of more generalizable assays. As whole-genome and transcriptomic technologies become more affordable and portable, CDx is poised to transform infectious disease care in several ways. (1) Personalized antimicrobial stewardship: By integrating pathogen resistance profiles with host pharmacogenomics, CDx can guide optimal drug selection and dosing; (2) Host-pathogen interaction profiling: Multi-omics approaches can help classify patients into endotypes, enabling stratified therapy (e.g., for sepsis or viral pneumonia); (3) Vaccine responsiveness: Genetic predictors of vaccine efficacy and adverse events could support tailored immunization strategies; (4) Clinical trials and drug development: CDx will increasingly be used in infectious disease drug trials for patient stratification and enrichment designs. The integration of companion diagnostics into infectious disease management represents a significant advance in precision medicine. The case of HLA-B*57:01 and abacavir hypersensitivity provides a compelling proof-of-concept, demonstrating how genetic screening can prevent severe adverse drug reactions and improve therapeutic outcomes. While challenges remain, the continued development and adoption of CDx tools will be critical in enhancing the safety, efficiency, and equity of infectious disease treatment. As the field moves forward, a multidisciplinary approach, that is, incorporating genomics, microbiology, bioinformatics, and clinical pharmacology, will be essential to realize the full potential of companion diagnostics in infectious diseases. Yi-Wei Tang: conceptualization (lead), writing – original draft (lead), writing – review and editing (equal). Eric von Hofe: conceptualization (supporting), writing – original draft (supporting), writing – review and editing (equal). The authors have nothing to report. The authors have nothing to report. The authors have nothing to report. The authors have nothing to report. Eric von Hofe is the President & Chief Scientific Officer at NuGenerex Immuno-Oncology. Both are excluded from all editorial decision-making related to the acceptance of this article. The authors have nothing to report.
Tang et al. (Sat,) studied this question.