Artificial intelligence’s potential in tailoring prescription of biologic therapy for chronic rhinosinusitis

To the Editor: We read with great interest the recent article by Goktas et al,1 suggesting a possible application of artificial intelligence (AI) in the field of Allergy and Immunology. We find that AI could also have an impactful use in the prescription of biological drugs in chronic rhinosinusitis (CRS). Artificial intelligence has emerged as a transformative technology, revolutionizing diagnostics and patient care in the field of medicine. It encompasses a range of computational algorithms that enable machines and computers to simulate human cognitive functions, including problem-solving and learning from experience. Artificial intelligence’s contributions to precision medicine, medical imaging interpretation,2 drug discovery,3 and personalized treatment optimization have been significant. Recently, AI has shown promise in tailoring treatment strategies for complex diseases by leveraging large datasets and sophisticated algorithms. This capability of AI holds great potential for improving the management of chronic conditions like CRS, which face challenges owing to disease heterogeneity and varied treatment options. Chronic rhinosinusitis is a complex disease characterized by diverse underlying pathophysiological mechanisms and varying treatment responses among patients. Notably, the heterogeneity of CRS, as classified in EPOS2020 based on clinical features, inflammatory patterns, and anatomical factors, presents a significant treatment obstacle.4 Particularly, the focus lies on type 2 endotypes, for which different biological drugs such as dupilumab, mepolizumab, and omalizumab have been introduced in recent years. Tailoring treatment to individual patient characteristics can enhance efficacy, reduce unnecessary costs, and minimize potential side effects. However, achieving personalized medicine in CRS poses several challenges because each phenotype may require a specific approach. This is where AI can play a crucial role by identifying patterns and strategies to develop more accurate treatment approaches. Artificial intelligence–powered decision support systems have the potential to provide clinicians with personalized treatment recommendations, increasing the likelihood of successful treatment outcomes. These systems should consider various factors such as anamnestic data, clinical features (eg, nasal polyp score, circulating and polyp biopsy eosinophil count, modified Lund-Kennedy score, Sino-Nasal Outcome Test-22 score), and comorbidities like asthma, atopic dermatitis, and pathologies associated with ipereosinophilia. Ongoing research on potential biomarkers such as interleukin-13,5 bone morphogenetic protein-2,6 interleukin-1β,6 and eosinophil count in nasal swab cytology7 also holds promise for their inclusion as valuable diagnostic tools. Furthermore, AI can be instrumental in identifying novel biomarkers that correlate with treatment response over time. Traditional approaches to identifying biomarkers are limited by a small set of candidates. In contrast, AI algorithms can analyze vast amounts of patient data, including genomics and proteomics, to uncover new relationships between biomarkers and treatment outcomes. This has the potential to significantly enhance overall treatment efficacy. In conclusion, AI has already demonstrated its value in medicine and its potential in tailoring the prescription of biologic drugs for CRS is substantial. Integrating AI-driven decision support systems and predictive models into clinical practice can improve treatment outcomes, enhance patient care, and optimize resource allocation. Collaboration among clinicians, researchers, and data scientists is crucial to fully harness the potential of AI and translate it into tangible benefits for patients with CRS.