Alfio Ferlito’s research while affiliated with University of Udine and other places

Artificial intelligence (AI) is revolutionizing head and neck oncology, offering innovations in tumor detection, treatment planning, and patient management. However, its integration into clinical practice is hindered by several limitations. These include clinician mistrust due to a lack of understanding of AI mechanisms, biases in algorithm development, and the potential over-reliance on technology, which may undermine clinical expertise. Data-related challenges, such as inconsistent quality and limited representativeness of datasets, further complicate AI’s application. Ethical, legal, and privacy concerns also pose significant barriers. Addressing these issues through transparent AI systems, clinician education, and clear regulations is essential for ensuring responsible, equitable use in head and neck oncology. This manuscript explores the limitations of AI in head and neck oncology.

Background Understanding the relationship between age and diagnostic challenges, as well as outcomes in older patients with head and neck cancer, is crucial. The significance of age as a prognostic factor and how it should influence treatment decisions is debated, highlighting the need for tailored insights to help physicians make more informed choices. Objective This scoping review aims to address the diagnostic challenges/differences and patient outcomes as infection, pain, treatment‐related complications (either postmedical or postoperative), treatment toxicities, disease control, morbidity, and/or survival/mortality in older patients (above the cut‐off age of 60 years) with head and neck cancer. Methods PubMed, Embase, Scopus, Web of Science, and LILACS were searched, as well as gray literature and reference lists of included articles. Results A total of 5029 articles were retrieved and 57 studies were included, among which 5 appraised evidence on diagnostic challenges/differences, 53 addressed clinical (as postoperative complications, the impact of comorbidities in the occurrence of complications and treatment, postoperative delirium, swallowing outcomes, the correlation between age and infections, pain and opioid use, morbidity rates, treatment‐related toxicities), and 26 reported survival. Evidence on diagnostic challenges and outcomes between older and younger patients is mixed, but age, comorbidities, and complications must be considered. While many studies show no significant differences in complications according to age, comorbidities often correlate with worse outcomes and are more frequent in the older group. Despite variable age cut‐offs, survival rates between older and younger patients are similar. Conclusion Treatment decisions should be based on a thorough medical evaluation and patient preferences rather than chronological age alone, as age itself does not account for outcome differences.

Despite decades of progress, laryngeal squamous cell carcinoma (SCC) is still associated with significant morbidity and mortality worldwide. Additional biomarkers are needed to apply precision medicine and predict the clinical course. We reviewed and summarised routinely reported histopathologic features (e.g. subtypes of laryngeal SCC) along with promising potential biomarkers not yet routinely assessed using international guidelines. These include extra- vs intratumoural vascular and perineural invasion, tumour budding, depth of invasion, and tumour-infiltrating lymphocytes. We also address the problem of specimen quality and type (open approach vs endoscopic surgery) and the related limitations. High-risk human papillomavirus infection is another controversial issue to be discussed, being rare in laryngeal SCC, with an indeterminate prognostic significance and less reliable p16 overexpression as a surrogate marker of HPV infection. Further studies are warranted to address the applicability and to see which of the described parameters may help to better stratify patients with laryngeal SCC and should therefore be included in the pathology report.

Objective: To evaluate the impact of postoperative radiotherapy (PORT) on oncological outcomes in node-negative early-stage oral squamous cell carcinoma (OSCC) with perineural invasion (PNI). Methods: A systematic review and meta-analysis was conducted using the PubMed, EMBASE, and Scopus databases for the period from 2000 to 2024. Studies comparing PORT versus observation in pN0 early-stage OSCC with PNI were included. Oncological outcomes assessed included overall survival (OS), disease-free survival (DFS), and local control (LC). A random-effects model was used to calculate log odds ratios, and heterogeneity was assessed using tau², chi², and I² statistics. Results: Seven retrospective studies comprising 522 patients (281 PORT, 241 no-PORT) were included. The 3-year overall survival (OS) was 86.3% in the PORT group compared to 71.1% in the no-PORT group (logOR = −1.03, p = 0.0012), while the 5-year OS was 88.1% versus 77.3% (logOR = −0.97, p = 0.0061). Disease-free survival (DFS) also favored PORT, with 3-year DFS at 86.3% versus 58.1% (logOR = −1.19, p < 0.001) and 5-year DFS at 86.3% versus 55% (logOR = −0.78, p = 0.003). Local control (LC) was higher in the PORT group, with 3-year LC rates of 89% compared to 72.2% in the no-PORT group (logOR = −1.13, p = 0.025). Conclusions: PORT significantly improves OS, DFS, and LC in node-negative early-stage OSCC with PNI as the sole adverse feature.

Sinonasal tumors with orbital involvement present significant management challenges due to the complex anatomy of the orbital and sinonasal complex and the potential for severe tumor-and treatment-related morbidities. This comprehensive review examines evidence-based decision-making guidelines in the context of treating sinonasal malignancies involving the orbit. We evaluate pre-treatment assessment strategies, discuss the indications for orbital preservation versus its exenteration, explore Adv Ther the role of neo-adjuvant therapies, and analyze reconstruction techniques with ensuing rehabilitation approaches and quality of life. Our review underscores the need for a patient-centered multidisciplinary approach that optimizes treatment outcomes and quality of life for patients suffering from such advanced sinonasal malignancies with orbital involvement.

Post/Long COVID (syndrome) is defined as a condition with symptoms persisting for more than 12 weeks after the onset of SARS-CoV-2 infection that cannot be explained otherwise. The prevalence of self-reported otorhinolaryngological Post/Long COVID symptoms is high. The aim of this review was to analyze the current literature regarding the actual prevalence, knowledge of the etiopathology, and evidence-based treatment recommendations of otorhinolaryngology-related Post/Long COVID symptoms. A systematic literature search of articles published since 2019 in PubMed and ScienceDirect was performed and resulted in 108 articles. These were the basis for this review and formed a comprehensive series of consented therapy statements on the most important of otorhinolaryngology-related Post/Long COVID symptoms. Otorhinolaryngological symptoms did not appear isolated but as part of a multi-organ syndrome. Self-reported otorhinolaryngology-related Post/Long COVID symptoms were often not confirmed by objective testing. The confirmed prevalence estimated for anosmia, dysgeusia, cough, facial palsy, hoarseness/dysphonia, acute hearing loss, tinnitus, and vertigo/dizziness was about 4%, 2%, 4–19%, 0%, 17–20%, 8%, 20%, and 5–26%, respectively. There are manifold theoretical concepts of the etiopathology of different symptoms, but there is no clear evidence-based proof. This certainly contributes to the fact that there is no effective specific treatment option for any of the symptoms mentioned. Healthcare pathways must be established so that otorhinolaryngological Post/Long COVID symptoms can be recognized and evaluated and otorhinolaryngologists can provide counseling. This would also help to establish and selectively include patients in clinical trials investigating specific therapeutic concepts.