Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 26 appraisals
Sleep
AI meets sleep surgery: assessing drug-induced sleep endoscopy interpretation with a large language model.
STUDY OBJECTIVES: To evaluate the performance and safety of a large language model in interpreting drug-induced sleep endoscopy (DISE) videos and providing treatment recommendations for obstructive sleep apnea, compared to expert human raters and the contemporaneous clinical report. METHODS: This prospective, blinded study included 16 adults undergoing drug-induced sleep endoscopy at a tertiary academic center. For each case, an anonymized procedural video, clinical vignette, and examination findings were independently reviewed by two sleep surgery experts, a senior otolaryngology resident, and a large language model. All the raters assessed the video quality, airway maneuvers, airway collapse at each anatomical site, and recommended therapy. Concordance with the clinical reference standard was evaluated using Cohen's kappa and intraclass correlation coefficients. Safety and reproducibility were assessed through subgroup and error-type analyses. RESULTS: Human raters demonstrated perfect or near-perfect agreement with the reference standard for the velum, oropharynx, tongue base, and jaw thrust response, and moderate agreement for epiglottic collapse. The large language model matched expert performance for all domains except the epiglottis, where moderate agreement was observed. Model-generated treatment recommendations were safe, consistent with accepted clinical practice, and highly reproducible between independent runs. No unsafe or discordant recommendations were identified. CONCLUSIONS: In this single-center study of 16 patients, a large language model accurately interpreted DISE videos and generated safe recommendations consistent with accepted clinical practice, approximating expert performance in most domains. Larger multicenter cohorts are needed to validate these findings and confirm generalizability. Statement of Significance This study demonstrates that artificial intelligence can interpret complex airway videos in sleep surgery and provide safe, expert-level treatment advice. By comparing the performance of a large language model with that of experienced clinicians, our findings suggest that advanced technology can help standardize decision-making in a highly subjective area of care. This work highlights the promise of artificial intelligence as an adjunct to clinical judgment, especially in settings where expert access is limited. However, important questions remain about the use of artificial intelligence in challenging cases and its integration into real-world practice. Future research should focus on larger, more diverse groups of patients and on ensuring ongoing oversight and safety.
14 July 2026
Read appraisal →Signal transduction and targeted therapy
Liposomal doxorubicin plus nab-paclitaxel with/without chemoradiotherapy in head and neck adenoid cystic carcinoma: single-arm phase II study
Given no standard therapies, this single-arm, open-label, multicenter phase II study was conducted in unresectable locally advanced or recurrent/metastatic adenoid cystic carcinoma of the head and neck (LA or R/M ACCHN). Patients received chemotherapeutic nanomedicines (20 mg/m2 liposomal doxorubicin plus 120 mg/m2 nab-paclitaxel on days 1 and 8) for three 21-day cycles. Subsequently, patients eligible for radiotherapy received concurrent chemoradiotherapy (cCRT; helical tomotherapy or intensity-modulated radiotherapy and nab-paclitaxel), whereas those ineligible for radiotherapy continued 2-3 cycles of original chemotherapy. Between May 27, 2021, and November 13, 2024, 31 enrolled patients were evaluable for efficacy and safety. The primary endpoint objective response rate (ORR) was 90.3% (95% CI, 74.2-98.0%), with 15 (48.4%) complete responses and 13 (41.9%) partial responses, meeting the prespecified primary endpoint. The median duration of response was 18.5 (95% CI, 0-45.8) months. With a median follow-up of 15.7 months by data cutoff (June 25, 2025), the median progression-free survival (PFS) was 25.7 (95% CI, 6.5-44.8) months; the 1-year PFS was 77.1%. The median overall survival (OS) was not reached (NR), with a 1-year OS of 90.0%. Patients who received cCRT had a prolonged PFS (NR vs. 12.8 months) and OS (NR vs. 21.2 months) compared to those who did not. Grade 3 treatment-related adverse events (TRAEs) occurred in 8 (25.8%) patients. No ≥grade 4 TRAEs or treatment-related deaths occurred. This study demonstrated tumor responses among the highest reported to date, with a manageable safety profile in unresectable LA or R/M ACCHN. Chemotherapeutic nanomedicines, especially sequential cCRT for appropriate patients, may represent an effective and safe option.
14 June 2026
Read appraisal →Journal of robotic surgery
Transoral robotic surgery using the da Vinci single-port system: current evidence and clinical indications - a systematic review and meta-analysis
The DaVinci Single-Port (SP) transoral robotic surgery (TORS) platform was developed to overcome spatial constraints and instrument collisions encountered with multiport systems in head and neck surgery. We systematically reviewed and meta-analyzed perioperative performance, safety, functional recovery, and oncologic outcomes of SP TORS. Following PRISMA 2020, we searched PubMed/MEDLINE, Embase, Cochrane Library, Web of Science, and the first 25 pages of Google Scholar through January 1, 2026. Peer-reviewed English-language clinical and cadaveric studies evaluating SP TORS were included. Three reviewers screened studies and extracted data. Random-effects meta-analyses were performed where appropriate. Fourteen studies (2019-2025) including 479 patients met inclusion criteria. Patient-level evidence was concentrated in oncologic cohorts, with additional smaller clinical applications in obstructive sleep apnoea-related tongue-base surgery (37 patients), hypopharyngeal foreign body removal (1 patient), parapharyngeal space surgery (2 patients), and SP-assisted reconstruction (3 patients), as well as preclinical feasibility studies. No study reported intraoperative conversion to multiport or open surgery. Pooled estimates were console time 60.04 min (95% CI 29.56-90.51; I² = 99.1%), docking time 10.13 min (0.47-19.79; I² = 97.9%), operative time 57.72 min (37.23-78.21; I² = 0.0%), length of stay 10.63 days (2.43-18.82; I² = 99.8%), and blood loss 24.16 mL (- 6.95 to 55.27; I² = 86.8%). Bleeding was the most consistently reported surgical complication, with major bleeding requiring operative management reported in larger cohorts but remaining uncommon. Functional recovery suggested early oral intake, although timing was inconsistently reported. Margin status was predominantly R0 in several oncologic cohorts, although one matched cohort reported a lower R0 rate. SP TORS appears feasible across indications, but current evidence is limited by nonrandomized designs, inconsistent long-term reporting, and between-study heterogeneity.
14 June 2026
Read appraisal →Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy
Multivariate feature analysis of early-stage laryngeal cancer serum components using surface-enhanced Raman spectroscopy
Laryngeal cancer is a common head-and-neck malignant tumor with geographically variable incidence. Its lack of specific early clinical symptoms often causes missed diagnosis, leading to most cases being identified at intermediate/advanced stages and thus poor treatment efficacy and prognosis. Early, accurate screening for high-risk populations is therefore critical to improving survival. This study systematically investigates the clinical application value of surface-enhanced Raman spectroscopy (SERS) combined with deep learning models for serum screening in early laryngeal cancer. It specifically examines the discriminative performance of one-dimensional convolutional neural network (1D-CNN) models, including CNN-attention with a fused attention mechanism and CNN-baseline without an attention mechanism, and compares these with traditional machine learning models.Serum samples were collected from three groups: Early laryngeal cancer, vocal fold polyps, and healthy controls. SERS was used to obtain serum molecular vibration fingerprint spectra. After spectral preprocessing, five models (CNN-attention, CNN-baseline, SVM, AdaBoost, KNN) were established to discriminate the three groups and compare performance. Leave-one-patient-out cross-validation (LOPO-CV) results indicate that the CNN-attention and CNN-baseline models significantly outperform other models in discrimination efficacy. Specifically, the CNN-attention model achieves a patient-level accuracy of 89.33%, while the CNN-baseline model reaches 88.67%. SVM, KNN, and AdaBoost had inferior performance. This study confirms that serum SERS-based 1D-CNN provides an efficient, accurate method for early laryngeal cancer screening, laying a foundation for clinical translation and offering new technical support for early screening.
7 June 2026
Read appraisal →Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
Structured yoga program as supportive care to enhance quality of life in head and neck cancer patients undergoing radiotherapy: a prospective matched controlled study
PURPOSE: Head and neck cancer (HNC) patients undergoing radiotherapy often experience significant deterioration in quality of life (QoL) due to physical and psychological challenges. This study evaluates the effect of a structured yoga program-including physical postures, breathing exercises, and meditation-on QoL among these patients. METHODS: A prospective matched controlled nonrandomized study was conducted at a Tertiary Care Centre in North East India. Then, 140 patients undergoing definitive or adjuvant radiotherapy were equally divided into experimental (yoga intervention) and control groups. QoL was assessed using the EORTC QLQ-C30 questionnaire at baseline, end of treatment, and two months posttreatment. The yoga intervention consisted of thrice-weekly sessions for two months. All primary statistical analysis was done using nonparametric methods. RESULTS: Statistically significant improvement was seen in the experimental group in global health status/QoL (p-value < 0.001), pain (< 0.001), financial difficulties (< 0.001), emotional functioning (0.001), and fatigue aspect (0.024). CONCLUSION: A structured yoga program significantly improved key QoL domains in head and neck cancer patients receiving radiotherapy. Yoga can be considered an effective supportive therapy in this population.
30 May 2026
Read appraisal →Otolaryngologia polska = The Polish otolaryngology
Leveraging large language models to audit perioperative total laryngectomy care: anesthetic, nutritional, and complication profiles in a real‑world cohort
<p><strong>Introduction: </strong>Total laryngectomy (TL) remains the standard treatment for advanced la‑ ryngeal cancer, and its perioperative management requires close collaboration within a multidisciplinary team, including anaesthesiologists and clinical nutritionists.</p><p><strong>Aim:</strong> The aim of this study was to analyse the perioperative course of patients un‑ dergoing TL from both an anaesthesiologist's and a nutritionist's perspective, with a particular focus on the feasibility and performance of structured data extraction from electronic medical records using large language models (LLMs).</p><p><strong>Material and methods:</strong> We performed a single‑centre retrospective analysis of 43 consecutive patients undergoing TL (ICD‑9 codes 30.32, 30.39, 30.41, 30.49) be‑ tween January 2024 and April 2026. Clinical data, including physician notes (n = 845), nursing notes (n = 2578), discharge summaries (n = 173), and consultations (n = 123), were subjected to structured extraction using large language models (gemma-3-31Bit/gemma-4-26B-it). Demographics, tumour stage, length of stay (LOS), periopera‑ tive nutrition, pharmacotherapy, postoperative complications, temporal laboratory trends, and rehabilitation were analysed.</p><p><strong>Results: </strong>The cohort comprised 35 men (81%) and 8 women (19%), with a mean age of 66.4 9.8 years; pT4a (49%) and pN0 (82%) predominated. Median LOS was 15.0 days (IQR 13.1-16.5). Enteral nutrition via nasogastric tube was used in 95% of patients, and oral intake was initiated at a median of postoperative day 13. Multimodal anal‑ gesia based on intravenous paracetamol (60%), ketoprofen (47%), and metamizole (40%) was routinely employed, while strong opioids were used in only 5% of pa‑ tients. Postoperative complications occurred in 22.5% of cases, with pharyngocuta‑ neous fistula (PCF) being the most frequent (12.5%). In‑hospital mortality was 0%, and overall mortality during a median 12‑month follow‑up was 2.3%. CRP peaked on postoperative day 3 (mean 114 mg/L). Pharmacologic VTE prophylaxis was admi‑ nistered in 65% of patients.</p><p><strong>Conclusions:</strong> In TL patients, enteral feeding forms the backbone of nutritional sup‑port, and multimodal analgesia provides effective pain control, whereas VTE pro‑phylaxis and other modifiable elements of perioperative care warrant optimization within prehabilitation and ERAS-based protocols.</p><p>&nbsp;</p>.
26 May 2026
Read appraisal →Head and neck pathology
Digital Pathology in Head and Neck Squamous Cell Carcinoma: Translational Advances and Clinical Integration for Pathologists, Oncologists, and Surgeons
PURPOSE: Artificial intelligence (AI)-enabled digital pathology has advanced rapidly in head and neck squamous cell carcinoma (HNSCC), but its readiness for clinical implementation remains uncertain. This review evaluates the translational maturity of AI-based digital pathology applications in HNSCC and their proximity to routine clinical use. METHODS: A structured literature search of PubMed and Scopus identified studies published between January 2020 and March 2026. Studies were included if they applied AI or machine learning to histopathologic or whole-slide images in HNSCC or related oral malignancies and reported diagnostic, prognostic, biomarker, or treatment-response outcomes. Given the narrative scope of this review, representative studies aligned with key clinical domains were selected for qualitative synthesis. A six-stage translational maturity framework was used to categorize applications based on development stage, validation, and clinical integration. RESULTS: Of 444 identified records, 358 unique studies were screened, from which representative HNSCC-specific AI digital pathology studies were selected for detailed analysis. Most applications-including diagnostic classification, dysplasia grading, prognostic modeling, and therapy-response prediction-remain at early translational stages (Stage 0-1), typically limited to retrospective, single-center cohorts with minimal external validation. Biomarker quantification and HPV prediction show relatively greater maturity, with some studies approaching Stage 2, reflecting external validation and early prospective evaluation. However, calibration, decision-curve analysis, and outcome-linked endpoints are rarely reported. No HNSCC-specific AI pathology tools have demonstrated outcome-linked deployment or achieved regulatory-grade implementation, and evidence for workflow integration, clinical decision impact, and patient-centered outcomes remains limited. CONCLUSION: AI-enabled digital pathology in HNSCC remains an early-stage field. Despite promising technical performance, most applications lack the prospective validation, workflow integration, and outcome-based evidence required for clinical adoption. Progress will depend on multi-institutional prospective studies, standardized reporting of clinical utility, and demonstration that AI-assisted decisions improve patient management and outcomes.
26 May 2026
Read appraisal →Journal of robotic surgery
Oncologic, functional, and economic outcomes of transoral robotic surgery for HPV-associated oropharyngeal squamous cell carcinoma: a systematic review and meta-analysis of pathology and treatment strategies
The incidence of human papillomavirus (HPV)-positive oropharyngeal squamous cell carcinoma (OPSCC) has increased substantially. Transoral robotic surgery (TORS) provides detailed pathological assessment that may support risk-adapted treatment strategies; however, integrated evidence on oncologic, functional, and economic outcomes remains limited. To systematically evaluate oncologic, functional, and economic outcomes of TORS-based treatment strategies in patients with HPV-positive OPSCC. A systematic review and meta-analysis were conducted in accordance with PRISMA 2020 guidelines. PubMed, Wiley Online Library, MDPI, Cureus, and medRxiv were searched for studies published between 2019 and 2025. Pre-specified subgroup meta-analyses (randomized/clinical trial versus observational studies, and primary versus salvage settings) were performed using random-effects models for overall survival (OS), postoperative bleeding, tracheostomy rate, and long-term gastrostomy dependence. Twenty-two studies (20 clinical and 2 economic) were included, with 14 contributing to quantitative synthesis. The pooled 2-year OS was 94% (95% CI: 0.90-0.98), the 3-year OS was 96% (95% CI: 0.92-0.99), and the 5-year OS was 92% (95% CI: 0.90-0.95). In the salvage setting, the pooled 2-year OS was 76% (95% CI: 0.71-0.80), with no observed statistical heterogeneity, although based on a limited number of studies. Subgroup analyses indicated higher survival estimates in randomized/clinical trial settings compared with observational cohorts. Functional outcomes were generally favorable, with pooled postoperative hemorrhage of 7% (95% CI: 0.04-0.11), tracheostomy rate of 9% (95% CI: 0.03-0.14), and long-term gastrostomy dependence of 2% (95% CI: 0.00-0.03).Across included studies, TORS provided detailed pathological information that may support risk stratification and inform adjuvant treatment decisions; however, the extent of treatment de-escalation varied across studies. TORS appears to be an effective treatment option in selected patients with HPV-positive OPSCC, with generally favorable oncologic and functional outcomes. However, interpretation of these findings is limited by heterogeneity in study design and patient populations. Further prospective studies are required to clarify its role in treatment de-escalation and long-term outcomes.Registration: PROSPERO (CRD420251240194).
22 May 2026
Read appraisal →Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
Nutritional prehabilitation in patients with head and neck cancer: an evidence mapping analysis
BACKGROUND: Head and neck cancer patients are often at high risk of malnutrition due to the disease and its treatment, and pretreatment nutritional prehabilitation is an important strategy for improving prognosis; however, the existing evidence still needs to be systematically synthesized. PURPOSE: To systematically describe and evaluate the evidence related to pretreatment nutritional prehabilitation in patients with head and neck cancer through evidence mapping to provide a basis for clinical practice. METHODS: Computerized searches of CNKI, Wan-fang, VIP, SinoMed, PubMed, Web of Science, Cochrane Library, Embase, and CINAHL were performed from the time of database construction to May 2025, to include systematic review/meta-analysis of nutritional prehabilitation of patients with head and neck cancer, randomized controlled trials, and observational studies. The Risk of Bias Tool, Newcastle-Ottawa Scale (NOS), and AMSTAR-2 scale were used to evaluate the methodological quality of the included studies, respectively, and the current status of the studies was demonstrated using a combination of text and graphs. RESULT: Then, 6 systematic evaluations/meta-analyses, 13 randomized controlled trials, and 4 observational studies were finally included. The largest number of studies was published in China, and all study populations consisted exclusively of patients with head and neck cancer. Interventions include pretreatment immunonutritional interventions, nutritional risk screening-guided interventions, standard enteral nutritional support, disease-specific nutritional formula support, nutritional counseling, and pretreatment swallow training. The outcome indicators mainly focused on the patients' nutritional status, complication rates, length of hospital stay, body weight, and BMI, etc. However, the methodological quality of the included studies needs to be improved. The results of the studies all showed that nutritional prehabilitation benefits the prognosis of patients with head and neck cancer. CONCLUSION: Overall evidence suggests that nutritional prehabilitation can improve patients' nutritional status and enhance their prognosis and quality of life, showing beneficial effects. More high-quality, large-sample clinical studies should be conducted in the future to guide the clinical implementation of nutritional prehabilitation.
21 May 2026
Read appraisal →Radiation oncology (London, England)
Comparative study between PET-CT simulation and CT for head and neck cancer treatment and their potential effect on the thyroid gland: a prospective comparative study
BACKGROUND: Radiotherapy (RT) plays a key role in multidisciplinary treatment of Head and neck cancers (HNCs). Over the past two decades, advances have improved RT's tumor control and toxicity profiles. Precise delineation of gross tumor volume and organs at risk using CT and PET-CT is crucial for safe and effective RT. Additionally, RT increases the risk of thyroid dysfunction, especially hypothyroidism. AIM: The present study compared PET-CT simulation and CT-alone planning in head and neck cancer (HNC) treatment, emphasizing their effects on thyroid gland function. METHODS: The patients were divided into two groups according to the simulation method (PET-CT or CT), two fresh blood samples (5 mL/ sample) were collected from each patient, before and after radiotherapy. TSH, FT3, and FT4 (Thyroid biomarkers) were assessed by electrochemiluminescent immunoassay (ECLIA). RESULTS: PET-CT simulation demonstrated a significant improvement in GTVp delineation accuracy by 71.8% compared with CT. After radiotherapy, TSH levels show minor decline changes, whereas FT3 and FT4 levels declined significantly, with greater declines observed in stages II and III. The most common tumor site was the larynx (60% in PET-CT, 75% in CT), followed by the tongue, nasopharynx, and parotid gland. CONCLUSION: PET-CT simulation offered superior GTVp delineation accuracy over CT, improving HNCs radiotherapy planning. Immediately after RT, radiotherapy caused significant declines in serum FT3 and FT4 levels, with only a minor, non-significant TSH declining, particularly in stage II/III disease. Notably FT4 is little more affected in PET-CT simulation patient group, highlighting the need for thyroid function monitoring. These results advocate PET-CT integration in planning and routine endocrine monitoring for treatment complications.
19 May 2026
Read appraisal →Computers in biology and medicine
Mapping intratumoral heterogeneity through PET-derived washout and deep learning after proton therapy
The distribution of produced isotopes during proton therapy can be imaged with Positron Emission Tomography (PET) to verify dose delivery. However, biological washout, driven by tissue-dependent processes such as perfusion and cellular metabolism, reduces PET signal-to-noise ratio (SNR) and limits quantitative analysis. In this work, we propose an uncertainty-aware deep learning framework to improve the estimation of washout parameters in post-proton therapy PET, not only enabling accurate correction for washout effects, but also mapping intratumoral heterogeneity as a surrogate marker of tumor status and treatment response. We trained the models on Monte Carlo-simulated data from eight head-and-neck cancer patients, and tested them on four additional head-and-neck and one liver patient. Each patient was represented by 75 digital twins with distinct tumoral washout dynamics and imaged 15 min after treatment, when slow washout components dominate. We also introduced washed-out maps, quantifying the contribution of medium and fast washout components to the loss in activity between the end of treatment and the start of PET imaging. Trained models significantly improved resolution and accuracy, reducing average absolute errors by 60% and 28% for washout rate and washed-out maps, respectively. For intratumoral regions as small as 5.1 mL, errors predominantly fell below thresholds for differentiating vascular status, and the models generalized across anatomical areas and acquisition delays. This study shows the potential of deep learning in post-proton therapy PET to non-invasively map washout kinetics and reveal intratumoral heterogeneity, supporting dose verification, tumor characterization, and treatment personalization. The framework is available at https://github.com/pcabrales/ppw.git.
17 May 2026
Read appraisal →Journal of robotic surgery
Longitudinal health-related quality of life after remote-access versus open thyroidectomy for thyroid cancer: a systematic review and meta-analysis
As remote-access thyroidectomy (RAT) becomes more widely used, evidence on patient- reported outcomes, particularly health-related quality of life (HRQoL), remains scarce and inconsistent. This meta-analysis compared postoperative HRQoL between thyroid cancer (TC) patients undergoing RAT and open thyroidectomy (OT) and assessed changes over time. A comprehensive search of five major databases was conducted from inception to August 2025. Studies reporting HRQoL after RAT or OT were included. Outcomes were grouped by postoperative timepoints. Outcomes were stratified across distinct postoperative timepoints to calculate pooled standardized mean difference (SMD) or mean difference (MD). Heterogeneity was explored through rigorous subgroup analyses encompassing surgical modalities, countries, and assessment instruments. Forty-one studies met the inclusion criteria, and 29 records were included in the quantitative synthesis. RAT demonstrated early advantages in comprehensive quality of life at 1 month and 3 months, and these advantages dissipated long term. Pain trajectories exhibited a biphasic pattern: RAT was associated with lower pain scores on postoperative day 1 but paradoxically higher scores during the 1-2 week period, with subsequent convergence. Cosmetic satisfaction and swallowing function consistently favored RAT from 1-2 weeks through 6 months, while voice outcomes showed no discernible differences. RAT appears to confer selected short- to medium-term patient-reported advantages over OT, particularly in cosmetic satisfaction and swallowing function. But these benefits are heterogeneous and not consistently maintained across all domains or timepoints. Future studies should standardize cross-culturally validated PRO instruments and adopt harmonized follow-up intervals and reporting guidelines to clarify the patient-centered value of RAT.
15 May 2026
Read appraisal →Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation
Responsiveness and clinical utility of patient-reported outcome instruments in head and neck cancer: a systematic review
PURPOSE: To synthesize and evaluate current evidence on the responsiveness, interpretability (MID/MIC), and clinical utility of validated health-related quality of life (HRQoL) patient-reported outcome (PRO) instruments used in adults with head and neck cancer (HNC). METHODS: A systematic search of PubMed, Scopus, Web of Science, Embase, and Cochrane Library (Central) identified studies published from January 2015 to July 2025. The review was registered in PROSPERO (CRD420251128978) and followed PRISMA 2020 guidelines. Observational and interventional studies assessing HRQoL using validated HNC instruments (EORTC QLQ-C30, QLQ-H&N35/H&N43, UW-QOL, FACT-H&N or MDADI) were included. Eligible studies reported responsiveness (ability to detect meaningful change) or clinical applicability. Methodological quality was evaluated using the COSMIN Risk of Bias checklist, and results were narratively synthesized due to methodological heterogeneity. RESULTS: Thirteen studies involving 8,075 patients met the inclusion criteria. The EORTC QLQ-C30 and H&N35/43, UW-QOL, FACT-H&N, and MDADI exhibited consistent responsiveness, capturing expected HRQoL deterioration during treatment and partial recovery within 6-12 months. Reported minimal important difference (MID) and minimal important change (MIC) values typically ranged from 4 to 15 points for improvement and 10-20 points for deterioration, supporting the interpretability of change scores. Several patient-reported outcome measures (PROMs) also demonstrated prognostic relevance, with early HRQoL changes associated with survival or recurrence. CONCLUSION: Validated HRQoL questionnaires in HNC demonstrate robust responsiveness and meaningful clinical applicability, supporting their integration into routine oncology practice. Their use may enhance early detection of complications, guide rehabilitative interventions, and facilitate more individualized, patient-centered care.
10 May 2026
Read appraisal →The Journal of clinical endocrinology and metabolism
An Interpretable Machine-Learning Model for Predicting Occult Central Lymph Node Metastasis in Papillary Thyroid Cancer
CONTEXT: Accurate preoperative prediction of occult lymph node metastasis (OLNM) in clinically lymph node negative (cN0) papillary thyroid carcinoma (PTC) is critical for optimizing therapeutic strategy, particularly for thermal ablation and active surveillance. OBJECTIVE: The aim of this study was to develop an interpretable machine-learning (ML) model to predict the risk of OLNM in cN0 PTC patients. METHODS: This retrospective study analyzed data of 961 cN0 PTC patients (August 2018-August 2023). Multivariable logistic regression identified independent risk factors for OLNM in cN0 PTC. The cohort was randomly divided into the training and test sets, and a subset of patients with tumors sized 1 cm or less was further extracted from the test set for internal validation. Eight ML models incorporating clinical, ultrasonographic, and molecular features were developed and evaluated. Shapley Additive exPlanations (SHAP) enhanced interpretability. RESULTS: RET fusion positivity and BRAF mutation positivity were identified as independent molecular risk factors for OLNM in cN0 PTC, alongside 6 clinical and ultrasonographic variables. Nine predictors were incorporated into the predictive model. The random forest (RF) model achieved optimal performance with an area under the curve (AUC) of 0.906 in the training set and 0.733 in the test set, along with the lowest Brier scores of 0.135 and 0.212, respectively. Analysis of tumors sized 1 cm or less internally validated the model's robustness with an AUC of 0.719. SHAP analysis identified size, age, and clustered punctate echogenic foci as the top predictors. CONCLUSION: This is the first study to identify RET fusion positivity as an independent OLNM risk factor in cN0 PTC. The developed RF model demonstrates moderate predictive performance for OLNM risk and provides a framework for integrating clinical, sonographic, and molecular data, and is deployed as a web calculator (https://predictingoccultlymphnodemetastasis.shinyapps.io/web3/).
23 Apr 2026
Read appraisal →Journal of clinical oncology : official journal of the American Society of Clinical Oncology
Pembrolizumab With or Without Lenvatinib as First-Line Therapy for Recurrent or Metastatic Head and Neck Squamous Cell Carcinoma: Phase III LEAP-010 Study
PURPOSE: The PD-1 inhibitor pembrolizumab is approved as first-line treatment for recurrent/metastatic head and neck squamous cell carcinoma (R/M HNSCC). In LEAP-010 (ClinicalTrials.gov identifier: NCT04199104), the multikinase inhibitor lenvatinib plus pembrolizumab was evaluated as first-line therapy in participants with PD-L1 combined positive score (CPS) ≥1 R/M HNSCC. METHODS: In this phase III, randomized, placebo-controlled, double-blind study, participants 18 years and older with PD-L1 CPS ≥1 R/M HNSCC deemed incurable by local therapy were randomly assigned 1:1 to lenvatinib 20 mg plus pembrolizumab 200 mg IV once every 3 weeks for ≤35 cycles or placebo orally once daily plus pembrolizumab 200 mg IV once every 3 weeks for ≤35 cycles. Primary end points were objective response rate (ORR), progression-free survival (PFS), and overall survival (OS). Per the prespecified analysis plan, ORR and PFS were reported from the first interim analysis (IA1; data cutoff: July 6, 2022), and OS from IA2 (data cutoff: May 30, 2023). RESULTS: Five hundred eleven participants were randomly assigned to lenvatinib plus pembrolizumab (n = 256) or placebo plus pembrolizumab (n = 255). The median time from random assignment to data cutoff was 11.5 months for IA1 and 21.3 months for IA2. At IA1, the median PFS was 6.2 months for lenvatinib plus pembrolizumab versus 2.8 months for placebo plus pembrolizumab (hazard ratio [HR], 0.64 [95% CI, 0.50 to 0.81]; P = .0001040); the ORR was 46.1% versus 25.4%, respectively (difference = 20.2% [95% CI, 10.5 to 29.6]; P = .0000251). At IA2, the median OS was 15.0 months for lenvatinib plus pembrolizumab versus 17.9 months for placebo plus pembrolizumab (HR,1.15 [95% CI, 0.91 to 1.45]; P = .882). At IA2, 170 (66.9%) participants receiving lenvatinib plus pembrolizumab had grade 3-4 all-cause adverse events compared with 97 (38.3%) participants on placebo plus pembrolizumab. CONCLUSION: In participants with PD-L1 CPS ≥1 R/M HNSCC, first-line lenvatinib plus pembrolizumab significantly improved ORR and PFS, but not OS, compared with placebo plus pembrolizumab. The safety profile was consistent with published data.
21 Apr 2026
Read appraisal →JMIR cancer
Exploring the Adoption of Mobile Health Apps Among Patients with Head and Neck Cancer After the COVID-19 Pandemic: Secondary Analysis of Cross-Sectional Survey
BACKGROUND: Patients with head and neck cancer (HNC) face significant challenges in accessing coordinated care due to the complexity and multimodality of their treatment and the impact on vital functions. The COVID-19 pandemic has disrupted cancer care while accelerating the adoption of digital health solutions. Mobile health (mHealth) apps offer potential solutions for remote symptom monitoring, communication between patients and providers, and continuity of care. Nevertheless, their acceptance among patients with HNC remains limited due to age-related digital divides and concerns about accessibility. OBJECTIVE: The aim of this study was to investigate the impact of the COVID-19 pandemic on patients with HNC and to explore their attitudes toward mHealth apps as a supplement to cancer treatment. METHODS: A secondary analysis of a cross-sectional survey was conducted between January 2023 and May 2024 at Julius Maximilian University of Würzburg in Germany. A total of 355 patients with HNC were recruited and completed the structured "Cancer and COVID-19" survey via the Corona Health app or in paper form. The 25-question survey assessed sociodemographic information, the impact of the pandemic on diagnosis/treatment/aftercare, and interest in mHealth apps. Descriptive statistics and bivariate correlations were used for the analysis. RESULTS: The cohort comprised 261 (74%) men and 94 (26%) women with an average age of 67 (SD 11.5) years. Most participants (n=264, 74%) stated that the pandemic had no impact on their cancer treatment, although 20% (n=71) experienced disruptions, particularly in follow-up appointments and treatment monitoring. Only 10% (n=36) currently used health apps, but 57% (n=203) expressed a willingness to use mHealth technologies. Younger patients, patients with higher education, and participants who were more affected by the pandemic showed greater openness to digital health solutions. The most significant barriers included age, digital literacy, and perceived usefulness, while preferred app features included interaction with physicians (n=160, 45%) and data sharing with researchers (n=153, 43%). CONCLUSIONS: Although the COVID-19 pandemic had only a limited direct impact on HNC care at the institution at hand, it revealed significant patient interest in mHealth apps. However, significant barriers remain, particularly among older adults with lower digital literacy. Future mHealth initiatives should focus on improving digital literacy, addressing privacy concerns, demonstrating clinical benefits, and developing personalized, accessible solutions to optimize cancer care for this vulnerable population.
17 Apr 2026
Read appraisal →Clinical cancer research : an official journal of the American Association for Cancer Research
Becotatug Vedotin in Patients with Recurrent or Metastatic Squamous Cell Carcinoma of the Head and Neck: A Multicenter, Phase IIa Trial
PURPOSE: This multicenter, phase IIa trial (NCT04868162) investigated the efficacy and safety of becotatug vedotin, an anti-epidermal growth factor receptor (EGFR) antibody-drug conjugate, in patients with recurrent or metastatic squamous cell carcinoma of the head and neck (R/M SCCHN) who have limited therapeutic options. PATIENTS AND METHODS: Patients with R/M SCCHN who progressed after platinum-based chemotherapy and/or programmed cell death 1 (PD-1)/programmed death ligand 1 (PD-L1) inhibitors, including those with multiple lines of prior therapy, were administered intravenous becotatug vedotin at either 2 or 2.3 mg/kg every 3 weeks. The primary endpoint was the objective response rate (ORR). RESULTS: Sixty-seven patients were enrolled (35 received 2 mg/kg and 32 received 2.3 mg/kg). The ORR was 20.9% [14/67, 95% confidence interval (CI), 11.9-32.6], with a median duration of response (DoR) of 10.9 months (95% CI, 2.6-15.1). The median progression-free survival (PFS) was 2.9 months (95% CI, 1.8-3.9), and the median overall survival (OS) was 6.7 months (95% CI, 5-8.9). Treatment-related adverse events (TRAE) were reported in 91% (61/67) patients, most commonly being rash (26.9%), pruritus (25.4%), constipation (23.9%), and anemia (20.9%). CONCLUSIONS: Becotatug vedotin demonstrated promising antitumor activity with a manageable safety profile in previously heavily treated R/M SCCHN, particularly at the recommended dose of 2.3 mg/kg, among patients who had failed platinum-based chemotherapy and PD-1/PD-L1 inhibitors (≤2 prior lines of therapy).
16 Apr 2026
Read appraisal →Journal of clinical oncology : official journal of the American Society of Clinical Oncology
Development of a Classifier for Metabolic Subtypes of Nasopharyngeal Carcinoma to Guide Personalized Immunotherapy Strategies: Biomarker Analysis of the Phase III CONTINUUM and DIPPER Trials
PURPOSE: Personalized immunotherapy strategies are urgently needed for patients with locoregionally advanced nasopharyngeal carcinoma (NPC). We aim to identify biomarkers predictive of immunotherapy benefits, using data from the phase III CONTINUUM (ClinicalTrials.gov identifier: NCT03700476) and DIPPER (ClinicalTrials.gov identifier: NCT03427827) randomized clinical trials. PATIENTS AND METHODS: Tumor samples from 407 patients in the CONTINUUM (discovery cohort) and DIPPER (validation cohort) trials were subjected to RNA sequencing. In the discovery cohort, metabolic gene-based consensus clustering was performed to determine subtypes. A machine learning-based classifier was subsequently developed in the discovery cohort and then applied to the validation cohort to assign metabolic subtypes. Gene set enrichment analyses were used to characterize the biological features of each metabolic subtype. The clinical end point was event-free survival (EFS). RESULTS: In the discovery cohort, three metabolic subtypes were identified with distinct tumor-intrinsic and immune features as well as differential EFS benefits from adding anti-PD-1 to chemoradiotherapy (CRT). Specifically, the MS1 subtype exhibited a significant improvement in 3-year EFS in the anti-PD-1 plus CRT arm compared with the CRT-alone arm (3-year EFS, 90.2% v 69.6%; hazard ratio, 0.27 [95% CI, 0.11 to 0.67]), whereas MS2 (3-year EFS, 94.1% v 93.8%) and MS3 subtypes (3-year EFS, 75.0% v 75.0%) derived no significant survival benefit. The subtype features were preserved in the validation cohort, with consistent prognostic and predictive value. A pooled analysis of both cohorts demonstrated the significant interaction between metabolic subtypes and the treatment effect (Pinteraction = 0.0074). CONCLUSION: In this biomarker study, we defined metabolic subtypes of NPC that predicted the EFS benefit from immunotherapy. This novel molecular classification provides a promising predictive biomarker for personalized treatment decision for patients with locoregionally advanced NPC.
11 Apr 2026
Read appraisal →PloS one
Development and evaluation of a multimodal feature-based predictive model for radiotherapy-induced oral mucositis in nasopharyngeal carcinoma
BACKGROUND: Accurate prediction of radiation-induced oral mucositis is crucial for personalized treatment in head and neck cancer. However, developing robust predictive models utilizing high-dimensional multimodal data (CT imaging, dose distribution, and clinical features) remains challenging, particularly in cohorts with limited sample sizes. OBJECTIVE: This study aimed to rigorously evaluate and compare the multi-class predictive performance of traditional machine learning algorithms and deep learning architectures under a small-cohort setting. METHODS: Multimodal data from 108 patients were collected. A comprehensive evaluation framework incorporating nine traditional machine learning algorithms and two deep learning models (a dimensionality-reduced 1D-CNN and a multimodal 3D-CNN) was established. To ensure robust evaluation, a stratified 5-fold cross-validation was employed. Model performance was comprehensively quantified using mean ± standard deviation (SD) across multiple metrics, including the Area Under the Curve (AUC), accuracy, and Matthews Correlation Coefficient (MCC). RESULTS: Inter-rater reliability for RIOM grading was excellent (Cohen's kappa = 0.82, 95% CI: 0.73-0.91). Among traditional machine learning approaches, the Extra Trees (ET) algorithm achieved the highest discriminative capacity (AUC: 0.956 ± 0.046), while Logistic Regression (LR) demonstrated optimal overall accuracy (0.832 ± 0.155) and stability. Regarding deep learning, the lightweight 1D-CNN utilizing fused low-dimensional features exhibited highly competitive and robust performance (AUC: 0.900 ± 0.072; Accuracy: 0.732 ± 0.140). In stark contrast, the high-dimensional multimodal 3D-CNN suffered from severe overfitting and mode collapse phenomenon, yielding significantly inferior results (AUC: 0.568 ± 0.090; MCC: -0.025 ± 0.031). CONCLUSIONS: For small-cohort radiomics and dosimetric analyses, ensemble learning models (e.g., ET) and appropriately regularized linear models (e.g., LR) remain highly effective. While deep learning holds promise, high-dimensional architectures like 3D-CNNs are highly susceptible to mode collapse without massive datasets. Instead, employing feature dimensionality reduction combined with lightweight networks (1D-CNN) is a vastly superior strategy to extract reliable predictive patterns from limited clinical data.
10 Apr 2026
Read appraisal →Physics in medicine and biology
Fraction-variant VMAT planning for patients with complex gynecological and head-and-neck cancer.
Objective. Increasing the number of arcs in volumetric modulated arc therapy (VMAT) allows for better intensity modulation and may improve plan quality. However, this leads to longer delivery times, which may cause patient discomfort and increase intra-fractional motion. In this study, it was investigated whether the delivery of different VMAT plans in different fractions may improve the dosimetric quality and delivery efficiency for the treatment of patients with complex tumor geometries.Approach. A direct aperture optimization algorithm was developed which allows for the simultaneous optimization of different VMAT plans to be delivered in different fractions, based on their cumulative physical dose. Each VMAT plan is constrained to deliver a uniform dose within the target volume, such that the entire treatment does not alter the fractionation scheme and is robust against inter-fractional setup errors. This approach was evaluated in-silico for five patients with gynecological and five patients with head-and-neck cancer.Main results. For all patients, fraction-variant treatments achieved significantly better target coverage and reduced the dose to critical organs-at-risk compared to fraction-invariant treatments that deliver the same plan in every fraction, where the dosimetric benefit was shown to increase with the number of different plans to be delivered. In addition, 1-arc and 2-arc fraction-variant treatments could approximate the dosimetric quality of 3-arc fraction-invariant treatments, while reducing the delivery time.Significance.Fraction-variant VMAT treatments may achieve excellent dosimetric quality for patients with complex tumor geometries, while keeping the delivery time per fraction viable.
9 Apr 2026
Read appraisal →Signal transduction and targeted therapy
Anti-PD-1 antibody penpulimab plus chemotherapy for recurrent or metastatic nasopharyngeal carcinoma: a randomized, double-blind phase 3 study
In this phase 3 trial, penpulimab combined with chemotherapy was assessed against a regimen of placebo plus chemotherapy for the first-line treatment of recurrent or metastatic nasopharyngeal carcinoma (R/M NPC). 291 patients were randomised and allocated in a 1:1 ratio in order to receive penpulimab (n = 144; 200 mg) or placebo (n = 147; 200 mg), plus chemotherapy (cisplatin/carboplatin and gemcitabine) every 3 weeks. Patients followed by maintenance therapy with penpulimab or placebo after 6 cycles. The primary endpoint of this study was progression-free survival (PFS) according to RECIST v1.1, and a significantly longer median PFS in the penpulimab arm versus the placebo arm (9.63 versus 7.00 months; hazard ratio 0.45, 95% CI: 0.33-0.62, P < 0.0001) was demonstrated in this prespecified interim analysis. The key secondary endpoint was the overall survival (OS). However, the OS data were still immature, and the median OS was not achieved (hazard ratio, 0.94; 95% CI: 0.63-1.40). The occurrence of treatment-related adverse events (grade ≥ 3) was 89.0% and 85.9% in two arms, with the most common being reduced the quantity of neutrophil (56.2% vs. 62.0%), reduced the quantity of white blood cell (54.1% vs. 54.9%), and anemia (45.2% vs. 38.7%). In the penpulimab arm, 6 patients (4.1%) experienced immune-related adverse events (grade ≥ 3). Adding penpulimab to chemotherapy led to a notable enhancement in PFS for the first-line R/M NPC treatment, alongside a safety profile that was both manageable and tolerable. ClinicalTrials.gov identifier NCT04974398.
8 Apr 2026
Read appraisal →PloS one
Utilizing the transformer mechanism to predict cervical lymph node metastasis in patients with papillary thyroid carcinoma
RATIONALE AND OBJECTIVE: The status of cervical lymph node metastasis(LNM) in Papillary thyroid carcinoma(PTC) can affect the patient's treatment plan and prognosis. This study aims to develop and validate the application value of Vision transformer (ViT) model in preoperatively predicting cervical LNM in PTC. MATERIALS AND METHODS: A total of 540 PTC patients were retrospectively reviewed from two hospitals from April 20,2022 to August 20,2023.The ViT model is built based on the two-dimensional rectangular ultrasound image of the primary thyroid tumor, and at the same time, to compare its performance, a deep learning model of the traditional Convolutional neural network (CNN) framework, a ultrasound radiomics combined model(Clinical-Rad model), and clinical model are built. RESULTS: The ViT model demonstrated an AUC of 0.807 (95% CI: 0.709-0.905) in the internal validation cohort and 0.809 (95% CI: 0.720-0.900) in the external validation cohort. The ViT model's AUC ranged from 0.807-0.814 across all cohorts, significantly exceeding the clinical model (AUC: 0.595-0.669, P<0.001). While the AUC of the ViT model in the training cohort was slightly lower than that of the combined ultrasound radiomics model (0.814 vs 0.828, P=0.491), it showed significantly higher AUC values in the internal (0.807 vs 0.718, P=0.049) and external validation cohorts (0.809 vs 0.691, P<0.001). Compared to the clinical and combined radiomics models, the ViT model exhibited stable and superior predictive performance for PTC cervical lymph node metastasis.In the internal validation cohort, Doctor C's net reclassification improvement (NRI) with the ViT model was 0.106 (P=0.022), and the integrated discrimination improvement (IDI) was 0.106 (P=0.023). Doctor D showed NRI and IDI values of 0.113 (P=0.022) and 0.106 (P=0.024), respectively. In the external validation cohort, Doctor C's NRI and IDI were 0.090 (P=0.024) and 0.106 (P=0.024), while Doctor D had values of 0.011 (P=0.013) and 0.106 (P=0.013). The ViT model enhanced the diagnostic capabilities of both Doctor C, with less clinical experience, and Doctor D, with extensive experience. CONCLUSION: The deep learning model based on the Transformer mechanism shows good performance in predicting LNM in PTC patients, which is superior to the clinical model, Clinical-Rad model, and traditional CNN model.
5 Apr 2026
Read appraisal →AJNR. American journal of neuroradiology
Temporomandibular Joint Lesions with Intracranial Extension: Illustrative Cases from a Systematic Review of the Literature and Our Institution
BACKGROUND: Intracranially extending temporomandibular joint (TMJ) lesions may be radiologically misinterpreted as primary intracranial or skull base pathologies, leading to diagnostic delays or inappropriate management. PURPOSE: This systematic review aimed to characterize the clinical and imaging features of such TMJ lesions and evaluate the impact of radiologic misclassification. We also aimed to develop a diagnostic framework for when to consider an intracranially extending TMJ lesion, based on clinical and radiologic features. DATA SOURCES: A comprehensive search of MEDLINE, Scopus, and EMBASE, conducted in accordance with PRISMA guidelines, yielded 2255 records. STUDY SELECTION: After screening with predetermined inclusion and exclusion criteria, 128 studies involving 152 patients were included in the final analysis. DATA ANALYSIS: Statistical analyses were performed using STATA software. We also identified 3 patient cases through our institutional neuroradiology practice who were clinically and radiologically assessed for intracranially extending TMJ lesions. DATA SYNTHESIS: Patients had symptoms for an average of 34 months before diagnosis (47% women, mean age 50 years). The most common pathologies were pigmented villonodular synovitis/tenosynovial giant-cell tumor (43%) and synovial chondromatosis (24%). Neurologic symptoms were reported in 48% of cases, most frequently hearing loss (70%). Nearly one-third (33%) of cases with an imaging differential did not list a TMJ pathology (18/55). In cases with accurate imaging diagnosis, 90% had both CT and MRI performed. Most lesions were nonenhancing (CT 83%, MRI 75%) and demonstrated no adjacent brain edema (96%). In 2 cases, a TMJ ganglion cyst and pseudogout were misdiagnosed as intracranial tumors, resulting in unnecessary intervention, including repeat craniotomy and radiotherapy. LIMITATIONS: There were inherent biases of case report literature, including variability in the reporting of the imaging and clinical features, management, and follow-up. CONCLUSIONS: TMJ lesions with intracranial extension often present with nonspecific symptoms and can mimic extra-axial tumors, leading to misdiagnosis on imaging. Recognition of hallmark imaging features, including lack of parenchymal invasion and distinct imaging patterns, may help improve radiologic accuracy and prevent overtreatment. We propose a diagnostic framework outlining when to suspect intracranially extending TMJ lesions based on clinical and imaging features, and how to avoid common diagnostic pitfalls.
4 Apr 2026
Read appraisal →Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
Preventive effects of dexamethasone oral solution on radiation-induced mucositis in patients with head and neck cancer: a randomized, triple-blind, parallel-group trial
PURPOSE: Radiation-induced oral mucositis is a frequent and debilitating complication of curative radiotherapy for head and neck cancer. This randomized, triple-blind, placebo-controlled, parallel-group trial evaluated whether a topical dexamethasone oral solution (0.5 mg/10 mL) used as a prophylactic mouthwash reduces the incidence and severity of radiation-induced oral mucositis. METHODS: Fifty-four patients with non-metastatic head and neck squamous cell carcinoma scheduled to receive ≥ 60 Gy radiotherapy were randomized 1:1 to dexamethasone mouthwash or placebo and instructed to gargle 10 mL four times daily during radiotherapy. The primary endpoint was the incidence and maximum severity of oral mucositis occurring during radiotherapy, defined as the highest grade recorded using the WHO Oral Toxicity Scale and NCI-CTCAE criteria v. 5. Secondary endpoints included patient-reported pain (VAS), dysphagia, treatment discontinuation, and mucositis status at 2 months. Fifty patients completed the study (25 per arm); four patients discontinued (reasons reported in Results). RESULTS: The dexamethasone group had significantly lower mucositis severity at all time points by WHO and CTCAE scales (all p < 0.01), a delayed peak severity (week 5 vs week 3), lower pain scores from weeks 4-7 (median VAS 0-3 versus 4-6; p < 0.01), and lower dysphagia at 2 months (32% vs 72%; p = 0.01). No grade 4 mucositis occurred in the dexamethasone arm; two grade 4 events occurred in placebo. There were no statistically significant differences in treatment discontinuation. Topical dexamethasone was well tolerated; no signal of treatment-related systemic steroid toxicity or increased oral infections was observed. CONCLUSION: In this phase II study, prophylactic topical dexamethasone mouthwash reduced the severity and delayed the onset of radiation-induced oral mucositis and improved patient-reported pain. These findings are promising but require confirmation in larger multicenter trials with longer follow-up and formal patient-reported outcomes (e.g., EORTC QLQ-H&N35, OMAS) and comparison with guideline-recommended strategies such as photobiomodulation. TRIAL REGISTRATION NUMBER: Iranian Registry of Clinical Trials; trial registration number: IRCT20240525061893N1; date of registration: 21 July 2024.
4 Apr 2026
Read appraisal →Journal of dentistry
A deep learning method for diagnosis of oral potentially malignant disorders
OBJECTIVES: This study aimed to develop and validate a two-stage deep learning method for diagnosing oral potentially malignant disorders (OPMDs). We also compared its diagnostic performance with that of clinicians at different levels of seniority and evaluated its utility as a clinical decision support tool. METHODS: A two-stage deep learning method was developed. CLA OPMD-OOML was designed to differentiate OPMDs from other oral mucosal lesions (OOML), while CLA OPMDs was designed to classify specific OPMD subtypes. The method was trained on an internal dataset (ZJUSS, n = 3305), and its generalizability was evaluated on two external multicenter datasets (WCHS and CS-SJTU, n = 1756). In a blinded, two-step comparative study, nine clinicians (junior, intermediate, and senior groups) performed diagnoses independently and then with AI assistance. RESULTS: The method significantly outperformed all clinician groups, achieving higher F1 scores in both CLA OPMD-OOML (89.9 % vs. 78.4-82.6 % for clinicians) and CLA OPMDs (92.2 % vs. 77.7-80.0 % for clinicians). It showed robust generalizability on the two external datasets, with F1 scores of 87.3 % and 86.9 % for CLA OPMD-OOML and 75.9-83.0 % for CLA OPMDs. With AI assistance, the diagnostic precision of junior and intermediate clinicians increased by 10.8 % and 5.9 %, respectively, raising the junior group's performance to the level of senior clinicians. CONCLUSIONS: The developed two-stage deep learning method demonstrated diagnostic performance comparable to or exceeding that of experienced clinicians in classifying OPMDs from clinical images. It functions as a powerful assistive tool that substantially enhances the diagnostic capabilities of junior and intermediate clinicians. CLINICAL SIGNIFICANCE: This AI method has the potential to serve as a reliable tool for large-scale early screening of OPMDs, particularly in regions with limited access to specialist dental care. It can also serve as a valuable training and decision support system for junior clinicians, helping to standardize diagnostic accuracy and improve patient outcomes through timely and accurate detection. A supplemental appendix to this article is available online.
4 Apr 2026
Read appraisal →Otolaryngology--head and neck surgery : official journal of American Academy of Otolaryngology-Head and Neck Surgery
Multicenter Clinical Validation of an Artificial Intelligence Diagnostic Classification Model for Laryngoscopy Images
OBJECTIVE: To develop and externally validate a computer-aided diagnosis (CADx) model using artificial intelligence (AI) for classifying laryngeal lesions from laryngoscopy images into high-risk (HR), low-risk (LR). STUDY DESIGN: Retrospective multicenter development of a CADx model and external validation on independent cohorts. SETTING: Multicenter tertiary referral hospitals (Italy, India, China, Greece, and Spain). METHODS: Over 20,000 images derived from laryngoscopic examinations were retrieved. Images were annotated based on histopathology or expert consensus. A deep learning model was trained using an internal dataset and evaluated on 2 external datasets to assess generalizability. The CADx model classifies only images containing visible lesions, discriminating between LR and HR categories. Diagnostic performance was measured using standard metrics, including accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC). Model performance was compared with physicians of varying expertise and ChatGPT-4o. RESULTS: The computer-aided diagnosis model achieved a similar performance across internal and external datasets in distinguishing HR from LR lesions, with accuracy/AUC of 0.90/0.89 internally, 0.85/0.85 on the Greek dataset, and 0.88/0.88 on the Spanish dataset. The model's accuracy was statistically noninferior to that of otolaryngologists and expert laryngologists, and superior to general practitioners and ChatGPT-4o. CONCLUSION: This is a large multicenter clinical validation of a CADx model for laryngeal endoscopy, demonstrating generalizability and performance comparable to clinicians in discriminating between LR and HR lesions. The model's success supports its potential role in augmenting diagnostic capabilities, especially in resource-limited settings. A prospective multicenter clinical trial is underway to assess real-world clinical implementation.
2 Apr 2026
Read appraisal →