Quality of Large Language Model Responses to Radiation Oncology Patient Care Questions.

Importance: Artificial intelligence (AI) large language models (LLMs) demonstrate potential in simulating human-like dialogue. Their efficacy in accurate patient-clinician communication within radiation oncology has yet to be explored. Objective: To determine an LLM's quality of responses to radiation oncology patient care questions using both domain-specific expertise and domain-agnostic metrics. Design, Setting, and Participants: This cross-sectional study retrieved questions and answers from websites (accessed February 1 to March 20, 2023) affiliated with the National Cancer Institute and the Radiological Society of North America. These questions were used as queries for an AI LLM, ChatGPT version 3.5 (accessed February 20 to April 20, 2023), to prompt LLM-generated responses. Three radiation oncologists and 3 radiation physicists ranked the LLM-generated responses for relative factual correctness, relative completeness, and relative conciseness compared with online expert answers. Statistical analysis was performed from July to October 2023. Main Outcomes and Measures: The LLM's responses were ranked by experts using domain-specific metrics such as relative correctness, conciseness, completeness, and potential harm compared with online expert answers on a 5-point Likert scale. Domain-agnostic metrics encompassing cosine similarity scores, readability scores, word count, lexicon, and syllable counts were computed as independent quality checks for LLM-generated responses. Results: Of the 115 radiation oncology questions retrieved from 4 professional society websites, the LLM performed the same or better in 108 responses (94%) for relative correctness, 89 responses (77%) for completeness, and 105 responses (91%) for conciseness compared with expert answers. Only 2 LLM responses were ranked as having potential harm. The mean (SD) readability consensus score for expert answers was 10.63 (3.17) vs 13.64 (2.22) for LLM answers (P < .001), indicating 10th grade and college reading levels, respectively. The mean (SD) number of syllables was 327.35 (277.15) for expert vs 376.21 (107.89) for LLM answers (P = .07), the mean (SD) word count was 226.33 (191.92) for expert vs 246.26 (69.36) for LLM answers (P = .27), and the mean (SD) lexicon score was 200.15 (171.28) for expert vs 219.10 (61.59) for LLM answers (P = .24). Conclusions and Relevance: In this cross-sectional study, the LLM generated accurate, comprehensive, and concise responses with minimal risk of harm, using language similar to human experts but at a higher reading level. These findings suggest the LLM's potential, with some retraining, as a valuable resource for patient queries in radiation oncology and other medical fields.

Local Therapies for Hepatocellular Carcinoma and Role of MRI-Guided Adaptive Radiation Therapy.

Hepatocellular carcinoma (HCC) is the most common liver tumor, with a continually rising incidence. The curative treatment for HCC is surgical resection or liver transplantation; however, only a small portion of patients are eligible due to local tumor burden or underlying liver dysfunction. Most HCC patients receive nonsurgical liver-directed therapies (LDTs), including thermal ablation, transarterial chemoembolization (TACE), transarterial radioembolization (TARE), and external beam radiation therapy (EBRT). Stereotactic ablative body radiation (SABR) is a specific type of EBRT that can precisely deliver a high dose of radiation to ablate tumor cells using a small number of treatments (or fractions, typically 5 or less). With onboard MRI imaging, MRI-guided SABR can improve therapeutic dose while minimizing normal tissue exposure. In the current review, we discuss different LDTs and compare them with EBRT, specifically SABR. The emerging MRI-guided adaptive radiation therapy has been reviewed, highlighting its advantages and potential role in HCC management.

Imaging of GBM in the Age of Molecular Markers and MRI Guided Adaptive Radiation Therapy.

Glioblastoma (GBM) continues to be one of the most lethal malignancies and is almost always fatal. In this review article, the role of radiation therapy, systemic therapy, as well as the molecular basis of classifying GBM is described. Technological advances in the treatment of GBM are outlined as well as the diagnostic imaging characteristics of this tumor. In addition, factors that affect prognosis such as differentiating progression from treatment effect is discussed. The role of MRI guided radiation therapy and how this technology may provide a mechanism to improve the care of patients with this disease are described.

Evolution of Radiation Therapy in Pancreas Cancer Management toward MRI-Guided Adaptive Radiation Therapy.

Pancreas cancer has a poor prognosis despite aggressive treatment and is the fourth leading cause of cancer death in the United States. At diagnosis, most patients have either metastatic or locally advanced disease. In this article, we review the evolution of treatments in locally advanced pancreas cancer (LAPC) and discuss the various radiation therapy fractionation schemes. Furthermore, we examine the data supporting dose escalation and the delivery of ablative biologically effective doses in the setting of LAPC. Finally, we review the role of MRI-guided radiation therapy in escalating dose while sparing organs at risk in the era of stereotactic magnetic resonance-guided adaptive radiation therapy.

Predicting Obstructive Sleep Apnea Status With the Reflux Symptom Index in a Sleep Study Population.

OBJECTIVES/HYPOTHESIS: Otolaryngologic symptoms of obstructive sleep apnea (OSA) and their diagnostic utility are not well studied. We aimed to elucidate the prevalence of otolaryngologic symptoms among patients being evaluated for OSA. Given findings that the Reflux Symptom Index (RSI) was strongly associated with OSA status, we evaluated the diagnostic utility of the RSI for predicting OSA status. STUDY DESIGN: Cross-sectional. METHODS: We recruited 101 adults presenting for ambulatory polysomnograms to the Northwestern Sleep Disorders Center from July 2017 to July 2018. The Epworth Sleepiness Scale, Pittsburgh Sleep Quality Index, Leicester Cough Questionnaire (LCQ), RSI, Gastroesophageal Reflux Disease Questionnaire, Sino-Nasal Outcome Test-22, Nasal Obstruction Symptom Evaluation, Eustachian Tube Dysfunction Questionnaire 7, and Headache Impact Test were administered. Polysomnogram results were subsequently obtained. Patients with OSA (apnea-hypopnea index ≥ 5) and without OSA were compared. RESULTS: Of the 101 participants, 98 had valid sleep study results. Of those, 72 were diagnosed with OSA and 26 were not. The two groups differed significantly in age and body mass index (BMI). Of the questionnaires, only the RSI and LCQ means differed significantly, with worse symptoms in the OSA group (P = .003 and .014, respectively). Upon univariate regression, age, BMI, and RSI were associated with OSA status. Using regression coefficients, a clinical score of 2 (RSI) + 1.5 (BMI) + age yielded a diagnostic model (C-statistic = 0.807, P < .001). A threshold score of 104.21 was 76.4% sensitive and 73.1% specific. CONCLUSIONS: Patients with OSA have worse symptoms of laryngopharyngeal reflux as measured by the RSI. The addition of the RSI to the recognized factors of age and BMI improves diagnostic utility for OSA. LEVEL OF EVIDENCE: 2 Laryngoscope, 2020.

Desmoglein 1 Regulates Invadopodia by Suppressing EGFR/Erk Signaling in an Erbin-Dependent Manner.

Loss of the desmosomal cell-cell adhesion molecule, Desmoglein 1 (Dsg1), has been reported as an indicator of poor prognosis in head and neck squamous cell carcinomas (HNSCC) overexpressing epidermal growth factor receptor (EGFR). It has been well established that EGFR signaling promotes the formation of invadopodia, actin-based protrusions formed by cancer cells to facilitate invasion and metastasis, by activating pathways leading to actin polymerization and ultimately matrix degradation. We previously showed that Dsg1 downregulates EGFR/Erk signaling by interacting with the ErbB2-binding protein Erbin (ErbB2 Interacting Protein) to promote keratinocyte differentiation. Here, we provide evidence that restoring Dsg1 expression in cells derived from HNSCC suppresses invasion by decreasing the number of invadopodia and matrix degradation. Moreover, Dsg1 requires Erbin to downregulate EGFR/Erk signaling and to fully suppress invadopodia formation. Our findings indicate a novel role for Dsg1 in the regulation of invadopodia signaling and provide potential new targets for development of therapies to prevent invadopodia formation and therefore cancer invasion and metastasis. IMPLICATIONS: Our work exposes a new pathway by which a desmosomal cadherin called Dsg1, which is lost early in head and neck cancer progression, suppresses cancer cell invadopodia formation by scaffolding ErbB2 Interacting Protein and consequent attenuation of EGF/Erk signaling.

Are chronic cough and laryngopharyngeal reflux more common in obstructive sleep apnea patients?

OBJECTIVES/HYPOTHESIS: To assess if there is a significant difference in the prevalence and severity of chronic cough symptoms in obstructive sleep apnea (OSA) patients versus non-OSA patients and examine this relationship in regard to laryngopharyngeal reflux (LPR) symptoms. STUDY DESIGN: Prospective cohort study. METHODS: Patients referred to Northwestern Medicine Sleep Lab for home sleep testing were enrolled. Patients filled out the Leicester Cough Questionnaire (LCQ) and Reflux Symptom Index (RSI) before completing sleep testing. Home sleep testing results were reviewed, and patients were separated into non-OSA and OSA groups by standard Apnea-Hypopnea Index (AHI) criteria. Demographic characteristics and questionnaire scores of the two groups were compared. The relationship between OSA severity, as determined by AHI, and LCQ and RSI scores was assessed. RESULTS: Of the 52 patients enrolled, 33 patients met criteria for OSA and 19 patients did not. Comparing patients without OSA versus those with OSA, there was a significant difference in mean LCQ score (129.9 vs. 120.0, respectively; P = .02), implying worse cough symptoms among OSA patients, and mean RSI score (3.2 vs. 11.2, respectively; P = .0013), implying worse upper-airway reflux symptoms among OSA patients. There was a significant correlation between LCQ score and AHI (r = -0.39, P = .0061) and between RSI score and AHI (r = 0.37, P = .0078). CONCLUSIONS: OSA patients demonstrate worse chronic cough and LPR-related quality of life versus non-OSA patients. Furthermore, the severity of these quality-of-life measures was correlated with the severity of the AHI. Chronic cough and particularly the pharyngeal LPR symptoms may be associated with the presence and severity of OSA. LEVEL OF EVIDENCE: 2 . Laryngoscope, 129:1244-1249, 2019.