2479 - Assessing Construct Validity of ESAS in Minority Populations and Exploring Temporal Variations in PROs by HPV Status Using PCA

Purpose/Objective(s): The ESAS is a widely used tool for quantifying symptom burden, validation in minority groups is limited. The use of Principal Component Analysis (PCA) for structural validation has been underexplored. Sociodemographic and biological differences between HPV-related (p16-+ve) and HPV-unrelated OPSCC pts, symptom perception and patient-reported outcomes (PROs) may differ. This study aims to assess ESAS' structural validity in a diverse cohort using PCA and explore temporal variations in symptom clustering based on HPV status.

Materials/Methods: ESAS scores were prospectively collected from consecutive OPSCC pts treated with concurrent chemoXRT at a single center. Thirteen symptoms were categorized into 3 domains: physical (pain, tiredness, nausea, drowsiness, shortness of breath, appetite, constipation, sleep), psychological (depression, anxiety, well-being), and social/emotional (family distress, spiritual distress). Symptoms were assessed at 7 time points. PCA was used for dimensionality reduction. A covariance matrix was created and the Kaiser-Meyer-Olkin (KMO) measure assessed sampling adequacy. 3 factors with the highest eigenvalues were selected, and a component matrix was generated using varimax rotation with Kaiser normalization. Cronbach’s alpha was used to examine the internal consistency of items. Composite mean scores were analyzed over time in HPV-+ve and HPV--ve groups.

Results: Among 169 head and neck cancer pts, 91 OPSCC cases (510 ESAS forms) were analyzed (299 p16-+ve, 211 p16--ve). Median age: 59 years; 79% male; racial distribution: 37% White, 30% Hispanic, 30% Black, 2% Asian, 1% other. Median radiation dose: 6996 cGy over 48 days. Most frequent symptoms: pain (57%), sleep (50%), tiredness (48%).

3 symptom components emerged:

Psychological: depression, anxiety, family/spiritual distress (CA,a = 0.903)

Physical: tiredness, drowsiness, well-being, shortness of breath, sleep, pain (CA,a = 0.848)

Gastrointestinal: nausea, appetite, constipation (CA,a = 0.685)

Temporal trends differed by HPV status:

Psychological: Peak earlier (4–6 wks) in p16-+ve (1.71, 2.05, 1.56) vs. later (7–10 wks) in p16--ve (1.46, 1.54, 2.07).

Physical: Higher for p16-+ve (3.08, 3.42, 3.69) vs. p16--ve (2.77, 3.02, 2.58).

Gastrointestinal: Peaked in both groups, but were higher in p16-+ve (2.71, 3.41, 3.21) vs. p16--ve (2.08, 1.86, 2.5).

Sleep clustered with physical symptoms, not psychological, suggesting a stronger association with treatment effects.

Conclusion: ESAS shows strong internal consistency in a diverse OPSCC cohort, with 3 distinct symptom components. Temporal symptom burden differed by HPV status. P16-+ve pts had higher and more prolonged physical and gastrointestinal symptoms, while p16--ve pts had a delayed peak in psychological distress. These findings highlight the need for HPV-specific supportive care strategies and suggest that biological or socioeconomic factors may contribute to the observed differences in PROs.