1177 - Patient Reported Symptom Trajectories during and after Radiotherapy (RT) for Human Papillomavirus-Associated Oropharyngeal Cancer (HPVOPC): A TROG 12.01 Secondary Analysis
Presenter(s)
J. Smith1, D. Rischin2, J. Corry3, M. Bressel4, L. Kenny5, S. Porceddu6, C. Wratten7, A. M. J. Macann8, J. E. Jackson9, and L. J. McDowell10; 1Princess Alexandra Hospital, Brisbane, Australia, 2Department of Medical Oncology, Peter MacCallum Cancer Centre, Melbourne, VIC, Australia, 3GenesisCare St. Vincent's Hospital, Melbourne, Australia, 4Sir Peter MacCallum Department of Oncology, The University of Melbourne, Melbourne, VIC, Australia, 5Department of Radiation Oncology, Royal Brisbane & Women’s Hospital, Brisbane, Australia, 6Peter MacCallum Cancer Centre, Melbourne, VIC, Australia, 7Department of Radiation Oncology, Calvary Mater Hospital and University of Newcastle, Newcastle, Australia, 8Department of Radiation Oncology, Auckland City Hospital, Auckland, New Zealand, 9Radiation Oncology Centers, Gold Coast, QLD, Australia, 10University of Queensland, Brisbane, Australia
Purpose/Objective(s):
To identify patient-reported symptom trajectories before, during and after RT for HPVOPC. Secondarily, we examined whether patient and treatment factors differed across symptom trajectory classes.Materials/Methods:
We performed a secondary analysis of the TROG 12.01 trial. Symptom severity (0-10) was extracted from a priori-selected MDASI-HN items: taste, swallowing, dry mouth, fatigue, and pain. Assessments were obtained before, during (week 1-7) and after RT (1,3,5,9 and 13 weeks and 6, 12 and 24 months). Taste-bud mucosa, pharyngeal constrictors, submandibular and parotid glands, brainstem, and cerebellum were contoured. Latent class growth mixture modelling (LCGMM) was utilised to identify distinct symptom trajectories; model selection followed pre-specified statistical criteria with clinical interpretability considered. Sociodemographic, treatment and dosimetric factors (mean dose, V10-V70) were compared across trajectory classes.Results:
LCGMM identified 2-4 trajectories for each symptom: taste (3), swallowing (3), dry mouth (3), pain (2) and fatigue (4). Trajectories differed by peak severity, recovery kinetics and residual long-term symptom burden. Almost all models had trajectories with minimal pre-RT symptom scores (=1). For taste, all three classes (TAS1-3) showed minimal pre-RT symptoms (mean score all <1): TAS1 (low acute, n=56), TAS2 (high peak, rapid recovery, n=92), and TAS3 (high peak, slow recovery, n=34). Peak symptoms occurred at the end of RT, and were substantially lower for TAS1 (mean 4.6, 95% CI 4.2-5.0) than TAS2 (9.0, 95% CI 8.6-9.3) and TAS3 (9.7, 95% CI 9.1-10). TAS1 recovered and stabilised by 9 weeks post RT (1.9, 95% CI 1.5-2.4) through to 24 months (1.6, 95% CI 1.0-2.2); TAS2 improved to 9 weeks post RT (2.9, 95% CI 2.5-3.3) with slower gains to 24 months (1.1, 95% CI 0.7-1.6), approximating TAS1 by 6 months post RT. TAS3 had the highest peak and a persistently poorer recovery, reporting significantly worse symptoms than TAS1/2 (9 weeks: 6.3, 95% CI 5.6-7.1; 24 months: 3.0, 95% CI 2.0-3.9). Age, ECOG, treatment arm (cetuximab vs. cisplatin) and baseline depression/anxiety differed across taste classes; taste bud dose did not. At acceptance, all symptom trajectories will be described in full. Briefly, swallowing classes differed by age, gender, ECOG, RT laterality and constrictor dose (V50/V60); dry mouth classes differed by age and contralateral parotid dose (V10, V20, V70). Pain classes varied only by baseline anxiety/depression. Fatigue classes differed by ECOG, treatment arm (cisplatin/cetuximab), baseline anxiety/depression scores, but not cerebellum or brainstem dose.Conclusion:
Distinct latent symptom trajectories after RT for HPVOPC provide clinically relevant prognostic information for counselling. Baseline anxiety and depression scores were commonly associated with subsequent symptom trajectory membership.