3746 - Depth of Dose-Response for Patient-Reported Outcomes in Two Elective Neck Ee-Escalation Trials

A. Tuo¹, D. H. Moon², V. Avkshtol¹, N. L. Pham Jr³, J. L. Shah⁴, and D. J. Sher²; ¹University of Texas Southwestern Medical Center, Dallas, TX, ²Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX, ³Department of Radiation Oncology, UT Southwestern Medical Center, Dallas, TX, ⁴Department of Radiation Oncology, University of Michigan, Ann Arbor, MI

Purpose/Objective(s): Patients with head and neck squamous cell carcinoma (HNSCC) receiving IMRT experience significant acute and late toxicities. The extent to which PROs may improve at lower doses is unclear, and such information is critical for optimal planning. In this study, we evaluated the relationship between PROs and OAR doses in two elective neck irradiation (ENI) de-escalation trials with resultant low dose distributions. Materials/

Methods: Eligible patients in this retrospective analysis had a diagnosis of oropharynx, larynx and hypopharynx HNSCC treated on two de-escalation trials, one reducing ENI dose (40 Gy) and volume and the other eliminating ENI. Instruments included the MD Anderson Dysphagia Inventory (MDADI) and EORTC QLQ-30 and HN35. Contoured OARs included oral cavity (OralCav), oral tongue (OTongue), floor of mouth, cervical esophagus (Eso), larynx, superior/middle constrictors (SMConstrict), inferior constrictors (InfConstrict), parotid glands (PG), submandibular glands (SMG), and tubarial glands (TG). OARs included PTV overlap. Pearson correlations analyzed continuous PRO and OAR metrics. OAR metrics were dichotomized at the median and in tertiles to determine PROs. The primary endpoint was 12-month PROs.
Results: The analysis include 130 patients (94 with oropharynx cancer [OPC]). The median/lower tertile cutpoints (Gy) were: OralCav (24.7/18.2), contralateral (CL) PG (14.6/12), ipsilateral (IL) PG (23.5/16.9), CL SMG (34.8/32), SMConstrict (44.6/40.1), InfConstrict (15.8/11.5), and Eso (6.8/4.6). MDADI, EORTC dry mouth (DM), and EORTC sticky saliva (SS) scores were improved with lower dose distributions to certain OARs in both the whole population and OPC-only (Table 1). MDADI scores were further improved for patients in both populations with doses in the lowest tertile for InfConstrict (91.5 vs. 86.0, p=0.03) and Eso (93.1 vs. 85.6, p<0.01). Only the lowest tertile of CL PG mean was associated with DM in the whole cohort. The lowest tertile of CL SMG mean was associated with SS in both populations, and the lowest tertile of OTongue mean was associated with SS among OPC patients. There were no significant relationships between TG and any outcome.
Conclusion: Even in a cohort of de-escalated patients with low doses to OARs, PROs improve with more favorable dose distributions. The novel TG was not associated with any PRO, but Eso may be an underappreciated OAR. More extreme dose reduction to a few OARs yielded further improvements in PROs, highlighting the need for innovation to further optimize post-treatment quality-of-life. Abstract 3746 – Table 1: Relationships between PRO severity and dose delivered to OARs

PRO		OAR (Threshold, Gy)	Score > median	Score < median	p-value
MDADI	Whole Cohort	SMConstrict (44.6)	94.0	87.6	<0.01
		Eso (6.8)	91.4	84.6	0.01
	OPC Only	OralCav (24.7)	94.6	87.6	<0.01
		SMConstrict (44.6)	94.0	87.2	<0.01
		Eso (6.8)	93.0	86.3	0.01
EORTC DM	Whole	CL PG (14.6)	40.2	29.6	0.048
		CL SMG (34.8)	41.9	28.3	<0.01
	OPC	CL SMG (34.8)	41.2	29.7	0.048
EORTC SS	OPC	CL SMG (34.8)	27.5	14.8	0.04