Randomized controlled trial of semi-individualized 3D-printed tissue retraction devices vs. standard shielding splints in head and neck cancer treated with intensity-modulated and particle radiotherapy (GUARD).

Background And Purpose: This randomized controlled trial evaluated whether tissue-retraction devices (TRDs) lessen severe radiation-induced oral mucositis (RIOM) versus standard flexible intraoral shielding splints (IOS) in patients receiving intensity-modulated radiotherapy (IMRT), carbon-ion RT, or combined-modality RT for head-and-neck cancer (HNC). Although TRDs may spare normal tissue, randomized evidence with an active control under contemporary RT techniques remains scarce.

Materials And Methods: HNC patients requiring RT at University Hospital Heidelberg (2020-2023) were randomized 1:1 to receive a semi-individualized, 3D-printed TRD or standard IOS. The primary outcome was severe RIOM (RTOG/CTCAE grade ≥3) at RT completion, assessed by video endoscopy in a predefined region adjacent/opposite to the clinical target volume. Secondary endpoints included acute and late toxicities, patient-reported outcomes (PROs), and dosimetric comparisons of organs at risk, including the tongue and palate.

Results: Of 34 randomized patients, 29 were evaluable (225 regions assessed). TRD significantly reduced the rate of RIOM grade ≥3 with 0/13 TRD vs. 9/16 IOS patients affected (p < 0.001). Grade 1-2 RIOM occurred in 7 (54 %) TRD vs. 2 (12 %) IOS. Among patients with tumors above the oral cavity, TRD significantly reduced tongue mean dose (median Dmean: 16.71 Gy [IQR: 2.41-20.71] vs. 29.26 Gy [IQR: 21.56-42.52]; p = 0.034). PROs trended toward less oral pain and swallowing difficulty with TRD.

Conclusion: TRD reduced severe RIOM in displaced normal tissues while improving tongue and palate dosimetry. Findings support TRD integration into clinical practice, though larger trials are needed to assess long-term patient-reported benefits and large-scale implementation.

Trial Registration: NCT04454697.