Combined Spatial and Dosimetric Recurrence Pattern Analysis in Head and Neck Squamous Cell Carcinoma Following Postoperative (Chemo)radiotherapy

TL;DR

This study examines recurrence patterns in head and neck cancer after surgery and radiotherapy to guide safer, more effective treatment strategies.

Contribution

The study identifies high-dose recurrence dominance and supports volumetric de-escalation in postoperative radiotherapy for HNSCC.

Findings

Most recurrences occurred in high-dose areas, suggesting biological resistance is a major cause of treatment failure.

Low rates of isolated nodal recurrences support the feasibility of reducing radiation in electively treated neck regions.

HPV-negative tumors may also benefit from de-escalation if accurate nodal staging is performed.

Abstract

Advancements in nodal staging for head and neck squamous cell carcinoma (HNSCC) have prompted radiotherapy de-escalation trials to reduce irradiation of electively treated neck regions, with the goal of improving treatment tolerability. While volumetric de-escalation has shown promise in definitive radiotherapy of HNSCC, limited data exist regarding its safety in the postoperative treatment setting. This study aimed to assess dose-level-specific locoregional recurrence patterns following standard postoperative (chemo)radiotherapy in a mixed HNSCC cohort to inform risk-adaptive radiotherapy strategies.

We retrospectively reviewed 203 HNSCC patients (75% HPV-negative, 25% HPV-positive) treated with curative intent postoperative (chemo)radiotherapy from 2017 to 2021. Recurrence imaging was co-registered with planning CT, and recurrent tumor volumes were dosimetrically compared to the target volume dose and spatially analyzed using a center-of-mass-based approach. We classified five recurrence types: A (central high-dose), B (peripheral high-dose), C (central intermediate- or low-dose), D (peripheral intermediate- or low-dose), and E (extraneous dose).

With a median follow-up of 39.7 months, the three-year local, regional, and distant control of HPV-negative HNSCC were 84%, 87%, and 87%, respectively. Of 56 recurrences, 17 were local, 13 regional, 3 locoregional, 9 combined local/regional with concomitant distant failure, and 14 distant only. Of 40 analyzed recurrences, we identified 47.5% as type A/B, 5% as type C/D intermediate-dose, and 20% as type E, half of which were secondary cancers. Among the 27.5% (11/40) type C/D low-dose recurrences in the elective target volume, 15% (6/40) were true nodal failures, resulting in an overall elective neck failure rate of 3% (6/203).

The predominance of high-dose recurrences suggests that biological tumor resistance is a key driver of treatment failure, highlighting the necessity to refine postoperative risk stratification and integrate tumor biology into dose escalation decisions. The low incidence of isolated nodal recurrences in electively treated neck regions supports the feasibility of volumetric de-escalation of postoperative radiotherapy. This approach might not only be feasible for HPV-associated oropharyngeal cancers but also for HPV-negative tumors, provided that accurate nodal staging has been conducted.

The online version contains supplementary material available at 10.1186/s13014-025-02641-8.

Figures