A Probabilistic Model of Bilateral Lymphatic Spread in Head and Neck Cancer

TL;DR

This paper presents a probabilistic model for predicting bilateral lymphatic spread in head and neck cancer, enabling personalized treatment planning by estimating occult disease risks in lymph node levels based on tumor characteristics.

Contribution

The study extends a hidden Markov model to include contralateral lymphatic spread, providing a data-driven tool for tailored radiotherapy target volume decisions in OPSCC.

Findings

Midline tumor extension is the main risk factor for contralateral involvement.

Occult disease in contralateral LNL III is rare if LNL II is negative.

Limiting CTV-N to LNL II in certain cases can reduce tissue exposure without compromising control.

Abstract

Current guidelines for elective nodal irradiation in oropharyngeal squamous cell carcinoma (OPSCC) recommend including large portions of the contralateral lymph system in the clinical target volume (CTV-N), even for lateralized tumors with no clinical lymph node involvement in the contralateral neck. This study introduces a probabilistic model of bilateral lymphatic tumor progression in OPSCC to estimate personalized risks of occult disease in specific lymph node levels (LNLs) based on clinical involvement, T-stage, and tumor lateralization. Building on a previously developed hidden Markov model for ipsilateral spread, we extend the approach to the contralateral neck. The model represents LNLs I, II, III, IV, V, and VII on both sides of the neck as binary hidden variables (healthy/involved), connected via arcs representing spread probabilities. These probabilities are learned using Markov chain Monte Carlo (MCMC) sampling from a dataset of 833 OPSCC patients, enabling the model to reflect the underlying lymphatic progression dynamics. The model accurately and precisely describes observed patterns of involvement with a compact set of interpretable parameters. Midline extension of the primary tumor is identified as the primary risk factor for contralateral involvement, with advanced T-stage and extensive ipsilateral involvement further increasing risk. Occult disease in contralateral LNL III is highly unlikely if upstream LNL II is clinically negative, and in contralateral LNL IV, occult disease is exceedingly rare without LNL III involvement. For lateralized tumors not crossing the midline, the model suggests the contralateral neck may safely be excluded from the CTV-N. For tumors extending across the midline but with a clinically negative contralateral neck, the CTV-N could be limited to LNL II, reducing unnecessary exposure of normal tissue while maintaining regional tumor control.