Integrating epidermal growth factor receptor assay with clinical parameters improves risk classification for relapse and survival in head-and-neck squamous cell carcinoma

Abstract

Purpose: Epidermal growth factor receptor (EGFR) overexpression has been consistently found to be an independent predictor of local-regional relapse (LRR) after radiotherapy. We assessed the extent by which it can refine risk classification for overall survival (OS) and LRR in patients with head-and-neck squamous cell carcinoma (HNSCC).

Methods and materials: EGFR expression in locally advanced HNSCC was measured by immunohistochemistry in a series of patients randomized to receive accelerated or conventional radiation regimens in a Phase III trial. Subsequently, data of the two series were pooled (N = 533) for conducting a recursive partitioning analysis that incorporated clinical parameters (e.g., performance status, primary site, T and N categories) and four molecular markers (EGFR, p53, Ki-67, and microvessel density).

Results: This study confirmed that patients with higher than median levels of tumor EGFR expression had a lower OS (relative risk [RR]: 1.90, p = 0.0010) and a higher LRR (RR: 1.91, p = 0.0163). Of the four markers analyzed, only EGFR was found to contribute to refining classification of patients into three risk classes with distinct OS and LRR outcomes. The addition of EGFR to three clinical parameters could identify patients having up to a fivefold difference in the risk of LRR.

Conclusions: Adding pretreatment EGFR expression data to known robust clinical prognostic variables improved the estimation of the probability for OS and LRR after radiotherapy. Its use for stratifying or selecting patients with defined tumor feature and pattern of relapse for enrollment into clinical trials testing specific therapeutic strategy warrants further investigation.

Figure 1

EGFR assay reproducibility and correlations between EGFR expression and treatment outcome. A: EGFR expression measured by mean optical density (MOD - optical densities measured over the labeled areas within the structure) and staining index (SI - the proportion of stained area relative to the total area of the structures). The continuous distributions of EGFR SI and MOD scores found in the samples from AFX-C arm were analogous to our previously reported series (13) from SFX arm (p=0.62 for SI; p=0.85 for MOD). B: Scatter plot showing that the vast majority of tumors with SI of over 80% recurred at local-regional sites after both fractionation regimens. C: The rates and relative risks of local-regional relapse in high versus low EGFR-expressing tumors after treatment with standard (SFX) or accelerated fractionation regimen (AFX-C).

Figure 2

Recursive partitioning analysis trees, A: without EGFR, and B: with EGFR.

Figure 3

A: Overall survival curves, and B: cumulative incidences curves for local-regional relapse of three distinct outcome classes (Class I – Best, Class II – Intermediate, and Class III – Worst) derived from RPA analysis in training and test sets.