Limitations of 8F1 Biomarker for Platinum-Resistance in HNSCC Are Explained by Co-detection of PCYT1A

Abstract

Objectives:

(1) Recognize how ERCC1 biomarker development in HNSCC has been impaired by use of 8F1, a nonspecific ERCC1 antibody, which also recognizes PCYT1A, a protein unrelated to DNA repair capacity. (2) Verify that specific ERRC1 antibodies, 4F9 and FL297, were more useful than 8F1 for prognosis of HNSCC patients undergoing cisplatin-radiation on a phase II trial. (3) Demonstrate the relative contribution of PCYT1A and ERCC1 in variation of 8F1 expression.

Methods:

We hypothesized that co-detection of PCYT1A could explain variation in 8F1 expression, and account for lost prognostic performance as compared with 4F9 and FL297. Pre-treatment tissue from 75 trial patients with previously characterized 4F9, FL297, and 8F1 expression was stained for PCYT1A. Quantitative digital IHC was performed, blinded to clinical outcomes. Expression was compared among the 4 antibodies.

Results:

High expression of 4F9 (hazard ratio [HR] 3.12; P = .02) and FL297 (HR 2.45; P = .025) were associated with poor progression-free survival (PFS). Neither 8F1 (HR 1.55; P = .29) nor PCYT1A expression (HR 1.89; P = .13) was associated with PFS. 4F9 and FL297 demonstrated high correlation (0.878). 8F1 demonstrated moderate correlation with 4F9 (0.55) and FL297 (0.467), and high correlation with PCYT1A (0.725). In a three-dimensional regression analysis, both 4F9 (T = 4.26, P < .001) and PCYT1A (T = 4.58, P < .001) were significant predictors of 8F1 variation, suggesting equivalent influence from ERCC1 and PCYT1A detection in 8F1 scoring.

Conclusions:

ERCC1 expression by specific antibodies is a resistance biomarker for platinum-based therapy in HNSCC. Although 8F1 detects ERCC1, cross-reactivity with PCYT1A confounds 8F1 scoring. Because of nonspecificity, 8F1 should be abandoned.