Simultaneous cytoplasmic and nuclear protein expression of melanoma antigen-A family and NY-ESO-1 cancer-testis antigens represents an independent marker for poor survival in head and neck cancer

Abstract

The prognosis of head and neck squamous cell carcinoma (HNSCC) patients remains poor. The identification of high-risk subgroups is needed for the development of custom-tailored therapies. The expression of cancer-testis antigens (CTAs) has been linked to a worse prognosis in other cancer types; however, their prognostic value in HNSCC is unclear because only few patients have been examined and data on CTA protein expression are sparse. A tissue microarray consisting of tumor samples from 453 HNSCC patients was evaluated for the expression of CTA proteins using immunohistochemistry. Frequency of expression and the subcellular expression pattern (nuclear, cytoplasmic, or both) was recorded. Protein expression of melanoma antigen (MAGE)-A family CTA, MAGE-C family CTA and NY-ESO-1 was found in approximately 30, 7 and 4% of tumors, respectively. The subcellular expression pattern in particular had a marked impact on the patients' prognosis. Median overall survival (OS) of patients with (i) simultaneous cytoplasmic and nuclear expression compared to (ii) either cytoplasmic or nuclear expression and (iii) negative patients was 23.0 versus 109.0 versus 102.5 months, for pan-MAGE (p < 0.0001), 46.6 versus 50.0 versus 109.0 for MAGE-A3/A4 (p = 0.0074) and 13.3 versus 50.0 versus 100.2 months for NY-ESO-1 (p = 0.0019). By multivariate analysis, these factors were confirmed as independent markers for poor survival. HNSCC patients showing protein expression of MAGE-A family members or NY-ESO-1 represent a subgroup with an extraordinarily poor survival. The development of immunotherapeutic strategies targeting these CTA may, therefore, be a promising approach to improve the outcome of HNSCC patients.

Keywords: NY-ESO-1; cancer-testis antigens; head and neck squamous cell carcinoma (HNSCC); melanoma antigen (MAGE); tissue microarray (TMA).