Correlation of MUC4 expression with clinical outcomes may establish MUC4 as a potential molecular prognostic marker for these tumors.

Median survival time was 24 months (range, 2-60 months) among the nine patients who died, whereas median follow-up time in the remaining 19 patients is 33.4 months (range, 4.7-73 months). A trend toward a reduction in MUC4 antigen expression in high-grade tumors (55% expression) compared with low-grade (91% expression) and intermediate-grade (100% expression) tumors is identified (chi square, p =.0975). Patients with tumors expressing MUC4 antigens are at reduced risk of death (hazard ratio [HR], 0.20; p =.0531). Adjustment for pathologic grade, T stage, and age results in a much higher risk of death for patients whose tumors do not express MUC4 antigens, although this does not meet statistical significance (HR, 26.6; p =.1). Analysis of recurrence adjusting for T stage reveals that patients whose tumors do not express MUC4 antigens are at increased risk of recurrence compared with patients whose tumor expresses MUC4 antigens (HR, 6.37; p =.03). ErbB2 receptor staining is noted in seven of 28 patients, with five of these seven showing 2+ and 3+ membrane-staining patterns. Adjustment for pathologic grade and age suggests that patients whose tumors express high levels of ErbB2 (2+, 3+) are at increased risk of death compared with patients with low or no expression of ErbB2 (HR, 2.29; p =.32). MUC4 antigen positivity is seen in two of the five cases with 2+ and 3+ staining for ErbB2. CONCLUSIONS.: These findings suggest MUC4 antigen positivity is associated with reduced risk of death and reduced risk of recurrence and may identify a subset of patients with more favorable prognosis. Although limited by small sample size, analysis reveals ErbB2 overexpression is not consistently associated with MUC4 antigen positivity and might be associated with increased risk of death.

No abstract

Chromosome l lq13 amplification has been found frequently in head and neck squamous cell carcinomas (HNSCCs), indicating poor prognosis. The EMS1 gene has been identified as one of the "active" oncogenes in this region because it is overexpressed in all carcinomas with EMS1 amplification, and this overexpression causes increased cell motility and invasion in vitro. However, little is known about the prognostic significance of individual EMS 1 amplification in HNSCC. Our objective was to further address the frequency of EMS 1 amplification in HNSCC and to evaluate its correlation with clinicopathologic variables, relapse, and survival. Methods: Tissue samples from 104 consecutive patients with primary HNSCC who underwent surgical resection of their tumors were studied. After DNA extraction, the EMS 1 copy number in tumor samples was estimated with the polymerase chain reaction method. The presence or absence of amplification was correlated with the anatomic site, T stage, nodal involvement, pathologic grade, disease recurrence, distant metastasis, clinical outcome (disease status), and survival. Patients were observed for at least 36 months (mean followup 54 months). Results: EMS1 amplification was identified in 21 cases (20%). Amplification was related with advanced T stages (P = 0.003), lymph node involvement (P = 0.02), and poor histologic differentiation (P = 0.018). Recurrent disease was identiffed in 76% (16/21) of cases with amplification and 39% (32/83) of cases without amplification (P = 0.006). Seventeen patients (81%) with EMS1 amplification and 35 patients (42%) without amplification died of disease (P = 0.003). EMS1 amplification status, individually, had a significant effect on survival (the 3-year survival was 54% in nonamplifled cases and 7% in amplified cases, P < 0.0001). In multivariate analysis (Cox regression model), EMS1 amplification was found to be an independent predictor of mortality by the tumor (P = 0.003). Conclusion: Amplification of the EMS1 gene seems to be an important biological marker indicating poor prognosis in HNSCC, independent of clinical stage, histologic differentiation, or anatomic site. This may confirm its role as an "active" oncogene in the llq13 region.