Background: Histologic assessment on the percentage of remaining cells in esophagectomy specimens has been shown to further stratify patients with an incomplete pathologic response after preoperative therapy. MIB-1 is an antibody to Ki-67, a marker for tumor proliferation and in pretreatment biopsies has been shown to be an independent predictor of pathologic response and survival.
Aims: To correlate MIB-1 indices with histological and pathological response of tumor cells in patients undergoing preoperative chemoradiation for EC and to determine if MIB-1 staining was associated with survival.
Methods: 172 consecutive patients with potentially curable, locally advanced (EC) were enrolled. 89 patients underwent esophagectomy. MIB-1 staining was performed on esophagectomy specimens by a single experienced pathologist and quantified both categorically and as the percent of cells staining positive. Specimens were assessed for a “sterilization effect” (microscopic nuclear, nucleolar, cytoplasmic, and chromatin changes when compared to apparently unaffected cancer cells). Chi-square or Fisher exact tests were used to assess the association between MIB-1 status and other categorical variables (gender, sterilization effect, and pathologic response). Kruskal-Wallis tests were used to assess differences in continuous variables between MIB-1 positive/negative patients and to assess the difference in MIB-1 % between levels of categorical variables. Correlation coefficients were used to quantitate the degree of association between MIB-1 percent and other continuous variables. Cox regression was used to determine the effect of MIB-1 on disease-free and overall survival.
Results: MIB-1 staining was performed on 32 patients: 23 positive, 5 negative, 4-non-reactive. MIB-1 was not significantly associated with age, race, gender, SUV, or change in SUV between PET scans. However, MIB-1 staining was significantly associated with sterilization effect (p = .001). The 17 patients with a positive sterilization effect had significantly lower median MIB-1 percent (5%) compared to the 15 patients with a negative sterilization effect (37.5%). The risk of death increased by 13% for every 10% increase in the percent positive MIB-1 staining, although this effect was not significant given our small sample size (p = .33).
Conclusions: Increased MIB-1 staining reflects residual tumor viability in esophageal cancer patients after preoperative therapy. MIB-1 can help determine which patients experience a sterilization effect after pre-operative radiochemotherapy for EC. In patients experiencing a sterilization effect, the low MIB-1 may indicate cell death in cycling cells.