PURPOSE: Molecular approaches as supplements to cytological examination of malignant ascites may play an important role in the clinical management of cancer patients. HLA-G is a potential tumor-associated marker and that one of its isoforms, HLA-G5, produces a secretory protein. This study is to assess the clinical utility of secreted HLA-G levels in differential diagnosis of malignant ascites. EXPERIMENTAL DESIGN: We used ELISA to assess whether secretory HLA-G (sHLA-G) could serve as a marker of malignant ascites in ovarian and breast carcinomas, which represent the most common malignant tumors causing ascites in women. RESULTS: On the basis of immunohistochemistry, 45 (61%) of 74 ovarian serous carcinomas and 22 (25%) invasive ductal carcinomas of the breast demonstrated HLA-G immunoreactivity ranging from 2 to 100% of the tumor cells. HLA-G staining was not detected in a wide variety of normal tissues, including ovarian surface epithelium and normal breast tissue. Revese transcription-PCR demonstrated the presence of HLA-G5 isoform in all of the tumor samples expressing HLA-G. ELISA was performed to measure the sHLA-G in 42 malignant and 18 benign ascites supernatants. sHLA-G levels were significantly higher in malignant ascites than in benign controls (P < 0.001). We found that the area under the receiver-operating characteristic curve for sHLA-G was 0.95 for malignant versus benign ascites specimens. At 100% specificity, the highest sensitivity to detect malignant ascites was 78% (95% confidence interval, 68-88%) at a cutoff of 13 ng/ml. CONCLUSIONS: Our findings suggest that measurement of sHLA-G is a useful molecular adjunct to cytology in the differential diagnosis of malignant versus benign ascites.
PURPOSE: Molecular approaches as supplements to cytological examination of malignant ascites may play an important role in the clinical management of cancerpatients. HLA-G is a potential tumor-associated marker and that one of its isoforms, HLA-G5, produces a secretory protein. This study is to assess the clinical utility of secreted HLA-G levels in differential diagnosis of malignant ascites. EXPERIMENTAL DESIGN: We used ELISA to assess whether secretory HLA-G (sHLA-G) could serve as a marker of malignant ascites in ovarian and breast carcinomas, which represent the most common malignant tumors causing ascites in women. RESULTS: On the basis of immunohistochemistry, 45 (61%) of 74 ovarian serous carcinomas and 22 (25%) invasive ductal carcinomas of the breast demonstrated HLA-G immunoreactivity ranging from 2 to 100% of the tumor cells. HLA-G staining was not detected in a wide variety of normal tissues, including ovarian surface epithelium and normal breast tissue. Revese transcription-PCR demonstrated the presence of HLA-G5 isoform in all of the tumor samples expressing HLA-G. ELISA was performed to measure the sHLA-G in 42 malignant and 18 benign ascites supernatants. sHLA-G levels were significantly higher in malignant ascites than in benign controls (P < 0.001). We found that the area under the receiver-operating characteristic curve for sHLA-G was 0.95 for malignant versus benign ascites specimens. At 100% specificity, the highest sensitivity to detect malignant ascites was 78% (95% confidence interval, 68-88%) at a cutoff of 13 ng/ml. CONCLUSIONS: Our findings suggest that measurement of sHLA-G is a useful molecular adjunct to cytology in the differential diagnosis of malignant versus benign ascites.
Authors: Donna E Hansel; Ayman Rahman; Robb E Wilentz; Ie-ming Shih; Michael T McMaster; Charles J Yeo; Anirban Maitra Journal: Int J Gastrointest Cancer Date: 2005