OBJECTIVE: To determine whether DNA aneuploidy is a sensitive and specific marker for the identification of tumor cells in effusions and whether the pattern of DNA aneuploidy can provide important information for the differential diagnosis of primary and secondary tumors of the serous membranes. STUDY DESIGN: One hundred eight malignant mesotheliomas as well as 102 metastatic carcinomas of the serous membranes were obtained from routine cytologic and histologic material. One hundred reactive effusions were investigated as controls. Nuclear DNA contents were measured after Feulgen staining using a TV image analysis system. RESULTS: DNA aneuploidy was assumed if abnormal DNA stemlines, a coefficient of variation of the first DNA stemline > or = 10%, or cells > 9c were observed. On this basis the prevalence of DNA aneuploidy in mesotheliomas was 83% for cytologic and 84% for histologic material. In effusions of metastatic carcinomas it was 100%. None of the 100 reactive effusions revealed DNA aneuploidy (prevalence, 0%). Positive predictive value for mesotheliomas was 100%; negative predictive value was 88% for cytologic and 82% for histologic material. Positive predictive value for metastatic carcinomas was 100%; negative predictive value was 100%. Seventy-two percent of the mesotheliomas revealed their greatest stemline within the range 1.80c-2.20c, whereas none of the metastatic carcinomas showed this stemline position. CONCLUSION: DNA image cytometry might be a very sensitive and highly specific, additional tool for identification of neoplastic cells in effusions as well as for the differential diagnosis of mesothelioma vs. metastatic carcinoma of the serous membranes.
OBJECTIVE: To determine whether DNA aneuploidy is a sensitive and specific marker for the identification of tumor cells in effusions and whether the pattern of DNA aneuploidy can provide important information for the differential diagnosis of primary and secondary tumors of the serous membranes. STUDY DESIGN: One hundred eight malignant mesotheliomas as well as 102 metastatic carcinomas of the serous membranes were obtained from routine cytologic and histologic material. One hundred reactive effusions were investigated as controls. Nuclear DNA contents were measured after Feulgen staining using a TV image analysis system. RESULTS: DNA aneuploidy was assumed if abnormal DNA stemlines, a coefficient of variation of the first DNA stemline > or = 10%, or cells > 9c were observed. On this basis the prevalence of DNA aneuploidy in mesotheliomas was 83% for cytologic and 84% for histologic material. In effusions of metastatic carcinomas it was 100%. None of the 100 reactive effusions revealed DNA aneuploidy (prevalence, 0%). Positive predictive value for mesotheliomas was 100%; negative predictive value was 88% for cytologic and 82% for histologic material. Positive predictive value for metastatic carcinomas was 100%; negative predictive value was 100%. Seventy-two percent of the mesotheliomas revealed their greatest stemline within the range 1.80c-2.20c, whereas none of the metastatic carcinomas showed this stemline position. CONCLUSION: DNA image cytometry might be a very sensitive and highly specific, additional tool for identification of neoplastic cells in effusions as well as for the differential diagnosis of mesothelioma vs. metastatic carcinoma of the serous membranes.