Journal of Clinical Oncology, Vol 6, 282-290, Copyright © 1988 by American Society of Clinical Oncology
Prediction of long-term survival by flow cytometric analysis of cellular DNA content in patients with advanced ovarian cancer
ML Friedlander, DW Hedley, C Swanson and P Russell
Ludwig Institute for Cancer Research (Sydney Branch), University of Sydney, NSW Australia.
The prognostic value of cellular DNA content in ovarian cancer (malignant
common epithelial tumors) was investigated by flow cytometric analysis of
paraffin-embedded tumor blocks from 128 previously untreated patients with
International Federation of Gynecology and Obstetrics (FIGO) stage III and
IV ovarian cancer entered in a prospective clinical trial of combination v
sequential therapy with chlorambucil and cisplatin. Seventy-three percent
of tumors were aneuploid and 27% were diploid. Multivariate analysis using
a Cox model showed that cellular DNA content (P less than .001) and FIGO
stage (P less than .02) were the only significant independent prognostic
variables. The median survival was 13 months for patients with aneuploid
tumors and 60 months for patients with diploid tumors (P less than .0001).
Further analysis indicated that the good prognosis associated with diploid
tumors was limited to patients with stage III disease, all patients with
stage IV (spread beyond the peritoneal cavity or liver metastases) disease
having a poor prognosis irrespective of ploidy. On pathological review,
nine borderline ovarian tumors (of low malignant potential) were
identified, and seven of these were diploid. These tumors have an unusually
favorable prognosis, despite apparent dissemination within the peritoneal
cavity, a paradox which is often difficult to explain using conventional
histological criteria. Although the vast majority of tumors in this study
(93%) were classified as invasive epithelial ovarian cancers, it is
possible that some of the patients with stage III diploid tumors may have
had malignancies that were predominantly of low malignant potential, thus
accounting in part for the prognostic significance of DNA content. By
incorporating flow cytometric DNA analysis with careful histopathological
assessment, it may be possible to better identify patients with an
inherently good prognosis. This assumes particular importance, as the
relatively favorable prognosis of patients with stage III diploid ovarian
tumors appears to be independent of the type of treatment.