Advanced-stage (FIGO III-IV) epithelial ovarian cancer: Multivariate analysis of prognostic factors in an area without a tertiary referral oncology center. A population-based study

Background: Most of the time, prognostic factors of patients with epithelia l ovarian cancer (EOC) have been evaluated in hospital-based populations, m ostly in secondary or tertiary referral centers. The aim of the study was t o analyze these factors in a well-defined and nonselected population with c omplete follow-up from a cantonal cancer registry and to assess the adequac y of treatment in an area which has no central institution for oncological treatment. Patients and Methods: From 1989 to 1995, the cantonal cancer reg istry of the Valais has registered 73 patients with ovarian epithelial canc ers of stages III and IV; 72 patients have been evaluable. We have calculat ed survival rates and analyzed variables such as age (<55 / 55-70 / >70 yea rs), stage (III/IV), DNA ploidy (diploid/aneuploid), residual tumor after s urgery (tumor less than or equal to 2 cm / less than or equal to 2 cm), his tologic grading (G1/G2/G3) and chemotherapy treatment (curative/palliative) . The median follow-up was 25 (range 18-101) months. Multivariate analyses (Cox's proportional hazard) were used to identify an independent effect of each variable on survival time. Survival rates were calculated according to the Kaplan and Meier method. Results:The Canton Valais has no tertiary cen ter and no trained gynecological (surgical) oncologist. The patients have b een operated in 7 regional hospitals and one private clinic by 16 different gynecologists and 8 general surgeons, chemotherapy regimens have been give n by 5 medical oncologists. Both treatments were very heterogeneous. The me dian age was 63 years. The estimated 5-year survival rate was 30%. Multivar iate analyses identified age and stage as the only significant prognostic f actors associated with survival. DNA ploidy, size of residual disease after primary surgery, histologic grading and chemotherapy had no significant im pact on survival. Conclusions:Variables such as primary treatment (surgical and chemotherapy) do not improve survival time, probably because patients had too heterogeneous surgical and chemotherapy treatments, the choice of w hich was influenced by the many treating physicians. Patients with ovarian cancer should be referred to central institutions where they are more likel y to have standardized and optimal surgery and chemotherapy treatment.