Survival and prognostic factors of early ovarian cancer.

Survival and prognostic factors were analysed in 150 patients with histologically confirmed epithelial ovarian cancer stage IA-IIA. The relapse-free and overall survival rates were, respectively, 81% and 88% after 3 and 74% and 84% after 5 years. The analysis of various prognostic factors indicates as the main factor the grade differentiation of the tumour.

Few studies have been conducted on prognostic factors in earlystage (I-IIA) ovarian cancer. Some authors have suggested that age, tumour grade, stage and rupture of the tumour capsule have a relevant impact on the survival of early ovarian cancer patients (Webb et al, 1973;Einhom et al, 1985). A recent study including 194 patients with stage I ovarian cancer identified grade, presence of ascites and surface tumour as independent prognostic factors (Ahmed et al, 1996). Other studies, however, did not confirm these indications (Dembo et al, 1986(Dembo et al, , 1990. To provide further information, we considered the prognostic factors in a series of patients with early ovarian cancer. Between 1981 and1996, 150 consecutive patients with histologically confirmed epithelial ovarian cancer stage IA-IIA (median age 53 years, range 23-76 years) were observed at the First Obstetric and Gynecologic Clinic of the University of Milan. At first diagnosis all patients underwent a total abdominal hysterectomy, bilateral salpingo-oophorectomy, omentectomy and at least eight multiple biopsies plus washing of the abdominal cavity. The patients with conservative surgery are not included in this analysis.

RESULTS
The patients' characteristics are shown in Table 1. Stage was IA in 51 cases, IB in ten cases, IC in 78 cases and IIA in 11 cases. After surgery, 75 patients were treated with chemotherapeutic regimens including cisplatinum alone or in combination with cyclophosphamide, 58 did not receive any treatment and in 17 patients treatment with P32 was performed. All patients treated with adjuvant chemotherapy were stage IA-IB G2-G3, or IC-IIA all grades.
Survival time was computed from the date of diagnosis until date of death (or of clinical or pathological relapse for disease-free survival) or censored to the last follow-up data. No patient was lost to follow-up. At the cut-off date, September 1996, the median of follow-up was 70 months (range 9-181 months). Survival Received 28 Feburary 1997Revised 19 May 1997 Accepted 10 June 1997 Correspondence to: F Parazzini probabilities were estimated according to Kaplan-Meier and compared using the log-rank test. The Cox model (Cox, 1972) was fitted to the data after graphical check of proportional hazards assumption to evaluate the prognostic value of considered factors, taking into account the potential reciprocal confounding effect.
At the time of the study analysis, 39 women had relapsed and 29 had died (27 out of the 39 who relapsed and 2 out of Ill with no recurrence of disease for causes unrelated to ovarian cancer).
The relapse-free and overall survival rates were, respectively, 81% and 88% after 3 years and 74% and 84% after 5 years.
Patients 53 years old or less tended to have a better 5-year survival rate (88%) than the older ones (79%; log-rank test P = 0.04, multivariate analysis P = 0.07). Women with clear-cell and undifferentiated tumour type had a 5year survival of 66% compared with 88% for women with other histological types (log-rank test P = 0.002), but this difference was not significant in the multivariate analysis (P = 0.59). Tumour grade 3 had a 69% 5-year survival rate, compared with 91% for grade 1-2 (log-rank test P = 0.0004; multivariate analysis, P = 0.04).
Compared with patients with stage IC, there was a significant difference in survival after 5 years between women with presence of tumour with or without ascites on the surface of the ovary (88%) or intraoperative rupture of the tumour capsule (72%) (data not shown in Table, multivariate analysis P = 0.03).
The site of relapse was the pelvis and/or the vaginal cuff in 16 (41%) cases and extrapelvic in 23 (59%) cases. No relationship emerged between site of relapse and age, histotype, grade, stage and tumour volume at first surgery.
Out of 39 patients with recurrent disease, 24 were treated with chemotherapy alone, eight with chemotherapy plus surgery and/or radiotherapy and three with radiotherapy alone; four patients had no treatment. The response to treatment at relapse was 'complete' in 11 patients (six out of the 16 patients with pelvic and/or vaginal cuff relapse and five out of the 23 patients with extrapelvic relapse), 'partial' in seven 'no change' in seven and ten cases progressed. Patients with recurrent disease had a 47% 3-year survival rate after relapse.

DISCUSSION
The results of this study show that the prognosis of patients with early-stage ovarian cancer is favourable. This finding is consistent with previous published studies that showed 5-year survival rates raging from 75% to 90% (Webb et al, 1973;Einhorn et al, 1985;Sevelda et al, 1989;Dembo et al, 1990;Lund & Williamson, 1991;Shueler et al, 1993;Ahmed et al, 1996). The analysis of various prognostic factors indicates as main factor the grade differentiation of the tumour. This finding is consistent with the results of previous studies (Webb et al, 1973;Sevelda et al, 1989;Ahmed et al, 1996). Thus, tumour grading is the main factor in the identification of patients with early ovarian cancer who could benefit from adjuvant chemotherapy.