The treatment of early stage epithelial ovarian cancer must be individualized. Thorough surgical exploration and staging are indicated for all patients with early stage disease. Adjuvant treatment with chemotherapy or radiotherapy is appropriate for those women at highest risk of recurrence.

Surgery
Properly staged early disease can be managed conservatively. The primary treatment for invasive stage I epithelial ovarian cancer is surgical, that is, the performance of a total abdominal hysterectomy, bilateral salpingo-oophorectomy, and surgical staging. In certain circumstances, a unilateral oophorectomy may suffice, as discussed below.

Radiotherapy
There have been two randomized studies of external beam radiotherapy in stage I epithelial ovarian cancer. Both compared pelvic radiotherapy with no postoperative treatment. These trials suggested that pelvic irradiation reduced the rate of pelvic relapses, but because relapses occurred throughout the peritoneal cavity, this was of no therapeutic benefit. Abdominopelvic radiotherapy has not been studied in a Phase III trial in stage I, but was compared with pelvic radiotherapy or to no treatment. No benefit was found in grade 1, where the risk of relapse was under 5% overall. In grades 2 and 3, a nonsignificant reduction in relapse risk was observed. A significant reduction in relapse risk was seen in patients whose tumors were densely adherent, but these patients are more correctly classified in stage II.

Chemotherapy
Some researchers have questioned the wisdom of chemotherapy in women with early stage disease, suggesting that the evidence for a durable impact on survival is marginal. Furthermore, the risks of leukemia with alkylating agents and cisplatin argue against the administration of adjuvant therapy unless there is a significant benefit. Two major prospective randomized clinical trials were conducted over a 10-year period that evaluated the efficacy of adjuvant therapy for localized ovarian cancer in patients with favorable and unfavorable prognoses. In the first trial, patients with favorable prognostic features were randomized either to no adjuvant therapy or to intermittent oral melphalan (0.2 mg/kg daily for 5 days every 4 to 6 weeks for up to 12 cycles). The second trial compared adjuvant melphalan (as described above) with a single intraperitoneal dose of 15 mCi of ₃₂P in patients with unfavorable prognostic features. Both trials were performed by the Gynecologic Oncology Group (GOG), required a comprehensive staging laparotomy, and were reported at a time when median follow-up was greater than 6 years. In these studies, conclusions were weakened by the exclusion of nearly half of the randomized patients from the analysis, but a nonsignificant advantage for the melphalan-treated patients was claimed. Two follow-up studies of comprehensively staged patients were subsequently performed. In the first study, 81 of 92 randomized patients were evaluable who had stage I cancers of grades 1 and 2, without ascites, tumor rupture, or excrescences. When observation was compared with melphalan, a small, nonsignificant benefit for melphalan was observed (a 93% versus an 88% 10-year relapse-free rate). In the second study, 143 patients were randomized, including all the remaining classes of stage I and stage II with small or no residual tumor. In assessing these patients, 141 were evaluable for the comparison of melphalan and intraperitoneal radiocolloid. No significant differences were observed (10-year survival rates were 65% and 69%, respectively), and two leukemic deaths occurred in the melphalan arm. The overall survival for all 222 patients entered on this study is shown in

Figure 118-6. There were no significant differences in 5-year survival for patients with a favorable prognosis who received no adjuvant therapy (94%) or melphalan (98%). In patients with unfavorable clinicopathologic features, overall survival was substantially worse. There were no differences, however, in 5-year survival for patients treated with intermittent melphalan or intraperitoneal ₃₂ P: 81% vs 78%, respectively.

In recent years, based upon experience in treating patients with advanced ovarian cancer, cisplatin, carboplatin, cyclophosphamide and paclitaxel have been administered, individually and in combination, to patients with early stage disease. Several series reported outcomes when cisplatin and/or cyclophosphamide have been used to treat patients with stage I disease. In a GOG trial that compared 3 cycles of cisplatin and cyclophosphamide to intraperitoneal ₃₂P in patients with Stage Ib and Ic disease, the progression-free survival of women receiving the platinum-based chemotherapy was 31% higher than that of women who received the radiocolloid. The Gruppo Italiano Collaborativo Oncologica Ginecologica (GICOG) has reported two studies that evaluated treatment with cisplatin for women with early stage disease. The first study included patients who had stage I cancers of grades 1 and 2, without ascites, positive cytology, rupture, or capsular penetration. When observation was compared to six cycles of cisplatin, the 4-year disease-free rates were 70% and 71% respectively. In the second study, patients with all other classes of stage I were randomized to intraperitoneal ₃₂P or six cycles of cisplatin. The 4-year disease-free survival rate with cisplatin was 79% compared to 69% with i.p. radiocolloid, but this difference did not achieve statistical significance. The GOG has subsequently completed an additional trial in which high-risk patients were randomized to three or six cycles of carboplatin and paclitaxel. The results of this trial are currently not available. The current GOG trial provides all patients at high risk of recurrence with 3 cycles of carboplatin and paclitaxel and participants are then randomized to observation or 6 months of weekly low-dose (40 mg/m₂) paclitaxel.

Outcomes of the randomized studies reported to date do not permit definitive treatment recommendations for stage I patients. Failure to prove a benefit for treatment may be the result of methodologic problems rather than ineffective treatment. None of the studies, for example, contained a sample size large enough to detect a 10% improvement in survival or freedom from relapse, despite durations of accrual that often exceeded 5 years. An improvement of 10% would, however, halve the number of treatment failures. In addition, several early studies included grade 1, stage I patients, whose relapse risk after surgery alone is about 5%. Most of these lesions are diploid, and tend to have an excellent prognosis. At present, it is possible to identify patients with borderline and well-differentiated lesions who do not require additional treatment. In the future, identification of these individuals might be made more precise by analysis of genomic and expression profiles, as well as by measurement of individual prognostic molecular markers. Patients with moderately and poorly differentiated lesions require further treatment, but optimal therapy remains to be defined.

Management of Early Stage Borderline Tumors
The principal treatment for borderline ovarian tumors is the surgical resection of the primary tumor. There is no evidence that either subsequent chemotherapy or radiation therapy improves survival. After performing a frozen section and determining that the histology is borderline, premenopausal patients who desire preservation of ovarian function may be managed with a conservative operation, such as a unilateral salpingo-oophorectomy. Thus, hormonal function and fertility can be maintained. In patients in whom an ovarian cystectomy has been performed and a borderline tumor is documented in the permanent pathology, no additional surgery is warranted.

There has been considerable controversy regarding the optimum treatment of patients with localized borderline ovarian tumors. This has been due, in part, to lack of unanimity regarding the histopathologic criteria for borderline tumors. For all stages of ovarian cancer, borderline tumors have a more favorable natural history than have invasive tumors. There have been conflicting reports regarding the efficacy and necessity of adjuvant chemotherapy for patients with localized stage I or stage II borderline tumors. In the large GOG trial described above, a total of 51 patients were reclassified as having borderline tumors. In these carefully staged patients, there have been no deaths directly attributable to cancer. While a substantial number of patients did receive adjuvant chemotherapy in these trials, there is no evidence that it was necessary or beneficial. If, after careful histologic review of multiple slides sectioned at 1-cm intervals, no evidence of stromal invasion is found, patients with localized borderline tumors should not routinely receive adjuvant therapy.

Management of Invasive Early Stage Low Risk Disease (Stage IA and IB, Low Grade)
In patients who have undergone a thorough staging laparotomy where there is no evidence of spread beyond the ovary, the performance of an abdominal hysterectomy and bilateral salpingo-oophorectomy is appropriate therapy. The uterus and contralateral ovary can be preserved in women with stage IA diploid lesions who wish to preserve fertility. These women should be followed carefully with periodic pelvic examinations and CA125 titers. Generally, the other ovary and uterus are removed at the completion of childbearing. In a recent report by Guthrie and colleagues, the outcome of 656 patients with early-stage epithelial ovarian cancer was studied. No patients who had a properly documented stage I, grade 1 cancer died of their disease; that is, there was a 100% survival in this condition when patients were surgically staged, and thus adjuvant radiation and chemotherapy appeared to be unnecessary.

Management of Invasive Early Stage High-Risk Disease (Stage IA and IB, High Grade, Stage IC and Stage II)
In patients whose disease is more poorly differentiated or in whom there are malignant cells either in ascitic fluid or peritoneal washings, additional therapy is indicated. Patients with grade 2 and grade 3 tumors, with densely adherent tumors, with large-volume ascites, and/or with positive peritoneal cytology, have a relapse risk of 20% to 45%, and postoperative treatment is warranted. Regrettably, it would appear that thorough staging with negative findings, including random peritoneal biopsies and lymph node sampling, does not eliminate the risk of relapse in patients with these characteristics. Although the optimal therapy for these patients is not known, treatment options include platinum based chemotherapy or abdominopelvic radiation therapy. Chemotherapy for patients with early stage high-risk epithelial ovarian cancer can be either single agent carboplatin or a combination of carboplatin and a taxane for 3 to 6 courses. Melphalan is not recommended due to its leukemogenic properties, its long-term compromise of marrow reserve, and its variable oral absorption, despite its ease of administration.

In the absence of definitive studies, the recommendation of adjuvant therapy is based on the assumption that chemotherapy or radiotherapy will cure a fraction of patients and that management of recurrent disease will not be as effective. In a recent report from the United Kingdom, 194 women with early stage disease were treated with surgery alone and received chemotherapy only at relapse. Among 44 assessable patients treated at recurrence with platinum based therapy, the response rate was 47%. Progression free and overall survival rates 5 years after salvage chemotherapy were 24% and 46% respectively.