Pathologists also classify adenocarcinomas according to the degree of histologic differentiation. Those tumors retaining clear-cut glandular features are considered grade 1, or well differentiated, whereas those that are largely composed of solid sheets of tumor are considered grade 3, or poorly differentiated. Tumors showing both glandular and solid areas are assigned to grade 2. The histologic grade seems to correlate roughly with biologic aggressiveness.
Stromal and germ-cell tumors Malignancies can also arise from the ovarian stroma or the primordial germ cells contained within the ovaries. Stromal tumors are often hormone-producing and include such types as the granulosa tumor, Sertoli-Leydig tumor, and several variants. Germ-cell tumors, which tend to be highly aggressive, include the dysgerminoma, endodermal sinus tumor, malignant teratoma, embryonal carcinoma, and rare primary choriocarcinoma of the ovaries. Malignant germ-cell tumors occur primarily in younger patients, with an average age at diagnosis of about 19 years.
Staging and prognosis
Staging system
The staging system for ovarian cancer shown in Table 1, developed by the International Federation of Gynecology and Obstetrics (FIGO), is used uniformly in all developed countries. It is based on the results of a properly performed exploratory laparotomy, a fact that bears emphasis, since inadequate surgical staging has been and continues to be a significant problem.
Surgical staging The surgical staging of ovarian cancer is based on an understanding of the patterns of disease spread and must be conducted in a systematic and thorough manner. It should include a complete evaluation of all visceral and parietal surfaces within the peritoneal cavity, omentectomy, and biopsy of aortic and pelvic lymph nodes. It generally includes removal of the internal reproductive organs as well, although exceptions to this rule can be made for younger women with limited disease who may wish to retain fertility.
TABLE 2Procedures for surgical staging for apparent early ovarian cancer
The issue of adequate surgical staging becomes particularly acute in just the patient population likely to be operated upon by individuals with no specialized training in gynecologic oncology: patients with adnexal masses that are not obvious cancers on preoperative evaluation. At the time of exploration, if the mass is shown to be malignant on frozen section and there is no obvious metastatic disease, a complete staging operation is essential to search for occult metastatic spread, which may be present in 20% to 30% of such cases. Also, if the tumor is documented to be stage IA by thorough staging and the patient wishes to preserve the potential for future fertility, it may be appropriate to conserve the uterus and uninvolved ovaries and fallopian tubes.
The elements of surgical staging for apparent early ovarian cancer are listed in Table 2.
Prognostic factors
The prognosis of epithelial ovarian cancer depends on a number of factors.
Disease stage Of primary importance is the disease stage, which, when properly determined, is of strong prognostic significance. The distribution of ovarian cancer cases by stage follows: stage I, 26%; stage II, 15%; stage III, 42%; stage IV, 17%. For patients with advanced ovarian cancer, the amount of residual tumor at the conclusion of the initial operation is of major importance. Patients with stage III disease who have minimal or no residual tumor may have a 30% to 50% chance of 5-year survival, whereas those patients with stage III disease left with bulky tumor masses have a 5-year survival rate of only about 10%.
Histologic grade and type Most studies have found the histologic grade of the tumor to have prognostic significance; the histologic cell type of the tumor is of less importance, although patients with clear-cell and possibly mucinous tumors may have a worse prognosis.
Molecular markers In recent years, a great deal of effort has been devoted to the identification of molecular markers of prognosis in ovarian cancer. Studies of HER2, p53, ras, and other oncogenes and tumor-suppressor genes have had varying results relative to prognostic significance. Currently, the assessment of molecular markers is ongoing in numerous studies in the hope of identifying clinically relevant targets that are susceptible to available agents. The continued progress in developing high-throughput techniques for determining gene and protein expression increases the likelihood that good candidates will be found.
Predictors of chemosensitivity Despite a continued effort to assess in vitro methods to predict the sensitivity or resistance of ovarian cancers to various chemotherapeutic drugs, the clinical usefulness of such an approach remains under investigation. The American Society of Clinical Oncology (ASCO) recently reviewed the relevant literature on the subject and reached the same conclusion for cancers in general.
Treatment
Surgery plays a crucial role in all phases of the management of ovarian cancer and, when applied as part of a multidisciplinary approach, affords patients the highest likelihood of a favorable outcome. For most patients with ovarian carcinoma, surgery is not curative due to dissemination of tumor cells throughout the abdominal cavity. Therefore, successful management generally requires additional treatment.
The use of postoperative chemotherapy is standard for all patients with advanced-stage disease and for many patients with early-stage disease. Adjunctive chemotherapy significantly prolongs survival, with most current data supporting the use of platinum- and taxane-based regimens.
Despite a long history of the use of radiation therapy in ovarian carcinoma, due to the relatively high sensitivity of ovarian cancer to radiation in general, opinions on indications for its use differ widely. Presumably, this controversy is due to the limited amount and adequacy of data comparing radiotherapy with modern chemotherapy regimens, as well as concerns regarding potential toxicities. Studies using radiation in both radical and palliative settings have been published, but standardization of the use of radiation varies significantly worldwide.
management of early-stage disease
Clearly, comprehensive surgical staging is necessary to properly identify patients with stages I and II ovarian carcinoma. Beyond surgery, the need for adjuvant treatment with chemotherapy has been recently supported, with the exception of patients with stage I disease and well-differentiated histology.
Surgery
Suspicious adnexal masses should be excised intact and submitted for frozen section. If a malignancy is confirmed and there is no obvious metastatic spread, complete surgical staging should be undertaken. As discussed previously (see section on "Staging and prognosis"), it is of critical importance that surgical staging be performed systematically and completely. Inadequate staging may result in inappropriate postoperative treatment, which can severely compromise the chances for cure.
Data from the American College of Surgeons community hospital-based tumor registry show that almost 75% of the primary surgeries for ovarian cancer performed in this country are done so without the involvement of a gynecologic oncologist. This finding is unfortunate given the fact that, with physical examination, measurement of CA-125 levels, and appropriate imaging tests, the majority of cases of ovarian cancer can be identified preoperatively. Results from other studies suggest that when a gynecologic oncologist is not present at the initial operation, staging is more often inadequate, cytoreduction is more often suboptimal, and long-term survival is poorer.
Conservation of reproductive organs In a woman of reproductive age with cancer limited to one ovary, it may be possible to conserve the uterus and opposite fallopian tube and ovary if she wishes to maintain the option of future fertility. To facilitate such intraoperative decision-making, it is essential that the surgeon's preoperative discussion with the patient and her family address the possibility of malignancy and review the surgical options for both benign and malignant diseases.
Operative laparoscopy Recent advances in the instrumentation for operative laparoscopy have led to an increase in the proportion of adnexal masses being managed with this technique. Physicians should exercise caution in selecting patients with adnexal masses for operative laparoscopic approaches. Unless the surgeon's laparoscopic skills are extraordinary, suspicious masses are best managed by laparotomy. For masses that are approached laparoscopically, the same surgical principles of removal without spill and complete surgical staging apply.
Systemic chemotherapy for early-stage disease
The current management of patients with early-stage disease focuses on comprehensive surgical staging and the identification of high-risk features. Patients with stage IA or IB tumors with well-differentiated histology have excellent 5-year survival rates, and adjuvant chemotherapy is generally not used in such patients. High-risk features include moderately to poorly differentiated tumors, stage IC or II disease, and clear-cell histology.
The reported survival rates of 60% to 80% in patients who have early-stage tumors with high-risk features suggested a potential role for adjuvant therapy. The IICG conducted two randomized trials to evaluate the role of adjuvant therapy in patients with stage I disease. The first trial compared cisplatin, 50 mg/ m² q28d × 6, with observation in 85 patients with stage IA or IB, grade 2–3 disease. The 5-year disease-free survival rate was higher in patients treated with cisplatin than in those who were observed (83% vs 63%), but the 5-year overall survival rate was similar in the two groups (88% vs 82%).
The second trial compared cisplatin (same dose) to phosphorus-32 (P-32) administration in 161 patients with stages IA–IB, grade 2, or stage IC disease. The 5-year disease-free survival rate again favored the platinum arm (85% vs 65%), but the 5-year overall survival rate was unchanged and similar to that reported in the previous trial. P-32 administration was associated with more long-term toxicity.
More recent data have provided support for a survival benefit to the immediate use of adjuvant chemotherapy in patients with early-stage disease. The results of the EORTC–ACTION trial and the ICON1 trial were combined and reported. A 5-year survival rate improvement of 8% was reported for those receiving immediate chemotherapy compared with reserving chemotherapy for those who relapsed (74% vs 82%; 95% CI: 2%–12%).
Improvements in systemic chemotherapy for advanced ovarian cancer with associated improvements in survival are relevant to the design of regimens for early-stage disease. GOG 157 evaluated 3 vs 6 cycles of paclitaxel and carboplatin in patients with stage IA or IB, grade 2–3; stage IC; or stage II disease. The trial completed accrual in 1995, and final results showed no significant benefit to the longer regimen. The GOG replacement trial evaluated 3 cycles of paclitaxel plus carboplatin with or without additional weekly paclitaxel (40 mg/ m²) in patients with early-stage disease. These data are reported in abstract form, but no benefit to extended-schedule paclitaxel was seen.
In the absence of additional data, taxane- and platinum-based systemic chemotherapy should be considered the standard approach for patients who have early-stage disease with the exception of well-staged IA or IB, grade 1 disease. The optimal number of cycles is currently unclear, but 3 cycles were considered the standard arm in the GOG 157 trial, although many clinicians offer 6 cycles of therapy in the absence of prohibitive toxicity.
Past GOG trials have established that patients with stages IA–IB, well-differentiated or moderately differentiated tumors have a 5-year survival rate of 90% to 98%, which does not seem to improve with adjuvant chemotherapy. However, patients with less favorable neoplasms by virtue of higher grade or stage have poorer outcomes (80% 5-year survival rate among treated patients).
Radiation therapy
Whole-abdominal irradiation A study by Hepp et al found whole-abdominal irradiation (WAI) to be an effective adjuvant therapy in patients with optimally debulked tumors. In a series of 60 patients, the 5-year survival rate was 55%, with a median follow-up of 96.5 months. Patients who received chemotherapy (n = 41) fared slightly worse than those who received radiation therapy only. The abdominal control rate was 83%, and the grade 3 and 4 late toxicity rates were 7% and 3%, respectively.
The findings indicate that 5- and 10-year survival rates obtained with WAI are at least equivalent to results obtained using modern systemic agents. However, in view of the recognized limitations of these trials, more rigorously gathered data will be required to establish the role of WAI in these patients.
management of Advanced disease
Surgery
In the majority of cases, surgeons operating on patients with ovarian cancer find obvious evidence of widespread metastatic disease. Ascites is often present, with diffuse peritoneal tumor studding and extensive omental involvement. In such cases, it is still important to document the surgical stage (usually a substage of stage III) and carefully evaluate and describe the extent and location of tumor identified at both the beginning and conclusion of surgery.
Optimal cytoreduction The primary function of surgery in patients with advanced ovarian cancer is cytoreduction or debulking. When surgery is performed by experienced gynecologic cancer surgeons, at least 50% of patients with stage III ovarian cancer can be left with "optimal" residual tumor (ie, ≤ 1 cm). The morbidity associated with such surgery is low, and operative mortality is rare.
Patients with optimally debulked disease have an increased likelihood of achieving a complete clinical response to chemotherapy. Disease progression-free interval, median survival, and long-term survival are all improved in patients who have optimal cytoreduction.
Even among patients with suboptimal residual disease (> 1 cm) after primary surgery, those left with smaller tumor volumes (1 to 2 cm) have a survival advantage over those with a larger residuum. It is thus clear that aggressive surgical cytoreduction, if successful in reducing tumor to small volumes, improves several measures of outcome.
Interval cytoreduction In an EORTC trial, 299 patients with suboptimal advanced ovarian cancer were randomized to receive 6 cycles of cisplatin plus cyclophosphamide with or without interval surgical cytoreduction after the third cycle. Median survival for patients who underwent interval debulking surgery was 27 months, vs 19 months for patients who did not have interval debulking (P = .01). The GOG then completed a randomized trial of interval cytoreduction using a cisplatin-paclitaxel chemotherapy regimen. These results show no benefit for interval cytoreduction (median overall survival, 32 vs 33 months). Taxane-based chemotherapy and more standardized aggressive initial debulking by experienced gynecologic oncologists in the GOG trial have been offered as possible explanations for the discordant outcomes. If an aggressive initial surgical attempt is provided by a gynecologic oncologist, interval surgical cytoreduction cannot be routinely recommended.
Chemotherapy
Two GOG studies have indicated that the major survival benefit of cytoreductive surgery was in those patients who were optimally cytoreduced to no macroscopic residual disease. In the first series, 1,895 patients with stage III disease received cisplatin and paclitaxel in standard IV regimens. Optimal cytoreduction to no macroscopic disease was achieved in 23%. This resulted in significantly better (P < .001) progression-free and overall survival rates (PFS = 33.0 months; OS = 71.9 months) vs patients with residual disease of 0.1 to 1 cm residual disease (PFS = 16.8 months; OS = 42.4 months) or residual disease greater than 1 cm (PFS = 14.1 months; OS = 35.0 months). In a subsequent GOG report on 360 patients with stage IV disease who received cisplatin and paclitaxel in standard IV regimens, 8% were optimally cytoreduced to no macroscopic residual tumor, yielding a PFS of 20.1 months and an OS of 64.1 months. For those patients with residual disease less than or equal to 1 cm, or between 1.1 and 5 cm, PFS and OS rates were similar (PFS = 13.0 months in both groups; OS = 28.7 and 31 months, respectively; Winter WE III et al: J Clin Oncol 25:3621–3627, 2007; Winter WE III et al: J Clin Oncol 26:83–89, 2008).
Primary treatment The results of two randomized trials support a survival advantage for patients treated with combinations of IV platinum and paclitaxel, as compared with those given a platinum plus cyclophosphamide. McGuire et al found a 37- vs 24-month median survival advantage for the platinum-paclitaxel arm. Similarly, an analysis of the intergroup trial by Piccart et al showed an improvement in median survival from 25 to 35 months (P = .001) in favor of the paclitaxel arm. In contrast, the initial analysis of the ICON 3 trial evaluating a control arm (carboplatin or CAP [cyclophosphamide, Adriamycin (doxorubicin), Platinol (carboplatin)] chemotherapy) vs paclitaxel and carboplatin has failed to show a survival advantage for the taxane-containing arm. Many factors in the study have been proposed to explain this difference, and for the present, taxane- and platinum-based therapy remains the standard.
A randomized trial (GOG 158) comparing paclitaxel (175 mg/ m² via a 3-hour infusion) plus carboplatin (dosed to achieve an area under the concentration-time curve [AUC] of 7.5) vs the standard regimen of paclitaxel (135 mg/ m² via a 24-hour infusion) plus cisplatin (75 mg/ m²) in patients with optimally debulked disease showed the shorter schedule with carboplatin to be as effective as the older regimen. Due to its decreased toxicity and ease of administration, the shorter schedule with carboplatin is the preferred treatment.
In addition, the SCOTROC trial suggested that, as primary treatment, docetaxel (Taxotere) and paclitaxel have similar efficacy when combined with carboplatin and that docetaxel produces less neuropathy.
A five-arm international randomized study of primary therapy for patients with stage III or IV disease evaluated carboplatin and paclitaxel as the control arm and studied two triplets (carboplatin + paclitaxel with either gemcitabine [Gemzar] or liposomal doxorubicin [Doxil]) and two sequential doublets (topotecan [Hycamtin]/carboplatin + carboplatin/paclitaxel or carboplatin/gemcitabine + carboplatin/paclitaxel). No difference in progression-free or overall survival rates was seen among the arms, and therefore paclitaxel and carboplatin remains the standard. Based on the variety of phase III trials employing IV paclitaxel and carboplatin therapy following maximal surgical cytoreduction, the expected progression-free and overall survival rates of stage III patients follow: stage III optimal (progression-free survival, 21 to 28 months; overall survival, 52 to 57 months) and stage III suboptimal (progression-free survival, 18 months; overall survival, 38 months).
Neoadjuvant chemotherapy In patients with an inadequate performance status to undergo aggressive primary debulking, a "neoadjuvant" approach with paclitaxel and carboplatin is often considered for several cycles prior to a maximal cytoreductive effort. A large randomized trial reported by Vergote et al evaluated the use of neoadjuvant chemotherapy in 718 patients with stages IIIC–IV ovarian cancer. Patients were randomized to undergo primary debulking followed by 6 courses of paclitaxel and carboplatin chemotherapy (arm A) vs 3 courses of neoadjuvant chemotherapy, interval debulking, and then 3 additional courses (arm B). In the reported data in abstract form, median overall survival was 29 and 30 months for arms A and B, respectively (HR, 0.098; CI: 0.85–1.14); the median progression-free survival was 11 months in both arms (HR, 0.99; CI: 0.87–1.13). The overall survival in both arms was lower than expected, but there was no difference between them. The approach is still controversial, and most gynecologic oncologists still consider primary surgical debulking the standard of care, reserving neoadjuvant therapy only for those patients in whom primary optimal debulking will not be achieved. Further developments in this area will be forthcoming.
The role of bevacizumab in primary treatment Bevacizumab (Avastin) has shown activity in patients with recurrent ovarian cancer. Response rates of 16% and 21% have been reported when it is used as a single agent and of 24% when it is used in conjunction with oral cyclophosphamide. The progression-free intervals across trials range from 4.4 months to 7.2 months, and the median overall survival ranges from 10.7 months to 17 months. An important and large first-line chemotherapy trial was reported at ASCO 2010. This trial randomized patients with stages III and IV ovarian cancer following surgical debulking to receive paclitaxel and carboplatin + placebo followed by placebo maintenance (total therapy, 15 months), paclitaxel and carboplatin + bevacizumab followed by placebo maintenance, or paclitaxel and carboplatin + bevacizumab with bevacizumab maintenance. This trial evaluated toxicity, progression-free, and overall survival. A total of 1,873 patients were enrolled in the study. The baseline clinical characteristics were well balanced. Adverse events were typical of those seen in other bevacizumab-containing studies. Hypertension was seen in 16%–22% of the bevacizumab-containing arms, and bowel perforations occurred in < 3%. The median PFS was 10.3 months for patients treated with chemotherapy alone vs 14.1 months for those receiving extended-schedule bevacizumab (HR, 0.717; CI: 0.625–0.824, P < .0001). No difference in overall survival was seen, although it is early for this assessment.
As a result of these preliminary data, there are many questions to consider. They will be answered in time with additional studies or clarified once details are available. Will there be an overall survival benefit? If not, is a strategy that prolongs only progression-free survival sufficient? How long should bevacizumab be given? Should it be continued longer than 15 months, until disease progression, or for life? What happens when bevacizumab is discontinued? Is the phenotype of relapsed disease on bevacizumab more aggressive? Is there rebound at its discontinuation? Finally, is the "cure" proportion improved with prolonged bevacizumab use?
Intraperitoneal chemotherapy
The randomized study by Armstrong et al employed intraperitoneal (IP) therapy as part of primary treatment. They showed a median overall survival of 65.6 months for the IP arm vs 49 months in the IV group. This finding represents the largest difference to date between two treatment arms in any study evaluating primary therapy. The study is the third in a series of studies supporting the IP administration of primary chemotherapy to optimally debulked stage III patients.
The first study by Alberts et al predated paclitaxel and carboplatin use and simply asked the question of whether the IV or IP administration of cisplatin was better, showing an advantage for the latter (median overall survival, 49 vs 41 months; P = .02) This first IP study has been criticized because it does not contain paclitaxel and thus does not reflect contemporary treatment.
Early treatment of relapse immediately after testing positive for elevated CA-125 levels in women with ovarian cancer in clinical complete remission does not improve survival over delaying treatment until the occurrence of clinical symptoms of relapse such as pelvic pain or bloating. This was according to data presented at ASCO 2009. In this study, investigators compared overall survival between 265 women with ovarian cancer in remission after initial chemotherapy who began second-line chemotherapy after experiencing a rise in CA-125 level, and 264 women with rising CA-125 level whose treatment was delayed until symptoms of relapse appeared. Second-line chemotherapy was started in the early treatment group a median 5 months before the delayed treatment group. At the time the study was presented, overall survival was the same between both groups (HR, 1.01, 95% CI: 0.82–1.25: P = .91). Researchers concluded that there was no survival benefit from early treatment based on a raised serum marker level alone. They added that consequently, there was no value in the routine measurement of CA-125 levels in the follow-up of ovarian cancer patients (Rustin GJ et al: J Clin Oncol 27[18S]: abstract P1, 2009).
The second study by Markman et al included paclitaxel, but the experimental arm not only included IP delivery of cisplatin but also added high-dose IV carboplatin in an attempt to "chemically debulk" the tumor prior to IP administration. The Markman study, while also showing an advantage for the IP-containing experimental arm (overall survival, 52 vs 63 months; P = .05), was criticized because more than one variable was changed and the benefit could not be directly attributed to IP therapy.
As previously discussed, the third and well-designed trial by Armstrong et al showed a median overall survival of 65.6 months for the IP arm vs 49 months for the IV group. This study used IV paclitaxel (135 mg/ m²) over 24 hours on day 1, IP cisplatin (100 mg/ m²) on day 2, and IP paclitaxel (60 mg/ m²) on day 8 for 6 total cycles. Due to increased toxicity in the IP arm (metabolic, neuropathy), only 42% of patients completed all 6 cycles. However, a quality-of-life analysis at 12 months showed no difference between the IP and IV groups, suggesting the toxicity was reversible.
