What is the benefit of bevacizumab combined with chemotherapy in patients with recurrent ovarian, fallopian tube or primary peritoneal malignancies?

The aim of this retrospective analysis was to investigate the efficacy and adverse effects of the monoclonal antivascular endothelial growth factor antibody bevacizumab (Avastin(R)) combined with chemotherapeutic agents in non-protocol patients with recurrent ovarian, fallopian tube, or primary peritoneal malignancies. Using our databases, we identified patients treated with bevacizumab combination therapy since June 2005. Responses were evaluated with Response evaluation Criteria in Solid tumors and serum CA125 Rustin criteria. Toxicity was assessed according to the Common toxicity Criteria (CTC) v.3.0. Data from 64 patients were included. The median patient age was 58 years, and they had undergone a median of 4.5 (range, 1-10) prior cytotoxic chemotherapy regimens. The median length of follow-up was 8 months (range, 2-29). The most commonly used combinations were bevacizumab plus taxanes (26.6%) and plus cyclophosphamide (26.6%). A median of 4 cycles of therapy with a median bevacizumab dose of 3,600 mg (range, 500-18,240) were administered. An overall response rate of 21.3% was observed in 13 patients with partial response, and another 42.6% of patients had stable disease. Among the patients with elevated pretreatment serum CA125 concentration, an overall response rate of 46.3% (25/54) was observed according to modification of the Rustin criteria. Fifteen (23.4%) patients had grades 3 or 4 adverse events. Gastrointestinal perforations occurred in 2 (3.1%) patients. Seventeen (26.6%) patients had improved performance status scores. Bevacizumab combined with chemotherapy showed promising clinical benefits, with significant response of serum CA125 concentration and moderate adverse effects.

Modified uterine manipulator and vaginal rings for total laparoscopic radical hysterectomy.

At present, there is no standard technique that allows surgeons performing total laparoscopic radical hysterectomy to complete the colpotomy and remove an adequate (2-cm) margin of upper vaginal tissue while maintaining adequate pneumoperitoneum. We evaluated the feasibility and safety of using a modified uterine manipulator for total laparoscopic radical hysterectomy in patients with cervical or endometrial cancer. A retrospective review was performed in all patients who underwent total laparoscopic radical hysterectomy using a modified uterine manipulator at our institution during the period April 2004 to December 2006. This analysis included 30 patients who underwent surgery with the modified uterine manipulator. There were no reports of difficulty with placement of the instrument, multiple attempts at placement, difficulty with uterine manipulation, or uterine perforation. In no patient was a vaginal incision or episiotomy required to fit the instrument through the introitus. In no case was there loss of pneumoperitoneum during colpotomy. Additional upper vaginal tissue had to be removed after intraoperative assessment of the adequacy of the surgical specimen in five (16.7%) of 30 patients. Use of the modified uterine manipulator according to our technique is safe and feasible, allowing for adequate vaginal resection and maintenance of pneumoperitoneum.

Time course and outcome of central recurrence after radiation therapy for carcinoma of the cervix.

We investigated the time course of central disease recurrence (CDR) in 2997 patients treated with radiation for stage I-II squamous cell carcinoma of the cervix. CDR rates were 6.8%, 7.8%, and 9.6%, at 5, 10, and 20 years, respectively. The risk of CDR was independently correlated with tumor size (P < 0.0001) but not with FIGO stage. The hazard rate peaked in the first year of follow-up and then fell steeply; after 3 years, the hazard rate was approximately constant at 0.2-0.4% per year. Although after 3 years the risk of CDR was low, it continued to be slightly greater for patients with tumors > or =5 cm than for those with smaller tumors (P= 0.001). Patients who had CDR < 36 months after treatment were less likely to be candidates for salvage therapy and had a poorer post-recurrence survival rate than those with recurrence > or =36 months after treatment (4.5% versus 42.1%, P < 0.0001). The higher rate of CDR in the first 3 years and the poor survival after early recurrence suggest that most early CDRs are true relapses. The relatively stable annual actuarial risk between 3 and 25 years and the better survival rate after late CDR suggest that most "recurrences" after 3 years are actually new neoplasms.

Phase II study of tirapazamine plus cisplatin in patients with advanced or recurrent cervical cancer.

The aim of this study was to evaluate the activity and toxicity of a tirapazamine (TPZ)/cisplatin drug combination in patients with stage IV or recurrent cervical cancer. The chemotherapy was administered for a maximum of eight cycles every 21 days. TPZ was administered intravenously at 330 mg/m(2) over a 2-h infusion, followed 1 h later by cisplatin intravenously at 75 mg/m(2) over 1 h on day 1. All patients received antiemetics including dexamethasone, ondansetron, and lorazepam. Subsequent doses were unchanged, reduced, or omitted according to observed toxicity and protocol guidelines. Response evaluation was performed every two cycles. Thirty-six patients with stage IV or recurrent cervical cancer were treated. Ninety-four percent of patients had prior radiotherapy. Two patients had prior chemotherapy. There were two complete responses and eight partial responses (27.8%). An additional 11 patients (30.6%) had stable disease as their best response. Response rate was greater in tumors outside of the previously radiated field (44.4% vs 11.1%). The median time to progression was 32.7 weeks. The most frequent grade 3 or 4 adverse events were nausea, vomiting, and fatigue, which occurred in 30.6%, 25%, and 22% of subjects, respectively. Anemia was the most frequent grade 3 or 4 hematologic toxicity at 8.3%. We conclude that the combination of cisplatin and TPZ was reasonably well tolerated in patients with recurrent or advanced cervical cancer. Further evaluation of this drug combination may be warranted.

Laparoscopic port-site metastases in patients with gynecological malignancies.

OBJECTIVE: The purpose of this study is to review all reported cases of laparoscopic port-site metastases in patients with gynecological malignancies. Potential etiologies as well as options for prevention are discussed. METHODS: We searched the Medline database for English-language articles presenting raw data on laparoscopic port-site metastases in patients with gynecological malignancies. RESULTS: We found 31 articles describing port-site metastases in 58 patients. Forty patients had low malignant potential (seven patients) or invasive ovarian carcinoma (33 patients). The median age of these patients was 50 years (range: 22-79), and 83% had advanced (stage III or IV) disease. Seventy-one percent of the patients (24 of 34) had ascites, and 97% (29 of 30) had carcinomatosis. Seventy-five percent of the laparoscopic procedures in this group were performed for diagnosis. Median time to diagnosis of port-site metastases was 17 days (range: 4-730). Seventy-one percent of port-site recurrences (15 of 21) were isolated to a tissue-manipulating port. Twelve patients had port-site metastases after laparoscopy for cervical cancer. The median age was 44 years (range: 31-74). Eighty percent of cases were squamous cell carcinoma. In 75% of the patients, laparoscopy was performed for therapeutic purposes. The median time to diagnosis of port-site metastases was 5 months (range: 1.5-19). Four patients had port-site metastases after laparoscopy for uterine cancer. The median age was 63 years (range: 56-72). The median time to diagnosis of metastases was 13.5 months (range: 6-21). Half of the recurrences were in the tissue-manipulating port. Port-site metastases after laparoscopy were reported for one patient each with a diagnosis of fallopian tube carcinoma and vaginal carcinoma. CONCLUSIONS: Laparoscopic port-site metastases are a potential complication of laparoscopy in patients with gynecological malignancies, even in patients with early-stage disease.

Third-line chemotherapy in platinum- and paclitaxel-resistant ovarian, fallopian tube, and primary peritoneal carcinoma patients.

Ovarian carcinoma is a malignant disease with a high rate of recurrence, necessitating repeated chemotherapy treatments. We conducted a retrospective study in patients with platinum- and paclitaxel-resistant ovarian, fallopian tubes and primary peritoneal carcinoma patients treated at M.D. Anderson Cancer Center. We evaluated the responses, progression-free intervals, and overall survival duration of 51 patients after third-line chemotherapy treatment. The overall response rate was 16% (eight cases) with 2% complete response rate (one case) and 14% partial response rate (seven cases). Stable disease was achieved in 31% (16 cases). The progression-free intervals of 24 patients who had response and stable disease was 7.4 months (range, 1.4-18.4 months). The median overall survival of all patients was 15.8 months (95% CI, 8.1-23.4 months). The median survival duration of eight responders was not significantly different from that of 43 nonresponders, 18.9 months (95% CI, 2.4-35.4 months) versus 15.8 months (95% CI, 6.4-25.2 months), respectively (P = 0.73). In conclusion, third-line chemotherapy in our study results in a modest response and prolongation of progression-free interval without obvious impact on survival. The decision to utilize third-line chemotherapy will be a balance of the limited efficacy, toxicity of the agents, and the expertise of the clinician.

Approaches for end-of-life care in the field of gynecologic oncology: an exploratory study.

We sought to explore the Society of Gynecologic Oncologists (SGO) members' opinions and decisions about end-of-life issues and incurable conditions. A survey was mailed to members of the SGO. Their responses were recorded on a Likert scale and entered into a database. The survey explored opinions, experiences, and decisions in managing terminally ill gynecologic oncology patients. Of 900 surveys, 327 were returned (response rate, 36%). Seventy-three percent were men, 89% were white, and 72% were of Christian denomination. Respondents believed that 97% of patients who are dying realize that they are dying but stated only 40% of these patients initiate conversations about end-of-life issues. In contrast, 92% of respondents stated that they initiate end-of-life discussions with patients. Ninety-two percent of respondents thought that the patients should be allowed to make end-of-life choices independently after the facts are given to them. However, 44% thought that it is important to influence the way information is presented, and 54% believe that the gynecologic oncologist (GO) controls the outcome of end-of-life discussions. Although the physicians' sex, race, religion, and age did not correlate with their treatment decisions, religion did correlate with less fear of death (P = 0.011) and less discomfort when talking with patients about death (P = 0.005). Fifty-four percent of respondents believed that the GO controls the outcome of end-of-life discussions, and 40% believe that their actions prolong the process of dying. Expanding our understanding of what motivates GOs to recommend continued treatment over palliation is important for preserving informed patient-motivated end-of-life decisions.

Role of surgical resection for lung, liver, and central nervous system metastases in patients with gynecological cancer: a literature review.

Many reports of ovarian, cervical, and uterine cancers metastatic to lung, liver, and brain have been published. A fewer number of them focused on the surgical treatment for these patients. We reviewed the published literature, regarding surgical management of metastatic disease in patients with gynecological cancer. Some prognostic factors in the patients with metastatic lesions from these three different cancers were found in common. Favorable prognostic factors for a prolonged survival were good performance status of the patients, long disease-free interval, absence of other systemic disease, and the resectability, preferably with a clear margin. These factors should be considered as the criteria for surgery. In well-selected patients, survival could be extended from the surgical procedure with minimal complications. Other types of treatment such as radiation therapy or chemotherapy could also be given in conjunction with surgery, depending on tumor type and disease status of the primary cancer, other systemic diseases, and residual metastatic lesions after surgery.

Analysis of 96 patients with advanced ovarian carcinoma treated with high-dose chemotherapy and autologous stem cell transplantation.

The purpose of this study was to identify characteristics significant to survival and progression-free survival in patients with advanced ovarian cancer receiving high-dose chemotherapy. In all, 96 patients received autologous stem cell transplantation. Regimens included paclitaxel with carboplatin (PC), topotecan, melphalan, cyclophosphamide (TMC) and cyclophosphamide, BCNU, thiotepa (CBT). At the time of transplantation, 43% of patients were in clinical CR, 34% were in clinical PR, 18% had progressive disease and 5% had stable disease. There were no treatment-related deaths. The 6-year survival by Kaplan-Meier was 38%. For patients who received transplantation for remission consolidation, the 6-year survival was 53% with a PFS of 29%. On univariate analysis, the CBT regimen, clear cell histology and disease status other than CR prior to treatment were statistically significant adverse prognostic factors. This analysis has demonstrated that patients in clinical remission are most likely to benefit from autologous transplantation, with the exception of patients with clear cell histology. The TMC combination appeared to be superior to the PC and CBT combinations. Comparative studies of different consolidation approaches will be necessary to determine if autologous transplantation is the preferred treatment for this patient population.

Mixed choriocarcinoma in a postmenopausal patient.

Gestational trophoblastic disease rarely presents in patients beyond the reproductive years. To our knowledge, this is the first case of mixed trophoblastic disease in a postmenopausal woman. We present here a case of a 60-year-old woman with evidence of a pelvic mass and pulmonary metastasis. Surgery revealed an 8 x 6 x 6 cm multinodular uterine tumor involving the right adnexa. Histologic review was consistent with choriocarcinoma with intermediate trophoblastic features. Postoperative beta-hCG was 381 561 mIU/ml. We conclude that maintaining a high index of suspicion facilitates the identification of postmenopausal patients with metastatic gestational trophoblastic disease. This case reconfirms the deceptive presentation of the "great masquerader".

Phase II study of carboplatin and liposomal doxorubicin in patients with recurrent squamous cell carcinoma of the cervix.

BACKGROUND: The activity of the combination of carboplatin and liposomal doxorubicin was tested in a Phase II study of patients with recurrent cervical carcinoma. METHODS: The combination of carboplatin (area under the concentration curve [AUC], 5) and liposomal doxorubicin (Doxil; starting dose, 40 mg/m(2)) was administered intravenously every 28 days to 37 patients with recurrent squamous cell cervical carcinoma to determine antitumor activity and toxicity profile. RESULTS: Twenty-nine patients were assessable for response, and 35 patients were assessable for toxicity. The overall response rate was 38%, the median time to response was 10 weeks, the median duration of response was 26 weeks, and the median survival was 37 weeks. The main toxic effect was myelosuppression, with Grade 3 and 4 neutropenia in 16 patients, anemia in 12 patients, thrombocytopenia in 11 patients, and neutropenic fever in 3 patients. Four patients had five infusion-related reactions during the infusion of liposomal doxorubicin, leading to treatment discontinuation in three patients. Grade > or = 2 nonhematologic toxicity included nausea in 17 patients, emesis in 14 patients, fatigue in 9 patients, mucositis and/or stomatitis in 8 patients, constipation in 6 patients, weight loss in 5 patients, hand-foot syndrome in 2 patients, and skin reactions in 3 patients. CONCLUSIONS: The combination of carboplatin and liposomal doxorubicin has modest activity in patients with recurrent cervical carcinoma.

Intraoperative lymphatic mapping and sentinel node identification with blue dye in patients with vulvar cancer.

OBJECTIVE: To determine the effectiveness of intraoperative lymphatic with blue dye alone as a means of localizing sentinel nodes in patients with vulvar cancer. METHODS: All patients undergoing primary surgical treatment for vulvar cancer were eligible for this prospective study. Isosulfan blue dye was injected intradermally at the edge of the primary tumor closest to the adjacent groin. Bilateral dye injections and groin dissections were performed if the tumor was within 2 cm of the midline. RESULTS: Fifty-two patients were enrolled in the study between 1993 and 1999. The median age was 58 years. Eighty-seven percent of the patients had T1 or T2 lesions, and 92% had nonsuspicious lymph nodes on palpation. Sixty-seven percent of the patients had squamous cell carcinoma; the remaining patients had melanoma or adenocarcinoma. The sentinel node was identified in 46 of the 52 patients (88%), comprising 22 of the 25 patients with lateral tumors and 24 of the 27 patients with midline lesions. The sentinel node was successfully identified in 57 of the 76 (75%) dissected groins. Sentinel node identification in the groin was hampered by the effects of prior excisional biopsy vs punch biopsy (11 of 25 vs 8 of 51, P = 0.007) and by the lateral vs midline location of the tumor (22 of 25 groins vs 35 of 51 groins, P = 0.067). During the first 2 years (1993-1994), a sentinel node could not be identified in 4 of the 25 (16%) patients and 13 of the 36 (36%) groins dissected, compared with 2 of the 27 (7%) of patients treated and 6 of the 40 (15%) groins dissected from 1995 through 1999 (P = 0.034). A total of 556 nodes were removed (median, 7 per groin), of which 83 (median, 1 per groin) were sentinel. The sentinel node was not identified in 2 of the 12 groins that proved to have metastatic disease. Both events occurred in the first 2 years of the study. There were no false-negative sentinel nodes. Since 1995, we have successfully identified the sentinel node in 16 of the 16 patients (25 of 25 groins) with T1 or T2 primary lesions, squamous histology, and nonsuspicious groin nodes on physical examination. CONCLUSIONS: Experience and careful patient selection can permit sentinel node identification with blue dye injection alone in more than 95% of patients with vulvar cancer.

Sentinel nodes in gynecologic malignancies.

The concept of sentinel node identification and lymphatic mapping is already established as part of standard practice in the surgical management of breast cancer and melanoma. To reduce extensive radical procedures and decrease morbidity in gynecologic malignancies, much effort is being focused on implementing less aggressive interventions. By combining the use of radioactive tracers and blue dyes, investigators are identifying sentinel nodes. In vulvar and cervical carcinomas, sentinel node identification may significantly reduce the number of patients undergoing unnecessary, extensive lymphadenectomy in the absence of disease. The addition of novel techniques, such as histopathologic ultrastaging, immunohistochemistry staining, and reverse transcriptase polymerase chain reaction assays, will help increase the accuracy and rate of detection of disease.

High-dose topotecan, melphalan, and cyclophosphamide (TMC) with stem cell support: a new regimen for the treatment of advanced ovarian cancer.

OBJECTIVE: The goal of this study was to determine the optimal dose of topotecan when used in combination with high-dose melphalan and cyclophosphamide (TMC), and to assess the toxicity and efficacy of the regimen in patients with advanced ovarian cancer. METHODS: Fifty-three patients with persistent or recurrent ovarian cancer were treated. Disease status at study entry included: platinum-sensitive recurrent disease (15 patients), platinum-resistant or refractory recurrent disease (15 patients), positive second-look surgery (16 patients), failure to achieve a primary clinical complete response (CR) (7 patients). Following stem cell mobilization and collection, patients were given cyclophosphamide 1 g/m(2)/day on Days -6, -5, -4; melphalan 70 mg/m(2)/day on Days -3, -2; and topotecan at escalating doses from 1.25 to 4.0 mg/m(2)/day on Days -6 to -2. Peripheral blood stem cells were infused on Day 0. RESULTS: The optimal topotecan dose selected for future trials was 4.0 mg/m(2)/day x 5 days. The regimen had acceptable toxicity with no regimen-related death. Toxicity (Bearman toxicity criteria) was limited mostly to grade 1-2 mucositis and diarrhea. The overall response rate of patients with measurable or evaluable disease was 93%. Median survival has not yet been reached, but with a median follow up of 18 months (range: 11-37) 77% of patients are alive. CONCLUSION: With a topotecan dose of 4.0 mg/m(2)/day x 5 days, the TMC regimen has acceptable toxicity and produces high response rates. In the setting of ovarian cancer, high-dose chemotherapy should be administered only as part of well-designed clinical trials. TMC should be considered a potential regimen for future randomized trials in patients with advanced ovarian cancer.

Sigmoid perforation following radiation therapy in patients with cervical cancer.

OBJECTIVE: We describe the clinical presentation, evaluation, management, and outcome of patients experiencing sigmoid perforation following radiation therapy for cervical cancer. METHODS: A database consisting of over 5000 patients with stage IB-IIIB cervix cancer treated between 1963 and 1992 revealed 35 patients with sigmoid perforation. Twenty-seven were diagnosed and managed at one institution, and they form the study group. RESULTS: The median age at the time of perforation was 50 years, and the median follow-up care was 78 months (range 6-396). The median time from completion of radiotherapy to perforation was 13 months (range 3-98). The mean interval from the first documented complaint to the index admission was 90 days. Nine (33%) of 27 patients were treated with high-dose radiation therapy. The most common complaint was abdominal pain in 25 (93%) patients, nausea occurred in 12 (44%) patients, weight loss in 12 (44%) patients, and vomiting in 10 (37%) patients. The pain was described as mild in 16 (73%) of 22 patients. Only 5 (18.5%) of 27 patients had physical signs of acute peritonitis, 8 (30%) of 27 patients had some form of tenderness, and 11 (41%) of 27 had a benign exam. A total of 20 (74%) patients had an abdominal radiograph, and 12 (44%) patients had a contrast enema for evaluation. Evidence of perforation was present in 5 (25%) of 20 plain abdominal radiographs and 1 (8%) of 12 contrast enemas. Following admission, 17 (63%) patients were observed initially with subsequent surgery after symptoms either failed to resolve or worsened. The median duration under observation was 4 days (range 1-23). Surgery was performed immediately in 8 patients (30%), and 2 (7%) were observed without operation. In these 2 patients, perforation was diagnosed postmortem. Seventeen (68%) of 25 patients had a localized abscess. Three of the patients who underwent immediate exploration and 7 who had surgery after a period of observation died postoperatively (10/25, 40%). Five (55%) of 9 patients in the group who received high-dose radiation therapy died because of sigmoid perforation. When the time frame of presentation was evaluated, we noted that 10 (50%) of 20 patients died between 1960 and 1979 and 1 (14%) of 7 died between 1980 and 1992. CONCLUSIONS: Sigmoid perforation following pelvic radiation for cervical cancer does not usually present with the typical signs of a ruptured viscus. A high degree of suspicion remains a priority in the care of radiated patients who present with abdominal pain given the atypical presentation of perforation in this group.

Treatment of uterine papillary serous carcinoma with paclitaxel.

OBJECTIVE: The aim of this study was to determine the effectiveness and toxicity of monthly treatment with intravenous paclitaxel for women with advanced or recurrent uterine papillary serous carcinoma (UPSC). METHODS: Consenting women with histologically confirmed advanced (FIGO stage III or IV) or recurrent UPSC were treated on an Institutional Review Board approved protocol of a 24-h intravenous infusion of 200 mg/m(2) of paclitaxel every 3 weeks. Both measurable and nonmeasurable disease cases were enrolled. Treatment was continued until disease progression, patient intolerance, or (in women with nonmeasurable disease) completion of six courses. RESULTS: Twenty patients received from 1 to 11 cycles of therapy. Two women died of disease after 1 cycle of therapy and were not evaluable for response. Among 13 women with measurable tumor receiving 2 or more cycles of therapy, 4 had a complete clinical response and 6 had a partial response (objective response rate, 77%). The median time to progression was 7.3 months (range, 2-21 months). All 3 remaining patients with measurable disease had stable disease for a median of 6 months. The 5 patients without evaluable disease received 5 to 6 cycles of adjuvant paclitaxel. Three developed recurrence (range, 4-10 months; median, 7.2 months). Neutropenia was the major toxicity. Eleven of the 20 patients required G-CSF support, and 9 were hospitalized for neutropenic fever. One woman had reversible cardiac symptoms, which might have been related to paclitaxel treatment. At the time of analysis (mean follow-up, 23 months; range, 4.3-59.9 months), 13 women had died of disease, 4 were alive with disease, and 2 were disease free. All 3 disease-free patients had been treated for nonmeasurable advanced stage disease. CONCLUSION: Paclitaxel appears to have excellent activity in the treatment of advanced or recurrent UPSC, an uncommon but aggressive malignancy. Longer survival appears to be more common among women with small-volume disease.

A Phase II study of topotecan in patients with squamous cell carcinoma of the cervix: a gynecologic oncology group study.

OBJECTIVE: The toxicity and activity of intravenous topotecan were assessed in a multicenter Phase II study (GOG 76-U) in patients with advanced, recurrent, or persistent squamous cell carcinoma of the uterine cervix. METHODS: Intravenous topotecan was administered at 1.5 mg/m2 per day for 5 consecutive days every 4 weeks in patients without prior chemotherapy, aside from chemosensitizing agents used in conjunction with radiotherapy. The study required histologic confirmation of primary diagnosis, adequate performance status, and measurable disease to assess response. A two-stage design for accrual was used to allow for early termination of the study should inadequate response or excessive toxicity be an issue. Modifications of dose were based on hematologic toxicity. Treatment was continued until progression of disease was documented or adverse effects prohibited further therapy. RESULTS: A total of 49 patients were entered on study: of these 5 were never treated, and 1 was not evaluable for response. More than 88% (38 of 43 patients) had received prior radiotherapy. A median of two courses were administered per patient with a range of 1 to 14 cycles. Grade 4 neutropenia occurred in 68% and grade 4 thrombocytopenia in 18% of patients. Nonhematologic toxic effects were infrequent and not dose-limiting. The overall response rate (complete and partial) was 18.6%. The median progression-free survival was 2.4 months. CONCLUSIONS: Topotecan administered at this dose and schedule demonstrated moderate activity albeit at a cost of substantial hematologic toxicity in patients with advanced, recurrent, or persistent squamous cell carcinoma of the cervix.

Secondary cytoreductive surgery for localized intra-abdominal recurrences in epithelial ovarian cancer.

OBJECTIVE: The aim of this study was to evaluate the role of secondary cytoreductive surgery in patients with recurrent epithelial ovarian cancer with an apparent solitary intra-abdominal focus. METHODS: We conducted a retrospective review of patients with epithelial ovarian cancer who underwent secondary cytoreduction for recurrence at the University of Texas M. D. Anderson Cancer Center between 1985 and 1994. Eligible patients included those who had a laparotomy to resect a tumor that was apparently solitary. Cytoreductive surgery was defined as optimal if the diameter of the largest residual tumor was < or =2 cm and suboptimal if >2 cm. RESULTS: Twenty-five patients met our eligibility criteria. Their mean age was 55 years (range, 35-73 years). The median time from primary diagnosis to recurrence was 37.6 months. Tumor was found to be confined to a solitary site in 15 patients (60%), to two sites in 6 (24%), and to three or more sites in 4 (16%). Surgical procedures included cytoreduction in 10 patients, intestinal resection in 8, splenectomy in 3, and limited biopsies in 4. Secondary cytoreduction was optimal in 18 of 25 patients (72%). The median postsecondary cytoreduction survival was 25.1 months for patients who had suboptimal secondary cytoreduction compared with 56.9 months for those who had optimal cytoreduction (P = 0.08). CONCLUSIONS: Secondary cytoreductive surgery for recurrent ovarian cancer at an apparently solitary intra-abdominal site resulted in optimal residual tumor in a high proportion of patients. Although there was no survival advantage for patients whose tumor was optimally debulked, there was a trend toward improved survival. A large prospective randomized trial of secondary cytoreduction for recurrence is recommended.