Exploring the use of chronic low-dose oral etoposide in ovarian cancer: is there a role for this "new drug" in the management of platinum-refractory disease?

Despite showing high objective response rates (70% to 80%) to cisplatin- or carboplatin-based chemotherapy, most patients with ovarian cancer ultimately die of complications of their disease. Etoposide, given either as a single agent or in combination with the organoplatinum compounds, has produced disappointingly low response rates in the salvage setting. Based on recent data that suggest chronic administration of oral etoposide is superior to single daily dosing every 3 to 4 weeks, and the failure of previous trials to evaluate etoposide's activity in cisplatin-resistant malignancies, we have begun a phase II trial of chronic, low-dose oral etoposide in patients with clinically defined, platinum-resistant ovarian cancer. Thus far, 11 patients have been entered into the study. Neutropenia, the most prevalent toxicity, has precluded several patients from receiving the full 20-day course of 50 mg/d. No responses to treatment have been observed in nine evaluable patients. The study continues to accrue patients. The final results of this study and other trials should help determine the efficacy of chronic low-dose oral etoposide administration in patients with ovarian cancer.

Impact on survival of surgically defined favorable responses to salvage intraperitoneal chemotherapy in small-volume residual ovarian cancer

PURPOSE

To evaluate the impact on survival of the attainment of surgically defined favorable responses (S-R) to salvage intraperitoneal (IP) chemotherapy after initial systemic cytotoxic drug delivery.

PATIENTS AND METHODS

We examined the survival of patients who were treated on one of three phase II IP trials that were conducted at the Memorial Sloan-Kettering Cancer Center. A total of 58 patients whose largest residual tumor masses measured less than or equal to 0.5 cm in maximum diameter at the initiation of this salvage therapy were assessable for response, 28 of whom (48%) demonstrated a S-R, which included 19 (33%) who achieved a surgically defined complete response (S-CR).

RESULTS

With a median follow-up of 43+ months (range, 33+ to 58+ months) from the initiation of IP therapy, 12 of 19 (63%) have recurred. The median duration of S-CR for the 10 patients with microscopic residual disease was 32 months compared with 15 months for the nine patients with macroscopic residual disease (largest tumor mass less than or equal to 0.5 cm; P greater than .1). For patients with microscopic residual disease who experienced a S-CR (n = 10) after salvage IP therapy, the median overall survival from the initiation of therapy has not been reached, but will exceed 4 years compared with a 25-month median survival for the nonresponding patients (n = 13; P = .004). The median survival for the 18 patients with small-volume macroscopic disease who responded to therapy was 40 months compared with 19 months for the nonresponders (P = .009).

CONCLUSION

Although the results of this evaluation are encouraging and suggest that the attainment of a S-R, particularly a S-CR, after IP chemotherapy may result in a clinically meaningful favorable impact on survival, a randomized controlled trial will be required to address definitively this important issue.

Association between pretreatment CA-125 levels and surgically documented complete responses in patients with ovarian cancer treated with second-line intraperitoneal therapy

Approximately 20%-40% of patients with small-volume residual ovarian cancer, following systemically administered platinum-based chemotherapy, will respond to a second-line intraperitoneal treatment regimen. In an effort to improve the selection criteria for patients being considered for this regional therapeutic approach, we retrospectively evaluated the influence of pretreatment CA-125 levels on the ability of a group of 70 patients with small-volume residual ovarian cancer (no tumor mass greater than 1 cm in diameter) to achieve a surgically defined complete response (S-CR) following treatment on one of three phase-2 intraperitoneal chemotherapy trials conducted at the Memorial Sloan-Kettering Cancer Center. Overall, 18/46 (39%) patients with normal pretreatment CA-125 levels (less than or equal to units/ml) achieved a S-CR, compared to only 4/24 patients (17%) with an elevated pretreatment value (chi 2 = 3.7, P greater than 0.5). Despite the lower S-CR rate in patients with elevated CA-125 levels, the duration of response and survival were similar in the two patient populations achieving a S-CR. Thus, we conclude that an elevated pretreatment CA-125 level in a patient with small-volume residual ovarian cancer should not be used by itself to disqualify an individual from consideration for a second-line intraperitoneal treatment regimen, although the finding suggests a reduced likelihood of achieving a S-CR with this therapeutic approach.

Salvage intraperitoneal therapy of small-volume residual ovarian cancer: impact of pretreatment finding of peritoneal carcinomatosis on the surgical complete response rate

Second-line intraperitoneal (i.p.) therapy has been demonstrated to result in surgically defined complete responses (S-CR) in 25%-40% of patients with small-volume residual ovarian cancer (microscopic disease or largest tumor mass less than 1 cm in diameter). To evaluate the influence of the surgical finding of diffuse peritoneal carcinomatosis on the S-CR rate to salvage i.p. therapy in this patient population, we retrospectively reviewed the operative reports of 70 patients with small-volume residual ovarian cancer treated on one of three phase-2 second-line i.p. trials at the Memorial Sloan-Kettering Cancer Center. Of the 11 patients with diffuse carcinomatosis, none achieved a S-CR compared to a S-CR rate of 37% (22/59) in patients without this surgical finding (chi 2 = 6.0; P less than 0.025). However, of the 7 patients with diffuse carcinomatosis treated on a cisplatin-based i.p. program, the only 2 who had previously responded to systemic platinum also experienced a response (partial) to the i.p. cisplatin regimen. In conclusion, while the surgical finding of diffuse peritoneal carcinomatosis indicates a poor prognosis in patients with small-volume residual ovarian cancer for response to i.p. chemotherapy, patients in this clinical setting with prior evidence of platinum sensitivity may experience some benefit from an i.p. cisplatin-based treatment strategy.

Impact of laparotomy finding of significant intraabdominal adhesions on the surgically defined complete response rate to subsequent salvage intraperitoneal chemotherapy

One possible explanation for the failure of the high concentrations of cytotoxic agents achieved following intraperitoneal (i.p.) drug delivery to produce a favorable response in patients with ovarian cancer is the inability of the drug-containing fluid to be adequately distributed throughout the peritoneal cavity, usually because of intraabdominal adhesion formation. To evaluate the influence of the severity of adhesions, observed at the time of laparotomy performed immediately preceding the initiation of i.p. therapy, on the ability to achieve a surgically defined complete response (S-CR), we retrospectively reviewed the operative reports of 70 patients with small-volume residual ovarian cancer treated on one of three phase-2 salvage i.p. trials at the Memorial Sloan-Kettering Cancer Center. The S-CR rate in the 36 patients with limited adhesion formation observed upon entering the peritoneal cavity was 28%, compared to 35% in the 34 patients with extensive adhesions (P greater than 0.05). In 33 patients treated with a phase-2 cisplatin-based i.p. program, who had previously responded to systemic platinum, 47% (8/17) and 44% (7/16) of those with limited and extensive adhesions, respectively, achieved a S-CR (P greater than 0.05). We conclude that the presence of extensive adhesions observed within the peritoneal cavity at the time of a laparotomy performed immediately prior to the initiation of i.p. therapy does not have a negative impact on the potential to achieve an S-CR, assuming it is technically feasible to lyse all significant adhesions prior to the completion of the operative procedure.

Ifosfamide and mesna in previously treated advanced epithelial ovarian cancer: activity in platinum-resistant disease

PURPOSE

There is a critical need to find new antineoplastic drugs that are active in platinum-refractory ovarian cancer. We conducted a phase II trial of single-agent ifosfamide with mesna uroprotection in patients with ovarian cancer previously treated with an organoplatinum compound to assess its activity in this clinical setting.

PATIENTS AND METHODS

Ifosfamide (1.0 or 1.2 g/m2/d for 5 days, delivered on a monthly schedule) was administered to the 57 patients entered onto this trial. Dose reductions were permitted for unacceptable toxicities.

RESULTS

Toxicity included severe bone marrow suppression (WBC count less than 1,000/microL and/or platelet count less than 50,000/microL), renal dysfunction (serum creatinine level greater than 2.0 mg/dL), and reversible CNS dysfunction (disorientation, hallucinations, somnolence, and agitation), which occurred in 20%, 14%, and 12% of patients, respectively. Of 41 patients with strictly defined platinum-refractory ovarian cancer, five (12%) demonstrated a partial (four) or complete (one) response to this treatment program.

CONCLUSION

Single-agent ifosfamide has modest but unequivocal activity in platinum-resistant ovarian cancer. Further studies of this drug used as a front-line agent along with an organoplatinum compound or as part of a dose-intensification program with bone marrow, peripheral stem cell, or colony-stimulating factor support are indicated. In addition, single-agent ifosfamide is a reasonable standard second-line treatment strategy in appropriately selected patients with platinum-refractory ovarian cancer.

Phase 2 trial of chronic low-dose oral etoposide as salvage therapy of platinum-refractory ovarian cancer

Eighteen previously treated patients with advanced ovarian cancer were entered into a phase 2 trial of chronic low-dose oral etoposide (50 mg/day for 20 days, repeated every 28 days) to determine the activity of this therapeutic strategy in organoplatinum-refractory disease. The treatment program was generally well tolerated, with mild neutropenia the most common side-effect encountered. One patient (6%; 95% confidence interval = 0-17%) achieved a partial response, which lasted for 11 months. Three additional patients (17%), who failed to meet the criteria of a partial response, demonstrated objective evidence of antineoplastic activity. Chronic low-dose oral etoposide administration is associated with definite, although modest, activity in platinum-refractory ovarian cancer.

Phase II trial of high-dose cisplatin with sodium thiosulfate nephroprotection in patients with advanced carcinoma of the uterine cervix previously untreated with chemotherapy

Cisplatin is one of the most active single agents in the treatment of advanced cancer of the cervix. The concurrent administration of the nephroprotective agent, sodium thiosulfate, has enabled exploitation of the therapeutic potential of cisplatin. To explore the role of cisplatin dose intensity in the treatment of patients with cancer of the uterine cervix, patients with persistent/recurrent measurable disease were treated with cisplatin at 200 mg/m2 as a 2-hr infusion with sodium thiosulfate given at 3.3 g/m2 1 hr prior to cisplatin and 6.6 g/m2 during the cisplatin infusion. Treatment was repeated monthly. Due to the known cumulative toxicity of cisplatin, treatment beyond two cycles (400 mg/m2) was given only to those patients who had at least demonstrated a PR. Audiologic evaluation was done prior to each cycle of treatment. Eleven patients were entered with a median age of 43 years (range, 25-57), a median KPS of 80% (range, 60-90%), and nine epidermoid and two adenocarcinoma, and all patients had received previous pelvic irradiation. Twenty-eight cycles of treatment were given: 1, five cycles; 3, three cycles; 7, two cycles. No greater than or equal to 3+ hematologic, neurologic, or renal toxicity was demonstrated. Ototoxicity was demonstrated in the mild to moderate hearing loss range (3000-8000 Hz). The greatest threshold shift occurred after the first course of cisplatin. There were three PRs with a maximum duration of 4 months. Due to the significant toxicities encountered, the low response rate, and the limited duration of responses, this trial was closed early to accrual.

Responses to second-line cisplatin-based intraperitoneal therapy in ovarian cancer: influence of a prior response to intravenous cisplatin

Phase II trials of second-line intraperitoneal (IP) cisplatin-based therapy in patients with ovarian cancer have demonstrated the ability of this approach to produce objective antitumor responses, including surgically defined complete responses (CRs), in individuals with persistent small-volume disease after front-line cisplatin-based intravenous (IV) treatment. To examine the influence of a prior response to systemic cisplatin on the activity of second-line IP cisplatin, we retrospectively analyzed two phase II trials of cisplatin-based IP therapy in persistent/recurrent ovarian cancer conducted at our institution. Of the 89 assessable patients on the two trials, 52 (58%) had previously responded to IV cisplatin. The overall response and CR rates to second-line IP cisplatin-based therapy in this previously responding population were 56% and 33%, respectively, compared with overall response and CR rates in the 37 nonresponders to IV cisplatin of 11% and 3%, respectively (P less than .001; chi 2, 1 df). In the 36 patients responding to systemic cisplatin and whose largest tumor mass measured less than 1 cm at IP cisplatin initiation, a 42% CR rate was observed, compared with a 7% CR rate in the 14 patients with the same bulk of disease who had previously failed to respond to systemic cisplatin (P less than .025). We conclude that a prior response to systemic cisplatin strongly influences the antineoplastic activity of second-line IP cisplatin in ovarian cancer.

Phase II trial of weekly or biweekly intraperitoneal mitoxantrone in epithelial ovarian cancer

Previous experimental and clinical evaluation has suggested that ovarian cancer is sensitive to the cytotoxic effects of mitoxantrone at concentrations achievable within the peritoneal cavity after intraperitoneal (IP) administration. Unfortunately, the use of the drug delivered IP at high doses (20 mg/m2 in 2 L normal saline [NS]) on a monthly schedule is compromised by severe local effects secondary to the irritant properties of the drug. To reduce toxicity and take advantage of minimal systemic drug exposure following IP administration, we treated 28 patients with a lower drug concentration of mitoxantrone (10 mg/m2 in 2 L NS), but on a weekly or every other week schedule (total, 12 courses). Compared with the monthly program, this regimen caused less pain, allowed for a higher cumulative dose of mitoxantrone to be delivered, and resulted in less serious treatment-related morbidity. Four of 13 assessable patients (31%) whose largest tumor was less than or equal to 1 cm in diameter demonstrated a surgically defined response. All responding patients had failed previously or exhibited a minimal response to cisplatin. Despite the improved toxicity profile of this regimen, the overall response rate was similar to the monthly program, probably secondary to inadequate IP drug distribution in many patients. Future investigative efforts using IP mitoxantrone as therapy for ovarian cancer might focus on developing methods to improve drug delivery to all sites of tumor within the peritoneal cavity (eg, intraoperative therapy, increased treatment volumes, and antiinflammatory agents to reduce adhesion formation).

Late effects of cisplatin-based chemotherapy on renal function in patients with ovarian carcinoma

While the acute toxicity of cisplatin on renal function is well described, the long-term effects have received little attention in the medical literature. In an effort to examine this important issue, we evaluated the renal function of a group of 60 women with ovarian cancer who received a second-line cisplatin-based chemotherapy program at the Memorial Sloan-Kettering Cancer Center longer than 6 months following the completion of their initial cisplatin-based chemotherapeutic regimen. These patients had not received any cisplatin for a median of 19 months (range, 6-57 months) prior to beginning the second-line program. Only 4 patients (7%) had a serum creatinine value greater than 1.3 mg/dl (upper limit of normal in our laboratory) at the time of initiation of the second-line program. In each case, the serum creatinine was less than or equal to 1.6 mg/dl. An additional 6 patients (10%) had a greater than 50% increase in serum creatinine above the initial baseline value prior to any treatment, but were still within the normal range. Following second-line cisplatin therapy, only 1 of the 60 patients (2%) experienced a serum creatinine rise to greater than 2.0 mg/dl. We conclude that a limited number of patients with ovarian cancer who are long-term survivors following cisplatin-based chemotherapy exhibit evidence of a mild persistent compromise of renal function and, in general, second-line cisplatin can be administered to previously treated patients without the development of serious renal dysfunction.

Second-line platinum therapy in patients with ovarian cancer previously treated with cisplatin

In an effort to critically define the incidence and clinical characteristics of secondary responses to cisplatin-based therapy in patients with ovarian cancer previously treated with a cisplatin-based program, a retrospective review was undertaken of patients at the Memorial Sloan-Kettering Cancer Center who received greater than or equal to two cisplatin/carboplatin-based programs. Eighty-two patients were identified who met the entry criteria of having had a cisplatin-free interval (CFI) of more than 4 months between the completion of their first regimen and the institution of a second cisplatin/carboplatin program. Of the 72 assessable patients (10 had no measurable disease, and a laparotomy was not performed to assess response), 31 (43%) responded, including 10 surgically defined complete responses (S-CRs). The overall response rates (and S-CR rate), based on duration of CFI, were 5 to 12 months, 27% (5%); 13 to 24 months, 33% (11%); and more than 24 months, 59% (22%). Twenty-nine patients (35%) received noncisplatin/carboplatin-containing treatments between the cisplatin programs. Patients without any treatment for more than 24 months from the completion of their initial therapy experienced a 77% (17 of 22) response rate and a 32% (seven of 22) S-CR rate. In conclusion, secondary responses to cisplatin/carboplatin-based treatment are common in patients with ovarian cancer who have previously responded to the agents and increase in frequency with greater distance from the initial therapy.

Persistent hypomagnesemia following cisplatin chemotherapy in patients with ovarian cancer

The development of acute hypomagnesemia following cisplatin administration is a well-recognized complication associated with the use of this chemotherapeutic agent. However, there is limited information available in the medical literature concerning how long this abnormality may persist following the discontinuation of cisplatin. Of 13 patients with ovarian cancer who had a baseline serum magnesium determination obtained prior to the initiation of a second-line cisplatin-based chemotherapy regimen, 9 (69%) were found to be hypomagnesemic (serum magnesium less than 1.4 mg/l), including 3 patients with serum magnesium values less than 1.0 mequiv/l. The median cisplatin-free interval for these 9 patients was 19 months (range 6-40 months). We conclude that persistent, and possibly permanent, hypomagnesemia is common following cisplatin chemotherapy.

Intraperitoneal cisplatin and cytarabine in the treatment of refractory or recurrent ovarian carcinoma

Preclinical evaluation has suggested impressive concentration-dependent cytotoxic synergy between cisplatin and cytarabine in ovarian carcinoma. To further evaluate the clinical relevance of these observations, 39 patients with refractory or recurrent ovarian carcinoma were entered onto a phase II trial of intraperitoneal (IP) cisplatin (100 to 105 mg/m2 per course) plus cytarabine (600 to 900 mg per course). Treatment was administered over 2 or 3 days for a maximum of five monthly courses, followed by surgical reevaluation in patients without clinical evidence of disease. The 3-day regimen was discontinued secondary to the development of severe thrombocytopenia (five of 12 courses platelets decreased to less than 50,000/mm3). Additional toxicities included abdominal pain (moderate to severe at some time during therapy in 46% of patients), fever without evidence of infection (44%), and bacterial peritonitis (10%). Three patients declined surgical reassessment. Fourteen of 36 (39%; 95% confidence interval [CI], 23% to 55%) assessable patients demonstrated surgically defined responses, including 12 of 23 (52%; 95% CI, 32% to 72%) patients with tumor nodules less than 1 cm in diameter and only two of 13 (15%; 95% CI, 0% to 34%) patients with any lesion greater than 1 cm. There were seven (30%; 95% CI, 11% to 49%) surgically defined complete responses (CRs) in patients with less than 1 cm disease and none in patients with larger tumor nodules. IP cisplatin/cytarabine results in a high surgically defined response rate in patients with minimal residual ovarian carcinoma, but activity is low in patients with bulky intraabdominal disease.

Phase II trial of intraperitoneal mitoxantrone in the management of refractory ovarian cancer

To define both the toxicity and efficacy of intraperitoneal mitoxantrone in the treatment of refractory ovarian carcinoma, 31 patients were entered onto a phase II trial of this agent delivered in a 2 L treatment volume on a monthly basis. Due to excessive local pain at the initial dose level (30 mg/m2), the amount of drug delivered with each treatment course was reduced to 20 mg/m2. Despite this reduction, 74% of patients required narcotic analgesia during treatment. In addition, there were four episodes of bowel obstruction (one requiring surgical intervention) during therapy, and two patients developed bowel obstruction and intraabdominal abscesses following the completion of treatment. Six of 18 evaluable patients (33%) whose largest tumor diameter was less than or equal to 1 cm at protocol initiation experienced surgically documented responses, compared with one of 11 patients (9%) whose largest tumor was greater than 1 cm in diameter. If the two patients exhibiting what we called a mixed response to treatment are included, seven of 21 patients previously treated with intraperitoneal cisplatin responded to this treatment program, including four patients who had failed to respond to intraperitoneal cisplatin. No responding patient has demonstrated clinical evidence of relapse with a median follow-up of 7 months (range, 3+ to 13+ months) from response laparotomy. Intraperitoneal mitoxantrone is an active treatment program in patients with small-volume refractory ovarian carcinoma, but local toxicity can be severe. Due to the toxicity encountered with this specific program, its use cannot be recommended for standard clinical practice. However, in view of the activity observed in refractory ovarian carcinoma, including responses in patients who had previously failed intraperitoneal cisplatin, it is important to continue to explore alternative therapeutic regimens using intraperitoneal mitoxantrone to reduce local toxicity while maintaining or improving efficacy.

Intraperitoneal cisplatin and etoposide in the treatment of refractory/recurrent ovarian carcinoma

To determine the efficacy of a 6-month course of combination intraperitoneal (IP) chemotherapy with cisplatin and etoposide in patients with refractory or recurrent advanced ovarian carcinoma, 67 patients were entered into this prospective, nonrandomized, single-institution trial. Cisplatin at 100 mg/m2 and etoposide at 200 mg/m2 were administered IP on day 1 every month for 6 months. Exploratory laparotomy was performed before protocol entry and was planned after the completion of 6 months of IP therapy to surgically document response. All patients had received prior intravenous (IV) chemotherapy with a cisplatin-based regimen. At protocol entry, 18 (27%) patients had surgically defined residual tumor (maximal tumor diameter) greater than 2.0 cm, 17 (25%) patients greater than 0.5 cm - less than or equal to 2.0 cm, and 32 (48%) patients less than or equal to 0.5 cm. Sixteen patients (24%) who had experienced a treatment-free interval of more than 1 year prior to study entry were considered as having recurrent disease and the remaining 51 (76%) patients were considered as having refractory disease. Toxicity was tolerable: four patients (6%) had nadir fever, three (4%) had culture-documented bacterial peritonitis, five (7%) had IP catheter-related complications, and 27 (40%) had an increase in serum creatinine greater than 1.5 mg/dL. Among the 57 patients who are fully evaluable for response, the overall surgically defined response rate, complete (CR), and partial response (PR), was 40% (23/57), and the CR rate was 21% (12/57). Among the patients with recurrent disease, eight of 13 (62%) responded, with responses seen among all categories of residual disease. Among the patients with refractory disease, 15 of 44 (34%) had surgically documented responses. However, responses were more frequent in patients with residual disease less than 0.5 cm; 11 of 20 (55%) versus four of 24 (17%) with residual greater than 0.5 cm, P = .019 (chi 2, one degree of freedom, Yates correction). The duration of the CRs ranges from 4 to 18+ months. Longer follow-up is needed to determine if there is any impact on survival.

Intraperitoneal therapy in the management of ovarian carcinoma

The intraperitoneal administration of chemotherapeutic and biological agents as therapy of ovarian carcinoma is based on both theoretical considerations and experimental evaluations which suggest that tumor present in the cavity can be exposed to higher concentrations of certain antineoplastic drugs than can be accomplished if the agents are administered systemically. Recent clinical data have confirmed both the safety and pharmacokinetic advantage associated with this approach. Surgically defined responses have been observed in patients with small-volume residual refractory ovarian carcinoma treated with several single-agent and combination intraperitoneal therapeutic programs. While significant activity has been noted in this clinical setting, a clearly defined role for intraperitoneal treatment in the standard management of ovarian carcinoma remains to be determined.

A phase I trial of intraperitoneal recombinant gamma-interferon in advanced ovarian carcinoma

The interferons are a class of biological agents that have demonstrated antineoplastic activity in a variety of tumors both in vitro and in vivo. Previous reports have suggested that interferons can be safely administered by the intraperitoneal (IP) route with a pharmacokinetic advantage for peritoneal cavity exposure compared with the systemic circulation and with objective antitumor activity being demonstrated. On the basis of these reports and laboratory data suggesting activity for recombinant gamma-interferon (r-GIFN) against several malignant cell lines, we treated 27 refractory ovarian carcinoma patients, including six with very-small-volume residual disease, with this agent delivered by the IP route. While r-GIFN was found to be remarkably well tolerated, with a 150- to 200-fold pharmacokinetic advantage for peak levels achieved in the peritoneal cavity compared with the plasma, no objective responses were observed. Despite the lack of demonstrated activity for single-agent IP-administered r-GIFN in this clinical setting, there remains considerable interest in this agent when delivered by the IP route because of in vitro data suggesting concentration-dependent synergy between r-GIFN and other biological agents.

Nonsquamous cancer of the vagina

This report retrospectively analyzes 48 cases of primary vaginal cancer treated at Walter Reed Army Medical Center and the Naval Hospital, Bethesda, from 1962 through 1983. There was an unusually high number of uncommon histologic types. Nine patients had an adenocarcinoma, 5 had a sarcoma, 3 had a melanoma, 2 had an adenosquamous carcinoma, 1 had a lymphoma, and 1 had a carcinoid tumor. The remaining 27 patients had a squamous cell carcinoma. This represents a 43% prevalence of nonsquamous lesions. Nonsquamous cancer of the vagina occurred in patients at an earlier age than squamous cell carcinoma. Presenting symptoms, the location of the tumor within the vagina, and survival rates were similar for both groups. The clinical characteristics and treatment of the patients with nonsquamous tumors are discussed.