Introduction

The standard treatment of primary ovarian cancer includes initial surgical cytoreduction followed by combination platinum and taxane-based chemotherapy.  Approximately 80% of patients will respond to this form of initial management.  Unfortunately, almost 50% of the responders will likely relapse, making their disease essentially incurable.  Strategies to improve quality of care and overall survival in ovarian cancer are directed toward improving the likelihood and duration of response to primary therapy, extending the disease free interval for patients in remission, and improving the likelihood and duration of response for patients with recurrent disease.

At ASCO 2004, there were no positive, pivotal trials in initial treatment for ovarian cancer. Results from two trials showed that there was no difference in outcomes with the addition of an anthracycline to the combination of platinum and taxane-based primary therapy in the initial treatment of stage IIB to IV ovarian cancer. Furthermore, an anti-MUC1, yitrium-90-labeled, monoclonal antibody (R1549) did not offer a survival advantage and high-dose chemotherapy with peripheral stem cell support does not appear to have a role in the management of ovarian cancer.

However, it is refreshing to note that advances in the treatment of recurrent disease were reported.  Results from a phase III randomized trial confirm that combination therapy in platinum-sensitive patients with recurrent disease is superior to single agents given alone. Also, several phase II trials evaluating Iressa® in combination with other agents show activity in platinum-sensitive patients. Researchers did report that prolonging Taxol® infusion does not appear to improve survival.

Triplet Therapy in First-Line Treatment

At ASCO 2003 (39th Annual Meeting), two studies concluded that the addition of topotecan (Hycamtin®) after completion of first-line therapy with platinum and taxane-based agents did not improve response rates or progression-free survival.[1],[2]  At ASCO 2004, two studies were presented that tested the hypothesis that the addition of an anthracycline as a third agent to the combination of platinum and taxane-based primary therapy would improve overall or progression-free survival. Both of these studies found no difference in outcome with the addition of an anthracycline in stage IIB to IV ovarian cancer. These results are discussed below and compared in accompanying tables.

Taxol®/Paraplatin® plus Ellence®: A phase III trial presented by Andreas du Bois on behalf of the AGO-GINECO Intergroup suggests that the addition of epirubicin (Ellence®) to platinum and taxane-based therapy in the upfront setting is not warranted, is more toxic, and does not improve outcome in stage IIB to IV ovarian cancer.[3]   Patients were randomized to receive six cycles of paclitaxel (Taxol®, 175mg/m2 over 3 hours) and carboplatin (Paraplatin®, AUC 5) with (TEC) or without (TC) Ellence® (60mg/m2). Both regimens were repeated every three weeks.  The maximum tolerable dose of Ellence® in combination with paclitaxel and carboplatin were previously determined in a Phase I/II trial published in 1999.  A predetermined subgroup analysis of optimally and suboptimally debulked patients was planned.  The primary endpoint was overall survival (OS). 

Over a 30 month period, 1282 patients were randomized, of which 1243 were treated and evaluable for overall and progression-free survival.  Nearly 75% of the patients in both groups had stage III disease and serous histology.  Optimal debulking was performed in approximately 70% of the patients in both arms.  Approximately 75% of the patients in each group completed six cycles of therapy.  TEC was found to be significantly more myelotoxic (see table 1) and had increased gastrointestinal toxicities, but did not increase other non-hematologic toxicities including cardiac toxicity, a concern with anthracyclines.  The overall response rate was similar between the two treatment arms and is shown in Table 2.  After a median follow-up of 52 months, the OS and progression-free survival (PFS) were not statistically different between the two groups (Table 3).  As expected, the planned subgroup analysis found that the OS for optimally debulked patients was better than the OS of suboptimally debulked patients.  Within both the optimally or suboptimally debulked subgroups, there was no difference in OS (Table 3). 

A similar study to the one outlined above was presented by Gunnar B. Kristensen on behalf of the Gynecologic Cancer Intergroup (GCIG) and showed that the addition of Ellence® to standard treatment did not improve PFS in patients with stage III disease.[4] Key differences between this previously published study and the AGO-GINECO Intergroup trial were that dose of Ellence® was higher and the primary endpoint was PFS.  Patients were randomized to receive six to nine cycles of Taxol® (175mg/m2 over 3 hours) and Paraplatin® (AUC 5) with (TEC) or without (TC) Ellence® (75mg/m2). Both regimens were repeated every three weeks.   A predetermined subgroup analysis of optimally and suboptimally debulked patients was planned. 

Over a 30 month period, 887 patients were randomized.  Approximately 73% of the patients had stage III disease and 67% had serous histology.  Optimal debulking was performed in approximately 40% of the patients in both arms, and 85% of the patients in each group completed at least six cycles of therapy.  TEC was found to be more toxic overall (see table 1), without increasing cardiac toxicity.  The overall response rate was similar between the two treatment arms and is shown in Table 2.  After a median follow-up of 30 months the PFS was not statistically different between the two groups (Table 3).  Neither the optimally or suboptimally debulked patients had a difference in PFS with regard to treatment arm (Table 3). 

Table 1  Grade 3 and 4 toxicities associated with TEC and TC regimens

* p <0.001 for each toxicity

** p <0.01

*** p <0.05

Table 2  Response rates for the TEC and TC regimens

Table 3  Overall (OS) and progression-free survival (PFS) for TEC and TC regimens

* Hazard ratio

** OS presented for AGO-GINECO study and PFS presented for GCIG study

New (and Old) Strategies in Consolidation

Though most ovarian cancer responds to standard front-line treatment, nearly half of these responders will develop recurrent disease.  Consolidation therapy is aimed both at extending the disease-free interval and curing patients who would otherwise recur.  Two studies presented at ASCO 2004 investigated different consolidation strategies that ranged from tolerable to aggressive in nature.  As was seen with the attempt to add a third drug to upfront treatment, neither of these consolidation strategies appear to be more affective than standard care.

The Study of Monoclonal Antibody Radioimmuntherapy (SMART): The SMART study capitalized on data from a Phase I/II study suggesting that an anti-MUC1, yitrium-90-labeled, monoclonal antibody (R1549) offered a survival advantage.[5]  Since over 90% of ovarian cancers express the MUC1 antigen, this seems to be a rationally designed intervention. However, results from the phase III SMART study suggest that treatment with R1549 immediately after a negative SLL does not improve overall or progression-free survival.[6] 

The SMART study involved patients with stage IC to IV ovarian cancer who had completed primary chemotherapy and had no clinical evidence of disease were randomized to a single intraperitoneal dose (IP) of R1549 via an IP catheter. Intraperitoneal delivery was utilized so that the drug could directly target locally persistent and microscopic disease. Treatment was typically administered the day after surgery, or whatever standard treatment was customary at the given institution.  All patients underwent a second look laparoscopy (SLL) and those found to have visible disease or whose adhesions prevented an adequate evaluation were removed from the study.  The remaining patients were treated per randomization.  The study was powered to detect a 15% improvement in overall survival. 

In total, 702 patients were randomized, 133 had an inadequate SLL, 122 had visible disease discovered at SLL, leaving 224 patients in the treatment arm and 223 patients in the control arm.  The two arms were well matched for most clinical variables.  Stage III or greater disease was present in 70% of the patients and 60% has serous histology.  All patients had received prior platinum-based therapy and over 90% had also received a taxane as part of their front-line treatment.  In the treatment arm 44% of the patients had residual disease after initial surgery as compared to 36% in the control arm.  Ten percent of the patients in both groups had microscopic disease found at the time of SLL. 

Adverse reactions in general were more common in the active treatment arm. The most common serious adverse event was intestinal obstruction which occurred in 10% of the patients in the treatment arm and 12.5% of patients in the control arm.  Approximately 5% of patients in both arms required a conversion from laparoscopy to laparotomy.  Five percent of patients in each arm had serious nausea and vomiting.  Minor adverse events occurred more commonly in the treatment arm.  Hematologic toxicity peaked at week 6 to 8 following treatment and approximately 80% of women in the active treatment arm had some grade of hematologic toxicity.  Grade 3 or 4 hematologic toxicity occurred in approximately 20% of patients in the treatment arm and infrequently in the control arm.  Thrombocytopenia was the most common hematologic toxicity. 

After a median follow-up of almost 3 years, 130 patients have died. Results showed that treatment with R1549 produced no improvement in OS or PFS. The hazard ratio for OS is 1.16 (0.8 – 1.6) and 0.9 (0.7 – 1.2) for PFS.  Furthermore, planned subgroup analyses failed to identify any group in which a benefit was identifiable.  Though no treatment benefit was appreciated, a large prospective database of patients with negative SLL has been created which will be useful for future studies. 

High-dose chemotherapy (HDC) versus conventional dose maintenance (CDM): A randomized phase III trial provided evidence that HDC does not have a role in advanced ovarian cancer, as has been found in other solid tumors.[7] Patients with stage III or IV disease were randomized to received HDC as Paraplatin® (400mg/m2) and cyclophosphamide (1500mg/m2) given daily for four days with peripheral blood stem cell (PBSC) support or CDM consisting of three cycles of carboplatin (Paraplatin®, 300mg/m2) and cyclophosphamide (500mg/m2) given every four weeks.  Eligibility criteria also included age less than 60 years, completion of 4 to 6 cycles of platinum-based chemotherapy, disease less than 2 cm at second look surgery or a pathologic complete response with high-risk features (such as suboptimal debulking, failure to normalize CA125 within the first three cycles of chemotherapy, etc). 

Over a five year period, 110 patients were enrolled.  Most patients had stage III disease and 40% had visible disease at the time of second look surgery.  After a median follow-up of 60 months, 94 patients have relapsed and 63 have died.  The median overall survival was 43 months (29–57) in the CDM arm and 50 months (30–69) in the HDC arm.  The progression-free survival in the CDM arm was 12 months (7–17) and 18 months (5–30) in the HDC arm.  Two patients died of treatment-related complications.  The median time to hematologic recovery was 11 days. 

Doublet Therapy in Recurrent Disease

The idea that single agents are equally efficacious as combination chemotherapy in the recurrent setting has recently been called into question.  Data presented at ASCO 2003 from the ICON-4 trial demonstrated that the combination of platinum with Taxol® given in recurrent platinum-sensitive patients is superior to platinum alone.[8]  At ASCO 2004, results from a Gynecologic Cancer Intergroup (GCIG) phase III randomized trial confirms that combination therapy in platinum-sensitive patients with recurrent disease is superior to single agents given alone.[9] 

The GCIG trial compared the combination of gemcitabine (Gemzar®) and Paraplatin®, a regimen that is potentially less neurotoxic than Paraplatin®/Taxol®, to Paraplatin® alone. This trial was conducted in patients who had evaluable recurrent disease and had not received their last platinum-containing regimen within the prior 6 months. Patients were randomized.to receive Paraplatin® (AUC 5) every 21 days for at least 6 cycles or Paraplatin® (AUC 4) given on day 1 in combination with Gemzar® (1000mg/m2) given on days 1 and 8 with both drugs repeated every 21 days for at least 6 cycles.  The primary outcome measure was progression-free survival.

Over a 30 month period, 356 patients were randomized. All patients had received front-line chemotherapy with a platinum-containing agent and 68% had also received Taxol®.  The platinum-free interval was greater than 12 months in 60% of the patients. 

The response rate and PFS were superior in the combination arm (Table 4). Patients in the GC arm also had more hematologic toxicity (Table 5); however, quality of life was superior in the GC arm as compared to the C agent arm.

After a median follow-up of 17 months, PFS in the GC arm was 8.6 months versus 5.8 months in the C arm (HR 0.72, p =0.003).  Overall survival was 17–18 months and not different between the two groups (HR =0.96, p =0.73), though the trial was not powered for OS.  Planned subgroup analyses revealed similar results when the PFS was compared between patients with a platinum-free interval of 6 to 12 months, more than 12 months, prior exposure to Taxol®, or measurable disease. 

Table 4  Response rates for the GC and C regimens

* p < 0.002

Table 5  Grade 3 and 4 toxicities associated with GC and C regimens

* p < 0.001

This study confirms that combination therapy in platinum-sensitive patients with recurrent disease is superior to single agents given alone.  Nonetheless, many questions are left unanswered.  Which of the tested combination regimens (GC or PC) or others are most efficacious and least toxic?  Would these combination therapies be as effective and less toxic if given as planned sequential therapy?  Can biologic agents improve outcomes even more if given in combination with these or other doublets?  The future will certainly contain additional trials aimed at answering these and other questions. It is refreshing to know that advances are being made in the treatment of recurrent disease as well as in primary care.

Phase II trials: Two smaller phase II trials, each with 105 patients that tested other doublets in platinum-sensitive patients with recurrent disease deserve mention.  Both Liposomal doxorubicin in combination with Paraplatin® (DC) produced an overall response rate (RR) of 63% with a PFS of 9.4 months and an OS of 32 months.[10]  Oxaliplatin (Eloxatin®) in combination with Taxol® (OP) an 88% RR with a PFS of 10.2 months and the OS was projected to be more than 30 months.[11] 

Taxol® 24-hour infusion vs 96-hour infusion : In addition to the trials discussed above, results of GOG 162 were presented by David Spriggs.  This randomized phase III trial compared cisplatin (Platinol®) given in combination with Taxol®, either as a 24-hour infusion or a 96-hour infusion, in suboptimally debulked advanced ovarian cancer.  This is one of the few trials restricted to the group of patients with the poorest prognosis, and it demonstrated that survival was not improved when the Taxol® infusion was prolonged.[12] 

Iressa® (gefitinib): There were also several phase II trials that studied Iressa® in combination with other agents in recurrent disease.  As expected, the responses were greater in platinum-sensitive patients than platinum-resistant patients, but it was unclear if these rates were any different than seen with single agents or doublets given to similarly pretreated patients.[13],[14],[15]

Conclusions

In conclusion, results reported at the 2004 ASCO meeting confirm previously reported findings:

• That the addition of an a third agent, Hycamtin® in 2003 studies and anthracyclines in the 2004 studies, to platinum and taxane-based combinations does not improve outcomes in the intial treatment of ovarian, and
• Doublets appear to improve outcomes over single-agent treatment in recurrent disease.

It will be of interest to see how molecularly targeted agents can be incorporated into the treatment of primary and recurrent ovarian cancer. While the researchers reported that an anti-MUC1, yitrium-90-labeled, monoclonal antibody (R1549) did not provide a survival advantage, smaller phase II trials reported that Iressa® produced some activity in recurrent disease.

References