|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology Current Status of Taxane and Platinum-Based Chemotherapy in Ovarian CancerFrom the Oncology Service Line, Franklin Square Hospital, Baltimore, MD. Address reprint requests to Oncology Service Line, Franklin Square Hospital, 9000 Franklin Square Drive, Baltimore, MD 21237; email: wmcguire52{at}aol.com.
Before 1993, the standard of care for the chemotherapeutic management of advanced ovarian cancer was cisplatin or carboplatin combined with a classic alkylating agent (typically cyclophosphamide). Studies in the 1990s have changed this standard to one of the platinum-containing agents combined with a taxane, paclitaxel, or docetaxel. This article reviews the pertinent studies regarding paclitaxel combined with cisplatin or carboplatin, discusses the remaining controversies surrounding how best to combine these agents, and provides opinions regarding the discordant outcomes noted in studies of the paclitaxel-platinum doublet. A separate article discusses the docetaxel-platinum doublet and how that might be considered an appropriate option for first-line therapy in patients with advanced, newly diagnosed ovarian cancer.
SINCE THE mid-1990s, some combinations of platinum and taxane have become the standard of care for advanced ovarian carcinoma (stages IIB to IV) in much of the world. The remainder of the decade of the 1990s focused on determining which platinum analog had a better therapeutic index and the proper dose and duration of infusion of paclitaxel. Several new drugs with unique mechanisms of action have been discovered and found to be active in recurrent ovarian cancer. During the next decade we will undoubtedly see studies aimed at determining the roles of these drugs as part of primary therapy.
In 1996, the results of a randomized Gynecologic Oncology Group (GOG) trial showed a significant outcome advantage in suboptimally debulked patients when cisplatin and paclitaxel were combined (paclitaxel 135 mg/m2 as a 24-hour infusion followed by cisplatin 75 mg/m2) compared with the previous standard of care1 of cyclophosphamide combined with either cisplatin or carboplatin. A replication of that study was performed by a European-Canadian consortium, and mature results of that study showed once again that the doublet of cisplatin and paclitaxel was superior to cyclophosphamide and cisplatin in advanced ovarian cancer.2 Differences between the GOG trial and the subsequent trial were that 35% of patients had optimally debulked disease in the consortium study compared with none in the GOG trial and that the European-Canadian trial used a 3-hour paclitaxel infusion at a dose of 175 mg/m2 rather than the 24-hour schedule and lower dose used in the GOG study. In the European-Canadian trial, 24% of patients developed grade 2 or greater peripheral neurotoxicity compared with only 4% in the GOG trial, demonstrating the effect of infusion duration on adverse events (paclitaxel is more neurotoxic in higher doses and shorter infusions). The median progression-free survival was identical to that seen in the GOG trial, although one third of the patients had optimal disease. This is more likely because of the inclusion of some optimal patients in the GOG study rather than inferiority of the 3-hour paclitaxel and carboplatin regimen compared with 24-hour paclitaxel and cisplatin. The combination of carboplatin and 3-hour paclitaxel was found to be a safe and highly active regimen in a phase 1 trial conducted by the GOG.3 On the basis of this small study and the ease and safety of outpatient administration, the combination had been adopted by most physicians even before any study was performed to demonstrate its equivalence with the longer infusion of paclitaxel with cisplatin. Although carboplatin was equivalent to cisplatin in combination with cyclophosphamide, assuming that the same will be true in combination with paclitaxel is somewhat dangerous. There is a well-known interaction between carboplatin and paclitaxel on the megakaryocyte such that carboplatin causes less thrombocytopenia in combination with paclitaxel than when carboplatin is used alone. A similar effect could also be operational at the level of the tumor cell. Thus, a GOG study was critically important; it compared these two regimens in patients with optimally debulked disease; that is, short-infusion (3-hour) paclitaxel 175 mg/m2 and carboplatin (area under the curve [AUC] 7.5) with long-infusion (24-hour) paclitaxel 135 mg/m2 and cisplatin 75 mg/m2. Recent data from that study strongly indicate that the carboplatin and paclitaxel doublet will not be less effective than the cisplatin and paclitaxel doublet and will have a better toxicity profile.4 Although the median follow-up is only 3 years, there are no differences in progression-free survival in the two groups, and these data are unlikely to change significantly with further maturity. In fact, the hazard ratio for the trial is 1.12, favoring the carboplatin arm. This study addressed paclitaxel infusion duration and platinum analog preference. Neurotoxicity in both arms was similar (16% and 19%, respectively, with grade 1 or higher), whereas the carboplatin arm showed significantly more thrombocytopenia and pain, and the cisplatin arm showed more fever, nausea, vomiting, and metabolic toxicity. A second study performed by a German consortium (Arbeitsgemeinschaft Gynakologische Onkologie) asked only the question of which platinum analog was preferred (stages IIB to IV), with all patients receiving paclitaxel 185 mg/m2 for 3 hours and either cisplatin 75 mg/m2 or carboplatin (AUC 6).5 Again, no significant outcome differences were noted, although the hazard ratio of 0.86 favored the cisplatin arm compared with the carboplatin arm, which is the reverse of the results seen in GOG-158. Some investigators have suggested that differences in dose-intensity of carboplatin in the two trials may explain these opposite trends. However, grade 1 or greater neurotoxicity was significantly greater in the cisplatin arm (56% v 35%) and was slower to resolve, demonstrating once again that the doublet of short-infusion paclitaxel with cisplatin is excessively neurotoxic and cannot be recommended. Confusing the issue regarding the standard of care was another recently reported GOG study, which compared long-infusion paclitaxel with cisplatin to each agent given as monotherapy in higher dose: cisplatin at 100 mg/m2 or paclitaxel at 200 mg/m2 as a 24-hour infusion.6 Response was significantly inferior with single-agent paclitaxel, but median survivals of all three regimens were identical, probably because of significant crossover of patients undergoing monotherapy to the other single agent. One could more readily administer six cycles of the combination than of either single agent (early progression occurs with paclitaxel monotherapy and early peripheral neurotoxicity occurs with cisplatin monotherapy). Nevertheless, in the European Canadian trial, significant crossover did not blur survival differences as it did in this trial. The only apparent explanation is the time of application of salvage therapy, which was early in the GOG trial (often after a positive second look and before clinical progression) and later in the OV10 trial (at clinical progression). A final trial (International Collaborative Ovarian Neoplasm Group) performed under the sponsorship the British Medical Research Council compared a regimen without paclitaxel that administered either single-agent carboplatin (AUC 5) or a three-drug regimen (cyclophosphamide, doxorubicin, and cisplatin) with a regimen containing carboplatin (AUC 5) and paclitaxel 175 mg/m2 for 3 hours.7 This large trial (n = 2,074) has a short follow-up (median, 30 months), but to date there has not been a significant difference in outcome between the paclitaxel-treated and non–paclitaxel-treated groups (survival hazard ratio, 0.98). Although the authors state that no outcome differences could be seen in any subgroup of patients, there appears to be some effect of the paclitaxel arm on survival in patients with bulky disease after primary debulking surgery (no statistics were provided for this cohort). The authors opine in their discussion that the reason for the different outcome of this trial compared with the two trials that led to acceptance of the platinum-taxane doublet as standard (GOG-111 and European-Canadian trials) was the use of an inferior control arm in the two previous studies. They suggest that cyclophosphamide had an antagonistic effect on cisplatin, leading to poor outcome in the control group and making the experimental arm appear superior. Interestingly, no citations were provided for this opinion. This is rather a curious opinion because the prior study conducted by this same group8 had compared single-agent carboplatin with the three-drug combination of cyclophosphamide, doxorubicin, and cisplatin. In that large trial, no outcome differences were seen between the two arms. Furthermore, a large meta-analysis9 showed that platinum in conjunction with other agents (mostly alkylating agents) was superior to single-agent platinum. Finally, two trials conducted in Italy,10,11 which compared single-agent cisplatin with cisplatin combinations including cyclophosphamide, demonstrated no differences in outcomes. The reason for the similarity in outcomes between the paclitaxel-based and the non–paclitaxel-based arms remains to be clarified. The authors state that 33% of patients undergoing the non–paclitaxel-based therapy received paclitaxel at some point in their disease process (timing not stated) such that early crossover may have blunted survival differences as it did in GOG-132. Another logical hypothesis is that treatment with carboplatin, when combined with paclitaxel, requires a higher dose of carboplatin to cause a similar cell kill than when carboplatin is used alone. This well-known effect was seen in the initial phase 1 trials.3 Furthermore, the discordant results in terms of outcome of the two trials mentioned previously (GOG-158 and the Arbeitsgemeinschaft Gynakologische Onkologie trial) indicate that when carboplatin is combined with paclitaxel, the carboplatin dose may require escalation to preserve efficacy. It has not been proved that the doublet of a taxane and platinum will improve outcome in patients with optimally debulked disease. Nevertheless, when platinum was incorporated into primary therapy in the 1980s, the greatest effect was seen in this population, and it is anticipated that the same effect will be observed with substitution of a taxane. A subgroup analysis of the OV10 trial shows a similar outcome effect in patients with both optimally and suboptimally debulked disease. It is my opinion that these data taken together indicate that taxane and platinum remain the treatment of choice for suboptimally debulked ovarian cancer. It may be possible to obtain similar results by administering the agents sequentially rather than concomitantly as long as the sequential administration is before clinical progression. The issue of dose-intensity of carboplatin in the carboplatin and paclitaxel doublet may have some bearing on outcome. This point may become moot as the docetaxel and carboplatin doublet gains acceptance, because there are no data to suggest a drug-drug interaction.
DR. CANNISTRA: Although survival appears slightly improved for the carboplatin arm in the Ozols study [Proc Ann Meet Am Soc Clin Oncol 18:A1373, 1999], GOG-158, the difference in the survival curves does not appear to be statistically significant. In a related issue, it’s not clear that carboplatin at an AUC of 7.5 is necessarily better than using carboplatin at lower AUCs, in the range of 5 to 6 for instance. DR. RUSTIN: Which AUC are you talking about? DR. CANNISTRA: The AUC that was used in GOG-158 is based on a calculated GFR using the Jelliffe formula [J Clin Oncol 14:1895, 1996]. In some of the UK studies, there was a proviso for actually measuring the AUC using EDTA clearance, for instance. DR. RUSTIN: In one of the latest SCOTROC studies where they used AUC 7 [Proc Am Soc Clin Oncol 21:810, 2002], which is probably more equivalent to your AUC 8, 75% of the patients had delays in the first four cycles of single-agent carboplatin. So you’re getting close to the maximum that’s achievable with a single agent. DR. CANNISTRA: How was the carboplatin AUC of 5 calculated in the SCOTROC trial? DR. VASEY: We used EDTA clearance in the majority of patients. The original investigations that looked at whether we should use an EDTA or a calculated clearance as opposed to this measured clearance led us to believe that there was a significant percentage increase in the AUC. However, last year at ASCO, there was a presentation that looked at all the formulas and compared them with EDTA [Proc Am Soc Clin Oncol 21:210A, 2002]. There didn’t appear to be a statistically important difference in the AUC, whichever way you do it. So this question of an AUC of 5 or 6 may be irrelevant. DR. CANNISTRA: I agree with you within the range of 5 to 6. However, if we’re using higher AUCs in the 7.5 range, we are risking more hematologic toxicity and perhaps neurotoxicity without definite proof that it’s superior to lower AUCs. DR. MARKMAN: The difference between an AUC of 6 and 7.5 is incredibly small. To argue that the slightest difference in AUC can make a difference at the level of a tumor cell is inconceivable to me. DR. McGUIRE: But the grade 3/grade 4 platelet toxicity in GOG-158 was 40% [Proc Ann Meet Am Soc Clin Oncol 18:A1373, 1999] and in AGO, in the carbo-paclitaxel arm, it was 7% [Proc Am Soc Clin Oncol 18:A1374, 1999]. DR. MARKMAN: Yes, you are going to get much more toxicity. People talk about the advantage of carboplatin being a reduction in neurotoxicity, but in GOG-158 there was no advantage at 3 versus 24 hours [Proc Ann Meet Am Soc Clin Oncol 18:A1373, 1999]. DR. CANNISTRA: In the US, what are common treatment practices with respect to carboplatin AUC dosing when it’s combined with paclitaxel? DR. McGUIRE: I usually try to start patients at 7.5. DR. CANNISTRA: I give carboplatin at an AUC of 5 to 6 based on the patient’s general level of health. I would start with AUC 5, especially if the patient has co-morbid disease and I’m concerned about her ability to tolerate chemotherapy. I can always increase the dose as I need to, depending on hematologic tolerance. But I usually don’t increase beyond AUC 6, due to problems with bone marrow suppression, dose delays, and neuropathy. DR. BOOKMAN: Keep in mind that the dose of 7.5 was based on a phase 1 trial (GOG-9202) initiated in 1992 when many of us believed that dose intensity was important [J Clin Oncol 14:1895–1902, 1996]. Therefore, a traditional phase 1, dose-escalating, small cohort trial was used to establish 7.5 as a maximally tolerated dose that could be recommended for phase 3 studies. However, over one third of the patients in that small trial had to have modifications made due to delays, thrombocytopenia, neutropenia, or addition of G-CSF. If I were designing GOG-158 today, I probably would not recommend that dose.
1. McGuire WP, Hoskins WJ, Brady MF, et al: Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 334:1–6, 1996
2. Piccart MJ, Bertelsen K, James K, et al: Randomized intergroup trial of cisplatin-paclitaxel versus cisplatin-cyclophosphamide in women with advanced epithelial ovarian cancer: Three-year results. J Natl Cancer Inst 92:699–708, 2000
3. Bookman MA, McGuire WP, Kilpatrick D, et al: Carboplatin and paclitaxel in ovarian carcinoma: A phase I study of the Gynecologic Oncology Group. J Clin Oncol 14:1895–1902, 1996 4. Ozols RF, Bundy BN, Fowler J, et al: Randomized phase III study of cisplatin/paclitaxel versus carboplatin/paclitaxel in optimal stage III epithelial ovarian cancer: A Gynecologic Oncology Group trial. Proc Am Soc Clin Oncol 18, 1999 (abstr 1373) 5. du Bois A, Lueck HJ, Meier W, et al: Cisplatin/paclitaxel vs. carboplatin/paclitaxel in ovarian cancer: Update of an Arbeitsgemeinschaft Gynaekolgische Onkologie study group trial. Proc Am Soc Clin Oncol 18, 1999 (abstr 1374)
6. Muggia FM, Braly PS, Brady MF, et al: Phase III randomized study of cisplatin versus paclitaxel versus cisplatin and paclitaxel in patients with suboptimal stage III or IV ovarian cancer: A Gynecologic Oncology Group study. J Clin Oncol 18:106–115, 2000 7. The ICON Group: Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide, doxorubicin, and cisplatin in women with ovarian cancer: ICON3 randomised trial. Lancet 360:505–515, 2002[CrossRef][Medline] 8. The ICON Collaborators: ICON2: Randomised trial of single-agent carboplatin against three drug combination of CAP (cyclophosphamide, doxorubicin, and cisplatin) in women with ovarian cancer. Lancet 352:1571–1576, 1998[CrossRef][Medline] 9. Advanced Ovarian Cancer Trialists Group: Chemotherapy in advanced ovarian cancer: An overview of randomized clinical trials. Br Med J 303:884–893, 1991
10. Bolis G, Favalli G, Danese S, et al: Weekly cisplatin given for 2 months versus cisplatin plus cyclophosphamide given for 5 months after cytoreductive surgery for advanced ovarian cancer. J Clin Oncol 15:1938–1944, 1997 11. Gruppo Interegionale Cooperativo Oncologic Ginecologia: Randomised comparison of cisplatin with cyclophosphamide/cisplatin and with cyclophosphamide/doxorubicin/cisplatin in advanced ovarian cancer. Lancet 2:353–359, 1987[CrossRef][Medline] Submitted January 10, 2003; accepted March 14, 2003.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2003 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
