Now, what do you put? Cisplatin or carboplatin? I hope to goodness you are not going to be asked to distinguish between those two. But there have been three studies of this issue now, and we actually have an answer. There was a Dutch trial that only had 182 patients on it, too few patients to allow any reasonable conclusions to be made. But there have been two subsequent studies, GOG protocol 158 and a German trial from the AGO group, both of which had almost 800 patients – 798 patients each. That’s enough patients to have equivalency conclusions to be drawn. Let’s look at those two. First the AGO study, presented at each of the last two ASCO’s. These studies randomized patients with IIb-IV disease – that is the both the large volume and small volume patients – to either Taxol over three hours plus cisplatin or Taxol over three hours plus carboplatin. And yes, they got a lot of neurotoxicity with three hour Taxol/cisplatin. They allowed up to six cycles of therapy. What they found was no difference. The trends all favored Taxol/cisplatin. The response rate 80% versus 70%. Clinical complete response rate 36% versus 30%, progression-free survival median 73 versus 69 weeks. Survival at one year 92% versus 91%. The trends here are very small, with cisplatin no significant difference.

GOG158 was done in a patient population with small volume residual stage III disease. Again, the randomization here was the same but the Taxol/cisplatin regimen used 24 hour Taxol plus cisplatin, which gives much less neurotoxicity. The Taxol/carboplatin regimen used an AUC of 7.5 for the carboplatin because that was the MTD on the phase I study. Each regimen given every three weeks for six cycles. Here are the results. Pathologic complete response rate, 45% with Taxol/cisplatin, 52% with Taxol/carboplatin. Progression-free survival essentially identical between the two arms. If anything here, the trends favor Taxol/carboplatin. And what that speaks to is there is literally no difference between a combination of Taxol/carboplatin and a combination of Taxol/cisplatin. The disappointing thing is that there is also no difference in neurotoxicity between the two. In the GOG study the neurotoxicity in these two arms was exactly the same. In fact there was slightly more nausea and vomiting on the Taxol/cisplatin arm. There was more nephrotoxicity on the Taxol/cisplatin arm. There was more myelosuppression on the Taxol/carboplatin arm. So that while Taxol/carboplatin did have an advantage in terms of therapeutic index, it was not as great as we would have anticipated going into the study.

The weight of evidence suggests that three hour Taxol/carboplatin is a reasonable treatment for ovarian carcinoma. In fact I can tell you that it is used by over 80% of you in treating ovarian carcinoma, based on a survey that was done by Bristol-Meyers Squibb. Now how much of these drugs should we use? Cisplatin, carboplatin and Taxol which are the major three drugs in ovarian carcinoma. Let’s look at each of the three. First, cisplatin. There have been two metaanalyses published on cisplatin; 2006 in the Journal of Clinical Oncology, 1993 in the Journal of the National Cancer Institute. What these two metaanalyses do, combined, is to establish a clear dose-response relationship up to a cisplatin dose of 60 mg per meter square but not beyond that. No advantage to increasing the cisplatin dose beyond 60 per meter square. There have been three randomized trials internationally now looking at the issue of 50 mg per meter square versus 100 mg per meter square of cisplatin. One GOG study and two Italian studies. No differences in response rate, no differences in survival by going up to the higher dose of cisplatin. So by the best evidence that we have to date there appears to be no advantage to going beyond a dose of cisplatin in the range of 50-60 mg per meter square every three weeks.