Case 3: The OCEANS and GOG-0213 Trials for Ovarian Cancer

Chad Hamilton, MD: The OCEANS study was a phase 3 randomized trial that assessed the combination of carboplatin/gemcitabine with or without bevacizumab in patients with platinum-sensitive recurrence who had not previously received bevacizumab. This was one of the early studies that helped inform this treatment paradigm. In OCEANS, there was a progression-free survival benefit with the addition of bevacizumab, to 12.4 months versus 8.4 months in the placebo arm of the trial. The final survival analysis didn’t show an increase in overall survival, which is a difficult end point to attain, and this got a category 2A recommendation ultimately in the latest guideline from the NCCN [National Comprehensive Cancer Network].

We’ve touched on this trial already. This is [Robert] Coleman, [MD’s,] herculean effort of GOG-0213, in which there were 2 randomizations: 1 to a surgery-versus-no-surgery arm in patients for whom it felt like they could be completely resected at the time of surgery, but we’ll focus on the other arm of the trial, which was carboplatin/Taxol [paclitaxel] alone versus carboplatin/Taxol [paclitaxel] with bevacizumab and then a bevacizumab maintenance arm as well.

This is the progression-free survival in the GOG-0213 trial. There was about a 4-month progression-free survival. This was not the primary end point of this trial; this was a secondary end point. The primary end point of GOG-0213 was overall survival, as I mentioned, a tough bar to meet. In the overall survival analysis, the combination yielded a 42.2-month versus 37.3-month benefit, with a hazard ratio that just crossed 1. It’s been characterized as a slightly increased survival versus a clinically meaningful survival, depending on your perspective.

Now, they went back at this and they did an audit in which there was inaccurate or incorrect treatment-free interval stratification in this trial. After they went back and did a sensitivity analysis, it did tip the overall survival into statistical significance. This is one of the rare trials that have given us a hint of an overall survival advantage, and that’s certainly something that nudges a lot of us into strongly considering this regimen.

That brings up a couple of questions. We’ve already touched on the first: is there an option for sequencing PARP inhibitor maintenance, so PARP-after-PARP? Shannon did a good job of addressing that already. What if this patient were BRCA wild type but HR [homologous recombination]-deficient? Would that change your treatment thoughts on this patient at all?

Shannon Westin, MD: In the secondary maintenance setting, the presence of any biomarker indicates that you want to consider PARP. We don’t know what the benefit will be in this patient. I liked the use of bevacizumab for this patient not only from a survival standpoint; when you add the bevacizumab, you get an increase in response, and a response is how you can get to maintenance. We want to get to maintenance, so I like the addition of bevacizumab for that patient. You then get into the question of switch maintenance. For this patient, who had a progression on olaparib, would it make sense to continue the bevacizumab alone versus switching to a PARP inhibitor? That would be a tough decision. Tom has already referenced some of the combinations of antiangiogenics and PARP that we’re still waiting to see as far as the recurrent setting if those get any kind of approvals: the olaparib/cediranib combination, AVANOVA trial, the bevacizumab and niraparib, the OVARIO. There are a lot of different combinations that are being looked at, but it would be a bit of a push out of the standard of care at this point to do a dual maintenance.

Thomas C. Krivak, MD: I agree. This patient was treated very appropriately with bevacizumab, and a clinical trial would be important. As we write these clinical trials for PARP-after-PARP, it’s going to be important to get these patients in. It seems like, with the data that Shannon presented earlier, a high percentage of our patients are going to be treated with PARP up front. This is going to be a clinical situation that we’re going to be faced with.

I don’t think it’s wrong to consider the retreatment, and then if they have a partial or complete response—this patient was treated with olaparib—to potentially try niraparib in maintenance or even in combination based off the AVANOVA. I don’t think that’s inappropriate whatsoever.

Chad Hamilton, MD: Shannon and Tom, I’m getting the impression that any time you ask a fellow or a trainee what to do, the first answer should be a clinical trial, right?

Shannon Westin, MD: That’s right. They can walk away with that; they’ll always get it right.

Transcript edited for clarity.