John L. Marshall, MD: We’re going to shift gears to where we’ve done science, and we’ve done some big randomized trials. The adjuvant data that we’re going to review—we were debating before—should have been a plenary paper. It’s really practice changing—big deal, big numbers out there, a lot to talk about in this space. We actually drew straws of who got to review it, and Toni, you were the short straw on that—or long straw. You won, so take it away. Tell us about the new data here at ASCO in the adjuvant setting.

Tanios S. Bekaii-Saab, MD: This is an interesting study, and we’ve been waiting for the results of this study for quite a while. I agree, this should have been a plenary session. It’s almost 500 patients. That’s a large adjuvant study in pancreas cancer, and this study was simple in its design. Patients would get randomized to FOLFIRINOX or to gemcitabine.

John L. Marshall, MD: Modified FOLFIRINOX.

Tanios S. Bekaii-Saab, MD: Modified FOLFIRINOX, as it should be in those patients. The primary endpoint was disease-free survival, and the secondary endpoint was overall survival, and there were other endpoints, as well. The disease-free survival was a little bit less than double but quite significantly improved—12.8 versus 21.6 months favoring FOLFIRINOX, with a hazard ratio of 0.59, so quite significant improvement. I mean, you don’t see that, and disease-free survival is primarily a measure of the success of the regimen. Median overall survival was also improved. It was about 34-plus months with gemcitabine, and it was 54-plus months with FOLFIRINOX—so a very positive study.

John L. Marshall, MD: So, both arms did pretty well, and the FOLFIRINOX arm does surprisingly well.

Tanios S. Bekaii-Saab, MD: Surprisingly well. You can blame some on good biology and selection, but there’s definitely a drug effect. When you look at the disease-free survival from the gemcitabine, it’s not much different from the CONCUR study; maybe 1 month more than that, at least histological. So, I think that when you look at this study, there are 2 effects. The biology effect—there’s definitely the FOLFIRINOX effect; of course, a little bit more toxic. We know these patients that go through surgical resections have a tough time to get even single-agent gemcitabine sometimes. This study just tells us 1 thing: that intensifying treatment is feasible in the adjuvant setting. But the way I look at this—again, to go back into the debate of pre versus post—it is most feasible when we actually give it ahead of surgery, and it probably makes much more sense now that we have all these data that are trickling in that intensification makes sense. So, limited intensification before surgery is more feasible and likely more yielding for a larger proportion of patients.

John L. Marshall, MD: Eileen, get in the weeds on this a little bit. I think about trying to give postoperative adjuvant FOLFIRINOX, even modified, to these patients would be challenging. Can you talk a little bit about who these people were and some characteristics of those folks?

Eileen M. O’Reilly, MD: I would echo Toni’s point. I think this is a very positive practice-changing study for us tomorrow in the clinic, for the next group of people that we see with adjuvant therapy. But who does it apply to? And that’s also going to be critical; it’s not for everybody. It was pretty clear this was a relatively highly selected patient population who were well enough to go to surgery, well enough to recover from surgery, well enough to get on to adjuvant therapy for 12 weeks. They also had to have a performance status of 0 to 1 and had a limitation on their carbohydrate antigen 19-9. So, we’re selecting out a lot of patients.

John L. Marshall, MD: And it’s a postoperative CA 19-9.

Eileen M. O’Reilly, MD: Postop CA 19-9, prestudy and preenrollment, and that was 180. Patients were stratified whether they were less than 90, or 90 to 180. So, it’s clear that this is selected, and it probably isn’t going to apply to our 85-year-old who has comorbidities and who has struggled through the postoperative course. There, I think either gemcitabine/capecitabine or single-agent gemcitabine will remain a standard. Here’s the good news: I think that we now have choices for people. The other big point to acknowledge, and Toni also hinted at this, is that overall survival is compelling. That’s almost certainly contributed to by multiagent cytotoxics that we have available now for treatment in the advanced disease setting, that we didn’t have in the era of other adjuvant studies.

John L. Marshall, MD: Kabir, go through when you’re giving FOLFIRINOX, for the folks out there that maybe don’t do it day in and day out. Modified is, what, 1 drop in the bolus? 150 mg/m²of irinotecan was the regimen they used here. Coming down to 85 mg/m² of oxaliplatin, the 2-day infusion every 2 weeks—maybe need growth factors, maybe not. Talk a little bit about managing that patient.

Kabir Mody, MD: So, I think it speaks to what a lot of us do in normal practice. With the advent of FOLFIRINOX in the metastatic disease setting, a lot of us realized we had to make dose modifications for that to be tolerable for a duration of time, for the patient to get benefit and avoid toxicity. So, as you said, dropping the 5-FU bolus really can contribute to helping with the myelosuppressive effect of the regimen. Dropping the irinotecan can help with the symptoms of diarrhea, which can be debilitating in some patients. And then oxaliplatin—obviously, the elephant in the room there is neuropathy. That can be sometimes delayed, and it can be really a problem for patients.

John L. Marshall, MD: I don’t think I’ve ever given 12 cycles of FOLFIRINOX, ever. I’ve never given much more than about 8 in most people. So, should we be doing a 3 versus 6 months of FOLFIRINOX in the adjuvant setting?

Kabir Mody, MD: Well, there are some studies that we’ve alluded to that have used different durations of chemotherapy—4 months or 6 months—so I think that’s quite a valid question to ask.

John L. Marshall, MD: Eileen and I maybe disagree on this, so I’ll ask George instead, because I don’t know what you’re going to say. I’m sure that with FOLFIRINOX, that we need all 3 bits of that. I keep wondering, is this just really well-given 5-FU? What is the relative contribution of the oxaliplatin and the irinotecan to this? Because we haven’t methodically studied this accumulated regimen. We just threw it at folks and compared it with gemcitabine. I worry that we are forever going to have FOLFIRINOX as this regimen and never really study the bits and components of it. What do you think?

George P. Kim, MD: Yes, so I’m pretty set on this. You need all the drugs.

John L. Marshall, MD: Why do you say that?

George P. Kim, MD: Because there’s a famous story about what was important in the regimen, and the lead PI said he had no idea, so he didn’t test it.

John L. Marshall, MD: But ProMACE-CytaBOM, we don’t give anymore, right?

Eileen M. O’Reilly, MD: Let me give you some data here just to support George’s point. So, FOLFOX—we’ve looked at it in untreated advanced pancreas cancer. The survivals are nowhere near what we’re seeing with FOLFIRINOX. The best we see is 7 to 8 months, so that, in my opinion, is imperfect. They’re not randomized data.

John L. Marshall, MD: But we have some positive and some negative data with oxaliplatin, right? I’m wondering if that’s the drug. I’ve got more consistent data with irinotecan in this space than I have with oxaliplatin, so that's my only argument.

Eileen M. O’Reilly, MD: But there are similar data with FOLFIRI and versions of FOLFIRI. The Europeans have studied this quite extensively; 7 to 8 months untreated metastatic disease, FOLFIRINOX for 11 months plus. Now we have data from a SWOG/Cooperative Group study. We’ll be talking about this when we come back to stromal modulation. The outcome there was 14 months. In a Cooperative Group setting, in North America, that to me is pretty convincing—that you need the combination of drugs for the optimal outcome for at least a select group of patients.

John L. Marshall, MD: But we deconstructed it in the metastatic setting.

Eileen M. O’Reilly, MD: You’re taking out the thrombosis patients.

George P. Kim, MD: Getting back to your question, we know that oxaliplatin didn’t work by itself—you had to give 5-FU. We know about IROX—the famous IROX regimen interactions between irinotecan and oxaliplatin. Bolus, continuous infusion, yes. So, I think you have to give all components, and if anything, we may try to dial it up. We may try to add maybe a biologic to that regimen. Unfortunately, the PEGPH20—that didn’t work, but you do need to continue to work on building regimens.

Transcript Edited for Clarity