John Marshall, MD: I went through a phase where, yes, surgery was exciting. Then I went through a fairly large part of my career where I thought, “What’s the benefit of surgery at all to even resectable patients?” I have to say, now that surgery has improved and the recoveries have improved, I see the value of that even in patients who relapsed later. There is a survival advantage to surgery. I’ve come around, myself, on the role of that.

Paul Oberstein, MD: I don’t know if this is going to hold true, but certainly the data presented today showed that patients who had surgery and went on the APACT trial—the adjuvant pancreatic cancer trial—in both arms did much better than patients have done historically. The optimistic view of that is that we’re doing something better. It’s not a fluke. Either the surgeons are better, the medical oncologists are selecting patients better, or we’re scanning better. I think it gives more confidence—to me at least, as an optimist—that it’s real and that we’re doing a better job at removing tumors from patients.

John Marshall, MD: Allyson, let me tag you on that—you jumped to that trial. Let’s talk about that study a little bit. I personally was disappointed in the results of that study, but review it for us, and discuss what it showed.

Allyson Ocean, MD: This was a large phase III study in patients who underwent resection, randomizing them to GEM [gemcitabine]/nab-paclitaxel versus GEM alone. The buzz going into this study, and the hope, was that it was going to be a positive study and provide another standard of care in the adjuvant setting. However, unfortunately, it was a negative study. What Paul just mentioned, which was the good that came out of it, is that the survival times for both arms were higher than what we’ve seen all along.

John Marshall, MD: What was the number. I forget it. Anybody remember it?

Shubham Pant, MD: Yes.

Edward Kim, MD: It was around 18 months for time to progression, or progression-free survival.

John Marshall, MD: Both arms? Wasn’t it better to add the nab-paclitaxel?

Allyson Ocean, MD: It wasn’t. You may ask, “What do you do for a patient who can’t get FOLFIRINOX [folinic acid, fluorouracil, irinotecan hydrochloride, oxaliplatin] in the adjuvant setting?” I have been giving GEM and CAPE [capecitabine], and that’s my standard. I was wondering if this was going to change it, but it didn’t.

John Marshall, MD: You guys are too young to remember that 5-FU [5-fluorouracil] is still the best drug, and it’s really the only 1 that’s consistently been shown effective in the adjuvant setting across a lot of our GI [gastrointestinal] cancers. This was non-fluoropyrimidine-based trial. I think GEM-CAPE [gemcitabine-capecitabine] is still quite a legitimate choice.

Edward Kim, MD: I think the primary endpoint was an independent review of disease-free survival. There’s a big discrepancy between what was found with independent review—a central review, looking just at imaging—and what was noted as recorded by investigators. Now, the investigator reported disease-free survival was much shorter, and the thought was you didn’t want to have the bias of the investigator. Potentially, things were more favorable. It was actually the opposite, and some of the thought is that if you have a central review, in the vacuum of just the image, are you able to truly capture these questionable cases of potential progression?

The clinician—the investigator who has CA19-9 [carbohydrate antigen 19-9], patient symptoms, and other factors to incorporate to say that the patients progress—clearly felt these patients had progressed sooner. That number was actually in line with historical controls of around 13 months. I think there’s still a lot out there to try to figure out exactly where this trial fits. They mentioned that there was a median overall survival. It wasn’t a primary endpoint.

What I think you were mentioning, Paul, about the outcomes that we’re seeing with adjuvant studies—in both the Purdue University study and the APACT study—the median overall survival for the GEM control arm was 35 months, which is over a year better than what we experienced with…

So is the question is this: “What is that due to? Is it the surgery?” I would argue it’s because of the subsequent therapies that we have as options.

John Marshall, MD: Also, I think it’s patient identification.

Shubham Pant, MD: Exactly. I agree with John. I think the big 1—in the APACT trial, they took patients who had a CA19-9 less than 100. Now we do CA19-9 post surgery. People have less than 100, but we have a number of patients who are more than 100. You’re selecting. There’s a selection bias happening there toward a population. You could say median overall survival wasn’t the primary endpoint, but the median overall survival in the G/A [gemcitabine-abraxane] group was 36 months; in the non-G/A group, it was about 30 months. That’s what we’re looking at. We’re looking at how to increase the survival of the patients. I think we are selecting better. I do think we’ve made advances; there’s no question about it. To make further advances, we’ve got be self-critical and look at data agnostically. We probably had a better patient selection in this trial with a CA19-9 less than 100 than we do when we have a 73-year-old with a CA19-9 at 300 post surgery, right?

Transcript Edited for Clarity