Progress and Promise in Advanced Pancreatic Cancer - Episode 9

Advanced Pancreatic Cancer: Options in the Adjuvant Setting

Transcript:George P. Kim, MD: You mentioned Matt Katz from MD Anderson, part of the Alliance Cooperative Group trial that was presented at ASCO. They took 22 patients; they gave them FOLFIRINOX, some got some radiation, and about 14 patients went to surgery. Most of those patients had portal vein involvement, not retraction of the tumor from arteries. When we’re talking about borderline, you’ve got to distinguish the portal vein and the arteries, and that’s really where the benefit was seen. Now, getting back to response rate, response rate with nab-paclitaxel and gemcitabine is also highest, in the 29%. You’ve got to remember the methodologies between the ACCORD trial and the MPACT trial were different. It’s reasonable to use nab-paclitaxel/gemcitabine in that borderline population. But most people are using FOLFIRINOX. The other point is...

Philip A. Philip, MD, PhD, FRCP: George, can I interrupt just for that? Surgeons don’t have biases usually, but they have this bias about FOLFIRINOX. When they send a patient to you, they say use FOLFIRINOX. I don’t know how they got to that conclusion.

George P. Kim, MD: Because they talk about it at their meetings.

Philip A. Philip, MD, PhD, FRCP: At their meetings, thank you. A lot of times the oncologists are pushed to use FOLFIRINOX because the surgeon says to. I had a couple of patients who came from Johns Hopkins because they are mentioning to patients to have treatment with me, and the surgeon says, “Make sure Dr. Philip gives you FOLFIRINOX.”

George P. Kim, MD: Yeah, but if you get a patient sick—you give them FOLFIRINOX and they don’t do well—they are no longer a surgical candidate.

Johanna Bendell, MD: Blame the surgeon.

George P. Kim, MD: The surgeons aren’t going to be happy with that either. So, you’ve got to take it both ways.

Philip A. Philip, MD, PhD, FRCP: I’m not defending the FOLFIRINOX. I’m just saying that’s a factor.

George P. Kim, MD: I agree with you, Philip, that post-op treatment is still the standard. The trials that are out there are nab-paclitaxel/gemcitabine. APAC and the French are using FOLFIRINOX in the adjuvant setting. They’re having a lot of trouble getting FOLFIRINOX. They had to get rid of the bolus. They had to bring the irinotecan down by 20%. So it’s going to be interesting to see which of those two regimens is most appropriate in the post-resection setting. But the problem...

Philip A. Philip, MD, PhD, FRCP: Can I interrupt you for a second? The problem of giving treatments that are toxic, more-than-average toxic, is that they compromise the delivery of the drugs and the patients end up getting less of systemic treatment.

George P. Kim, MD: Exactly.

Philip A. Philip, MD, PhD, FRCP: So that’s my concern. Can I just continue for a second? I saw a patient, an engineer in our area, who came for a second opinion who was in his mid-50s and he had T3 N0, and he went to his oncologist and the oncologist in the community gave him an option of either gemcitabine or FOLFIRINOX—not gemcitabine/nab-paclitaxel—one or the other. It’s possible that nab-paclitaxel in the adjuvant setting would be an insurance issue. Maybe that’s why it was deleted. That might be the reason, I’m not sure. And he came to me for the second opinion. I said, “Gemcitabine, that’s the standard of care,” and then he sent me an e-mail and said, “I’m very much into science, statistics. My statistics are better with FOLFIRINOX. I’m just going to go back to FOLFIRINOX with that doctor.” Now, where he got this evidence, I don’t know, but he was extrapolating from the metastatic disease. He probably read the original paper. But, one thing advised to people who want to give FOLFIRINOX in the adjuvant setting out of protocol is that we tried it, FOLFIRI in colon cancer, it didn’t work.

Thomas A. Abrams MD: Oh, I completely agree with you on that.

Philip A. Philip, MD, PhD, FRCP: Combination with bevacizumab in colon cancer, EGFR blocker, ECF in gastric cancer.

Thomas A. Abrams MD: It is inappropriate to extrapolate data from the metastatic setting to the adjuvant setting; we know that. It doesn’t work. As appealing, intellectually, as that would be, it doesn’t translate into good practice and we know that from many, many trials. The neoadjuvant setting is qualitatively different from the adjuvant setting for the reasons that you enumerated. Patients are really beaten up after a Whipple. They are not able to get the same degree of treatment that they can pre-operatively and that’s why pre-operative treatment is very appealing. And I’m glad that we are devoting more of our resources to studying the neoadjuvant setting.

Johanna Bendell, MD: So, outside of the trial, both ya’ll are giving gemcitabine adjuvantly as a single agent?

Thomas A. Abrams MD: Yes.

Philip A. Philip, MD, PhD, FRCP: Adjuvant?

Johanna Bendell, MD: Yes.

Philip A. Philip, MD, PhD, FRCP: Yes, I’m giving gemcitabine. One has to keep in mind something, which Caio brought up: if you give capecitabine, it’s also correct because the studies haven’t shown really that gemcitabine is better than fluoropyrimidines.

Caio Rocha Lima, MD: That’s right.

Philip A. Philip, MD, PhD, FRCP: Ironically, it is easier to use. But, certainly, gemcitabine has become our standard and we use it for clinical trials, control arms, etc. But the reality is that fluoropyrimidines are as good as gemcitabine.

Johanna Bendell, MD: George, I know something’s coming…

George P. Kim, MD: When I was at Mayo, we gave FOLFIRINOX, we gave gemcitabine/Abraxane in the adjuvant setting, so we’ve done it before. It’s like the patient you described who went through the data, and he was very good. He went through the data and he said, “This is what I want.” And it gets back to the earlier discussion, what does the patient want? We think we know. These drugs are going to fail in the adjuvant setting. But then why are we doing the trials if we think they’re going to fail in the adjuvant setting? Because we don’t know. We’re waiting for the trials to read out. The Abraxane trial, the FOLFIRINOX trial, and these may be advances. So to dismiss them, I don’t know.

Then why are we bothering with all these big trials? We’re doing them because we want to advance the science, we want to advance the field. And let’s face it, gemcitabine is not the greatest drug, and I’m not going to give them gemcitabine/erlotinib. So what are you going to do? You know they’re going to come back. They have lymph node—positive disease, they have some tumor that makes you nervous. What are you going to do? You’re really going to just stick to gemcitabine right now? The FDA approved these trials. These trials are ongoing. They’re already accrued, and the RTOG is doing the adjuvant trial and they’re bringing in other chemotherapies like gemcitabine/Abraxane. So there’s some precedent to do that, but also I agree with you: you’ve got to be careful and you want to make sure the patient is a partner in making that decision.

Philip A. Philip, MD, PhD, FRCP: But, this is an interesting statement. I mean, if a patient comes to me and says, “I want FOLFIRINOX,” I have to say I don’t have data. And, in this case, his wife said, “Well, if you don’t have data, it doesn’t mean it does not work.” So are we going to start thinking that way? Because you know, we don’t do that.

George P. Kim, MD: What are you going to tell the RTOG when they want to use gemcitabine/Abraxane?

Philip A. Philip, MD, PhD, FRCP: But they’re doing it as part of the clinical trial.

Thomas A. Abrams MD: Right. That is the huge distinction. To bring a clinical trial experimental arm into your practice off of protocol, I wouldn’t do it. You run the risk of harming the patient, and to me, as much as we think that gemcitabine is not a great adjuvant therapy, gemcitabine/Abraxane could be worse. FOLFIRINOX could be worse and then you would really be doing the patient a disservice.

Transcript Edited for Clarity

Johanna Bendell, MD: Incredible controversy in a setting with limited treatment options, very difficult disease. And the answer is going to be in the clinical trials.