Panelists: Johanna C. Bendell, MD, Sarah Cannon Research Institute; Winson Y. Cheung, MD, MPH, University of Calgary and CancerControl Alberta; Manuel Hidalgo, MD, PhD, Harvard Medical School and Beth Israel Deaconess Medical Center; Ramesh K. Ramanathan, MD, Mayo Clinic; Tanios Bekaii-Saab, MD, FACP, Mayo Clinic; Thomas Seufferlein, MD, University of Ulm
Johanna C. Bendell, MD: Unfortunately, 80% of the patients who come to us with pancreas cancer come with metastatic disease. It used to be that we would quote patients as 4 to 6 months’ average survival once they were diagnosed with metastatic disease. But certainly, we now have new chemotherapy combinations and options that are extending that survival to maybe even 10 to 12 months (if we’re looking with our rose-colored glasses). Tony, can you tell us a little about some of these combinations and what do you do in this setting?
Tanios Bekaii-Saab, MD, FACP: It definitely has shifted (at least since 2010) to a little bit more positive. We’re not quite there yet, but I think as we understand a little more how to divvy up pancreas cancer into subgroups, we will continue advancing. Today, very simply, we have 2 standard regimens for patients who are relatively well performing. Those patients who have a performance status of 0 or 1 are those that seem to benefit the most from chemotherapy. Those with a performance status of 2—borderline. I know we’ll discuss this a little bit more. But clearly, for most of our patients who are eligible for treatment, there are 2 regimens. One is a 3-drug regimen. The other one is a 2-drug regimen.
The 3-drug regimen is FOLFIRINOX. Then, gemcitabine and nab-paclitaxel. And the question is, how do you pick between the 2? You have this patient coming to your clinic, and you have to make that decision. They’ve never really been compared head-to-head. We make all the assumptions that the 3-drug regimen is probably better because, essentially, the data look a little bit better, historically (although there’s a lot of biases that get introduced into these types of cross-study analyses). The reality is, in the real world, when we look at a lot of the databases in the more recent studies, they look kind of similar in terms of their outcomes. The key is to think, now, more about how we allow patients to get exposed, like we do, with colon cancer, lung cancer, and others (although we’re limited in the choices here). How can we expose patients or think about sequencing rather than to throw the whole gauntlet?
So, in my practice, for most patients, the first-line option is going to be gemcitabine and nab-paclitaxel—thinking about, potentially, a second-line option with 5-FU and nanoliposomal irinotecan. FOLFIRINOX, in all frankness, I’ve used very selectively, either on trials or for those patients where I suspect, perhaps, some element of BRCA or BRCA-ness. We know irinotecan and oxaliplatin both have significant activity. Taxanes or gemcitabine don’t. So, with that, which amounts to about 5% to 10% of patients, perhaps FOLFIRINOX is a better choice. But for the others, frankly, I don’t see much of a difference between the two. In fact, my preference has been to be, I would call it, a little bit more merciful on my patients. I essentially think sequentially rather than throw the whole kitchen sink on them.
Johanna C. Bendell, MD: I always find this fun to query medical oncologists, especially gastrointestinal oncology specialists, on which they believe in. I find that when you have your regimen, it tends to be a very passionate belief. So, I’m going to perform my experiment here. Thomas, FOLFIRINOX or gemcitabine and nab-paclitaxel?
Thomas Seufferlein, MD: Well, we would probably use a little more FOLFIRINOX. It’s about 30% FOLFIRINOX and 70% gemcitabine and nab-paclitaxel in those who can receive active chemotherapy first-line. We have good experience with FOLFIRINOX, and there are some patients who really tolerate it well. The problem is predictability. We had young patients who actually couldn’t stand it and had very serious side effects, and we had more elderly patients who had no problems at all. So, it’s very hard to predict who is going to do well. Whereas with gemcitabine and nab-paclitaxel, usually there is a much broader tolerance. We didn’t see these extremes that we’ve seen with FOLFIRINOX, but we have around about 30% to 70%—this is the ratio in which we use it.
Johanna C. Bendell, MD: Do you modify the FOLFIRINOX at all?
Thomas Seufferlein, MD: Well, we do in the way that we do not use the bolus and we de-escalate after, let’s say, 3 to 4 months of treatment. We take out the oxaliplatin, and that works well in our hands.
Johanna C. Bendell, MD: Very good. Do you use growth factor support?
Thomas Seufferlein, MD: Well, if required, yes. As was revealed in a clinical trial, probably 30% of patients will definitely need it. We don’t do it automatically, but we are, particularly in the elderly population, very cautious. And in that case, we also would use it in the preventive phase.
Johanna C. Bendell, MD: Good. I don’t feel so wimpy anymore. Very good. Ramesh, what’s your choice?
Ramesh K. Ramanathan, MD: I think it’s similar. Obviously, in the Mayo Clinic system, there are differences, even there. There’s Arizona, Rochester, and Florida. We have some regional and intrastate differences. I tend to use about 20% to 30% of modified FOLFIRINOX without the bolus, gemcitabine and Abraxane in 60%, and maybe 10%, 15% of single-agent gemcitabine. The FOLFIRINOX is obviously modified—no bolus. A lot of times, I will dose-reduce the irinotecan and oxaliplatin by 20%.
Johanna C. Bendell, MD: Winson, the Canadians?
Winson Y. Cheung, MD, MPH: Well, it’s an interesting question. In Canada, we have very strong population-based data. So, 2, 3 years ago, when gemcitabine and nab-paclitaxel first became available, we looked at our experience and it was a 50/50 split (between FOLFIRINOX and gemcitabine/nab-paclitaxel). However, a very small minority was using gemcitabine as monotherapy. And recently, we looked at the same data, or new data, and the tide is really shifting. The breakdown, now, is closer to 60% for gemcitabine and nab-paclitaxel and 40% for FOLFIRINOX. But to others’ points, for FOLFIRINOX, there’s a huge tendency to also dose reduce and modify. So, it calls into question whether we’re still really giving the same benefit when we’re really modifying to the nth degree.
Johanna C. Bendell, MD: That’s a great question. Actually, I’m going to throw the vote now to Manuel, but I’m going to ask you a question about dose modification because you were so involved in the original studies with gemcitabine and nab-paclitaxel.
Manuel Hidalgo, MD, PhD: I think in the United States, the trend is GA, or gemcitabine/Abraxane. GA is probably 60%, 70% of utilization now. If you ask me what happens at the Harvard Cancer Center, I’ll say that most people use FOLFIRINOX (for the patient who is fit) with the different modifications. We have discussed the key ones. I personally use more gemcitabine/Abraxane, and what I find very interesting is if you look at the last studies, mostly single-center experiences (retrospective studies, but also the control arms of randomized clinical trials in which patients were highly selected), GA is giving you the same 11 or 12 months’ median survival as FOLFIRINOX. So, I personally think that the IMPaCT trial was loaded, perhaps, with patients who had a lower performance status.
Tanios Bekaii-Saab, MD, FACP: Or no second-line options, too. And to that point (and I would like to second that), there was a recent publication from the United States Oncology database that looked at that experience with both gemcitabine/Abraxane and FOLFIRINOX in terms of utilization in terms of outcomes. And, intriguingly, they found the same trends. Regardless, patients do about the same in terms of their survival in outcomes whether you choose FOLFIRINOX or gemcitabine/Abraxane. So, I concur 100%. I don’t really think that there is much of a difference as long as you have access to salvage, second-line therapies. The sequential approach is what we do in all other cancers (except in this one, where we tend to believe that maybe we shouldn’t).