Johanna C. Bendell, MD: We talk about toxicities, and we think about gemcitabine and nab-paclitaxel. We do see toxicities with gemcitabine and nab-paclitaxel (such as cytopenia), which tend to alter our dosing. And there have been some papers that have looked at whether to dose reduce (in these patients) or go to an every-other-week treatment regimen. What’s your take on that?

Manuel Hidalgo, MD, PhD: You can do either. There is probably more data with a dose-reduced regimen than with an every-other-week scheduled regimen. In my own experience, the every-other-week schedule works very well and is well tolerated. I tend to avoid day 8 and treat on day 1 and day 15 (in patients who do not tolerate the standard schedule), rather than dose reduce them to 100 mg of nab-paclitaxel.

Johanna C. Bendell, MD: Yes, that’s what I do as well. Do you all do the same thing?

Ramesh K. Ramanathan, MD: If the treatment causes neutropenia and thrombocytopenia, I tend to skip treatment.

Tanios Bekaii-Saab, MD, FACP: We have published data on the every-other-week regimen—retrospective, in our experience. Again, the survival was close to 10.5 months. We know that gemcitabine is a tricky drug. It has really been designed as a weekly drug, but it ends up being used, mostly, in an every-other-week or 2-weeks on, 1-week off drug regimen. It’s very tough to keep going with the weekly gemcitabine regimen. So, I concur. I think I’d rather keep the dose intensity but shift to the every-other-week regimen. We have biweekly regimens in colorectal cancer, and we know that they work better than the weekly regimens. We have proof of concept to essentially cut down on the dose intensity. Patients have a hard time with weekly regimens.

Manuel Hidalgo, MD, PhD: It’s easier. They don’t need to come in on day 8. And actually, when you get the dose right, you don’t need to check counts on day 8. It’s very convenient.

Johanna C. Bendell, MD: Yes, it’s nice for the patients.

Tanios Bekaii-Saab, MD, FACP: Yes, I agree.

Johanna C. Bendell, MD: Speaking of trying to be nice to patients or bring them back to the clinic more often, we have these chemotherapy options available to us, and we’ve been trying to see, “Can we give the chemotherapy in a smarter way?” We’ve seen some recent data on concurrent versus sequential gemcitabine and nab-paclitaxel administration. Ramesh, can you tell us a little bit about that?

Ramesh K. Ramanathan, MD: This goes back 20, 30 years. When I was a Fellow, we used to give paclitaxel before carboplatin on day 1 and day 2. This is nothing new. There are pharmacodynamic and pharmacokinetic data.

There is pharmacologic validity to this. In cells lines, this does work better. This also goes back, maybe, a decade, to the fixed-dose gemcitabine regimen. Gemcitabine is a prodrug. It has to be metabolized by cytidine deaminase to the active component, and that’s a rate-limiting example. So, when you give a bolus of 1500 mg of gemcitabine over 30 minutes, in theory, some of it is wasted. If you give it slower, it is a more active drug. The fixed-dose rate of gemcitabine administration does increase the response rate, as well as, maybe, progression-free survival, but is more toxic, and randomized studies did not prove for it to be better.

We do have an interesting paper (I think it was presented at the ASCO Annual Meeting) on giving nab-paclitaxel on day 1 and gemcitabine on day 2 because nab-paclitaxel (in preclinical models) inhibits cytidine deaminase. Again, this is a randomized, small phase II study that showed a little bit of improvement in response rate. It did seem that there was a little more toxicity, but in practice, it's recommended to give nab-paclitaxel just before gemcitabine, in any case. I think it’s an interesting study, but I don’t think it’s practical to bring patients in for 2 days every week.

Tanios Bekaii-Saab, MD, FACP: For a very little difference, if any.

Johanna C. Bendell, MD: It looks promising, though—the feasibility of having a patient in for 2 days in a row. We consider the cost and the time to the patients, and this is probably something that we’ll keep an eye on, but it is not quite prime time yet. Manuel, you’ve done some beautiful work. These patients with pancreas cancer can often come in with a poor performance status. What do you do when you see those patients, and how do you treat them?

Manuel Hidalgo, MD, PhD: We did a study that actually is being presented here at The European Society for Medical Oncology (ESMO) 2017 Congress. In the study, we first tried to evaluate different dose regimens—every other week or dose reduce on days 1, 8, and 15 in patients who were considered pure ECOG 2. And then, that was followed by a randomized phase II trial of 2 of these schemes. Here, we treated about 210 patients. The results are very interesting. They are basically similar (a little worse but not much) to the IMPaCT trial. There is a 5- to 6-month progression-free survival rate, and around a 7- to 8-month median survival rate in that population. I think the number of patients with a compromised performance status is probably larger than what we think, and that goes together with the better data in the studies that have been done, now, where patients are 0 and 1 when selected to be part of the control arm of a randomized clinical trial.

So, based on that data, we tend to treat patients with ECOG 2 with the combination, more and more, and we rarely use gemcitabine alone. Gemcitabine alone would be for patients who really cannot tolerate the combination. And these patients, indeed, are probably better candidates for supportive care before anything else.

Johanna C. Bendell, MD: That’s very interesting. So many times, patients come in to see you and they have this poor performance status. You get scared to administer combination therapy. So, you default to gemcitabine alone. But you also know that gemcitabine alone is not very helpful (as compared to the combination regimen).

Manuel Hidalgo, MD, PhD: We’ve taken the other approach, which is to support the patient in the first few days by controlling pain and nutrition. We get everything that we can fixed, and then we dose them at 100 mg and 1000 mg and follow them very closely. Some patients improve, and some others do not. If they do not, then they go on to palliative care. But if they improve, in some of these patients, you can rescue. So, the approach has changed a little bit.

Thomas Seufferlein, MD: Manuel, if I’m right, you opted (in your study) for a weekly schedule?

Manuel Hidalgo, MD, PhD: It was a weekly schedule.

Thomas Seufferlein, MD: Would you say a biweekly schedule would work as well?

Manuel Hidalgo, MD, PhD: Well, it’s hard to say. The phase I component of that study is small. The way we defined what to pick was based on 30-day mortality and cumulative grade 3/4 toxicity. So, we applied a few criteria, and for small differences, the conventional schedule was selected. Personally, I don’t think there is a big difference.

Thomas Seufferlein, MD: OK.

Tanios Bekaii-Saab, MD, FACP: Relative to the discussion of gemcitabine as a single agent, there was a meta-analysis a while ago that essentially suggested for patients with a performance status of 2, there’s zero benefit from single-agent gemcitabine. In fact, it ends up being a toxic placebo, at best. I do agree that you can either treat patients with a doublet or a triplet regimen (when eligible) or you can go the route of best supportive care.

Transcript Edited for Clarity