We started this study because locally advanced pancreatic cancer represents 30-35% of all pancreatic cancers and it’s a condition that we really don’t know how to optimally treat. There have been very few studies that have really made major progress in locally advanced pancreatic cancer and the median survival has ranged somewhere between 9-19 months across several prospective studies. 

The modern systemic therapy regimes of FOLFIRINOX and gemcitabine/nab-paclitaxel are used routinely in locally advanced pancreatic cancer, even though there has really been only one prospective study with either of those regimens, that’s the LAPACT study with gemcitabine /nab-paclitaxel. About 10-15% of patients do get rendered operable but the vast majority of patients remain incurable. Whilst most of us will incorporate systemic therapy, we really don’t know. We would like to be able to incorporate local therapy as well but that has been very controversial. In fact, radiation therapy for locally advanced pancreatic cancer, even though many of us it, has never been proven to be of benefit for that disease.

So part of the reason that we wanted to go through this intra-arterial approach is really based on the scientific premise that has been well established that pancreatic tumours, particularly the primary tumour in the pancreas, is notoriously resistant to chemotherapy, in part because the chemotherapy doesn’t get delivered very effectively to the cancer cells. The tumour is very dense, has a very fibrotic stroma, it’s actually very minimally cellular but also there’s a high degree of interstitial pressure. So the chemotherapy that is given intravenously doesn’t actually get very effectively to the tumour. 

So this tissue microperfusion system was developed to try and increase the concentration of chemotherapy within the local tumour microenvironment and has been shown in early studies in preclinical models to increase the concentration of, in this case, gemcitabine locally to the area of the tumour by about a hundred-fold.

What was the methodology?

We wanted to design this study to do two things. One is to cut through some of the controversies, and I mentioned the controversy about the role of radiation therapy, the controversy about which systemic chemotherapy regimen to use and the controversy about how long to continue chemotherapy. So one of the ways that we thought we would circumvent those problems is really to go for a gold standard endpoint of overall survival. So all of these other issues and how we chose to pursue the study, again, have some degree of controversy but, at the end of the day, we’re asking a fairly simple question which is does the addition of intra-arterial gemcitabine instead of continuation of the IV gemcitabine /nab-paclitaxel improve overall survival for these patients?

So the study design was a randomised phase III trial. It had a power of 80% to detect a hazard ratio of 0.06. With that study design, 114 patients would ultimately need to be randomised to detect the improved overall survival. But we also designed the study in a very real-world way, meaning that we know that local therapies generally are being used after a period of systemic chemotherapy now. So the way the study is designed is patients get started on induction gemcitabine/nab-paclitaxel, completely standard of care. In fact, actually we allowed patients to start gemcitabine/nab-paclitaxel prior to enrolment and then they would still be eligible if they came to a study centre to join the study because many times patients will see us for a second opinion after they’ve already started treatment. So we’ve allowed them to start the gemcitabine/nab-paclitaxel and then enrol on the study.

So, they get an induction period of chemotherapy, systemic chemotherapy, with gemcitabine /nab-paclitaxel. We also added in, during the induction phase, radiation therapy, specifically SBRT, stereotactic body radiotherapy. The reason for that is, not to dive deep into the controversy about the role of radiation, but it’s actually based on the observation in the original registry study done with this tissue microperfusion system that the patients who got radiation therapy prior to getting the intra-arterial gemcitabine had a significantly longer overall survival. In some of the preclinical models this is really borne out with the science because essentially what happens is this intra-arterial gemcitabine is injected into the artery most proximal to the tumour and we are anticipating that it diffuses or that there is a higher concentration that’s getting to the tumour. Some of that gemcitabine will leak off into the vasa vasorum, the small little blood vessels that actually feed the artery itself. The radiation therapy actually damages those vasa vasorum and that’s really focussing the gemcitabine to get just primarily to the tumour. So during induction patients also get SBRT.

After four total months of induction, patients then enter the randomisation phase where they get randomised to either get continuation of IV gemcitabine/nab-paclitaxel for four months, standard of care dosing, or they get up to four months of intra-arterial gemcitabine which is given intra-arterially once every two weeks for a total of up to eight treatments. That four months is really the study period, the randomisation period. After that period if patients continue to have non-progressive disease they then go on to continuation therapy. Again, making it real world appropriate, we gave physicians the choice to either continue the IV gemcitabine/nab-paclitaxel or to move on to maintenance capecitabine, which is what I tend to do, so it was my influence in that study. Then we ultimately follow the patients for overall survival. 

Again, the study needs 114 patients ultimately to be randomised but the FDA allowed for two interim analyses and this report was the first interim analysis where 26 events had occurred, in this case events being death or withdrawal from the study for any reason. At the point of the first interim analysis where these 26 events had occurred, 45 patients total had been randomised in the study.

What were the findings?

This was a true locally advanced patient population, locally advanced by NCCN guidelines which are now considered the gold standard across the world. What we found at the first interim analysis was, first of all, that actually, to my surprise, to be perfectly frank, the period of randomisation to intra-arterial gemcitabine was actually better tolerated. When you look at all adverse events the rate of adverse events was 65% lower for patients who received intra-arterial gemcitabine. When you really break it down by each individual adverse event, the intra-arterial gemcitabine group tolerated things better, for the most part, with the exception of abdominal pain and nausea which tended to be transient, better than the continuation IV gemcitabine/nab-paclitaxel.

In terms of the hard outcome of survival at this first interim analysis, we were very pleased to see that there was an improvement in overall survival. So the median overall survival for patients who were randomised to the intra-arterial gemcitabine arm was 15.7 months compared to 10 months, roughly, for the patients who were continuation of IV gemcitabine /nab-paclitaxel. Then, similarly, the progression free survival was also greater, it was about 14.5 months for the IA gemcitabine versus only about 7.6 months for the IV gemcitabine /nab-paclitaxel. 

Now, these numbers don’t sound great but actually it really has to take into consideration that we’re counting the clock from the period of randomisation and, on average across the groups, the patients who get to the point of randomisation were in the induction phase 5.5 months from the time of diagnosis. So, really, when you look at it from the time of diagnosis, the overall survival for the IV gemcitabine/nab-paclitaxel arm was about 15.5 months, which is right in line with previous prospective studies, but the intra-arterial gemcitabine arm was 21.5 months which is about six months longer.

How can these results impact the treatment of advanced pancreatic cancer?

I definitely will be very clear that this is just the first interim analysis and I’m holding my breath. I would love for this to be a positive study, just to have another option for patients, but I also know the nature of pancreas cancer trials and how many of them starting as promising phase I or phase II studies ended up being negative phase III studies. So this is a little bit different, it’s an interim analysis of the phase III study but, nevertheless, we really have to wait for the final results to come out.

We are expecting the second interim analysis in late 2024 and the final study to be reported out in 2025. If this holds true, if the current overall survival benefit holds true, it would definitely open up an option for patients to incorporate this into the treatment for their pancreatic cancer. The observation, and we actually are doing a bunch of healthcare-related quality of life assessments, if the observation that the adverse events are less, that the patients actually tolerate it better, then it really paints a picture where patients basically get a break from chemotherapy, from ongoing weekly, three weeks on, one week off, IV gemcitabine/nab-paclitaxel and instead can get this period where they’re just getting the once very two weeks procedure. It is a procedure that has to be done in the interventional radiology suite, but it seems like patients actually are tolerating it pretty well. So that’s one nice thing, it gives them this break off of systemic chemotherapy. 

The other thing is it opens the door. A lot of people have asked us about what about other chemotherapeutic agents that could be delivered intra-arterially. Certainly, there are drugs that are being investigated in addition to gemcitabine to be used intra-arterially.

Of course, with the NAPOLI-3 study, we can have a whole separate discussion of how to interpret that data, but one of the interpretations that’s perfectly reasonable, and Dr O’Reilly actually said this clearly in her recent ASCO presentation, whatever the debate of whether we should be using nanoliposomal irinotecan or not, one thing that is clear is that three-drug chemotherapy has now been shown to be better than two-drug chemotherapy, i.e. NALIRIFOX is better than gemcitabine/nab-paclitaxel. So with that foundation, it opens up the door to saying, okay, if in this intra-arterial gemcitabine paradigm the overall survival holds true, then we really should also be testing this in the context of a FOLFIRINOX-based regimen which, actually, really is a regimen that needs a break. I treat many patients with IV gemcitabine/nab-paclitaxel and don’t really mind continuing it on an ongoing basis but you can’t give patients FOLFIRINOX for more than about 4-6 months without really needing to drop the doses, change the schedule, something along those lines. So to have a period of a respite with 4 months of IA gemcitabine would be a real pleasure to incorporate into the treatment of locally advanced pancreatic cancer for patients treated with FOLFIRINOX. But, of course, that’s probably something that needs to be established in its own independent study.