Tanios Bekaii-Saab, MD, discusses emerging treatments offering hope to patients with pancreatic cancer.
Tanios Bekaii-Saab, MD
The increasing incidence of pancreatic cancer could make the malignancy the second-leading cause of cancer-related death in the United States, making it imperative to find new treatment options, explains Tanios Bekaii-Saab, MD.
PARP inhibitors are one of the treatment strategies being developed for patients with BRCA1/2-mutated pancreatic cancer, as those mutations are known to be sensitive to PARP inhibition across several cancer types. Three PARP inhibitors currently being investigated include veliparib (ABT-888), olaparib (Lynparza), and rucaparib (Rubraca), particularly in the maintenance setting.
Other approaches are also being explored, noted Saab, pointing to ongoing studies of PEGPH20 and stemness inhibitors.
In an interview during the 2018 OncLive® State of the Science Summit™ on Gastrointestinal Cancers, Bekaii-Saab, a professor of medicine at Mayo Clinic, discussed emerging treatments offering hope to patients with pancreatic cancer.Bekaii-Saab: Pancreatic cancer is leading the trend in terms of increasing incidence. The mortality of pancreatic cancer continues to be on the rise. In fact, if nothing is done to change this, it will be the second cancer killer in the United States, creating an urgency to find new treatments.
Over the last few years, we have developed new strategies to extend survival for our patients. Half of the patients are crossing the 1-year survival line, with 25% crossing the 2-year survival line. We have patients who survive 3, 4, and 5 years, but it is less than 5% of patients. However, this would be unheard of 5 or 6 years ago, so we are doing better. We are on the right path to ultimately bring pancreatic cancer into the mainstream.
The key in treating pancreatic cancer is to make sure that we have a strategy. The nihilism that surrounds this cancer and its high mortality has left many physicians to overlook sequencing. The thought was to find the best treatment first and figure out the rest later. Like other cancers, we need to start thinking about sequencing.
There are 2 main pathways for this cancer. My preferred sequence is to start with gemcitabine and nab-paclitaxel (Abraxane), typically biweekly, as it preserves the outcome and seems to significantly improve the toxicity profile. Patients are less likely to tolerate this with a different dose. When patients fail gemcitabine and nab-paclitaxel, then they can receive 5-fluorouracil (5-FU) and MM-398 (irinotecan liposome injection; Onivyde) based on the NAPOLI-1 trial. Then in the third-line setting, we move to a platinum-based regimen such as FOLFOX (5-FU, leucovorin, and oxaliplatin) or cisplatin.
The reason why FOLFOX has been pushed further down the line is because there were 2 conflicting studies; 1 was from Canada and 1 was from Europe. One showed some benefit, whereas the other showed detriment. I do not buy the detriment study, but it shows that there is no benefit.
We published a meta-analysis that showed all of the experience with phase III studies of oxaliplatin and irinotecan. What was found was that the irinotecan class of compounds tends to improve outcomes with progression-free survival (PFS) and overall survival (OS). The oxaliplatin-based regimens improved PFS, but did not improve OS. They are less likely to benefit patients.
Overall, the sequence in that group is gemcitabine and nab-paclitaxel, 5-FU, and if patients get to the third-line, oxaliplatin. However, if there is a clinical trial, that takes precedence.
The other way to treat patients is to start with FOLFIRINOX (5-FU, leucovorin, irinotecan, and oxaliplatin). That is an aggressive regimen. Even for the younger and more performant patients, I argue that you should always consider 1 strategy or the other because we do not have any trial that has compared the 2.
Also, the toxicities can be quite tough and cumbersome on patients. However, if you choose to go down that path with FOLFIRNOX in the first-line setting, the second-line setting is more obscure because there are not a lot of data. We need to take inference from smaller studies. Gemcitabine and nab-paclitaxel could be an option for some patients if they do not have the stamina, but gemcitabine as a single agent does not have much activity.
In the third-line space, there are not many options left. That is assuming the patients have a performance status of 1. If the patient has a performance status of 2 and above, those patients do not seem to benefit from any therapy and seem to be candidates for best supportive care and hospice.
With that being said, 1 of the elements that is important to keep in mind is that many physicians confuse performance status with symptoms that are not well controlled. Once you optimize the symptom control for patients, then you can better assess performance status. If a patient has a poor performance status, they often have pain and are not eating well. However, if they are put on enzymes, pain medications, and occasionally steroids, it can improve their appetite and activity.
Some patients went from a performance status of 2 or 3 to a performance status of 0 or 1. Many of them benefitted from treatment so it is important to keep in mind that performance status is a big detriment of who should get aggressive therapy versus no therapy. However, we don't make conclusions about performance status until we optimize the symptoms of the patient. Pancreatic cancer is now seeing more of an understanding of the molecular and genetic drivers. We are understanding what subsets of patients may benefit from certain treatments. For patients with microsatellite instability-high (MSI-H) tumors, immunotherapy works best. That subgroup was included in the study that led to the approval of pembrolizumab (Keytruda) in MSI-H cancers. There were a few patients with pancreatic cancer who had incredible responses. One patient continues to be in a complete remission from that treatment.
Unfortunately, less than 1% of all patients with pancreatic cancer have MSI-H tumors. However, in my clinic, every single patient with pancreatic cancer gets checked for MSI because if it is 1 out of 100, then if you don't check 100 patients you're not going to find the 1 [who has an MSI-H tumor]. I want to find that 1 patient because they may not even need chemotherapy and can go straight to pembrolizumab.
The other group of patients who are showing promise are patients with BRCA1/2 and PALB2 mutations and homologous recombination deficiency (HRD) as they may benefit from platinum-based treatment and PARP inhibitors. There are studies evaluating different PARP inhibitors in pancreatic cancer. The 3 PARP inhibitors that have been studied the most are veliparib, olaparib, and rucaparib. Veliparib appears to be the least potent. There are rare reports of any single-agent activity of this drug in pancreatic cancer. I don't think that this agent should be developed for pancreatic cancer. I am not even sure that it has enough potency as a PARP inhibitor.
The other 2 PARP inhibitors, olaparib and rucaparib, have shown single-agent activity in refractory patients. I don't believe that single-agent PARP inhibitors are going to be the answer for an aggressive disease like pancreatic cancer, but they would be optimally combined for maintenance phases. There are maintenance studies being done where patients receive FOLFIRINOX and then go on to receive olaparib after 4 months. Patients are randomized to olaparib versus observation.
The other strategy that is more interesting, in my opinion, is combining these agents when feasible with regimens containing either irinotecan-based compounds and then a liposomal irinotecan or a platinum-based agent, such as cisplatin. Preclinical data suggest that those agents synergize better with those inhibitors than they do with a platinum-based agent.
These strategies are being looked at in various studies. There are studies with oxaliplatin, cisplatin, veliparib, and rucaparib. One study that we will be launching soon is looking at 5-FU in combination with rucaparib. This will be a phase I trial that will lead into a phase Ib and II study for unselected patients and then eventually in BRCA1/2-mutant and HRD patients.
I am hopeful that this area with show benefit for at least 20% of patients so we can have around 20% to 25% of patients, including those with MSI-H tumors, with stronger remissions and more durable responses.There are 2 other agents and pathways that are showing promise. The first is PEGPH20, which is an agent that targets the stroma and hyaluronan (HA). It breaks down that stroma to allow chemotherapy into the tumor to maximize the kill. It is a great concept and looks very promising. There were 2 studies running parallel with each other; one is with gemcitabine and nab-paclitaxel and the other with FOLFIRINOX. Both were randomized trials.
The development of these drugs saw [adverse events] with blood clots—some being lethal—which changed the whole paradigm. Patients receiving the drug had to receive an anticoagulant low-molecular-weight heparin to help prevent the clotting events. This was smoother in the second stage. The study with gemcitabine and nab-paclitaxel in the exploratory analysis suggested that there may be a benefit for those patients who have HA-high tumors.
Then comes the disappointing study from SWOG with FOLFIRINOX with or without PEGPH20 in unselected patients. The biggest surprise was not just that there was no benefit for adding PEGPH20 to FOLFIRINOX, but there was a detriment. In fact, there was a 50% loss of survival . It was 15.1 months on FOLFIRINOX alone and 7 months on FOLFIRINOX plus PEGPH20, which is around what we see with gemcitabine alone.
We have seen that before with agents that target stroma. Stroma has been characterized as both a friend and a foe. Stroma contains the cancer from spreading out, it prevents things from going in, and keeps the center of the cancer very immunosuppressive, which is why we can kill the cancer. There seems to be a benefit from breaking down that stroma, but we need to think about how to do it best. Can we break it enough to be able to kill the cancer? If we break it too much are we releasing those cancer cells? That certainly remains a concern, but we will see what the results of the phase III studies show.
The last agent that holds promise is napabucasin (BBI-608), which is a stem cell inhibitor. That compound relies on the fact that pancreatic cancer cells tend to have stem cells that are incredibly stubborn and hard to kill with chemotherapy and radiation. When we use chemotherapy, the sensitive cancer cells will die off, but some of them will be stressed and revert to a stemless phenotype. That is why when we went from the first-line to second-line to third-line setting, we start losing benefit of chemotherapy. It doesn't matter how powerful the chemotherapy is because you lose power as you go down and enrich the cancer with these stemless-like cells.
Napabucasin seems to inhibit and facilitate killing lowering the chemosensitizing capacity against these pancreatic cancer cells. If you give it with chemotherapy and gemcitabine, nab-paclitaxel, or paclitaxel, you seem to kill those cancer cells more efficiently and prevent them from reverting to the stemless phenotype.
We did this in a phase Ib study of around 67 patients with the objective response rate being 55%. There were 2 complete responses in that group of patients and we continue to see positive trends with PFS and OS. That has taken napabucasin to the phase III CanStem111P trial, which is randomizing patients to gemcitabine and nab-paclitaxel versus gemcitabine and nab-paclitaxel plus napabucasin. This ongoing study is accruing well, and we hope to create a new standard of care for pancreatic cancer.