During an OncLive Peer Exchange®, a panel of pancreatic cancer experts discuss how they select between first- and subsequent-line cytotoxic regimens and strategies they use to reduce toxicity, including dose modifications and alternative combinations. They also provide insights on some of the emerging therapies in clinical trials for advanced pancreatic cancer.
John L. Marshall, MD
Despite a move away from chemotherapy for many solid organ cancers, cytotoxic regimens remain the standard of care for patients with metastatic pancreatic cancer. However, frequently used regimens only modestly improve survival, are associated with considerable toxicity, and cause treatment resistance in many patients early on. Investigators are focusing on approaches for improving the tolerability to cytotoxic agents and on identifying novel anticancer therapies that may lead to long-term survival, if not cure, with minimal toxicity.
During an OncLive Peer Exchange®, a panel of pancreatic cancer experts discussed how they select between first- and subsequent-line cytotoxic regimens and strategies they use to reduce toxicity, including dose modifications and alternative combinations. They also provided insights on some of the emerging therapies in clinical trials for advanced pancreatic cancer. “I think we are at the tip of the iceberg. There are so many new agents being tested for the first time,” panelist Shubham Pant, MD, said.
First-Line Cytotoxic Regimens and Maintenance Therapy
In patients with treatment-naïve metastatic pancreatic adenocarcinoma who have a good performance status, the National Comprehensive Cancer Network (NCCN) guidelines recommend FOLFIRINOX/modified FOLFIRNOX for those with an Eastern Cooperative Oncology Group (ECOG) score of 0 to 1 and gemcitabine/nab-paclitaxel (Abraxane) for those with an ECOG score of 0 to 2. FOLFIRINOX consists of folinic acid, fluorouracil (5-FU), irinotecan hydrochloride, and oxaliplatin; the modified form involves reducing the initial dose of 5-FU and irinotecan.1
“We do not have, at this point, head-tohead data to compare the 2,” panelist Edward Kim, MD, PhD, said, noting that this makes selecting between the recommended first-line options challenging. He said he ultimately decides based on patients’ fitness and goals. “We have to recognize the really difficult position these patients are in with metastatic disease, oftentimes with not-great performance status, needing something for not just treatment but palliation. We can’t lose sight of that aspect of it,” he said.
There are no biomarkers to universally guide treatment, but the panelists noted that the presence of a BRCA mutation may help in therapy selection for some patients. “I think [it’s] the best biomarker we could use right now, and if they’re BRCA-positive, use platinum,” Pant said. The NCCN recommends platinum-based regimens including FOLFIRINOX/modified FOLFIRNOX or gemcitabine/cisplatin.1
Table. PRODIGE 35-PANOPTIMOX Results2 (Click to Enlarge)
In addition to improving the toxicity profile of cytotoxic regimens, investigators have sought ways to modify these regimens to enable other nonchemotherapeutic agents to be incorporated, with a goal to further improve outcomes. The panelists discussed one such approach presented by Picozzi and colleagues at the 2019 Gastrointestinal Cancers Symposium, which assessed alternating gemcitabine/nab-paclitaxel with FOLFIRI.3
The study reported a median overall survival (OS) of 14.1 months, which is competitive with other reports, and the investigators noted the regimen had a more favorable toxicity profile while enabling a broader platform for the addition of other nonchemotherapeutic agents to the base regimen.3 However, it remains unclear which nonchemotherapeutic agents may work synergistically with this regimen to improve outcomes.
When discussing maintenance approaches, the panelists said their decision largely hinges on patients’ clinical picture and preferences. “If they can handle it, I like to continue FOLFIRINOX,” panelist Paul E. Oberstein, MD, said. “I agree with FOLFIRINOX,” panelist Allyson Ocean, MD, said, with Kim concurring.
In patients with nausea and diarrhea, Oberstein said he may use 5-FU alone. In contrast, Pant said he prefers using fewer drugs during maintenance therapy, particularly because of the neurotoxicity associated with FOLFIRINOX. “To keep FOLFIRINOX or FOLFOX [folinic acid, 5-FU, oxaliplatin] going for some time is fairly challenging in these patients, so I tend to go more toward—depending on the disease control—either FOLFIRIbased or 5-FU—based therapy,” he said, noting that the data on 5-FU have shown similar survival to FOLFIRINOX.
The panelists proceeded to discuss a new and potentially exciting noncytotoxic maintenance therapy approach using olaparib (Lynparza) following first-line platinum-based therapy in patients with germline BRCA mutations. The approach was assessed in POLO, the first phase III trial to evaluate the efficacy of maintenance treatment with a PARP inhibitor in pancreatic cancer.4
“[The study investigators] looked at over 3000 patients internationally and identified around 150 who had a germline mutation, so around 7% or 8% [of patients],” Oberstein said, emphasizing that blood samples were all it took to identify these patients.
Patients with BRCA mutations who completed 4 months of platinum-based chemotherapy without progression were randomly assigned to receive olaparib or placebo. The median progression-free survival (PFS) with the PARP inhibitor was 7.4 months compared with 3.8 months with placebo (HR, 0.53; 95% CI, 0.35-0.82; P = .0038). In addition, after 2 years, 22.1% of patients had no disease progression versus 9.6% of those who received placebo.4
“Those who got the PARP inhibitor had around a doubling of the time until their tumor recurred compared with those who got essentially nothing, which I think was validation of this concept that a PARP inhibitor in the patient with the right selection, which is a very small population, can provide benefit,” Oberstein said.
Kim cautioned that although PFS was significantly improved with olaparib, the interim OS was similar between arms. “I think we have to think very carefully about how to interpret that…for me, I’m not sure it changes what I would practice right now,” he said.
However, because the survival data were still immature, Ocean was hopeful. “In the PARP arm, the survival time was over a year of disease, and the survival data were only less than half mature, so we may see something more come out of this,” she said. “I feel that where the trial should go now, in this population, is, how can we harness the DDR [DNA damage response] pathway even more using combinations with chemotherapy and immunotherapy or even vaccines,” she added.
Regardless of the panelists’ enthusiasm levels for the data, they all agreed that the trial had a revolutionary design and indicated that more studies like it are needed to find subsets of patients who may benefit from targeted agents and other novel approaches, which may even include taking a break from treatment.
“It opens the door for further combination studies…I’m very optimistic,” Pant said.
Oberstein pointed out that even patients who received the placebo had good survival. “That sort of argues that maybe it is OK to stop everything [in a group with BRCA mutations that has done well], which is something I would have never done,” he said.
However, POLO trial data need to mature and other randomized studies conducted before more definitive conclusions can be reached regarding the viability of giving patients a break from therapy, particularly because it can be difficult to rescue patients once their disease progresses.
In the second-line setting, the preferred regimen is typically whichever regimen was not administered in the first-line setting.1 “In a patient who has had 5-FU—based therapy first line, we sequence usually to gemcitabine- based therapy, whether gemcitabine/ nab-paclitaxel, gemcitabine alone, or sometimes gemcitabine/capecitabine. In the patients who do get gemcitabine/nab-paclitaxel first line, we have liposomal irinotecan [Onivyde] in combination with 5-FU as a standard option,” Ocean said.
She explained that the liposomal irinotecan tends to be less toxic, particularly regarding neurotoxicity, indicating it’s a “non-neuropathy—causing, active second-line option,” which is important because a lot of patients may have already developed neuropathy during first-line therapy. However, she reported observing higher rates of bone marrow suppression and diarrhea in patients receiving this regimen, making it imperative to monitor blood cell counts weekly.
The panelists proceeded to dissect the question of whether it becomes possible to deconstruct second-line therapy to avoid such effects and reduce overall toxicity, such as by giving liposomal irinotecan alone. Pant said that the original trial examining this, NAPOLI-1, did not find benefit with liposomal irinotecan monotherapy and that the final OS analysis showed liposomal irinotecan plus 5-FU/leucovorin to maintain an OS advantage over 5-FU/leucovorin during extended follow-up; the estimated 1-year OS rates were 26% and 16%, respectively.5
“It has to be 5-FU and liposomal irinotecan. The big thing about giving this drug is that you just need to talk to the patient about the toxicities and follow them very closely,” he said. Kim concurred and said that liposomal irinotecan plus 5-FU should be the standard of care after gemcitabine-based therapy.
The panelists noted that superiority of liposomal irinotecan plus 5-FU/leucovorin has also been shown compared with some other second-line regimens, including oxaliplatin-based chemotherapy. In one such study presented at the 2019 Gastrointestinal Cancers Symposium, the median PFS in patients who received liposomal irinotecan plus 5-FU/leucovorin after gemcitabine-based chemotherapy was 4.49 months, whereas those who received oxaliplatin plus fluoropyrimidines had a median PFS of 3.44 months (HR, 0.47; 95% CI, 0.27- 0.81; P = .007).6 Median OS also improved. In patients with a CA 19-9 level >1000 kU/L at the beginning of second-line treatment, it was 9.31 versus 6.16 months, respectively (HR, 0.43; 95% CI, 0.18-1.02; P = .0386).
A variety of noncytotoxic agents are being actively studied in clinical trials. One such agent the panelists discussed is pegvorhyaluronidase alfa, or PEGPH20, which is being studied in the phase II HALO 202 trial.7 PEGPH20 helps degrade hyaluronan (HA), a polysaccharide that is found in soft tissues throughout the body but accumulates in excess in the tumor microenvironment, leading to elevated interstitial pressure and impairing perfusion, hindering drug delivery to the tumor.7
In HALO 202, treatment-naïve patients were randomly assigned to receive PEGPH20 plus nab-paclitaxel/gemcitabine (PAG) or nab-paclitaxel/gemcitabine. PAG was found to improve PFS, with the largest improvement observed in patients with HA-high tumors (ie, extracellular matrix HA staining ≥50% of tumor surface at any intensity).
However, a challenge with PAG is its high risk of thromboembolic (TE) events. “The trial stopped for a period of time because the rate of thrombosis was close to 50%,” Kim said, noting that even the nab-paclitaxel/gemcitabine arm had a high rate of TE at 25%. “Both groups were subsequently treated with low— molecular weight heparin, and that lowered their rates to <10%,” he said.
The panelists discussed whether there is value in providing thromboprophylaxis to patients with pancreatic cancer given the high risk of TE events in these patients. Pant said that the results of a subset analysis of the CASSINI trial showed rivaroxaban thromboprophylaxis decreases blood clots without increasing the risk of major bleeding in patients with pancreatic cancer.8 “We don’t know if [thromboprophylaxis] makes patients live longer, but we know that if we want to give it, you have an argument to give it,” he said.
In other cancers, the presence of certain molecular aberrations, such as high microsatellite instability (MSI) and NTRK fusions, can open the door to new and transformative treatments. The FDA has approved 2 drugs, larotrectinib (Vitrakvi) and entrectinib (Rozlytrek), in solid tumor settings with these aberrations.9 Although these aberrations have been observed in pancreatic cancer, they are exceedingly rare, leading the panel to discuss whether it is worthwhile to test patients for them.
Ocean said she may consider testing in patients with a strong family history of cancer. However, Kim indicated that this might change because some markers can now be detected on liquid biopsies, potentially negating the need for complicated tissue biopsies. “I recently diagnosed a patient, through a liquid biopsy, of MSI high,” he said.
As a final part of the discussion, the panelists examined whether there is any role for immunotherapy. Thus far, they noted that there is no activity. “With the immune checkpoint inhibitors, the response is 0% as a single agent in pancreatic cancer—you cannot go lower than that,” Pant said.
However, he explained that numerous efforts are underway to try to turn these cold tumors hot, such as by using colony-stimulating factor 1 receptor agents, RIP1 kinase inhibition, and CD40 antibody. “We just need to keep on looking and keep on putting more patients on trial, and maybe earlier on, not when they’re in third line and they’ve got a low albumen and they’re getting taps,” he said.
In their concluding remarks, the panelists emphasized the importance of enrolling patients in clinical trials. “We need smarter trials. We need more adaptive designs to get the answers early. I think that’s where we really move the needle in this disease,” Pant said.
However, a major challenge is keeping patients on FOLFIRNOX long enough to elicit a response, since it is a difficult regimen to tolerate. The panelists proceeded to discuss approaches that have been assessed to make FOLFIRNOX more tolerable. Pant reviewed the phase II PRODIGE 35-PANOPTIMOX trial, which compared 3 approaches: 6 months of FOLFIRINOX (arm A); 4 months of FOLFIRINOX followed by 5-FU/leucovorin, with FOLFIRINOX reintroduced if patients progressed (arm B); or gemcitabine monotherapy alternating with folinic acid, 5-FU, and irinotecan hydrochloride (FOLFIRI; arm C).2 Outcomes were comparable between arms A and B, with outcomes in arm C inferior to those observed in the other arms (Table).2 Based on these data, Pant said he would “start with FOLFIRINOX, [complete] the 4 months, and then try to do a maintenance approach.”