Findings from the phase 3 NAPOLI 3 trial support the NALIRIFOX regimen as a first-line treatment for patients with metastatic pancreatic ductal adenocarcinoma.
Treatment with liposomal irinotecan (Onivyde) plus 5-fluorouracil (5-FU), combined with leucovorin, and oxaliplatin (NALIRIFOX) led to improvements in overall survival (OS) and progression-free survival (PFS) compared with nab-paclitaxel (Abraxane) plus gemcitabine in patients with metastatic pancreatic ductal adenocarcinoma (PDAC).1
These findings come from the phase 3 NAPOLI 3 trial (NCT04083235) presented at the 2023 Gastrointestinal Cancers Symposium. Data showed that at a median follow-up of 16.1 months (95% CI, 15.3-16.8), patients in the intent-to-treat (ITT) population administered liposomal irinotecan plus NALIRIFOX (n = 383) experienced a median OS of 11.1 months (95% CI, 10.0-12.1) compared with 9.2 months (95% CI, 8.3-10.6) for those given nab-paclitaxel plus gemcitabine (n = 387; HR, 0.83; 95% CI, 0.70-0.99; P = .04).
“These results support the NALIRIFOX regimen as a new regimen for the first-line treatment of patients with metastatic PDAC. [The regimen represents] something we can hopefully build off in the future,” lead study author Zev A. Wainberg, MD, a professor of medicine at University of California, Los Angeles (UCLA) and co-director of the UCLA GI Oncology Program, said in a presentation of the data.
Based on the findings of NAPOLI 3, Ipsen plans to file a supplemental new drug application seeking the approval of liposomal irinotecan plus NALIRIFOX in this patient population in the United States.2 In October 2015, the FDA approved liposomal irinotecan plus 5-FU/leucovorin for the treatment of patients with metastatic PDAC after progression with gemcitabine-based treatment.3
The randomized, open-label NAPOLI 3 study evaluated the efficacy and safety of liposomal irinotecan plus NALIRIFOX vs nab-paclitaxel plus gemcitabine in patients with confirmed metastatic PDAC who were previously untreated in the metastatic setting. Patients were required to have metastatic disease diagnosed no more than 6 weeks prior to screening, at least 1 metastatic lesion measurable by MRI or CT scan per RECIST v1.1 criteria, and an ECOG performance status of 0 or 1.
Enrolled patients were randomly assigned 1:1 to 50 mg/m2 of liposomal irinotecan plus 2400 mg/m2 of 5-FU, 400 mg/m2 of leucovorin, and 60 mg/m2 of oxaliplatin on days 1 and 15 of each 28-day cycle, or 1000 mg/m2 of gemcitabine plus 125 mg/m2 of nab-paclitaxel on days 1, 8, and 15 of each 28-day cycle. Treatment continued until disease progression, unacceptable toxicity, or study withdrawal. Tumor assessment was conducted every 8 weeks per RECIST v1.1 criteria, and follow-up continued until death or study conclusion.
Stratification factors included ECOG performance status (0 vs 1), region, and presence or absence of liver metastases.
OS served as the trial’s primary end point. Secondary end points comprised investigator-assessed PFS and overall response rate (ORR) per RECIST v1.1 criteria and safety. Health-related quality of life and biomarker assessments were exploratory end points.
Statistical analysis consisted of a stratified long-rank test with a 2-sided alpha of 0.05 and 90% power. Analysis occurred after at least 543 OS events in the ITT population in a stepwise hierarchical test for efficacy. The P boundary for efficacy at the final analysis was <.048.
At the July 23, 2022, data cutoff, 85.1% of patients in the experimental arm discontinued treatment compared with 96.1% in the control arm. The primary reasons for discontinuation in the liposomal irinotecan/NALIRIFOX arm were progressive disease (48%), adverse effects (AEs; 14.1%), death (8.4%), or other (14.7%). In the nab-paclitaxel/gemcitabine arm, those rates were 45.7%, 23.8%, 6.7%, and 19.9%, respectively.
Baseline characteristics were similar across both treatment arms. The median age was 64 years (range, 20-85) and 65 years (range, 36-82) in the liposomal irinotecan/NALIRIFOX and nab-paclitaxel/gemcitabine arms, respectively. The majority of patients in both arms were male (53.3% and 59.4% for the experimental and control arms, respectively), White (82.2% and 83.7%), had an ECOG performance status of 1 (58.0% and 56.6%), had at least 3 metastatic sites (38.9% and 36.4%), had liver metastases (80.2% and 80.4%), were treated in the rest of the world outside of North America and East Asia (65.8% and 65.6%), had their main tumor located outside the head of the pancreas (61.6% and 59.7%), and had a baseline CA 19-9 level of at least 37 U/mL (83.8% and 81.7%).
The median time from metastatic diagnosis at study entry was 3.00 weeks (range, 0.6-9.1) for the liposomal irinotecan/NALIRIFOX group and 3.57 weeks (range, 0.4-10.9) for the nab-paclitaxel/gemcitabine group.
Additional data showed the patients in the liposomal irinotecan/NALIRIFOX arm achieved a median PFS of 7.4 months (95% CI, 6.0-7.7) compared with 5.6 months (95% CI, 5.3-5.8) for the nab-paclitaxel/gemcitabine arm (HR, 0.69; 95% CI, 0.58-0.83; P < .0001).
Liposomal irinotecan/NALIRIFOX elicited an ORR of 41.8% (95% CI, 36.8%-46.9%) vs 36.2% (95% CI, 31.4%-41.2%) for nab-paclitaxel/gemcitabine. The complete response rate was 0.3% in both arms. In the liposomal irinotecan/NALIRIFOX arm, the rates of partial response, stable disease, and progressive disease were 41.5%, 25.8%, and 9.9%, respectively. In the nab-paclitaxel/gemcitabine arm, those rates were 35.9%, 26.1%, and 14.5%, respectively. Additionally, 22.5% and 23.3% of patients from the liposomal irinotecan/NALIRIFOX and nab-paclitaxel/gemcitabine arms, respectively, were not evaluable for response.
In the experimental arm, 50.5% of patients underwent subsequent anticancer therapy, including 50.5% who had systemic anti-neoplastic therapy, 0.3% who had surgery, and 0.5% who received radiotherapy. In the nab-paclitaxel/gemcitabine arm, 54.4% of patients received subsequent anticancer therapy, including 54.1% who had systemic anti-neoplastic therapy, 0.5% who had surgery, and 1.1% who had radiotherapy.
Among patients evaluable for safety, patients treated with liposomal irinotecan/NALIRIFOX (n = 370) had a median treatment duration of 24.29 weeks (range, 0.4-100.9) compared with 17.57 weeks (range, 0.7-81.7) for those in the nab-paclitaxel/gemcitabine arm (n = 379).
Any-grade treatment-emergent AEs (TEAEs) occurred in 99.7% and 99.2% of patients in the experimental and control arms, respectively, and 95.1% and 92.9% had any-grade TEAEs related to the treatment regimen. Grade 3 or higher TEAEs were reported in 87% of patients in the liposomal irinotecan/NALIRIFOX arm and 86% of patients in the nab-paclitaxel/gemcitabine arm, and 70.8% and 68.1%, respectively, had grade 3 or higher TEAEs related to treatment.
Serious TEAEs occurred in 54.3% of patients given liposomal irinotecan/NALIRIFOX, and 26.5% experienced a serious TEAE related to treatment. In those administered nab-paclitaxel/gemcitabine, 51.5% had serious TEAEs, and 19% had serious TEAEs related to treatment. TEAEs leading to death occurred in 5.9% of patients in the liposomal irinotecan/NALIRIFOX group, including 1.6% of patients who died due to a TEAE related to treatment. In the nab-paclitaxel/gemcitabine group, 6.1% of patients experienced a TEAE leading to death, and 2.1% of patients died due to TEAEs related to treatment.
“When one looks at the nuances in the patients and the toxicity profiles, we can see these 2 regimens have very different toxicity profiles,” Wainberg said.
The most common hematologic TEAEs in the experimental and control arms, respectively, included neutropenia (any grade, 29.5% vs 31.9%; grade 3/4, 14.1% vs 24.5%), decreased neutrophil count (20.5% vs 18.7%; 9.7% vs 13.5%), febrile neutropenia (2.4% vs 2.6%; 2.4% vs 2.4%), anemia (26.2% vs 40.4%; 10.5% vs 17.4%), thrombocytopenia (13.5% vs 22.7%; 0.8% vs 3.7%), and decreased platelet count (10.5% vs 17.9%; 0.8% vs 2.4%).
The most common non-hematologic TEAEs in the liposomal irinotecan/NALIRIFOX and nab-paclitaxel/gemcitabine groups, respectively, included diarrhea (any grade, 70.5% vs 36.7%; grade 3/4, 20.3% vs 4.5%), nausea (59.5% vs 42.7%; 11.9% vs 2.6%), vomiting (39.7% vs 26.4%; 7.0% vs 2.1%), hypokalemia (31.6% vs 12.9%; 15.1% vs 4.0%), peripheral neuropathy (17.8% vs 17.4%; 3.2% vs 5.8%), peripheral sensory neuropathy (15.1% vs 13.5%; 3.5% vs 2.9%), paresthesia (11.9% vs 8.7%; 0.3% vs 0.5%), and pyrexia (10.5% vs 23.0%; 0.8% vs 1.6%).
In a discussion of the presentation, Laura Goff, MD, MSCI, an associate professor of medicine, and the executive medical director for the Cancer Patient Care Center at Vanderbilt Ingram Cancer Center, referenced the phase 2 SWOG S1505 trial (NCT02562716) of mFOLFIRINOX vs nab-paclitaxel/gemcitabine in patients with resectable PDAC. She noted that the resectable setting is clinically different from the metastatic setting, and although the randomized SWOG S1505 study was not designed as a head-to-head trial, the similar outcomes between the arms led to the belief that there was not a difference between the regimens.
“I would argue that this randomized trial of NAPOLI 3 should supersede our opinion about the efficacy of these regimens [mFOLFIRINOX and nab-paclitaxel/gemcitabine] in the metastatic setting,” Goff said.
Goff also pointed to safety findings observed with both liposomal irinotecan/NALIRIFOX and nab-paclitaxel/gemcitabine during NAPOLI 3. “High rates of toxicity were seen in both arms, despite the reputation that gemcitabine/nab-paclitaxel is a significantly easier regimen to tolerate. I would argue that the data do not necessarily support that,” Goff said.