Results of this study demonstrated that HR070803 combined with 5-FU and leucovorin significantly extended OS among patients with unresectable locally advanced, or metastatic PDAC who have failed gemcitabine-based therapy, as compared with placebo plus 5-FU and leucovorin. The HR070803 group exhibited a 37% reduction in the risk of death and a median OS extension of 2.4 months relative to the placebo group. Moreover, patients assigned to HR070803 group showed numerically superior PFS, ORR, TTF, and CA19-9 response. The safety profile of HR070803 plus 5-FU and leucovorin was manageable.

Both this study and the NAPOLI-1 study recruited patients with metastatic PDAC who were previously treated with gemcitabine-based therapies.14,15 In the NAPOLI-1 study, treatment with Onivyde (nanoliposomal irinotecan [70 mg/m2] combined with FU and folic acid) demonstrated advantages in OS (median 6.2 versus 4.2 months, HR 0.63), PFS (median 3.1 versus 1.5 months, HR 0.56), and TTF (2.3 versus 1.4 months, HR 0.6) compared to FU plus folic acid group, and thus became the standard of care in this patient population.14,15 Our study indicated that the efficacy of HR070803 combined with 5-FU and leucovorin was comparable to this standard therapy, despite a lower dose of liposomal irinotecan (56.5 mg/m2). The promising efficacy of HR070803 may partially be attributed to the long median drug exposure time (17.4 weeks), as well as the small liposome particle size (approximately 80–90 nM), which could improve penetration of the drug to the target tumor lesion and thus contribute to a favorable efficacy profile. While cross-trial comparisons should be interpreted with caution, our results suggest that HR070803 plus 5-FU and leucovorin might be a promising alternative in regions where Onivyde is not readily available.

AEs that occurring more frequently (≥ 10%) in the HR070803 group compared to the placebo group included nausea, vomiting, diarrhea, decreased white blood cell count, decreased neutrophil count, loss of appetite, and fatigue. These AEs were consistent with those known for irinotecan, and no new toxicities were observed. The slow-release nature of HR070803 liposome maintains the blood concentrations of the active metabolite SN-38 and total irinotecan at stable lower levels, resulting in relative lower incidences of decreased neutrophil count, diarrhea, and cholinergic syndrome compared to that of irinotecan hydrochloride (54%–96.9%, 72.4%–88%, and 28.3%, respectively).21,22,23

The AE profile in our trial aligned with previous reports for irinotecan liposome.14,15,16,17,18 In our study, neither treatment discontinuation nor serious events resulting from neutropenia and diarrhea were observed. The incidence of diarrhea and decreased neutrophil count with HR070803 combination treatment was 45.6% (grade ≥ 3: 4.1%) and 33.3% (grade ≥ 3: 12.9%), respectively, compared to 59% (grade ≥ 3: 13%) and 39% (grade ≥ 3: 27%) in NAPOLI-1.14,15 Additionally, the proportion of patients who required drug discontinuation and dose reduction due to AEs in our study was 4.1% and 21.1%, respectively, while in the NAPOLI-1 study, it was 11% and 33%, respectively.14,15 The low incidence of toxicity observed suggests that HR070803 may offer an extended therapeutic window, as evidenced by a median drug exposure duration of 17.4 weeks in our study and 8.7 weeks in the NAPOLI-1 trial. This extension of the therapeutic window could potentially enhance the antitumor efficacy of HR070803-based therapy.

UGT1A1 is a crucial enzyme in irinotecan metabolism. Gene mutations in the UGT1A1 gene and decreased enzyme activity can lead to an increased incidence of diarrhea and decreased neutrophil count caused by irinotecan.24,25,26 In our study, six patients in the HR070803 group had UGT1A1 homozygous mutation, with four having UGT1A1*28 mutation and two having UGT1A1*6 mutation. Among them, only one patient with UGT1A1*28 mutation experienced neutropenia event (grade 3), indicating that the UGT1A1 polymorphism had a limit impact on neutropenia incidence in our study.

Despite the increased incidence of certain AEs associated with HR070803, there was no significant difference in quality of life between the HR070803 and placebo groups. This further suggested that the AEs related to HR070803 combination therapy are acceptable and manageable.

We selected irinotecan placebo for the control group to avoid the possibility of unblinding due to inconsistent appearance between the control drug and the HR070803, and therefore to achieve a double-blind design, and to minimize bias to the greatest extent.

One of the limitations in this study is the small number of patients with UGT1A1 mutations, which did not allow for the exploration of the necessity of dose adjustment of irinotecan hydrochloride liposome in this population. Another limitation of this study is that the current dosing regimen of 2000 mg/m2 intravenously over 46 h every two weeks for 5-FU may result in less potent outcomes when extrapolating the efficacy of our HR070803 combination therapy to a global population. This is because the standard of care (Onivyde/FU/leucovorin combination therapy), approved based on the NAPOLI-1 study, uses a dose of 2400 mg/m2 of FU under the same administration schedule, although a different dose of FU was administered in the control arm in the NAPOLI-1 study. In our study, the dose and schedule of 5-FU in both arms (same dose and schedule to facilitate a double-blind design, enhance patient compliance, and strengthen the rigor of the study design) were chosen based on clinical practice in Chinese pancreatic cancer patients, who generally have poor tolerance, to improve patient tolerance while ensuring efficacy. Additionally, 5-FU is a time-dependent chemotherapy drug and was administered by continuous infusion over 46 h every two weeks in this study, so we speculate that the efficacy will not be compromised and the toxicities will be more tolerable by reducing the dose of 5-FU to 2000 mg/m2. Due to the lack of a direct comparison between our study and NAPOLI-1, the use of this HR070803 combination therapy in racially diverse populations remains to be further clarified.

In conclusion, HR070803 plus 5-FU and leucovorin significantly extended OS and improved other efficacy endpoints in unresectable, locally advanced, or metastatic PDAC who have received gemcitabine-based therapy, compared to placebo plus 5-FU and leucovorin. The safety was acceptable and manageable. This HR070803 combination therapy might represent a novel standard second-line treatment option for this patient population.