Metastatic pancreatic cancer is one of the most lethal malignancies, with very limited treatment options (1). Nanoliposomal irinotecan (nal-IRI) is mainly used for patients with metastatic pancreatic cancer (1), because NAPOLI-1, the global randomized phase 3 trial, has revealed a survival benefit with the regimen consisted of nal-IRI plus 5-fluorouracil (5-FU) and leucovorin for patients with metastatic pancreatic cancer after previous gemcitabine-based therapy (2). Based on this trial result, the regulatory agency in Japan has approved the nal-IRI plus 5-FU and L-leucovorin (nal-IRI/FL) combination therapy. Particularly interesting studies focused on predictors of treatment-response by nal-IRI-based therapy in patients with pancreatic cancer have indicated that C-reactive protein/albumin ratio (CAR), neutrophil/lymphocyte ratio (NLR), Glasgow prognostic score (GPS), total bilirubin, carcinomatosis, and previous irinotecan treatment are significantly associated with overall survival (3-8).

The most common grade ≥3 treatment-emergent adverse event reported in Asian patients is neutropenia (54.5%) in the subgroup analysis of NAPOLI-1 study (9). Leukopenia/ neutropenia (76.1%/71.7%) and diarrhea (58.7%) are commonly seen nal-IRI/FL treatment-emergent adverse events in the Japanese randomized phase 2 trial (10).

Irinotecan released from nal-IRI in the body is converted to SN-38, which is demonstrated up to 1000-fold higher topoisomerase I inhibitory activity versus irinotecan (11). SN-38 is inactivated by uridine diphosphate-glucuronosyltransferase (UGT) 1A1 (11). Patients with UGT1A1*6 and *28 genotypes have demonstrated decreased metabolic activity (12,13). Allele frequencies of UGT1A1*6 and *28 are 0.157 and 0.097, respectively, in Japanese, and 0.007 and 0.388, respectively, in Caucasians (14). The initial dose of nal-IRI has to be adjusted according to the UGT1A1 genotype (15). This is consistent with NAPOLI-1 trial, in which patients homozygous for the UGT1A1*28 allele initially receive nal-IRI at a 20 mg/m² dose reduction from the usual 70 mg/m² (2). In the Japanese clinical study, all patients with UGT1A1 (UGT1A1*6/*6 or *6/*28) mutations in the nal-IRI/FL arm exhibited grade ≥3 decreased neutrophil count in spite of the dose reduction (15). A previous study in Taiwan has revealed a significantly higher incidence of grade ≥3 neutropenia and diarrhea in patients with UGT1A1*6 or *28 homozygosity/compound heterozygosity than in those with single heterozygosity/wild type (16). However, no study has reported the risk factors for the adverse events during nal-IRI/FL treatment based on analysis of factors other than the UGT1A1 polymorphism in clinical settings.

The present study aimed to investigate the risk factors for leukopenia, neutropenia, and diarrhea during nal-IRI/FL treatment in Japanese patients based on their characteristics, such as UGT1A1 polymorphism, laboratory data before nal-IRI/FL initiation, and tumor location.

Ethics. This study was designed and implemented under the Declaration of Helsinki and its amendments, and was approved by the Wakayama Medical University Ethics Committee (No. 3,828). Information regarding the conduct of the study was disclosed on the website, and the research participants were given the option to refuse study participation.

Patient and safety evaluation. Patients with metastatic pancreatic cancer, who started nal-IRI/FL treatment at Wakayama Medical University Hospital, were retrospectively recruited from June 2021 to May 2022. Patient backgrounds, including the laboratory data before nal-IRI/FL initiation, were collected from the electronic medical records. White blood cell count nadir, neutrophil count nadir, and onset of diarrhea were evaluated during nal-IRI/FL treatment and graded according to the Common Terminology Criteria for Adverse Events version 5.0.

Statistical analysis. Statistical Package for the Social Sciences (SPSS^® Statistics 28.0; IBM Japan, Tokyo, Japan) was used for all statistical analyses. Data are expressed as the number of patients or the median with an interquartile range. Kruskal–Wallis tests and the post hoc Mann–Whitney U-tests with a Bonferroni correction were used to compare the medians of continuous values among groups, whereas chi-square tests were utilized to test categorical data distribution. Multivariate logistic regression analysis was conducted to identify factors affecting the onset of diarrhea, in which the onset of diarrhea was the dependent variable and all the independent variables with p-values of <0.2 defined in the univariate logistic regression analysis were tested. Multiple regression analyses were performed to identify factors related to the nadir count of white blood cell and neutrophil, in which the nadir count was the dependent variable and the independent variables with p-values of <0.2 defined in the single regression analysis were selected and tested. p-values of <0.05 were considered statistically significant.

Thirty six patients with metastatic pancreatic cancer were recruited in this study. Table I summarizes the characteristics of patients who received nal-IRI/FL treatment. There were 16, 16, and four patients with UGT1A1*6 or *28 wild-type (–/–), heterozygous (+/–), and homozygous (+/+) genotypes, respectively. The UGT1A1*6 and *28 alleles were detected in three and 12 patients in the UGT1A1*6 or *28 (+/–) group, respectively, and the allele detail of one patient was unknown. The UGT1A1*6/*6, *6/*28, and *28/*28 genotypes were detected in one, one, and two patients in the UGT1A1*6 or *28 (+/+) group, respectively. Patients with UGT1A1*6 or *28 (+/+) showed higher total bilirubin levels in their pretreatment stages. The initial nal-IRI doses of patients with UGT1A1*6 or *28 (–/–), (+/–), and (+/+) were 56 (56-56), 56 (48-67), and 48 (40-49) mg/m² (median with the interquartile range), respectively. The prescribed nal-IRI doses for patients with UGT1A1*6 or *28 (+/+) were lower than those for other patients (p=0.011). Conversely, factors other than total bilirubin level and initial nal-IRI dose were not significantly different among the UGT1A1 genotype groups.

Diarrhea during nal-IRI/FL treatment was observed in seven (24.1%) of 29 patients, who had not been prescribed any antidiarrhea medication at the commencing time of nal-IRI. The rest of seven patients were excluded, since they received loperamide for diarrhea symptoms at the start of therapy. The incidence of grade ≥1 diarrhea in patients with UGT1A1*6 or *28 (–/–), (+/–), and (+/+) were 27.3%, 21.4%, and 25.0%, respectively (Figure 1). Multivariate analysis did not extract any risk factors, which significantly affected the development of diarrhea possibly induced by nal-IRI/FL (Table II).

The nadir counts of white blood cells in patients with UGT1A1*6 or *28 (–/–), (+/–), and (+/+) were 3,340 (2,950–5,120), 3,220 (2,520-3,923), and 2,050 (1,550-2,513)/mm³, respectively (median with the interquartile range) (Figure 2A). The nadir counts of neutrophil in patients with UGT1A1*6 or *28 (–/–), (+/–), and (+/+) were 1,930 (1,540-2,782), 1,285 (1,093-2,113), and 860 (407-1,616)/mm³, respectively (median with the interquartile range) (Figure 2B). The nadir counts of white blood cells and neutrophils were significantly lower in patients with UGT1A1*6 or *28 (+/+) (p=0.033, 0.043, respectively) (Figure 2A and B). Grade 3 or 4 leukopenia and neutropenia in the UGT1A1*6 or *28 (+/+) group demonstrated an incidence rate of 50.0%, indicating a higher tendency than those in UGT1A1*6 or *28 (–/–) and (+/–) groups at 7.1% and 12.5%, respectively (p=0.085 for each) (Figure 2C and D). Multiple regression analyses revealed that the UGT1A1*6 or *28 (+/+) genotype, high AST value before the nal-IRI/FL therapy, and pancreatic head cancer were significantly associated with decreased white blood cell count (p=0.009, 0.019, and 0.030, respectively) (Table III), and that of UGT1A1*6 or *28 (+/+) was significantly related to decreased neutrophil count (p=0.017) (Table IV).

This study revealed the risk factors for leukopenia and neutropenia during nal-IRI/FL treatment in Japanese patients based on patient characteristics such as UGT1A1 polymorphism, laboratory data before nal-IRI/FL initiation, and tumor location.

This study did not extract any risk factors that significantly influenced the development of diarrhea during nal-IRI/FL treatment (Table II). A previous study has revealed a significantly higher incidence of grade ≥3 diarrhea in patients with UGT1A1*6 or *28 homozygosity/compound heterozygosity than in those with single heterozygosity/wild type at 33.3% versus 9.5%, respectively (16). This study surveyed the incidence of grade ≥1 diarrhea, and we excluded seven cases because of regular loperamide administration from the start of nal-IRI/FL. This exclusion is necessary to accurately assess the adverse effects of nal-IRI/FL, but the influence of the UGT1A1 polymorphism might not have been detected due to the small sample size.

By this study, we identified a 14.7% incidence rate of grade ≥3 neutropenia during nal-IRI/FL treatment, although an incidence of 45.7% of grade ≥3 neutropenia has been reported by a Japanese phase 2 trial (10). The standard nal-IRI/FL regimen consists of 70 mg/m² nal-IRI, 200 mg/m² L-leucovorin, and 2,400 mg/m² 5-FU, and patients with UGT1A1*6 or *28 (+/+) should be prescribed nal-IRI at a 20 mg/m² dose reduction (15). However, as shown in this study, the dose and its reduction were at the attending physician’s discretion rather than the standard reduction rule (Table I). Thus, the lower incidence of grade ≥3 neutropenia found in this study might be associated with a reduced treatment dose. The incidence of grade ≥3 neutropenia in a previous study has been reported to be significantly higher in patients with UGT1A1*6 or *28 homozygosity/compound heterozygosity than in those with single heterozygosity/wild type; occurrence rates are 73.3 and 38.1%, respectively (16). Patients with UGT1A1*6 and *28 genotypes have demonstrated decreased metabolic activity (12,13). In particular, the phase 1 studies of nal-IRI-based therapy have indicated that patients with UGT1A1*6/*28 show a higher area under the plasma concentration–time curve of SN-38 than that of the other patients and experience grade 4 neutropenia (17,18). The neutropenia is associated with unencapsulated SN-38 maximum plasma concentration (C_max), and Asians show higher unencapsulated SN-38 C_max compared with Caucasians in a pharmacokinetic analysis (19). Nal-IRI is less toxic than conventional irinotecan in vivo (20). The grade ≥3 neutropenia and leukopenia in the UGT1A1*6 or *28 (+/+) group exhibited a high incidence (Figure 2), indicating that unencapsulated SN-38 C_max might have been increased in this group.

The results of this study revealed that increased AST value before the nal-IRI/FL therapy and pancreatic head cancer were significantly associated with leukopenia (Table III). A previous study on FOLFIRINOX, which includes conventional irinotecan and provides similar prognoses to nal-IRI/FL as second-line therapy (21), has revealed the significant risk factor of grade 4 neutropenia: tumor location in the head of the pancreas (odds ratio=1.96) (22). Biliary obstruction due to a tumor in the head of the pancreas increases bilirubin levels, and total bilirubin higher than the upper limit of normal range is also a risk factor for grade 4 neutropenia (23). We believe that biliary obstruction and decreased liver function before nal-IRI/FL initiation might be associated with developing leukopenia through elevated unencapsulated SN-38 C_max. There are many reports on predictors of treatment-response by nal-IRI-based therapy. In particular, CAR, NLR, GPS, total bilirubin, carcinomatosis, and previous treatment with irinotecan have been significantly associated with overall survival (3-8). However, no study has reported the risk factors for the adverse events during nal-IRI/FL treatment based on the analysis of factors other than the UGT1A1 polymorphism in clinical settings. The results of this study indicated that UGT1A1*6 or *28 (+/+), high AST value before the nal-IRI/FL therapy and pancreatic head cancer may be risk factors for myelosuppression during nal-IRI/FL treatment (Table III, Table IV).

This study has several limitations. This is a retrospective study based on medical records. Further, we did not measure the plasma concentration of irinotecan and its active metabolite SN-38. Therefore, a prospective study with a larger number of patients needs to clarify the influence of factors affecting the pharmacokinetics of nal-IRI.

Patients with UGT1A1*6 or *28 (+/+) should be especially concerned about neutropenia and leukopenia during nal-IRI/FL treatment. Further, high AST value and pancreatic head cancer may be risk factors for leukopenia during nal-IRI/FL treatment.

The Authors declare no competing interests in relation to this study.

Conceptualization: T.I. and M.S.; Data curation: T.I., H.E., S.H., K.K., and S.S.; Formal analysis: T.I. and M.S.; Supervision: R.A., M.K., and K.M.; Writing – original draft: T.I., M.S., and K.M.; Writing – review & editing: R.A. and M.K.; All Authors read and approved the final manuscript.

The Authors would like to thank Enago (www.enago.jp) for the English language review.

No external funding was received for this study.