Abstract
Background/Aim: Nanoliposomal irinotecan with 5-fluorouracil and L-leucovorin (nal-IRI/FL) is the standard regimen for metastatic pancreatic cancer, but there are no reports on prediction of early discontinuation. In this study, we investigated predictive factors of early discontinuation of nal-IRI/FL. Patients and Methods: The study included 36 patients who received nal-IRI/FL at Wakayama Medical University Hospital between June 2021 and May 2022. Those with time-to-treatment failure (TTF) ≤28 days were defined as the early discontinuation group (group ED), and those with TTF >28 days were placed in the continuation group (group C). Laboratory data were collected just before and every 14 days after initiation of nal-IRI/FL treatment. Results: There were six patients (16.7%) in group ED and 30 patients (83.3%) in group C. The lymphocyte×albumin (LA) score before therapy was significantly lower in group ED (p=0.005). In receiver operating characteristic analysis, pre-treatment LA was the best predictor for early discontinuation, with a cutoff value of 4,142 (sensitivity: 1.00, specificity: 0.77, p=0.004). In group C, LA was significantly lower at 28 days before nal-IRI/FL treatment failure compared to the value before the start of therapy [median with range: 3,299 (1,478-6,994) vs. 4,304 (2,085-8,085), p=0.006]. Conclusion: The LA score is a useful marker for evaluating treatment continuity, and especially early discontinuation, of nal-IRI/FL in patients with pancreatic cancer.
Pancreatic cancer is an increasingly common cause of mortality with a 5-year survival rate of about 10%, and is becoming a global burden (1). The poor prognosis of pancreatic cancer is due to limited treatment options (2, 3). A recent global phase 3 trial showed that addition of nanoliposomal irinotecan (nal-IRI) to 5-fluorouracil (5-FU) and leucovorin for patients with metastatic pancreatic cancer after gemcitabine-based therapy increased the median survival time from 4.2 to 6.1 months (4). Based on clinical data, nal-IRI with 5-FU and L-leucovorin (nal-IRI/FL) is now the standard second-line treatment after a gemcitabine-based regimen for metastatic pancreatic cancer (2).
The efficacy and toxicity of nal-IRI-based therapy in patients with pancreatic cancer vary among individuals, and several predictive factors for outcomes have been identified (5-13). Thus, total bilirubin, carcinomatosis, and previous irinotecan treatment are significantly associated with overall survival (OS) (5). The genotype of uridine diphosphate-glucuronosyltransferase 1A1*6 or *28 (+/+), high aspartate aminotransferase before therapy, and pancreatic head cancer are related to severity of myelosuppression (13). Several particularly interesting recent studies have indicated that inflammatory and nutritional factors are also associated with OS (6-12). Thus, increases in serum systemic inflammatory markers, such as neutrophils, platelets, monocytes, and CRP, and decreases in lymphocyte count and serum albumin have been connected with disease progression (14). By combining these variables, indicators such as the CRP-to-albumin ratio (CAR), neutrophil-to-lymphocyte ratio (NLR), and Glasgow prognostic score (GPS) are used as prognostic factors.
In the clinical setting, early discontinuation of nal-IRI/FL therapy is inevitable in some patients with metastatic pancreatic cancer. The predictive factors described above may also predict early discontinuation of nal-IRI/FL, but there is currently no evidence to substantiate this hypothesis. Identification of factors that predict early discontinuation is important for planning appropriate treatment strategies. In this study, our objective was to identify predictive factors of early discontinuation of nal-IRI/FL treatment in patients with pancreatic cancer, and to examine changes in the identified factors over the treatment period.
Patients and Methods
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. 3828). Information regarding the conduct of the study was disclosed on a website, and research subjects were given the opportunity to refuse enrollment in the study. The need for informed consent was waived by the Wakayama Medical University Ethics Committee.
Patients. The subjects were 36 patients with metastatic pancreatic cancer who received nal-IRI/FL therapy at Wakayama Medical University Hospital between June 2021 and May 2022. Patient information including time-to-treatment failure (TTF) and OS were collected from electronic medical records in the hospital system. TTF was specified as the period from initiation to discontinuation of nal-IRI/FL treatment for any reason. OS was defined as the time from initiation of nal-IRI/FL to death from any cause. Early discontinuation was defined as TTF ≤28 days (i.e., two courses of nal-IRI/FL) (group ED) and TTF >28 days was defined as treatment continuation (group C).
Inflammatory and nutritional indicators. Laboratory data were collected just before and every 14 days (i.e., one course of nal-IRI/FL) after initiation of nal-IRI/FL treatment. The albumin-to-globulin ratio (AGR), CAR, lymphocyte×albumin (LA) score, lymphocyte-to-monocyte ratio (LMR), NLR, and platelet-to-lymphocyte ratio (PLR) were calculated. The albumin level, lymphocyte count, and total cholesterol were used to determine the controlling nutrition status (CONUT), and the modified CONUT (mCONUT) score was calculated from the albumin level, lymphocyte count, and hemoglobin level (15). The GPS and modified GPS (mGPS) were determined by combining CRP and albumin levels (16, 17). The formulas for calculating the Geriatric Nutritional Risk Index (GNRI), prognostic nutritional index (PNI), systemic immune-inflammation index (SII), and systemic inflammation response index (SIRI) were: GNRI=14.89×albumin (g/dl)+41.7×body mass index (kg/m2)/22 (18); PNI=10×albumin (g/dl)+0.005×lymphocyte count (19); SII=platelet (109/l)×neutrophils (109/l)/lymphocytes (109/l) (20); and SIRI=monocytes (109/l)×neutrophils (109/l)/lymphocytes (109/l) (21).
Statistical analysis. All statistical analyses were performed using JMP ver. 16.0.0 (SAS Institute Inc, Cary, NC, USA). Data are expressed as the number of patients or the median with a range. Fisher exact tests were used to examine the distribution of categorical data. Mann-Whitney U-tests were used to compare medians of continuous values between two groups, whereas paired data were compared by Wilcoxon signed rank test. Receiver operating characteristic (ROC) curves were used to determine the optimal cutoff values of predictive factors for early discontinuation of nal-IRI/FL based on maximum values for the Youden index (sensitivity+specificity−1). Time-to-event analyses were conducted using the Kaplan-Meier method and compared by log-rank test. p<0.05 was considered significant in all analyses.
Results
The demographic and clinical characteristics of the patients with pancreatic cancer who received nal-IRI/FL therapy are shown in Table I for group ED (n=6, 16.7%) and group C (n=30, 83.3%). Age, sex, and body weight did not differ significantly between the groups. The median OS was 263 days [95% confidence interval (CI)=182–not reached] in all patients (Figure 1A) and was significantly shorter in group ED [63 (37-224) vs. 357 (185–not reached) days, p=0.002] (Figure 1B). The median TTF was 94 days (95%CI=56-140) in all patients (Figure 1C) and was also significantly shorter in group ED [14 (14-28) vs. 102 (74-231) days, p<0.001] (Figure 1D). The reasons for discontinuation in group ED were disease progression (n=3), malaise (n=1), pneumonitis (n=1), and a composite of pancytopenia, acute kidney injury, and diarrhea (n=1).
Patient characteristics.
Kaplan-Meier curves for overall survival (OS) and time-to-treatment failure (TTF). (A) OS of all patients. (B) OS in early discontinuation and continuation cases. (C) TTF of all patients. (D) TTF in early discontinuation and continuation cases. Time-to-event analyses were performed using the Kaplan-Meier method. p-Values were calculated by log-rank test.
Among inflammatory and nutritional factors measured just prior to initiation of nal-IRI/FL therapy, group ED had a significantly lower lymphocyte count (p=0.036), albumin level (p=0.041), GNRI (p=0.044), LA score (p=0.005), LMR (p=0.015), and PNI (p=0.010), and significantly higher mCONUT (p=0.023) and GPS (p=0.025) scores compared to group C (Table II). In ROC analysis of early discontinuation, LA was the best predictor among all investigated factors, with a cutoff value of 4142 (area under the ROC curve: 0.872, sensitivity: 1.00, specificity: 0.77, p=0.004) (Table III). Patients with a LA score <4,142 before nal-IRI/FL treatment had significantly shorter OS [148 (60–not reached) vs. 357 (244–not reached) days, p=0.018] (Figure 2A) and shorter TTF [42 (14-74) vs. 140 (94-264) days, p<0.001] (Figure 2B) compared with those with a LA score ≥4,142.
Inflammatory and nutritional factors obtained just prior to initiation of nal-IRI/FL therapy that may be associated with early discontinuation.
Receiver operating characteristic analysis to evaluate predictive factors for early discontinuation of nal-IRI/FL.
Kaplan-Meier curves for each group divided by the lymphocyte×albumin (LA) score before treatment. (A) Overall survival. (B) Time-to-treatment failure. Time-to-event analyses were performed using the Kaplan-Meier method. p-Values were calculated by log-rank test.
The change in LA score after initiation of treatment in group C is shown in Figure 3A. LA varied over the nal-IRI/FL period, but was significantly lower in group C at 28 days before nal-IRI/FL treatment failure compared to before treatment [3,299 (1,478-6,994) vs. 4,304 (2,085-8,085), p=0.006] (Figure 3B).
Change in lymphocyte×albumin (LA) scores after initiation of treatment in continuation cases. (A) Rate of change in LA during treatment. (B) Comparison of LA scores before treatment with those 28 days before treatment failure. Rate of change in LA score was defined as LA on the day of nal-IRI/FL infusion or the day before infusion divided by that before treatment. Ten cases were excluded because of missing information for LA at 28 days before treatment failure (B). Gray lines indicate the change in LA for each patient. The box plot shows minimum, first quartile, median, third quartile, and maximum values. The central rectangle spans the first to third quartile (interquartile range). An outlier is shown as an open circle. The p-value was calculated by Wilcoxon signed rank test.
Discussion
The results of this study suggest that the LA score before initiation of nal-IRI/FL therapy provides insight into the probability of treatment continuity in patients with pancreatic cancer. Thus, LA prior to therapy may be a biomarker to determine the feasibility of nal-IRI/FL treatment.
Despite the limited treatment options in pancreatic cancer (2), 6 of 36 patients (16.7%) in the current study discontinued nal-IRI/FL treatment early. Age, sex, and body weight could not predict early discontinuation of nal-IRI/FL. The median OS of patients with early discontinuation was significantly shorter than that in continuation cases, which strongly suggests the need to predict early discontinuation of nal-IRI/FL and then switch to another therapeutic option. Among the factors examined in the study, laboratory values for eight inflammatory and nutritional factors differed significantly between early discontinuation and continuation cases. Factors, such as CAR, NLR, and GPS have been shown to affect OS in pancreatic cancer patients treated with chemotherapy, including nal-IRI-based regimen (6-12, 22), but predictors for early discontinuation of nal-IRI/FL have not been identified. The results of this study show that inflammation and nutritional status prior to nal-IRI/FL therapy is related to early treatment discontinuation.
LA was the best predictor for early discontinuation of nal-IRI/FL, with a cutoff value of 4142, and the prognosis for patients with a low LA score before treatment was significantly worse than that for those with a high score. Half of the early discontinuation cases terminated nal-IRI/FL therapy due to disease progression, and the other half stopped treatment due to adverse events. Serum systemic inflammatory markers, such as neutrophils, platelets, monocytes, and CRP, are elevated with disease progression, whereas lymphocyte count and serum albumin decrease (14). The combination of lymphocyte count and albumin level was found to be the best predictor of early discontinuation of nal-IRI/FL in this study.
LA is a recently proposed predictive marker, with lower LA scores found to be significantly related to poor OS and recurrence-free survival in patients with stage II/III rectal cancer (23). The mechanism linking the lower LA score with early discontinuation of nal-IRI/FL treatment is not clear, but we propose the following hypothesis. Lymphocytes play an important role in host cell-mediated cytotoxic immune responses to tumors, and a high density of tumor-infiltrating lymphocytes is strongly correlated with a favorable prognosis in pancreatic cancer (24). This suggests that low lymphocyte counts may be associated with a poor antitumor immune response. Since albumin levels reflect nutritional status, hypoalbuminemia may affect immunity in the tumor microenvironment. Low albumin levels are also associated with a higher risk of grade 3 or higher chemotherapy toxicity in older patients with solid tumors (25). Thus, we believe that the LA score is a useful tool to predict both early disease progression and occurrence of adverse events.
Regarding the change in LA after initiation of treatment in continuation cases, we examined whether a decrease in LA during nal-IRI/FL therapy could predict treatment failure. LA scores at 28 days before nal-IRI/FL treatment failure in these cases were significantly lower than those before treatment initiation. Interestingly, the median LA score for continuation cases was over the cutoff value of 4,142 obtained from laboratory data prior to treatment; however, the score at 28 days before treatment failure was lower than 4,142. This finding strongly indicates that monitoring the LA score during treatment can help to predict the risk for discontinuation of nal-IRI/FL therapy.
This study has some limitations, including its retrospective design based on medical records and relatively small sample size. Thus, a prospective study with a larger number of patients is needed to validate the results of the study.
Conclusion
The LA score is a useful biomarker for evaluating treatment continuity, and especially early discontinuation, of nal-IRI/FL therapy in patients with pancreatic cancer. The LA score may be useful as a key prognostic factor to plan appropriate treatment strategies for patients receiving nal-IRI/FL.
Footnotes
Authors’ Contributions
Conceptualization: T.I. and M.Suno; Data curation: T.I., M.Shintani, and A.I.; Formal analysis: T.I., M.Suno, M.Shintani, and A.I.; Supervision: R.A., M.K., and K.M.; Writing – original draft: T.I., M.Suno, and K.M.; Writing – review & editing: R.A. and M.K.; All Authors read and approved the final manuscript.
Funding
This study was not supported by any sponsor or funder.
Conflicts of Interest
The Authors have no conflicts of interest to declare in relation to this study.
- Received July 19, 2024.
- Revision received August 4, 2024.
- Accepted August 5, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).









