Second-Line Fluoropyrimidine-Based Chemotherapy in Advanced Biliary Tract Cancer: A Meta-analysis Based on Individual Patient-Level Data of Randomized Trials

Article information

J Korean Cancer Assoc. 2024;.crt.2024.652

Publication date (electronic) : 2024 October 17

doi : https://doi.org/10.4143/crt.2024.652

Jaewon Hyung¹, Minsu Kang², Ilhwan Kim³, Kyu-pyo Kim¹, Baek-Yeol Ryoo¹, Jaekyung Cheon^1,4, Hyewon Ryu⁵, Ji Sung Lee^6,7, Ji-Won Kim², In Sil Choi⁸, Jin Hyun Park⁸, Ghassan K. Abou-Alfa^9,10,11, Jin Won Kim^,2, Changhoon Yoo^,1

¹Department of Oncology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea

²Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Korea

³Division of Oncology, Department of Internal Medicine, Haeundae Paik Hospital, Inje University College of Medicine, Busan, Korea

⁴Division of Hematology and Oncology, Department of Internal Medicine, Ulsan University Hospital, University of Ulsan College of Medicine, Ulsan, Korea

⁵Division of Hematology and Oncology, Department of Internal Medicine, Chungnam National University Hospital, Chungnam National University College of Medicine, Daejeon, Korea

⁶Clinical Research Center, Asan Institute for Life Sciences, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea

⁷Department of Clinical Epidemiology and Biostatistics, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea

⁸Department of Internal Medicine, Seoul Metropolitan Government Seoul National University Boramae Medical Center, Seoul National University Colledge of Medicine, Seoul, Korea

⁹Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA

¹⁰Department of Medicine, Weill Medical College at Cornell University, New York, NY, USA

¹¹Department of Medicine, Trinity College Dublin Medical School, Dublin, Ireland

Correspondence: Jin Won Kim, Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, 82 Gumi-ro 173 Beon-gil, Bungdang-gu, Seongnam 13602, Korea Tel: 82-31-787-7053 E-mail: kimjw05@snu.ac.kr

Co-correspondence: Changhoon Yoo, Department of Oncology, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea Tel: 82-2-3010-1727 E-mail: yooc@amc.seoul.kr

*Jaewon Hyung and Minsu Kang contributed equally to this work.

Received 2024 July 16; Accepted 2024 October 15.

Abstract

Introduction

Biliary tract cancer (BTC) is an epithelial malignancy that arises from the intra- and extrahepatic bile ducts and gallbladder [1]. It is a complex disease with diverse clinical and molecular characteristics, often leading to a poor prognosis, and majority of patients are diagnosed at an advanced stage [1,2]. Systemic treatment is the primary approach for managing unresectable or metastatic BTC. In the first-line treatment setting, recent randomized phase 3 trials have shown clinical benefits of adding anti–programmed death-1 (pembrolizumab) or programmed death-ligand 1 (durvalumab) to gemcitabine plus cisplatin (GemCis) [3,4]. However, the majority of patients eventually experience disease progression on these treatments and require subsequent therapy. Based on the results of previous randomized trials, fluoropyrimidine-based regimens are recommended as second-line treatment in the global guidelines if actionable targets such as isocitrate dehydrogenase 1 (IDH1), fibroblast growth factor receptor 2 (FGFR2) gene fusion, or human epidermal growth factor receptor 2 (HER2) amplification are not identified in the tumors [5-13].

Fluorouracil and leucovorin (5-FU/LV) plus oxaliplatin (FOLFOX) is recommended based on the results of the phase 3 ABC-06 study, which demonstrated a significantly improved overall survival (OS) compared to active symptom control [14]. Liposomal irinotecan (nal-IRI) plus 5-FU/LV was investigated in the phase 2b NIFTY trial and has shown significant benefits in terms of both progression-free survival (PFS) and OS compared to 5-FU/LV alone [15,16]. In the phase 2 FIReFOX study, modified FOLFOX (mFOLFOX) was compared with modified 5-FU/LV plus irinotecan (mFOLFIRI) as second-line therapy, and there were no significant differences in PFS and OS between the two regimens [17]. Although nal-IRI plus 5-FU/LV, mFOLFOX, mFOLFIRI, and 5-FU/LV are suggested for second-line therapy in the global guidelines, there have been no studies that compared these regimens head-to-head [5,6].

Here, we conducted an individual patient-level pooled analysis of the NIFTY (NCT03524508) and FIReFOX (NCT03464968) trials to compare the efficacy outcomes of second-line 5-FU–based chemotherapy regimens after progression on first-line GemCis in patients with advanced BTC.

Materials and Methods

Results

1. Patients

Among the ITT populations of both trials, all 178 patients from the NIFTY trial and 99 patients from the FIReFOX trial were included in the current analysis after excluding 19 patients due to primary Ampulla of Vater cancer (n=6) and ECOG performance status 2 (n=13) (Fig. 1). There were no significant differences in baseline characteristics between the two trials, except for the higher proportion of patients with ECOG performance status 0 in the nal-IRI plus 5-FU/LV group (26.1%) and the 5-FU/LV group (17.8%) compared to the mFOLFOX group (6.1%) and the mFOLFIRI group (8.0%) (p=0.006) (Table 1).

Fig. 1.

Study flow. AoV, ampulla of Vater; ECOG PS, Eastern Cooperative Oncology Group performance status; mFOLFIRI, modified fluorouracil and leucovorin plus irinotecan; mFOLFOX, modified fluorouracil and leucovorin plus oxaliplatin; nal-IRI, nanoliposomal irinotecan; SNUBH, Seoul National University Bundang Hospital; 5-FU/LV, fluorouracil and leucovorin.

Table 1.

Baseline characteristics

2. Nal-IRI plus 5-FU/LV vs. mFOLFOX or mFOLFIRI

Efficacy outcomes including ORR, PFS, and OS of the four treatment groups are summarized in the S1 Table. Tumor response according to RECIST v1.1 are shown in the S2 Table. The nal-IRI plus 5-FU/LV group showed significantly better PFS and OS compared to the mFOLFOX group (pairwise log-rank p=0.01 and p=0.02, respectively) and the mFOLFIRI group (p=0.04 and p=0.03, respectively) (Fig. 2). In the analysis involving the combined mFOLFOX or mFOLFIRI group, the nal-IRI plus 5-FU/LV group showed significantly better PFS (pairwise log-rank p=0.02) and OS (p=0.02) compared to the combined mFOLFOX or mFOLFIRI group (S3 Fig.). In the multivariable analysis adjusted for the effect of potential prognostic variables and each trial, similar trends for PFS and OS benefit of nal-IRI plus 5-FU were observed when compared with the mFOLFOX group (vs. nal-IRI plus 5-FU/LV, PFS; adjusted HR 1.46 [95% confidence interval (CI), 1.00 to 2.13; p=0.05], and OS; 1.39 [95% CI, 0.93 to 2.07; p=0.11]) and the mFOLFIRI group (vs. nal-IRI plus 5-FU/LV, PFS; 1.31 [95% CI, 0.89 to 1.92; p=0.17], and OS; 1.36 [95% CI, 0.92 to 2.03; p=0.13]), although these were not statistically significant (Table 2). In comparison involving the combined mFOLFOX or mFOLFIRI group, the nal-IRI plus 5-FU/LV group showed significantly better PFS (vs. nal-IRI plus 5-FU/LV, adjusted HR 1.38 [95% CI, 1.01 to 1.89; p=0.04]), and there was a marginal association with improved OS (adjusted HR 1.37 [95% CI, 0.99 to 1.91; p=0.06]) (S4 Table).

Fig. 2.

Kaplan-Meier estimates of survival outcomes in the nanoliposomal irinotecan (nal-IRI) plus fluorouracil and leucovorin (5-FU/LV), modified fluorouracil and leucovorin plus oxaliplatin (mFOLFOX), modified fluorouracil and leucovorin plus irinotecan (mFOLFIRI), and 5-FU/LV groups. (A) Progression-free survival (PFS). (B) Overall survival (OS). CI, confidence interval.

Table 2.

Cox proportional hazards modeling with shared frailty according to treatment groups and key prognostic variables

4. Safety profiles

After excluding four patients in the 5-FU/LV group, one patient in the mFOLFOX group, and one patient in the mFOLFIRI group who did not receive any study treatment after randomization, 271 patients in the ITT population were included in the safety analysis set (Fig. 1). Most patients experienced adverse events (n=87 [98.7%] in the nal-IRI plus 5-FU/LV group; n=47 [97.9%] in the mFOLFOX group; n=47 [95.9%] in the mFOLFIRI group; and n=77 [89.5%] in the 5-FU/LV group) (Table 3). Grade 3-4 adverse events occurred most frequently in the nal-IRI plus 5-FU/LV group (n=68, 77.3%) followed by the mFOLFOX group (n=29, 60.4%), the mFOLFIRI group (n=27, 55.1%) and the 5-FU/LV monotherapy group (n=29, 33.7%). Neutropenia was the most common grade 3 or higher adverse event among those who received doublet regimens (n=21, 23.9% in the nal-IRI plus 5-FU/LV group; n=13, 27.1% in the mFOLFOX group; and n=14, 28.6% in the mFOLFIRI group), while its incidence was 1.2% (n=1) in the 5-FU/LV group. Otherwise, the most common grade 3-4 adverse events were fatigue (n=11, 12.5%) and nausea (n=5, 5/7%) in the nal-IRI plus 5-FU/LV group, anemia (n=6, 12.5%), thrombocytopenia (n=6, 12.5%) and peripheral neuropathy (n=3, 6.3%) in the mFOLFOX group, and anemia (n=10, 20.4%), thrombocytopenia (n=5, 10.2%) and fatigue (n=3, 6.1%) in the mFOLFIRI group.

Table 3.

Adverse events that occurred in > 5% of patients regardless of causality

Discussion

Here, we conducted a pooled analysis using individual patient data from two randomized trials of fluoropyrimidine-containing second-line chemotherapy regimens for patients with unresectable or metastatic BTC after progression on first-line GemCis. Previously, in the NIFTY trial, nal-IRI plus 5-FU/LV showed significantly better PFS and OS than 5-FU/LV, while the FIReFOX trial showed similar PFS and OS for mFOLFOX and mFOLFIRI [15-17]. We compared the key efficacy outcomes of nal-IRI plus 5-FU/LV, mFOLFOX, mFOLFIRI, and 5-FU/LV, which are guideline-recommended and commonly used regimens in daily practice for biomarker-unselected patients who had previously failed on prior GemCis [5,6,19-23]. To the best of our knowledge, this is the first study to compare the efficacy outcomes of chemotherapy regimens, including nal-IRI plus 5-FU/LV and mFOLFOX. As these two trials share similarities in terms of region/ethnicity (South Korea), inclusion/exclusion criteria, and study period, it may be advantageous to compare the efficacy outcomes of these regimens with reduced unintended confounding biases, such as the pattern of daily practice for BTC.

In the current analysis, the nal-IRI plus 5-FU/LV group showed significantly better PFS and OS compared to both the mFOLFOX group and the mFOLFIRI group, as well as the combined mFOLFOX and mFOLFIRI group, according to the pairwise log-rank test. Consistent trends favoring nal-IRI plus 5-FU/LV for PFS and OS were observed in multivariable analyses, which were adjusted for the effect of potential prognostic variables and trials, although statistical significance (p < 0.05) was not achieved. There has been a lack of studies comparing the clinical outcomes of nal-IRI plus 5-FU/LV with those of other widely used fluorouracil doublets, such as mFOLFOX or mFOLFIRI. Due to the small sample size, especially for the mFOLFOX and mFOLFIRI groups, and the marginal significance in the multivariable analyses for nal-IRI plus 5-FU/LV vs. mFOLFIRI or mFOLFOX, the results of this analysis are not conclusive and further investigations are necessary.

The other key focus in the current analysis was the comparison of the efficacy outcomes between mFOLFOX or mFOLFIRI, and 5-FU/LV monotherapy. The mFOLFOX, mFOLFIRI, or combined mFOLFOX and mFOLFIRI groups showed trends for better PFS compared to the 5-FU/LV group. However, there was no difference in terms of OS among these regimens, which indicates that oxaliplatin- or irinotecan-containing doublet may provide benefit in terms of PFS but have a limited impact on OS compared to 5-FU/LV monotherapy. FOLFOX has been regarded as the global standard second-line therapy following progression on GemCis based on the results of the ABC-06 trial [14]. However, in this trial, FOLFOX was compared with active supportive care, and there are no other prospective data demonstrating the impact of adding oxaliplatin to 5-FU/LV monotherapy in the second-line setting [14]. Our findings are partially in line with previous retrospective studies. In a multicenter retrospective study conducted in France and Italy, which included 525 patients, OS did not show a significant difference among fluoropyrimidine monotherapy, irinotecan-fluoropyrimidine, and platinum-fluoropyrimidine doublets [19]. Furthermore, in a retrospective study in Korea (n=321), platinum-fluoropyrimidine doublets was associated with a higher ORR than fluoropyrimidine monotherapy, while there were no significant differences in PFS and OS [22]. The remaining issue is whether nal-IRI plus 5-FU/LV is superior to 5-FU/LV in global patient populations. In contrast to the positive results of the NIFTY trial, the recent phase 2 NALIRICC trial conducted in Germany showed similar PFS and OS between nal-IRI plus 5-FU/LV and 5-FU/LV monotherapy, despite an improved ORR with nal-IRI plus 5-FU/LV compared to 5-FU/LV monotherapy (14.3% vs. 3.9%) [24]. As this randomized study included a relatively small number of patients (n=100), its findings do not definitively exclude the potential effectiveness of nal-IRI plus 5-FU/LV over 5-FU/LV monotherapy in the Western patient population. Further investigation, such as meta-analysis and real-world evidence, is required [25].

The profiles of adverse events in each regimen were in line with those in the previous studies [14,26,27]. Hematological adverse events were the most frequent, particularly in the doublet combination regimens. The incidences of grade 3-4 neutropenia (23.9%-28.6%) and anemia (9.1%-20.4%) were similar across the nal-IRI plus 5-FU/LV, mFOLFOX, and mFOLFIRI groups. In contrast, 5-FU/LV monotherapy exhibited a more favorable profile of hematological adverse events (grade 3-4 neutropenia, 1.2%; anemia, 3.5%). Among non-hematological adverse events, fatigue was the most common adverse event of any grade (30.7%-52.1%) in patients receiving doublet combination regimens. Diarrhea occurred more frequently in the nal-IRI plus 5-FU/LV group (any grade, 22.7%; grade 3-4, 4.6%), while peripheral neuropathy was more common in the mFOLFOX group (any grade, 41.7%; grade 3-4, 6.3%).

Because there is no large global phase 3 trial for head-to-head comparison among these agents, there still exists uncertainty about which regimen is the best chemotherapy option for patients who progress on first-line GemCis, if they are not biologically eligible for targeted therapy. As patients with advanced BTC who have progressed on first-line therapy are often medically fragile, the toxicity profile of each regimen should be the key consideration, along with efficacy data, when selecting fluoropyrimidine-based second-line therapy.

Our study has several limitations. This study is a analysis that exclusively included patients of East Asian ethnicity. The NIFTY and FIReFOX trials did not exclude patients with actionable genomic alterations as the studies were performed prior to the approval of targeted agents [15,17]. Although most patient characteristics were well balanced in each treatment group in this study, there was an imbalance in ECOG performance status between the two trials; however, ECOG performance status was not significantly associated with survival outcomes in the multivariable analysis, indicating limited impact on the results of the current analysis. Moreover, the sample sizes in the current analysis may be insufficient to yield sufficient statistical power for comparing each treatment regimen and subgroup analyses. Despite these limitations, the current individual patient-level pooled analysis compared the efficacy of second-line treatments based on two of the largest randomized trials, including the two active agents proven effective in previous randomized trials (FOLFOX and nal-IRI plus 5-FU/LV).

In conclusion, nal-IRI plus 5-FU/LV showed favorable survival outcomes compared to mFOLFOX, mFOLFIRI, or 5-FU/LV alone. Although nal-IRI plus 5-FU/LV may be a preferable second-line cytotoxic chemotherapy regimen for medically fit patients with advanced BTC after progression on first-line GemCis, the safety profiles of these regimens should be considered along with efficacy outcomes when selecting second-line therapy for patients without targetable genetic alterations.

Electronic Supplementary Material

Article information Continued