Abstract
Background/Aim: Recent advances in chemotherapy have made significant progress in conversion surgery (CS) for unresectable pancreatic cancer (uPC). However, the success rate and efficacy of CS have not been fully demonstrated in patients with uPC treated with FOLFIRINOX (FFX) or gemcitabine plus nab-paclitaxel (GnP). Patients and Methods: We retrospectively reviewed the records of 318 patients with uPC who received FFX or GnP as first-line chemotherapy. The efficacy in the CS group, defined as undergoing complete resection after chemotherapy, was analyzed, and compared with the non-CS group; then, contributing factors to achieving CS were extracted. We also analyzed differences in the efficacy of CS between locally advanced pancreatic cancer (LAPC) and metastatic pancreatic cancer (MPC). Results: Overall, CS was achieved in 4.3% of cases, eight patients (13.3%) with LAPC and five (2.1%) with MPC. Contributing factors to CS were LAPC, no liver metastasis, CA19-9 ≤37, and chemotherapy response. After adjusting for these, overall survival was significantly better in the CS group than in the non-CS group [median of 32.9 vs. 11.3 months; adjusted hazard ratio (HR)=0.32; 95% confidence interval (CI)=0.14-0.70; p<0.01]. Median relapse-free survival duration after CS was 19.1 and 18.1 months in the LAPC-CS and MPC-CS group, respectively (p=0.84). The median post-conversion survival was 27.6 months in the entire CS group, 43.8 months in the LAPC-CS group and 21.3 months in the MPC-CS group. Conclusion: CS was achieved in 13.3% of LAPC and 2.1% of MPC cases. If possible, CS can markedly improve prognosis, even in MPC.
Pancreatic cancer is the fourth leading cause of cancer-related death in Japan and the United States and is estimated to become the second leading cause of death by 2030 in the latter (1, 2). This is due in part to the difficulty in diagnosing pancreatic cancer early. Surgical resection is the only curative treatment for pancreatic cancer, but it is estimated that 80% of patients are unresectable at the time of diagnosis (3). However, chemotherapy for advanced pancreatic cancer has achieved remarkable progress in recent years. The PRODIGE-4/ACCORD-11 trial in metastatic pancreatic cancer (MPC) showed that the fluorouracil, leucovorin, irinotecan, and oxaliplatin (FOLFIRINOX; FFX) regimen was superior to gemcitabine monotherapy in terms of response rate, progression-free survival, and overall survival (OS) (4). The MPACT study showed that the combination of gemcitabine and albumin-bound paclitaxel (nab-paclitaxel) was significantly superior to gemcitabine monotherapy in terms of response rate, progression-free survival, and OS (5). On the basis of this evidence, FFX and gemcitabine plus nab-paclitaxel (GnP) are now recommended as the standard treatment for patients with advanced pancreatic cancer who can tolerate these regimens (6-8).
Locally advanced pancreatic cancer (LAPC) was previously considered unresectable and an indication for palliative chemotherapy. However, recent advances in chemotherapy have led to increased use of a strategy that includes neoadjuvant chemotherapy (NAT) followed by conversion surgery (CS) (9). There are also increasing reports of CS for metastatic pancreatic cancer. However, there has been considerable heterogeneity among the available studies regarding the inclusion/resectability criteria, neoadjuvant regimens used, criteria for surgical exploration after NAT, extent of surgery, and as a consequence, resectability rates and survival outcomes (10). This heterogeneity can be attributed to the fact that the data of these studies include populations that vary from those assembled for CS to those that are clearly unresectable and undergoing palliative treatment. In the real world, the extent to which CS improves prognosis, the characteristics of cases in which CS is possible, and the rate at which CS can be achieved have not yet been fully elucidated.
Therefore, we investigated the characteristics and efficacy of CS cases in patients treated with GnP and FFX as palliative chemotherapy using real-world data in this multicenter retrospective study.
Patients and Methods
Study design. This study was performed as a post-hoc analysis of a multicenter retrospective study of GnP or FFX in patients with unresectable or recurrent pancreatic cancer (NAPOLEON study) conducted by specialists in medical oncology and gastroenterology in the Kyushu Region, Japan. The NAPOLEON study was conducted to the ethical guideline of the Declaration of Helsinki and was centrally approved by the Institutional review board of Saga Medical Center Koseikan (study ID 17-09-01-02), and also approved by the Institutional review boards or ethics committee of other institutions.
We reviewed the records of 318 consecutive patients with unresectable and recurrent pancreatic cancer who started FFX or GnP as first-line chemotherapy from December 2013 to March 2017 as the study group. LAPC and MPC were determined according to the criteria of each institution. Patients who underwent surgery for complete resection after the first chemotherapy induction were extracted as the CS group; CS treatment included resection of metastases and ablation of metastases in addition to primary tumor resection. Indications for CS were determined by surgeons at each institution based on pancreatic cancer guidelines (11). We then compared the CS group and patients who did not achieve CS (non-CS group) and analyzed the efficacy and contributing factors to attaining CS. We also analyzed differences in the efficacy of CS in patients with LAPC and MPC.
The FFX group consisted of patients who received the original or the modified regimen. The original FFX regimen corresponded to oxaliplatin at a dose of 85 mg/m2 intravenously (IV) over 2 h, followed by l-leucovorin at a dose of 200 mg/m2 IV over 2 h, with an additional dose 30 min later. Irinotecan at 180 mg/m2 IV over 90 min, followed by a bolus dose of fluorouracil at 400 mg/m2 IV, followed by a continuous dose of fluorouracil at 2400 mg/m2 IV over 46 h every 2 weeks (4). In contrast, modified-FFX consists of oxaliplatin at 85 mg/m2 IV over 2 h, followed by l-leucovorin at 200 mg/m2 IV over 2 h, irinotecan at 150 mg/m2 IV over 90 min, and then after 30 min, fluorouracil at 2400 mg/m2 IV over 46 h (12). GnP consisted of 125 mg/m2 nab-paclitaxel IV over 30 min followed by 1,000 mg/m2 gemcitabine IV over 30 min every 4 weeks on days 1, 8, and 15 (5).
Assessments. The primary endpoint of the study was OS. Other endpoints included the proportion of patients who achieved an objective response, disease control, time to conversion, relapse-free survival after CS, and post-conversion survival. Radiological data, including computed tomography or magnetic resonance imaging, were reviewed in accordance with the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 (13). An objective response was defined as a complete or partial response, and disease control was defined as a complete or partial response with stable disease as the best response. OS was calculated from the date of administration of first-line chemotherapy to the date of death from any cause or was censored at the final follow-up examination. Relapse-free survival after CS was calculated from the date of performing of CS to the date of relapse or death from any cause, whichever was earlier, or was censored at the final follow-up examination. Post-conversion survival was calculated from the date of performing CS to the date of death from any cause or was censored at the final follow-up examination.
Statistical analyses. The proportions and antitumor effects were compared using the Mann–Whitney test for continuous data and the χ2 test for categorical data. Survival was estimated using the Kaplan–Meier method, and comparisons of the probability of survival were performed using the log-rank test and the Cox proportional hazards model. Hazard ratios (HRs) are expressed with 95% confidence intervals (95% CIs). Differences with values of p<0.05 were considered significant. Univariate logistic regression analysis of contributing factors for CS implementation was performed on the following factors: age, sex, Eastern Cooperative Oncology Group performance status (ECOG PS), body mass index, stage, pancreatic tumor location, presence of biliary drainage, histology, liver metastasis, peritoneal dissemination, number of metastases, maximal tumor diameter, serum albumin, lactate dehydrogenase (LDH), C-reactive protein (CRP), carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA19-9), neutrophil-to-lymphocyte ratio (NLR), and primary regimen. Statistical analyses were performed using R v.3.6.3 (R Foundation for Statistical Computing, Vienna, Austria).
Results
CS versus non-CS. The median follow-up duration was 11.1 months (95% CI=10.4-12.1 months). Fifteen patients with ongoing first-line chemotherapy or patients who were untraceable were excluded. CS was achieved in eight patients (13.3%) with LAPC and five (2.1%) with MPC, with a total of 4.3% among all cases (Figure 1). Here, we retrospectively analyzed these 13 patients as the CS group and the other 290 patients as the non-CS group. The baseline characteristics for each group are shown in Table I. The CS group had more LAPC cases, fewer liver metastases, and lower CA19-9 levels than the non-CS group. The CS group had significantly better best response, response rate, and maximum rate of tumor shrinkage compared to the non-CS group (Table II). Median OS was 38.9 and 11.1 months in the CS and non-CS groups, respectively, which was significantly better in the CS group (HR=0.21; 95% CI=0.10-0.45; p<0.01) (Figure 2A).
A flow diagram of the study. FFX: FOLFIRINOX; GnP: gemcitabine plus nab-paclitaxel.
Patient baseline characteristics.
Comparison of the antitumor effects of chemotherapy in the CS versus non-CS groups.
Kaplan–Meier curves for crude (A) and adjusted (B) overall survival. Panel (B) shows curves adjusted for stage, liver metastasis, CA19-9, and chemotherapy response. CS: Conversion surgery; HR: hazard ratio; CI: confidence interval; CA19-9: carbohydrate antigen 19-9.
Next, chemotherapy responses were univariately analyzed, and the following factors were extracted (p<0.05): LAPC, no liver metastasis, CA19-9 within normal limit (≤37 U/ml), and chemotherapy response (Table III). These four factors are generally thought to be clinically meaningful and contribute to prolonged survival. So, we next compared OS of the CS and non-CS groups after adjusting for these factors. In the adjusted comparison, median OS was 32.9 months for the CS group and 11.3 months for the non-CS group, again showing a significantly better prognosis for the CS group (adjusted HR=0.32; 95% CI=0.14-0.70; p<0.01) (Figure 2B).
Univariate analysis of contributing factors for conversion surgery (CS).
LAPC vs. MPC in the CS group. Among the 13 patients in the CS group, eight cases were LAPC (LAPC-CS) and five were MPC (LAPC-CS). All of the LAPC-CS cases were adenocarcinoma, but two of the MPC-CS cases were not adenocarcinoma (1 acinar cell carcinoma and 1 anaplastic carcinoma). The median CA19-9 levels were 22 U/ml in LAPC-CS and 134 U/ml in MPC-CS, and although it was higher in MPC-CS, the number of cases was too small to examine statistically. The median time to CS was 6.5 months in LAPC-CS and 11.8 months in MPC-CS (Table I).
All patients in MPC-CS responded to chemotherapy, and the maximum tumor shrinkage rate was higher than that in LAPC-CS (Table II, Figure 3A). The median relapse-free survival after CS was 19.1 months in LAPC-CS and 18.1 months in MPC-CS, which did not represent a statistically significant difference (HR=0.87; 95% CI=0.24-3.12; p=0.84) (Figure 3B).
(A) Waterfall plot for changes in target tumor size after chemotherapy. (B) Kaplan–Meier curves for relapse-free survival and (C) post-conversion survival. LAPC: Locally advanced pancreatic cancer; MPC: metastatic pancreatic cancer; CS: conversion surgery; HR: hazard ratio; CI: confidence interval.
The median post-conversion survival was 27.6 months in the entire CS group, 43.8 months in LAPC-CS and 21.3 months in MPC-CS. There was no statistically significant difference in OS after CS between LAPC-CS and MPC-CS (HR=0.75; 95% CI=0.16-3.40; p=0.71) (Figure 3C). Characteristics of the CS group are shown in Table IV.
Characteristics of the conversion surgery (CS) group.
Discussion
This study showed that CS after FFX or GnP treatment could improve the prognosis of patients with advanced pancreatic cancer. We also found that CS was achieved in 13.3% of inoperable LAPC and 2.1% of inoperable MPC cases in the real world.
Even in the era of FFX, the reported CS rates for advanced pancreatic cancer range from 6.3% to 78% (14-18). The reasons for this may be that most of them are limited to retrospective reports of a small number of cases, selection bias, and differences in CS indications. In recent years, several large studies have been reported: Hackert et al. reported on 575 chemotherapy-treated advanced pancreatic cancers; CS was performed in 292 patients (50.8%) and the median OS was 15.3 months after CS, compared with 8.5 months for non-resected patients (16). Gemenetzis et al. reported on 415 cases of LAPC; 84 (20%) CS cases were characterized by good PS, small tumor size, and low CA19-9 levels; the median OS was 35.3 months after CS versus 16.3 months for non-resection (17). Maggino et al. reported a prospective study in 2019 on an intention-to-treat basis of 680 patients, 267 (39.3%) of whom were borderline resectable and 413 (60.7%) who had LAPC; 92.9% (n=570) received chemotherapy with overall resection. The overall resection rate was 15.1%, and the resection rate for LAPC was 9% (18). The rate of CS varies among studies, and at least a part of the unresectable advanced pancreatic cancer population is CS-capable, and the rate of such patients is expected to increase with the progress of chemotherapy.
In this study, stage, low CA19-9 level, no liver metastasis, and response to chemotherapy were associated with the performance of CS. These factors are also prognostic factors and may be biased; the prognosis of the CS group was still favorable even after adjusting for these factors. CS therapy might contribute to improved prognosis if the response to chemotherapy is successful.
It is noteworthy that in this study, MPC-CS had the same relapse-free survival rate and post-CS survival rate as LAPC-CS, even though it was a study with a small number of patients. MPC-CS is a “super responder” to primary chemotherapy, but similar results were found when adjusting for the factors mentioned above, including treatment effect, suggesting that CS contributes to improved prognosis. Although CS is more difficult in MPC than in LAPC, cases in which CS is possible may have a prognosis similar to that of LAPC. Additionally, the achievement rate of CS has not been fully demonstrated in real-world patients who were diagnosed with unresectable pancreatic cancer by surgeon or physician and treated with FFX or GnP. In this study, CS was achieved in 13.3% of LAPC and 2.1% of MPC, suggesting that CS is a real-world treatment option.
The limitations of this study are that it was retrospective, had a small sample size in the CS group, and was lacking clear criteria for CS. The duration of chemotherapy treatment up to CS is something that should be examined in future work. In 2020, nanoliposomal irinotecan plus 5-fluorouracil and leucovorin (Nal-IRI+5-FU/LV) was introduced as second-line therapy for unresectable pancreatic cancer (19). Otsu et al. reported that Nal-IRI+5-FU/LV was better tolerated than FOLFIRINOX, although its prognosis was comparable to that of FOLFIRINOX (20). In addition, NAPOLI 3 reported that liposomal irinotecan + 5-FU + LV + oxaliplatin (NALIRIFOX) as first-line therapy showed an improvement in OS and PFS compared to GnP in patients with metastatic pancreatic cancer (21). With the advent of these newer regimens, CS rates may increase in the future. In this study, the median duration of chemotherapy treatment to CS was 6.6 months, and all patients in the MPC-CS group had a duration of chemotherapy treatment of 6 months or longer. A multicenter study by The Japanese Society of Hepato-Biliary Pancreatic Surgery reported that patients who received chemotherapy for more than 6 months had a significantly better OS (22). Despite the lack of clear criteria at this time, CS might be actively considered for patients with no signs of tumor progression after more than 6 months of chemotherapy.
Conclusion
Real-world CS achievement rates in patients with unresectable pancreatic cancer were 13.3% for LAPC and 2.1% for MPC. Patients with locally advanced disease, CA19-9 ≤37 U/ml, antitumor response, and without liver metastasis were associated with CS. Even after adjustment for these associated factors, CS was associated with a better prognosis than non-CS. The rate leading to CS was higher in LAPC than in MPC, but the prognosis of patients with MPC who underwent CS was comparable to that of those with LAPC. If possible, CS could contribute to improving the prognosis of patients with unresectable pancreatic cancer.
Acknowledgements
We thank all of the patients, their families and all of the investigators at the 14 institutions in the NAPOLEON study. We would like to give thanks for the cooperation of the Fukuoka Medical Oncology Group - Kyushu Yamaguchi Total Oncology Group (FMOG-KYTOG) and the Saga Study Group of Liver Disease (SASLD). We are indebted to Dr. Yoshinobu Okabe of the Kurume University Hospital, Dr. Yasunori Kawaguchi of the Saga Medical Centre Koseikan, and Dr. Masato Uenomachi of the Hamanomachi Hospital for their assistance in data collection and discussion. We thank James P. Mahaffey, PhD, from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.
Footnotes
Authors’ Contributions
YI, TO, NU, TM, KM and TsS worked on the study concepts. YI, TO, MS, MF, AM, NU, TM, KM, and TsS contributed to the study design. Data acquisition was performed by YI, TO, FK, YU, JN, AK, SO, SA, MF, AM, YS, SU, HT, TH, TaS, KN, NU, TM, KM and TsS. Quality control of data and algorithms was performed by YI, TO, MS, KM and TsS. YI, TO, MS, TM, KM and TsS performed data analysis and interpretation. YI, TO, MS, KM and TsS conducted statistical analyses. The first draft of the manuscript was written by YI, TO, MS, KM and TsS. All Authors commented on previous versions of the manuscript. All Authors read and approved the final manuscript.
Conflicts of Interest
A.M. received a personal fee from Eli Lilly Japan, Daiichi-Sankyo pharmaceutical, Taiho pharmaceutical, Bristol Myers Squibb, and Ono pharmaceutical. The other authors have no competing interests or financial disclosures to declare.
- Received January 10, 2023.
- Revision received January 21, 2023.
- Accepted January 23, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.