Advertisement

Impact of first-line FOLFIRINOX versus Gemcitabine/Nab-Paclitaxel chemotherapy on survival in advanced pancreatic cancer: Evidence from the prospective international multicentre PURPLE pancreatic cancer registry

  • Jordan Santucci
    Affiliations
    Walter & Eliza Hall Institute of Medical Research, VIC, Australia

    The Department of Medicine, St Vincent's Hospital Melbourne, VIC, Australia

    University of Melbourne, Faculty of Medicine, Dentistry and Health Sciences, VIC, Australia
    Search for articles by this author
  • Mark Tacey
    Affiliations
    The Department of Medical Oncology, Northern Health, VIC, Australia
    Search for articles by this author
  • Benjamin Thomson
    Affiliations
    University of Melbourne, Faculty of Medicine, Dentistry and Health Sciences, VIC, Australia

    The Department of Surgery, The Royal Melbourne Hospital, VIC, Australia
    Search for articles by this author
  • Michael Michael
    Affiliations
    The Department of Medical Oncology, Peter MacCallum Cancer Centre, VIC, Australia
    Search for articles by this author
  • Rachel Wong
    Affiliations
    The Department of Medical Oncology, Eastern Health, VIC, Australia

    The Department of Medical Oncology, Epworth Health, VIC, Australia

    Eastern Health Clinical School, Monash University, Melbourne, VIC, Australia
    Search for articles by this author
  • Julia Shapiro
    Affiliations
    The Department of Medicine, Alfred Hospital, VIC, Australia

    Cabrini Haematology and Oncology Centre, Cabrini Health, VIC, Australia
    Search for articles by this author
  • Ross Jennens
    Affiliations
    The Department of Medical Oncology, Peter MacCallum Cancer Centre, VIC, Australia

    The Department of Medical Oncology, Epworth Health, VIC, Australia
    Search for articles by this author
  • Kate Clarke
    Affiliations
    The Department of Medical Oncology, Wellington Hospital, New Zealand
    Search for articles by this author
  • Sharon Pattison
    Affiliations
    The Department of Medical Oncology, Dunedin University Hospital, New Zealand
    Search for articles by this author
  • Matthew Burge
    Affiliations
    The Department of Medical Oncology, Royal Brisbane and Women's Hospital, QLD, Australia
    Search for articles by this author
  • Rob Zielinski
    Affiliations
    The Department of Medical Oncology, Orange and Dubbo Base Hospitals, NSW, Australia
    Search for articles by this author
  • Mehrdad Nikfarjam
    Affiliations
    The Department of Surgery, Austin Health, VIC, Australia
    Search for articles by this author
  • Sumitra Ananda
    Affiliations
    Walter & Eliza Hall Institute of Medical Research, VIC, Australia

    The Department of Medical Oncology, Peter MacCallum Cancer Centre, VIC, Australia

    The Department of Medical Oncology, Epworth Health, VIC, Australia

    The Department of Medical Oncology, Western Health, VIC, Australia
    Search for articles by this author
  • Lara Lipton
    Affiliations
    Walter & Eliza Hall Institute of Medical Research, VIC, Australia

    The Department of Medical Oncology, Peter MacCallum Cancer Centre, VIC, Australia

    The Department of Medical Oncology, Western Health, VIC, Australia

    Cabrini Haematology and Oncology Centre, Cabrini Health, VIC, Australia
    Search for articles by this author
  • Peter Gibbs
    Affiliations
    Walter & Eliza Hall Institute of Medical Research, VIC, Australia

    University of Melbourne, Faculty of Medicine, Dentistry and Health Sciences, VIC, Australia

    The Department of Surgery, The Royal Melbourne Hospital, VIC, Australia

    The Department of Medical Oncology, Western Health, VIC, Australia
    Search for articles by this author
  • Belinda Lee
    Correspondence
    Corresponding author: Walter & Eliza Hall Institute of Medical Research, VIC, Australia. Tel.: +61 432 860 813.
    Affiliations
    Walter & Eliza Hall Institute of Medical Research, VIC, Australia

    University of Melbourne, Faculty of Medicine, Dentistry and Health Sciences, VIC, Australia

    The Department of Surgery, The Royal Melbourne Hospital, VIC, Australia

    The Department of Medical Oncology, Peter MacCallum Cancer Centre, VIC, Australia

    The Department of Medical Oncology, Northern Health, VIC, Australia
    Search for articles by this author
Published:August 18, 2022DOI:https://doi.org/10.1016/j.ejca.2022.06.042

      Highlights

      • Analysis of optimal therapeutic sequencing in advanced pancreatic cancer (n = 615).
      • 1st-line FFX efficacy comparable to GEM-P (mOS 11.3 versus 12.3 months P = 0.37).
      • Longer mOS with 2nd-line chemotherapy versus 1st-line only (17.4 versus 8.2 months P < 0.001).
      • 2nd-line 5FU versus GEM-based treatment mOS similar (17.3 versus 15.9 months P = 0.92).

      Abstract

      Background

      First-line palliative chemotherapy regimens in advanced pancreatic ductal adenocarcinoma (PDAC) have not been compared in head-to-head phase III randomised controlled trials (RCT). Data on optimum first-line treatment and subsequent sequencing is lacking.

      Objective

      To compare overall survival (OS) between first-line treatment regimens in a real-world population to determine if an optimal therapeutic sequence is associated with survival benefit.

      Methods

      A retrospective analysis of prospectively collated data from the Australasian PURPLE pancreatic cancer registry was undertaken.

      Findings

      From 2016 to 2020, of 1551 pancreatic cancer patients, 615 received palliative-intent chemotherapy. Patients with early-stage resected disease without recurrence (n = 369), radiotherapy alone (n = 43), received supportive care alone (n = 458) or had less than 3 months follow-up (n = 66) were excluded. Median OS was comparable between patients receiving first-line Gemcitabine/Nab-Paclitaxel (n = 376) and those receiving FOLFIRINOX (n = 73) (11.3 versus 12.3 months, P = 0.37), with 38% proceeding to second-line chemotherapy which was associated with longer mOS compared to first-line treatment alone (17.4 versus 8.2 months, P < 0.001). With second-line treatment following prior FOLFIRINOX (n = 29) or Gemcitabine/Nab-Paclitaxel (n = 101), mOS did not differ significantly (17.3 versus 15.9 months, P = 0.92), respectively, whilst median progression-free survival was longer with prior FOLFIRINOX (5.2 versus 2.9 months, P = 0.03).

      Conclusion

      There was no significant difference in overall survival between either first-line chemotherapy choice, despite patients receiving FOLFIRINOX being younger, fitter, and more likely to have localised disease. However, FOLFIRINOX was associated with delayed progression. In the absence of phase III RCT data, clinicians should be comfortable using either Gemcitabine/Nab-Paclitaxel or FOLFIRINOX as first-line therapy in advanced PDAC.

      Keywords

      1. Introduction

      Pancreatic cancer is projected to become the second leading cause of cancer-related death by 2030, with the global burden doubling by 2060. In the last 25 years, the incidence, prevalence and mortality of pancreatic ductal adenocarcinoma (PDAC) has increased by 55%, 63% and 53% [
      • Lippi G.
      • Mattiuzzi C.
      The global burden of pancreatic cancer.
      ]. These statistics reflect the late stage of detection in the majority of patients (∼80%) who present with either locally-advanced unresectable or metastatic disease and an urgent need to optimised and improve outcomes [
      • Kamisawa T.
      • Wood L.D.
      • Itoi T.
      • Takaori K.
      Pancreatic cancer.
      ]. Combination chemotherapy regimens are standard-of-care in advanced disease but are associated with substantial toxicity [
      • Von Hoff D.D.
      • Ervin T.
      • Arena F.P.
      • Chiorean E.G.
      • Infante J.
      • Moore M.
      • et al.
      Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine.
      ,
      • Conroy T.
      • Desseigne F.
      • Ychou M.
      • Bouche O.
      • Guimbaud R.
      • Becouarn Y.
      • et al.
      FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer.
      ]. Median overall survival (mOS) remains dismal at less than 12 months for metastatic PDAC (mPDAC) [
      • Hall B.R.
      • Cannon A.
      • Atri P.
      • Wichman C.S.
      • Smith L.M.
      • Ganti A.K.
      • et al.
      Advanced pancreatic cancer: a meta-analysis of clinical trials over thirty years.
      ]. Predictive biomarkers to guide optimal selection of combination chemotherapy in advanced PDAC are lacking. This paper seeks to explore comparative outcomes in a real-world setting. Deciphering patterns of response and optimal first-line choice of chemotherapy in this poor prognostic cancer is warranted.
      Current National Comprehensive Cancer Network (NCCN) guidelines recommend either chemotherapy combinations, gemcitabine with nab-paclitaxel (Gem/Nab-P) or FOLFIRINOX (folinic acid, 5FU, irinotecan, oxaliplatin), as first-line palliative treatment over gemc-itabine monotherapy [
      • Chiorean E.G.
      • Cheung W.Y.
      • Giordano G.
      • Kim G.
      • Al-Batran S.E.
      Real-world comparative effectiveness of nab-paclitaxel plus gemcitabine versus FOLFIRINOX in advanced pancreatic cancer: a systematic review.
      ]. The PRODIGE-4/ACCORD-11 trial randomised 342 patients with mPDAC to receive either first-line FOLFIRINOX or gemcitabine, illustrating convincing benefit from FOLFIRINOX (mOS 11.1 versus 6.8 months, P < 0.001) [
      • Conroy T.
      • Desseigne F.
      • Ychou M.
      • Bouche O.
      • Guimbaud R.
      • Becouarn Y.
      • et al.
      FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer.
      ]. In the MPACT trial involving 861 mPDAC patients, significant survival advantage (mOS 8.5 versus 6.7 months, P < 0.001) was also reported with Gem/Nab-P compared to gemcitabine monotherapy. Although the magnitude of benefit was less than observed with FOLFIRINOX in PRODIGE-4/ACCORD-11, an overall survival beyond three years was demonstrated in some patients in the advanced disease setting for the first time in this illness, which no prior gemcitabine-based studies have previously reported [
      • Von Hoff D.D.
      • Ervin T.
      • Arena F.P.
      • Chiorean E.G.
      • Infante J.
      • Moore M.
      • et al.
      Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine.
      ]. FOLFIRINOX is often chosen for younger, fitter patients since this triplet regimen is associated with greater risk of toxicity, whereas Gem/Nab-P is more likely to be considered when age or performance status is more limiting [
      • Lambert A.
      • Schwarz L.
      • Borbath I.
      • Henry A.
      • Van Laethem J.L.
      • Malka D.
      • et al.
      An update on treatment options for pancreatic adenocarcinoma.
      ,
      • Ducreux M.
      • Cuhna A.S.
      • Caramella C.
      • Hollebecque A.
      • Burtin P.
      • Goere D.
      • et al.
      Cancer of the pancreas: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up.
      ,
      • Sohal D.P.S.
      • Kennedy E.B.
      • Khorana A.
      • Copur M.S.
      • Crane C.H.
      • Garrido-Laguna I.
      • et al.
      Metastatic pancreatic cancer: ASCO clinical practice guideline update.
      ,
      National Comprehensive Cancer Network
      Clinical practice guidelines in oncology: pancreatic adenocarcinoma. Version 1.
      ].
      In the locally advanced unresectable disease setting, there is no consensus on chemotherapeutic management. Based on studies including LAP07, an international, open-label, phase III randomised trial comparing chemoradiotherapy and chemotherapy alone in patients with locally advanced PDAC [
      • Hammel P.
      • Huguet F.
      • van Laethem J.L.
      • Goldstein D.
      • Glimelius B.
      • Artru P.
      • et al.
      Effect of chemoradiotherapy vs chemotherapy on survival in patients with locally advanced pancreatic cancer controlled after 4 months of gemcitabine with or without Erlotinib: the LAP07 randomized clinical trial.
      ], the European Society of Medical Oncology (ESMO) recommends 6 months of chemotherapy with gemcitabine alone as the standard-of-care [
      • Ducreux M.
      • Cuhna A.S.
      • Caramella C.
      • Hollebecque A.
      • Burtin P.
      • Goere D.
      • et al.
      Cancer of the pancreas: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up.
      ,
      • Hammel P.
      • Huguet F.
      • van Laethem J.L.
      • Goldstein D.
      • Glimelius B.
      • Artru P.
      • et al.
      Effect of chemoradiotherapy vs chemotherapy on survival in patients with locally advanced pancreatic cancer controlled after 4 months of gemcitabine with or without Erlotinib: the LAP07 randomized clinical trial.
      ]. However, in a subgroup of patients, adding capecitabine with radiotherapy may be considered for those without systemic progression. By contrast, NCCN extrapolates from findings in metastatic disease to recommend either FOLFIRINOX or Gem/Nab-P in locally-advanced PDAC, with consideration of chemoradiation after 3-months of induction chemotherapy [
      • Tempero M.A.
      • Malafa M.P.
      • Al-Hawary M.
      • Asbun H.
      • Bain A.
      • Behrman S.W.
      • et al.
      Pancreatic adenocarcinoma, version 2.2017, NCCN clinical practice guidelines in oncology.
      ]. Additional data are required to evaluate chemotherapy combinations in locally-advanced PDAC.
      In this retrospective analysis of prospectively collected clinical registry data, we report outcomes in advanced PDAC patients treated with palliative chemotherapy. The primary aim was to compare survival outcomes between first-line treatment regimens and subsequent therapy in a real-world population to determine if an optimal therapeutic sequence is associated with survival benefit. Secondarily, this study aims to describe treatment patterns in a large, representative cohort of patients with advanced pancreatic cancer treated across Australia and New Zealand.

      2. Methods

      2.1 Study population and description

      This observational cohort analysis utilised the Pancreatic cancer: Understanding Routine Practice & Lifting End results (PURPLE) translational registry to identify consecutive patients with locally advanced unresectable, recurrent, or de novo metastatic PDAC, who received palliative-intent chemotherapy between 2016 and 2020. Data on tumour (T), nodal (N), and metastases(M) (TNM) histopathological classification was not included as most locally advanced/metastatic patients did not undergo resections. Chemotherapy was selected at treat-ing physician discretion per standard-of-care guidelines. Human Research Ethics Committees at each of the 14 participating institutions approved the study. Patients were excluded from this study if they received first-line palliative-intent radiotherapy, only received best supportive care or if there was insufficient treatment information recorded.
      The primary endpoint, overall survival (OS), was defined as the time between the date of diagnosis of advanced unresectable PDAC to death. PFS was defined as the time from commencement of first-line treatment to disease progression or death.

      2.2 Statistical analysis

      Clinico-pathological variables were compared by Chi-square tests and the Mann–Whitney (rank sum) tests for categorical and continuous variables, respectively. Survival estimates were calculated by Kaplan–Meier method with log-rank test for survival comparisons. Patients were censored at last follow-up. Hazard ratios were estimated by univariate Cox regression. Variables with univariate P values <0.20 were considered in backward stepwise multivariate Cox regression to adjust for potential confounders. P values <0.05 were considered statistically significant. Statistical analyses were performed with SPSS (Macintosh v26).

      3. Results

      3.1 Patient and treatment details

      At the time of data analysis, 1551 patients with PDAC from 14 sites were registered within the PURPLE registry. Of these, 615 satisfied inclusion criteria of locally advanced unresectable or metastatic pancreatic cancer receiving first-line palliative combination chemotherapy (Fig. 1). Patients who received first-line palliative-intent radiotherapy (n = 43), received best supportive care alone (n = 458), or had less than 3 months of therapeutic follow up data (n = 66) were excluded. From the final cohort of 615 patients, 197 (32%) presented with locally advanced or borderline-resectable disease that was not resected, 320 (52%) presented with de novo mPDAC, and 98 (16%) had recurrence following resection of localised disease. Of those included post-resection, the site of recurrence was local for 15 (15%) patients, distant (lymph nodes or metastasis) for 58 (59%), local and distant for 13 (13%), and unknown in 12 (12%).
      Fig. 1
      Fig. 1Flow diagram representing the selection of patients for this study after screening for inclusion from the available PURPLE database then application of exclusion criteria. Abbreviations: PDAC = pancreatic ductal adenocarcinoma. FOLFIRINOX = Folinic acid, 5FU, irinotecan, oxaliplatin. 5FU = 5-fluorouracil.
      The frequencies of first- and second-line palliative chemotherapies received are summarised in Fig. 2. Patients receiving first-line Gem/Nab-P (n = 376) were older (median 67 versus 59 years, P < 0.001), had higher Charlson Comorbidity Index (P = 0.002), and had poorer performance status (ECOG ≤ 1, P = 0.01; ECOG 2, P = 0.04) compared to the FOLFIRINOX group (n = 73), but otherwise shared similar baseline characteristics (Table 1). Similar proportions of the Gem/Nab-P and FOLFIRINOX groups received second-line chemotherapy (37% versus 44% P = 0.29, respectively). Gemcitabine was used as first-line treatment in 75 patients and 91 patients received one of many alternative chemotherapeutic regimens included for descriptive purposes but not for survival analysis.
      Fig. 2
      Fig. 2(A) Frequency of patients receiving each first-line palliative chemotherapy regimen in the entire cohort of 615 patients with advanced pancreatic ductal adenocarcinoma. (B) Frequency of patients receiving each second-line palliative chemotherapy regimen amongst the 376 patients who received first-line Gem/Nab-P. (C) Frequency of patients receiving each second-line palliative chemotherapy regimen amongst the 73 patients who received first-line FOLFIRINOX. Abbreviations: FOLFIRINOX = folinic acid (=FOL), 5-fluorouracil [5FU] (=F), irinotecan (=IRIN), oxaliplatin (=OX). Gem/Nab-P = gemcitabine, nab-paclitaxel. FOLFIRI = folinic acid, 5FU, irinotecan. FOLFOX = folinic acid, 5FU, oxaliplatin.
      Table 1Comparison of clinico-pathological characteristics of patients receiving first-line Gem/Nab-P versus those receiving first-line FOLFIRINOX.
      Clinico-pathological characteristicGem/Nab-P (n = 376)FOLFIRINOX (n = 73)P value
      Age at diagnosis, median years (IQR)67 (60–83)59 (54–65)<0.001∗
      Male sex n (%)198 (52.7)44 (60.3)0.47
      ECOG performance status at first presentation n (%)0.17
       ≤1336 (89.4)72 (98.6)(0.01)
       231 (8.2)1 (1.4)(0.04)
       37 (1.9)0
       41 (0.3)0
      Obstructive jaundice at first presentation n (%)96 (25.5)17 (23.3)0.69
      Charlson Comorbidity Index score at first presentation n (%)0.002
       0210 (55.9)56 (76.7)(0.001)
       196 (25.5)15 (20.5)(0.001)
       ≥269 (18.4)2 (2.7)
      Primary pancreatic tumour location n (%)0.90
       Unknown14 (3.7)3 (4.1)
       Body80 (21.3)14 (19.2)
       Head209 (55.6)40 (54.8)
       Tail70 (18.6)16 (21.9)
       Whole organ3 (0.8)0
      Stage at first presentation n (%)0.77
       Resectable58 (15.4)9 (12.3)
       Locally-advanced/borderline-resectable109 (29.0)23 (31.5)
       Metastatic209 (55.6)41 (56.2)
      Number of metastatic sites at onset of advanced disease n (%)0.97
       0113 (30.1)24 (32.9)
       1154 (41.0)29 (39.7)
       281 (21.5)15 (20.5)
       ≥328 (7.4)5 (6.8)
      Metastatic site at onset of advanced disease n (%)
       Liver188 (50.0)33 (45.2)0.45
       Lung59 (15.7)11 (15.1)0.89
       Peritoneum or malignant ascites39 (10.4)7 (9.6)0.84
       Bone7 (1.9)2 (2.7)0.62
       Lymph nodes70 (18.6)11 (15.1)0.47
       Other12 (3.2)5 (6.8)0.13
      Prior treatment n (%)
       Neoadjuvant chemotherapy17 (4.5)2 (2.7)0.49
       Surgical resection of primary tumour47 (12.5)10 (13.7)0.78
       Adjuvant chemotherapy45 (12.0)9 (12.3)0.93
       Biliary stent111 (29.5)23 (31.5)0.73
      First-line palliative treatment duration, median months (IQR)4.0 (0.2–29.9)3.2 (0.5–13.5)0.30∗
      Received second-line chemotherapy n (%)140 (37.2)32 (43.8)0.29
      Received third-line chemotherapy n (%)43 (11.4)8 (11.0)0.91
      Median follow-up time, months
      All patients were followed until death (maximum follow-up time = 40.5 months) except for 100 censored patients with median (IQR) follow-up of 8.7 (3.7–16.8) months.
      All patients were followed until death (maximum follow-up time = 36.8 months) except for 25 censored patients with median (IQR) follow-up of 6.3 (2.2–12.1) months.
      Number and site of metastases reported refer to metastatic burden at diagnosis of advanced disease. One patient treated with first-line Gem/Nab-P had unknown presenting ECOG performance status. Treatment duration was calculated as the time between the treatment start and end dates.
      Chi-square test was used to compare proportions between treatment groups and P values for significantly different column proportions according to the Z test are provided in brackets. Each metastatic site and prior treatment was analysed as a binary nominal variable (“yes” versus “no”), hence a Chi-square P value is provided for each.
      ∗P value for Mann–Whitney U test comparing non-parametric continuous variables between the two first-line treatment groups.
      Bolded P values are statistically significant.
      Abbreviations: ECOG = Eastern Cooperative Oncology Group. FOLFIRINOX = folinic acid, 5-fluorouracil, irinotecan, oxaliplatin. Gem/Nab-P = gemcitabine/nab-paclitaxel. IQR = interquartile range.
      a All patients were followed until death (maximum follow-up time = 40.5 months) except for 100 censored patients with median (IQR) follow-up of 8.7 (3.7–16.8) months.
      b All patients were followed until death (maximum follow-up time = 36.8 months) except for 25 censored patients with median (IQR) follow-up of 6.3 (2.2–12.1) months.

      3.2 First-line treatment efficacy

      In the entire cohort, mOS was 11.4 (95% confidence interval [CI] 10.3–12.4) months and the one-year survival rate was 36%. The mOS (11.3 versus 12.3 months, hazard ratio [HR] 1.15, 95% CI 0.85–1.57, P = 0.37), mPFS (5.7 versus 5.1 months, HR 0.92, 95% CI 0.69–1.22, P = 0.54), and ORR (21.8% versus 20.5%, P = 0.81) were not significantly different between patients receiving first-line Gem/Nab-P as compared with FOLFIRINOX, respectively (Fig. 3A and B, and Table 2). Significantly improved OS was observed with both Gem/Nab-P and FOLFIRINOX compared to first-line gemcitabine monotherapy (n = 75, mOS 7.3 [95% CI 4.8–9.8] months, P = 0.04 and P = 0.03 for Gem/Nab-P and FOLFIRINOX, respectively). On univariate and subsequent multivariate analysis (Table 2), the choice between Gem/Nab-P versus FOLFIRINOX as first-line palliative regimen in the entire cohort was not significantly associated with progression (HR 1.04, 95% CI 0.78–1.38, P = 0.81) nor OS (HR 0.99, 95% CI 0.64–1.53, P = 0.96).
      Fig. 3
      Fig. 3(A) Kaplan–Meier estimates of overall survival (OS) in advanced pancreatic ductal adenocarcinoma (PDAC) for 376 patients receiving Gem/Nab-P (median OS 11.3 [95% CI 10.3–12.3] months) versus 73 patients receiving FOLFIRINOX (median OS 12.3 [95% CI 8.3–16.2] months) for first-line palliative-intent treatment, Log rank P = 0.37. (B) Kaplan–Meier estimates of first-line progression-free survival (PFS) for all advanced PDAC patients receiving first-line palliative Gem/Nab-P (median PFS 5.7 [95% CI 5.1–6.4] months) versus FOLFIRINOX (median PFS 5.1 [95% CI 3.6–6.6] months), Log rank P = 0.54. (C) Kaplan–Meier estimates of OS for 101 advanced PDAC patients receiving first-line Gem/Nab-P and second-line 5FU-based treatment (median OS 15.9 [95% CI 13.5–18.3] months) versus 29 patients receiving FOLFIRINOX followed by gemcitabine-based treatment (median OS 17.3 [95% CI 12.8–21.8] months), Log rank P = 0.92. (D) Kaplan–Meier estimates of second-line PFS for all advanced PDAC patients receiving 5FU-based treatment as second-line therapy after first-line Gem/Nab-P (median PFS 2.9 [95% CI 2.4–3.3] months) versus patients receiving second-line gemcitabine-based treatment after first-line FOLFIRINOX (median PFS 5.2 [95% CI 2.5–8.0] months), Breslow P = 0.03. (E) Kaplan–Meier estimates of OS for 109 locally-advanced PDAC patients receiving first-line palliative Gem/Nab-P (median OS 13.2 [95% CI 11.3–15.1] months) versus 23 patients receiving FOLFIRINOX (median OS 11.4 [95% CI 4.8–17.9] months), Log rank P = 0.96. (F) Kaplan–Meier estimates of OS for 209 de novo metastatic PDAC patients receiving first-line palliative Gem/Nab-P (median OS 9.0 [95% CI 7.8–10.3 months) versus 41 patients receiving FOLFIRINOX (median OS 12.1 [95% CI 7.6–16.5] months), Log rank P = 0.24. Abbreviations: CI = confidence interval FOLFIRINOX = folinic acid (=FOL), 5-fluorouracil [5FU] (=F), irinotecan (=IRIN), oxaliplatin (=OX). Gem/Nab-P = gemcitabine, nab-paclitaxel. OS = overall survival. PDAC = pancreatic ductal adenocarcinoma. PFS = progression-free survival.
      Table 2Overall and progression-free survival analysis by clinico-pathological variables, first-line palliative regimen, and first-then-second-line treatment sequence.
      VariableUnivariate Cox regression analysisMultivariate Cox regression analysis
      HR95% CIPHR95% CIP
      Overall survival analysis by clinico-pathological variable and treatment
      Age at diagnosis1.0151.005–1.0240.003
      Sex, female versus male1.170.73–1.870.53
      ECOG status at diagnosis
       ECOG 2 versus ≤11.871.36–2.56<0.0011.571.13–2.160.01
      Obstructive jaundice
       At presentation yes versus no1.120.91–1.370.29
      Stage at presentation
       Localised versus metastatic0.760.69–0.84<0.0010.660.54–0.81<0.001
      CCI comorbidity score, ≥2 versus ≤11.100.86–1.410.44
      Primary tumour location
       Pancreatic head versus other0.940.85–1.030.17
      Number of metastatic sites at onset of advanced disease
       ≤1 versus ≥20.770.69–0.85<0.0010.610.49–0.77<0.001
      Prior neoadjuvant treatment
       Yes versus no0.930.76–1.140.51
      First-line chemotherapy
       FOLFIRINOX versus Gem/Nab-P0.870.64–1.180.360.990.64–1.530.96
      First-then-second-line treatment sequence
       SEQ1 versus SEQ21.030.65–1.610.920.850.45–1.600.61
      First-line progression-free survival analysis by clinico-pathological variable and treatment
      Age at diagnosis1.011.00–1.010.23
      Sex, female versus male0.980.70–1.380.93
      ECOG status at diagnosis
       ECOG 2 versus ≤11.401.04–1.890.03
      Obstructive jaundice
       At presentation yes versus no1.010.92–1.110.85
      Stage at presentation
       Localised versus metastatic0.840.77–0.92<0.0010.760.63–0.920.004
      CCI comorbidity score, ≥2 versus ≤10.890.71–1.130.34
      Primary tumour location
       Pancreatic head versus other0.960.88–1.040.31
      Number of metastatic sites at onset of advanced disease
       ≤1 versus ≥20.860.78–0.940.0010.800.66–0.980.03
      Prior neoadjuvant treatment
       Yes versus no1.040.87–1.250.67
      First-line chemotherapy
       FOLFIRINOX versus Gem/Nab-P1.090.82–1.450.561.040.78–1.380.81
      Second-line progression-free survival analysis by clinico-pathological variable and treatment
      Age at diagnosis1.000.98–1.010.67
      Sex, female versus male1.020.52–1.980.96
      ECOG status at diagnosis
       ECOG 2 versus ≤11.010.53–1.910.98
      Obstructive jaundice
       At presentation yes versus no1.100.94–1.280.25
      Stage at presentation
       Localised versus metastatic0.840.74–0.970.010.740.55–0.990.045
      CCI comorbidity score, ≥2 versus ≤10.900.56–1.440.66
      Primary tumour location
       Pancreatic head versus other0.980.85–1.120.75
      Number of metastatic sites at onset of advanced disease
       ≤1 versus ≥20.820.70–0.950.010.680.49–0.960.03
      Prior neoadjuvant treatment
       Yes versus no1.090.78–1.530.61
      First-then-second-line treatment sequence
       SEQ1 versus SEQ20.770.50–1.180.220.640.41–1.000.05
      The final set of variables included in the multivariate Cox regression models were based on a backward stepwise method with initial inclusion of treatment, age, sex and other variables with P < 0.20 on univariate analysis. All variables included in multivariate analyses were independent. Correlation between covariates was tested by use of a correlation matrix automatically performed along with the SPSS Cox regression function. Bolded P values are statistically significant.
      Abbreviations: CCI, Charlson Comorbidity Index; CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; HR, hazard ratio; SEQ1, first-line FOLFIRINOX then second-line gemcitabine-based treatment; SEQ2, first-line Gem/Nab-P then second-line 5FU-based treatment.
      In the subset of patients with locally-advanced unresectable disease, 23 patients received first-line palliative FOLFIRINOX and 109 received Gem/Nab-P with similar OS outcomes (mOS 11.4 versus 13.2 months, respectively; HR 1.20, 95% CI 0.70–2.06, P = 0.96) (Fig. 3E) but both combination regimens were associated with significantly longer mOS than the 8.9 (95% CI 4.2–13.7) months observed with first-line gemcitabine alone (P = 0.04 and P = 0.01, respectively). Similarly, in patients with de novo metastatic PDAC first-line FOLFIRINOX (n = 41) and Gem/Nab-P (n = 209) had comparable OS outcomes (mOS 12.1 versus 9.0 months, respectively; HR 0.93, 95% CI 0.83–1.05, P = 0.24) (Fig. 3F).

      3.3 Impact of number of lines of therapy received

      In a minority of patients who were well enough to receive second-line chemotherapy, the mOS was 17.4 months compared to 8.2 months in those who only received first-line treatment and <6 months if untreated. This suggests that some patients are able to have significant tumour control with chemotherapy (17.4 [95% CI 15.4–19.4] versus 8.2 [95% CI 7.3–9.1] months, P < 0.001).

      3.4 Efficacy of different treatment sequences using gemcitabine-based versus 5FU-based regimens

      The mOS did not differ significantly between first-line Gem/Nab-P then second-line 5FU-based regimens (n = 101) as compared with FOLFIRINOX then gemcitabine-based (n = 29) sequences (15.9 [95% CI 13.5–18.3] versus 17.3 [95% CI 12.8–21.8] months P = 0.92, respectively; Fig. 3C). However, mPFS of second-line 5FU-based treatment after first-line Gem/Nab-P was significantly shorter at 2.9 (95% CI 2.4–3.3) months compared to 5.2 (95% CI 2.5–8.0) months with second-line gemcitabine-based treatment after FOLFIRINOX (P = 0.03; Fig. 3D). Multivariate analysis (Table 2) showed no significant difference in risk of death between sequencing through FOLFIRINOX then gemcitabine-based treatment as compared with Gem/Nab-P then 5FU-based regimens (HR 0.85, 95% CI 0.45–1.60, P = 0.61). Stage at presentation and number of metastatic sites at advanced disease onset were the only two statistically significant predictors of second-line progression on multivariate analysis (Table 2). The sequence of FOLFIRINOX followed by gemcitabine-based treatment trended towards lower risk of second-line progression compared to Gem/Nab-P followed by 5FU-based treatment in multivariate analysis (HR 0.64, 95% CI 0.41–1.00), showing a trend but marginally fell short of statistical significance (P = 0.05).

      4. Discussion

      Without clinical trials comparing FOLFIRINOX and Gem/Nab-P or different treatment sequences, there is uncertainty around optimal treatment selection and sequencing strategies. Real-world studies evaluating standard-of-care options may help inform these decisions. In this registry-based analysis of advanced PDAC patients, real-world survival outcomes between first-line FOLFIRINOX and Gem/Nab-P, and the sequences and choice of second-line chemotherapy were compared. In this study, first-line FOLFIRINOX and Gem/Nab-P provided similar survival advantages over gemcitabine monotherapy. First-line FOLFIRINOX followed by gemcitabine-based treatment was associated with significantly longer mPFS of 5.2 months versus 2.9 months compared with first-line Gem/Nab-P followed by combination 5FU-based therapy (P = 0.03). However, this did not translate to significant OS improv-ement.
      The one-year survival rate of 36% and mOS of 11.4 months for the entire cohort in this study, corresponds closely with published outcomes in other population-based observational series of advanced PDAC [
      • Kieler M.
      • Unseld M.
      • Bianconi D.
      • Schindl M.
      • Kornek G.V.
      • Scheithauer W.
      • et al.
      Impact of New chemotherapy regimens on the treatment landscape and survival of locally advanced and metastatic pancreatic cancer patients.
      ,
      • Abrams T.A.
      • Meyer G.
      • Meyerhardt J.A.
      • Wolpin B.M.
      • Schrag D.
      • Fuchs C.S.
      Patterns of chemotherapy use in a U.S.-Based cohort of patients with metastatic pancreatic cancer.
      ,
      • Cartwright T.H.
      • Parisi M.
      • Espirito J.L.
      • Wilson T.W.
      • Pelletier C.
      • Patel M.
      • et al.
      Clinical outcomes with first-line chemotherapy in a large retrospective study of patients with metastatic pancreatic cancer treated in a US community oncology setting.
      ]. Compared to those receiving first-line FOLFIRINOX, patients receiving Gem/Nab-P were significantly older, had worse medical comorbidity, and were more likely to have an ECOG performance status of ≥2. These observations suggest that perceived fitness to tolerate toxicities and performance status significantly influenced physicians' and patients’ preferences between the two first-line regimens, which is consistent with the higher risk of adverse effects from FOLFIRINOX [
      • Ducreux M.
      • Cuhna A.S.
      • Caramella C.
      • Hollebecque A.
      • Burtin P.
      • Goere D.
      • et al.
      Cancer of the pancreas: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up.
      ,
      National Comprehensive Cancer Network
      Clinical practice guidelines in oncology: pancreatic adenocarcinoma. Version 1.
      ].
      Despite differences in baseline characteristics, there was no statistically significant difference in outcomes between first-line Gem/Nab-P and FOLFIRINOX reported in this study. FOLFIRINOX and Gem/Nab-P appear to have comparable efficacy in real-world observational series despite the OS differences reported in their respective clinical trials [
      • Cartwright T.H.
      • Parisi M.
      • Espirito J.L.
      • Wilson T.W.
      • Pelletier C.
      • Patel M.
      • et al.
      Clinical outcomes with first-line chemotherapy in a large retrospective study of patients with metastatic pancreatic cancer treated in a US community oncology setting.
      ,
      • Kang J.
      • Hwang I.
      • Yoo C.
      • Kim K.P.
      • Jeong J.H.
      • Chang H.M.
      • et al.
      Nab-paclitaxel plus gemcitabine versus FOLFIRINOX as the first-line chemotherapy for patients with metastatic pancreatic cancer: retrospective analysis.
      ,
      • Wang Y.
      • Camateros P.
      • Cheung W.Y.
      A real-world comparison of FOLFIRINOX, gemcitabine plus nab-paclitaxel, and gemcitabine in advanced pancreatic cancers.
      ,
      • Suker M.
      • Beumer B.R.
      • Sadot E.
      • Marthey L.
      • Faris J.E.
      • Mellon E.A.
      • et al.
      FOLFIRINOX for locally advanced pancreatic cancer: a systematic review and patient-level meta-analysis.
      ,
      • Gresham G.K.
      • Wells G.A.
      • Gill S.
      • Cameron C.
      • Jonker D.J.
      Chemotherapy regimens for advanced pancreatic cancer: a systematic review and network meta-analysis.
      ]. These findings may in part be due to patient demographic factors and more frequent dose modifications or down-grading of FOLFIRINOX to doublet combinations, which may attenuate efficacy in the real-world [
      • Wang Y.
      • Camateros P.
      • Cheung W.Y.
      A real-world comparison of FOLFIRINOX, gemcitabine plus nab-paclitaxel, and gemcitabine in advanced pancreatic cancers.
      ]. Data were not available in our study to evaluate the effect of dose modifications. Patients treated with FOLFIRINOX in our cohort were of a similar age range and ECOG performance status compared to the PRODIGE-4/ACCORD-11 trial where the median age was 61 and excluded patients >75 years or ECOG ≥ 2.
      Our study included a substantial population (32%) with locally advanced unresectable PDAC. As in metastatic disease, FOLFIRINOX and Gem/Nab-P had comparable outcomes in patients with locally advanced PDAC (mOS 11.4 versus 13.2 months, P = 0.96) and demonstrated greater efficacy compared to gemcitabine alone (mOS 8.9 months, P = 0.04 and P = 0.01, respectively). Therefore, our results support the National Comprehensive Cancer Network guidelines, which extrapolate from trials in metastatic patients to recommend first-line combination chemotherapy in locally advanced PDAC.
      The choice of sequence between FOLFIRINOX followed by gemcitabine-based treatment versus the alternative of Gem/Nab-P followed by 5FU-based combi-nation chemotherapy did not significantly impact OS with mOS of 17.3 and 15.9 months, respectively. Second-line clinical trials to date have focused on 5FU-based combinations after gemcitabine-based treatment. Our results suggest that second-line treatment with gemcitabine-based regimens after first-line FOLFIRINOX is similarly efficacious, even after multivariate analysis adjusting for confounding factors. Other studies of treatment sequencing support our findings. Vogl et al. reported in a small retrospective cohort study of 48 patients no significant difference in survival between FOLFIRINOX followed by Gem/Nab-P or vice versa (mOS 13.8 vs 13.7 months, P = 0.90) [
      • Vogl U.M.
      • Andalibi H.
      • Klaus A.
      • Vormittag L.
      • Schima W.
      • Heinrich B.
      • et al.
      Nab-paclitaxel and gemcitabine or FOLFIRINOX as first-line treatment in patients with unresectable adenocarcinoma of the pancreas: does sequence matter?.
      ]. Likewise, Glassman et al. found similar survival rates with Gem/Nab-P followed by FOLFIRI (mOS 23.0 months) versus FOLFIRINOX followed by gemcitabine-based treatment (mOS 25.5 months) [
      • Glassman D.C.
      • Palmaira R.L.
      • Covington C.M.
      • Desai A.M.
      • Ku G.Y.
      • Li J.
      • et al.
      Nanoliposomal irinotecan with fluorouracil for the treatment of advanced pancreatic cancer, a single institution experience.
      ].
      We observed similar survival rates between the different second-line 5FU-based combinations after first-line Gem/Nab-P. However, our comparative analyses were limited by relatively smaller numbers of patients in each sequence sub-group. The use of the second-line nano-liposomal variant of irinotecan with 5FU and leucovorin was not assessed in this study, as it was infrequently administered in our cohort. Its administration in Australia, whilst approved by the Australian Therapeutic Goods Administration, is not reimbursed and the choice of 5FU as a comparator rather than FOLFIRI or OFF regimens limit the interpretation of this data.
      Given the typically aggressive course of PDAC, progression is often closely followed by deterioration in performance status [
      • Gresham G.K.
      • Wells G.A.
      • Gill S.
      • Cameron C.
      • Jonker D.J.
      Chemotherapy regimens for advanced pancreatic cancer: a systematic review and network meta-analysis.
      ]. Only 38% of our cohort were considered suitable to receive second-line therapy. This highlights the clinical importance of our finding that the sequence of FOLFIRINOX then second-line gemcitabine-based treatment was associated with significantly prolonged second-line PFS as compared with second-line 5FU-based treatment after Gem/Nab-P. Encouragingly, our study demonstrates that patients who proceed to second-line therapy can achieve an improvement in median overall survival of up to 17.3 months. This real-world data provides information that can aid consultations. This incremental improvement in mOS warrants consideration for all patients fit enough to receive second-line therapy [
      • Conroy T.
      • Desseigne F.
      • Ychou M.
      • Bouche O.
      • Guimbaud R.
      • Becouarn Y.
      • et al.
      FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer.
      ,
      • Hall B.R.
      • Cannon A.
      • Atri P.
      • Wichman C.S.
      • Smith L.M.
      • Ganti A.K.
      • et al.
      Advanced pancreatic cancer: a meta-analysis of clinical trials over thirty years.
      ].
      This study has several limitations. Firstly, this was a retrospective registry analysis of non-randomised cohorts thus influenced by selection bias, most evident in expected differences in age and fitness by indication between FOLFIRINOX and Gem/Nab-P groups. Nonetheless, we attempted to adjust for confounders through multivariate Cox regression modelling. Secondly, dose modifications, number of treatment cycles, and toxicities were not compared. This is particularly pertinent to FOLFIRINOX treatment, which frequently requires dose modification potentially altering its efficacy. However, we did not observe significantly earlier treatment cessation with first-line FOLFIRINOX compared to Gem/Nab-P (median treatment duration 3.2 versus 4.0 months, P = 0.30). Thirdly, we acknowledge that there is a possible selection bias that has not been accounted for given the lack of data on time-lag between first- and second-line treatments as well as the scanning frequency used to monitor progression, which was performed at physician discretion in a real-world setting. Considering our results, further prospective randomised clinical trial studies are warranted to delineate differences in efficacy between alternative treatment sequencing approaches.
      In comparing our registry-based study with previous trial data, we also acknowledge that direct cross-study comparisons can potentially lead to misleading conclusions as this does not account for difference in study design. However, without randomised trial head-to-head data comparing FOLFIRINOX and Gem/Nab-P, this study offers insight into real-world outcomes for patients with data from the M-PACT and PRODIGE-4/ACCORD-11 clinical trials used as the best available yard stick to measure improvements in survival [
      • Conroy T.
      • Desseigne F.
      • Ychou M.
      • Bouche O.
      • Guimbaud R.
      • Becouarn Y.
      • et al.
      FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer.
      ,
      • Hall B.R.
      • Cannon A.
      • Atri P.
      • Wichman C.S.
      • Smith L.M.
      • Ganti A.K.
      • et al.
      Advanced pancreatic cancer: a meta-analysis of clinical trials over thirty years.
      ].
      Despite the limitations to our study, this analysis reports, to the best of our knowledge, prospectively collected data from one of the largest single cohort of PDAC patients treated across multiple institutions, public and private practices, and geographical regions, to provide a unique population-based perspective and description of real-world treatment patterns and outcomes in Australasia.

      5. Conclusion

      Our results suggest that first-line FOLFIRINOX and Gem/Nab-P have similar efficacy, despite patients offered FOLFIRINOX being younger and fitter. The sequences of each first-line combination regimen with best-practice second-line therapy resulted in similar OS outcomes. Head-to-head randomised clinical trials are needed to make firm conclusions regarding the optimal initial treatment and sequence of regimens for each patient subset.

      Author contributions

      Jordan Santucci is the first author. Dr Belinda Lee was the supervisor and senior author for this project and provided extensive feedback and input into the study design, data collection, analysis and manuscript, in collaboration with clinicians and researchers from each contributing site. These additional contributors include: Dr Matthew Burge, Dr Sharon Pattison, Dr Julia Shapiro, Dr Kate Clarke, Dr Rob Zielinski, Dr Mehrdad Nikfarjam, Associate Professor Benjamin Thomson, Associate Professor Michael Michael, Associate Professor Rachel Wong, Associate Professor Ananda Sumitra, Associate Professor Lara Lipton, Professor Peter Gibbs.
      BioGrid Australia provided technical support to the PURPLE Translational Registry platform from 14 health services including: Cabrini Health, Epworth Health, Austin Health, Melbourne Health, Peter MacCallum Cancer Centre, Northern Health, Eastern Health, Western Health, Wellington Hospital, Dunedin University Hospital, Royal Brisbane and Women's Hospital, and Orange, Bathurst and Dubbo Base Hospitals.

      Role of funding source

      Dr Belinda Lee is the recipient of Hemstritch Centenary Fellowship and receives funding from The Pancare Foundation. Providers of funding had no role in study design, data collection or analysis.
      This project was hosted and supported by Gibbs laboratory in the Division of Personalised Oncology, Walter and Elizabeth Hall Institute of Medical Rese-arch.

      Ethics committee approval

      Melbourne Health Human Research Ethics Committee (EC00243). Approval Number: 2016.200.

      Conflict of interest statement

      The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

      References

        • Lippi G.
        • Mattiuzzi C.
        The global burden of pancreatic cancer.
        Arch Med Sci. 2020; 16: 820-824
        • Kamisawa T.
        • Wood L.D.
        • Itoi T.
        • Takaori K.
        Pancreatic cancer.
        Lancet. 2016; 388: 73-85
        • Von Hoff D.D.
        • Ervin T.
        • Arena F.P.
        • Chiorean E.G.
        • Infante J.
        • Moore M.
        • et al.
        Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine.
        N Engl J Med. 2013; 369: 1691-1703
        • Conroy T.
        • Desseigne F.
        • Ychou M.
        • Bouche O.
        • Guimbaud R.
        • Becouarn Y.
        • et al.
        FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer.
        N Engl J Med. 2011; 364: 1817-1825
        • Hall B.R.
        • Cannon A.
        • Atri P.
        • Wichman C.S.
        • Smith L.M.
        • Ganti A.K.
        • et al.
        Advanced pancreatic cancer: a meta-analysis of clinical trials over thirty years.
        Oncotarget. 2018; 9: 19396-19405
        • Chiorean E.G.
        • Cheung W.Y.
        • Giordano G.
        • Kim G.
        • Al-Batran S.E.
        Real-world comparative effectiveness of nab-paclitaxel plus gemcitabine versus FOLFIRINOX in advanced pancreatic cancer: a systematic review.
        Ther Adv Med Oncol. 2019; 111758835919850367
        • Lambert A.
        • Schwarz L.
        • Borbath I.
        • Henry A.
        • Van Laethem J.L.
        • Malka D.
        • et al.
        An update on treatment options for pancreatic adenocarcinoma.
        Ther Adv Med Oncol. 2019; 111758835919875568
        • Ducreux M.
        • Cuhna A.S.
        • Caramella C.
        • Hollebecque A.
        • Burtin P.
        • Goere D.
        • et al.
        Cancer of the pancreas: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up.
        Ann Oncol. 2015; 26: v56-v68
        • Sohal D.P.S.
        • Kennedy E.B.
        • Khorana A.
        • Copur M.S.
        • Crane C.H.
        • Garrido-Laguna I.
        • et al.
        Metastatic pancreatic cancer: ASCO clinical practice guideline update.
        J Clin Oncol. 2018; 36: 2545-2556
        • National Comprehensive Cancer Network
        Clinical practice guidelines in oncology: pancreatic adenocarcinoma. Version 1.
        ([Internet]. [cited 2020 Feb 10] [Available from:)
        • Hammel P.
        • Huguet F.
        • van Laethem J.L.
        • Goldstein D.
        • Glimelius B.
        • Artru P.
        • et al.
        Effect of chemoradiotherapy vs chemotherapy on survival in patients with locally advanced pancreatic cancer controlled after 4 months of gemcitabine with or without Erlotinib: the LAP07 randomized clinical trial.
        JAMA. 2016; 315: 1844-1853
        • Tempero M.A.
        • Malafa M.P.
        • Al-Hawary M.
        • Asbun H.
        • Bain A.
        • Behrman S.W.
        • et al.
        Pancreatic adenocarcinoma, version 2.2017, NCCN clinical practice guidelines in oncology.
        J Natl Compr Cancer Netw. 2017; 15: 1028-1061
        • Kieler M.
        • Unseld M.
        • Bianconi D.
        • Schindl M.
        • Kornek G.V.
        • Scheithauer W.
        • et al.
        Impact of New chemotherapy regimens on the treatment landscape and survival of locally advanced and metastatic pancreatic cancer patients.
        J Clin Med. 2020; 9
        • Abrams T.A.
        • Meyer G.
        • Meyerhardt J.A.
        • Wolpin B.M.
        • Schrag D.
        • Fuchs C.S.
        Patterns of chemotherapy use in a U.S.-Based cohort of patients with metastatic pancreatic cancer.
        Oncologist. 2017; 22: 925-933
        • Cartwright T.H.
        • Parisi M.
        • Espirito J.L.
        • Wilson T.W.
        • Pelletier C.
        • Patel M.
        • et al.
        Clinical outcomes with first-line chemotherapy in a large retrospective study of patients with metastatic pancreatic cancer treated in a US community oncology setting.
        Drugs Real World Outcomes. 2018; 5: 149-159
        • Kang J.
        • Hwang I.
        • Yoo C.
        • Kim K.P.
        • Jeong J.H.
        • Chang H.M.
        • et al.
        Nab-paclitaxel plus gemcitabine versus FOLFIRINOX as the first-line chemotherapy for patients with metastatic pancreatic cancer: retrospective analysis.
        Invest N Drugs. 2018; 36: 732-741
        • Wang Y.
        • Camateros P.
        • Cheung W.Y.
        A real-world comparison of FOLFIRINOX, gemcitabine plus nab-paclitaxel, and gemcitabine in advanced pancreatic cancers.
        J Gastrointest Cancer. 2019; 50: 62-68
        • Suker M.
        • Beumer B.R.
        • Sadot E.
        • Marthey L.
        • Faris J.E.
        • Mellon E.A.
        • et al.
        FOLFIRINOX for locally advanced pancreatic cancer: a systematic review and patient-level meta-analysis.
        Lancet Oncol. 2016; 17: 801-810
        • Gresham G.K.
        • Wells G.A.
        • Gill S.
        • Cameron C.
        • Jonker D.J.
        Chemotherapy regimens for advanced pancreatic cancer: a systematic review and network meta-analysis.
        BMC Cancer. 2014; 14: 471
        • Vogl U.M.
        • Andalibi H.
        • Klaus A.
        • Vormittag L.
        • Schima W.
        • Heinrich B.
        • et al.
        Nab-paclitaxel and gemcitabine or FOLFIRINOX as first-line treatment in patients with unresectable adenocarcinoma of the pancreas: does sequence matter?.
        BMC Cancer. 2019; 19: 28
        • Glassman D.C.
        • Palmaira R.L.
        • Covington C.M.
        • Desai A.M.
        • Ku G.Y.
        • Li J.
        • et al.
        Nanoliposomal irinotecan with fluorouracil for the treatment of advanced pancreatic cancer, a single institution experience.
        BMC Cancer. 2018; 18: 693