ResProt JMIR Res Protoc JMIR Research Protocols 1929-0748 JMIR Publications Toronto, Canada v6i3e43 28351831 10.2196/resprot.7201 Protocol Protocol Protocol for Combined Analysis of FOXFIRE, SIRFLOX, and FOXFIRE-Global Randomized Phase III Trials of Chemotherapy +/- Selective Internal Radiation Therapy as First-Line Treatment for Patients With Metastatic Colorectal Cancer Eysenbach Gunther Kennedy Andrew Helmberger Thomas Virdee Pradeep S BSc, MSc 1 http://orcid.org/0000-0002-3006-8730 Moschandreas Joanna BSc, MSc, PhD 1 http://orcid.org/0000-0002-7169-0892 Gebski Val MStat 2 http://orcid.org/0000-0001-8748-1844 Love Sharon B BSc 1 http://orcid.org/0000-0002-6695-5390 Francis E Anne BSc, D. Phil 3 http://orcid.org/0000-0003-4618-774X Wasan Harpreet S MBBS, FRCP 4 http://orcid.org/0000-0002-6268-2030 van Hazel Guy MBBS, FRACP, FRCP 5 http://orcid.org/0000-0001-9717-7993 Gibbs Peter MBBS, FRACP, MD 6 http://orcid.org/0000-0003-1423-4484 Sharma Ricky A MD, PhD 7
National Institute for Health Research University College London Hospitals Biomedical Research Centre UCL Cancer Institute University College London London, United Kingdom 44 020 3108 2212 44 020 7679 6817 ricky.sharma@oncology.ox.ac.uk
8 http://orcid.org/0000-0002-4873-9918 1 Centre for Statistics in Medicine University of Oxford Oxford United Kingdom 2 NHMRC Clinical Trials Centre University of Sydney Sydney Australia 3 Oncology Clinical Trials Office Department of Oncology University of Oxford Oxford United Kingdom 4 Imperial College Healthcare NHS Trust Hammersmith Hospital London United Kingdom 5 University of Western Australia Perth Australia 6 Western Hospital Footscray Victoria Australia 7 National Institute for Health Research University College London Hospitals Biomedical Research Centre London United Kingdom 8 Cancer Research United Kingdom–Medical Research Council Oxford Institute for Radiation Oncology University of Oxford Oxford United Kingdom Corresponding Author: Ricky A Sharma ricky.sharma@oncology.ox.ac.uk 032017 28 03 2017 6 3 e43 20 12 2016 11 1 2017 19 1 2017 21 1 2017 ©Pradeep S Virdee, Joanna Moschandreas, Val Gebski, Sharon B Love, E Anne Francis, Harpreet S Wasan, Guy van Hazel, Peter Gibbs, Ricky A Sharma. Originally published in JMIR Research Protocols (http://www.researchprotocols.org), 28.03.2017. 2017

This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be included.

 Background

In colorectal cancer (CRC), unresectable liver metastases are associated with a poor prognosis. The FOXFIRE (an open-label randomized phase III trial of 5-fluorouracil, oxaliplatin, and folinic acid +/- interventional radioembolization as first-line treatment for patients with unresectable liver-only or liver-predominant metastatic colorectal cancer), SIRFLOX (randomized comparative study of FOLFOX6m plus SIR-Spheres microspheres versus FOLFOX6m alone as first-line treatment in patients with nonresectable liver metastases from primary colorectal carcinoma), and FOXFIRE-Global (assessment of overall survival of FOLFOX6m plus SIR-Spheres microspheres versus FOLFOX6m alone as first-line treatment in patients with nonresectable liver metastases from primary colorectal carcinoma in a randomized clinical study) clinical trials were designed to evaluate the efficacy and safety of combining first-line chemotherapy with selective internal radiation therapy (SIRT) using yttrium-90 resin microspheres, also called transarterial radioembolization.

 Objective

The aim of this analysis is to prospectively combine clinical data from 3 trials to allow adequate power to evaluate the impact of chemotherapy with SIRT on overall survival.

 Methods

Eligible patients are adults with histologically confirmed CRC and unequivocal evidence of liver metastases which are not treatable by surgical resection or local ablation with curative intent at the time of study entry. Patients may also have limited extrahepatic metastases. Final analysis will take place when all participants have been followed up for a minimum of 2 years.

 Results

Efficacy and safety estimates derived using individual participant data (IPD) from SIRFLOX, FOXFIRE, and FOXFIRE-Global will be pooled using 2-stage prospective meta-analysis. Secondary outcome measures include progression-free survival (PFS), liver-specific PFS, health-related quality of life, response rate, resection rate, and adverse event profile. The large study population will facilitate comparisons of low frequency adverse events and allow for more robust safety analyses. The potential treatment benefit in those patients who present with disease confined to the liver will be investigated using 1-stage IPD meta-analysis. Efficacy will be analyzed on an intention-to-treat basis.

 Conclusions

This analysis will assess the impact of SIRT combined with chemotherapy on overall survival in the first-line treatment of metastatic CRC. If positive, the results will change the standard of care for this disease.

 Trial Registration

FOXFIRE ISRCTN Registry ISRCTN83867919; http://www.isrctn.com/ISRCTN83867919 (Archived by WebCite at http://www.webcitation.org/6oN7axrvA). SIRFLOX ClinicalTrials.gov NCT00724503; https://clinicaltrials.gov/ ct2/show/NCT00724503 (Archived by WebCite at http://www.webcitation.org/6oN7lEGbD). FOXFIRE-Global ClinicalTrials.gov NCT01721954; https://clinicaltrials.gov/ct2/show/NCT01721954 (Archived by WebCite at http://www.webcitation.org/ 6oN7vvQvG).

 colorectal neoplasms liver neoplasm metastasis radiation oncology survival analysis meta-analysis Introduction

The 5-year overall survival (OS) of metastatic colorectal cancer (mCRC) patients, who constitute 21% of all CRC diagnoses, is approximately 13% [1]. The liver is the dominant site of metastases in CRC; liver metastases are the most common cause of death for patients with CRC [2,3]. To improve outcomes in mCRC, efforts have been made to increase the proportion of patients eligible for surgical resection, which is currently 20% [4-8]. The use of down-staging neoadjuvant chemotherapy in clinical studies has suggested that 10%-20% of patients with inoperable liver disease may be converted to candidates for curative resection [9]. Furthermore, a statistical correlation between tumor response and resection rates has been found across clinical studies [10].

Among the liver-directed therapies that may control liver metastases, selective internal radiation therapy (SIRT) is one option for patients with liver-only or liver-dominant disease [11,12]. SIR-Spheres (Sirtex Medical Limited) containing the β-emitter, yttrium-90, are delivered into the arterial supply of the liver under fluoroscopic guidance. The delivery of the resin microspheres into branches of the hepatic artery, which supplies the majority of blood to liver tumors, results in selective targeting of the tumor by high-dose radiotherapy, whereas the healthy liver is supplied predominantly by the portal venous system and is therefore relatively spared from radiation treatment [12].

The FOXFIRE (an open-label randomized phase III trial of 5-fluorouracil, oxaliplatin, and folinic acid +/- interventional radioembolization as first-line treatment for patients with unresectable liver-only or liver-predominant metastatic colorectal cancer), SIRFLOX (randomized comparative study of FOLFOX6m plus SIR-Spheres microspheres versus FOLFOX6m alone as first-line treatment in patients with nonresectable liver metastases from primary colorectal carcinoma), and FOXFIRE-Global (assessment of overall survival of FOLFOX6m plus SIR-Spheres microspheres versus FOLFOX6m alone as first-line treatment in patients with nonresectable liver metastases from primary colorectal carcinoma in a randomized clinical study) clinical trials were designed to study SIRT in combination with chemotherapy, specifically the modified fluorouracil, leucovorin, and oxaliplatin (FOLFOX) regimen, compared with FOLFOX alone as first-line therapy for mCRC [13,14]. Eligibility criteria and trial designs were similar for the 3 trials so that they could be prospectively combined. The primary and some secondary endpoints of SIRFLOX have been published, representing the largest published, randomized, multicenter trial of any liver-directed therapy in patients with mCRC [15]. The results showed increased progression-free survival (PFS) in the liver with the addition of SIRT but no effect of SIRT on overall PFS [15].

The primary objective of the combined analysis is to report the planned meta-analysis (MA) of individual participant data (IPD) from the FOXFIRE, SIRFLOX, and FOXFIRE-Global studies for the primary endpoint, OS, and for secondary outcomes including PFS, liver-specific PFS, quality of life measures, response rate, resection rate, and adverse event profile. The IPD-MA will assess the treatment effects on clinical outcomes in the entire subject group and in key subgroups.

 Methods Study Design

As the results on survival benefits for each of the 3 studies are still blinded, the statistical design is a prospective MA based on randomized IPD from the FOXFIRE, SIRFLOX, and FOXFIRE-Global clinical trials.

The protocols of the FOXFIRE [13] and SIRFLOX [14] trials have been previously published. All 3 clinical trials were open-label multicenter, randomized, 2-arm trials comparing chemotherapy plus SIRT with chemotherapy alone as first-line treatment for patients with mCRC. The design of the 3 trials is summarized in Table 1 and Figure 1.

Characteristics of studies included in the meta-analysis: FOXFIRE [13], SIRFLOX [14], and FOXFIRE-Global.

FOXFIREa SIRFLOXb FOXFIRE-Globalc
Start of recruitment Nov 13, 2009 Oct 11, 2006 May 20, 2013
End of recruitment Oct 31, 2014 Apr 25, 2013 Dec 23, 2014
End of follow-up Oct 31, 2016 Apr 25, 2018 Dec 23, 2019
Number of recruiting centers 28 87 69
Primary objective Overall survival Progression-free survival Overall survival
Secondary objectives Progression-free survival Progression-free survival in the liver Toxicity and safety Tumor response rate Quality of life Liver resection rate Health costs/economics Proportion of patients receiving second line treatment Time to second line treatment Overall survival Progression-free survival in the liver Toxicity and safety Tumor response rate Quality of life Liver resection rate Hepatic and extrahepatic recurrence rate Progression-free survival Progression-free survival in the liver Toxicity and safety Tumor response rate Quality of life Liver resection rate Hepatic and extrahepatic recurrence rate Health economics
Sample size accrued 364 530 209
Accrual period November 2009–October 2014 October 2006–April 2013 May 2013–December 2014
Follow-up period November 2014–October 2016 April 2013–April 2018 December 2014–December 2019
Randomization 1:1 with minimization 1:1 with minimization 1:1 with minimization
Minimization factors Liver only versus extrahepatic metastases Extent of tumor involvement of the liver (≤25% or >25% determined by CTd scan) Planned use of biological agent (from March 2011 on) Investigational center Liver only versus extrahepatic metastases Extent of tumor involvement of the liver (≤25% or >25% determined by CT scan) and based upon the tumor involvement groupings used by Gray et al [16] Planned use of bevacizumab with chemotherapy Investigational center Liver only versus extrahepatic metastases Extent of tumor involvement of the liver (≤25% or >25% determined by CT scan) and based upon the tumor involvement groupings used by Gray et al [16] Planned use of bevacizumab with chemotherapy Investigational center
Recruiting countries/regions United Kingdom (England, Northern Ireland, Scotland, Wales) Australia, Europe, Israel, New Zealand, and United States Australia, Europe, Israel, Korea, New Zealand, Singapore, Taiwan and United States

aAn open-label randomized phase III trial of 5-fluorouracil, oxaliplatin, and folinic acid +/- interventional radioembolization as first-line treatment for patients with unresectable liver-only or liver-predominant metastatic colorectal cancer.

bRandomized comparative study of FOLFOX6m plus SIR-Spheres microspheres versus FOLFOX6m alone as first-line treatment in patients with nonresectable liver metastases from primary colorectal carcinoma.

cAssessment of overall survival of FOLFOX6m plus SIR-Spheres microspheres versus FOLFOX6m alone as first-line treatment in patients with nonresectable liver metastases from primary colorectal carcinoma in a randomized clinical study.

dComputed tomography.

Schema for the FOXFIRE, SIRFLOX, and FOXFIRE-Global trials.

 Patients and Treatment

All randomized patients from the SIRFLOX, FOXFIRE, and FOXFIRE-Global clinical trials will be included in the combined analysis.

Inclusion Criteria

Histologically confirmed CRC with liver-only or liver-dominant metastases not amenable to surgical resection, primary tumour in situ permitted (FOXFIRE); histologically confirmed CRC with liver-only or liver-dominant metastases (SIRFLOX); histologically confirmed adenocarcinoma of the colon or rectum with or without primary tumor in situ (FOXFIRE-Global)

Unequivocal and measurable computed tomography (CT) evidence of liver metastases not treatable by surgical resection or local ablation with curative intent at time of trial entry

Chemotherapy-naïve for mCRC, but previous adjuvant systemic chemotherapy for primary CRC or neoadjuvant chemoradiotherapy to the pelvis more than 6 months before recruitment is permitted (SIRFLOX, FOXFIRE-Global); eligible for systemic chemotherapy as first-line treatment for metastatic CRC (FOXFIRE)

Additional limited extrahepatic metastases in the lung or lymph nodes permitted (SIRFLOX, FOXFIRE-Global); patients are permitted to have limited extrahepatic disease (FOXFIRE)

Age ≥18 years

World Health Organization performance status of 0-1

Life expectancy >3 months

Adequate hematological, renal, and hepatic function

Female patients must be postmenopausal or using adequate contraception if premenopausal, and male patients must be using an appropriate method of contraception if with a premenopausal partner (FOXFIRE); female patients must either be postmenopausal, sterile (surgically or chemically or radiation-induced) or if sexually active using an acceptable method of contraception, and male patients must be surgically sterile or if sexually active and have a premenopausal partner must be using an acceptable method of contraception (SIRFLOX, FOXFIRE-Global)

Suitable for all aspects of treatment determined by clinical assessment undertaken by investigator

Willing and able to provide written informed consent

FOXFIRE-Global and SIRFLOX:

All imaging evidence used as part of the screening process must be within 28 days prior to randomization

 Exclusion Criteria

Evidence of ascites, cirrhosis, portal hypertension; tumour involvement of, or thrombosis leading to occlusion of the main portal vein

Previous radiotherapy delivered to upper abdomen or upper lumbar spine (FOXFIRE); previous radiotherapy delivered to the upper abdomen (SIRFLOX); previous radiotherapy delivered to the liver (FOXFIRE-Global)

Nonmalignant disease that would render patient ineligible at the discretion of the investigator

Dose-limiting toxicity associated with previous 5-fluorouracil or oxaliplatin chemotherapy

Peripheral neuropathy higher than grade 1 (National Cancer Institute–Common Terminology Criteria for Adverse Events [NCI-CTCAE] version 3)

Pregnant or breastfeeding

Previous chemotherapy for any malignancy; adjuvant chemotherapy for colorectal cancer is not an exclusion criterion provided that it was completed more than 6 months prior to entry into the study (SIRFLOX, FOXFIRE-Global). Previous chemotherapy for metastatic colorectal cancer; adjuvant chemotherapy for colorectal cancer is not an exclusion criterion provided that it was completed more than 6 months prior to entry into the study (FOXFIRE)

Concurrent or prior history of cancer other than adequately treated nonmelanoma skin cancer or carcinoma in situ of the cervix (SIRFLOX, FOXFIRE-Global); other active malignancy within last 5 years excluding colorectal cancer and other nonmelanoma skin cancers (FOXFIRE)

SIRFLOX and FOXFIRE-Global:

Allergy to nonionic contrast agents

FOXFIRE only:

Liver metastases amenable to curative resection at time of study entry

Equivocal, immeasurable, or nonevaluable liver metastases

Unequivocal evidence of bone metastasis

All patients received first-line chemotherapy for mCRC. In FOXFIRE, the chemotherapy received was oxaliplatin, 5-fluorouracil, and leucovorin/folic acid in the OxMdG regime [17]. In SIRFLOX and FOXFIRE-Global, the chemotherapy consisted of the same drugs in the mFOLFOX6 regime [18]. Patients were randomized 1:1 to the addition of SIRT using yttrium-90 resin microspheres, using minimization. The minimization factors are given in Table 1. Biological agents (cetuximab or bevacizumab in FOXFIRE; bevacizumab in SIRFLOX) were permitted to be added to chemotherapy at the treating investigators’ discretion. Protocol treatment in each study was commenced within 28 days of randomization. Further details on randomization by minimization and treatment regimens for FOXFIRE and SIRFLOX have been published previously [13,14]. FOXFIRE-Global used the same randomization method and treatment regimen as that used for the SIRFLOX study.

In all 3 trials, patients were assessed every 2 weeks during chemotherapy treatment. In SIRFLOX and FOXFIRE-Global, patients were assessed every 12 weeks during the postprogression follow-up period. FOXFIRE patients were assessed every 8 weeks following completion of treatment up to 18 months and then every 12 weeks thereafter. In all 3 trials, patients were followed until death or for a period of at least 2 years after randomization. Patients undergoing surgical resection after trial entry were also followed up until trial closure or until death.

Screening and follow-up assessments included clinical assessment and laboratory analyses, recording concurrent medications, CT scan of the chest/abdomen/pelvis with or without magnetic resonance imaging, recording adverse events, assessment for resection or ablation, and questionnaire-based assessments of quality of life.

 Study Outcomes Primary Outcome of the Combined Analysis

The primary outcome of OS is the time from randomization to death from any cause, with patients still alive being censored at their last known follow-up date.

 Secondary Outcomes of the Combined Analysis

The secondary outcomes are PFS at any site, liver-specific PFS, health-related quality of life (HRQoL), response rate, resection rate, and the safety profiles. PFS is defined as the time from randomization to radiological progression according to Response Evaluation Criteria In Solid Tumors (RECIST version 1.0) or death from any cause, whichever is sooner. Patients who are not observed to progress or die during the course of the trial will be censored at last known progression-free follow-up date. Patients who withdraw from study treatment prior to documented progression will be censored at the time they commence nonstudy treatment. Patients who become eligible for resection will be considered as still being on study until progression is documented or last follow-up. Scans and tumor response will be centrally reviewed in FOXFIRE and SIRFLOX.

Liver-specific PFS is defined as the time from randomization to radiological progression in the liver (hepatic progression) according to RECIST version 1.0. Progression outside the liver (extrahepatic progression) and deaths prior to progression will be considered as competing risks for failure in the liver. Hepatic progression will be assumed to have occurred immediately before extrahepatic progression in patients who have identical hepatic and extrahepatic progression dates. Patients who withdraw from study treatment prior to any documented progression will be censored at the time they commence nonstudy treatment. Patients who become eligible for resection will be considered as still being on study until progression is documented or last follow-up.

HRQoL will be assessed using the EuroQol 5 dimensions questionnaire (EQ-5D). The EQ-5D comprises the following 5 dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each of the 5 dimensions is scored, generating a profile. A single index score or utility value, representing population-derived preferences for different health states, will be attached to each profile.

The response rate (objective response) is defined as the number of patients achieving a complete or partial response according to RECIST version 1.0 divided by the number of patients in each treatment arm. Early death by any cause and unknown responses will be included in the denominator. Response will be assessed over 2 time periods: within 12 months of randomization and over follow-up.

The resection rate in each treatment arm is defined as the number of patients undergoing resection of their liver metastases divided by the number of patients randomized in each arm. Patients undergoing treatment are assessed by an experienced liver multidisciplinary team or equivalent for eligibility for resection at the discretion of the treating physician.

The safety profile is assessed by the collection of adverse events (AEs) and serious adverse events (SAEs) at any time during treatment with grading using the NCI-CTCAE version 3.

 Sample Size Calculation

For the primary combined OS analysis, the FOXFIRE study protocol details a calculated 1075 patients required using a protocol-specified hazard ratio (HR) of 0.8, a 36-month accrual period, and 18-month minimum follow-up with 2-sided 5% significance, 80% power, and allowing for noncompliance. A total of 710 deaths are expected. The required sample size had been previously calculated to be 1022 [13] but was updated to reflect an estimated increase in median OS in the control group. As the intervention is a local treatment to the liver only, it is anticipated that even if there is no OS benefit in the whole population, a demonstrated survival benefit in the liver-only subgroup would be clinically meaningful and could change practice. It is anticipated that, with a 6-month increase in OS in the SIRT treated liver-only patients compared to those randomized to chemotherapy only, this would require 463 events in the liver-only subgroup.

 Closure of Study

The final analysis will be undertaken when a minimum of 710 deaths overall and 463 deaths in the liver-only subgroup have been observed in the pooled dataset and there has been a minimum follow-up of 2 years since the last patient was randomized into the 3 trials.

 Data Monitoring Committee and Interim Analyses

The independent data and safety monitoring committee (IDSMC) consisted of the same representatives for the FOXFIRE, SIRFLOX, and FOXFIRE-Global clinical trials. Formal interim monitoring of the accumulating data was performed at regular intervals (approximately every 6 months) by the IDSMC for each trial separately. This information included results from other relevant trials but not the analysis of primary or secondary objectives or outcomes by treatment groups apart from the prespecified interim analyses. As part of the review, the IDSMC was asked to justify continued recruitment of further patients or further follow-up. The IDSMC advised on the frequency of future reviews of the data on the basis of accrual and event rates.

The IDSMC reviewed the combined safety data from FOXFIRE and SIRFLOX at the interim analyses. The following planned interim analyses were undertaken using combined data from the FOXFIRE and SIRFLOX trials:

Analysis of toxicity and safety: 8 months after at least 80 patients were randomized in total (a minimum of 40 per trial)

Analysis of toxicity and safety: 8 months after at least 300 patients were randomized in total (a minimum of 120 patients per trial)

The 3 clinical trials that constitute this MA were conducted in accordance with the Declaration of Helsinki and current Good Clinical Practice guidelines, and all participating centers obtained the relevant ethics committee approval before patient enrollment. FOXFIRE was approved by the National Research Ethics Service Committee South Central – Berkshire Research Ethics Committee reference 09/H0505/1 and sponsored by the University of Oxford. SIRFLOX and FOXFIRE-Global were approved by the relevant ethics committees for each center and sponsored by Sirtex Technology Pty Ltd.

 Statistical Analysis Descriptive Summaries

Summaries of baseline factors, including minimization factors, and percentages of missing data will be reported. Losses to follow-up will be reported for each trial and combined. Median follow-up time will be calculated using the reverse Kaplan-Meier method. Dose of trial-specific treatment delivered and treatment received subsequent-to-trial treatment will be described.

 Efficacy

All efficacy analyses will be performed on an intention-to-treat (ITT) basis. OS and PFS estimates will be obtained using Kaplan-Meier survival curves, unadjusted log-rank tests, and survival models. A 2-stage inverse-variance weighted IPD-MA will be performed for both OS and PFS, with the first stage consisting of trial-specific analyses to obtain efficacy estimates (HRs) and the second stage being a pooled analysis of the separate trial-specific HRs. As a sensitivity analysis, a 1-stage IPD-MA using regression models stratified by study will be performed to confirm the results obtained from the 2-stage IPD-MA. For each outcome, multivariable models will be used in the 1-stage IPD-MA to account for baseline covariates.

Liver-specific PFS will be analyzed using cumulative incidence curves, Gray’s test, and competing risks regression. This analysis will be performed on the pooled dataset with models stratified by trial. For OS, PFS, and liver-specific PFS, the potential treatment benefit in those patients who present with disease confined to the liver will be investigated. This prespecified subgroup analysis will be performed on the pooled dataset. The analysis strategy will include calculating HR for the effect of treatment in the liver-only subgroup using survival models stratified by trial.

A separate landmark analysis of OS will start at the 15-month time point (after randomization) and therefore exclude those who have died/withdrawn within 15 months of randomization. This time point has previously been considered of value in clinical trials of the treatment of mCRC [19]. Based on expert opinion, it is anticipated that 15 months from baseline allows sufficient time for events to occur in those with both diagnosed extrahepatic metastases and subclinical extrahepatic metastases at baseline, and therefore this analysis may be of value in detecting any differential survival impact of SIRT in patients with liver-only metastases.

The EQ-5D data will be merged across all 3 trials and summary data prepared on the mean EQ-5D utility score in each trial arm by time period, with appropriate tests for difference. Resection rate and response rate will be analyzed using chi-square tests and odds ratios (ORs) for the individual trials and a 2-stage IPD-MA.

 Safety

Safety analyses will be performed on patients who received at least 1 dose of chemotherapy in either arm and on an as-treated basis (safety population). AEs experienced up to 28 days after the end of trial treatment or 7 months postrandomization, whichever was earlier, will be included. Descriptive summaries of the frequency and severity of AEs and the numbers of patients experiencing AEs of grade 3 or higher between treatment arms will be presented overall and by system organ class. The Medical Dictionary for Regulatory Activities version 16.1 is used to categorize the AEs. Univariate comparisons will be made.

All hypothesis tests will be 2-sided. A significance level of .05 will be used to indicate statistical significance. No missing data imputation is intended. Missing days in dates may be appropriately imputed.

 Results

Data from the final data lock will become available in January 2017. Data analysis will take place in 2017 with results being disseminated via peer-reviewed journals in 2017 and 2018.

 Discussion

Optimizing outcomes of treatment in patients with nonresectable liver metastases was identified in an international consensus expert statement as a key clinical need to be addressed [20]. This is the first extensive investigation of SIRT in the first-line setting for liver metastases from mCRC adequately powered to address an overall survival endpoint.

This prospective MA of 3 phase III studies will provide comprehensive evidence of the safety and potential efficacy of SIRT in the first-line setting for patients with liver metastases from mCRC. The results of the SIRFLOX trial published so far have suggested that the addition of SIRT to chemotherapy can improve liver-specific PFS [15]. The number of patients included in this combined analysis and the long-term follow-up, unprecedented in the field of interventional oncology, will provide adequate power to address a survival endpoint.

Although the SIRFLOX trial reported that liver-specific PFS was longer in the SIRT arm than in the control arm, the PFS at any site was similar between the SIRT arm and the control arm [15]. This can be accounted for by the fact that SIRT is a locoregional treatment to the liver only and therefore will not influence the progression of extrahepatic metastases or extrahepatic subclinical micrometastases.

It is generally accepted that successful resection of liver metastases correlates with improved overall survival, particularly in the context of neoadjuvant or adjuvant chemotherapy [21,22]. Surgery and perioperative chemotherapy are therefore routinely offered to patients with mCRC if the liver metastases are resectable. Similarly, the addition of radiofrequency ablation of liver metastases to standard chemotherapy appears to improve clinical outcomes for patients compared to those patients receiving standard chemotherapy alone [23,24]. These findings with liver-directed therapies suggest that improvement of liver-specific PFS in patients with liver metastases may correlate to improvement in OS, an important hypothesis to be tested for SIRT for the first time in this combined MA of the SIRFLOX, FOXFIRE, and FOXFIRE-Global trials.

The use of an IPD rather than an aggregate data approach to systematic review and MA of randomized controlled trials enables the standardization of outcomes across trials and detailed data checking, providing a more in-depth exploration and more robust MA results [25]. The proposed MA, by prospectively combining IPD from 3 trials, goes beyond classical MA aims and overrides drawbacks from single trials. The prospective approach allows for consistent inclusion and exclusion criteria between studies. The statistical analysis is going to be standardized across studies, and using IPD-MA will provide sufficient statistical power to draw conclusions from subgroup analyses that are generally undermined by low sample size. The use of the 2-stage approach for the main analyses of OS and PFS is prespecified, as the 2-stage approach might produce different parameter estimates to the 1-stage approach, although the estimates are usually similar [26,27].

The results from the combined analysis of the SIRFLOX, FOXFIRE, and FOXFIRE-Global trials will provide valuable clinical information on the efficacy and toxicity profile of SIRT combined with chemotherapy as a first-line regimen for liver metastases from mCRC that will guide clinicians in the use of this technology to treat their patients. The IPD-MA will allow comparisons of less common AEs that would not be possible in a smaller population. Furthermore, this prospective MA provides sufficient power to determine whether SIRT provides a significant survival benefit for patients with metastases confined to the liver and no clinically detectable extrahepatic disease, an important research question among clinicians treating mCRC. When reported, the results of this combined analysis will define the use of SIRT in the treatment of mCRC.

 Abbreviations AE

adverse event

 CRC

colorectal cancer

 CT

computed tomography

 EQ-5D

EuroQol 5 dimensions questionnaire

 FOLFOX

modified fluorouracil, leucovorin, and oxaliplatin regimen

 FOXFIRE

An open-label randomized phase III trial of 5-fluorouracil, oxaliplatin, and folinic acid +/- interventional radioembolization as first-line treatment for patients with unresectable liver-only or liver-predominant metastatic colorectal cancer

 FOXFIRE-Global

Assessment of overall survival of FOLFOX6m plus SIR-Spheres microspheres versus FOLFOX6m alone as first-line treatment in patients with nonresectable liver metastases from primary colorectal carcinoma in a randomized clinical study

 HR

hazard ratio

 HRQoL

health-related quality of life

 IDSMC

independent data and safety monitoring committee

 IPD

individual participant data

 IPD-MA

Individual participant data meta-analysis

 ITT

intention-to-treat

 MA

meta-analysis

 mCRC

metastatic colorectal cancer

 NCI-CTCAE

National Cancer Institute–Common Terminology Criteria for Adverse Events

 OR

odds ratio

 OS

overall survival

 PFS

progression-free survival

 RECIST

Response Evaluation Criteria In Solid Tumors

 SAE

serious adverse events

 SIRFLOX

Randomized comparative study of FOLFOX6m plus SIR-Spheres microspheres versus FOLFOX6m alone as first-line treatment in patients with nonresectable liver metastases from primary colorectal carcinoma

 SIRT

selective internal radiation therapy

The FOXFIRE trial was developed by the National Cancer Research Institute Colorectal Clinical Study Group and is sponsored by the University of Oxford. It is supported by the Bobby Moore Fund of Cancer Research United Kingdom (Clinical Trials Awards and Advisory Committee reference number CRUK/A16630), an educational grant from Sirtex Medical Ltd, and by the National Institute for Health Research Biomedical Research Center Oxford. The SIRFLOX and FOXFIRE-Global studies are sponsored by Sirtex Technology Pty Ltd.

The FOXFIRE Trial Management and Quality Assurance Group: Adil Al-Nahhas, Dave Berry, Ian Chau, Luise Dunham, David Kerr, Nas Khan, Val Lewington, Rachel Midgley, Bruno Morgan, Sarah Pearson, Anne Roberts, Will Steward, Paul Tait, Greg Wilson, Andy Wotherspoon.

FOXFIRE Health Economists: Alastair Gray, Jane Wolstenholme.

The FOXFIRE Study Group: Richard Adams, Andrew Bateman, Claire Blesing, Ewan Brown, Ian Chau, Sebastian Cummins, David Cunningham, Stephen Falk, Maher Hadaki, Marcia Hall, Tamas Hickish, Joanne Hornbuckle, Fiona Lofts, Sarah Lowndes, Astrid Mayer, Matthew Metcalfe, Gary Middleton, Jamie Mills, Amir Montazeri, Rebecca Muirhead, Andreas Polychronis, Colin Purcell, Paul Ross, Liz Sherwin, David Smith, Rubin Soomal, Daniel Swinson, Axel Walther, Andrew Weaver, Charles Wilson, Greg Wilson. We would also like to thank the University of Oxford statisticians Susan Dutton and Peter Dutton for their contributions to the FOXFIRE study.

The SIRFLOX and FOXFIRE-Global Study Group: Steven Ades, Morteza Aghmesheh, Pradip Amin, Bruna Angelelli, Miklos Auber, Jacques Balosso, Alex Beny, Daniel Bloomgarden, Patrick Boland, Eveline Boucher, Christoph Bremer, Michael Brown, Harald-Robert Bruch, James Bui, Matthew Burge, Giuseppe Cardaci, James Carlisle, Ruiz Casado, Yi-Jen Chen, Patrick Chevallier, Michael Chuong, Stephen Clarke, Prasad Cooray, Andrew Coveler, Michel Craninx, Thierry Delanoit, Amélie Deleporte, Kyran Dowling, Aurelie Durand, Paul Eliadis, Francis Facchini, Kynan Feeney, Thomas Ferguson, Michel Ferrante, Aurelie Ferru, Michael Findlay, Maria Fragoso, Gary Frenette, Jacob Frick, Vinod Ganju, Michael Garofalo, Karen Geboes, Gerald Gehbauer, Benjamin George, Ravit Geva, Michael Gordon, Cristina Granetto, Kate Gregory, Seza Gulec, Pascal Hammel, James Hannigan, Norman Heching, Volker Heinemann, Thomas Helmberger, Alain Hendlisz, Koen Hendrickx, Matthew Holtzman, Ayala Hubert, Richard Isaacs, Christopher Jackson, Philip James, Adeel Kaiser, Chris Karapetis, Andreas Kaubisch, Yeul Hong Kim, Yon-Dschun Ko, Todd Kooy, Hendrik Kröning, Frank Lammert, Jin Tung Liang, Winston Liauw, Lionel Lim, Yoo Joo Lim, Jin Hwang Liu, Samy Louafi, Marc de Man, Jeffrey Margolis, Robert Martin, Andrea Martoni, Gavin Marx, Marco Matos, Els Monsaert, Veerle Moons, Stefania Mosconi, Louise Nott, Arnd Nusch, Anne O’Donnell, Howard Ozer, Siddarth Padia, Nick Pavlakis, Marc Peeters, David Perez, Stefan Pluntke, Marc Polus, Alex Powell, Marc Pracht, Timothy Price, Jorge Ramon, David Ransom, Christine Rebischung, Jens Ricke, Karsten Ridwelski, Hanno Riess, Jorge Ramon Riera, William Rilling, Bridget Robinson, Javier Rodríguez, Federico Sanchez, Tilmann Sauerbruch, Michael Savin, Klemens Scheidhauer, Elyse Schneiderman, Grant Seeger, Eva Segelov, Einat Shaham Schmueli, Adi Shani, Jenny Shannon, Navesh Sharma, Stephen Shibata, Nimit Singhal, Denis Smith, Randall Smith, Salomon Stemmer, Oliver Stötzer, Andrew Strickland, Julien Taieb, Iain Tan, Klaus Tatsch, Eric Terrebonne, Thomas Tichler, Antonio Trogu, Craig Underhill, Daniel Van Daele, Ursula Vehling-Kaiser, Ruth Vera-Garcia, Caterina Vivaldi, Thomas Vogl, Euan Walpole, Eric Wang, Mark Westcott, Samuel Whiting, Ido Wolf.

The authors acknowledge Bruce Gray for his involvement in the development of the initial SIRFLOX protocol and the editorial assistance provided by Martin Gilmour and Thierry Deltheil of ESP Bioscience (Crowthorne, United Kingdom) during the preparation of this manuscript.

RAS and HSW conceived the FOXFIRE study and secured the funding for the investigator-initiated study. SBL and JM provided statistical advice. All authors read and approved the final draft of the manuscript for submission. PG, VG, and GVH were substantially involved in the conception and design of the SIRFLOX and FOXFIRE-Global studies. EAF coordinated the FOXFIRE study during the patient follow-up period. All authors were involved in critically reviewing the manuscript for important intellectual content, and all authors read and approved the final draft of the manuscript for submission.

RAS has received research funding, honoraria, and consultancy fees from Sirtex Medical Ltd. PG has received honoraria from Sirtex for participation in advisory boards and for giving presentations. VG and GVH have received compensation for participation in advisory committees from Sirtex. PSV, JM, HSW, EAF, and SBL declare no competing interests.

 Surveillance, Epidemiology, and End Results Stat Fact Sheets: Colon and Rectum Cancer, 2003-2009 2013 2016-12-15 https://seer.cancer.gov/statfacts/html/colorect.html Abbas S Lam V Hollands M Ten-year survival after liver resection for colorectal metastases: systematic review and meta-analysis ISRN Oncol 2011 10.5402/2011/763245 22091431 PMC3200144 Helling TS Martin M Cause of death from liver metastases in colorectal cancer Ann Surg Oncol 2014 02 21 2 501 506 10.1245/s10434-013-3297-7 24081807 Rothbarth J van de Velde CJH Treatment of liver metastases of colorectal cancer Ann Oncol 2005 16 Suppl 2 ii144 ii149 10.1093/annonc/mdi702 15958446 16/suppl_2/ii144 Berber E Pelley R Siperstein AE Predictors of survival after radiofrequency thermal ablation of colorectal cancer metastases to the liver: a prospective study J Clin Oncol 2005 03 01 23 7 1358 1364 10.1200/JCO.2005.12.039 15684312 JCO.2005.12.039 Navarra G Ayav A Weber J Jensen SL Smadga C Nicholls JP Habib NA Jiao LR Short- and long-term results of intraoperative radiofrequency ablation of liver metastases Int J Colorectal Dis 2005 11 20 6 521 528 10.1007/s00384-005-0743-4 15864606 Van Cutsem E Nordlinger B Adam R Köhne C Pozzo C Poston G Ychou M Rougier P European Colorectal Metastases Treatment Group Towards a pan-European consensus on the treatment of patients with colorectal liver metastases Eur J Cancer 2006 09 42 14 2212 2221 10.1016/j.ejca.2006.04.012 16904315 S0959-8049(06)00518-1 Jones RP Poston GJ Resection of liver metastases in colorectal cancer in the era of expanding systemic therapy Annu Rev Med 2016 09 28 10.1146/annurev-med-062415-093510 27686016 Alberts SR Horvath WL Sternfeld WC Goldberg RM Mahoney MR Dakhil SR Levitt R Rowland K Nair S Sargent DJ Donohue JH Oxaliplatin, fluorouracil, and leucovorin for patients with unresectable liver-only metastases from colorectal cancer: a North Central Cancer Treatment Group phase II study J Clin Oncol 2005 12 20 23 36 9243 9249 10.1200/JCO.2005.07.740 16230673 JCO.2005.07.740 Folprecht G Grothey A Alberts S Raab H Köhne C Neoadjuvant treatment of unresectable colorectal liver metastases: correlation between tumour response and resection rates Ann Oncol 2005 08 16 8 1311 1319 10.1093/annonc/mdi246 15870084 mdi246 Nicolay NH Berry DP Sharma RA Liver metastases from colorectal cancer: radioembolization with systemic therapy Nat Rev Clin Oncol 2009 12 6 12 687 697 10.1038/nrclinonc.2009.165 19884901 nrclinonc.2009.165 Wang LM Jani AR Hill EJ Sharma RA Anatomical basis and histopathological changes resulting from selective internal radiotherapy for liver metastases J Clin Pathol 2013 03 66 3 205 211 10.1136/jclinpath-2012-201231 23162108 jclinpath-2012-201231 PMC3595145 Dutton SJ Kenealy N Love SB Wasan HS Sharma RA FOXFIRE Protocol Development Group and the NCRI Colorectal Clinical Study Group FOXFIRE protocol: an open-label, randomised, phase III trial of 5-fluorouracil, oxaliplatin and folinic acid (OxMdG) with or without interventional Selective Internal Radiation Therapy (SIRT) as first-line treatment for patients with unresectable liver-only or liver-dominant metastatic colorectal cancer BMC Cancer 2014 07 09 14 497 10.1186/1471-2407-14-497 25011439 1471-2407-14-497 PMC4107961 Gibbs P Gebski V Van Buskirk M Thurston K Cade DN Van Hazel GA Selective Internal Radiation Therapy (SIRT) with yttrium-90 resin microspheres plus standard systemic chemotherapy regimen of FOLFOX versus FOLFOX alone as first-line treatment of non-resectable liver metastases from colorectal cancer: the SIRFLOX study BMC Cancer 2014 12 01 14 897 10.1186/1471-2407-14-897 25487708 1471-2407-14-897 PMC4289171 van Hazel GA Heinemann V Sharma NK Findlay MPN Ricke J Peeters M Perez D Robinson BA Strickland AH Ferguson T Rodríguez J Kröning H Wolf I Ganju V Walpole E Boucher E Tichler T Shacham-Shmueli E Powell A Eliadis P Isaacs R Price D Moeslein F Taieb J Bower G Gebski V Van Buskirk M Cade DN Thurston K Gibbs P SIRFLOX: randomized phase III trial comparing first-line mFOLFOX6 (plus or minus bevacizumab) versus mFOLFOX6 (plus or minus bevacizumab) plus selective internal radiation therapy in patients with metastatic colorectal cancer J Clin Oncol 2016 05 20 34 15 1723 1731 10.1200/JCO.2015.66.1181 26903575 JCO.2015.66.1181 Gray B Van Hazel G Hope M Burton M Moroz P Anderson J Gebski V Randomised trial of SIR-Spheres plus chemotherapy versus chemotherapy alone for treating patients with liver metastases from primary large bowel cancer Ann Oncol 2001 12 12 12 1711 1720 11843249 Shimizu T Satoh T Tamura K Ozaki T Okamoto I Fukuoka M Nakagawa K Oxaliplatin/fluorouracil/leucovorin (FOLFOX4 and modified FOLFOX6) in patients with refractory or advanced colorectal cancer: post-approval Japanese population experience Int J Clin Oncol 2007 06 12 3 218 223 10.1007/s10147-007-0658-x 17566846 Braun MS Adab F Bradley C McAdam K Thomas G Wadd NJ Rea D Philips R Twelves C Bozzino J MacMillan C Saunders MP Counsell R Anderson H McDonald A Stewart J Robinson A Davies S Richards FJ Seymour MT Modified de Gramont with oxaliplatin in the first-line treatment of advanced colorectal cancer Br J Cancer 2003 10 06 89 7 1155 1158 10.1038/sj.bjc.6601237 14520437 6601237 PMC2394314 Allegra CJ Yothers G O'Connell MJ Sharif S Petrelli NJ Colangelo LH Atkins JN Seay TE Fehrenbacher L Goldberg RM O'Reilly S Chu L Azar CA Lopa S Wolmark N Phase III trial assessing bevacizumab in stages II and III carcinoma of the colon: results of NSABP protocol C-08 J Clin Oncol 2011 01 01 29 1 11 16 10.1200/JCO.2010.30.0855 20940184 JCO.2010.30.0855 PMC3055856 Clary BM Grothey A Kopetz S Marsh RDW Systemic cytotoxic and biologic therapies for colorectal cancer liver metastases: expert consensus statement HPB (Oxford) 2013 02 15 2 116 118 10.1111/j.1477-2574.2012.00598.x 23297722 S1365-182X(15)31365-4 PMC3719917 Nordlinger B Sorbye H Glimelius B Poston GJ Schlag PM Rougier P Bechstein WO Primrose JN Walpole ET Finch-Jones M Jaeck D Mirza D Parks RW Mauer M Tanis E Van Cutsem CE Scheithauer W Gruenberger T EORTC Gastro-Intestinal Tract Cancer Group Cancer UK Arbeitsgruppe Lebermetastasen und tumoren in der Chirurgischen Arbeitsgemeinschaft Onkologie (ALM-CAO) Australasian Gastro-Intestinal Trials Group (AGITG) Fédération Francophone de Cancérologie Digestive (FFCD) Perioperative FOLFOX4 chemotherapy and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC 40983): long-term results of a randomised, controlled, phase 3 trial Lancet Oncol 2013 11 14 12 1208 1215 10.1016/S1470-2045(13)70447-9 24120480 S1470-2045(13)70447-9 Bokemeyer C Van Cutsem E Rougier P Ciardiello F Heeger S Schlichting M Celik I Köhne CH Addition of cetuximab to chemotherapy as first-line treatment for KRAS wild-type metastatic colorectal cancer: pooled analysis of the CRYSTAL and OPUS randomised clinical trials Eur J Cancer 2012 07 48 10 1466 1475 10.1016/j.ejca.2012.02.057 22446022 S0959-8049(12)00209-2 Ruers T Punt C Van Coevorden F Pierie JPEN Borel-Rinkes I Ledermann JA Poston G Bechstein W Lentz MA Mauer M Van Cutsem E Lutz MP Nordlinger B EORTC Gastro-Intestinal Tract Cancer Group Arbeitsgruppe Lebermetastasen und tumoren in der Chirurgischen Arbeitsgemeinschaft Onkologie (ALM-CAO) National Cancer Research Institute Colorectal Clinical Study Group (NCRI CCSG) Radiofrequency ablation combined with systemic treatment versus systemic treatment alone in patients with non-resectable colorectal liver metastases: a randomized EORTC Intergroup phase II study (EORTC 40004) Ann Oncol 2012 10 23 10 2619 2626 10.1093/annonc/mds053 22431703 mds053 PMC3457746 Ruers T Cornelis J Punt C Van Coevorden F Pierie J Rinkes I Ledermann J Poston G Bechstein W Lentz M Mauer M Van Cutsem E Lutz M Nordlinger B Radiofrequency ablation (RFA) combined with chemotherapy for unresectable colorectal liver metastases (CRC LM): Long-term survival results of a randomized phase II study of the EORTC-NCRI CCSG-ALM Intergroup 40004 (CLOCC) J Clin Oncol 2015 33 Suppl A3501 Tierney JF Vale C Riley R Smith CT Stewart L Clarke M Rovers M Individual participant data (IPD) meta-analyses of randomised controlled trials: guidance on their use PLoS Med 2015 07 12 7 e1001855 10.1371/journal.pmed.1001855 26196287 PMEDICINE-D-13-04070 PMC4510878 Debray TPA Moons KGM Abo-Zaid GMA Koffijberg H Riley RD Individual participant data meta-analysis for a binary outcome: one-stage or two-stage? PLoS One 2013 8 4 e60650 10.1371/journal.pone.0060650 23585842 PONE-D-13-01399 PMC3621872 Stewart GB Altman DG Askie LM Duley L Simmonds MC Stewart LA Statistical analysis of individual participant data meta-analyses: a comparison of methods and recommendations for practice PLoS One 2012 7 10 e46042 10.1371/journal.pone.0046042 23056232 PONE-D-12-16086 PMC3463584