Clinical outcomes of treatments for patients with refractory metastatic colorectal cancer: a systematic literature review and network meta-analysis

Abstract

Background:

For metastatic colorectal cancer (mCRC) patients whose disease is refractory to standard therapies, treatment with regorafenib, fruquintinib, or trifluridine/tipiracil (FTD/TPI; also known as TAS-102) with or without bevacizumab is recommended. Recent clinical trial evidence indicates that FTD/TPI + bevacizumab has superior efficacy to FTD/TPI for these patients, although its relative efficacy compared with other treatment regimens requires further evidence.

Objectives:

A systematic literature review (SLR) and network meta-analysis (NMA) were performed to evaluate the relative efficacy of treatments for patients with refractory mCRC.

Design:

SLR and NMA.

Data sources and methods:

Databases were searched for clinical trials evaluating treatments for refractory mCRC patients. NMA using random- and fixed-effects models was performed to estimate the relative treatment effects of FTD/TPI + bevacizumab compared with other treatment regimens on overall survival (OS) and progression-free survival (PFS).

Results:

Twenty-eight randomized controlled trials evaluating treatment regimens in patients with refractory mCRC were identified. Of these, 16 trials connected in a network with FTD/TPI + bevacizumab. NMA including these trials using random-effects models showed that the efficacy of FTD/TPI + bevacizumab was statistically favorable relative to that of other treatment regimens including placebo/best supportive care (hazard ratio (95% credible interval); OS: 0.41 (0.28, 0.58), PFS: 0.21 (0.14, 0.31)), FTD/TPI monotherapy (OS: 0.59 (0.43, 0.79), PFS: 0.46 (0.34, 0.64)), cetuximab (OS: 0.47 (0.29, 0.73), PFS: 0.32 (0.20, 0.53)), cetuximab + irinotecan (PFS: 0.43 (0.19, 0.98), panitumumab (OS: 0.46 (0.28, 0.72), PFS: 0.35 (0.22, 0.59)), regorafenib (OS: 0.60 (0.38, 0.95), PFS: 0.49 (0.31, 0.84)), and fruquintinib (OS: 0.62 (0.39, 0.94)).

Conclusion:

NMA results suggest that FTD/TPI + bevacizumab confers clinically relevant improvements in OS and PFS compared with other treatment regimens for refractory mCRC patients, supporting the use of this combination therapy in the third-line treatment setting.

Keywords

combination therapy network meta-analysis refractory metastatic colorectal cancer survival outcomes systematic review

Introduction

Colorectal cancer (CRC) is the third most commonly diagnosed cancer and the second leading cause of cancer-related death worldwide.^1,2 Although the incidence of CRC is declining in certain countries and regions, its global incidence is on the rise due to greater environmental exposures resulting from the adoption of Western diets and more sedentary lifestyles, with 3.2 million new cases expected in 2040.^1,2

Approximately 50%–80% of CRC patients experience cancer metastasis.³ Patients with metastatic CRC (mCRC) typically receive first- and second-line treatment with oxaliplatin- or irinotecan-containing regimens, which may also include targeted therapies (e.g., bevacizumab, cetuximab, panitumumab) based on the tumor molecular status and primary tumor location. In mCRC patients whose disease is refractory to standard therapies, treatment with trifluridine/tipiracil (FTD/TPI; also known as TAS-102) with or without bevacizumab is recommended with the highest level of evidence and magnitude of clinical benefit.^3–6 Regorafenib and, more recently, fruquintinib have also been recommended as treatments for refractory mCRC with lower magnitudes of clinical benefit.^3–5,7–9

FTD/TPI treats cancer by inhibiting the cell cycle through its incorporation into DNA.¹⁰ Although randomized controlled trials (RCTs) such as RECOURSE demonstrate the clinical efficacy of FTD/TPI monotherapy versus placebo in prolonging the overall survival (OS) of refractory mCRC patients,^11,12 additional evidence suggests that the anti-cancer efficacy of FTD/TPI can be synergistically boosted by its delivery in combination with other agents, such as bevacizumab.¹³ Indeed, a phase II clinical trial shows that FTD/TPI + bevacizumab was more effective than FTD/TPI monotherapy in prolonging OS among third-line mCRC patients (9.4 vs 6.7 months, respectively).¹⁴ Furthermore, the superiority of FTD/TPI + bevacizumab over FTD/TPI monotherapy was confirmed by SUNLIGHT, a phase III RCT that enrolled patients with refractory mCRC treated mainly in the third-line setting. In this trial, median progression-free survival (PFS) and OS were 5.6 and 10.8 months in the FTD/TPI + bevacizumab arm versus 2.4 and 7.5 months in the FTD/TPI arm, respectively.¹⁵

Although clinical trial evidence suggests that FTD/TPI + bevacizumab is an effective approach to treating mCRC patients whose disease is refractory to standard therapies, it has not been compared against all other treatment regimens in head-to-head RCTs. Network meta-analysis (NMA) is a statistical method that allows for indirect comparisons between treatments where head-to-head evidence from RCTs may not be available. Specifically, NMA can be used to combine direct and indirect evidence regarding any interventions that form a connected network of trials wherein each trial has at least one intervention in common with another trial and all trials are sufficiently similar in terms of trial characteristics, treatment characteristics, distribution of baseline patient characteristics, and outcome definitions.^16,17 Thus, we estimated the relative efficacy of FTD/TPI + bevacizumab in terms of OS and PFS compared with other treatments for refractory mCRC patients through a systematic literature review (SLR) and NMA.

Materials and methods

Systematic literature review

The SLR was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and National Institute for Health and Care Excellence guidance for systematic reviews.^18,19 (PRISMA 2020 checklist, Supplemental Table 1).

Study selection was performed using pre-specified population, interventions, comparators, outcomes, and study design (PICOS) criteria (Supplemental Table 2). Clinical trials were eligible for inclusion if they enrolled patients with unresectable adenocarcinoma of the colon or rectum who received two prior chemotherapy regimens for the treatment of advanced or mCRC and demonstrated progressive disease or intolerance to the last regimen. Although the target population was patients undergoing third-line treatment for mCRC, it was anticipated that most trials evaluating relevant comparators would be conducted in a broader population; therefore, trials were included if they enrolled any proportion of third-line patients, and the plausibility of comparing trials with different proportions of third-line patients was evaluated in the feasibility assessment. Interventions of interest were the following treatments delivered alone or in combination with each other: aflibercept, best supportive care (BSC), bevacizumab, capecitabine, cetuximab, encorafenib, fluorouracil, fruquintinib, FTD/TPI, irinotecan, oxaliplatin, panitumumab, ramucirumab, or regorafenib. Comparators of interest were placebo, any intervention of interest, or investigator’s choice of therapy if the options were among the interventions of interest. Although any type of clinical trials, including RCTs, non-randomized multi-trials, and single-arm trials, were eligible for inclusion in the SLR, only RCTs were considered for inclusion in the NMA.

Relevant trials were identified by searches of Excerpta Medica database (Embase), Medical Literature Analysis and Retrieval System Online (MEDLINE), and Cochrane Central Register of Controlled Trials (CENTRAL) via Ovid on October 12, 2023 (full search strategies provided in Supplemental Tables 3–5). Scottish Intercollegiate Guidelines Network search filters (https://www.sign.ac.uk/what-we-do/methodology/search-filters/) were used to limit Embase and MEDLINE search results to RCTs and were modified to also include single-arm trials.²⁰ As clinical trials of third-line treatments for mCRC patients did not appear until after 2010,^5,21 search results were limited to publications from 2010 to present.

Conference abstracts from the American Society of Clinical Oncology (ASCO) Annual Meeting (2021–2023), ASCO Gastrointestinal Cancers Symposium (2021–2023), European Society of Medical Oncology (ESMO) Congress (2021–2022), and ESMO World Congress on Gastrointestinal Cancer (2021–2023) were hand-searched to identify trial results not yet published as journal articles. To mitigate the risk of publication bias (i.e., missing clinical trials with null or negative results not published as journal articles or conference abstracts), the US National Institute of Health Clinical Trials Registry and European Union Clinical Trial Registry were searched to identify completed trials with results available.

Title/abstract screening, full-text screening, and data extraction were performed by two independent reviewers, and any discrepancies between reviewers were resolved through discussion or by involving a third reviewer. The risk of bias in included RCTs was assessed using the Cochrane Collaboration’s Risk of Bias tool, version 2.²²

Feasibility assessment

NMA is an extension of pairwise meta-analysis that allows indirect comparison of interventions that have not been studied in a head-to-head fashion.²³ As the validity of an NMA depends on the absence of systematic differences in treatment effects across trials,^16,23–26 a feasibility assessment was performed before commencing the NMA to (1) determine whether RCT evidence for the interventions of interest formed one connected network where each trial has at least one intervention (active or placebo) in common with another trial for each outcome; and (2) assessment of trial design, treatment, patient, and outcome characteristics that may affect treatment effects across direct comparisons within the evidence networks.²⁴

To further explore whether the number of prior treatment regimens received by patients modified the treatment effect for FTD/TPI versus placebo, post hoc analysis of data from the RECOURSE trial (ClinicalTrials.gov identifier: NCT01607957)^12,27–30 was conducted. Three methods were used to investigate the treatment effect by the number of prior regimens: (1) interaction term analysis in a univariate model; (2) interaction term analysis in a multivariate model including treatment (FTD/TPI vs placebo), number of prior regimens, KRAS status, time since diagnosis of metastasis, region of the world, primary tumor site, ECOG status at baseline, and number of metastatic sites; and (3) stratification analysis in a multivariate model including treatment (FTD/TPI vs placebo), KRAS status, time since diagnosis of metastasis, region of the world, race, ECOG status at baseline, and number of metastatic sites.

Network meta-analysis

After feasibility assessment, NMA was performed for OS and PFS. NMA is typically performed using a random-effects model (which assumes that the effect size for each trial in the network can vary due to heterogeneity among trial design, treatment, patient, and/or outcome characteristics) or a fixed-effects model (which assumes that one effect size underlies all trials in the network). Although random-effects models are generally preferred because they account for uncertainty arising from heterogeneity among trials, fixed-effects models may be appropriate when the number of trials for each treatment comparison is too small (i.e., several connections in the network supported by only one trial) to obtain an accurate estimate of between-trial variance.³¹ Thus, as some treatment comparisons in the present evidence network were based on multiple trials and others were based on single trials, NMA was performed using both random-effects and fixed-effects models. Statistical tests conducted to assess the proportional hazards assumption (e.g., log cumulative hazard, Schoenfeld residuals, smoothed hazard) indicated violations for only 3 of the 16 connected trials; therefore, constant hazards analysis was conducted. Proportional hazards between treatments were assumed using regression models with a contrast-based normal likelihood for the log hazard ratio (HR; and corresponding standard error) for each trial.³²

Normal non-informative prior distributions for the parameters were estimated with a mean of 0 and a variance of 10,000.³² NMA results are presented as OS and PFS HRs estimating the treatment effect of each intervention relative to the reference treatment. The posterior distributions of relative treatment effects are summarized by the median and 95% credible interval (CrI), which reflects a 95% probability that the estimate is within the specified range. Unlike confidence intervals (CIs), which are used in frequentist analysis to determine whether the true estimate of effect lies within a specified interval (giving rise to a dichotomized interpretation of “significance” vs “non-significance”), CrIs are used in Bayesian analysis to reflect the probability that the true estimate of effect lies with the specified interval (with no dichotomized interpretation).³³ As such, HRs <1 with 95% CrIs that do not cross 1 are interpreted as “statistically favorable,” whereas HRs <1 with 95% CrIs that do cross 1 are interpreted as “numerically favorable.” Analyses were conducted using R version 4.2.2 (http://www.r-project.org/) and JAGS version 4.3.1 (OpenBUGS Project Management Group).

Results

SLR and feasibility assessment

Study selection

The literature search yielded 8112 records, of which 6005 titles/abstracts and 246 full-texts were screened (Supplemental Figure 1). Twenty-six additional records were identified from conference websites, clinical trial registries, and the SUNLIGHT clinical study report. In total, 84 trials were included in the SLR (28 RCTs, 2 non-randomized multi-arm trials, and 54 single-arm trials; Supplemental Table 6). Of the 28 RCTs, 19 evaluated at least 2 treatment arms of interest. Of these 19 RCTs, 16 trials connected in a network with FTD/TPI + bevacizumab (Figure 1). Thus, the feasibility assessment focused on the trial design characteristics, treatment characteristics, baseline patient characteristics, and outcome availability/definitions of these 16 connected RCTs to evaluate the feasibility of conducting a credible NMA.

Figure 1.

Network of evidence for OS and PFS.

Trial design characteristics

Of the 16 connected RCTs, 10 trials were phase III, and 6 trials were phase II (Table 1; Supplemental Table 7). One trial was quadruple-blind, one was triple-blind, seven were double-blind, and seven were open-label. Several trials were multinational, whereas some trials were conducted across the Asia continent or within single countries. All trials had a low risk or only some concerns of bias (Supplemental Figure 2).

Table 1.

Trial characteristics.

Trial ID	Phase	Masking	Geographical region	Multi-center	Treatments	N
20020408 (NCT00113763)^34–42	II	Open-label	Multinational	Yes	Panitumumab + BSC vs BSC	463
ASPECCT^40–44	III	Open-label	Multinational	Yes	Panitumumab vs Cetuximab	999
BOND-3 (NCT02292758)⁴⁵	II	Double-blind	US	Yes	Cetuximab + irinotecan + bevacizumab vs Cetuximab + irinotecan + placebo	36
CO.17^46–53	III	Open-label	Multinational	Yes	Cetuximab + BSC vs BSC	572
CONCUR^54,55	III	Triple-blind	Asia	Yes	Regorafenib vs Placebo	136
CORRECT (NCT01103323)^56–59	III	Quadruple-blind	Multinational	Yes	Regorafenib vs Placebo	760
FRESCO^60–63	III	Double-blind	China	Yes	Fruquintinib vs Placebo	416
FRESCO-2 (NCT04322539)^64,65	III	Double-blind	Multinational	Yes	Fruquintinib vs Placebo	691
ICECREAM⁶⁶	II	Open-label	Multinational	Yes	Cetuximab vs Cetuximab + irinotecan	53
Pfeiffer 2020^14,67	II	Open-label	Denmark	Yes	FTD/TPI vs FTD/TPI + bevacizumab	93
RECOURSE (NCT01607957)^12,27–30	III	Double-blind	Multinational	Yes	FTD/TPI vs Placebo	800
SUNLIGHT^15,68–70	III	Open-label	Multinational	Yes	FTD/TPI + bevacizumab vs FTD/TPI	492
TERRA (NCT01955837)⁷¹	III	Double-blind	Asia	Yes	FTD/TPI vs Placebo	406
VELO^72–75	II	Open-label	Italy	Yes	FTD/TPI + panitumumab vs FTD/TPI	62
Xu 2017 (NCT02196688)⁷⁶	II	Double-blind	China	Yes	Fruquintinib + BSC vs Placebo + BSC	71
Yoshino 2012⁷⁷	II	Double-blind	Japan	Yes	FTD/TPI vs Placebo	169

BSC, best supportive care; FTD/TPI, trifluridine/tipiracil; US, United States.

Treatment characteristics

Across trials, the most commonly evaluated treatment regimens were placebo/BSC (n = 10 trials), FTD/TPI monotherapy (n = 6 trials), cetuximab (n = 3 trials), fruquintinib (n = 3 trials), regorafenib (n = 2 trials), FTD/TPI + bevacizumab (n = 2 trials), and panitumumab (n = 2 trials). Regarding trials evaluating FTD/TPI-containing treatment regimens, all trials used similar FTD/TPI delivery schedules, except the Yoshino 2012 trial that did not specify the days of FTD/TPI delivery (Supplemental Table 8). Regarding trials evaluating cetuximab-containing treatment regimens, three trials (ASPECCT, CO.17, and ICECREAM) evaluated a higher initial dose (400 mg/m²) followed by lower subsequent doses (250 mg/m²) of cetuximab on a weekly basis, whereas BOND-3 evaluated a constant higher dose (500 mg/m²) of cetuximab on a twice monthly basis. The doses and delivery schedules for bevacizumab-, panitumumab-, regorafenib-, and fruquintinib-containing regimens were similar across trials.

Patient characteristics

Median patient age and the proportion of male patients were similar across trials (Supplemental Table 9). The distribution of patient race/ethnicity varied among trials, particularly the proportions of patients who were White or Asian. However, similar OS and PFS reported for Whites and Asians in the CORRECT and RECOURSE trials implied that ethnicity was not a treatment effect modifier for this population. All trials enrolled mostly patients with an ECOG performance status (or equivalent) of 0 or 1, although CO.17 and 20020408 enrolled >10% of patients with a performance status of 2. Trials varied in terms of the number of prior lines of therapy received for advanced/metastatic disease and in the proportion of patients who had received certain prior treatment regimens (Supplemental Tables 10 and 11). National Comprehensive Cancer Network (NCCN) and ESMO treatment guidelines state that cetuximab and panitumumab should only be used for patients with KRAS/NRAS/BRAF wild-type and left-sided tumors, as patients with KRAS/NRAS/BRAF mutant or right-sided tumors are unlikely to respond to these agents.^3–5 However, there was minimal reporting and/or between-trial heterogeneity in these patient characteristics among trials evaluating cetuximab- or panitumumab-containing treatment regimens (Supplemental Tables 12 and 13).

Reported outcomes

Ten trials reported investigator-assessed PFS, one trial (20020408) reported independent review committee (IRC)-assessed PFS, and one trial (Yoshino 2012) reported both investigator-assessed and IRC-assessed PFS (Supplemental Table 14). Four trials (BOND-3, CO.17, FRESCO, and ICECREAM) did not describe the method of PFS assessment. Definitions of OS and PFS were similar across all trials. Reported HRs for OS and PFS are presented in Supplemental Tables 15 and 16.

Summary of feasibility assessment

Although the target population was patients undergoing third-line treatment for mCRC, most trials were not conducted exclusively in the third-line setting. To investigate the impact of line of treatment on relative treatment effects for different trials included in the NMA, the HRs for different lines of treatment within the same trial were compared when available; in all, there was wide overlap in the CIs and no conclusive trend in the impact of different lines of treatment on the relative treatment effects for trials included in the feasibility assessment. Additionally, in post hoc analysis of data from RECOURSE, the interaction terms between treatment (FTD/TPI vs placebo) and number of prior regimens were not statistically significant in the univariate model (2 vs ⩾4 prior regimens: p = 0.172; 3 vs ⩾4 prior regimens: p = 0.400); the interaction terms between treatment and number of prior regimens were not statistically significant in the multivariate model (2 vs 3 or ⩾4 prior regimens: p = 0.433; 3 vs 2 or ⩾4 prior regimens: p = 0.607; ⩾4 vs 2 or 3 prior regimens, p = 0.292); and the HRs were similar in multivariate analyses stratified by 2 (HR = 0.81), 3 (HR = 0.71), or 4 (HR = 0.62) prior regimens, implying that line of treatment is not an important treatment effect modifier. Thus, there was no conclusive evidence that the line of treatment modifies the relative treatment effects of trials included in the feasibility assessment and, therefore, it was feasible to include trials evaluating different proportions of patients undergoing third-line treatment in the same network.

Overall, considering trial designs, baseline patient characteristics, and outcome definitions, the feasibility assessment revealed no critical dissimilarities among connected trials that would prohibit their inclusion in the NMA.

Network meta-analysis

In the networks for OS and PFS (Figure 1), BSC and placebo were treated as the same node, as they were assumed to have equivalent efficacy. Also, two different dosage regimens for cetuximab (i.e., a higher initial dose (400 mg/m²) followed by lower subsequent doses (250 mg/m²) on a weekly basis vs a constant higher dose (500 mg/m²) on a twice monthly basis) were treated as the same node in the network based on evidence of similar pharmacokinetics in modeling analyses and effectiveness in real-world patient populations.⁷⁸

Overall survival

FTD/TPI + bevacizumab had statistically favorable effects on OS relative to FTD/TPI (HR: 0.59, 95% CrI: 0.43–0.79), cetuximab (HR: 0.47, 95% CrI: 0.29–0.73), panitumumab (HR: 0.46, 95% CrI: 0.28–0.72), regorafenib (HR: 0.60, 95% CrI: 0.38–0.95), fruquintinib (HR: 0.62, 95% CrI: 0.39–0.94), and placebo/BSC (HR: 0.41, 95% CrI: 0.28–0.58) in the random-effects NMA model (Table 2, top). Similar results were found in the fixed-effects NMA model (FTD/TPI + bevacizumab vs FTD/TPI (HR: 0.59, 95% CrI: 0.49–0.73), cetuximab (HR: 0.47, 95% CrI: 0.36–0.62), cetuximab + irinotecan (HR: 0.49, 95% CrI: 0.26–0.94), panitumumab (HR: 0.47, 95% CrI: 0.36–0.62), regorafenib (HR: 0.59, 95% CrI: 0.44–0.79), fruquintinib (HR: 0.63, 95% CrI: 0.48–0.83), and placebo/BSC (HR: 0.42, 95% CrI: 0.33–0.53; Supplemental Table 17). FTD/TPI + bevacizumab also had numerically favorable effects on OS relative to cetuximab + irinotecan and FTD/TPI + panitumumab in the random-effects model and relative to FTD/TPI + panitumumab in the fixed-effects model.

Table 2.

Results of random-effects NMA for OS (top) and PFS (bottom).

OS	FTD/TPI + bevacizumab	FTD/TPI	Cetuximab	Cetuximab + irinotecan	Panitumumab	Cetuximab + irinotecan + bevacizumab	Regorafenib	Fruquintinib	FTD/TPI + panitumumab	Placebo/BSC
FTD/TPI + bevacizumab	1	0.59 (0.43, 0.79)	0.47 (0.29, 0.73)	0.49 (0.22, 1.09)	0.46 (0.28, 0.72)	1.19 (0.30, 4.18)	0.60 (0.38, 0.95)	0.62 (0.39, 0.94)	0.57 (0.28, 1.16)	0.41 (0.28, 0.58)
FTD/TPI		1	0.79(0.55, 1.12)	0.83(0.40, 1.72)	0.78(0.54, 1.09)	2.01(0.54, 6.77)	1.01(0.73, 1.45)	1.05(0.75, 1.42)	0.96(0.51, 1.87)	0.70 (0.56, 0.86)
Cetuximab			1	1.06(0.56, 2.00)	0.99(0.74, 1.28)	2.55(0.73, 8.32)	1.28(0.88, 1.93)	1.33(0.92, 1.89)	1.22(0.59, 2.59)	0.88(0.66, 1.16)
Cetuximab + irinotecan				1	0.93(0.46, 1.87)	2.41(0.82, 6.70)	1.21(0.58, 2.55)	1.26(0.60, 2.61)	1.14(0.44, 3.08)	0.83(0.42, 1.66)
Panitumumab					1	2.59(0.72, 8.64)	1.30(0.90, 1.99)	1.35(0.94, 1.94)	1.23(0.60, 2.64)	0.89(0.68, 1.19)
Cetuximab + irinotecan + bevacizumab						1	0.50(0.15, 1.88)	0.52(0.15, 1.93)	0.47(0.12, 2.15)	0.35(0.10, 1.25)
Regorafenib							1	1.04(0.70, 1.45)	0.95(0.46, 1.96)	0.69 (0.51, 0.88)
Fruquintinib								1	0.92(0.45, 1.92)	0.66 (0.53, 0.84)
FTD/TPI + panitumumab									1	0.72(0.36, 1.40)
Placebo/BSC										1
PFS	FTD/TPI + bevacizumab	FTD/TPI	Cetuximab	Cetuximab + irinotecan	Panitumumab	Cetuximab + irinotecan + bevacizumab	Regorafenib	Fruquintinib	FTD/TPI + panitumumab	Placebo/BSC
FTD/TPI + bevacizumab	1	0.46 (0.34, 0.64)	0.32 (0.20, 0.53)	0.43 (0.19, 0.98)	0.35 (0.22, 0.59)	0.68(0.19, 2.49)	0.49 (0.31, 0.84)	0.71(0.45, 1.15)	0.96(0.49, 1.96)	0.21 (0.14, 0.31)
FTD/TPI		1	0.69(0.46, 1.01)	0.93(0.43, 1.99)	0.76(0.51, 1.12)	1.47(0.41, 5.14)	1.05(0.74, 1.60)	1.53 (1.07, 2.16)	2.07 (1.11, 3.92)	0.45 (0.35, 0.57)
Cetuximab			1	1.35(0.70, 2.64)	1.09(0.81, 1.50)	2.13(0.64, 7.14)	1.51 (1.03, 2.42)	2.22 (1.50, 3.31)	3.00 (1.46, 6.32)	0.65 (0.48, 0.88)
Cetuximab + irinotecan				1	0.81(0.39, 1.68)	1.58(0.57, 4.31)	1.13(0.52, 2.52)	1.64(0.76, 3.57)	2.23(0.83, 5.97)	0.48 (0.23, 0.99)
Panitumumab					1	1.95(0.57, 6.69)	1.39(0.94, 2.20)	2.03 (1.35, 3.03)	2.75 (1.34, 5.77)	0.59 (0.43, 0.81)
Cetuximab + irinotecan + bevacizumab						1	0.72(0.20, 2.62)	1.04(0.29, 3.69)	1.42(0.34, 5.89)	0.30(0.09, 1.06)
Regorafenib							1	1.46(0.95, 2.07)	1.97(0.93, 4.04)	0.43 (0.31, 0.55)
Fruquintinib								1	1.36(0.66, 2.81)	0.29 (0.23, 0.38)
FTD/TPI + panitumumab									1	0.22 (0.11, 0.42)
Placebo/BSC										1

Each cell represents the comparison (HR and 95% CrI) of the row treatment versus the column treatment. All bolded values are statistically meaningful at the 0.05 significance level. OS: deviance information criterion: 27.41; deviance: 15.09; standard deviation: 0.12. PFS: deviance information criterion: 28.03; deviance: 15.28; standard deviation: 0.14.

BSC, best supportive care; CrI, credible interval; FTD/TPI, trifluridine/tipiracil; HR, hazard ratio; NMA, network meta-analysis; OS, overall survival; PFS, progression-free survival.

Progression-free survival

FTD/TPI + bevacizumab had statistically favorable effects on PFS relative to FTD/TPI (HR: 0.46, 95% CrI: 0.34–0.64), cetuximab (HR: 0.32, 95% CrI: 0.20–0.53), cetuximab + irinotecan (HR: 0.43, 95% CrI: 0.19–0.98), panitumumab (HR: 0.35, 95% CrI: 0.22–0.59), regorafenib (HR: 0.49, 95% CrI: 0.31–0.84), and placebo/BSC (HR: 0.21, 95% CrI: 0.14–0.31) in the random-effects NMA model (Table 2, bottom). Similar results were found in the fixed-effect model (FTD/TPI + bevacizumab vs FTD/TPI (HR: 0.45, 95% CrI: 0.38–0.54), cetuximab (HR: 0.32, 95% CrI: 0.25–0.42), cetuximab + irinotecan (HR: 0.43, 95% CrI: 0.23–0.82), panitumumab (HR: 0.34, 95% CrI: 0.26–0.45), regorafenib (HR: 0.46, 95% CrI: 0.35, 0.60), fruquintinib (HR: 0.70, 95% CrI: 0.54–0.91), and placebo/BSC (HR: 0.21, 95% CrI: 0.17–0.26; Supplemental Table 17). FTD/TPI + bevacizumab also had numerically favorable effects on PFS relative to cetuximab + irinotecan + bevacizumab and fruquintinib in the random-effects model and relative to cetuximab + irinotecan + bevacizumab in the fixed-effects model, but these comparisons did not achieve statistical importance.

Discussion

Converging evidence indicates that the anti-cancer efficacy of FTD/TPI can be synergistically boosted by its delivery in combination with bevacizumab.¹³ A preclinical study using CRC xenografts reports that FTD/TPI + bevacizumab inhibits tumor growth to a greater extent than FTD/TPI or bevacizumab alone, possibly because bevacizumab promotes the accumulation and phosphorylation of FTD in tumors by normalizing tumor vasculature.⁷⁹ The recent phase III SUNLIGHT trial demonstrated that refractory mCRC patients treated with FTD/TPI + bevacizumab had significantly longer PFS and OS than patients treated with FTD/TPI alone.¹⁵ Also, meta-analyses of real-world studies support the effectiveness of FTD/TPI + bevacizumab outside the clinical trial setting.^80,81 Based on this evidence, FTD/TPI with or without bevacizumab is recommended by current NCCN and ESMO guidelines, offering the highest level of evidence and magnitude of clinical benefit among available treatments.^3–6 However, the relative efficacy of FTD/TPI + bevacizumab compared with other treatment regimens for patients with refractory mCRC requires further evidence. Therefore, the objective of this SLR and NMA was to identify and quantitatively synthesize evidence on the relative efficacy of FTD/TPI + bevacizumab compared with other treatments for refractory mCRC patients.

Of the 28 RCTs included in the SLR, 16 trials connected in a network with FTD/TPI + bevacizumab and were deemed sufficiently similar for inclusion in NMA. The results of this NMA showed that FTD/TPI + bevacizumab had statistical superiority over placebo/BSC, FTD/TPI, cetuximab, cetuximab + irinotecan, panitumumab, and regorafenib for both OS and PFS in both random-effects and fixed-effects models. FTD/TPI + bevacizumab also had statistical superiority over fruquintinib for OS in both random-effects and fixed-effects models and PFS in the fixed-effects model as well as over cetuximab + irinotecan for OS in the fixed-effects model. In addition, FTD/TPI + bevacizumab had a numerically favorable effect over FTD/TPI + panitumumab for OS and over cetuximab + irinotecan + bevacizumab for PFS in both random-effects and fixed-effects models. Thus, although the results of random-effects models are generally preferred because their assumptions (e.g., variance in effect size due to between-trial heterogeneity) are more plausible, we note that both random-effects and fixed-effects models produced similar results, underscoring the robustness of the present NMA results to different modeling approaches.

Because HRs between treatments evaluated in a network of evidence may change over time, the assumption of proportional hazards may over- or under-estimate the advantage of different treatments at particular timepoints. However, as only 3 of the 16 trials in the network showed violations of the proportional hazards assumption, and long-term OS in mCRC is generally low,⁸² this assumption may be unlikely to have a major impact on the NMA results. In addition, the results of the present NMA align with those of a recently published NMA that used a time-varying approach to evaluate the comparative effectiveness of later-line treatments for mCRC in terms of 12-month PFS and OS, which reports that FTD/TPI + bevacizumab is the most dominant treatment option compared with regorafenib, FTD/TPI monotherapy, fruquintinib, atezolizumab + cobimetinib, and atezolizumab monotherapy.⁸³

This study has several strengths that maximize its comprehensiveness and robustness. The SLR involved highly sensitive searches in the peer-reviewed literature as well as searches of recent conferences and clinical trial registries. The review process was guided by pre-defined eligibility criteria established in the SLR protocol. Data quality was ensured through the involvement of two reviewers in the study selection and data extraction phases of the project. After identifying relevant trials in the SLR, detailed feasibility assessment was carried out to identify trials that could serve as sources of bias in the NMA. Although between-trial heterogeneity modifying the relative treatment effects of included trials has the potential to bias the analysis, efforts were made to explore potential effect modifiers such as ethnicity and line of treatment to ensure differences between trials did not bias the results or contribute to inconsistency in direct versus indirect evidence of treatment effects, and random-effects models were run in addition to fixed-effect models to account for any uncertainty arising from between-trial heterogeneity.

However, the conclusions drawn from this study may be limited by some considerations. First, although current NCCN and ESMO guidelines recommend the use of cetuximab and panitumumab only for CRC patients with left-sided and KRAS/NRAS/BRAF wild-type tumors,^3–5 these patient characteristics were not reported by all trials evaluating cetuximab- or panitumumab-containing regimens, and available data indicate that at least some trials did not deliver cetuximab or panitumumab according to the recommendations. Second, two included trials (CORRECT and CONCUR) evaluating the same comparison, regorafenib versus placebo, reported substantially different results (HRs for OS: 0.55 vs 0.77, respectively), implying some differences between the trial design or population and potential effect modification that may give rise to additional uncertainty in NMA results for this comparison (i.e., wider CrIs). Third, although there is some evidence for the efficacy of chemotherapy re-challenge in the third-line setting,⁸⁴ there are currently no RCTs comparing chemotherapy re-challenge regimens with other treatments for refractory mCRC patients. Thus, the relative efficacy of chemotherapy re-challenge could not be evaluated in the present NMA. Fourth, although we performed a detailed feasibility assessment exploring potential treatment effect modifiers to ensure that differences between trials did not bias the NMA results, some residual confounding may have been present in the analysis due to the lack of reporting on certain patient characteristics by all trials. Fifth, as with any SLR, the evidence base continues to evolve quickly and, as such, any trials published after the search date were not captured. Six, there is a risk of publication bias, as some clinical trial results fail to be published. However, our study included an extensive search of conference abstract and clinical trial registries to mitigate the potential impact of publication bias. Seventh, the search and selection of relevant trials were restricted to publications in English, although most major clinical trials are expected to be published in English. Finally, the analysis did not include quality of life, which is an essential measure for evaluating the treatment of patients with refractory metastatic disease.

Conclusion

Overall, the results of this SLR and NMA suggest that FTD/TPI + bevacizumab confers important improvements in efficacy in terms of OS and PFS compared with other treatment regimens for refractory mCRC patients. These findings provide additional support for current guideline recommendations for the use of this combination therapy in the third-line treatment setting.

Supplemental Material

sj-docx-1-tam-10.1177_17588359251400882 – Supplemental material for Clinical outcomes of treatments for patients with refractory metastatic colorectal cancer: a systematic literature review and network meta-analysis

Supplemental material, sj-docx-1-tam-10.1177_17588359251400882 for Clinical outcomes of treatments for patients with refractory metastatic colorectal cancer: a systematic literature review and network meta-analysis by Julien Taieb, Sana Yahiaoui, Elias Choucair, Weiyu Yao, Ole Hauch, Katherine G. Akers, Andrew M. Frederickson and Aziz Zaanan in Therapeutic Advances in Medical Oncology

Footnotes

Acknowledgements

We thank Dweeti Nayak (Precision AQ) and Ina Zhang (Precision AQ) for their contributions to this study.

Declarations

ORCID iD

Katherine G. Akers

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Supplemental material

Supplemental material for this article is available online.

References

Keum

Giovannucci

Global burden of colorectal cancer: emerging trends, risk factors and prevention strategies. Nat Rev Gastroenterol Hepatol 2019; 16: 713–732.

Global colorectal cancer burden in 2020 and projections to 2040. Transl Oncol 2021; 14: 101174.

NCCN. NCCN clinical practice guidelines in oncology: colon cancer, Version 5.2024, https://www.nccn.Org/professionals/physician_gls/pdf/colon.pdf (2024, accessed 22 August 2024).

NCCN. NCCN clinical practice guidelines in oncology: rectal cancer, Version 4.2024, https://www.nccn.Org/professionals/physician_gls/pdf/rectal.pdf (2024, accessed 22 August 2024).

Cervantes

Adam

Roselló

, et al. Metastatic colorectal cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol 2023; 34: 10–32.

ESMO. ESMO-MCBS scorecards: trifluridine/tipiracil (TAS-102), SUNLIGHT, https://www.esmo.org/guidelines/esmo-mcbs/esmo-mcbs-for-solid-tumours/esmo-mcbs-scorecards/scorecard-382-1 (2024, accessed 22 August 2024).

ESMO. ESMO-MCBS scorecards: fruquintinib. FRESCO, https://www.esmo.org/guidelines/esmo-mcbs/esmo-mcbs-for-solid-tumours/esmo-mcbs-scorecards/scorecard-413-1 (2023, accessed 22 August 2024).

ESMO. ESMO-MCBS scorecards: fruquintinib. FRESCO-2, https://www.esmo.org/guidelines/esmo-mcbs/esmo-mcbs-for-solid-tumours/esmo-mcbs-scorecards?scorecard=412 (2023, accessed 22 August 2024).

ESMO. ESMO-MCBS scorecards: regorafenib. CORRECT, https://www.esmo.org/guidelines/esmo-mcbs/esmo-mcbs-for-solid-tumours/esmo-mcbs-scorecards/scorecard-26-1 (2024, accessed 22 August 2024).

10.

Uboha

Hochster

HS.

TAS-102: a novel antimetabolite for the 21st century. Future Oncol 2016; 12: 153–163.

11.

Mayer

Ohtsu

Yoshino

, et al. TAS-102 versus placebo plus best supportive care in patients with metastatic colorectal cancer refractory to standard therapies: final survival results of the phase III RECOURSE trial. J Clin Oncol 2016; 34: 634.

12.

Mayer

Van Cutsem

Falcone

, et al. Randomized trial of TAS-102 for refractory metastatic colorectal cancer. N Engl J Med 2015; 372: 1909–1919.

13.

Peeters

Cervantes

Moreno Vera

, et al. Trifluridine/tipiracil: an emerging strategy for the management of gastrointestinal cancers. Future Oncol 2018; 14: 1629–1645.

14.

Pfeiffer

Yilmaz

Möller

, et al. TAS-102 with or without bevacizumab in patients with chemorefractory metastatic colorectal cancer: an investigator-initiated, open-label, randomised, phase 2 trial. Lancet Oncol 2020; 21: 412–420.

15.

Prager

Taieb

Fakih

, et al. Trifluridine–tipiracil and bevacizumab in refractory metastatic colorectal cancer. N Engl J Med 2023; 388: 1657–1667.

16.

Dias

Welton

Sutton

, et al. Evidence synthesis for decision making 4: inconsistency in networks of evidence based on randomized controlled trials. Med Decis Making 2013; 33: 641–656.

17.

Jansen

Naci

Is network meta-analysis as valid as standard pairwise meta-analysis? It all depends on the distribution of effect modifiers. BMC Med 2013; 11: 159.

18.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021; 372: n71.

19.

NICE. Appendix B methodology checklist: systematic reviews and meta-analyses. The Social Care Guidance Manual, 2016.

20.

O’Rourke

Ward

Thompson

Single-arm trials in systematic reviews: are they adequately captured by existing search filters?

Value Health 2016; 19: A358.

21.

Walter

Hawkins

Pollock

, et al. Systematic review and network meta-analyses of third-line treatments for metastatic colorectal cancer. J Cancer Res Clin Oncol 2020; 146: 2575–2587.

22.

Sterne

JAC

Savović

Page

, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019; 366: l4898.

23.

Cope

Zhang

Saletan

, et al. A process for assessing the feasibility of a network meta-analysis: a case study of everolimus in combination with hormonal therapy versus chemotherapy for advanced breast cancer. BMC Med 2014; 12: 1–17.

24.

Cooper

Sutton

Morris

, et al. Addressing between-study heterogeneity and inconsistency in mixed treatment comparisons: application to stroke prevention treatments in individuals with non-rheumatic atrial fibrillation. Stat Med 2009; 28: 1861–1881.

25.

Jansen

Fleurence

Devine

, et al. Interpreting indirect treatment comparisons and network meta-analysis for health-care decision making: report of the ISPOR Task Force on Indirect Treatment Comparisons Good Research Practices: part 1. Value Health 2011; 14: 417–428.

26.

Dias

Welton

Sutton

, et al. NICE DSU technical support document 2: a generalised linear modelling framework for pairwise and network meta-analysis of randomised controlled trials. London: National Institute for Health and Care Excellence, https://www.ncbi.nlm.nih.gov/books/NBK310366/ (2014, accessed 22 August 2024).

27.

Longo-Muñoz

Argiles

Tabernero

, et al. Efficacy of trifluridine and tipiracil (TAS-102) versus placebo, with supportive care, in a randomized, controlled trial of patients with metastatic colorectal cancer from Spain: results of a subgroup analysis of the phase 3 RECOURSE trial. Clin Transl Oncol 2017; 19: 227–235.

28.

Van Cutsem

Mayer

Laurent

, et al. The subgroups of the phase III RECOURSE trial of trifluridine/tipiracil (TAS-102) versus placebo with best supportive care in patients with metastatic colorectal cancer. Eur J Cancer 2018; 90: 63–72.

29.

Tabernero

Argiles

Sobrero

, et al. Effect of trifluridine/tipiracil in patients treated in RECOURSE by prognostic factors at baseline: an exploratory analysis. ESMO Open 2020; 5: e000752.

30.

Pietrantonio

Fucà

Manca

, et al. Validation of the Colon Life nomogram in patients with refractory metastatic colorectal cancer enrolled in the RECOURSE trial. Tumori 2021; 107: 353–359.

31.

Dettori

Norvell

Chapman

. Fixed-effect vs random-effects models for meta-analysis: 3 points to consider. Global Spine J 2022; 12: 1624–1626.

32.

The R Core Team. R: A language and environment for statistical computing, Version 4.5.2 (2025-10-31). R Foundation for Statistical Computing, https://cran.r-project.org/doc/manuals/r-release/fullrefman.pdf (2025, accessed 22 August 2024).

33.

Hespanhol

Vallio

Costa

, et al. Understanding and interpreting confidence and credible intervals around effect estimates. Braz J Phys Ther 2019; 23: 290–301.

34.

Van Cutsem

Peeters

Siena

, et al. Open-label phase III trial of panitumumab plus best supportive care compared with best supportive care alone in patients with chemotherapy-refractory metastatic colorectal cancer. J Clin Oncol 2007; 25: 1658–1664.

35.

Siena

Peeters

Van Cutsem

, et al. Association of progression-free survival with patient-reported outcomes and survival: results from a randomised phase 3 trial of panitumumab. Br J Cancer 2007; 97: 1469–1474.

36.

Amado

Wolf

Peeters

, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol 2008; 26: 1626–1634.

37.

Peeters

Siena

Van Cutsem

, et al. Association of progression-free survival, overall survival, and patient-reported outcomes by skin toxicity and KRAS status in patients receiving panitumumab monotherapy. Cancer 2009; 115: 1544–1554.

38.

Wang

Zhao

Barber

, et al. A Q-TWiST analysis comparing panitumumab plus best supportive care (BSC) with BSC alone in patients with wild-type KRAS metastatic colorectal cancer. Br J Cancer 2011; 104: 1848–1853.

39.

Poulin-Costello

Azoulay

Van Cutsem

, et al. An analysis of the treatment effect of panitumumab on overall survival from a phase 3, randomized, controlled, multicenter trial (20020408) in patients with chemotherapy refractory metastatic colorectal cancer. Target Oncol 2013; 8: 127–136.

40.

Price

Kim

, et al. Final results and outcomes by prior bevacizumab exposure, skin toxicity, and hypomagnesaemia from ASPECCT: randomized phase 3 non-inferiority study of panitumumab versus cetuximab in chemorefractory wild-type KRAS exon 2 metastatic colorectal cancer. Eur J Cancer 2016; 68: 51–59.

41.

Price

Peeters

Kim

, et al. Panitumumab versus cetuximab in patients with chemotherapy-refractory wild-type KRAS exon 2 metastatic colorectal cancer (ASPECCT): a randomised, multicentre, open-label, non-inferiority phase 3 study. Lancet Oncol 2014; 15: 569–579.

42.

Kim

Peeters

Thomas

, et al. Impact of emergent circulating tumor DNA RAS mutation in panitumumab-treated chemoresistant metastatic colorectal cancer. Clin Cancer Res 2018; 24: 5602–5609.

43.

Peeters

Price

Boedigheimer

, et al. Evaluation of emergent mutations in circulating cell-free DNA and clinical outcomes in patients with metastatic colorectal cancer treated with panitumumab in the ASPECCT study. Clin Cancer Res 2019; 25: 1216–1225.

44.

Price

Ang

Boedigheimer

, et al. Frequency of S492R mutations in the epidermal growth factor receptor: analysis of plasma DNA from patients with metastatic colorectal cancer treated with panitumumab or cetuximab monotherapy. Cancer Biol Ther 2020; 21: 891–898.

45.

Lipsyc-Sharf

F-S

Yurgelun

, et al. Cetuximab and irinotecan with or without bevacizumab in refractory metastatic colorectal cancer: BOND-3, an ACCRU network randomized clinical trial. Oncologist 2022; 27: 292–298.

46.

Jonker

O’Callaghan

Karapetis

, et al. Cetuximab for the treatment of colorectal cancer. N Engl J Med 2007; 357: 2040–2048.

47.

Karapetis

Jonker

Daneshmand

, et al. PIK3CA, BRAF, and PTEN status and benefit from cetuximab in the treatment of advanced colorectal cancer—results from NCIC CTG/AGITG CO. 17. Clin Cancer Res 2014; 20: 744–753.

48.

Asmis

Powell

Karapetis

, et al. Comorbidity, age and overall survival in cetuximab-treated patients with advanced colorectal cancer (ACRC)—results from NCIC CTG CO.17: a phase III trial of cetuximab versus best supportive care. Ann Oncol 2011; 22: 118–126.

49.

Harbison

Horak

Ledeine

J-M

, et al. Validation of companion diagnostic for detection of mutations in codons 12 and 13 of the KRAS gene in patients with metastatic colorectal cancer: analysis of the NCIC CTG CO.17 trial. Arch Pathol Lab Med 2013; 137: 820–827.

50.

Karapetis

Khambata-Ford

Jonker

, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med 2008; 359: 1757–1765.

51.

Brule

Jonker

Karapetis

, et al. Location of colon cancer (right-sided versus left-sided) as a prognostic factor and a predictor of benefit from cetuximab in NCIC CO.17. Eur J Cancer 2015; 51: 1405–1414.

52.

Loree

Dowers

, et al. Expanded low allele frequency RAS and BRAF V600E testing in metastatic colorectal cancer as predictive biomarkers for cetuximab in the randomized CO.17 trial. Clin Cancer Res 2021; 27: 52–59.

53.

Jonker

Karapetis

Harbison

, et al. Epiregulin gene expression as a biomarker of benefit from cetuximab in the treatment of advanced colorectal cancer. Br J Cancer 2014; 110: 648–655.

54.

Qin

, et al. Regorafenib plus best supportive care versus placebo plus best supportive care in Asian patients with previously treated metastatic colorectal cancer (CONCUR): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol 2015; 16: 619–629.

55.

Qin

, et al. Regorafenib in Chinese patients with metastatic colorectal cancer: subgroup analysis of the phase 3 CONCUR trial. J Gastroenterol Hepatol 2020; 35: 1307–1316.

56.

Grothey

Van Cutsem

Sobrero

, et al. Regorafenib monotherapy for previously treated metastatic colorectal cancer (CORRECT): an international, multicentre, randomised, placebo-controlled, phase 3 trial. Lancet 2013; 381: 303–312.

57.

Tabernero

Lenz

H-J

Siena

, et al. Analysis of circulating DNA and protein biomarkers to predict the clinical activity of regorafenib and assess prognosis in patients with metastatic colorectal cancer: a retrospective, exploratory analysis of the CORRECT trial. Lancet Oncol 2015; 16: 937–948.

58.

Ricotta

Verrioli

Ghezzi

, et al. Radiological imaging markers predicting clinical outcome in patients with metastatic colorectal carcinoma treated with regorafenib: post hoc analysis of the CORRECT phase III trial (RadioCORRECT study). ESMO open 2016; 1: e000111.

59.

Vienot

Vernerey

Bouard

, et al. SO-20 stanniocalcin 1 (STC1) in patients with refractory colorectal cancer (CRC) treated with regorafenib: an exploratory analysis of the CORRECT trial. Ann Oncol 2022; 33: S365.

60.

Qin

R-H

, et al. Effect of fruquintinib vs placebo on overall survival in patients with previously treated metastatic colorectal cancer: the FRESCO randomized clinical trial. JAMA 2018; 319: 2486–2496.

61.

Guo

Bai

, et al. Safety profile and adverse events of special interest for fruquintinib in Chinese patients with previously treated metastatic colorectal cancer: analysis of the phase 3 FRESCO trial. Adv Ther 2020; 37: 4585–4598.

62.

Qin

R-H

Shen

, et al. Subgroup analysis by liver metastasis in the FRESCO trial comparing fruquintinib versus placebo plus best supportive care in Chinese patients with metastatic colorectal cancer. Onco Targets Ther 2021; 14: 4439–4450.

63.

Qin

Guo

, et al. Subgroup analysis by prior anti-VEGF or anti-EGFR target therapy in FRESCO, a randomized, double-blind, phase III trial. Future Oncol 2021; 17: 1339–1350.

64.

Dasari

Lonardi

Garcia-Carbonero

, et al. Subgroup analyses of safety and efficacy by number and types of prior lines of treatment in FRESCO-2, a global phase III study of fruquintinib in patients with refractory metastatic colorectal cancer. J Clin Oncol 2023; 41: 3604–3604.

65.

Dasari

Lonardi

Garcia-Carbonero

, et al. Fruquintinib versus placebo in patients with refractory metastatic colorectal cancer (FRESCO-2): an international, multicentre, randomised, double-blind, phase 3 study. Lancet 2023; 402: 41–53.

66.

Segelov

Thavaneswaran

Waring

, et al. Response to cetuximab with or without irinotecan in patients with refractory metastatic colorectal cancer harboring the KRAS G13D mutation: Australasian Gastro-Intestinal Trials Group ICECREAM Study. J Clin Oncol 2016; 34: 2258–2264.

67.

Pfeiffer

Winther

Möller

, et al. P-241 Neutropenia and survival outcomes in metastatic colorectal cancer patients treated with trifluridine/tipiracil plus bevazicumab: final analysis from a Danish randomised study. Ann Oncol 2023; 34: S101.

68.

Tabernero

Prager

Fakih

, et al. Trifluridine/tipiracil plus bevacizumab for third-line treatment of refractory metastatic colorectal cancer: the phase 3 randomized SUNLIGHT study. J Clin Oncol 2023; 41: 4.

69.

Servier. Clinical study report, 2022.

70.

Taieb

Prager

Fakih

, et al. Effect of trifluridine/tipiracil in combination with bevacizumab on ECOG-PS in refractory metastatic colorectal cancer: an analysis of the phase 3 SUNLIGHT trial. J Clin Oncol 2023; 41: 3594–3594.

71.

Kim

Shen

, et al. Results of a randomized, double-blind, placebo-controlled, phase III trial of trifluridine/tipiracil (TAS-102) monotherapy in Asian patients with previously treated metastatic colorectal cancer: the TERRA study. J Clin Oncol 2018; 36: 350–358.

72.

Napolitano

De Falco

Martini

, et al. Panitumumab plus trifluridine/tipiracil as third-line anti-EGFR rechallenge therapy in chemo-refractory RAS WT metastatic colorectal cancer: the VELO randomized phase II clinical trial. J Clin Oncol 2023; 41: 129.

73.

Ciardiello

Napolitano

De Falco

, et al. SO-24 Panitumumab plus trifluridine/tipiracil as anti-EGFR rechallenge therapy in patients with refractory RAS wild-type metastatic colorectal cancer: overall survival and subgroup analysis of the randomized phase 2 VELO trial. Ann Oncol 2023; 34: S171–S172.

74.

Napolitano

Ciardiello

De Falco

, et al. Panitumumab plus trifluridine/tipiracil as anti-EGFR rechallenge therapy in patients with refractory RAS wild-type metastatic colorectal cancer: overall survival and subgroup analysis of the randomized phase II VELO trial. Int J Cancer 2023; 153: 1520–1528.

75.

Napolitano

De Falco

Martini

, et al. Panitumumab plus trifluridine-tipiracil as anti-epidermal growth factor receptor rechallenge therapy for refractory RAS wild-type metastatic colorectal cancer: a phase 2 randomized clinical trial. JAMA Oncol 2023; 9(7): 966–970.

76.

Bai

, et al. Safety and efficacy of fruquintinib in patients with previously treated metastatic colorectal cancer: a phase Ib study and a randomized double-blind phase II study. J Hematol Oncol 2017; 10: 22.

77.

Yoshino

Mizunuma

Yamazaki

, et al. TAS-102 monotherapy for pretreated metastatic colorectal cancer: a double-blind, randomised, placebo-controlled phase 2 trial. Lancet Oncol 2012; 13: 993–1001.

78.

FDA. FDA approves new dosing regimen for cetuximab, https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-new-dosing-regimen-cetuximab (2021, accessed 22 August 2024).

79.

Tsukihara

Nakagawa

Sakamoto

, et al. Efficacy of combination chemotherapy using a novel oral chemotherapeutic agent, TAS-102, together with bevacizumab, cetuximab, or panitumumab on human colorectal cancer xenografts. Oncol Rep 2015; 33: 2135–2142.

80.

Voutsadakis

IA.

A systematic review and meta-analysis of trifluridine/tipiracil plus bevacizumab for the treatment of metastatic colorectal cancer: evidence from real-world series. Curr Oncol 2023; 30: 5227–5239.

81.

Aquino de Moraes

Dantas Leite Pessôa

Duarte de Castro Ribeiro

, et al. Trifluridine–tipiracil plus bevacizumab versus trifluridine–tipiracil monotherapy for chemorefractory metastatic colorectal cancer: a systematic review and meta-analysis. BMC Cancer 2024; 24: 674.

82.

Biller

Schrag

Diagnosis and treatment of metastatic colorectal cancer: a review. JAMA 2021; 325: 669–685.

83.

Obeng-Kusi

Martin

Roe

, et al. Comparative efficacy of later-line therapies for metastatic colorectal cancer: a network meta-analysis of survival curves. Expert Rev Pharmacoecon Outcomes Res 2024; 24(8): 923–932.

84.

Duan

Zhu

Shi

, et al. Chemotherapy re-use versus anti-angiogenic monotherapy as the third-line treatment of patients with metastatic colorectal cancer: a real-world cohort study. BMC Cancer 2024; 24: 302.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.49 MB