Efficacy and Safety of Trifluridine/Tipiracil-Containing Combinations in Colorectal Cancer and Other Advanced Solid Tumors: A Systematic Review

Abstract We performed a systematic literature review to identify and summarize data from studies reporting clinical efficacy and safety outcomes for trifluridine/tipiracil (FTD/TPI) combined with other antineoplastic agents in advanced cancers, including metastatic colorectal cancer (mCRC). We conducted a systematic search on May 29, 2021, for studies reporting one or more efficacy or safety outcome with FTD/TPI-containing combinations. Our search yielded 1378 publications, with 38 records meeting selection criteria: 35 studies of FTD/TPI-containing combinations in mCRC (31 studies second line or later) and 3 studies in other tumor types. FTD/TPI plus bevacizumab was extensively studied, including 19 studies in chemorefractory mCRC. Median overall survival ranged 8.6-14.4 months and median progression-free survival 3.7-6.8 months with FTD/TPI plus bevacizumab in refractory mCRC. Based on one randomized and several retrospective studies, FTD/TPI plus bevacizumab was associated with improved outcomes compared with FTD/TPI monotherapy. FTD/TPI combinations with chemotherapy or other targeted agents were reported in small early-phase studies; preliminary data indicated higher antitumor activity for certain combinations. Overall, no safety concerns existed with FTD/TPI combinations; most common grade ≥ 3 adverse event was neutropenia, ranging 5%-100% across all studies. In studies comparing FTD/TPI combinations with monotherapy, grade ≥ 3 neutropenia appeared more frequently with combinations (29%-67%) vs. monotherapy (5%-41%). Discontinuation rates due to adverse events ranged 0%-11% for FTD/TPI plus bevacizumab and 0%-17% with other combinations. This systematic review supports feasibility and safety of FTD/TPI plus bevacizumab in refractory mCRC. Data on non-bevacizumab FTD/TPI combinations remain preliminary and need further validation.


Introduction
Fluoropyrimidines, including 5-fluorouracil and capecitabine, alone or as part of combination regimens, have formed the mainstay in treating gastrointestinal cancers. 1,2However, resistance to fluoropyrimidines remains a considerable barrier to effective treatment. 2rifluridine/tipiracil (FTD/TPI or TAS-102; Taiho Oncology, Inc., Princeton, NJ, USA) is an oral cytotoxic agent comprising e602 The Oncologist, 2024, Vol. 29, No. 5   trifluridine, a thymidine analog, and tipiracil, a thymidine phosphorylase inhibitor. 3FTD/TPI has a unique mechanism of action distinguishing it from other fluoropyrimidines. 2 FTD is incorporated into DNA, causing DNA dysfunction, and tipiracil increases the oral bioavailability of FTD.
In the phase III randomized RECOURSE trial, FTD/TPI monotherapy significantly improved survival versus placebo in patients with chemorefractory metastatic colorectal cancer (mCRC; after ≥ 2 prior systemic regimens).Median overall survival (OS) was 7.1 vs. 5.3 months for FTD/TPI versus placebo (hazard ratio [HR] 0.68 (95% confidence interval [CI], 0.58 − 0.81; P < .001).FTD/TPI also demonstrated a manageable safety profile, with hematologic and gastrointestinalrelated adverse events (AEs) being the most common. 4In the phase III randomized TAGS trial, FTD/TPI monotherapy was associated with a significant survival benefit versus placebo (median OS, 5.7 vs. 3.6 months [95% CI, 4.8-6.2];HR, 0.69 [95% CI, 0.56-0.85];P < .001) in patients with metastatic gastric or gastroesophageal junction cancer (mGC/ GEJC) whose disease progressed after ≥ 2 prior chemotherapy (chemo) regimens. 5As a result, FTD/TPI was approved as third-or later-line treatment for patients with mCRC (in 2015) and mGC/GEJC (in 2019). 6n addition to these two trials, multiple studies over recent years have evaluated the combination of FTD/TPI with targeted therapies, other chemotherapeutic agents, and immunotherapeutic agents, both in mCRC and other cancer types.The combination of FTD/TPI with the anti-vascular endothelial growth factor (VEGF) antibody bevacizumab (BEV) has shown promising results in patients with refractory mCRC in clinical trials, 7,8 and FTD/TPI + BEV is now recommended in the National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines in Oncology (NCCN Guidelines) as a treatment option for patients with this disease. 9,10hile there are systematic literature reviews (SLRs) and meta-analyses summarizing FTD/TPI monotherapy in CRC, [11][12][13] systematic reviews evaluating FTD/TPI-containing combination regimens across tumor types are rare.This SLR's objective is to identify and summarize data from studies reporting clinical efficacy and safety outcomes for FTD/TPI in combination with other antineoplastic agents in various cancers, including CRC.

Methods
Methods used in this unregistered SLR were prespecified and documented in a study protocol (supporting information).PRISMA reporting guidelines for systematic reviews 14,15 and the Cochrane Handbook for Systematic Reviews of Interventions, version 5.1.0guided reporting. 16

Screening and Study Selection
Inclusion and exclusion criteria developed using the PICOS approach were applied to shortlist publications of interest for studies reporting clinical efficacy, safety, and health-related quality of life (HRQoL) outcomes for FTD/TPI combined with other antineoplastic agents to treat patients with cancer (Supplementary Table S4).Using these criteria, 2 researchers independently screened abstracts and then full-text articles in a 2-stage process, with a third reviewer adjudicating any differences.Only studies that fulfilled the inclusion and exclusion criteria and reported one or more defined outcomes were included in the analysis.Identification of records is shown (Fig. 1), aligned with PRISMA statement recommendations. 14

Qualitative Synthesis of Data
Findings were tabulated and summarized.One reviewer extracted data from the included records into Microsoft Excel tables, which were then checked and validated by a second reviewer.Outputs included a trial design overview, patient population (tumor type and stage), sample size, follow-up duration, intervention, comparator, primary and secondary endpoints including but not limited to clinical efficacy, safety, and HRQoL.Results were collated and studies were characterized by cancer type, line of treatment, and finally by the type of FTD/TPI combination partner.As this analysis was designed to provide a qualitative review, median values and ranges were provided for survival outcomes.
The study followed a prespecified protocol as a qualitative rather than quantitative analysis.A quantitative metaanalysis was not conducted due to considerable differences between the studies in terms of study design, disease, intervention, and line of treatment.For various combinations of disease, interventions, and lines of treatment, there was an insufficient number of studies for all the groups other than for the FTD/TPI and BEV combination in 2L+ or 3L+ mCRC.Patient characteristics in these individual studies were sporadically reported, and for those that were reported, too diverse to be appropriately combined in a meta-analysis.The outcomes being reported were summarized as median timeto-event endpoints (OS/PFS), which often follow skewed distributions and assumptions valid in meta-analyses of means that may not have been appropriate for medians.For these reasons, a qualitative synthesis of the data was performed.

Risk of Bias Quality Assessment
Criteria from the Cochrane risk-of-bias tool for RCTs 16,17 and the Downs and Black checklist for non-RCTs 17 were used to assess the risk of bias in RCTs and non-RCTs, respectively.In addition, the Newcastle-Ottawa Scale was used to evaluate single-arm observational studies. 18Two reviewers assessed risk of bias, with disagreement resolved by discussion.

Study Selection
In total, 1378 records were identified across all databases searched (Fig. 1).Of these, 366 were duplicates between Overall, 31 records included studies of FTD/TPI-containing combinations used in second-or third-line or later (2L+ or 3L+) settings in mCRC, 7,8, 4 records detailed studies of FTD/TPI combinations in the first-line (1L) setting in mCRC (n = 3) [48][49][50] or where treatment setting was not reported in mCRC (n = 1), 51 2 were studies of FTD/TPI-containing combinations used in the 2L + setting in mGC/GEJC, 52,53 and one study evaluated FTD/TPI plus temozolamide (TEM) in patients with advanced neuroendocrine tumors (previous treatment unspecified). 54 Daa in mCRC, and particularly those in previously treated mCRC, were summarized separately for this analysis, as this constituted the largest subset of the results.
The triplet combination of BEV + FTD/TPI + another chemotherapeutic agent was evaluated in three small studies, with two reporting efficacy outcomes.A median OS of 15.1 months and median PFS of 6.3 months were noted with FTD/TPI + BEV + OXA in 3L+ mCRC (Figs. 2 and 3; Supplementary Tables S7 and S8), 32 and a median PFS of 7.9 months was noted with FTD/TPI + BEV + IRI in 2L+ mCRC. 40CRs ranged from 83% to 89% among patients treated with FTD/TPI + BEV + chemo in the 2L+ and 3L+ settings (Supplementary Fig. S2B).
The combination of FTD/TPI + NIVO + OXA was associated with a median PFS of six months, and an ORR of 7%, whereas FTD/TPI + NIVO was associated with a median PFS of 2.8 months and a 0% ORR (Figs. 2 and 3; Supplementary Tables S7 and S8). 32,456,47 In the phase II RCT comparing FTD/TPI + BEV with FTD/ TPI monotherapy, the incidence of grade 3 or 4 neutropenia was higher with FTD/TPI + BEV (67%) than with FTD/ TPI alone (38%). 8However, in this study, the overall rates of serious AEs were similar in the FTD/TPI and FTD/TPI + BEV groups, and similar numbers of patients discontinued because of AEs.Across retrospective observational studies that evaluated both FTD/TPI and FTD/TPI + BEV in patients with previously treated mCRC, 22,28,29,33,36,38 rates of grade 3 or higher neutropenia ranged from 5% to 41% in the FTD/TPI group and 29% to 53% in the FTD/TPI + BEV group.Other key AEs experienced by patients receiving FTD/TPI + BEV included fatigue, nausea, and diarrhea; however, incidences of grade ≥ 3 events of these AEs remained low (Table 2).The proportions of patients discontinuing treatment due to an AE were low, ranging from 0% to 11% across various studies.Among patients treated with other FTD/TPI combinations in both 2L+ and 3L+ settings, neutropenia was again the most frequent grade ≥ 3 AE observed (incidence ranging from 17% to 100%), and the proportions of patients discontinuing treatment due to an AE ranged from 0% to 17% (Table 3). 19,26,40,42,45Grade ≥ 3 neutropenia was reported at an Figure Median OS with FTD/TPI-containing combination regimens in patients with metastatic colorectal cancer in second-or third-line settings.*Data pertains to patients who received FTD/TPI ± BEV.Abbreviations: 2L, second line; 3L, third line; BEV, bevacizumab; chemo, chemotherapy; FTD/TPI, trifluridine/tipiracil; IRI, irinotecan; NA, not applicable; NIN, nintedanib; NIVO, nivolumab; OS, overall survival; OXA, oxaliplatin; PAN, panitumumab.

e608
The Oncologist, 2024, Vol. 29, No. 5 incidence of 100% with the combination of FTD/TPI with IRI at the highest dose level evaluated in a dose-escalation study. 19he standard recommended dose of FTD/TPI requires either no or slight adjustment when combined with other antineoplastic agents.A dose escalation study indicated that FTD/TPI is safe in treating mCRC at the recommended dose of 35 mg/m² bid in combination with 85 mg/m² of OXA Q2W, 26 while other studies indicated a lower maximum tolerated dose of FTD/TPI (25 mg/m² bid) when combined with 180 mg/m² IRI Q2W 40 or with 120 mg REG daily. 35

FTD/TPI + BEV in 1L mCRC
Two studies evaluated FTD/TPI + BEV in the 1L setting in mCRC (Table 1).In the noncomparative phase II RCT TASCO1, in patients with mCRC ineligible for full-dose combination chemotherapy with irinotecan or oxaliplatin or for curative resection of metastatic lesions, FTD/TPI + BEV (n = 77) and capecitabine + BEV (n = 76) were respectively associated with a median PFS of 9.2 and 7.8 months, median OS of 18 and 16.2 months, ORRs of 34% and 30%, and DCRs of 86% and 78%.In the FTD/TPI + BEV and capecitabine + BEV groups, most frequent grade ≥ 3 AEs, were neutropenia (47% and 5%), hand-foot syndrome (0% and 12%) and diarrhea (0% and 8%). 48In extended follow-up, median OS was 22.3 months with FTD/TPI + BEV and 17.7 months with capecitabine + BEV. 50,55Overall, these data indicated clinical activity of the FTD/TPI + BEV regimen in untreated mCRC, efficacy similar to that of capecitabine + BEV.In a smaller phase II trial in patients with mCRC aged ≥ 70 years, including those ineligible or eligible for (but opted not to receive) oxaliplatin-or irinotecan-containing regimens (n = 37), FTD/TPI + BEV as IL treatment resulted in a median OS of 22.4 months, median PFS of 9.4 months, an ORR of 41%, and a DCR of 87%.Most (72%) patients experienced grade ≥ 3 neutropenia. 49

FTD/TPI-Containing Combinations in Other Tumor Types
Two studies evaluated FTD/TPI-containing combination regimens in mGC/GEJC in the 2L+ setting.A phase II clinical trial that evaluated FTD/TPI + ramucirumab in patients with mGC/GEJC who were previously treated (n = 64) resulted in an ORR of 13% and a DCR of 81%.In total, 78% of patients experienced grade ≥ 3 neutropenia.The efficacy and safety data were consistent regardless of previous ramucirumab exposure, and the authors concluded that this regimen had clinical activity in this population. 53Preliminary results from a phase I/II clinical trial assessing FTD/TPI + IRI in patients with previously treated GC (n = 11) indicated a median PFS of 3 months, and a median OS of 10.2 months.Overall, 91% of patients had grade ≥ 3 neutropenia. 52eparately, one phase I dose-escalation trial evaluated FTD/ TPI + TEM in patients with advanced neuroendocrine tumors (n = 15; prior treatment unspecified).In this trial, an ORR of 8% and a DCR of 92% was observed; 33% experienced grade ≥ 3 neutropenia. 54

Discussion
To our knowledge, this SLR is the first to summarize all published studies of FTD/TPI-containing combination regimens across tumors.Most studies evaluating FTD/TPI-containing  The Oncologist, 2024, Vol. 29, No. 5
In contrast to data in 2L+ and 3L+ mCRC, FTD/TPI + BEV did not improve outcomes compared with standard of care in untreated mCRC. 48In the noncomparative phase II TASCO1 study, promising clinical activity and tolerability was noted with FTD/TPI + BEV 48,50,55 ; however, preliminary results from the ongoing comparative phase III SOLSTICE trial of FTD/ TPI + BEV versus capecitabine + BEV in previously untreated mCRC indicated that FTD/TPI + BEV was not superior to capecitabine + BEV in the 1L setting; median PFS was similar with both regimens (9.3 vs. 9.4 months; HR 0.87, 95% CI, 0.75-1.02). 56tudies evaluating FTD/TPI in combination with targeted agents other than BEV were less common 21,30,35,44 and among these, encouraging results were observed with FTD/ TPI + PAN, an epidermal growth factor receptor (EGFR) antibody, with a median PFS of ≈ 6 months and median OS of ≈ 14 months. 44Ongoing phase II/III studies are evaluating FTD/TPI + PAN in both 1L (NCT05007132 EudraCT: 2019-004223-20) and 3L+ settings (NCT05468892) in mCRC (Table 4).While results from other studies included in this analysis were less conclusive, 21,30,51 the combination of FTD/ TPI with non-BEV targeted therapies is an area of active research and ongoing studies are exploring FTD/TPI in combination with agents targeting EGFR, MEK, VEGFR, and HER2 receptors (Table 4).
Other FTD/TPI-containing combinations were less effective and are not being pursued in phase II/III studies.Although phase I studies showed that FTD/TPI combined with OXA or IRI were tolerable in 2L+/3L+ mCRC, preliminary activity reported in 3 studies of FTD/TPI + OXA was not favorable (median PFS, ≈2 months; ORR, 0%-4%). 26,39,42Efficacy data with FTD/TPI + IRI are largely lacking or inconclusive. 19imilarly, combining immunotherapeutic agents with FTD/ TPI has not proven to be efficacious: phase II studies of FTD/ TPI + NIVO 45 and FTD/TPI + NIVO + OXA 32 were prematurely halted because of lack of efficacy.
Our analysis identified few published studies of FTD/TPI combinations tumor types other than CRC, such as gastroesophageal cancers, which is not surprising given that FTD/ TPI monotherapy was only granted approval for this indication in 2019. 6Following the trend seen in mCRC, newly initiated or ongoing phases I and II studies in mGC/GEJC    4) indicated that adding BEV to FTD/TPI did not improve efficacy outcomes compared with FTD/TPI monotherapy for patients with pretreated metastatic esophagogastric adenocarcinoma, although OS and PFS benefits were seen with both regimens (median PFS, 3.7-3.9months; OS, 9.0-9.9months). 57hile FTD/TPI combinations in other tumor types are being explored, such as two prospective studies of FTD/ TPI + irinotecan in pretreated cholangiocarcinoma (Table 4), these efforts remain relatively rare.

Study
The safety profiles of FTD/TPI-containing combinations mostly showed a higher incidence of grade ≥ 3 hematologic AEs.Neutropenia was the most common grade ≥ 3 AE associated with FTD/TPI-containing combinations, with varying incidence across different regimens and studies.Our analysis indicated that grade ≥ 3 neutropenia may occur in up to 72% of patients in the 2L+ setting, 7,8,46 and up to 50% of patients in the 3L+ setting receiving FTD/TPI in combination with BEV. 28,29,37,38,43,47However, while neutropenia was more common with this combination than with FTD/TPI monotherapy, 8,38 the difference was not statistically significant, 38 and overall, treatment-related discontinuations were low in patients receiving FTD/TPI + BEV.Neutropenia has shown a dose-response relationship with FTD exposure, and higher FTD/TPI plasma levels are associated with improved OS and PFS, and a reduced time to performance status deterioration. 36,58][60] The incidence of non-hematologic grade ≥ 3 events remained low across studies in this analysis.Even in the recently reported phase III SOLSTICE trial, although the incidence of neutropenia was higher with the FTD/ TPI + BEV compared with capecitabine + BEV, the nonhematologic safety profile of the former was more favorable. 56Importantly, cardiotoxicity was not a concern with FTD/TPI-containing combinations, unlike with fluoropyrimidines. 61In phase III studies, the overall incidence of grade ≥ 3 cardiac events was rare 4,5 and a recent metaanalysis reported no increased cardiotoxicity risk with FTD/TPI compared with placebo. 62These data lend support for FTD/TPI as a drug of choice or as a good backbone for combination therapeutic regimens for patients with cardiac disease or cardiac side effects from previous chemotherapeutic regimens. 63Another patient subset that may benefit from FTD/TPI are those with dihydropyrimidine dehydrogenase (DPD) deficiency, as these patients are at risk of severe life-threatening AEs with 5-FU-containing regimens. 64As FTD/TPI is not metabolized by DPD, 65 FTD/ TPI is presumably safe for these patients.However, evidence of FTD/TPI's safety in patients with DPD deficiency is limited to a few case reports [66][67][68] and needs further clinical evaluation.
Quality-of-life (QoL) data were generally not reported in most studies included in this analysis, as these were early feasibility studies.One exception was the TASCO1 study of FTD/TPI + BEV in 1L mCRC.In this study, QoL was generally maintained during treatment, with no clinically relevant changes from baseline observed in global health status, functioning scales, and most symptom scales. 48Further data are needed to more fully establish the QoL impact of different FTD/TPI-based regimens, particularly among patients who were pretreated and/or over 65 years.
Limitations of this review were that the analysis captured phases I and II data only, with all but 2 studies being nonrandomized trials or retrospective observational studies.These studies were heterogeneous for combinations, treatment settings, and patient populations, which, in addition to the relatively small sample sizes (range, 9 to 97 patients), limited our ability to draw definite conclusions.Upcoming data from the phase III randomized SOLSTICE, 56 SUNLIGHT (NCT04737187), and COLSTAR (NCT05223673) trials, which will comprise larger datasets, should help further elucidate the role of FTD/TPI combinations in mCRC.Data on non-BEV combinations in mCRC were also limited.This analysis lacked data from other tumor types, particularly mGC/GEJC; however, the bulk of these studies are underway or were recently reported, including the phase III study of FTD/TPI + bevacizumab in platinum-refractory mGC/ GEJC (EudraCT: 2018-004845-18).Another possible limitation was that two-thirds (21/31) of the included studies were conducted in Japanese populations, highlighting the need for more data in diverse populations.Fortunately, a good number of ongoing FTD/TPI combination therapy trials are being conducted among populations in the US and Europe (Table 4).
Taken together, this comprehensive SLR consolidates the current body of evidence regarding FTD/TPI combinations in metastatic solid tumors and supports the feasibility and safety of certain FTD/TPI-containing combinations, such as FTD/ TPI + BEV, in the guideline-recommended setting of refractory mCRC.

Table 1 .
Studies of FTD/TPI-containing combination regimens in metastatic colorectal cancer.

phase Population specifics Intervention 1, n Intervention 2, n Treatment setting Reported study endpoints a Study location Median follow-up, mo
FTD/TPI + BEV

Table 2 .
Summary of safety in patients with metastatic colorectal cancer treated with FTD/TPI plus bevacizumab in second-or third-line settings.

Table 3 .
Summary of safety in patients with metastatic colorectal cancer treated with other FTD/TPI combinations in second-or third-line settings.

Table 4 .
Ongoing and recently reported of FTD/TPI in combination with other agents in mCRC and other advanced solid tumors (up to date July 2022).