Effectiveness of primary tumor resection for survival after first-line cetuximab or bevacizumab in KRAS wild-type metastatic colorectal cancer treated with subsequent trifluridine/tipiracil or regorafenib

Yu-Hsun Chen; Chih-Chien Wu; Chien-Chou Su; Pei-Ting Lee; Yi-Chia Su

doi:10.3393/ac.2025.00759.0108

Articles

Page Path: HOME > Ann Coloproctol > Volume 42(1); 2026 > Article

Original Article Colorectal cancer Effectiveness of primary tumor resection for survival after first-line cetuximab or bevacizumab in KRAS wild-type metastatic colorectal cancer treated with subsequent trifluridine/tipiracil or regorafenib: Yu-Hsun Chen¹, Chih-Chien Wu^1,2, Chien-Chou Su³, Pei-Ting Lee^4,5, Yi-Chia Su^4,6,7; Annals of Coloproctology 2026;42(1):127-140.
DOI: https://doi.org/10.3393/ac.2025.00759.0108
Published online: February 23, 2026

¹Division of Colorectal Surgery, Department of Surgery, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan

²Department of Surgery, School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan

³Clinical Innovation and Research Center, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan

⁴Department of Pharmacy, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan

⁵Department of Public Health, College of Medicine, National Cheng Kung University, Tainan, Taiwan

⁶Department of Pharmacy, School of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan

⁷Department of Nursing, Shu-Zen Junior College of Medicine and Management, Kaohsiung, Taiwan

Correspondence to: Yi-Chia Su, PhD Department of Pharmacy, Kaohsiung Veterans General Hospital, No. 386, Dazhong 1st Rd, Zuoying District, Kaohsiung 813414, Taiwan Email: jia.love70@vghks.gov.tw

• Received: June 14, 2025 • Revised: September 30, 2025 • Accepted: September 30, 2025

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

prev next

493 Views
22 Download

Full Article

Download PDF

Abstract
INTRODUCTION
METHODS
RESULTS
DISCUSSION
ARTICLE INFORMATION
REFERENCES

Abstract

Purpose
The optimal sequencing of targeted therapies and the role of primary tumor resection (PTR) in KRAS wild-type metastatic colorectal cancer (mCRC) remain unclear. This study compared survival outcomes in patients treated with first-line cetuximab plus FOLFIRI (folinic acid, 5-fluorouracil, and irinotecan) versus bevacizumab plus FOLFIRI, followed by second-line oxaliplatin-based chemotherapy and later-line trifluridine/tipiracil or regorafenib.
Methods
This retrospective cohort study used Taiwan’s National Health Insurance Research Database and the Taiwan Cancer Registry. Patients diagnosed with mCRC between 2013 and 2019 were included if they received first-line cetuximab or bevacizumab plus FOLFIRI, followed by later-line trifluridine/tipiracil or regorafenib. Patients were stratified by PTR status. Primary endpoints were overall survival and survival during trifluridine/tipiracil or regorafenib treatment. Secondary endpoints included time to treatment discontinuation (TTD) and TTD during trifluridine/tipiracil or regorafenib therapy. Stabilized inverse probability of treatment weighting was used for adjustment.
Results
Among 559 patients, 278 were assigned to the non-PTR group and 281 to the PTR group. In the non-PTR group, the cetuximab cohort demonstrated significantly longer survival during trifluridine/tipiracil or regorafenib therapy (6.2 months vs. 4.9 months; hazard ratio [HR], 0.72) and longer TTD1 (the interval between initiation of first-line therapy and the start of second-line chemotherapy; 11.8 months vs. 9.5 months; HR, 0.67) than the bevacizumab cohort. Survival differences between regimens were less pronounced among patients who underwent PTR.
Conclusion
First-line cetuximab plus FOLFIRI may confer a survival advantage over bevacizumab in patients with KRAS wild-type mCRC without PTR, including during later-line therapy with trifluridine/tipiracil or regorafenib, whereas bevacizumab appears to provide more consistent benefits in those with PTR.
Keywords: Colorectal neoplasms; Cetuximab; Bevacizumab

INTRODUCTION

Colorectal cancer (CRC) is a major global health concern and remains a leading cause of cancer-related mortality worldwide [1]. At diagnosis, approximately 20% to 25% of patients present with metastatic disease, most commonly involving the liver and lungs [2]. Among these individuals, about 80% to 90% are considered ineligible for surgical resection of either the primary tumor or distant metastases [3, 4]. Therefore, the selection of effective systemic therapy is crucial for managing patients with unresectable metastatic CRC (mCRC).

For newly diagnosed patients with KRAS wild-type unresectable mCRC, the standard initial approach typically involves targeted therapy in combination with chemotherapy. Cetuximab and bevacizumab are the 2 principal targeted agents currently approved for first-line use [5]. These agents are commonly combined with FOLFIRI (folinic acid, 5-fluorouracil, and irinotecan), which represents one of the most widely adopted first-line therapeutic strategies for this population [5–7]. Upon disease progression after first-line therapy, the recommended second-line regimen generally includes FOLFOX (5-fluorouracil and leucovorin combined with oxaliplatin), with or without continuation of the previously used targeted agent. In patients initially treated with bevacizumab plus FOLFIRI, bevacizumab plus FOLFOX or FOLFOX alone is often used as second-line therapy. After further progression, cetuximab is typically reimbursed as a third-line agent. In this treatment sequence, trifluridine/tipiracil or regorafenib is used as fourth-line therapy [8]. Conversely, for those who receive first-line cetuximab plus FOLFIRI followed by cetuximab plus FOLFOX or FOLFOX alone as second-line therapy, trifluridine/tipiracil or regorafenib serves as a third-line option [8]. Despite the clinical importance of treatment sequencing, evidence comparing downstream outcomes and optimal sequencing of trifluridine/tipiracil or regorafenib following first-line cetuximab-based versus bevacizumab-based therapy in KRAS wild-type mCRC remains limited. Two pivotal head-to-head trials, CALGB 80405 and FIRE-3, have directly compared these regimens [9, 10]. In both studies, roughly 80% of patients underwent primary tumor resection (PTR) before initiating targeted therapy, while the remaining 20% did not.

To date, no study has comprehensively examined treatment sequences extending to trifluridine/tipiracil or regorafenib after first-line cetuximab- or bevacizumab-based therapy in patients with KRAS wild-type mCRC. However, PTR status [11, 12] and variability in subsequent treatment strategies are recognized as key determinants of overall survival (OS). Therefore, this study aimed to evaluate the impact of PTR on survival outcomes, including OS and time to treatment discontinuation (TTD), in patients with KRAS wild-type mCRC who received first-line cetuximab plus FOLFIRI or bevacizumab plus FOLFIRI, followed by second-line oxaliplatin-based chemotherapy and subsequent trifluridine/tipiracil or regorafenib. The study also assessed how PTR differentially influenced outcomes across treatment sequences, with the goal of providing insights to guide clinical decision-making and optimize therapeutic strategies in this patient population.

METHODS

Ethics statement

The study protocol was approved by the Institutional Review Board of Kaohsiung Veterans General Hospital (No. KSVGH22-EM10–01). The requirement for informed consent was waived because all data from Taiwan’s National Health Insurance Research Database (NHIRD) are standardized, encrypted, and deidentified. All datasets were fully anonymized, and the study adhered to the ethical standards outlined in the principles of the Declaration of Helsinki. This study followed the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines.

Data source

The NHIRD in Taiwan is derived from the country’s single-payer National Health Insurance (NHI) system, which covers more than 99.99% of the population. This database provides comprehensive records of medication use, including bevacizumab, cetuximab, fluorouracil, irinotecan, and oxaliplatin, as well as data on surgical procedures. The Taiwan Cancer Registry (TCR), administered by Taiwan Ministry of Health and Welfare, provides cancer-related data with an approximate coverage rate of 97%. To ensure data accuracy, the TCR conducts an internal validation process using standardized logic checks at the central office to identify and correct inconsistencies prior to final entry. Both NHIRD and TCR were curated by the Health and Welfare Data Science Center under Taiwan Ministry of Health and Welfare supervision. Mortality data, including cause of death, were tracked through December 31, 2020.

Study population and therapies

The diagnosis of mCRC was based on the International Classification of Diseases for Oncology, 3rd edition codes C180–C189, C199, and C209. Eligible patients were those diagnosed between January 1, 2013, and December 31, 2019, recorded in the TCR, and who received first-line therapy with bevacizumab or cetuximab plus FOLFIRI. The index date for first-line therapy was defined as the date of the initial administration of bevacizumab or cetuximab combined with FOLFIRI. The index date for second-line therapy was defined as the date of the first administration of oxaliplatin-based chemotherapy. Second-line therapy was defined as any chemotherapy regimen differing from the first-line treatment. According to the national reimbursement policy, first-line chemotherapy for mCRC is limited to FOLFIRI-based regimens combined with targeted agents [8]. Therefore, the first-line regimen in this study consisted of a targeted agent plus FOLFIRI, while second-line therapy was defined as a switch to a non-FOLFIRI regimen, most commonly oxaliplatin-based. For patients receiving first-line bevacizumab plus FOLFIRI, third-line therapy was defined as the date of the first administration of a cetuximab-containing regimen, and fourth-line therapy as the date of the first administration of trifluridine/tipiracil or regorafenib. Conversely, for patients treated with first-line cetuximab plus FOLFIRI, third-line therapy was defined as the date of the first administration of either trifluridine/tipiracil or regorafenib.

Patients were included if they completed at least 6 cycles of first-line therapy with bevacizumab or cetuximab plus FOLFIRI, maintained intervals of fewer than 60 days between cycles, and received only one second-line regimen. Exclusion criteria were as follows: (1) age <20 years; (2) synchronous left- and right-sided tumors; (3) receipt of targeted therapy within 1 year before diagnosis; (4) metastasectomy prior to the index date; (5) KRAS gene mutation or missing KRAS data; (6) fewer than 6 consistent cycles of first-line therapy; (7) follow-up duration <3 months; (8) intercycle intervals >60 days; or (9) no receipt of trifluridine/tipiracil or regorafenib (Fig. 1).

Study variables and outcomes

Demographic variables included the year of targeted therapy, age, sex, histological grade and type (mucinous, signet-ring cell, or adenocarcinoma), primary tumor location, stage (IVA, IVB, or IVC), tumor size, carcinoembryonic antigen (CEA) status, bowel obstruction, bowel perforation, radiotherapy, metastasectomy type, surgical status prior to index date, and Charlson Comorbidity Index score based on the International Classification of Diseases, 9th Revision (ICD-9) [13, 14] and ICD-10 codes recorded within 1 year before the index date. Additional variables included hospital level (medical center or regional hospital), timing of metastasectomy, total duration and number of second-line oxaliplatin cycles, and whether second-line oxaliplatin was combined with a targeted agent. Patients with KRAS wild-type mCRC were stratified into 2 groups: those who underwent PTR and those who did not.

The primary outcome was OS, categorized as OS1, OS2, OS3, and OS during trifluridine/tipiracil or regorafenib therapy. OS1, OS2, and OS3 were defined as the duration from initiation of first-, second-, and third-line therapies, respectively, to death, censoring, or the end of 2020. OS for patients receiving trifluridine/tipiracil or regorafenib was defined as the time from initiation of either treatment to death, censoring, or the end of 2020. The secondary outcome was TTD. TTD1 was defined as the interval between initiation of first-line therapy and the start of second-line chemotherapy, and TTD2 as the interval between the start of second-line therapy and the start of third-line chemotherapy. TTD for trifluridine/tipiracil or regorafenib was defined as the interval from initiation of second-line therapy to the first administration of either agent.

Statistical analysis

Descriptive statistics were computed for demographic and tumor characteristics. A P-value of <0.05 was used to identify baseline imbalances between cetuximab plus FOLFIRI and bevacizumab plus FOLFIRI in patients with KRAS wild-type unresectable mCRC. OS and TTD were estimated and compared using the Kaplan-Meier method and log-rank test to assess unadjusted survival differences. Adjusted comparisons were performed using multivariate Cox proportional hazards models, with hazard ratios (HRs) and 95% confidence intervals (CIs) reported. Covariates were entered into a logistic regression model to generate propensity scores for treatment allocation. A Cox proportional hazards model adjusted for propensity scores and baseline characteristics was then applied to compare survival between treatment groups. Inverse probability weighting was performed using stabilized inverse probability of treatment weights (SIPTW) to control for confounding and achieve balance between cohorts. Covariates associated with outcomes, including comorbidities and tumor characteristics, were included in the propensity score model. SIPTW was employed to minimize sample loss and estimate average treatment effects.

All analyses were conducted using SAS ver. 9.4 (SAS Institute) and RStudio ver. 2024.09.1 + 341 (Posit Software). A 2-sided P-value of <0.05 was considered statistically significant for all hypotheses. Missing data were managed using multivariate imputation by chained equations with fully conditional specification. Incomplete variables were imputed using a Bayesian model [15]. The imputed variables included histological grade, tumor size, CEA status, and hospital level.

RESULTS

Cohort characteristics

A total of 13,473 patients with mCRC were identified during the study period (Fig. 1). Among these, 559 patients who received targeted therapy combined with chemotherapy as the first-line regimen, oxaliplatin-based chemotherapy as the second-line regimen, and trifluridine/tipiracil or regorafenib as third- or fourth-line therapy were included. Among the 278 patients who did not undergo PTR before the index date of first-line therapy, 164 received bevacizumab plus FOLFIRI as first-line therapy, followed by second-line oxaliplatin-based therapy with bevacizumab, third-line cetuximab-based therapy, and fourth-line trifluridine/tipiracil or regorafenib. The remaining 114 patients received cetuximab plus FOLFIRI as first-line therapy, followed by second-line oxaliplatin-based therapy with cetuximab and third-line trifluridine/tipiracil or regorafenib. Among the 281 patients who underwent PTR before the index date, 162 received bevacizumab plus FOLFIRI as first-line therapy, followed by second-line oxaliplatin-based therapy with bevacizumab, third-line cetuximab, and fourth-line trifluridine/tipiracil or regorafenib. The remaining 119 patients received cetuximab plus FOLFIRI as first-line therapy, followed by second-line oxaliplatin-based therapy with cetuximab and third-line trifluridine/tipiracil or regorafenib (Tables 1, 2).

OS and TTD

The OS and TTD outcomes are presented in Figs. 2–5. Results of the subgroup analyses based on tumor sidedness are shown in Figs. 4 and 5. Among patients who did not undergo PTR, 219 of 278 (78.7%) died during follow-up: 134 of 164 (81.7%) in the first-line bevacizumab plus FOLFIRI group (with subsequent oxaliplatin-based therapy ±bevacizumab, third-line cetuximab-based therapy, and fourth-line trifluridine/tipiracil or regorafenib) and 85 of 114 (74.6%) in the first-line cetuximab plus FOLFIRI group (with subsequent oxaliplatin-based therapy ±cetuximab and third-line trifluridine/tipiracil or regorafenib). The median OS1 did not significantly differ between the cetuximab plus FOLFIRI group (29.6 months; 95% CI, 26.2–31.8 months) and the bevacizumab plus FOLFIRI group (31.0 months; 95% CI, 29.0–33.3 months; crude HR, 0.98 [95% CI, 0.74–1.28]; adjusted HR, 0.99 [95% CI, 0.73–1.34]). The median OS2 was not significantly shorter in the cetuximab group (16.4 months; 95% CI, 13.7–18.2 months) than in the bevacizumab group (19.2 months; 95% CI, 17.2–20.5 months; crude HR, 1.18 [95% CI, 0.90–1.55]; adjusted HR, 1.24 [95% CI, 0.91–1.68]). However, the median OS3 was significantly shorter in the cetuximab group (6.2 months; 95% CI, 5.6–8.0 months) compared with the bevacizumab group (13.6 months; 95% CI, 12.5–15.0 months; crude HR, 1.99 [95% CI, 1.51–2.62]; adjusted HR, 2.43 [95% CI, 1.77–3.35]). Conversely, median OS following trifluridine/tipiracil or regorafenib was significantly longer in the cetuximab group (6.2 months; 95% CI, 5.6–8.0 months) compared with the bevacizumab group (4.9 months; 95% CI, 4.3–6.2 months; crude HR, 0.74 [95% CI, 0.56–0.97]; adjusted HR, 0.72 [95% CI, 0.54–0.98]) (Figs. 2, 4). The median TTD1 was significantly longer in the cetuximab group (11.8 months; 95% CI, 10.5–13.3 months) than in the bevacizumab group (9.5 months; 95% CI, 8.7–10.6 months; crude HR, 0.72 [95% CI, 0.57–0.92]; adjusted HR, 0.67 [95% CI, 0.52–0.88]). Similarly, the median TTD2 was longer in the cetuximab group (8.8 months; 95% CI, 7.7–9.6 months) than in the bevacizumab group (4.4 months; 95% CI, 3.9–5.6 months; crude HR, 0.52 [95% CI, 0.41–0.66]; adjusted HR, 0.45 [95% CI, 0.34–0.59]). In contrast, the median TTD for trifluridine/tipiracil or regorafenib was shorter in the cetuximab group (8.8 months; 95% CI, 7.7–9.6 months) compared with the bevacizumab group (12.5 months; 95% CI, 11.6–13.7 months; crude HR, 1.97 [95% CI, 1.55–2.51]; adjusted HR, 2.12 [95% CI, 1.62–2.77]).

Among patients who underwent PTR, 224 of 281 (79.7%) died during follow-up: 128 of 162 (79.0%) in the first-line bevacizumab plus FOLFIRI group (with subsequent oxaliplatin-based therapy ± bevacizumab, third-line cetuximab-based therapy, and fourth-line trifluridine/tipiracil or regorafenib) and 96 of 119 (80.7%) in the first-line cetuximab plus FOLFIRI group (with subsequent oxaliplatin-based therapy ± cetuximab and third-line trifluridine/tipiracil or regorafenib). The median OS1 was significantly different between the cetuximab plus FOLFIRI group (29.7 months; 95% CI, 26.9–34.0 months) and the bevacizumab plus FOLFIRI group (36.6 months; 95% CI, 34.1–38.3 months; crude HR, 1.33 [95% CI, 1.02–1.73]; adjusted HR, 1.41 [95% CI, 1.04–1.92]). The median OS2 was significantly shorter in the cetuximab group (17.0 months; 95% CI, 14.7–21.2 months) than in the bevacizumab group (23.5 months; 95% CI, 19.9–26.9 months; crude HR, 1.50 [95% CI, 1.15–1.96]; adjusted HR, 1.64 [95% CI, 1.20–2.24]). Likewise, the median OS3 was significantly shorter in the cetuximab group (6.1 months; 95% CI, 5.2–8.0 months) compared with the bevacizumab group (16.3 months; 95% CI, 13.9–18.4 months; crude HR, 2.89 [95% CI, 2.19–3.80]; adjusted HR, 3.25 [95% CI, 2.36–4.48]). The median OS following trifluridine/tipiracil or regorafenib was not significantly shorter in the cetuximab group (6.1 months; 95% CI, 5.2–8.0 months) than in the bevacizumab group (6.7 months; 95% CI, 5.4–7.9 months; crude HR, 1.05 [95% CI, 0.80–1.37]; adjusted HR, 0.99 [95% CI, 0.73–1.34]) (Figs. 2, 4). The median TTD1 did not differ significantly between the cetuximab group (10.6 months; 95% CI, 10.0–11.5 months) and the bevacizumab group (11.4 months; 95% CI, 10.4–12.5 months; crude HR, 0.93 [95% CI, 0.74–1.19]; adjusted HR, 0.87 [95% CI, 0.67–1.14]). The median TTD2 was significantly longer in the cetuximab group (9.0 months; 95% CI, 7.9–10.7 months) compared with the bevacizumab group (5.1 months; 95% CI, 4.3–6.0 months; crude HR, 0.53 [95% CI, 0.42–0.67]; adjusted HR, 0.50 [95% CI, 0.38–0.66]) Conversely, the median TTD for trifluridine/tipiracil or regorafenib was shorter in the cetuximab group (9.0 months; 95% CI, 7.9–10.7 months) compared with the bevacizumab group (14.4 months; 95% CI, 12.7–15.7 months; crude HR, 1.66 [95% CI, 1.30–2.11]; adjusted HR, 1.83 [95% CI, 1.39–2.41]).

DISCUSSION

This study analyzed patients with KRAS wild-type mCRC who received first-line treatment with either cetuximab plus FOLFIRI or bevacizumab plus FOLFIRI, followed by later-line therapy with trifluridine/tipiracil or regorafenib. Among patients in the bevacizumab plus FOLFIRI group, those without PTR had a median OS of 4.9 months after receiving trifluridine/tipiracil or regorafenib, compared with 6.7 months in those with PTR. The corresponding TTD1 values were 9.5 and 11.4 months, respectively. In contrast, patients treated with cetuximab plus FOLFIRI demonstrated a consistent median OS of 6.0 months following later-line therapy, regardless of PTR status. Similarly, TTD1 remained relatively stable at 11.8 months in patients without PTR and 10.6 months in those with PTR. These findings suggest that the clinical efficacy of cetuximab-based treatment is less influenced by PTR status, whereas the therapeutic effect of bevacizumab-based therapy appears to depend more strongly on whether PTR was performed [16]. Across all survival outcomes evaluated, OS following trifluridine/tipiracil or regorafenib and TTD1 were the least affected by treatment sequencing. These parameters may therefore represent more stable indicators of therapeutic efficacy. The observed differences likely reflect the distinct mechanisms of action of the 2 targeted agents. Bevacizumab primarily acts by modifying the tumor microenvironment through the inhibition of angiogenesis [12, 17]. In patients without PTR, a greater tumor vascular burden may diminish the therapeutic benefit of bevacizumab, whereas surgical resection could reduce this burden, thereby enhancing efficacy and contributing to improved survival outcomes [17, 18]. Among patients who did not undergo PTR, TTD1 analysis showed a shorter treatment duration before failure in the bevacizumab plus FOLFIRI group compared with the cetuximab plus FOLFIRI group, suggesting that bevacizumab plus FOLFIRI may be less effective in controlling early tumor progression, potentially leading to earlier transition to second-line therapy. In contrast, cetuximab directly inhibits tumor cell proliferation and induces apoptosis by blocking the EGFR–RAS–RAF–MEK–ERK signaling pathway [19, 20]. As its mechanism acts directly on tumor cells, its efficacy is less dependent on the tumor microenvironment or vascularization. This may explain the consistent benefit observed even in patients who did not undergo PTR. In the non-PTR subgroup, patients initially treated with cetuximab plus FOLFIRI exhibited superior OS following later-line trifluridine/tipiracil or regorafenib compared with those initially treated with bevacizumab plus FOLFIRI. These findings indicate that the survival benefit associated with cetuximab extends into later treatment lines and underscore the importance of selecting an optimal first-line strategy in this patient population [10].

In this study of KRAS wild-type mCRC patients who did not undergo PTR, both the cetuximab plus FOLFIRI and bevacizumab plus FOLFIRI groups eventually received trifluridine/tipiracil or regorafenib as later-line therapy. When analyzing OS1, no significant difference was observed between the 2 groups. However, OS2 and OS3 were significantly shorter in patients initially treated with cetuximab plus FOLFIRI compared with those receiving bevacizumab plus FOLFIRI, and this trend was even more pronounced in patients who underwent PTR. This difference can be largely attributed to the NHI reimbursement policy [8]. Under current regulations, patients who receive cetuximab plus FOLFIRI as first-line therapy must switch to an oxaliplatin-based regimen for second-line treatment. The use of bevacizumab in the second line is permitted only as a self-pay option, which is not captured in the NHIRD. After failure of second-line therapy, these patients proceed directly to trifluridine/tipiracil or regorafenib. In contrast, patients who begin with bevacizumab plus FOLFIRI can transition to an oxaliplatin-based regimen as second-line therapy and, following further progression, become eligible for third-line cetuximab-based treatment, with trifluridine/tipiracil or regorafenib administered thereafter. Consequently, the treatment sequence in the bevacizumab plus FOLFIRI group delays the initiation of trifluridine/tipiracil or regorafenib and extends survival during the third-line phase. However, this sequence may also introduce immortal time bias, as patients must survive long enough to receive second-line oxaliplatin-based therapy and subsequently progress before becoming eligible for cetuximab. This largely explains why OS2 and OS3 favored bevacizumab. Conversely, outcomes following the initiation of trifluridine/tipiracil or regorafenib (OS beyond the third line, as well as TTD1 and TTD2) tended to favor cetuximab, primarily because both groups had to survive sufficiently long to reach these later lines of therapy, thereby reducing immortal time bias. For later-line TTD, bevacizumab again showed an advantage, which may be explained by the fact that patients initially treated with cetuximab could receive oxaliplatin-based therapy combined with bevacizumab during the second line. Unlike the bevacizumab-first group, which required additional progression before starting trifluridine/tipiracil or regorafenib, these patients transitioned more directly to later-line therapy. Importantly, this same pattern was observed in both the non-PTR and PTR groups, supporting the consistency of these findings across subgroups. Additionally, OS1 for first-line bevacizumab plus FOLFIRI was superior to that of first-line cetuximab plus FOLFIRI, likely because surgical resection reduces tumor vascular burden, thereby enhancing the efficacy of bevacizumab and improving survival outcomes [17, 18].

This study had several strengths that enhance its scientific and clinical relevance. First, the use of real-world data from a large and heterogeneous patient population ensures strong external validity and supports applicability to routine clinical practice. Second, the large sample size provides statistical robustness and improves the reliability of observed differences between treatment strategies. Third, this study presents a comprehensive evaluation of treatment sequencing in KRAS wild-type mCRC, including patients both with and without PTR. By comparing first-line cetuximab plus FOLFIRI and bevacizumab plus FOLFIRI while incorporating later-line regimens such as trifluridine/tipiracil or regorafenib, this analysis provides a broad overview of treatment outcomes across multiple lines of therapy. Fourth, integration of clinical and molecular data, including KRAS status, enables a more nuanced interpretation of treatment effects. Fifth, extended follow-up and assessment of long-term endpoints, specifically OS and TTD, offer valuable insight into the durability of treatment benefits. Finally, the application of rigorous statistical methods, including multivariate adjustment and SIPTW, minimizes confounding and enhances precision. Collectively, these strengths make this study a valuable contribution to evidence-based oncology and provide guidance for optimizing treatment selection and sequencing in KRAS wild-type mCRC.

Despite these strengths, several limitations should be acknowledged. First, the study population was limited to Taiwanese patients. Although this homogeneous cohort supports internal validity, the generalizability of the findings to other populations may be restricted. In particular, NHI policies governing drug access and reimbursement may have influenced outcomes such as OS1, OS2, OS3, and TTD. Nevertheless, the results are broadly consistent with international studies comparing cetuximab plus FOLFIRI and bevacizumab plus FOLFIRI as first-line treatments, suggesting wider applicability. Second, the retrospective design introduces the potential for selection bias, as treatment allocation was not randomized and may have been influenced by clinical factors. To mitigate this, propensity score–based adjustments using SIPTW were implemented to balance baseline differences. Third, although real-world data provide practical insights, they may include incomplete or missing variables. Factors such as disease burden, number and location of metastases, resectability of primary and metastatic lesions prior to surgery, microsatellite instability, and NRAS/BRAF mutation status were not available and may have introduced unmeasured confounding. As a partial corrective, we adjusted for clinical stage (IVA, IVB, or IVC) as a surrogate for disease extent; however, this may not fully capture the impact of PTR on survival. Consequently, potential imbalances in disease severity could have influenced the true effect of PTR, as patients with potentially curable disease may not have been equally distributed across groups, despite SIPTW adjustment for other available variables. Fourth, although OS and TTD were included as long-term outcomes, other clinically meaningful endpoints such as progression-free survival and quality of life were not assessed, limiting the comprehensiveness of the evaluation. Fifth, under the previous NHI policy in Taiwan, marked differences in treatment sequencing between the cetuximab and bevacizumab groups may have influenced OS and TTD. Specifically, patients in the cetuximab group typically received only 2 lines of treatment before initiating regorafenib or trifluridine/tipiracil, whereas those in the bevacizumab group received 3. However, the database lacked a reliable variable to accurately assess the clinical impact of these sequencing differences.

Despite these limitations, this study provides meaningful real-world insights into the treatment of KRAS wild-type mCRC. Among patients who did not undergo PTR, first-line cetuximab plus FOLFIRI was associated with longer treatment duration and improved OS, including during later-line therapy with trifluridine/tipiracil or regorafenib, compared with bevacizumab plus FOLFIRI. Conversely, bevacizumab demonstrated more consistent benefits in patients who underwent PTR.

ARTICLE INFORMATION

Conflict of interest

No potential conflict of interest relevant to this article was reported.

Funding

This study was supported by Kaohsiung Veterans General Hospital (No. KSVGH-113-129 and KSVGH 115-073).

Acknowledgments

The authors wish to thank the Health Data Science Center, National Cheng Kung University Hospital for providing administrative and technical support and Ms. Hsiao-Ling Chiu of the Cancer Center, Kaohsiung Veterans General Hospital for her advisory comments on interpreting the results.

Author contributions

Conceptualization: YCS, YHC, CCW; Data curation: YHC , YCS, CCW; Formal analysis: YCS, CCS, PTL, CCW; Funding acquisition: CCW; Investigation: YCS , PTL, CCW; Methodology: YCS , CCS, CCW; Project administration: PTL; Visualization: CCW; Writing the original draft: YCS, CCW; Writing–review & editing: all authors. All the authors read and approved the final manuscript.

Fig. 1.

Flowchart of cohort selection. mCRC, metastatic colorectal cancer; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan; PTR, primary tumor resection.

Fig. 2.

Kaplan-Meier survival estimates of overall survival (OS) for (A–D) patients without PTR and (E–G) patients with PTR. (A, E) OS1 (duration from start of first-line therapy). (B, F) OS2 (duration from start of second-line therapy). (C, G) OS3 (duration from start of third-line therapy). (D, H) OS during trifluridine/tipiracil or regorafenib therapy. CI, confidence interval; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan.

Fig. 3.

Kaplan-Meier estimates of time to treatment discontinuation (TTD) for (A–C) patients without PTR and (D–F) patients with PTR. (A, D) TTD1 (the interval between initiation of first-line therapy and the start of second-line chemotherapy). (B, E) TTD2 (the interval between the start of second-line therapy and the start of third-line chemotherapy). (C, F) TTD for trifluridine/tipiracil or regorafenib (the interval from initiation of second-line therapy to the first administration of either agent). CI, confidence interval; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan.

Fig. 4.

Hazard ratios for overall survival (OS) in patients who did not undergo primary tumor resection (PTR) and those who underwent PTR prior to index date, including the left-sided tumor subgroup. Cet, cetuximab; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan; Bev, bevacizumab; CI, confidence interval; SIPTW, stabilized inverse probability of treatment weights.

Fig. 5.

Hazard ratios for time to treatment discontinuation (TTD) in patients who did not undergo primary tumor resection (PTR) and those who underwent PTR prior to index date, including the left-sided tumor subgroup. Cet, cetuximab; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan; Bev, bevacizumab; CI, confidence interval; SIPTW, stabilized inverse probability of treatment weights.

Table 1.

Baseline characteristics of the study cohort stratified by PTR status

Characteristic	Without PTR			With PTR
Characteristic	Cetuximab+FOLFIRI (n=114)	Bevacizumab+FOLFIRI (n=164)	P-value	Cetuximab+FOLFIRI (n=119)	Bevacizumab+FOLFIRI (n=162)	P-value
Death	85 (74.6)	134 (81.7)	0.2	96 (80.7)	128 (79.0)	0.7
Male sex	73 (64.0)	108 (65.9)	0.8	69 (58.0)	94 (58.0)	>0.99
Age (yr)	59±11	57±11	0.4	57±12	57±12	0.7
Year of targeted therapy			0.2			0.7
2013	6 (5.3)	19 (11.6)		9 (7.6)	23 (14.2)
2014	11 (9.6)	15 (9.1)		16 (13.4)	25 (15.4)
2015	23 (20.2)	26 (15.9)		23 (19.3)	27 (16.7)
2016	18 (15.8)	33 (20.1)		27 (22.7)	37 (22.8)
2017	18 (15.8)	33 (20.1)		20 (16.8)	23 (14.2)
2018	26 (22.8)	24 (14.6)		17 (14.3)	18 (11.1)
≥2019	12 (10.5)	14 (8.5)		7 (5.9)	9 (5.6)
Radiotherapy	18 (15.8)	36 (22.0)	0.2	8 (6.7)	10 (6.2)	0.9
Charlson Comorbidity Index	8±2	9±2	0.1	8±2	8±2	0.8
Stage			<0.05			0.9
IVA	70 (61.4)	83 (50.6)		66 (55.5)	87 (53.7)
IVB	35 (30.7)	77 (47.0)		46 (38.7)	67 (41.4)
IVC	9 (7.9)	4 (2.4)		7 (5.9)	8 (4.9)
Histologic			0.5			0.5
Adenocarcinoma	114 (100)	164 (100)^a		111 (93.3)	154 (95.1)
Signet-ring cell carcinoma or mucinous	0 (0)			8 (6.7)	8 (4.9)
Tumor differentiation grade			0.6			0.8
Well differentiated	4 (3.5)	7 (4.3)		101 (84.9)	136 (84.0)
Moderately differentiated	100 (87.7)	136 (82.9)
Poorly differentiated, undifferentiated, or anaplastic	10 (8.8)	21 (12.8)		18 (15.1)	26 (16.0)
Tumor sidedness			0.1			0.4
Right	11 (9.6)	26 (15.9)		22 (18.5)	36 (22.2)
Left	103 (90.4)	138 (84.1)		97 (81.5)	126 (77.8)
Obstruction	53 (46.5)	64 (39.0)	0.2	67 (56.3)	91 (56.2)	>0.99
Positive CEA	105 (92.1)	143 (87.2)	0.2	98 (82.4)	140 (86.4)	0.3
Hospital level (medical center)	78 (68.4)	101 (61.6)	0.2	83 (69.7)	104 (64.2)	0.3
Metastasectomy			<0.05			0.5
Liver resection	39 (34.2)	24 (14.6)		36 (30.3)	37 (22.8)
Lung resection	4 (3.5)	5 (3.0)		10 (8.4)	11 (6.8)

Values are presented as number (%) or mean±standard deviation. Missing values were imputed using multivariate imputation by chained equations.

The proportions of missing values in the non-PTR population were as follows: tumor differentiation grade, 33.1%; tumor size, 29.1%; tumor sidedness, 0%; bowel obstruction, 0%; bowel perforation, 0%; CEA status, 2.5%; and hospital level, 8.3%. The proportions of missing values in the PTR population were: tumor differentiation grade, 2.5%; tumor size, 1.1%; tumor sidedness, 0%; bowel obstruction, 0%; bowel perforation, 0%; CEA status, 5.3%; and hospital level, 5.0%.

PTR, primary tumor resection, FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan; CEA, carcinoembryonic antigen.

^aUnder Taiwan’s National Health Insurance Research Database policy, case numbers for any item less than 3 (but not 0) must be combined with another category to ensure data release. Thus, the accurate numbers for this category are 163 (99.4%) and 1 (0.6%) or 162 (98.8%) and 2 (1.2%). The P-value is based on the original data (i.e., 163 or 162 vs. 1 or 2).

Table 2.

Time of metastasectomy and duration or cycles of anticancer medicines in the study cohort, stratified by PTR status

Characteristic	Without PTR			With PTR
Characteristic	Cetuximab+FOLFIRI (n=114)	Bevacizumab+FOLFIRI (n=164)	P-value	Cetuximab+FOLFIRI (n=119)	Bevacizumab+FOLFIRI (n=162)	P-value
Time of metastasectomy			<0.05			<0.05
Between first- and second-line therapy	37 (32.5)	19 (11.6)		35 (29.4)	28 (17.3)
Between second- and third-line therapy	3 (2.6)	6 (3.7)		5 (4.2)	6 (3.7)
After third-line therapy	0 (0)	3 (1.8)		0 (0)	8 (4.9)
Total second oxaliplatin duration (mo)	5.1 (2.8–6.8)	2.8 (2.0–5.1)	<0.05	5.1 (2.9–6.9)	3.3 (2.1–5.5)	<0.05
Total No. of second oxaliplatin cycles	8 (5–12)	6 (4–8)	<0.05	8 (6-11)	6 (4-9)	<0.05
Second oxaliplatin + index targeted therapy	35 (30.7)	53 (32.3)	0.8	42 (35.3)	29 (17.9)	<0.05
Duration of second original targeted therapy (mo)	2.1 (1.1–4.4)	2.1 (1.1–3.3)	0.8	3.6 (1.9–5.6)	1.6 (0.4–3.5)	<0.05
Total No. of cycles of second original targeted therapy	4 (3–7)	4 (3–6)	0.8	6.5 (4-9)	3 (2–6)	<0.05
Second bevacizumab use in first cetuximab group	20 (17.5)	-		14 (11.8)	-
Second bevacizumab duration (mo)	4.3 (2.8–7.2)	-		7.5 (3.9–9.7)	-
Total No. of second bevacizumab cycles	8.5 (6–13)	-		12 (7–16)	-
Third cetuximab duration in first bevacizumab group (mo)	-	5.3 (2.8–7.2)		-	5.4 (3.6–7.0)
Total No. of third cetuximab cycles in first bevacizumab group	-	10 (6–13)		-	10 (6–12)
Total trifluridine/tipiracil or regorafenib duration (mo)	1.9 (0.7–4.4)	1.8 (0.7–4.3)	0.6	2.1 (1.2–5.3)	2.4 (0.9–5.6)	0.8

Values are presented as number (%) or median (interquartile range).

PTR, primary tumor resection, FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan.

REFERENCES

1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394–424. Article PubMed PDF
2. Vatandoust S, Price TJ, Karapetis CS. Colorectal cancer: metastases to a single organ. World J Gastroenterol 2015;21:11767–76. Article PubMed PMC
3. Nordlinger B, Van Cutsem E, Gruenberger T, Glimelius B, Poston G, Rougier P, et al. Combination of surgery and chemotherapy and the role of targeted agents in the treatment of patients with colorectal liver metastases: recommendations from an expert panel. Ann Oncol 2009;20:985–92. Article PubMed
4. Folprecht G, Grothey A, Alberts S, Raab HR, Köhne CH. Neoadjuvant treatment of unresectable colorectal liver metastases: correlation between tumour response and resection rates. Ann Oncol 2005;16:1311–9. Article PubMed
5. National Comprehensive Cancer Network (NCCN). About [Internet]. NCCN; [2025 Feb 2]. Available from: https://www.nccn.org/home/about
6. Botrel TE, Clark LG, Paladini L, Clark OA. Efficacy and safety of bevacizumab plus chemotherapy compared to chemotherapy alone in previously untreated advanced or metastatic colorectal cancer: a systematic review and meta-analysis. BMC Cancer 2016;16:677.Article PubMed PMC PDF
7. Wu CC, Wang JH, Lin PC, Liang CA, Huang CY, Lien HC, et al. Tumor sidedness and efficacy of first-line therapy in patients with RAS/BRAF wild-type metastatic colorectal cancer: a network meta-analysis. Crit Rev Oncol Hematol 2020;145:102823.Article PubMed
8. National Health Insurance Administration. [The drugs payment guideline for anti-neoplastic agents] [Internet]. Taiwan Ministry of Health and Welfare; 2024 [cited 2025 Jul 1]. Taiwanese. Available from: https://www.nhi.gov.tw/ch/dl-55685-99c675b771ab4b2789c891bc8db447ce-1.pdf
9. Venook AP, Niedzwiecki D, Lenz HJ, Innocenti F, Fruth B, Meyerhardt JA, et al. Effect of first-line chemotherapy combined with cetuximab or bevacizumab on overall survival in patients with KRAS wild-type advanced or metastatic colorectal cancer: a randomized clinical trial. JAMA 2017;317:2392–401. Article PubMed PMC
10. Heinemann V, von Weikersthal LF, Decker T, Kiani A, Vehling-Kaiser U, Al-Batran SE, et al. FOLFIRI plus cetuximab versus FOLFIRI plus bevacizumab as first-line treatment for patients with metastatic colorectal cancer (FIRE-3): a randomised, open-label, phase 3 trial. Lancet Oncol 2014;15:1065–75. Article PubMed
11. Wu CC, Su CC, Lee PT, Chen YH, Hsu CW, Su YC. Comparative effectiveness of primary tumor resection versus chemotherapy in patients with asymptomatic unresectable metastatic colorectal cancer: a retrospective cohort study using the target trial emulation framework. Am J Cancer Res 2024;14:2172–86. Article PubMed PMC
12. Ferrara N, Kerbel RS. Angiogenesis as a therapeutic target. Nature 2005;438:967–74. Article PubMed PDF
13. Charlson ME, Charlson RE, Peterson JC, Marinopoulos SS, Briggs WM, Hollenberg JP. The Charlson comorbidity index is adapted to predict costs of chronic disease in primary care patients. J Clin Epidemiol 2008;61:1234–40. Article PubMed
14. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992;45:613–9. Article PubMed
15. van Buuren S, Groothuis-Oudshoorn K. mice: multivariate imputation by chained equations in R. J Stat Soft 2011;45:1–67.Article
16. Su YC, Wu CC, Su CC, Hsieh MC, Cheng CL, Kao Yang YH. Comparative effectiveness of bevacizumab versus cetuximab in metastatic colorectal cancer patients without primary tumor resection. Cancers (Basel) 2022;14:2118.Article PubMed PMC
17. Jain RK. Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science 2005;307:58–62. Article PubMed
18. Van Cutsem E, Rivera F, Berry S, Kretzschmar A, Michael M, DiBartolomeo M, et al. Safety and efficacy of first-line bevacizumab with FOLFOX, XELOX, FOLFIRI and fluoropyrimidines in metastatic colorectal cancer: the BEAT study. Ann Oncol 2009;20:1842–7. Article PubMed
19. Karapetis CS, Khambata-Ford S, Jonker DJ, O'Callaghan CJ, Tu D, Tebbutt NC, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med 2008;359:1757–65. Article PubMed
20. Ciardiello F, Tortora G. EGFR antagonists in cancer treatment. N Engl J Med 2008;358:1160–74. Article PubMed

Figure & Data

References

Citations

Citations to this article as recorded by

Cite this Article

Cite this Article: export Copy Download Format; Close

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

RIS — For EndNote, ProCite, RefWorks, and most other reference management software
BibTeX — For JabRef, BibDesk, and other BibTeX-specific software

Include:

Citation for the content below
Citation and abstract for the content below

Effectiveness of primary tumor resection for survival after first-line cetuximab or bevacizumab in KRAS wild-type metastatic colorectal cancer treated with subsequent trifluridine/tipiracil or regorafenib

Ann Coloproctol. 2026;42(1):127-140. Published online February 27, 2026

DOI: https://doi.org/10.3393/ac.2025.00759.0108

XML Download

Figure

Related articles

Effectiveness of oxaliplatin-based second-line therapy following cetuximab+FOLFIRI or bevacizumab+FOLFIRI in KRAS wild-type metastatic colorectal cancer without primary tumor resection

Fig. 1. Flowchart of cohort selection. mCRC, metastatic colorectal cancer; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan; PTR, primary tumor resection.

Fig. 2. Kaplan-Meier survival estimates of overall survival (OS) for (A–D) patients without PTR and (E–G) patients with PTR. (A, E) OS1 (duration from start of first-line therapy). (B, F) OS2 (duration from start of second-line therapy). (C, G) OS3 (duration from start of third-line therapy). (D, H) OS during trifluridine/tipiracil or regorafenib therapy. CI, confidence interval; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan.

Fig. 3. Kaplan-Meier estimates of time to treatment discontinuation (TTD) for (A–C) patients without PTR and (D–F) patients with PTR. (A, D) TTD1 (the interval between initiation of first-line therapy and the start of second-line chemotherapy). (B, E) TTD2 (the interval between the start of second-line therapy and the start of third-line chemotherapy). (C, F) TTD for trifluridine/tipiracil or regorafenib (the interval from initiation of second-line therapy to the first administration of either agent). CI, confidence interval; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan.

Fig. 4. Hazard ratios for overall survival (OS) in patients who did not undergo primary tumor resection (PTR) and those who underwent PTR prior to index date, including the left-sided tumor subgroup. Cet, cetuximab; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan; Bev, bevacizumab; CI, confidence interval; SIPTW, stabilized inverse probability of treatment weights.

Fig. 5. Hazard ratios for time to treatment discontinuation (TTD) in patients who did not undergo primary tumor resection (PTR) and those who underwent PTR prior to index date, including the left-sided tumor subgroup. Cet, cetuximab; FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan; Bev, bevacizumab; CI, confidence interval; SIPTW, stabilized inverse probability of treatment weights.

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Characteristic	Without PTR			With PTR
Characteristic	Cetuximab+FOLFIRI (n=114)	Bevacizumab+FOLFIRI (n=164)	P-value	Cetuximab+FOLFIRI (n=119)	Bevacizumab+FOLFIRI (n=162)	P-value
Death	85 (74.6)	134 (81.7)	0.2	96 (80.7)	128 (79.0)	0.7
Male sex	73 (64.0)	108 (65.9)	0.8	69 (58.0)	94 (58.0)	>0.99
Age (yr)	59±11	57±11	0.4	57±12	57±12	0.7
Year of targeted therapy			0.2			0.7
2013	6 (5.3)	19 (11.6)		9 (7.6)	23 (14.2)
2014	11 (9.6)	15 (9.1)		16 (13.4)	25 (15.4)
2015	23 (20.2)	26 (15.9)		23 (19.3)	27 (16.7)
2016	18 (15.8)	33 (20.1)		27 (22.7)	37 (22.8)
2017	18 (15.8)	33 (20.1)		20 (16.8)	23 (14.2)
2018	26 (22.8)	24 (14.6)		17 (14.3)	18 (11.1)
≥2019	12 (10.5)	14 (8.5)		7 (5.9)	9 (5.6)
Radiotherapy	18 (15.8)	36 (22.0)	0.2	8 (6.7)	10 (6.2)	0.9
Charlson Comorbidity Index	8±2	9±2	0.1	8±2	8±2	0.8
Stage			<0.05			0.9
IVA	70 (61.4)	83 (50.6)		66 (55.5)	87 (53.7)
IVB	35 (30.7)	77 (47.0)		46 (38.7)	67 (41.4)
IVC	9 (7.9)	4 (2.4)		7 (5.9)	8 (4.9)
Histologic			0.5			0.5
Adenocarcinoma	114 (100)	164 (100)^a		111 (93.3)	154 (95.1)
Signet-ring cell carcinoma or mucinous	0 (0)			8 (6.7)	8 (4.9)
Tumor differentiation grade			0.6			0.8
Well differentiated	4 (3.5)	7 (4.3)		101 (84.9)	136 (84.0)
Moderately differentiated	100 (87.7)	136 (82.9)
Poorly differentiated, undifferentiated, or anaplastic	10 (8.8)	21 (12.8)		18 (15.1)	26 (16.0)
Tumor sidedness			0.1			0.4
Right	11 (9.6)	26 (15.9)		22 (18.5)	36 (22.2)
Left	103 (90.4)	138 (84.1)		97 (81.5)	126 (77.8)
Obstruction	53 (46.5)	64 (39.0)	0.2	67 (56.3)	91 (56.2)	>0.99
Positive CEA	105 (92.1)	143 (87.2)	0.2	98 (82.4)	140 (86.4)	0.3
Hospital level (medical center)	78 (68.4)	101 (61.6)	0.2	83 (69.7)	104 (64.2)	0.3
Metastasectomy			<0.05			0.5
Liver resection	39 (34.2)	24 (14.6)		36 (30.3)	37 (22.8)
Lung resection	4 (3.5)	5 (3.0)		10 (8.4)	11 (6.8)

Characteristic	Without PTR			With PTR
Characteristic	Cetuximab+FOLFIRI (n=114)	Bevacizumab+FOLFIRI (n=164)	P-value	Cetuximab+FOLFIRI (n=119)	Bevacizumab+FOLFIRI (n=162)	P-value
Time of metastasectomy			<0.05			<0.05
Between first- and second-line therapy	37 (32.5)	19 (11.6)		35 (29.4)	28 (17.3)
Between second- and third-line therapy	3 (2.6)	6 (3.7)		5 (4.2)	6 (3.7)
After third-line therapy	0 (0)	3 (1.8)		0 (0)	8 (4.9)
Total second oxaliplatin duration (mo)	5.1 (2.8–6.8)	2.8 (2.0–5.1)	<0.05	5.1 (2.9–6.9)	3.3 (2.1–5.5)	<0.05
Total No. of second oxaliplatin cycles	8 (5–12)	6 (4–8)	<0.05	8 (6-11)	6 (4-9)	<0.05
Second oxaliplatin + index targeted therapy	35 (30.7)	53 (32.3)	0.8	42 (35.3)	29 (17.9)	<0.05
Duration of second original targeted therapy (mo)	2.1 (1.1–4.4)	2.1 (1.1–3.3)	0.8	3.6 (1.9–5.6)	1.6 (0.4–3.5)	<0.05
Total No. of cycles of second original targeted therapy	4 (3–7)	4 (3–6)	0.8	6.5 (4-9)	3 (2–6)	<0.05
Second bevacizumab use in first cetuximab group	20 (17.5)	-		14 (11.8)	-
Second bevacizumab duration (mo)	4.3 (2.8–7.2)	-		7.5 (3.9–9.7)	-
Total No. of second bevacizumab cycles	8.5 (6–13)	-		12 (7–16)	-
Third cetuximab duration in first bevacizumab group (mo)	-	5.3 (2.8–7.2)		-	5.4 (3.6–7.0)
Total No. of third cetuximab cycles in first bevacizumab group	-	10 (6–13)		-	10 (6–12)
Total trifluridine/tipiracil or regorafenib duration (mo)	1.9 (0.7–4.4)	1.8 (0.7–4.3)	0.6	2.1 (1.2–5.3)	2.4 (0.9–5.6)	0.8

Table 1. Baseline characteristics of the study cohort stratified by PTR status

Values are presented as number (%) or mean±standard deviation. Missing values were imputed using multivariate imputation by chained equations.

PTR, primary tumor resection, FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan; CEA, carcinoembryonic antigen.

Table 2. Time of metastasectomy and duration or cycles of anticancer medicines in the study cohort, stratified by PTR status

Values are presented as number (%) or median (interquartile range).

PTR, primary tumor resection, FOLFIRI, folinic acid, 5-fluorouracil, and irinotecan.