Abstract
Background:
Raltitrexed plus S-1 (RS) and regorafenib both showed considerable efficacy for metastatic colorectal cancer (mCRC) patients. This study aims to compare the effectiveness and safety of two different regimens in patients with refractory mCRC.
Methods:
This retrospective cohort study included mCRC patients who were treated with RS or regorafenib from February 2017 to June 2021. A propensity score matching (PSM) analysis was conducted to balance the baseline characteristics of all patients. Progression-free survival (PFS), overall survival (OS), tumor response, and safety of two regimens were evaluated.
Results:
A total of 187 patients were included in our study, with 107 patients in the RS group and 80 patients in the regorafenib group. After PSM, 78 pairs were recognized. Patients treated with RS had a semblable PFS compared to those treated with regorafenib before PSM (4.8 months vs 5.5 months, p = 0.400) and after PSM (4.7 months vs 5.4 months, p = 0.430). Patients in the RS group were associated with a longer OS than those in the regorafenib group (13.4 months vs 10.1 months, p = 0.010). A similar trend of OS was also obtained in the matched cohort (13.3 months vs 10.0 months, p = 0.024). Both objective response rate (12.8% vs 5.1%, p = 0.093) and disease control rate (53.8% vs 46.2%, p = 0.337) in the RS cohort were higher than those in the regorafenib group, without significant differences. Adverse events (AEs) of each group were well tolerated.
Conclusion:
Patients treated with RS demonstrated a longer OS than those treated with regorafenib and had manageable AEs, which could be recognized as a primary choice for refractory mCRC.
Plain Language Summary
Both raltitrexed plus S-1 (RS) and regorafenib showed considerable efficacy for metastatic colorectal cancer (mCRC) patients. No study has compared the two regimens yet. Therefore, we compare the efficacy and safety between RS and regorafenib to provide an optimal treatment option. We retrospectively included patients with mCRC who failed at least two standard treatments. All enrolled patients received RS or regorafenib treatments. We conducted a propensity score matching to eliminate differences in the enrolled patients. After the analysis, we found no significant differences in progression-free survival in patients between the two groups. However, patients treated with RS had a longer OS than those treated with regorafenib, whether before matching (13.4 months vs 10.1 months, p = 0.010) or after matching (13.3 months vs 10.0 months, p = 0.024). In addition, the adverse effects caused by cancer-related therapy were tolerable for the patient. Certainly, this is a non-randomized retrospective study with a small sample size, so we conducted a propensity score matching to minimize potential bias. Importantly, this is the first research comparing the two treatments, and we believe that the results of this article could present a primary choice for clinical doctors dealing with patients with standard treatments that failed mCRC.
Introduction
Colorectal cancer (CRC) is the third-most commonly diagnosed cancer worldwide and the fourth leading cause of cancer-related death globally. 1 As early diagnosis of colorectal cancer is complex and difficult, 25% of patients present with metastatic colorectal cancer (mCRC) when diagnosed, 2 with a poor 5-year survival of less than 15%. Standard treatments consisting of cytotoxic drugs (fluorouracil, irinotecan, and oxaliplatin) combined with targeted drugs (bevacizumab or cetuximab) can achieve a particular effect in the initial stage. 3 However, as the disease progresses, the effectiveness of existing treatments is limited. Therefore, effective drugs are not available at present, and further therapeutic regimens are required to improve the survival of mCRC.
Regorafenib was demonstrated to significantly improve the progression-free survival (PFS) and overall survival (OS) of patients with refractory mCRC in the CONCUR 4 and CORRECT 5 phase-III trials. And it was recommended as a standard posterior line treatment for patients with mCRC who have previously received fluorouracil, oxaliplatin, and irinotecan-based chemotherapy. 6 S-1, a fluorouracil derivative, could effectively hinder the growth and proliferation of new vascular endothelial cells in tumor tissue, thus inhibiting tumor. S-1 monotherapy or combination therapy has demonstrated a considerable benefit for CRC patients.7,8 Raltitrexed, as a specific thymidylate synthase inhibitor, could inhibit the activity of the target enzyme of fluorouracil, thus demonstrating a synergistic effect with fluorouracil. 9 Our previous studies have demonstrated the effectiveness and safety of the combination of raltitrexed plus S-1 (RS) with or without bevacizumab in patients with refractory mCRC.10,11 However, no studies have compared the RS and regorafenib. Herein, we aimed to make a comparison between RS and regorafenib in patients with mCRC who failed the previous standard treatment to explore the efficacy and safety of these two regimens.
Materials and methods
Study design and patients
This retrospective cohort study consecutively collected patients diagnosed with colorectal cancer who received RS or regorafenib after the failure of at least two standard treatments from February 2017 to June 2021 in our hospital. The reporting of this study conformed to the STROBE statement. 12 Patients would be included if they met the following criteria: (1) pathologically confirmed unresectable or metastatic colorectal adenocarcinoma with at least one measurable disease according to the Response Evaluation Criteria in Solid Tumors (v1.1); 13 (2) age was or older than 18 years; (3) previous standard therapy failed. Patients were excluded if they lack complete clinical materials; were combined with other targeted therapy; had undergone local treatment on measurable diseases before the initial evaluation; or were complicated with other serious physical illnesses. Notably, the crossover administration was also excluded in this study. The study was approved by the Institutional Review Board of West China Hospital, Sichuan University, Chengdu, China [Approval number: 2021 Review (NO.1416)] on 10 November 2021. All the patients were collected from the database in West China Hospital, Sichuan University, and patients’ details were hidden. Exemption of informed consent was granted by Institutional Review Board of West China Hospital, Sichuan University.
Endpoint’s definition
OS was defined as the time between the onset of treatment and death by any cause. PFS was measured from the time starting the treatment to the date of disease progression or death of any cause. Objective response rate (ORR) was defined as a complete response (CR) or partial response (PR), while disease control rate (DCR) was defined as a CR, PR, or stable disease. The primary survival endpoint was OS, and the secondary endpoints were PFS, ORR, DCR, and safety. We performed the follow-up every 2 to 3 months. Most were followed up by telephone, and a few were followed up with the assistance of local departments of the census. Adverse events (AEs) were assessed in accordance with the Common Terminology Criteria for Adverse Events, version 4.03.
Propensity score matching
Propensity score matching (PSM) for the two groups was performed using a 1:1 ratio by R statistical programming language software version 4.1.1. To make the differences in patients’ baseline characteristics minimal, we included the variables with potential effects on prognosis, including primary tumor location, RAS status, BRAF status, surgery, first-line therapy with target drugs, second-line therapy with target drugs, and therapy line in this propensity score model. Chi-square tests or Fisher exact tests were performed to assess the statistical significance of differences in covariates between treatment groups before and after matching. PSM of the two cohorts was then conducted with a caliper of 0.24 on the propensity scale.
Statistical analysis
Statistical analysis was performed using SPSS version 26.0 (IBM Corporation, Armonk, NY) and R version 4.1.1. The Kaplan–Meier curves were used in performing the survival, including PFS and OS. Hazard ratios (HRs) with 95% confidence intervals (CIs) were used to compare the risk of the two regimens. P-value less than 0.05 in a two-tailed test was used to define statistical significance.
Results
Patients and baseline characteristics
A total of 187 patients meeting inclusion and exclusion criteria in West China hospital from February 2017 to June 2021 were collected, with 107 in the RS cohort and 80 in the regorafenib cohort. The flow chart for patient selection is shown in Figure 1. The baseline characteristics of patients treated with RS were generally similar to those treated with regorafenib (Table 1). Concretely, tumors in both groups were predominantly located on the left side, with mostly increased levels of CEA and CA199 before the treatment. Liver metastases were the most common metastatic lesions, occurring in about 70% of patients, followed by lung, lymph nodes, and bone metastases. A majority of patients were complicated by multiple metastases. Besides, RAS mutation, including KRAS and NRAS, appeared in 36.2–40.2% of the included population, while BRAF mutation occurred in 5.6% of patients in the RS group and 3.8% of patients in the regorafenib group, respectively. Poor pathological differentiation accounted for a small ratio in the RS group (23.4%) and the regorafenib group (18.8%). The proportion of patients in the T4 stage (55%) in the regorafenib group was higher than that in the RS group (41.1%). Nonetheless, the difference was not significant. As for the treatment, the majority of patients had previously undergone surgery. They received similar targeted therapies including anti-epidermal growth factor receptor and anti-vascular endothelial growth factor drugs (Supplementary Table 1). Importantly, RS or regorafenib was selected as the third-line therapy for most patients with mCRC, and only a small part of patients received it as their fourth or higher-line treatment. Most of them did not receive any subsequent therapy (Supplementary Table 2). After balancing the patients’ baseline characteristics, 78 cases in each cohort were identified in the PSM (Figure 1). In the matched cohort, no significant differences were discovered between the RS and the regorafenib cohort in all variables (Table 1).
Flow diagram of patient selection.
Baseline characteristics of all eligible patients.
RS, raltitrexed plus S-1.
Treatment regimen which includes oxaliplatin, such as FOLFOX, XELOX, and CAPOX.
Treatment regimen which includes irinotecan, such as XELIRI and FOLFIRI.
Efficacy
The median follow-up period was 17.3 (14.2–20.6) months in the RS group and 11.5 (9.3–11.7) months in the regorafenib group. Before PSM, patients in the RS group had a comparable median PFS to those in the regorafenib group (4.8 months vs 5.5 months, p = 0.40). Nevertheless, patients in the RS group had a superior OS than those in the regorafenib group (13.4 months vs 10.1 months, p = 0.01). Similar prognostic trends for PFS and OS were observed in the matched cohort: patients treated with RS had a semblable PFS to those treated with regorafenib (4.7 months vs 5.4 months, p = 0.43). Patients in the RS cohort were associated with a longer median OS than those in the regorafenib cohort (13.3 months vs 10.0 months, p = 0.02; Figure 2).
Kaplan–Meier survival curves for patients before matching and after matching: (a) progression-free survival of patients before matching, (b) overall survival of patients before matching, (c) progression-free survival of patients in the matched cohort, and (d) overall survival of patients in the matched cohort.
The tumor response of all the patients before and after matching was summarized in Table 2. No patient had a CR in either group. Before PSM, 12.1% (13/107) patients achieved PR in the RS group, and 5.0% (4/80) patients obtained PR in the regorafenib group. No significant difference was found in ORR between the two groups (p = 0.092). In addition, patients in the RS group displayed a similar DCR compared with those in the regorafenib group (54.2% vs 45.0%, p = 0.21). After PSM, 12.8% (10/78) and 5.1% (4/78) patients showed PR in the RS and regorafenib groups, respectively. Both the ORR (12.8% vs 5.1%, p = 0.093) and the DCR (53.8% vs 46.2%, p = 0.34) in the RS group were quantitatively higher than those in the regorafenib group. However, the difference was not significant either in the unmatching cohort or the matching cohort.
The overall response of all the patients before and after matching.
DCR, disease control rate; ORR, objective response rate; PD, progressive disease; PR, partial response; RS, raltitrexed plus S-1; SD, stable disease.
Subgroup analysis
Furthermore, to better define prognostic differences between RS and regorafenib groups, we performed a subgroup analysis for PFS and OS in the matched cohort (Figure 3). No statistical differences were obtained between these two groups in PFS for nearly all parameters analyzed. As for OS, patients in the RS group had a superior prognosis with following beneficial factor, including younger age (<60 years old), higher CEA levels at baseline (>10 times of normal level), absence of lung metastasis, mutated RAS gene, wild BRAF gene, low pathological differentiation, lymph node invasion, and combination with target therapy in the first- or second-line treatment.
Subgroup analyses for progression-free survival and overall survival: (a) progression-free survival and (b) overall survival.
Adverse events
AEs in all patients were summarized in Table 3. In this study, nearly all patients underwent AEs caused by cancer-related therapy. Generally, it was universal for AEs in the RS cohort than that in the regorafenib group. The hematological AEs were quite common in the RS group, including anemia (45%), leucopenia (40%), AST/ALT increased (40%), neutropenia (38%), as well as thrombocytopenia (24%). Some unexceptional AEs during the tumor therapy, like fatigue (59%), anorexia (35%), vomit (30%), and diarrhea (22%), also occurred. Notably, hyperpigmentation happened to 46% population in the RS cohort while hand–foot skin reaction only happened to 19%. On the contrary, no hyperpigmentation was observed in the regorafenib group, while hand–foot skin reaction (69%) was the most common AE. The secondary common AEs were hyperbilirubinemia (37%), followed by AST/ALT increased (27%), vomit (26%), hypertension (24%), anorexia (20%), and so on. It was worthy to note that the incidence of grade ⩾ 3 hand–foot skin reaction and hypertension in the regorafenib group could reach 15% and 10%, respectively.
Adverse events of all the included patients.
RS, raltitrexed plus S-1.
Discussion
Our previous phase-II trial10,11 suggested that RS with or without bevacizumab was a promising third- or later-line treatment option in patients with refractory mCRC. Regorafenib has been approved worldwide as a recommended drug for the third-line therapy of mCRC.6,14 Criteria for the appropriate selection of RS or regorafenib have not yet been established. Thus, we explored retrospectively the efficacy and toxicity of RS versus regorafenib in mCRC patients who were intolerant to standard therapies. This study, for the first time, concluded that in patients with standard treatments who failed mCRC, the RS has an OS advantage over regorafenib despite its obvious but controllable AEs. Although no statistical difference was obtained, ORR in the RS group was twice as high as that in the regorafenib group. In terms of PFS and DCR, these two regimens are comparable.
In this study, the PFS and OS of regorafenib were 5.4 and 10.0 months, respectively, which are longer than those in the CONCUR 4 (3.2 and 8.8 months) and CORRECT 5 (1.9 and 6.4 months) trials. Nonetheless, the DCR in this study (46.2%) was comparable to that in the CONCUR (51.0%) and CORRECT (41.0%) trials. A relatively higher proportion of the third-line treatments and a longer efficacy evaluation and patients’ follow-up period in this study may contribute to the above outcomes. Our study indicated a similar DCR (53.8%) in the RS group to that in the previous study (54.3%). 10 But the PFS (4.7 months) and OS (13.3 months) of RS in this study are a little longer than those in that study with 3.6 and 12.4 months, respectively. The likely reason is that the proportion of patients treated with targeted therapies, either cetuximab or bevacizumab, in this study (about 60%) was higher than that in that study (49.9%). Although, previous studies suggested that the effect of regorafenib might be affected by previous treatments. 4 Some studies proved that trifluridine/tipiracil (TAS-102) was an effective treatment option for refractory colorectal cancer,15,16 while combination with bevacizumab would make survival benefit more prominent.17–19 The PFS time was similar to that in our study. However, we did not explore the effects of combination therapy in this research, and more studies are needed in the future. The therapy lines in the two groups are well balanced in the current research (Table 1). Crossover administration is excluded in this study, and the proportion of posterior line treatment is also semblable between the two groups, making the outcomes more credible.
Up to now, the subgroups of patients who can benefit from RS or regorafenib remain unclear, and the prognostic factor of these two regimens is not very pronounced. Previous researches indicated that the density reduction in lung metastases, 20 a good Eastern Cooperative Oncology Group performance status, a long history of metastatic disease (⩾18 months), and the presence of a limited lung metastatic disease are associated with better prognosis. 21 Some studies also pointed out that Cancer-Inflammation Prognostic Index has an excellent discriminatory power in predicting mortality for patients treated with regorafenib. 22 As for the prognostic predictors of the RS regimen, there have not been reported because of its inprevalence in mCRC. In our study, the following characteristics like old age (year > 60), less than 10 times CEA levels at baseline, lung metastasis, wild RAS, higher differentiation, without target therapy in the first- or second-line treatment, and lymph node negative seems to be associated with a better prognosis for regorafenib, because patients with the above factors in the regorafenib group had a similar OS to those in the RS group.
As for AEs, we concluded that RS was associated with higher toxicity than regorafenib overall. RS was generally well tolerated, and the common toxicities observed in our study resembled those reported in our published study. 10 In the RS group, the most common AEs were fatigue, bone marrow depression, liver dysfunction, and hyperpigmentation. The neutropenia (8%), leucopenia (7%), thrombocytopenia (6%), anorexia (6%), and fatigue (6%) were the most common grade 3/4 toxicities.
The AEs of regorafenib reported here are consistent with those reported in other clinical trials.5,23,24 In the regorafenib group, the most common grade 3/4 toxicities were hand–foot skin reaction (15%), hypertension (10%), and hyperbilirubinemia (6%), which were significantly higher than those in the RS group. However, the incidences of bone marrow depression and gastrointestinal AEs in the regorafenib group were lower than that in the RS group. As a result, the early and proactive prophylaxis, and management of AEs, especially neutropenia, leucopenia, thrombocytopenia, liver function test abnormalities, hand–foot skin reaction, and hypertension, which were the most common AEs needing therapeutic intervention, are essential to ensure patients can keep on therapy.
The study had the following limitations that should be considered: first, this was a non-randomized retrospective research with a small sample size, and the treatment regimen was chosen by different researchers, which resulted in potential selection bias. Therefore, we conducted PSM analysis between the RS and the regorafenib group. Second, patients’ quality of life, one of the crucial factors in salvage-line setting, could not be evaluated. Third, all the enrolled patients in our study were Chinese. Nevertheless, there were no ethnic differences in previous phase-III trials on the safety and efficacy of RS and regorafenib; therefore, our outcomes may be applied to the majority regardless of ethnicity. Further prospective clinical trials in larger cohorts are warranted to validate this study results.
Conclusion
Our study suggests that RS therapy is associated with a longer OS, with manageable toxicities, compared with regorafenib as the primary treatment choice in patients with standard treatments who failed mCRC.
Supplemental Material
sj-docx-1-tag-10.1177_17562848221098246 – Supplemental material for Efficacy and safety of raltitrexed plus S-1 versus regorafenib in patients with refractory metastatic colorectal cancer: a real-world propensity score matching study
Supplemental material, sj-docx-1-tag-10.1177_17562848221098246 for Efficacy and safety of raltitrexed plus S-1 versus regorafenib in patients with refractory metastatic colorectal cancer: a real-world propensity score matching study by Yu-Wen Zhou, Jia-Ling Wang, Qing-Fang Li, Yuan-Lin He, Lin-Juan Li, Rui-Zhi Liu, Ye Chen, Shuang Zhang, Meng Qiu and Ji-Yan Liu in Therapeutic Advances in Gastroenterology
Footnotes
Acknowledgements
The authors thank all medical oncologists for participating in the study, as well as the colleagues, and nurses involved.
Ethics approval and consent to participate
The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice. The study was approved by the Institutional Review Board of West China Hospital, Sichuan University, Chengdu, China [(Approval number: 2021 Review (NO.1416)] on 10 November 2021. Patients’ details were hidden. Exemption of informed consent was granted by Institutional Review Board of West China Hospital, Sichuan University.
Consent for publication
Not applicable
Author contribution(s)
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Sichuan Science and Technology Department Key Research and Development Project (grant no. 2019YFS0539), 1.3.5 Project for Disciplines of Excellence, West China Hospital, Sichuan University (grant no. ZYJC18022), Sichuan Provincial Science and Technology Department (grant no. 22ZDYF1228), Chengdu Science and Technology Project (grant no. 2021-YF05-01982-SN).
Conflict of interest statement
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Availability of data and materials
The data analyzed in this study are available from the corresponding author on reasonable request.
Supplemental material
Supplemental material for this article is available online.
References
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.
