Abstract
Background/Aim: This study investigated the treatment patterns and prognosis of patients with metastatic or unresectable colorectal cancer (mCRC) treated with chemotherapy with targeting agents. Patients and Methods: This longitudinal multicenter study included 963 patients with mCRC who were treated in Korea between 2016 and 2020. Treatment patterns and efficacy were compared according to the mutation status and clinical factors. Results: As first-line therapy, most of the patients (83.5%) received FOLFOX plus bevacizumab (35.4%), followed by FOLFIRI plus bevacizumab (18.8%), FOLFIRI plus cetuximab (17.0%), and FOLFOX plus cetuximab (12.3%). Bevacizumab was the most frequent agent (78.8%) combined with chemotherapy in RAS-mutated CRC, while cetuximab (57.2%) in RAS wild-type CRC. Cetuximab was frequently combined with a doublet regimen in patients with left-sided CRC than in those with right-sided CRC (34.4% vs. 16%). As second-line therapy, most patients (63.4%) also received doublet regimens with bevacizumab, and FOLFIRI plus aflibercept was administered in 15.1%. The objective response rate with FOLFIRI plus cetuximab was significantly higher in patients with left-sided CRC than in those with right-sided CRC (59.2% vs. 30.8%, p=0.008) and marginally higher in patients with RAS wild-type CRC than in those with RAS-mutated CRC (55.6% vs. 0.0%, p=0.092). Progression-free survival (PFS) with FOLFOX plus bevacizumab was significantly shorter than that with FOLFIRI plus bevacizumab (p=0.030) in RAS-mutated CRC, whereas there were no significant differences between regimens in RAS wild-type CRC. Conclusion: In patients with unresectable metastatic colorectal cancer, doublet chemotherapy with targeting agents is the most common therapy and efficacy depends on the mutation status as well as clinical factors.
In the past two decades, the treatment of metastatic colorectal cancer (mCRC) has evolved significantly. The standard palliative treatment for mCRC is fluorouracil-based combination chemotherapy (with oxaliplatin or irinotecan), with or without biological agents targeting vascular endothelial growth factor (VEGF) or epidermal growth factor receptor (EGFR) (1), and the median overall survival (OS) has improved by approximately 20-30 months (2-6). The latest version of the National Comprehensive Cancer Network (NCCN) guidelines recommends the use of anti-EGFR agents based on the RAS and BRAF mutation status or primary tumor location (1). Treatment strategies using molecular tumor tests to select tailored therapies have gradually gained attention with promising results. For instance, immune checkpoint inhibitors (ICI) (7-10) and BRAF inhibitors (11, 12) have emerged as treatment options for microsatellite instability-high (MSI-H) CRC and BRAF-mutated CRC, respectively. Anti-HER2 therapy can also be considered for HER2-amplified CRC without RAS and BRAF mutations (13, 14), and NTRK fusion is a new target for the treatment of advanced solid cancers, including CRC (15).
However, real-world practice, where all molecular tests were not routinely performed, could affect adherence to this molecular testing guide (16), and the reimbursement of anticancer therapy by the National Health Insurance (NHI) needs to be considered when deciding on treatment options. In Korea, the health care system is implemented under the NHI program, a universal social insurance program that covers the entire population and is compulsory by law. Doublet chemotherapy (fluorouracil plus oxaliplatin or irinotecan) has been widely used to treat mCRC since the early 2000s. Cetuximab in combination with doublet chemotherapy as first-line treatment for KRAS wild-type CRC, and bevacizumab in combination with doublet chemotherapy as first- or second-line treatment for mCRC are reimbursed since 2014. Aflibercept in combination with fluorouracil plus irinotecan as second-line therapy is reimbursed for mCRC progressing on oxaliplatin-based chemotherapy since 2017. The treatment patterns for mCRC may change over time, and the effects of regimen choices and patterns on patient survival remain unknown. Moreover, data on how these agents are used in routine clinical practice are limited. Therefore, we conducted a longitudinal observational study to investigate the treatment patterns and prognosis of patients with mCRC treated with chemotherapy with targeting agents.
Patients and Methods
Study design and patients. This multicenter, longitudinal, observational study evaluated the treatment patterns and prognosis of patients with mCRC who started first-line palliative chemotherapy between December 2016 and July 2019 at 27 institutions in the Republic of Korea. Eligible patients were those aged 20 years, with histologically or cytologically confirmed metastatic or unresectable adenocarcinoma of the colon or rectum, and no prior history of palliative systemic anticancer treatment. This study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice, and was approved by the Institutional Review Board (IRB) of each participating center. All information was obtained with the appropriate IRB reporting.
Data collection. We collected data on patients’ demographics; body mass index, Eastern Cooperative Oncology Group Performance Status (ECOG PS); family history of colorectal cancer or adenomatous polyps in first relatives; primary tumor location; sites of metastasis (liver, lung, peritoneum, bone, brain, or lymph node); KRAS, NRAS, and BRAF mutation status; serum carcinoembryonic antigen (CEA) level; treatment patterns; and clinical outcomes. Right-sided CRC included cecum, ascending, and transverse colon cancer, and left-sided CRC included descending, sigmoid, and rectal cancers. The functional assessment of cancer therapy-colorectal (FACT-C) scores were calculated to evaluate the quality of life during chemotherapy at baseline, after two months of treatment, and after one year of treatment. The subscales included physical well-being (PWB), social well-being (SWB), emotional well-being (EWB), and functional well-being (FWB) (17).
Statistical analysis. Categorical and quantitative data were compared using the chi-square test or Fisher’s exact test and Mann-Whitney U-test, respectively. Treatment patterns were examined according to lines of therapy, and the frequency of each chemotherapy regimen was estimated. Treatment patterns were compared according to the RAS or BRAF mutation status, primary tumor sidedness (right-sided vs. left-sided), and age (<70 years vs. ≥70 years). The objective response rate (ORR) and disease control rate (DCR) were compared between the frequently used chemotherapy regimens, and the association of RAS mutation status or sidedness of the primary tumor (right-sided vs. left-sided) with ORR and DCR was investigated. Progression-free survival (PFS) was calculated from the date of systemic anticancer treatment initiation to the date of disease progression or death from any cause, whichever occurred first. The OS was calculated from the date of systemic anticancer treatment initiation to the date of death from any cause. Survival curves were estimated using the Kaplan-Meier method and compared using log-rank tests between the most frequently used chemotherapy regimens. The associations of RAS mutation status, primary tumor sidedness (right-sided vs. left-sided), and age (<70 years vs. ≥70 years) with PFS and OS were investigated. Cox proportional hazards models were used for univariate and multivariate analyses, and multivariate analysis included potential variables (p<0.1) in the univariate analyses. A p-value <0.05 was considered statistically significant. Statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA), and all tests were two-sided with 5% defined as the level of significance.
Results
Patient characteristics. A total of 963 patients with histologically or cytologically confirmed mCRC were included in this study. Table I summarizes the baseline characteristics of the patients. The median age was 62 years (range=22-90 years), and 62.7% of the patients were men. Approximately one-third of the patients (29.3%) were elderly (>70 years), and most patients (72.8%) had left-sided CRC. The cohort included patients with KRAS-mutated CRC (n=369/831, 44.4%), NRAS-mutated CRC (n=37/797, 4.6%), and BRAF-mutated CRC (n=28/422, 6.6%). Only 24 patients underwent MSI testing, which identified all as having microsatellite stable (MSS) CRC.
Baseline characteristics.
Treatment patterns of first-line chemotherapy for mCRC. Cytotoxic chemotherapy was administered to all 963 patients as first-line treatment (Table II). Approximately half of the patients were treated with fluorouracil and oxaliplatin (FOLFOX)-based chemotherapy (n=541, 56.2%) and less than half of the patients were treated with fluorouracil and irinotecan (FOLFIRI)-based chemotherapy (n=408, 42.4%). Most patients (n=804, 83.5%) received doublet regimens (FOLFOX or FOLFIRI) with targeting agents (bevacizumab or cetuximab), specifically, FOLFOX plus bevacizumab (n=341, 35.4%), followed by FOLFIRI plus bevacizumab (n=181, 18.8%), FOLFIRI plus cetuximab (n=164, 17.0%), and FOLFOX plus cetuximab (n=118, 12.3%). Bevacizumab was the most frequently combined targeting agent in the doublet regimen (n=522, 54.2%), and cetuximab was administered to 282 (29.3%) patients. There were 144 (15.0%) patients treated with FOLFOX or FOLFIRI without targeting agents and 13 (1.3%) treated with capecitabine monotherapy. Among the 963 patients, 101 (10.5%) underwent metastasectomy during chemotherapy. Of these, 68 were treated with FOLFOX with targeting agents; 22 with FOLFIRI with targeting agents; six with FOLFIRI alone; and five with FOLFOX alone.
Treatment regimens from first-line to second-line chemotherapy in patients with metastatic colorectal cancer.
Treatment patterns of first-line chemotherapy in each patient subgroup. The frequency of first-line chemotherapy regimens was investigated according to the mutation status, primary tumor location, and age (Figure 1). In RAS-mutated CRC, FOLFOX or FOLFIRI plus bevacizumab was the most frequent regimen (n=313, 78.8%), and only six (1.5%) patients were administered cetuximab as the targeting agent. In RAS wild-type CRC, FOLFOX or FOLFIRI plus cetuximab was the most frequent regimen (n=261, 57.2%), and 142 (31.1%) patients received FOLFOX or FOLFIRI plus bevacizumab. However, the frequency of bevacizumab or cetuximab combined with a doublet regimen was not different between BRAF-mutated and wild-type CRC: bevacizumab (64.3% vs. 63.7%) and cetuximab (28.6% vs. 27.7%), respectively. According to the primary tumor location, cetuximab was frequently combined with a doublet regimen in patients with left-sided CRC, compared with those with right-sided CRC (34.4% vs. 16%). FOLFOX or FOLFIRI plus bevacizumab was the most common regimen, regardless of the primary tumor location. In older patients with CRC (≥70 years), FOLFOX or FOLFIRI with bevacizumab or cetuximab still remained the most frequent regimen. However, FOLFOX or FOLFIRI alone or capecitabine were administered more frequently in older patients with CRC (≥70 years) than in non-elderly patients (<70 years) (22.3% vs. 13.8%).
Treatment regimens of first-line chemotherapy according to the mutation status and clinical factors. FOLFOX: Fluorouracil and oxaliplatin; FOLFIRI: fluorouracil and irinotecan; FOLFOXIRI: fluorouracil, oxaliplatin, and irinotecan.
Treatment patterns of second- or third-line chemotherapy for mCRC. Of the 963 patients, 405 (42.1%) experienced disease progression, and 172 (17.9%) received second-line chemotherapy (Table II). Most of these patients (n=109, 63.4%) received doublet regimens (FOLFOX or FOLFIRI) with bevacizumab, and FOLFIRI plus aflibercept was administered to 26 (15.1%) patients. Among the 172 patients, 12 (7.0%) underwent metastasectomy during chemotherapy. Furthermore, twenty-nine patients received third-line chemotherapy, and the most frequent regimens included capecitabine (n=11, 37.9%) and regorafenib (n=7, 24.1%).
Treatment responses and survival with frequent first-line chemotherapy regimens. The ORR ranged from 37.9% to 54.1% with first-line use of FOLFOX or FOLFIRI plus bevacizumab or cetuximab and was lower in patients receiving FOLFOX or FOLFIRI without targeting agents (22.4%-28.8%) (Table III). The DCR ranged from 74.2% to 91.4% with first-line chemotherapy (Table III). There were no significant differences in the ORR and DCR with most first-line chemotherapy regimens according to the RAS mutation status and primary tumor location, respectively (Table IV). However, the ORR with FOLFIRI plus cetuximab was significantly higher in patients with left-sided CRC than in those with right-sided CRC (59.2% vs. 30.8%, p=0.008) and higher in patients with RAS wild-type CRC than in those with RAS-mutated CRC (55.6% vs. 0.0%, p=0.092) (Table IV).
Clinical response with frequent chemotherapy regimen.
Objective response and disease control rate with first-line chemotherapy according to the RAS status and primary tumor location.
During the median follow-up duration of 18.8 months, the median PFS was 12.9 months and the median OS was not reached. Patients treated with FOLFOX plus bevacizumab had a significantly shorter PFS than those treated with FOLFIRI plus bevacizumab (p=0.009) and FOLFOX or FOLFIRI plus cetuximab (p=0.036 and p=0.047), respectively (Figure 2A). Patients who received FOLFOX or FOLFIRI without targeting agents seemed to have the shortest PFS (Figure 2A). However, the chemotherapy regimen did not remain a significant factor of PFS in the multivariate analysis (Table V). The median OS was the worst in patients receiving first-line FOLFOX or FOLFIRI without targeting agents (Figure 2B). The OS with first-line FOLFIRI plus cetuximab appeared to be shorter than that with FOLFOX or FOLFIRI with targeting agents (Figure 2B). Multivariate analysis revealed that higher ECOG PS (2), right-sided CRC, and elevated baseline CEA (5.0 ng/ml) were independent, significant factors for poor PFS, whereas older age (70 years), higher ECOG PS (2), NRAS-mutated CRC, and FOLFOX or FOLFIRI chemotherapy alone (vs. FOLFOX or FOLFIRI with bevacizumab) were independent significant factors for poor OS (Table V).
The Kaplan-Meier curves of (A) progression-free survival (PFS) with frequent first-line chemotherapy and (B) overall survival (OS).
Univariate and multivariate analysis for progression-free survival with first-line chemotherapy and overall survival.
The PFS with frequent first-line chemotherapy was also compared according to the RAS mutation status and primary tumor location. The PFS with FOLFOX plus bevacizumab was significantly shorter than that with FOLFIRI plus bevacizumab (p=0.030) in RAS-mutated CRC, whereas there were no significant differences between chemotherapy regimens in RAS wild-type CRC (Figure 3A and B). The PFS with FOLFOX plus bevacizumab was the shortest in left-sided CRC, whereas it seemed longer in right-sided CRC (Figure 3C and D).
The Kaplan-Meier curves of (A) progression-free survival (PFS) with frequent first-line chemotherapy and (B) overall survival (OS) according to the RAS mutation status, primary tumor location and age.
Additionally, PFS with FOLFIRI plus cetuximab appeared to be longer than with other chemotherapy regimens in left-sided CRC, but not in right-sided CRC (Figure 3C and D). For left-sided CRC, patients treated with FOLFIRI plus bevacizumab had longer PFS than those treated with other regimens (Figure 3D). There were similar trends in PFS between chemotherapy regimens in non-elderly patients (Figure 3E). In older patients with CRC, PFS with FOLFIRI plus cetuximab was significantly the shortest, similar to that with FOLFOX or FOLFIRI alone, whereas PFS with FOLFOX or FOLFIRI plus bevacizumab was the longest (Figure 3F).
FACT-C scales for quality of life during chemotherapy. Detailed FACT-C subscale scores are presented in Table VI. There were no significant differences in the PWB, SWB, EWB, and FWB scores between baseline, 2-month, and 1-year chemotherapy. The median values of PWB and EWB ranged from 4.0 to 6.0, and those of SWB and FWB ranged from 16.0 to 18.7 at each time point (Table VI). Similar trends were observed for PWB, SWB, EWB, and FWB scores, regardless of chemotherapy regimen.
FACT-C scales for quality of life during chemotherapy.
Discussion
The current study showed the treatment patterns and efficacy of chemotherapy with or without targeting agents in a large number of patients with mCRC treated in Korea. Common regimens, such as FOLFOX or FOLFIRI with bevacizumab or cetuximab and the use of targeting agents according to the RAS mutation status or sidedness of the primary tumor were in accordance with the international practice guidelines for the management of mCRC (1). This study also confirmed the comparable efficacy of chemotherapy in patients with mCRC in real-world practice to those in pivotal trials (2-6).
The demographic and clinical characteristics of our patients were similar to those reported in previous studies. However, molecular tumor tests, particularly those of the MSI status and BRAF mutations, were not performed sufficiently in this study. In the KEYNOTE-164 and CheckMate-142 trials, pembrolizumab and nivolumab, anti-programmed cell death protein 1 inhibitors, demonstrated improved ORR and PFS in previously treated patients with mismatch repair-deficient (dMMR)/MSI-H mCRC (8, 10). In the recent KEYNOTE-177 trial, pembrolizumab led to a significantly longer PFS than chemotherapy as the first-line treatment for MSI-H/dMMR CRC (9, 18). In addition, the BEACON trial demonstrated that a combination of encorafenib, cetuximab, and binimetinib significantly improved the OS and ORR compared to standard chemotherapy in patients with mCRC harboring the BRAF V600E mutation (11, 12). However, ICIs and BRAF inhibitors have not yet been reimbursed in Korea, which may contribute to the lower performing the abovementioned molecular tests prior to treatment choice. A relatively lower proportion of patients underwent the NRAS mutation test compared to the KRAS mutation test because it has been reimbursed since October 2017 in Korea. As the role of targeting therapy has become increasingly prominent in the treatment of mCRC, molecular tests have gradually become more common as a routine workup.
There was a substantial degree of overlap among the common chemotherapy regimens (FOLFOX or FOLFIRI) administered as first- and second-line regimens. This suggests that patients switched to the opposite regimen (FOLFOX to FOLFIRI or FOLFIRI to FOLFOX) as a backbone chemotherapy. Bevacizumab was the preferred biological agent used in combination with doublet chemotherapy, which may be due to its convenience of use—regardless of the mutation status—and relatively lower toxicity. Bevacizumab was more frequently combined with FOLFOX than with FOLFIRI chemotherapy, and aflibercept plus FOLFIRI was administered to patients whose cancer progressed on the previous oxaliplatin-based chemotherapy. Surgical resection during first- or second-line therapy was performed in 10.5% and 7% of patients, respectively, which appears to be higher than the 3%-10% reported in previous studies (2-5). This may have contributed to the longer median PFS (12.9 months) compared to that of previous studies. Metastasectomy can be actively considered regardless of the treatment line for selected patients in real-world practice. In the multivariate analysis, several well-known factors associated with prognosis, such as age, performance, sidedness of the primary tumor, RAS mutation status, and serum CEA, were also found in this study.
Cetuximab was mainly used as first-line treatment for RAS wild-type CRC, and bevacizumab was administered as the targeting agent in 31.3% of the patients with RAS wild-type CRC. Because cetuximab cannot be reimbursed as second-line treatment for RAS wild-type CRC, oncologists may prefer cetuximab in combination with doublet chemotherapy as first-line treatment, considering the subsequent use of bevacizumab after disease progression. The ORR with FOLFIRI plus cetuximab was marginally higher in patients with RAS wild-type CRC than in those with RAS-mutated CRC (55.6% vs. 0.0%, p=0.092), consistent with previous studies (5, 6). Despite the small number of patients, eight (28.6%) of the patients with BRAF-mutated CRC were treated with cetuximab plus doublet chemotherapy. Prior to the BEACON trial (11, 12), the BRAF mutation test may not have been actively performed to determine treatment choice, and the NCCN guidelines did not recommend the use of cetuximab based on the BRAF mutation status.
When the frequency of targeting agents was examined according to the sidedness of the primary tumor, bevacizumab was the most common agent regardless of the primary tumor location, and cetuximab was more frequently combined with doublet chemotherapy in patients with left-sided CRC than in those with right-sided CRC (34.4% vs. 16%). Patients with left-sided CRC had a significantly higher ORR with FOLFIRI plus cetuximab than those with right-sided CRC (59.2% vs. 30.8%, p=0.008). Patients with left-sided CRC may receive less benefit from FOLFOX plus bevacizumab in terms of PFS. These findings were in accordance with those of previous trials (19-22). The phase III CALGB/SWOG 80405 trial reported that OS with cetuximab plus chemotherapy was longer than that with bevacizumab plus chemotherapy (36.0 months vs. 31.4 months) in patients with left-sided CRC. The FIRE-3 trial reported similar results, with an OS of 38.3 months in patients on cetuximab plus chemotherapy vs. 28.0 months in those on bevacizumab plus chemotherapy (23). Among the patients with right-sided CRC, 16% received cetuximab because the primary tumor location was not considered when selecting targeting agents, based on the 2016 version of the NCCN guidelines. In addition, PFS with cetuximab plus doublet chemotherapy did not seem worse than that with other regimens in right-sided CRC. In a retrospective study, there was no significant difference in the ORR or PFS with cetuximab-containing chemotherapy (6.1 months vs. 4.2 months; p=0.278) between left- and right-sided CRC (24). This suggests that cetuximab can be cautiously used as a targeting agent in certain patients with right-sided CRC, and sidedness of the primary tumor along with the RAS mutation status is more significant for treatment choice than sidedness of the primary tumor alone.
Approximately one-third of the patients (29.3%) in this study were older (age ≥70 years). Similar trends were observed in first-line chemotherapy regimens between elderly and non-elderly patients; most older patients were also treated with FOLFOX or FOLFIRI plus bevacizumab or cetuximab. Despite the conflicting results, several previous studies have failed to demonstrate the superiority of doublet chemotherapy over fluoropyrimidine monotherapy (25-27) and there have been concerns regarding the toxicity of irinotecan in older or frail patients (28). However, oncologists’ preferences regarding treatment choices did not depend on age alone, and performance status was a more consistently significant factor for both PFS and OS in the multivariate analysis. Furthermore, the PFS with FOLFOX or FOLFIRI plus bevacizumab was longer than that with FOLFOX or FOLFIRI without targeting agents in these patients.
There were no significant differences in the median values of FACT-C scale scores at each time point, although the scores numerically decreased during chemotherapy. This suggests that oncologists actively monitor and manage patient toxicity during chemotherapy in routine practice, and the development of supportive care may have contributed to these results. In addition, specific toxicity itself may not disturb quality of life during chemotherapy, because there were similar trends in scores between chemotherapy regimens.
Study limitations. First, the follow-up duration was too short to analyze the patterns of second- or third-line therapies and OS. Second, because of the observational nature of the study, the results of molecular tests for mCRC were incomplete, particularly those of the BRAF mutation status and MSI. Third, this study could not include patients treated with ICIs, BRAF inhibitors, or novel agents because they were not covered by the NHI or not approved in Korea at that time; thus, their use was limited. Despite these limitations, the strengths of this study are the large number of patients with mCRC and reliable insight into treatment patterns and prognosis in real-world clinical practice.
In conclusion, the current study showed that doublet chemotherapy (FOLFOX or FOLFIRI) plus bevacizumab or cetuximab was the most common first- or second-line anticancer therapy and confirmed these regimens’ efficacy in patients with mCRC in real practice. Treatment choice may depend on the molecular mutation status and clinical factors.
Footnotes
Authors’ Contributions
JHK: Conceived and designed the analysis, collected data, contributed data or analysis tools, performed the analysis, and wrote the article. YC, SJS, YSP, JHK, CK, SHL, MHK, JGK, IGH, JWC, SHS, SYK, SCL, STL, JSK, HCJ, MHK, ISC, HWR, KHL, MHL, JYL, JHP, SYJ, NL, and CYP: Collected data and contributed data or analysis tools. YHK: Conceived and designed the analysis, collected data, and wrote the article.
Funding
Boryung Corporation was the funding source and was involved in all stages of the study conduct and analysis.
Conflicts of Interest
All Authors have no conflicts of interest to declare in relation to this study.
- Received November 27, 2023.
- Revision received December 5, 2023.
- Accepted December 15, 2023.
- Copyright © 2024 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
