Abstract
Background/Aim: The clinical significance of many RAS-family mutations in colorectal cancer (CRC) remains unclear. The purpose of this study was to investigate the relationship of RAS mutations on an exon basis (i.e., mutations in KRAS exons 2, 3, and 4 and in NRAS) with clinicopathological features and prognosis in CRC. Patients and Methods: We performed a retrospective cohort study of the medical records and frozen tissue samples of 268 consecutive patients with stage I-III CRC who underwent curative resection at a single institution between 2014 and 2018. Results: The RAS mutation rate was significantly associated with age and histology. Patients with KRAS exon 2 mutations exhibited shorter recurrence-free survival compared to those with KRAS wild-type, KRAS exon 3 mutations, KRAS exon 4 mutations, and NRAS mutations (73.0% vs. 85.5%, 86.7%, 85.7%; p=0.031). Age and histology were independent risk factors for RAS mutations. RAS mutations were independent prognostic factors with respect to recurrence-free survival in patients with stage I-III CRC. Conclusion: In stage I-III CRC patients, KRAS exon 2 mutations had the worst prognosis, whereas KRAS wild type, exon 3 mutations, exon 4 mutations, and NRAS mutations had better prognoses.
Colorectal cancer (CRC) is one of the most frequently diagnosed malignancies worldwide, and the second most common cause of cancer-associated deaths in Japan (1). Cancer treatments have made great progress in recent years. Comprehensive systemic treatment for unresectable advanced cancer is effective and important. Biomarkers that are highly predictive of the prognosis of cancer patients are in great demand (2, 3). Identifying predictors of recurrence can improve prognosis in CRC patients undergoing curative surgery.
In recent years, nutritional markers, inflammatory markers, and gene mutations have been attracting attention as factors for predicting prognosis (4-8). Detection of RAS mutations has emerged as an important assessment method for patients with CRC due to its clinical value in predicting prognosis (9). Anti-epidermal growth factor receptor (EGFR) antibodies play an important role in the therapy for CRC. Investigating biomarkers, is imperative for choosing the most appropriate drug for CRC treatment (10-12).
KRAS exons 2, 3, and 4 mutations or NRAS mutations are common in CRC (13). Mutations occur in about 20-25% of human cancers; in CRC, KRAS mutations are found in about 40% of cases and more frequently affect exon 2 (96%). NRAS mutations are found in about 3-5% of CRC, and are more frequently located in exon 3 (60%) (11).
KRAS mutations are routinely tested for metastatic CRC in clinical practice. Mutations in KRAS have been widely reported to be strongly associated with resistance to anti-EGFR therapy (14, 15). However, KRAS exon 3, KRAS exon 4, and NRAS mutations occur in 3-5% of CRC and have not been extensively studied due to their low mutation rates (16); thus, the prognostic value of these mutations remains unclear and no consensus has been reached on the clinicopathologic features and prognosis of patients with mutations according to exon (9). We previously reported the poor prognosis of KRAS G12V and G12C mutations (7). The aim of this study was to assess the prognostic value of RAS mutations by exon and the clinicopathologic features of the patients carrying these mutations.
Patients and Methods
Patient selection. We retrospectively analyzed 286 consecutive patients with CRC who underwent curative resection at Teikyo University Hospital, Japan, from 2015 through 2018. This study included patients identified as having pathological stages I-III CRC according to the 8th edition of the American Joint Committee on Cancer (AJCC) staging system (7).
The exclusion criteria were: 1) multiple primary malignancies, 2) history of familial adenomatous polyposis or Lynch syndrome, 3) previous chemoradiation for rectal cancer. Standard demographic and clinicopathologic data were collected for each patient, including information regarding tumor (T) node (N), and metastasis (M) stage. Tissue samples were surgically excised only after obtaining informed consent from each patient. The present study was approved by the Ethics Committee of the Teikyo University (Registration Number: 19-153).
Follow-up. Curative surgical resection was defined as gross and histological complete tumor clearance with no evidence of distant metastasis. The patient was followed every 3 months for the first 3 years and every 6 months for the next 2 years. All follow-ups included physical examination and testing for the tumor markers serum carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9). Computer tomographic scans of the chest and abdomen were usually taken every 6 months. At 1 year postoperative follow-up, all patients underwent complete colonoscopy. After that, patients after rectal cancer surgery underwent colonoscopy once a year, and patients with colon cancer underwent colonoscopy once every two years. Cancer recurrence was defined as the appearance of clinical, radiological, and/or pathological diagnosis of tumor locally or distant from the original site (7).
RAS mutation analysis. Testing for RAS/BRAF mutations was performed at Hoken Kagaku Laboratories (Kanagawa, Japan) using samples collected from tumor tissues. The selected area of formalin-fixed paraffin-embedded samples (FFPE) was deparaffined, followed by DNA isolation from the samples using a QIAamp DNA FFPE Tissue Kit (Qiagen, Manchester, UK) according to the manufacturer’s instructions. DNA quantification was performed on a NanoDrop 2000c (ThermoFisher Scientific, Waltham, MA, USA). The obtained DNA was amplified by PCR-reverse sequence-specific oligonucleotide (PCR-rSSO) on an Applied Biosystems VeritiTM 200 Thermal Cycler (Thermo Fisher Scientific) using MEBGENTM RASKET-B kit (MBL, Tokyo, Japan). Cycling conditions were the following: 5 min at 40°C, 2 min at 95°C, followed by 10 cycles at 95°C for 20 s and 62°C for 30 s, and then 45 cycles at 90°C for 20 s, 60°C for 30 s and 72°C for 30 s. Finally, 1 min at 72°C, followed by 95°C before the product was cooled down at 4°C. The amplified PCR product was then hybridized with probes in Beads Mix in Hybridization Buffer provided with the MEBGENTM RASKET-B kit. The reaction was performed at 95°C for 2 min followed by 55°C for 20 min. The product was purified according to the manufacturer’s instructions and incubated with fluorescent Phycoerythrin-labelled streptavidin. Fluorescence was measured by flow cytometry on a Luminex 100/200 System (Luminex, Austin, TX, USA), and the data were analyzed with the associated UniMAG software (Luminex) (7).
Statistical analysis. Between-group comparisons were performed with the chi-square test or Fisher’s exact test for proportions, and the Mann–Whitney U-test for continuous variables. The 3-year recurrence-free survival (RFS) was defined as the period between the date of surgery and the date of any tumor recurrence within 3 years after surgery. Survival was compared by determining the Kaplan–Meier curves, and the differences in survival were evaluated using the log-rank test. Cox regression analysis was used to identify factors significantly associated with RFS. Differences with a p-value of <0.05 were considered significant in all analyses. All statistical analyses were performed using JMP 15 software (SAS Institute Inc., Cary, NC, USA) and GraphPad Prism v5.0 (GraphPad Software Inc., La Jolla, CA, USA).
Results
Patient characteristics and frequency of RAS mutations. All 268 patients enrolled in the present study were diagnosed with either CRC stage I 30.6% (n=82), stage II 39.6% (n=106), or stage III 29.9% (n=80) (Table I). The rate of RAS mutations in stage I CRC was 28.1%, in stage II was 40.6%, and in stage III was 31.3% (Table I).
Mutation rate according to TNM stage.
Mutation characteristics of the RAS gene. As Table II shows, in the 268 tumor samples, a total of 128 RAS mutations (47%; 128/268) were detected. 93.7% (120/128) of the samples involved KRAS mutations and 6.3% (8/128) NRAS mutations. The KRAS exon 2 mutation was the most common mutation, seen in 82% (105/128) of cases (Figure 1).
Association of clinicopathological features with mutational status.
Frequencies of RAS mutation subtypes.
Clinicopathological characteristics of the RAS group. We investigated the relationship between RAS mutations and age, sex, the location of the primary tumor, histology, depth of tumor invasion, lymph node metastasis, lymph invasion, venous invasion, CEA level, and CA19-9 level (Table II). In the RAS mutant group, age at diagnosis was higher than that in the wild-type group (p=0.037). Regarding histology, compared to the RAS wild type group, the mutant group had primarily poorly differentiated CRC rather than well-differentiated or moderately differentiated tumors (p=0.023) (Table II).
Relapse-free survival of patients in relation to RAS status. Fifty-three patients (19.8%; 53/268) developed recurrence after a median of 1,315 days postoperatively. As Table III shows, univariate analysis identified histology, depth of tumor invasion, lymph node metastasis, lymph invasion, venous invasion, CEA level, CA19-9 level, and RAS mutation status as predictive of 3-year RFS. In multivariate analysis and while controlling for other factors, histology, depth of tumor invasion, lymph node metastasis, CEA level, and RAS mutation status remained statistically significant (p=0.010, 0.009, 0.001, 0.01, 0.006) (Table III).
Univariate and multivariate analysis of 3-year relapse-free survival stratified by clinicopathological features.
Three-year RFS rates of patients in the wild-type group and mutant group were 85.8% and 75.5%, respectively (p=0.018) (Figure 2). Wild-type CRC had a significantly better prognosis than mutant CRC.
Kaplan–Meier curve of 3-year relapse-free survival of patients with colorectal cancer stratified by RAS status (mutated or wild type).
Clinicopathological characteristics and survival according to the location of the RAS mutation. We divided the status into KRAS exon 2, KRAS exon 3 or exon 4 mutations, and NRAS mutations, and examined clinicopathological factors and prognosis. Regarding clinicopathological factors, depth of tumor invasion and venous invasion were significantly different (p=0.033, 0.043) (Table IV).
Comparison between KRAS exon 2 mutations, KRAS exon 3 mutations or KRAS exon 4 mutations, and NRAS mutations.
We analyzed the 3-year RFS in each mutation group by exon. Three-year RFS rates in patients in the wild-type group, KRAS exon 2, and KRAS exon 3, exon 4 or NRAS mutation group were 85.5%, 73.0%, and 86.7% (p=0.016) (Figure 3). The KRAS exon 2 mutant group had the shortest 3-year RFS and the difference was significant.
Kaplan–Meier curve of 3-year relapse-free survival of patients with colorectal cancer stratified according to the specific RAS mutation (KRAS exon 2 mutation vs. KRAS exon 3 or exon 4 mutation or NRAS mutation vs. Wild, *KRAS exon 2 mutation vs. KRAS exon 3 or exon 4, **KRAS exon 2 mutation vs. Wild type).
Discussion
We investigated the clinicopathological factors and prognostic value of RAS mutations in primary tumors from 268 consecutive patients with stage I-III CRC. Our analyses showed a significant proportion of the patients (47.8%) had tumors bearing RAS mutations, and the majority (105/128) were in exon 2. Notably, RAS mutated CRC did not show significant differences in terms of clinical and pathological characteristics, except for a higher prevalence of mucinous histology and higher age. RAS mutations in exon 2 were associated with worse RFS whereas the KRAS exon 3 mutations, KRAS exon 4 mutations, and NRAS mutations were not significantly different from wild type, and thus considered low-risk.
The most common oncogenic mutations in CRC are KRAS point mutations at positions 12, 13, and 61, approximately 90% of which are in exon 2 (17). These mutations reduce GTPase hydrolase activity and preserve protein activity. Within our population, KRAS exon 2 mutations were independently associated with worse 3-year RFS compared to wild-type. In contrast, exon 3, exon 4, and NRAS mutations had 3-year RFS similar to that of wild type. Mutations in exon 2 were found to increase the risk of recurrence two-fold over the wild-type, and these findings are consistent with those of other reports (9, 18, 19).
A study analyzing the behavior of KRAS exon 2 mutants found that the G12V variant had reduced GTPase activity, 25% of that of the G12D mutant and 10% of that of wild type (7, 20, 21). The exon 2 mutations also had reduced affinity for binding GTPase-activating proteins, further reducing GTPase function (21). This phenomenon alters the threshold at which cancer apoptosis is induced, potentially enhancing the transforming capacity of cells and avoiding apoptosis (22).
This study has three limitations. First, it had a retrospective design and consisted of patients from a single institution. Second, the sample size was relatively small (n=268). Third, patients underwent various invasive surgical procedures for CRC. Mortality and morbidity due to surgical techniques were not considered. Our findings warrant further review and validation in CRC patients from many centers and countries.
This retrospective study demonstrated that patients with stage I-III CRC with KRAS exon 2 mutations had the worst prognosis, whereas KRAS exon 3, exon 4, and NRAS mutations predicted a better prognosis. Thus, further studies on treatment efficacy should evaluate patients with KRAS exon 2 mutations separately from those with other RAS mutations and the specific mutation should be considered when predicting the clinical outcome of patients and individualizing therapies. The results of this study may constitute an additional prognostic factor in CRC and should be further explored in future studies.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number JP 22K08784 and ACRO Research Grants of Teikyo University. The Authors are grateful to Prof. Kazuaki Yokoyama and Dr. Kotaro Hama for their assistance in this study.
Footnotes
Authors’ Contributions
Study design: KA, TH, YH, KN and KM. Data collection, and analysis: TH, TM, KA, RS, YF, KM and TF.
Drafting of manuscript: TH, YH.
Conflicts of Interest
The Authors have no conflicts of interest to declare in relation to this study.
- Received January 30, 2023.
- Revision received February 13, 2023.
- Accepted February 15, 2023.
- Copyright © 2023 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- Chemotherapy Efficacy in Patients With Colorectal Cancer Experiencing Early Recurrence During or After Adjuvant Chemotherapy
- PIK3CA Mutated Colorectal Cancers Without KRAS, NRAS and BRAF Mutations Possess Common and Potentially Targetable Mutations in Epigenetic Modifiers and DNA Damage Response Genes