Background
Colorectal cancer was the third most common cancer in men (746,000 cases, 10.0% of the total) and the second in women (614,000 cases, 9.2% of the total) worldwide in 2012 [
1]. Mutations in
KRAS exon 2 occur in ~35% of all metastatic colorectal cancers (mCRCs) [
2,
3], and constitutively activate the mitogen-activated protein kinase (MAPK) pathway [
4,
5]. These mutations are validated biomarkers for resistance to anti-epidermal growth factor receptor (EGFR) therapy in patients with mCRC [
6-
11]. Although conventional
KRAS tests are useful to exclude patients without benefit from anti-EGFR therapy, response rates and disease control rates to anti-EGFR antibody monotherapy among patients with
KRAS exon 2 wild-type tumors are only 13–17% and 51%, respectively [
6,
7]. Therefore, more accurate patient selection requires identification of other predictive factors to improve the risk–benefit profile of anti-EGFR therapy.
Until recently, there have been no validated biomarkers other than
KRAS exon 2 mutations. Recently, several reports have shown that other
KRAS (exons 3 or 4) and
NRAS mutations (exons 2– 4) occur in ~20% of mCRC patients with
KRAS exon 2 wild-type tumors, which are associated with resistance to anti-EGFR therapy for mCRC [
12-
18].
BRAF mutations were detected in 5–10% of patients with mCRC with V600E as a hot spot.
BRAF is a downstream molecule of
KRAS and the clinical data suggest that
BRAF V600E mutations are associated with poor prognosis in patients with mCRC [
11,
12,
19-
24]. However, the relationship between
BRAF mutations and the efficacy of anti-EGFR therapy remains controversial [
19-
22]. Besides the KRAS–BRAF pathway, the other major downstream signaling pathway activated by EGFR is the PI3K–AKT signaling pathway.
PIK3CA mutations, most of which were in exons 9 and 20, were detected in 10–15% of patients with mCRC. According to a European Consortium report [
19],
PIK3CA mutations in exon 20 but not in exon 9 were associated with resistance to anti-EGFR therapy for mCRC. However, in other studies, no clear correlation between
PIK3CA mutations and the efficacy of anti-EGFR therapy has been observed [
21,
22]. Meanwhile, targeting agents for these mutations are under development.
We previously reported that a multi-gene cancer panel with Luminex technology (GENOSEARCH Mu-PACK, MBL, Japan) is useful for detection of 36 mutations in
KRAS exons 3 or 4,
NRAS,
BRAF and
PIK3CA in a single reaction using 50-ng template DNA from formalin-fixed, paraffin-embedded (FFPE) specimens [
25]. Importantly, the analysis of 82 samples was fully concordant with conventional direct sequencing. However, information about the frequencies and clinicopathological features of these mutations in clinical practice, including the relationship between mutation status and the efficacy of anti-EGFR therapy, especially among Asian populations, is still limited.
In the present study, we evaluated the frequencies andclinicopathological features of KRAS, NRAS, BRAF and PIK3CA mutations in Japanese mCRC patients, and assessed their corresponding effects on the efficacy of anti-EGFR therapy.
Methods
Patients
We have conducted a retrospective observational study in our institution to evaluate the frequencies and clinicopathological features of KRAS, NRAS, BRAF and PIK3CA mutations in Japanese mCRC patients. Principal inclusion criteria were as follows: histologically confirmed adenocarcinoma of the colon or rectum; and presence of unresectable metastatic disease.
Between January 2013 and June 2014, we analyzed 264 patients with mCRC who met the inclusion criteria. The study was conducted with the approval of the National Cancer Center Institutional Review Board. Written informed consent was obtained from as many patients as possible. For the deceased patients and their relatives, we also disclosed the study design at the website of National Cancer Center and gave them the opportunity to express their wills in accordance with the Epidemiological Study Guideline of Ministry of Health, Labour and Welfare in Japan.
Molecular profiling and data analysis
Genomic DNA was extracted from FFPE cancer specimens (239 primary tumors and 25 metastases). A total of 36 mutations were analyzed using Luminex (xMAP) technology (GENOSEARCH Mu-PACK, MBL), including: KRAS codon 61 (Q61K, Q61E, Q61L, Q61P, Q61R and Q61H); KRAS codon 146 (A146T, A146S, A146P, A146E, A146V and A146G); NRAS codon 12 (G12S, G12C, G12R, G12D, G12V and G12A), codon 13 (G13S, G13C, G13R, G13D, G13V and G13A); codon 61 (Q61K, Q61E, Q61L, Q61P, Q61R and Q61H); BRAF codon 600 (V600E); PIK3CA exon 9 codon 542 (E542K); codon 545 (E545K); codon 546 (E546K); and exon 20 codon 1047 (H1047R, H1047L). The lower limit of the percentage of mutant allele in the tumor samples accepted by the study was 5%. Initially, 50-ng samples of template DNA were collected from FFPE tissue samples and were amplified using polymerase chain reactions (PCRs) with a biotin-labeled primer. Subsequently, PCR products and fluorescent Luminex beads were bound to oligonucleotide probes that were complementary to wild-type and mutant genes, and were hybridized and labeled with streptavidin–phycoerythrin. Subsequently, the products were processed according to Luminex assays and data were analyzed using UniMAG software (MBL). The procedure took ~4.5 h. The status of KRAS exon 2 (codons 12 and 13) was evaluated by amplification using a refractory mutation system–Scorpion assay with 1% sensitivity in a central vendor laboratory.
Patient characteristics, including age, sex, site of primary lesion, histology, site of metastases, and treatment results, were collected from medical records. Sites of primary lesions were divided into right colon, left colon, and rectum. Right-sided tumors were defined as those arising anywhere from the cecum to the transverse colon, and left-sided tumors as those arising anywhere from the splenic flexure to the rectosigmoid junction. The efficacy of anti-EGFR therapy was evaluated according to gene status in patients who met the following inclusion criteria: Eastern Cooperative Oncology Group performance status (ECOG PS) score ≤ 2, KRAS exon 2 wild type, at least one prior chemotherapy regimen, treatment with anti-EGFR either as monotherapy or in combination with irinotecan or FOLFIRI (5-FU, L-leucovorin and irinotecan), baseline computed tomography (CT) performed within 28 days of anti-EGFR therapy, initial evaluation of treatment effect via CT scan within 3 months of initial anti-EGFR therapy and adequate hematological, hepatic and renal function.
Statistical methods
Gene mutation frequencies and associations of RAS or BRAF mutations with clinicopathological features were estimated in mCRC patients.
Response rate (RR) and disease control rate (DCR; including complete or partial response and stable disease) were evaluated for anti-EGFR therapy according to the Response Evaluation Criteria in Solid Tumors (RECIST; version 1.1). Progression-free survival (PFS) was defined as the time from initial administration of anti-EGFR regimens until the first objective evidence of disease progression or death from any cause. Overall survival (OS) was defined as the time from initial administration of anti-EGFR regimens until death from any cause. For PFS or OS, patients were censored at the time of their last follow-up if they were free of disease progression or alive, respectively. PFS and OS rates were estimated using the Kaplan–Meier method, and differences among the groups according to KRAS, NRAS, BRAF and PIK3CA gene status were identified by univariate and multivariate analyses using Cox proportional hazards models and presented as hazard ratios (HRs) with 95% confidence intervals (CIs). Confounders in univariate and multivariate analyses included ECOG PS (0 vs. 1 and 2), numbers of metastatic sites (1 vs. ≥ 2), treatment line of anti-EGFR regimens (2nd vs. 3rd) and types of anti-EGFR regimens (monotherapy vs. combination therapy).
The χ2 test, Fisher’s exact test, Mann–Whitney U test, or Kruskal–Wallis test was used to compare patient characteristics and treatment response, as appropriate. Statistical analyses were performed using IBM SPSS Statistics version 21 (IBM Corporation, Armonk, NY, USA). All tests were two-sided, and differences were considered significant when P was < 0.05.
Discussion
We elucidated the prevalence of KRAS, NRAS, BRAF and PIK3CA mutations in Japanese mCRC patients, and clarified the relationship between gene status and clinicopathological features, including the efficacy of anti-EGFR therapy. To date, clinical evidence about these mutations in mCRC has been based on clinical studies in western countries. The present study is believed to be the first to provide information on frequency and type of KRAS, NRAS, BRAF and PIK3CA mutations in Japanese patients with mCRC. In addition, the clinical feasibility of the present novel multiplex kit was demonstrated.
In our patient cohort, the frequency of patients with
KRAS exon 2 (34.1%) mutant tumors was similar to that in previous studies [
2-
4]. A total of 12.1% of patients without
KRAS exon 2 mutations had other
RAS mutations, which was lower than that in recent studies from western countries, which showed 15–26% of these mutations [
12-
18]. Another previous study from Japan showed that other
RAS mutations were detected in seven (12.7%) of 55 samples without
KRAS exon 2 mutations with 3–13% sensitivity [
26], which was similar to our result. Several possible explanations for the relatively lower frequency of other
RAS mutations in our study compared with western studies might be considered. First, there were some differences in detectable
RAS mutations by multiplex kit between our study and western studies. In our study, we did not analyze
KRAS codons 59 and 117 and
NRAS codons 59, 117 and 146, while these codons were analyzed in most western studies. Although the frequencies of these mutations are considered to be low, it might be one of the causes of the lower frequency in our patient cohort. Second, the sensitivity of
RAS mutation analysis may vary among studies. In the present study, all mutations were detectable with 5–10% sensitivity. In contrast, Surveyor Scan Kits, BEAMing technology and pyrosequencing were used in pivotal studies, and
RAS mutations were detected with 1–10% sensitivity [
12-
18]. A recent multicenter study in Japan, including our institution, showed that other
RAS mutations were detected in 15% of patients with
KRAS exon 2 wild type, using a newer multiplex kit (MEBGEN RASKET Kit) [
27]. This method detected 48
RAS mutations in exon 2 (codons 12 and 13), exon 3 (codons 59 and 61) and exon 4 (codons 117 and 146), with 1–5% sensitivity in a single reaction using 50–100-ng DNA from FFPE tissue without manual dissection. Given these methodological differences, further studies are required to confirm differences in the prevalence of other
RAS mutations between Asian and western populations. In this study, we detected
BRAF mutations in 5.4% of patients. The prevalence of
BRAF mutation might be dependent on the patient population studied. mCRC patients with
BRAF mutant tumors have a poor prognosis, so the prevalence of
BRAF mutant populations may decline in pretreated patients compared with chemonaïve patients. The prevalence of
BRAF mutations in our patient cohort was similar to that of previous studies of pretreated patients with mCRC [
11,
12,
19-
24].
We also investigated the clinicopathological features of mCRC patients with respect to
RAS and
BRAF mutations. Primary rectal tumor tends to be more frequently observed in
KRAS exon 2 and other
RAS mutant tumors rather than
RAS wild-type tumors, although this was not statistically significant. Previous studies showed that
KRAS exon 2 mutation was significantly higher in the right colon [
28,
29], in disagreement with our analysis. No significant differences in other clinicopathological features such as age, sex, primary lesion, histology, and site of metastasis were observed between
KRAS exon 2 and other
RAS mutant tumors, which is similar to previous studies [
30]. Regardless of these clinicopathological features, it is reported that other gene expression profiles based on The Cancer Genome Atlas appear to be similar in patients with
KRAS and
NRAS mutant mCRC, suggesting that treatment selection based on molecular profile is important [
30]. In accordance with previous reports [
23,
24],
BRAF mutant tumors are more likely to develop in the right colon, and to have poorly differentiated or mucinous adenocarcinoma, and peritoneal metastasis in comparison with
BRAF wild-type tumors.
In agreement with previous studies [
19,
25], mutations in
KRAS exons 3 or 4,
NRAS,
BRAF or
PIK3CA were not associated with clinical benefits from anti-EGFR therapy in the present cohort. On the basis of recent prospective and retrospective randomized trials of anti-EGFR therapy [
12-
18], the National Comprehensive Cancer Network (NCCN) recommends anti-EGFR therapy for mCRC patients without other
RAS mutant tumors or
KRAS exon 2 mutant tumors [
31]. The Japanese Society of Medical Oncology (JSMO) also recommends testing for all
RAS mutations in patients with mCRC before anti-EGFR therapy. In contrast, whether
BRAF and
PIK3CA mutations are predictive of the efficacy of anti-EGFR therapy remains controversial [
19-
22]. Previous trials suggest that intensive combination chemotherapy with FOLFOXIRI (5-FU, L-leucovorin, irinotecan, and oxaliplatin) and bevacizumab might be especially effective for
BRAF mutant mCRC [
32]. Recently, the combination of
BRAF inhibitors and anti-EGFR monoclonal antibodies, with or without PI3K inhibitors or MEK inhibitors, has shown promising results in phase I trials in patients with
BRAF mutant CRC [
33,
34]. Patients with
BRAF mutant CRC are often refractory to systematic chemotherapy and have poor prognosis, therefore, screening for
BRAF mutations is important during recruitment of patients for these clinical trials. Accordingly, we conducted a multi-institutional screening (GI-SCREEN) study using the present multiplex kit to elucidate the nationwide prevalence of these targetable mutations.
There were several methodological limitations to the present study. First, not all of the patients in our study period were evaluated for their RAS gene status. Thus, the analysis may have been subject to some selection bias. Second, the small sample size and single-center population were other major limitations. Owing to the overall small number of patients with KRAS exon 3 or 4, NRAS, BRAF or PIK3CA mutations, we could not evaluate the impact of each gene mutation on the efficacy of anti-EGFR therapy. In addition, our analyses were explorative and hypothesis generating. This issue should be analyzed in a larger cohort.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
KS and TY (Yoshino) conceived the study design. AK carried out the majority of molecular genetic studies and analyses of the clinical data. KS, SF, YK, HB, WO, TK, NF, TD, and TY (Yoshino) provided clinical data and helped collect tumor tissues. TY (Yamanaka) statistically analyzed the clinical data. AO coordinated the study and helped to draft the manuscript. All authors have read and approved the final manuscript.