Main

Although the small intestine accounts for 90% of the mucosal surface area of the gastrointestinal tract, small intestinal adenocarcinoma is rare and accounts for only approximately 2% of gastrointestinal malignancies.1 Epidemiologic observations have shown an increasing incidence of small intestinal adenocarcinoma in recent years, with an estimated 9160 new cases in 2014 in the United States.2

Small intestinal adenocarcinoma has a poor prognosis at all stages, which appears to be intermediate between that of colon and gastric cancer.3 Lymph node metastasis and distal location (jejunum and ileum) are the most important independent prognostic factors.4 As a result of the nonspecific nature of presenting clinical manifestations and the lack of effective tools for exploring the small intestine, small intestinal adenocarcinomas are usually diagnosed at an advanced stage,5 at which chemotherapy can only prolong overall survival. The cytotoxic agents used are generally the same as those used to treat patients with advanced colorectal cancers, and include oral or intravenous fluoropyrimidines, platinum and irinotecan.5 Unfortunately, there is no standard treatment and the place of new targeted therapies has still to be defined.1 Moreover, the number of tumor samples collected remains limited because of the low incidence of the disease, which explains the lack of information regarding somatic genetic alterations that occur during small intestinal carcinogenesis.

The RAS-RAF mitogen-activated protein kinase cascade has a crucial role in tumor cell proliferation. The KRAS oncogene is mutated in approximately 30–40% of colorectal cancers, mainly at codon 12 or 13. KRAS mutations are predictive of a lack of efficacy of anti-epidermal growth factor receptor (EGFR) monoclonal antibodies, such as panitumumab and cetuximab, in patients with metastatic colorectal cancer.6 Recently, several case reports have described the effect of anti-EGFR therapy in patients with wild-type KRAS small intestinal adenocarcinomas.7, 8 Moreover, a number of studies of combined treatments, such as CAPOX (capecitabine and oxaliplatin) with bevacizumab, CAPOX with irinotecan, CAPOX with panitumumab, GEMOX (gemcitabine and oxaliplatin) with erlotinib and Nab-paclitaxel, are exploring the effects of anti-EGFR agents in small intestinal adenocarcinoma.9 BRAF mutations, which always occur in the absence of KRAS mutations, have also been associated with resistance to anti-EGFR treatment in colorectal cancers.

KRAS and BRAF are two of the most frequently studied oncogenes, but they have been little studied in small intestinal adenocarcinoma.1, 3, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 In addition, because most of these few analyses were performed on only small numbers of small intestinal adenocarcinoma patients, they need to be validated.

In this study, we analyze the mutational status of KRAS and BRAF in small intestinal adenocarcinomas and consider the utility of targeted therapy in the light of their frequency. In addition, we evaluate their clinicopathologic and prognostic significance.

Materials and methods

Study Population

This study was approved by the Institutional Review Board of Incheon St. Mary’s Hospital (OC14OIMI0133). A total cohort of 197 surgically resected primary small intestinal adenocarcinoma cases was collected from the surgical pathology archives of 22 South Korean institutions by the Korean Small Intestinal Cancer Study Group, as previously reported.21 Carcinomas extending from the surrounding gastrointestinal tract organs, such as the stomach, ampulla of Vater, pancreas, cecum or appendix, into the small bowel were excluded from the analysis. Clinical and pathologic data collected as part of a previous study were used again in this study.21

Histologic types and tumor grading were classified according to the WHO classification.22 Briefly, tumor grading was classified as low-grade (well (>95% with gland formation) and moderately differentiated (50–95% with gland formation) adenocarcinomas) and high-grade (poorly differentiated (1–49% with gland formation) and undifferentiated (no gland formation)) carcinomas.22

Molecular Analysis

Ten sections with 10 μm in thickness from formalin-fixed paraffin-embedded tissue blocks were used and manually dissected to extract genomic DNA. Genomic DNA was extracted with a QIAmp DNA Mini Kit (Qiagen, Valencia, CA, USA) following the manufacturer’s protocol. Mutations in codons 12 and 13 of KRAS exon 1 and codon 600 of BRAF exon 15 were identified by cycle sequencing. The KRAS and BRAF genes were PCR amplified with primers for KRAS (F: 5′-TGACATGTTCTAATATAGTCAC-3′, R: 5′-ACAAGATTTACCTCTATTGTT-3′) and primers for BRAF (F: 5′-TCATAATGCTTGCTCTGATAGGA-3′, R: 5′-GGCCAAAAATTTAATCAGTGGA-3′). In general, PCR reactions were run in a total volume of 25 μl with 0.3 μM of each primer using AmpliTaq Gold PCR Master Mix (Applied Biosystems, Foster City, CA, USA). Samples were subjected to initial denaturation at 95 °C for 15 min, 40–45 cycles at 95 °C for 50 s, annealing for 50 s and elongation at 72 °C for 1 min, followed by final elongation at 72 °C for 7 min. PCR products were column-purified using a QIAquick PCR Purification Kit (Qiagen) or enzymatically treated with ExoSAP-IT (USB, Cleveland, OH, USA). The sequencing primers were identical to the PCR primers, and all samples were sequenced in both directions using a BigDye Terminator Cycle Sequencing Kit, version 1.1 (Applied Biosystems). The sequencing reactions were analyzed on an ABI Prism 3100 Genetic Analyzer with Sequencing Analysis software, version 3.7 (Applied Biosystems).

Statistical Analysis

Statistical analyses were performed using SPSS software (version 17.0; SPSS, Chicago, IL, USA). Student’s t-test, the χ2 test and Fisher’s exact test were used to examine associations between categorical variables. Multivariate relationships were estimated by fitting logistic regression models. Overall patient survival was defined as the time from surgical resection to death or last follow-up examination. Survival rates were calculated by the Kaplan–Meier method. Associations between survival rates and various clinicopathologic factors were assessed using the log-rank test. We also investigated the significance of any prognostic factors using Cox proportional hazards modeling. P-values <0.05 were considered to denote statistical significance.

Results

Patient Characteristics

Of the total of 197 patients, 190 with interpretable KRAS or BRAF sequencing results were included in our study. The mutation status of KRAS and BRAF was evaluable in 190 and 178 patients, respectively. We could not perform complete sequencing on the other samples because of a lack DNA or degradation of the DNA.

Patient ages ranged between 23 and 86 years (mean, 59.0 years; s.d., 14.0 years). The male to female ratio was 1.7. The median follow-up period after surgical resection was 28.6 months (range, 0–168.4 months). Predisposing conditions were observed in 22 cases (12%), including 16 cases of sporadic adenoma, 3 of Peutz–Jeghers syndrome, 2 of Meckel’s diverticulum and 1 of Crohn’s disease. Forty patients were classified as ‘suspected Lynch syndrome’ based on revised Bethesda guideline. There were no patients with familial adenomatous polyposis, Gardner syndrome, gluten-sensitive enteropathy or intestinal duplication. Tumors were located in the duodenum in 105 cases (55%), the jejunum in 56 cases (30%) and the ileum in 29 cases (15%).

KRAS and BRAF Mutations

KRAS mutations were observed in 32% (61/190 cases) of the small intestinal adenocarcinoma patients, whereas BRAF mutation was detected in only 1% (2/178 cases). The KRAS and BRAF mutations were mutually exclusive.

The KRAS mutations were mostly in codon 12 (Table 1). Fifty patients (50/61 cases, 82%) had KRAS mutations in codon 12, and 11 patients (11/61 cases, 18%) had KRAS mutations in codon 13. Based on the mutation sequence of KRAS, the most frequent KRAS mutation subtype—expressed in the recommended genetic nomenclature23—was p.G12D (31/61 cases, 51%), followed by p.G13D (11/61, 18%), p.G12C (7/61, 11%), p.G12V (6/61, 10%), p.G12A (4/61, 6%), p.G12R (1/61, 2%) and p.G12S (1/61, 2%). There was a significant difference in the mutation sequences of KRAS according to tumor location (P=0.042): p.G12D, p.G12C and p.G13D were more common in the proximal (duodenal) adenocarcinomas. p.G12A was detected only in distal adenocarcinomas, whereas p.G12V, p.G12R and p.G12S were found only in proximal adenocarcinomas. The most common base substitution in the KRAS mutations was a G to A transition (G>A; 43/61 cases, 71%), followed by G>T (13/61, 21%) and G>C (5/61, 8%). An association between the type of base substitution and tumor location was also observed (P=0.038): G>A and G>T substitutions were common in the proximal small intestine, whereas G>C was common in the distal small intestine.

Table 1 KRAS mutation status among small intestinal adenocarcinomas
Full size table

The Association Between KRAS or BRAF Mutations and Clinicopathologic Factors

The associations between KRAS mutation and clinicopathologic factors are summarized in Table 2. The patients with mutant KRAS tended to have higher T classifications (P=0.034) and more frequent pancreatic invasion (P=0.020) than those with wild-type KRAS. However, high-grade carcinomas were more common among adenocarcinomas with wild-type KRAS (P=0.036). There was no association between KRAS mutation and other clinicopathologic factors, including age and gender, growth pattern, tumor location, histologic type, nodal metastasis, other loop invasion, retroperitoneal seeding, and perineural and lymphovascular invasion.

Table 2 Correlation between clinicopathological factors and KRAS mutation status in small intestinal adenocarcinomas
Full size table

Similarly, patients with mutant KRAS or BRAF had higher T classification (P=0.040) and more frequent pancreatic invasion (P=0.017), whereas high-grade carcinomas were more frequent in adenocarcinomas with wild-type KRAS and BRAF (P=0.024).

Association Between KRAS Mutation and Clinicopathologic Factors According to Tumor Location

We analyzed in detail the relationship between pathologic factors and KRAS mutation according to tumor location (Table 3). Proximal adenocarcinoma patients with KRAS mutations tended to have tumors with high T classifications (P=0.022) and more frequent pancreatic invasion (P=0.045), whereas distal cases with KRAS mutations had lower-grade tumors (P=0.028).

Table 3 Correlation between clinicopathologic features and KRAS mutation status according to tumor location
Full size table

We next evaluated the relationship between pathologic factors and KRAS mutation subtypes according to tumor location (Tables 4 and 5). In the proximal adenocarcinomas, KRAS G>A and G12D mutations were closely associated with late T classification (P=0.015 and 0.009, respectively) and pancreatic invasion (P=0.011 and 0.003, respectively). By contrast, neither KRAS G>A nor G12D mutations were associated with any pathologic factors among the distal adenocarcinomas.

Table 4 Correlation between clinicopathologic features and the presence of KRAS G>A mutations according to tumor location
Full size table
Table 5 Correlation between clinicopathologic features and the presence of KRAS G12D mutations according to tumor location
Full size table

Association Between Subtypes of KRAS Mutations and Clinicopathologic Factors

We assessed the correlation of KRAS mutational subtype with clinicopathologic variables (Tables 6 and 7). In a univariate association analysis stratified by KRAS mutation subtype, KRAS G>A mutation was significantly more frequent in tumors with a higher T classification (P=0.015) and pancreatic invasion (P=0.017) than in KRAS wild-type tumors and those with mutations other than G>A. The KRAS G>A mutation was not related to tumor location or differentiation status. In a multivariate logistic regression analysis including location, differentiation, T stage and pancreatic invasion, the KRAS G>A mutation remained significantly more frequent in late T stage tumors (odds ratio=1.588, 95% CI: 1.085–2.324; P=0.017).

Table 6 Correlation between clinicopathologic features and the presence of KRAS G>A mutations
Full size table
Table 7 Correlation between clinicopathologic features and the presence of KRAS G12D mutations
Full size table

In a univariate analysis stratified by KRAS mutation subtype, the KRAS G12D mutation was more closely related to higher T classification (P=0.004) and pancreatic invasion (P=0.016) than the KRAS wild-type and mutations other than G12D. In a multivariate logistic regression analysis including location, differentiation, T classification and pancreatic invasion, KRAS G12D mutation remained significantly more frequent in adenocarcinomas with advanced T classification (odds ratio=0.519, 95% CI: 0.324–0.833; P=0.007).

Survival Analysis

Small intestinal adenocarcinoma patients with KRAS mutations (median survival time, 21.5 months) had shorter overall survival outcomes than those with wild-type KRAS, but the difference did not reach statistical significance (38.5 months; P=0.116, log-rank test; Figure 1a). The survival times of the two BRAF mutation carriers were also relatively short (P=0.675). There were no significant differences in survival time distribution between the patients with KRAS or BRAF mutations (median, 22.6 months) and those with wild-type KRAS and BRAF (38.5 months; P=0.148; Figure 1b).

Figure 1
figure 1

Survival of patients with small intestinal adenocarcinomas based on KRAS or BRAF mutational status. (a) Small intestinal adenocarcinoma patients with KRAS mutations tended to have relatively shorter overall survival outcomes (median, 21.5 months) than those with wild-type KRAS (38.5 months), but this effect does not reach statistical significance (P=0.116, log-rank test). (b) Small intestinal adenocarcinoma patients having either KRAS or BRAF mutation (median, 22.6 months) also tended to have shorter survival times than those with wild-type KRAS and BRAF (38.5 months; P=0.148). There were no significant differences in the survival time distributions of (c) patients with KRAS codon 12 mutations (median, 21.0 months) and those with codon 13 mutations (21.5 months; P=0.805), (d) patients with KRAS G>A mutations (median, 21.0 months) and those with KRAS wild-type or mutations other than G>A (36.5 months; P=0.398), and (e) patients with KRAS G12D (median, 17.3 months) and KRAS wild-type patients and those with mutations other than G12D (30.7 months; P=0.507).

Full size image

When overall survival times were compared based on mutational status, the median survival time of the patients with codon 12 KRAS mutations was not significantly different (21.0 months) from those with codon 13 mutations (21.5 months; P=0.805; Figure 1c). Small intestinal adenocarcinoma patients with KRAS G>A mutations were associated with shorter survival time (median, 21.0 months) than those with KRAS wild-type and those with other than G>A mutations (36.5 months), but this effect did not reach statistical significance (P=0.398; Figure 1d). There were no significant differences in survival time distributions between carriers with KRAS G12D (median, 17.3 months) and KRAS wild-type patients and those with mutations other than G12D (30.7 months; P=0.507; Figure 1e).

In terms of tumor location, no relation between KRAS mutation and overall survival was seen in proximal versus distal adenocarcinomas. In addition, KRAS mutation subtype was also not related to the patient survival in either proximal or distal adenocarcinomas.

When survival of small intestinal adenocarcinoma patients was compared depending on T classifications, significant survival differences were observed. In lower T classification (pT1–pT3) group, the survival time for patients with KRAS mutation (median, 46.0 months) was significantly shorter than those with wild-type KRAS (85.4 months; P=0.046; Figure 2a). However, in higher T classification (pT4) group, there was no significant difference in survival time distribution between the patients with KRAS mutations (median, 50.1 months) and those with wild-type KRAS (54.8 months; P=0.792; Figure 2b).

Figure 2
figure 2

Survival of small intestinal adenocarcinomas patients with KRAS mutation based on T classification. (a) In lower pT classification (pT1-pT3) subgroup, the survival time for patients with KRAS mutation (median, 46.0 months) was significantly shorter than those with wild-type KRAS (85.4 months; P=0.046, log-rank test). (b) In higher pT classification subgroup (pT4), there was no significant difference in survival time distribution between the patients with KRAS mutations (median, 50.1 months) and those with wild-type KRAS (54.8 months; P=0.792).

Full size image

After grouping according to tumor location, we also compared the survival of the patients. In the proximal adenocarcinomas, KRAS mutation was not related to the patient survival in either lower or higher pT classification. However, in the distal adenocarcinomas, the patients with KRAS mutations (median, 33.0 months) had significantly shorter overall survival outcomes than those with wild-type KRAS (78.7 months; P=0.026) in a lower pT classification group. On the contrary, in higher pT classification group of distal adenocarcinomas, the patients with KRAS mutation (median, 30.1 months) had slightly longer survival time than those with wild-type KRAS (22.1 months), but it was not statistically significant (P=0.428).

Prognostic Significance of KRAS Mutation in Lower pT Classification Tumors

The independent prognostic significance of KRAS mutation and other clinicopathologic factors in small intestinal adenocarcinomas of lower pT classification, which were considered significant by univariate analyses, was further evaluated using the Cox proportional hazards modeling (Table 8). According to this multivariate analysis, KRAS mutation (P=0.005) and the presence of lymphovascular invasion (P<0.0001) are poor independent prognostic predictors of overall survival in small intestinal adenocarcinoma patients with lower pT classification.

Table 8 Multivariate analysis of the small intestinal adenocarcinoma patients with lower T classifications (pT1–pT3)
Full size table

Discussion

We have sequenced the KRAS and BRAF genes of a large number of patients (n=190) in this study and found KRAS mutations in 32% (61 cases). Although a few previous studies of small intestinal adenocarcinomas have examined KRAS and/or BRAF mutations, which are well-known oncogenes in colorectal cancer, most of these analyses were performed on small numbers of patients (Table 9).1, 3, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 The frequency of KRAS mutations among small intestinal adenocarcinomas ranged from 9 to 83% in the previous studies. Previous studies from Japan with homogeneous ethnic groups showed a wide range of frequencies of KRAS mutation (9–71%),13, 14, 15, 19 and other studies with heterogeneous ethnic groups, such as studies from USA, also reported various frequencies of KRAS mutation (14–83%).10, 17, 18, 20 Various KRAS mutation frequencies in the previous studies may be related with different detection methods rather than diverse ethnic background. Several different techniques for detecting KRAS and/or BRAF mutation were used, including restriction fragment length polymorphism, allele-specific oligonucleotide hybridization, PCR-single-strand conformation polymorphism, allele discrimination assay and Sanger sequencing.1, 3, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 In addition to different detection techniques, differences in sample size may contribute to the wide range of KRAS mutation frequencies observed. The overall prevalence of KRAS mutations in the previous studies was 38% (150/400). In those studies with >35 KRAS mutations tested,3, 10, 11 the prevalence of KRAS mutation was reported to be between 31 and 43%. This is similar to the prevalence of KRAS mutations in this study. In addition, the prevalence of KRAS mutations among small intestinal adenocarcinomas in the previous studies was similar to that among colorectal cancers.3, 6

Table 9 Previous studies of KRAS and/or BRAF mutations in small intestinal adenocarcinoma
Full size table

On the other hand, BRAF mutations occur at a lower rate (0–13%) than KRAS mutations in colorectal cancers.23 BRAF mutations are also rarely encountered among small intestinal adenocarcinomas, where the prevalence of BRAF mutation ranges from 3 to 14%.1, 3, 11, 12 In this study, BRAF mutations were observed in only 1% of the small intestinal adenocarcinomas. In accord with previous reports for colorectal cancers23 and small intestinal adenocarcinomas,1, 3, 12 KRAS and BRAF mutations were mutually exclusive ways in our observation as the same with results of the previous studies.3, 12 Although one study reported concurrent KRAS and BRAF mutation in squamous cell carcinoma of the lung,24 to the best our knowledge, there is no report of concurrent KRAS and BRAF mutation in small intestinal adenocarcinomas. Further studies with larger number of cases are needed for identifying the existence of concurrent KRAS and BRAF mutations in small intestinal adenocarcinomas. In summary, the molecular alterations in small intestinal adenocarcinomas are close to those in colorectal cancers, with a high frequency of KRAS mutations and infrequent BRAF mutations.

Several retrospective studies have indicated that chemotherapy prolongs overall survival in patients with advanced small intestinal adenocarcinomas, but there is no standard frontline regimen owing to a lack of randomized trials.5, 25 Patients with advanced small intestinal adenocarcinomas are often treated with the same chemotherapy regimens as patients with advanced colorectal cancers or gastric cancers, especially 5-fluorouracil (FU) or 5-FU-based schedules.9 The combination of 5-FU and a platinum-based agent has been considered more effective than other regimens such as oxaliplatin-based chemotherapy.25, 26 The main challenge for the near future is to identify molecular markers involved in small bowel carcinogenesis that predict chemosensitivity, and thus to improve survival.5 Overman et al27 observed that a high proportion of small intestinal adenocarcinomas express both EGFR and vascular endothelial growth factor (VEGF), suggesting that these patients may benefit from therapeutic strategies targeting EGFR and VEGF. Targeted therapies such as monoclonal antibodies against VEGF or EGFR combined with chemotherapy have already exhibited significant efficacy in metastatic colorectal cancers.5 Given the similar prevalence of KRAS mutations in small intestinal adenocarcinomas and colorectal cancers, we think that targeted therapy with anti-EGFR monoclonal antibody would be particularly appropriate for small intestinal adenocarcinoma patients with wild-type KRAS and BRAF as it is for colorectal cancer patients.

Different genetic mutations may be responsible for different biological effects. In the literature on colorectal cancer, codon 12 mutations of the KRAS are associated with a mucinous phenotype.23 By contrast, codon 13 mutations tend to be nonmucinous and are characterized as more aggressive with a greater metastatic potential.23 In this study, we did not find any difference in clinicopathologic findings in terms of mucinous/nonmucinous histologic subtype with respect to codon 12/13 mutations of KRAS among the small intestinal adenocarcinomas.

Patients with KRAS and/or BRAF mutations had shorter survival than those with wild-type genes, but the effect was not statistically significant. There has been considerable controversy regarding the prognostic significance of KRAS/BRAF mutations.3, 10 Many studies have investigated the prognostic value of KRAS/BRAF mutations on retrospectively collected cohorts of colorectal cancers patients, with conflicting results.23, 28, 29, 30, 31, 32 Initially, KRAS was found to be an important prognostic indicator, but this finding was later restricted to G12V mutations.31 Colorectal cancer patients with G>A and G>C mutations tended to have a worse prognosis than those with G>T mutations.30 In addition, mutation of G>T but not G>A or G>C increased the risk of recurrence and death of colorectal cancer patients.32 To the best of our knowledge, there have been only two previously reported studies of small intestinal adenocarcinomas concerning the prognostic significance of KRAS mutations in Western populations,3, 10 and no reports about the prognostic significance of BRAF mutations. Aparicio et al3 reported that mutated KRAS status in small intestinal adenocarcinoma was an independent predictor of longer overall survival, but this was meaningful only in stage IV patients. Whereas about a third of study population (20/63, 32%) was stage IV patients in the study by Aparicio et al, no case of stage IV was present in our study. Fu et al10 proposed that KRAS G>A mutation was significantly correlated with late disease stage and poor tumor differentiation, and carriers had an increased risk of distant metastasis and relapse, as well as significantly shorter overall survival. However, in their study, only duodenal adenocarcinomas were included. In this study, KRAS G>A mutation as well as G12D mutation was closely related to advanced T classification, as in the study of Fu et al.

In addition, we found that mutated KRAS status was an independent predictor of poor survival in small intestinal adenocarcinoma, particularly in patients of lower pT classification. In the literature about colorectal cancers, KRAS mutations have been detected in the earliest neoplastic lesions found in colonic mucosa, and appear to exert a strong influence on the growth of polyps and early cancers. Nash et al33 reported that the patients with mutant KRAS and microsatellite stability had significantly worse survival than other groups among the early-stage colorectal cancer (stages I and II). To the best of our knowledge, this study is the first report to identify the prognostic significance of KRAS mutations in lower pT classification of small intestinal adenocarcinomas.

Codons 61 and 146 mutation are additional hotspots for KRAS mutations, and data from a small number of studies of colorectal cancers suggest that mutation at these sites predicts resistance to anti-EGFR therapy.34 Despite their growing clinical relevance, the clinicopathological and molecular features of colorectal cancers with KRAS codon 61 or 146 mutations remain largely unknown. A few studies have encountered mutations of codons 61 and 146 of KRAS in small intestinal adenocarcinomas.3, 15, 17, 18, 19 Arai et al19 described only one case of a codon 61 KRAS mutation among eight cases of mutant KRAS genes in small intestinal adenocarcinomas. Aparicio et al3 only documented KRAS codon 61 (n=2) and codon 146 (n=1) mutations in a total of 21 cases of mutated KRAS. As we only examined codons 12 and 13 KRAS mutations, we do not have information regarding the frequency of KRAS mutations of codons 61 and 146. Therefore, further studies are needed to define the predictive value of these mutations in small intestinal adenocarcinomas.

In conclusion, our data from 190 small intestinal adenocarcinoma patients demonstrate that KRAS and, infrequently, BRAF mutations are observed in a subset of small intestinal adenocarcinomas. KRAS mutations are associated with higher pT classification and pancreas invasion. Small intestinal adenocarcinoma patients with either KRAS or BRAF mutation have a tendency toward shorter survival than those with wild-type KRAS. Mutation of KRAS oncogene is a worse prognostic predictor in small intestinal adenocarcinomas of lower pT classification. Our observations suggest that targeted therapies such as anti-EGFR chemotherapy could be beneficial in the two-thirds of small intestinal adenocarcinoma patients with wild-type KRAS and BRAF if they have metastatic disease.