Abstract
Objective
To determine the impact of RAS mutation status on survival and patterns of recurrence in patients undergoing curative resection of colorectal liver metastases (CLM) after preoperative modern chemotherapy.
Summary Background Data
RAS mutation has been reported to be associated with aggressive tumor biology. However, the effect of RAS mutation on survival and patterns of recurrence after resection of CLM remains unclear.
Methods
Somatic mutations were analyzed using mass spectroscopy in 193 patients who underwent single-regimen modern chemotherapy before resection of CLM. The relationship between RAS mutation status and survival outcomes was investigated.
Results
Detected somatic mutations included RAS (KRAS/NRAS) in 34 patients (18%), PIK3CA in 13 (7%), and BRAF in 2 (1%). At a median follow-up of 33 months, 3-year overall survival (OS) rates were 81% in patients with wild-type vs 52.2% in patients with mutant RAS (P=0.002); 3-year recurrence-free survival (RFS) rates were 33.5% with wild-type vs 13.5% with mutant RAS (P=0.001). Liver and lung recurrences were observed in 89 and 83 patients, respectively. Patients with RAS mutation had a lower 3-year lung RFS rate (34.6% vs 59.3%, P<0.001), but not a lower 3-year liver RFS rate (43.8% vs 50.2%, P=0.181). In multivariate analyses, RAS mutation predicted worse OS (hazard ratio [HR] 2.3, P=0.002), overall RFS (HR 1.9, P=0.005), and lung RFS (HR 2.0, P=0.01), but not liver RFS (P=0.181).
Conclusions
RAS mutation predicts early lung recurrence and worse survival after curative resection of CLM. This information may be used to individualize systemic and local tumor-directed therapies and follow-up strategies.
INTRODUCTION
For patients with colorectal liver metastases (CLM), hepatic resection combined with systemic therapy has been reported to be the most effective treatment in terms of improving survival.[1] To date, various prognostic scoring systems have been proposed to stratify patient prognosis after resection of CLM.[2–4] However, in the era of modern systemic therapy, these conventional scoring systems are becoming less relevant because of inconsistent predictive power and lack of reproducibility due to selection bias.[5]
Preoperative systemic therapy has increasingly been used as part of a multidisciplinary approach for patients with CLM to test the biologic aggressiveness of the tumor and select optimal candidates for surgery.[6] Our group has reported that pathologic response[7–9] and radiologic response[10, 11] to preoperative chemotherapy are powerful predictors of long-term outcomes for patients with CLM. However, in patients with CLM, there is strong variability in clinical presentation, biologic aggressiveness, and patterns of treatment failure, and no biomarker predicts these phenotypic differences.
During the past decade, mutation status of RAS family genes (predominantly KRAS and NRAS) has been shown to correlate with the effectiveness of anti-EGFR agents against unresectable metastatic colorectal cancer,[12, 13] and a possible prognostic role of these somatic gene mutations after resection of CLM in an era predating the use of preoperative chemotherapy has been reported.[14] Our group has recently reported that metachronous CLM detected after modern chemotherapy for the primary colorectal tumor are associated with a higher incidence of somatic gene mutations and worse survival.[15] In addition, while recently reported studies have indicated a per-patient concordance of mutation type between primary tumor and metastases, a higher rate of mutation has been identified in patients with metastases at particular sites (e.g., peritoneum, lung and brain metastases), suggesting that tumors with mutations have a propensity to metastasize to the lungs and brain.[16–19]
On the basis of these clinical findings, we hypothesized that somatic mutation status predicts survival outcomes and types of recurrence after curative resection of CLM. In this study, we investigated the impact of RAS mutation status on survival and patterns of recurrence in patients who underwent curative resection of CLM after preoperative modern chemotherapy.
PATIENTS AND METHODS
Study Population
The Institutional Review Board of The University of Texas MD Anderson Cancer Center approved this study protocol (PA11-0653). The prospectively maintained liver resections database of the Department of Surgical Oncology was queried to identify all patients who underwent liver resection for CLM during the period from November 1997 through October 2011. We studied patients undergoing curative hepatectomy without concomitant radiofrequency ablation. Patients with a history of previous treatment for metastatic disease (chemotherapy, radiofrequency ablation, or resection) were excluded.
All patients included in this study received preoperative oxaliplatin- or irinotecan-based chemotherapy including the anti-VEGF agent bevacizumab. Patients who received anti-EGFR agents were excluded. Patients who received 2 or more regimens of preoperative chemotherapy because of disease progression during first-line chemotherapy were excluded from the current study, as were patients who died within 90 days after hepatectomy and patients who had less than 5% viable tumor cells in the CLM specimen.
Preoperative, Intraoperative, and Postoperative Management
Before operation, all patients underwent a medical history, physical examination, laboratory evaluation, and imaging studies, including helical computed tomography of the chest, abdomen, and pelvis with a triphasic liver protocol. In selected patients, fluorodeoxyglucose positron emission tomography was used to rule out extrahepatic disease and confirm the metastatic nature of atypical lesions. Only patients with hepatic and extrahepatic disease amenable to complete and safe resection were considered for hepatectomy. In patients with an anticipated insufficient future liver remnant, preoperative portal vein embolization was used to induce hypertrophy.
During laparotomy, intraoperative sonography of the liver was performed to confirm the location of known CLM and their relation to the portal pedicles or the hepatic veins and to rule out the presence of previously undetected CLM. Parenchymal transection was carried out under total or selective hepatic inflow occlusion using the Cavitron Ultrasonic Surgical Aspirator (Valleylab, Boulder, CO) and saline-linked cautery (dissecting sealer DS 3.0, Tissuelink Medical, Inc, Dover, NH) as reported previously.[20]
All specimens were subjected to histologic evaluation to confirm the diagnosis of metastatic colorectal cancer and determine the width of the tumor-free surgical margin. The degree of pathologic response of CLM to preoperative chemotherapy was defined according to the percentage of the CLM tumor surface area composed of viable tumor cells: major pathologic response was defined < 50% viable cells and minor pathologic response ≥ 50% viable cells.[7] After surgery, chemotherapy was usually reintroduced to complete a total of 12 cycles including both preoperative and postoperative chemotherapy. Patients were reassessed every 4 months after completion of the second stage of liver resection. Radiological evidence or positive biopsy was required to confirm recurrence, and time and site of relapse were systematically recorded. Further treatment was decided according to the findings at reassessment.
DNA Extraction and Somatic Gene Mutation Profiling
Hematoxylin-eosin-stained slides from all CLM were reviewed by a gastrointestinal pathologist (DMM). Tumor viability in the CLM specimens was checked in order to exclude specimens with tumor viability less than 5%. Areas with maximum amount of available tumor were selected for macrodissection. Tumor tissue was scraped from the glass slides under direct visualization or under a dissecting microscope, and DNA was extracted from tumor tissue using a QIAmp DNA Mini Kit (Qiagen, Valencia, CA).
Somatic gene mutations were assessed using mass spectrometry. DNA extracted from formalin-fixed, paraffin-embedded resected CLM was quantified and analyzed with Sequenom MassARRAY technology (Sequenom, Inc., San Diego, CA).[21] Sequenom’s MassARRAY system utilizes polymerase chain reaction amplification and single-base primer extension for mutation detection.[22–24] The MassARRAY system offers a highly effective method for profiling hundreds of somatic mutations in parallel. A high-throughput analysis of 159 point mutations in 33 genes commonly involved in solid tumors was performed in MD Anderson’s Characterized Cell Line Core Facility. The genes tested for this study were AKT1, AKT2, AKT3, ALK, BRAF, CDK4, CTNNB1, DEAR1, EGFR, ERA, FRAP, GNAS, HIF1A, IDH1, IDH2, IGFR1R, JAK2, KIT, KRAS, MEK1, MET, NRAS, PDGFRA, PDPK1, PHLPP2, PIK3CA, PIK3R1, PRKAG1, PRKAG2, RET, RICTOR, STK11, and TNK2.
Statistical Analysis
Qualitative variables were expressed as frequencies. Patients were stratified according to KRAS or NRAS mutation status into 2 groups: mutant RAS and wild-type RAS. Clinicopathological features were compared between these 2 groups using chi-square or Fisher’s exact tests, as appropriate. Overall survival (OS), overall recurrence-free survival (RFS), lung RFS, and liver RFS were calculated using the Kaplan–Meier method from the date of liver resection to the date of death, first recurrence at any site, lung recurrence, or liver recurrence, respectively. Patients without an event during the follow-up period were censored at the date of last follow-up. These survival outcomes were compared using the log-rank test.
To identify factors associated with OS and RFS in the entire study cohort, we evaluated the following clinicopathologic variables in a univariate analysis: disease-free interval after the primary tumor diagnosis (<12 months vs ≥12 months), primary tumor location (rectum vs colon), regional lymph node status of the primary tumor (positive vs negative), RAS mutation status (mutant vs wild-type), number of cycles of preoperative chemotherapy for CLM (>6 vs ≤6), pathologic response to preoperative chemotherapy (major vs minor), number of CLM in the pathologic specimen (multiple vs solitary), diameter of the largest of the CLM in the pathologic specimen (>5 vs ≤5 cm), major postoperative complications were defined as complications of grade 3 or higher (necessitating a surgical, endoscopic or radiological procedure) in the Dindo classification[25], and liver resection margin status on microscopic analysis (positive vs negative),.
All variables associated with OS, overall RFS, lung RFS, and liver RFS with P<0.1 in the univariate proportional hazards models were entered into a Cox multivariate regression model with backward elimination. P<0.05 was considered statistically significant. Statistical analyses were performed using the software IBM SPSS Statistics, version 19 (IBM, Armonk, NY).
RESULTS
Among 1406 consecutive patients treated for CLM at MD Anderson during the study period, 621 were excluded because of concomitant radiofrequency ablation or nonreceipt of preoperative chemotherapy. An additional 497 patients were excluded because they received multiple lines of preoperative chemotherapy, did not receive bevacizumab, received anti-EGFR agents before or after liver resection, had less than 5% viable tumor cells in the specimen, died within 90 days after surgery, or underwent a noncurative hepatectomy. Among the 288 patients eligible for the genetic testing, 95 patients were excluded because they did not have available paraffin blocks or had insufficient DNA for genetic analysis. The remaining 193 patients were studied in detail (Figure 1).
FIGURE 1.
Selection of study population.
Somatic Gene Mutation Status
Of the 193 patients included in the study, 43 (22.3%) had one or more somatic mutations in tested genes. Thirty-four patients (17.6%) had RAS mutations (27 KRAS and 7 NRAS), 13 patients (6.7%) had PIK3CA mutations, 2 patients (1%) had BRAF mutations, and 2 patients (1%) had rare mutations—one had a CTNNB1 mutation, and the other had an AKT1 mutation. Among the 34 patients with RAS mutations, 29 (85%) exhibited a mutation at codon 12 (nucleotide changes: G→A in 17 patients, G→T in 8 patients, and G→C in 4 patients), 3 (9%) exhibited a mutation at codon 61 (A→G in 1 patient, A→C in 1 patient, and C→G in 1 patient), and 2 (6%) exhibited a mutation at codon 13 (G→A in both of them). Among the 13 patients with PIK3CA mutations, 6 (46%) exhibited a mutation at codon 20 and 7 (54%) at codon 9.
Patient Characteristics by RAS Mutation Status
Clinicopathologic characteristics by RAS mutation status are shown in Table 1. Patients with RAS mutation had a lower rate of major pathologic response (< 50% viable tumor cells) than patients with wild-type RAS (38.2% vs 58.5%; P=0.037). The remaining characteristics did not differ significantly between patients with mutant and wild-type RAS.
Table 1.
Clinicopathologic Characteristics by RAS Mutation Status
| Characteristic | Wild-type RAS (n=159) |
MutantRAS (n=34) |
Univariate Analysis P Value |
|---|---|---|---|
| Disease-free interval <12 months, no. (%) (n=130) | 107 (67.3) | 23 (67.6) | 0.968 |
| Rectal primary tumor, no. (%) (n=39) | 32 (20.1) | 7 (20.6) | 0.951 |
| Primary tumor positive nodal status, no. (%) (n=133) | 107 (67.3) | 26 (76.5) | 0.294 |
| No. of cycles of preoperative chemotherapy, median (range) | 6 (2–23) | 6 (3–24) | 0.345 |
| % viable tumor cells <50%, no. (%) (n=106) | 93 (58.5) | 13 (38.2) | 0.037 |
| No. of CLM, median (range) | 2 (1–80) | 3 (1–18) | 0.569 |
| Diameter of largest of CLM, mm, median (range) | 25 (5–150) | 22.5 (5–100) | 0.484 |
CLM indicates colorectal liver metastases.
Long-term Survival and Predictors of Outcomes
At a median follow-up time of 33 months, 51 patients had died (48 of cancer and 3 of other causes), 54 were alive with disease recurrence, and 88 were alive with no evidence of disease at last follow-up. For the entire cohort, 3- and 5-year RFS rates were 29.9% and 26.9%, respectively, and 3- and 5-year OS rates were 76.4% and 61.8%, respectively. Patients with mutant RAS had worse long-term outcomes than those with wild-type RAS (3-year RFS: 13.5% vs 33.5%, P=0.001; 3-year OS: 52.2% vs 81%, P=0.002) (Figure 3).
FIGURE 3.
Overall survival (OS) (A) and recurrence-free survival (RFS) (B) according to RAS mutation status.
In multivariate analysis, independent risk factors of worse RFS were RAS mutation (hazard ratio [HR], 1.9; 95% CI, 1.2–3.0; P=0.005) and minor pathologic response to preoperative chemotherapy (HR, 2.0; 95% CI, 1.4–3.0; P<0.001) (Table 2). Independent risk factors of worse OS also were RAS mutation (HR, 2.3; 95% CI, 1.1–4.5; P=0.002) and minor pathologic response (HR, 2.1; 95% CI, 1.1–4.0; P=0.022) (Table 3).
Table 2.
Univariate and Multivariate Analysis of Factors Associated with Recurrence-free Survival (RFS) in 193 Patients Who Underwent Somatic Gene Mutation Analysis
| Factor | 3-year RFS (%) |
5-year RFS (%) |
Univariate Analysis P Value |
Multivariate Analysis | |
|---|---|---|---|---|---|
| P Value | Hazard Ratio (95% CI) | ||||
| Disease-free intervala | |||||
| <12 months (n=130) | 27.8 | 22.4 | 0.147 | ||
| ≥12 months (n=63) | 34.2 | 34.2 | |||
| Primary tumor location | |||||
| Rectum (n=39) | 16.7 | 16.7 | 0.063 | NS | |
| Colon (n=154) | 32.5 | 26.7 | |||
| Primary tumor nodal status | |||||
| Positive (n=133) | 24.6 | 24.6 | 0.067 | NS | |
| Negative (n=60) | 42 | 33.5 | |||
| RAS mutation status | |||||
| Mutant (n=34) | 13.5 | - | 0.001 | 0.005 | 1.92 (1.21 – 3.03) |
| Wild-type (n=159) | 33.5 | 32.1 | |||
| No. of cycles of preoperative chemotherapy | |||||
| >6 (n=65) | 21.5 | 16.6 | 0.027 | NS | |
| ≤6 (n=128) | 34.5 | 27.6 | |||
| Pathologic response | |||||
| %VTC ≥50% (n=87) | 13.5 | 13.5 | <0.001 | <0.001 | 2.02 (1.36 – 2.99) |
| %VTC <50% (n=106) | 33.5 | 30.5 | |||
| No. of CLM | |||||
| Multiple (n=125) | 24.9 | 23.4 | 0.007 | NS | |
| Single (n=68) | 41.3 | 36.1 | |||
| Diameter of largest of CLM | |||||
| >5 cm (n=24) | 25.2 | 16.8 | 0.240 | ||
| ≤5 cm (n=169) | 30.7 | 28.3 | |||
| Major complication | |||||
| Yes (n=32) | 16.7 | 11.1 | 0.085 | NS | |
| No (n=161) | 32.8 | 28.6 | |||
| Surgical margin | |||||
| Poitive (n=19) | 24.6 | 24.6 | 0.833 | ||
| Negative (n=174) | 30.1 | 26.7 | |||
CLM indicates colorectal liver metastases; NS, not significant; %VTC, percentage viable tumor cells.
From diagnosis of primary tumor to diagnosis of CLM.
Table 3.
Univariate and Multivariate Analysis of Factors Associated with Overall Survival (OS) in 193 Patients Who Underwent Somatic Gene Mutation Analysis
| Factor | 3-year OS (%) |
5-year OS (%) |
Univariate Analysis P Value |
Multivariate Analysis | |
|---|---|---|---|---|---|
| P Value | Hazard Ratio (95% CI) | ||||
| Disease-free intervala | |||||
| <12 months (n=130) | 72.7 | 58.3 | 0.197 | ||
| ≥12 months (n=63) | 84 | 69 | |||
| Primary tumor location | |||||
| Rectum (n=39) | 71.8 | 54.2 | 0.611 | ||
| Colon (n=154) | 77.2 | 63.4 | |||
| Primary tumor nodal status | |||||
| Positive (n=133) | 72.5 | 53 | 0.004 | NS | |
| Negative (n=60) | 85 | 81.3 | |||
| RAS mutation status | |||||
| Mutant (n=34) | 52.2 | 44.7 | 0.002 | 0.002 | 2.26 (1.13 – 4.51) |
| Wild-type (n=159) | 81 | 65.4 | |||
| No. of cycles of preoperative chemotherapy | |||||
| >6 (n=65) | 65.5 | 51.2 | 0.017 | NS | |
| ≤6 (n=128) | 83.2 | 69 | |||
| Pathologic response | |||||
| %VTC ≥50% (n=87) | 65.2 | 45.2 | <0.001 | 0.022 | 2.10 (1.11 – 3.97) |
| %VTC <50% (n=106) | 86.1 | 64.5 | |||
| No. of CLM | |||||
| Multiple (n=125) | 73 | 58.4 | 0.160 | ||
| Single (n=68) | 82.8 | 67.8 | |||
| Diameter of largest of CLM | |||||
| >5 cm (n=24) | 58.9 | 46.4 | 0.030 | NS | |
| ≤5 cm (n=169) | 79.1 | 64.3 | |||
| Major complication | |||||
| Yes (n=32) | 64.1 | 53 | 0.108 | ||
| No (n=161) | 77.6 | 62 | |||
| Surgical margin | |||||
| Negative (n=174) | 75.1 | 59.1 | 0.312 | ||
| Positive (n=19) | 86.5 | 86.5 | |||
CLM indicates colorectal liver metastases; NS, not significant; %VTC, percentage viable tumor cells.
From diagnosis of primary tumor to diagnosis of CLM.
Patterns of Recurrence and Predictors of Recurrence Pattern
Of the 126 patients who had tumor recurrence during the follow-up period, 83 patients had a lung recurrence and 89 patients had a liver recurrence. Patients with mutant RAS had worse 3-year lung-RFS than those with wild-type RAS (34.6% vs 59.3%, P<0.001) (Figure 4A, Table 4). In contrast, 3-year liver-RFS was not influenced by RAS mutation (43.8% for mutant RAS vs 50.2% for wild-type RAS, P=0.181) (Figure 4B, table 4). At the last follow-up, lung recurrence was observed in 64.7% of patients with RAS mutation (22/34) versus 38.3% of patients with wild-type RAS (61/159) (P=0.005). The incidence of liver recurrence did not correlate significantly with RAS mutation status (44.7% of patients with RAS mutation vs 52.9% of those with wild-type RAS; P=0.379).
FIGURE 4.
Lung (A) and liver (B) recurrence-free survival (RFS) according to RAS mutation status.
Table 4.
Univariate and Multivariate Analysis of Factors Associated with Lung Recurrence after Resection of CLM
| Factor | 3-year Lung RFS (%) |
Univariate Analysis P Value |
Multivariate Analysis | |
|---|---|---|---|---|
|
P Value |
Hazard Ratio (95% CI) | |||
| Disease-free intervala | ||||
| <12 months (n=130) | 52.7 | 0.680 | ||
| ≥12 months (n=63) | 59.7 | |||
| Primary tumor location | ||||
| Rectum (n=39) | 34.3 | 0.016 | 0.069 | |
| Colon (n=154) | 59.4 | |||
| Primary tumor nodal status | ||||
| Positive (n=133) | 52.8 | 0.511 | ||
| Negative (n=60) | 59.7 | |||
| RAS mutation status | ||||
| Mutant (n=34) | 34.6 | <0.001 | 0.01 | 2.01 (1.20 – 3.41) |
| Wild-type (n=159) | 59.3 | |||
| No. of cycles of preoperative chemotherapy | ||||
| >6 (n=65) | 54.6 | 0.976 | ||
| ≤6 (n=128) | 55.9 | |||
| Pathologic response | ||||
| %VTC ≥50% (n=87) | 39.7 | 0.001 | 0.009 | 1.91 (1.17 – 3.10) |
| %VTC <50% (n=106) | 64.5 | |||
| No. of CLM | ||||
| Multiple (n=125) | 50.3 | 0.062 | NS | |
| Single (n=68) | 66 | |||
| Diameter of largest of CLM | ||||
| >5 cm (n=24) | 45.5 | 0.04 | NS | |
| ≥5 cm (n=169) | 56.4 | |||
| Major complication | ||||
| Yes (n=32) | 34.2 | 0.164 | ||
| No (n=161) | 57.7 | |||
| Surgical margin | ||||
| Positive (n=19) | 81.3 | 0.295 | ||
| Negative (n=174) | 73.1 | |||
CLM indicates colorectal liver metastases; NS, not significant; RFS, recurrence-free survival; %VTC, percentage viable tumor cells.
From diagnosis of primary tumor to diagnosis of CLM.
Multivariate analysis indicated that RAS mutation (HR, 2.0; 95% CI, 1.1–3.4; P=0.01) and minor pathologic response (HR, 1.9; 95% CI, 1.1–3.0; P=0.009) were independent predictors of lung RFS, while minor pathologic response was the only independent predictor of liver RFS (HR, 2.2; 95% CI, 1.4–3.5; P=0.001) (Tables 4 and 5).
Table 5.
Univariate and Multivariate Analysis of Factors Associated with Liver Recurrence after Resection of CLM
| Factor | 3-year Liver RFS (%) |
Univariate Analysis P Value |
Multivariate Analysis | |
|---|---|---|---|---|
| P Value | Hazard Ratio (95% CI) | |||
| Disease-free intervala | ||||
| <12 months (n=130) | 45 | 0.016 | NS | |
| ≥12 months (n=63) | 58.4 | |||
| Primary tumor location | ||||
| Rectum (n=39) | 44.3 | 0.533 | ||
| Colon (n=154) | 50.3 | |||
| Primary tumor nodal status | ||||
| Positive (n=133) | 42.3 | 0.018 | 0.072 | 1.64 (0.96 – 2.83) |
| Negative (n=60) | 64.6 | |||
| RAS mutation status | ||||
| Wild-type (n=159) | 43.8 | 0.181 | ||
| Mutant (n=34) | 50.2 | |||
| No. of cycles of preoperative chemotherapy | ||||
| >6 (n=65) | 43.2 | 0.164 | ||
| ≤6 (n=128) | 52.6 | |||
| Pathologic response | ||||
| %VTC ≥50% (n=87) | 33.1 | <0.001 | 0.001 | 2.22 (1.41 – 3.52) |
| %VTC <50% (n=106) | 58.8 | |||
| No. of CLM | ||||
| Multiple (n=125) | 42.1 | 0.015 | NS | |
| Single (n=68) | 63.9 | |||
| Diameter of largest of CLM | ||||
| >5 cm (n=24) | 35.1 | 0.127 | ||
| ≥5 cm (n=169) | 51.2 | |||
| Major complication | ||||
| Yes (n=32) | 37.1 | 0.066 | ||
| No (n=161) | 51.7 | |||
| Surgical margin | ||||
| Positive (n=19) | 38.2 | 0.402 | ||
| Negative (n=174) | 49.1 | |||
CLM indicates colorectal liver metastases; NS, not significant; RFS, recurrence-free survival; %VTC, percentage viable tumor cells.
From diagnosis of primary tumor to diagnosis of CLM.
DISCUSSION
In this study, we analyzed the prognostic impact of RAS mutation status in 193 patients who underwent curative resection of CLM after single-regimen modern systemic therapy. Consistent with previous oncogene profiling studies for primary colorectal cancers, RAS, BRAF, and PIK3CA were identified as the most common point-mutated genes in CLM.[23, 26] Our analysis indicates that RAS mutation status is an independent predictor of OS, overall RFS, and lung RFS, but not liver RFS after resection of CLM. The current study also confirms our previous study indicating the pre-eminence of response to chemotherapy as a dynamic biological predictor outcome superior to traditional clinical pathological predictors such as number of liver metastases, size of liver metastases, lymph node status of primary tumor or surgical margins.[7, 27] Over the past decade, the prognostic role of somatic mutations has been actively investigated, especially in primary colorectal cancer. The most common somatic gene mutations reported in previous studies were KRAS, NRAS, PIK3CA, and BRAF, with mutation rates around 40%, 2.6%, 14.5%, and 4.7%, respectively.[13] The previous studies of molecular alterations in colorectal cancer included patients with advanced and unresectable metastatic disease. The current study indicates that KRAS mutation rates are lower in patients selected to undergo curative resection of CLM than in medical series reporting on KRAS mutation in patients with advanced metastatic colorectal cancer not amenable to curative resection. This suggests that among patients affected by metastatic colorectal cancer, those amenable to curative resection of CLM represent a preselected population with better tumor biology and longer survival. This hypothesis is supported by the fact that the only study reporting on rates of KRAS mutations indicated a 25% rate of mutations in patients undergoing resection after receiving preoperative modern chemotherapy.[28]
In this study, we analyzed the mutation status of multiple somatic genes and found that of 43 patients with at least one somatic gene mutation, the majority had a RAS mutation (n=34), while only a small minority had mutations of PIK3CA (n=5), BRAF (n=2), CTTNB1 (n=1), and AKT1 (n=1). This finding confirms the dominance of RAS mutations in colorectal cancer and suggests that the less common mutations (of PIK3CA, BRAF, CTTNB1, and AKT1) are unlikely to contribute significantly to future overall outcome analyses of patients undergoing resection of CLM. However, it is still possible that, in future studies, these rare mutations could help to predict outcomes of specific subsets of patients with poor prognosis or specific metastatic patterns.
Only a few prior studies[16–18] have focused on mutation rates and recurrence patterns in patients with colorectal cancer. Tie et al[16] reported higher rates of RAS mutations in colorectal lung metastases than in CLM. In the current study, we specifically looked at patterns of recurrence following hepatectomy. The results indicate that, compared to wild-type RAS, RAS mutation is associated with a shorter 3-year lung RFS rate (34.6% vs 59.3%, P<0.001), but not with a shorter liver RFS rate (43.8% vs 50.2%, P=0.181) (Figure 4). These results suggest a propensity for RAS-mutated tumors to metastasize to lungs and are in line with the results of previous studies that have shown higher KRAS mutation rates in lung (62%) and brain (57%) metastases from colorectal cancer than in CLM (32%).[16, 17] This discordance in mutation rates suggests the possibility that KRAS-mutant tumors are biologically versatile and able to grow in different visceral organs and have higher capacity for systemic vascular (as opposed to portal vascular) tumor spread. These findings argue in favor of studying molecular heterogeneity and differences in biologic interaction between colon cancer tumor cells and host organ factors at different metastatic sites, in in vitro and in vivo models, comparing KRAS-mutant and KRAS-wild-type colorectal cancer.
In a previous study, indeterminate lung nodules were reported in 43% of patients undergoing chest computed tomography before resection of CLM. Only 35% of indeterminate nodules proved to be lung metastases, and their presence was not associated with worse survival, leading the authors to conclude that the presence of indeterminate lung nodules should not preclude resection of CLM.[29] The findings from our current study, while supporting the role of preoperative chest computed tomography before resection of CLM, suggest that nonspecific lung nodules in patients with RAS mutations may be more likely to represent metastatic disease. This information may help physicians select, among patients with multiple and bilobar CLM requiring extensive resection, those who may benefit from an aggressive surgical approach. In addition, our data favor the use of chest computed tomography surveillance after resection of CLM in patients harboring a RAS mutation.
The RAS mutation rate in the current study (18%) is lower than the 47% RAS mutation rate reported in our previous study in patients undergoing resection of metachronous CLM.[15] We interpret the findings of our previous study as indicating that adjuvant chemotherapy for 6 months after resection of the primary tumor resulted in a selection pressure favoring the onset of metachronous liver metastases enriched for KRAS mutations and prevented metastases in a number of patients with primary tumors with wild-type KRAS - similar to the phenomenon by which antibiotic therapy can select for treatment-resistant bacteria. These interpretations are consistent with findings from other studies indicating that adjuvant FOLFOX for the primary tumor does not cause mutations in CLM.[30, 31] and that the mutation types remain concordant between the primary tumor and CLM in more than 90% of patients when the primary and the metastases are compared.[32–36]
In contrast with our previous study, our current work focused on patients who underwent single-regimen preoperative modern chemotherapy for 2–3 months and excluded patients who had received multiple lines of chemotherapy. Therefore, the current study focused on a clinically preselected population of patients with favorable tumor biology accounting for the low mutation rate. In addition, the analysis of clinical and pathological differences between patients with mutant RAS and those with wild-type RAS indicated similar median numbers of chemotherapy cycles in the two groups (Table 1). This data supports the concept that preoperative chemotherapy for liver metastases does not affect the RAS mutation rate.[37] Taken together, these mutational data do not argue against the use of modern chemotherapy as adjuvant therapy for primary tumors or neoadjuvant therapy for liver metastases.
The limitations of this study include its retrospective nature and the selected patient population due to limited availability of specimens suitable for genetic analysis. However, the present analysis was based on a patient population with similar pathologic and clinical characteristics, and RAS mutation status well stratified patients with respect to prognosis and patterns of recurrence, even after exclusion of patients with very good pathologic response to chemotherapy (% of residual tumor cells < 5%), in whom genetic profiling was impossible due to very low proportion of residual viable tumor cells in the specimen and consequent insufficient DNA for genetic analysis.
In conclusion, RAS mutation status is a powerful predictor of OS, RFS, and lung recurrence after curative resection of CLM. These data indicate that the genetic profile of CLM can be used to improve selection of patients with CLM for surgery and predict outcome of patients with CLM. In addition, the finding of a higher rate of pathologic response in patients with wild-type RAS sets the stage for further studies focusing on somatic gene mutations and pattern of response associated with preoperative chemotherapy.
FIGURE 2.
Somatic gene mutations.
ACKNOWLEDGEMENTS
The authors particularly thank Ruth J. Haynes and Stephanie Deming for editing.
Source of Funding: This research was supported in part by the National Institutes of Health through MD Anderson’s Cancer Center Support Grant, CA016672.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of Interest The authors report no conflicts of interest relevant to this article.
REFERENCES
- 1.Kopetz S, Chang GJ, Overman MJ, et al. Improved survival in metastatic colorectal cancer is associated with adoption of hepatic resection and improved chemotherapy. J Clin Oncol. 2009;27:3677–3683. doi: 10.1200/JCO.2008.20.5278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg. 1999;230:309–318. doi: 10.1097/00000658-199909000-00004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nordlinger B, Guiguet M, Vaillant JC, et al. Surgical resection of colorectal carcinoma metastases to the liver. A prognostic scoring system to improve case selection, based on 1568 patients. Association Francaise de Chirurgie. Cancer. 1996;77:1254–1262. [PubMed] [Google Scholar]
- 4.Rees M, Tekkis PP, Welsh FK, et al. Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer: A multifactorial model of 929 patients. Ann Surg. 2008;247:125–135. doi: 10.1097/SLA.0b013e31815aa2c2. [DOI] [PubMed] [Google Scholar]
- 5.George B, Kopetz S. Predictive and prognostic markers in colorectal cancer. Curr Oncol Rep. 2011;13:206–215. doi: 10.1007/s11912-011-0162-3. [DOI] [PubMed] [Google Scholar]
- 6.Kopetz S, Vauthey JN. Perioperative chemotherapy for resectable hepatic metastases. Lancet. 2008;371:963–965. doi: 10.1016/S0140-6736(08)60429-8. [DOI] [PubMed] [Google Scholar]
- 7.Blazer DG, 3rd, Kishi Y, Maru DM, et al. Pathologic response to preoperative chemotherapy: a new outcome end point after resection of hepatic colorectal metastases. J Clin Oncol. 2008;26:5344–5351. doi: 10.1200/JCO.2008.17.5299. [DOI] [PubMed] [Google Scholar]
- 8.Maru DM, Kopetz S, Boonsirikamchai P, et al. Tumor thickness at the tumor-normal interface: a novel pathologic indicator of chemotherapy response in hepatic colorectal metastases. Am J Surg Pathol. 2010;34:1287–1294. doi: 10.1097/PAS.0b013e3181eb2f7b. [DOI] [PubMed] [Google Scholar]
- 9.Brouquet A, Abdalla EK, Kopetz S, et al. High survival rate after two-stage resection of advanced colorectal liver metastases: response-based selection and complete resection define outcome. J Clin Oncol. 2011;29:1083–1090. doi: 10.1200/JCO.2010.32.6132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Chun YS, Vauthey JN, Boonsirikamchai P, et al. Association of computed tomography morphologic criteria with pathologic response and survival in patients treated with bevacizumab for colorectal liver metastases. JAMA. 2009;302:2338–2344. doi: 10.1001/jama.2009.1755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Shindoh J, Loyer EM, Kopetz S, et al. Optimal morphologic response to preoperative chemotherapy: an alternate outcome end point before resection of hepatic colorectal metastases. J Clin Oncol. 2012;30:4566–4572. doi: 10.1200/JCO.2012.45.2854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Van Cutsem E, Kohne CH, Hitre E, et al. Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer. N Engl J Med. 2009;360:1408–1417. doi: 10.1056/NEJMoa0805019. [DOI] [PubMed] [Google Scholar]
- 13.De Roock W, Claes B, Bernasconi D, et al. Effects of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in chemotherapy-refractory metastatic colorectal cancer: a retrospective consortium analysis. Lancet Oncol. 2010;11:753–762. doi: 10.1016/S1470-2045(10)70130-3. [DOI] [PubMed] [Google Scholar]
- 14.Nash GM, Gimbel M, Shia J, et al. KRAS mutation correlates with accelerated metastatic progression in patients with colorectal liver metastases. Ann Surg Oncol. 2010;17:572–578. doi: 10.1245/s10434-009-0605-3. [DOI] [PubMed] [Google Scholar]
- 15.Andreou A, Kopetz S, Maru DM, et al. Adjuvant chemotherapy with FOLFOX for primary colorectal cancer is associated with increased somatic gene mutations and inferior survival in patients undergoing hepatectomy for metachronous liver metastases. Ann Surg. 2012;256:642–650. doi: 10.1097/SLA.0b013e31826b4dcc. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tie J, Lipton L, Desai J, et al. KRAS mutation is associated with lung metastasis in patients with curatively resected colorectal cancer. Clin Cancer Res. 2011;17:1122–1130. doi: 10.1158/1078-0432.CCR-10-1720. [DOI] [PubMed] [Google Scholar]
- 17.Kim MJ, Lee HS, Kim JH, et al. Different metastatic pattern according to the KRAS mutational status and site-specific discordance of KRAS status in patients with colorectal cancer. BMC Cancer. 2012;12:347. doi: 10.1186/1471-2407-12-347. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cejas P, Lopez-Gomez M, Aguayo C, et al. KRAS mutations in primary colorectal cancer tumors and related metastases: a potential role in prediction of lung metastasis. PLoS One. 2009;4:e8199. doi: 10.1371/journal.pone.0008199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tran B, Kopetz S, Tie J, et al. Impact of BRAF mutation and microsatellite instability on the pattern of metastatic spread and prognosis in metastatic colorectal cancer. Cancer. 2011;117:4623–4632. doi: 10.1002/cncr.26086. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Aloia TA, Zorzi D, Abdalla EK, et al. Two-surgeon technique for hepatic parenchymal transection of the noncirrhotic liver using saline-linked cautery and ultrasonic dissection. Ann Surg. 2005;242:172–177. doi: 10.1097/01.sla.0000171300.62318.f4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pearce M, Ehrich M. In: Somatic Mutation Analysis in Tumor Samples using the Sequenom. MassARRAY® System, editor. San Diego: Sequenom; 2011. [Google Scholar]
- 22.Gabriel S, Ziaugra L, Tabbaa D. SNP genotyping using the Sequenom MassARRAY iPLEX platform. Curr Protoc Hum Genet. 2009;Chapter 2(Unit 2):12. doi: 10.1002/0471142905.hg0212s60. [DOI] [PubMed] [Google Scholar]
- 23.Thomas RK, Baker AC, Debiasi RM, et al. High-throughput oncogene mutation profiling in human cancer. Nat Genet. 2007;39:347–351. doi: 10.1038/ng1975. [DOI] [PubMed] [Google Scholar]
- 24.Pearce M, Ehrich M. Somatic Mutation Analysis in Tumor Samples using the Sequenom. MassARRAYR® System. Nat. Methods. 2010 http://www.nature.com/appnotes/nmeth/2010/101304/full/an7545.html. [Google Scholar]
- 25.Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240:205–213. doi: 10.1097/01.sla.0000133083.54934.ae. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.MacConaill LE, Campbell CD, Kehoe SM, et al. Profiling critical cancer gene mutations in clinical tumor samples. PLoS One. 2009;4:e7887. doi: 10.1371/journal.pone.0007887. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Andreou A, Aloia TA, Brouquet A, et al. Margin status remains an important determinant of survival after surgical resection of colorectal liver metastases in the era of modern chemotherapy. Ann Surg. doi: 10.1097/SLA.0b013e318283a4d1. Epub: 2013 Feb 19; [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Stremitzer S, Stift J, Gruenberger B, et al. KRAS status and outcome of liver resection after neoadjuvant chemotherapy including bevacizumab. Br J Surg. 2012;99:1575–1582. doi: 10.1002/bjs.8909. [DOI] [PubMed] [Google Scholar]
- 29.Maithel SK, Ginsberg MS, D'Amico F, et al. Natural history of patients with subcentimeter pulmonary nodules undergoing hepatic resection for metastatic colorectal cancer. J Am Coll Surg. 2010;210:31–38. doi: 10.1016/j.jamcollsurg.2009.09.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Kawamoto Y, Tsuchihara K, Yoshino T, et al. KRAS mutations in primary tumours and post-FOLFOX metastatic lesions in cases of colorectal cancer. Br J Cancer. 2012;107:340–344. doi: 10.1038/bjc.2012.218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Vauthey JN, Kopetz S, Aloia TA, et al. KRAS mutation in colorectal cancer metastases after adjuvant FOLFOX for the primary. Br J Cancer. 2012;107:1442–1443. doi: 10.1038/bjc.2012.419. author reply 1444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Oudejans JJ, Slebos RJ, Zoetmulder FA, et al. Differential activation of ras genes by point mutation in human colon cancer with metastases to either lung or liver. Int J Cancer. 1991;49:875–879. doi: 10.1002/ijc.2910490613. [DOI] [PubMed] [Google Scholar]
- 33.Etienne-Grimaldi MC, Formento JL, Francoual M, et al. K-Ras mutations and treatment outcome in colorectal cancer patients receiving exclusive fluoropyrimidine therapy. Clin Cancer Res. 2008;14:4830–4835. doi: 10.1158/1078-0432.CCR-07-4906. [DOI] [PubMed] [Google Scholar]
- 34.Santini D, Loupakis F, Vincenzi B, et al. High concordance of KRAS status between primary colorectal tumors and related metastatic sites: implications for clinical practice. Oncologist. 2008;13:1270–1275. doi: 10.1634/theoncologist.2008-0181. [DOI] [PubMed] [Google Scholar]
- 35.Italiano A, Hostein I, Soubeyran I, et al. KRAS and BRAF mutational status in primary colorectal tumors and related metastatic sites: biological and clinical implications. Ann Surg Oncol. 2010;17:1429–1434. doi: 10.1245/s10434-009-0864-z. [DOI] [PubMed] [Google Scholar]
- 36.Knijn N, Mekenkamp LJ, Klomp M, et al. KRAS mutation analysis: a comparison between primary tumours and matched liver metastases in 305 colorectal cancer patients. Br J Cancer. 2011;104:1020–1026. doi: 10.1038/bjc.2011.26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Vakiani E, Janakiraman M, Shen R, et al. Comparative genomic analysis of primary versus metastatic colorectal carcinomas. J Clin Oncol. 2012;30:2956–2962. doi: 10.1200/JCO.2011.38.2994. [DOI] [PMC free article] [PubMed] [Google Scholar]