18F-FDG PET/CT imaging in rectal cancer: relationship with theRAS mutational status
Pierre Lovinfosse,MD
Benjamin Koopmansch,PhD
Frederic Lambert,MD
Sébastien Jodogne,PhD
Gaelle Kustermans,PhD
Mathieu Hatt,PhD
Dimitris Visvikis,PhD
Laurence Seidel,MSc
Marc Polus,MD
Adelin Albert,PhD
Philippe Delvenne,MD, PhD
Roland Hustinx,MD, PhD
Address correspondence to: Dr Pierre Lovinfosse. E-mail:pierre.lovinfosse@chu.ulg.ac.be
Corresponding author.
Received 2016 Mar 7; Revised 2016 Apr 25; Accepted 2016 May 4; Issue date 2016 Jul.
Abstract
Objective:
Treating metastatic colorectal cancer with anti-EGFR monoclonal antibodies is recommended only for patients whose tumour does not harbour mutations ofKRAS orNRAS. The aim of this study was to investigate the biology of rectal cancers and specifically to evaluate the relationship between fluorine-18 fludeoxyglucose (18F-FDG) positron emission tomography (PET) intensity and heterogeneity parameters and their mutational status.
Methods:
151 patients with newly diagnosed rectal cancer were included in this retrospective study. All patients underwent a baseline18F-FDG PET/CT within a median time interval of 27 days of tumour tissue sampling, which was performed before any treatment. Standardized uptake values (SUVs), volume-based parameters and texture analysis were studied. We retrospectively performedKRAS genotyping on codons 12, 13, 61, 117 and 146,NRAS genotyping on codons 12, 13 and 61 andBRAF on codon 600. Associations between PET/CT parameters and the mutational status were assessed using univariate and multivariate analysis.
Results:
83 (55%) patients had anRAS mutation: 74KRAS and 9NRAS, while 68 patients had no mutation (wild-type tumours). No patient hadBRAF mutation. First-order features based on intensity histogram analysis were significantly associated withRAS mutations: maximum SUV (SUVmax) (p-value = 0.002), mean SUV (p-value = 0.006), skewness (p-value = 0.049), SUV standard deviation (p-value = 0.001) and SUV coefficient of variation (SUVcov) (p-value = 0.001). Both SUVcov and SUVmax showed an area under the curve of 0.65 with sensitivity of 56% and 69%, respectively, and specificity of 64% and 52%, respectively. None of the volume-based (metabolic tumour volume and total lesion glycolysis), nor local or regional textural features were associated with the presence ofRAS mutations.
Conclusion:
Although rectal cancers withKRAS orNRAS mutations display a significantly higher glucose metabolism than wild-type cancers, the accuracy of the currently proposed quantitative metrics extracted from18F-FDG PET/CT is not sufficiently high for playing a meaningful clinical role.
Advances in knowledge:
RAS-mutated rectal cancers have a significantly higher glucose metabolism. However, the accuracy of18F-FDG PET/CT quantitative metrics is not as such as the technique could play a clinical role.
INTRODUCTION
Colorectal cancer (CRC) is the third most common cause of cancers worldwide in both males and females.1 In terms of mortality, it ranks third in the USA and fourth in the world.1,2 CRC usually develops from adenomatous polyps and results from a succession of genetic alterations, leading to inactivation of tumour suppressor genes and DNA repair genes and activation of oncogenes.3,4 TheRAS–RAF–MAPK pathway is one of the main intracellular signalling roads that activate the growth and division cell mechanisms, a cascade triggered by the binding of the epidermal growth factor on its membrane epidermal growth factor receptor (EGFR).5 Downstream in the cascade,RAS proteins are essential for intracellular signal transmission. Mutations of the genes that encodeKRAS andNRAS, described in about 30–40% and 10% of CRC, respectively, are responsible for the constitutive activation of this pathway independently of EGFR binding.6–12 Thus, targeted treatments with anti-EGFR monoclonal antibodies (cetuximab and panitumumab) are recommended only for metastatic patients whose tumours do not harbour mutations ofKRAS orNRAS, also called “wild-type” tumours (WT).6,13–15 Furthermore,RAS andRAF mutations are known to be mutually exclusive and seem to have the same tumorigenic effects.16
Several studies have explored the association between the tumoral metabolism evaluated by fluorine-18 fludeoxyglucose (18F-FDG) positron emission tomography (PET) and the presence of mutations in various types of cancers, including thyroid cancer,17,18 lung cancer19–24 and CRC,25–27 but with conflicting results. The aim of the present study was to clarify in a large cohort the overall concordance rate between the most common CRC mutations influencing the therapeutic strategy and the glucose metabolism and thus investigate whether18F-FDG PET/CT could play a role in the prediction of the mutational status of rectal cancers in clinical routine.
METHODS AND MATERIALS
All procedures were performed in accordance with the principles of the 1964 Declaration of Helsinski and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study to use rectal cancer biopsies.
This retrospective study was approved by the ethics committee of our institution under protocol B707201420492. 151 patients with newly diagnosed rectal or rectosigmoid junction cancers diagnosed between September 2009 and January 2014 were included in the study. The median age was 66 years (range: 32–92 years), and there were 95 males and 56 females. According to the seventh edition of the American Joint Committee on Cancer staging classification, there were 7 (4.6%) patients with Stage I, 16 (10.6%) patients with Stage II, 99 (65.6%) patients with Stage III and 29 (19.2%) patients with Stage IV disease.
Mutational status analysis
All sample specimens analyzed were retrieved from the biobank of our institution, where they were kept for scientific research after patients provided their informed consent. These samples were obtained prior to any treatment from biopsies in 144 patients and from surgical resections in 7 patients. Each formalin-fixed, paraffin-embedded (FFPE) tumour tissue section was assessed by a fully trained pathologist, who confirmed the diagnosis and the histological type of rectal cancer and ensured that the tissue sections used for DNA extraction were significantly infiltrated by the tumour, according to the threshold of detection of our genotyping test. DNA was extracted from FFPE tumour tissue sections using a Maxwell® 16 FFPE Plus LEV DNA Purification Kit (Promega, Madison, WI). PCR and pyrosequencing were performed forKRAS (codons 12, 13, 61, 117 and 146),NRAS (codons 12, 13 and 61) andBRAF (codon 600). ForKRAS codons 12 and 13 andBRAF codon 600, previously described methods were used.28,29 All other “positions” (codons) were screened following a home-made genotyping procedure, described in thesupplementary material. For each patient, positions were tested in a stepwise manner according to their prevalence until a first mutation was found. We then stopped the screening.
Fluorine-18 fludeoxyglucose positron emission tomography/CT acquisition
All patients underwent18F-FDG PET/CT before any treatment for tumoral staging or target volume delineation, within a median time interval of 27 days (range: 1–87 days) of the sample.18F-FDG PET/CT was acquired at our institution on two cross-calibrated three-dimensional combined PET/CT scanners (Philips Medical Systems, Cleveland, OH), 70 ± 15 min after the injection of 282 ± 54 Mbq of18F-FDG. Patients fasted for at least 6 h before the injection, and median glycaemia was 101 mg dl−1. A low-dose CT (5-mm slice thickness; tube voltage: 120 kV; and tube current–time product: 50–80 mAs depending on the patient weight) was performed without the injection of the i.v. contrast agent, followed by the PET emission scan of 90 s per bed position performed from the upper thigh to the base of the skull. Images were reconstructed on a 144 × 144 matrix, leading to voxels of 4 × 4 × 4 mm3, using an iterative list mode time-of-flight algorithm and corrections for attenuation, dead time, random and scatter events were applied. There was no change of scan protocol during the period of patient inclusion.
Fluorine-18 fludeoxyglucose positron emission tomography/CT analysis
Only the primary lesion was included in the analysis, which was performed by an experienced nuclear physician unaware of the mutational status and the patient outcome. A volume of interest (VOI) was drawn manually around the tumour and special care was given to avoid the inclusion of other structures, particularly the bladder and adjacent lymph nodes. An automatic segmentation was performed inside these VOIs using a fuzzy locally adaptive Bayesian algorithm.30–32 Metabolic tumour volume (MTV), maximum standardized uptake value (SUVmax) and mean SUV (SUVmean) were measured inside these segmented VOIs, and the total lesion glycolysis (TLG) was calculated according to the formula: TLG = SUVmean × MTV.33
Several textural features were studied to provide a measure of tumour heterogeneity, using a home-made software implemented with the cross-platform Python language. The algorithms used to measure these parameters are given in thesupplementary data. We used a quantization into 64 grey levels for all the analyses, as previously recommended.34 We calculated several first-order features based on histogram analysis, which identifies the intensity distribution on the original image: SUVmin, SUVmax, mean SUVmean, SUV standard deviation (SUVsd), SUV coefficient of variation (SUVcov), skewness and excess kurtosis. Local features were measured using grey-level co-occurrence or with neighbourhood intensity difference matrices. Grey-level co-occurrence quantifies the relationships between pairs of voxels, providing several second-order textural features. We considered angular second moment, contrast, entropy, correlation, homogeneity and dissimilarity.35–37 Neighbourhood intensity difference matrices quantifies the differences between each voxel and its 27 neighbours, providing several other second-order textural features: coarseness, contrast acquired with neighbourhood intensity difference matrices (contrastNGTDM) and busyness.35,38 Finally, we included two regional textural features calculated in the intensity–size–zone matrix that study characteristics of homogeneous zones: intensity variability and size–zone variability.35
Statistical analysis
Results were expressed as mean and standard deviation. The Spearman correlation coefficient was used to measure the relation between two parameters. For each parameter, we compared the three groups (KRAS mutated,NRAS mutated and WT) by the non-parametric Kruskal–Wallis test. The same test was used to look for associations between18F-FDG PET/CT parameters and the presence of mutations (RAS-mutated tumoursvs WT). We then performed the same analysis focusing on a subgroup of tumours with an MTV >10 cm3. We then carried out a stepwise multivariate logistic regression analysis including only parameters with univariatep-values ≤ 0.2. For significant parameters associated with the presence of any mutations, we determined the optimal cut-off maximizing the couple sensibility/specificity using receiver-operating characteristic analysis. Finally, we searched whether the mutation rate was different between rectal and rectosigmoid locations, using Fisher's exact test.
Results were considered statistically significant at the 5% critical level (p-value < 0.05). All statistical analyses were carried out using the SAS® statistical package v. 9.4 (SAS Institute Inc., NC).
RESULTS
151 patients with 151 rectal cancers were included in the study. We were able to detect a mutation in 83 (55%) patients (KRAS:n = 74;NRAS:n = 9; andBRAF:n = 0), while 68 patients were considered as wild type.KRAS mutations were found on codons 12 (n = 51), 13 (n = 13), 61 (n = 5), 117 (n = 1) and 146 (n = 4) andNRAS mutations were found on codons 12 (n = 2), 13 (n = 4) and 61 (n = 3). There was no difference of mutation rate between rectal (65 WT/75RAS-mutated) and rectosigmoid (4 WT/8RAS-mutated) locations (p-value = 0.55). We did not detect any patient with aBRAF mutation in codon 600. Clinical features are summarized inTable 1 and all the18F-FDG PET/CT parameters results are shown inTable 2.
Table 1.
Study patient characteristics
| Characteristics | Entire cohortN = 151 | MutatedN = 83 (55%) | Wild typeN = 68 (45%) |
|---|---|---|---|
| Age (years) | 65.8 ± 11.7 | 66.0 ± 11.9 | 65.4 ± 11.7 |
| Gender M/F | 95/56 (62.9/37.1%) | 50/33 (60.2/39.8%) | 45/23 (66.2/33.8%) |
| T stage | |||
| T1 | 5 (3.3%) | 0 (0.0%) | 5 (7.4%) |
| T2 | 25 (16.6%) | 16 (19.3%) | 9 (13.2%) |
| T3 | 111 (73.5%) | 62 (74.7%) | 49 (72.1%) |
| T4 | 10 (6.6%) | 5 (6.0%) | 5 (7.4%) |
| N stage | |||
| N0 | 23 (15.2%) | 12 (14.5%) | 11 (16.2%) |
| N1 | 85 (56.3%) | 49 (59%) | 36 (52.9%) |
| N2 | 43 (28.5%) | 22 (26.5%) | 21 (30.9%) |
| M stage | |||
| M0 | 122 (80.8%) | 66 (79.5%) | 56 (82.4%) |
| M1 | 29 (19.2%) | 17 (20.5%) | 12 (17.6%) |
| Stage | |||
| I | 7 (4.6%) | 4 (4.8%) | 3 (4.4%) |
| II | 16 (10.6%) | 8 (9.6%) | 8 (11.8%) |
| III | 99 (65.6%) | 54 (65.1%) | 45 (66.2%) |
| IV | 29 (19.2%) | 17 (20.5%) | 12 (17.6%) |
F, female; M, male.
There was no significant difference between mutated and wild-type groups for any of these parameters.
Table 2.
Fluorine-18 fludeoxyglucose positron emission tomography intensity, volume-based and textural analysis feature results (n = 151)
| Parameters | Mean | SD | Min | Max |
|---|---|---|---|---|
| SUVmin | 5.4 | 2.5 | 1.3 | 19.9 |
| SUVmax | 16.0 | 8.2 | 3.8 | 51.1 |
| SUVmean | 8.8 | 4.3 | 2.7 | 33.8 |
| Skewness | 0.63 | 0.28 | 0.04 | 1.72 |
| SUVsd | 2.3 | 1.3 | 0.5 | 8.3 |
| Excess kurtosis | −0.14 | 0.82 | −1.24 | 4.40 |
| SUVcov | 0.26 | 0.03 | 0.18 | 0.33 |
| MTV (cm3) | 29.9 | 25.3 | 2.9 | 139.1 |
| TLG | 262.9 | 244.6 | 10.5 | 1644.8 |
| ASM | 0.0036 | 0.0031 | 0.0008 | 0.0223 |
| Contrast | 289.1 | 111.3 | 80.4 | 735.3 |
| Entropy | 5.95 | 0.66 | 3.83 | 7.38 |
| Correlation | 0.33 | 0.12 | 0.02 | 0.60 |
| Homogeneity | 0.15 | 0.03 | 0.10 | 0.24 |
| Dissimilarity | 13.2 | 2.8 | 6.7 | 22.0 |
| Coarseness | 5.08 | 1.40 | 1.55 | 9.49 |
| ContrastNGTDM | 1.30 | 1.02 | 0.22 | 9.04 |
| Busyness | 0.11 | 0.08 | 0.03 | 0.45 |
| Intensity variability | 11.3 | 10.2 | 2.0 | 61.3 |
| Size–zone variability | 0.0046 | 0.0037 | 0.0004 | 0.0238 |
ASM, angular second moment; cm3, cubic centimetres; contrastNGTDM, contrast acquired with neighbourhood intensity difference matrices; Max, maximum; Min, minimum; MTV, metabolic tumour volume; SD, standard deviation; SUVcov, standardized uptake value coefficient of variation; SUVmax, maximum SUV; SUVmean, mean SUV; SUVmin, minimum SUV; SUVsd, SUV standard deviation; TLG, total lesion glycolysis.
The SUVmin (p-value = 0.005), SUVmax (p-value = 0.002), SUVmean (p-value = 0.006), skewness (p-value = 0.049), SUVsd (p-value = 0.001) and SUVcov (p-value = 0.001) were significantly higher in rectal tumours with RAS mutation than in WT (Figure 1). The volume-based parameters and the other textural analysis features were not associated with the presence of any kind of mutation. No significant difference was found between the two populations in terms of age (p-value = 0.76), gender (p-value = 0.50), T stage (p-value = 0.07), N stage (p-value = 0.75), metastatic disease (p-value = 0.68) or tumoral stage (p-value = 0.96). These results are presented inTable 3. The multivariate analysis showed that only SUVcov was significantly associated with the presence of mutations (p-value < 0.001 for tumoral stage).
Figure 1.
Means and standard deviation of minimum standardized uptake value (SUVmin), maximum SUV (SUVmax), mean SUV (SUVmean), skewness, SUV standard deviation (SUVsd) and coefficient of variation (SUVcov) forRAS-mutated tumours,KRAS-mutated tumours,NRAS-mutated tumours and wild-type tumours (WT) (n = 151). * indicates the significant difference between groups and WT.
Table 3.
Parameter distributions according toRAS-mutated status (n = 151)
| Parameters | Wild type (n = 68) | RAS mutated (n = 83) | KRAS mutated (n = 74) | NRAS mutated (n = 9) |
|---|---|---|---|---|
| Age | 65.4 ± 11.7 | 66.0 ±11.9 | 66.3 ± 11.9 | 63.9 ± 13.0 |
| Gender (M/F) | 66%/34% | 60%/40% | 61%/39% | 56%/44% |
| Stage I/II/III/IV | 4%/12%/66%/18% | 5%/10%/65%/20% | 5%/11%/62%/22% | 0%/0%/89%/11% |
| SUVmin | 4.81 ± 2.05 | 5.94 ±2.75a | 5.97 ± 2.79a | 5.73 ± 2.53a |
| SUVmax | 13.74 ± 6.77 | 17.80 ± 8.83a | 17.93 ± 8.98a | 16.73 ± 7.93a |
| SUVmean | 7.73 ± 3.43 | 9.61 ± 4.70a | 9.66 ± 4.77a | 9.22 ± 4.40a |
| Skewness | 0.57 ± 0.24 | 0.68 ± 0.31a | 0.68 ± 0.30a | 0.68 ± 0.36a |
| SUVsd | 1.93 ± 1.00 | 2.58 ± 1.38a | 2.60 ± 1.39a | 2.45 ± 1.40a |
| Excess kurtosis | −0.27 ± 0.63 | −0.03 ± 0.93 | −0.05 ± 0.91 | 0.07 ± 1.21 |
| SUVcov | 0.25 ± 0.03 | 0.26 ± 0.03a | 0.26 ± 0.03a | 0.26 ± 0.04a |
| MTV (cm3) | 31.7 ± 28.3 | 28.4 ± 22.7 | 29.5 ± 23.6 | 19.3 ± 7.8 |
| TLG | 254.5 ± 265.1 | 269.9 ± 227.9 | 282.6 ± 237.3 | 165.6 ± 70.0 |
| ASM | 0.0035 ± 0.0026 | 0.0036 ± 0.0034 | 0.0036 ± 0.0035 | 0.0040 ± 0.0026 |
| Contrast | 282.2 ± 99.8 | 294.7 ± 120.2 | 293.9 ± 119.6 | 301.3 ± 132.8 |
| Entropy | 5.96 ± 0.67 | 5.93 ± 0.66 | 5.96 ± 0.67 | 5.73 ± 0.53 |
| Correlation | 0.33 ± 0.12 | 0.32 ± 0.13 | 0.32 ± 0.13 | 0.31 ± 0.12 |
| Homogeneity | 0.153 ± 0.024 | 0.153 ± 0.027 | 0.153 ± 0.027 | 0.153 ± 0.031 |
| Dissimilarity | 13.11 ± 2.55 | 13.32 ± 2.95 | 13.29 ± 2.93 | 13.51 ± 3.29 |
| Coarseness | 5.28 ± 1.53 | 4.91 ± 1.27 | 4.92 ± 1.27 | 4.83 ± 1.39 |
| ContrastNGTDM | 1.22 ± 0.70 | 1.37 ± 1.23 | 1.36 ± 1.24 | 1.48 ± 1.14 |
| Busyness | 0.103 ± 0.071 | 0.122 ± 0.081 | 0.13 ± 0.08 | 0.08 ± 0.04 |
| Intensity variability | 11.26 ± 10.51 | 11.29 ± 9.92 | 11.76 ± 10.36 | 7.43 ± 3.36 |
| Size–zone variability | 0.0046 ± 0.0034 | 0.0046 ± 0.0040 | 0.0045 ± 0.0041 | 0.0051 ± 0.0033 |
ASM, angular second moment; cm3, cubic centimetres; contrastNGTDM, contrast acquired with neighbourhood intensity difference matrices; F, female; M, male; MTV, metabolic tumour volume; SUVcov, standardized uptake value of coefficient of variation; SUVmax, maximum SUV; SUVmean, mean SUV; SUVmin, minimum SUV; SUVsd, SUV standard deviation; TLG, total lesion glycolysis.
p-value < 0.05 in comparison with wild-type tumours.
ConsideringKRAS andNRAS mutations, there was no difference in terms of age (p-value = 0.59), gender (p-value = 0.99), T stage (p-value = 0.19), N stage (p-value = 0.23), metastatic disease (p-value = 0.99) and tumoral TNM stage (p-value = 0.82). Intensity, volume-based and textural features were also similar between these two groups of mutated rectal cancers.
The same analyses were then performed including only the lesions with MTV >10 cm3 (n = 128). In this population, SUVmax (p-value = 0.033), skewness (p-value = 0.014), SUVsd (p-value = 0.033) and SUVcov (p-value = 0.003) remained higher inRAS-mutated than in WT tumours. Only SUVmin and SUVmean were not significantly different onRAS-mutated tumours. All the volume-based and the other textural features were similar between groups (p-value > 0.05). These results are presented inTable 4.
Table 4.
Parameter distributions according toRAS-mutated status for lesions with metabolic tumour volume (MTV) > 10 cm3 (n = 128)
| Parameters | Wild type (n = 56) | RAS mutated (n = 72) | KRAS mutated (n = 65) | NRAS mutated (n = 7) |
|---|---|---|---|---|
| Age | 64.8 ± 10.8 | 66.8 ± 11.39 | 67.4 ± 11.1 | 60.5 ± 12.9 |
| Gender (M/F) | 70%/30% | 65%/35% | 66%/34% | 57%/43% |
| Stage I/II/III/IV | 0%/9%/71%/20% | 1.5%/10%/65%/23.5% | 1.5%/11%/63%/24.5% | 0%/0%/86%/14% |
| SUVmin | 5.13 ± 2.07 | 5.95 ± 2.78 | 6.02 ± 2.85 | 5.30 ± 2.09 |
| SUVmax | 14.82 ± 6.84 | 18.07 ± 9.00a | 18.36 ± 9.19a | 15.42 ± 6.91 |
| SUVmean | 8.29 ± 3.44 | 9.62 ± 4.77 | 9.75 ± 4.88 | 8.44 ± 3.70 |
| Skewness | 0.55 ± 0.24 | 0.69 ± 0.32a | 0.69 ± 0.31a | 0.67 ± 0.41a |
| SUVsd | 2.08 ± 1.02 | 2.57 ± 1.39a | 2.62 ± 1.42a | 2.11 ± 1.10 |
| Excess Kurtosis | −0.30 ± 0.62 | 0.04 ± 0.98a | 0.02 ± 0.94a | 0.21 ± 1.36a |
| SUVcov | 0.25 ± 0.02 | 0.26 ± 0.03a | 0.26 ± 0.03a | 0.25 ± 0.03a |
| MTV (cm3) | 36.8 ± 28.6 | 31.8 ± 22.5 | 32.8 ± 23.4 | 22.6 ± 4.5 |
| TLG | 300.5 ± 270.8 | 300.7 ± 228.9 | 313.1 ± 237.0 | 185.2 ± 58.2 |
| ASM | 0.0025 ± 0.0011 | 0.0026 ± 0.0010 | 0.0025 ± 0.0010 | 0.0028 ± 0.0005 |
| Contrast | 257.9 ± 80.9 | 260.5 ± 81.5 | 262.7 ± 83.6 | 239.8 ± 58.9 |
| Entropy | 6.18 ± 0.48 | 6.13 ± 0.43 | 6.15 ± 0.45 | 5.98 ± 0.19 |
| Correlation | 0.36 ± 0.10 | 0.35 ± 0.11 | 0.35 ± 0.11 | 0.36 ± 0.08 |
| Homogeneity | 0.160 ± 0.024 | 0.160 ± 0.024 | 0.160 ± 0.024 | 0.160 ± 0.025 |
| Dissimilarity | 12.54 ± 2.24 | 12.53 ± 2.24 | 12.58 ± 2.28 | 12.07 ± 1.89 |
| Coarseness | 5.65 ± 1.40 | 5.23 ± 1.00 | 5.20 ± 1.03 | 5.46 ± 0.66 |
| ContrastNGTDM | 0.99 ± 0.38 | 1.01 ± 0.38 | 1.02 ± 0.39 | 0.95 ± 0.25 |
| Busyness | 0.12 ± 0.07 | 0.13 ± 0.08 | 0.14 ± 0.083 | 0.09 ± 0.037 |
| Intensity variability | 12.96 ± 10.86 | 12.57 ± 10.06 | 13.00 ± 10.46 | 8.57 ± 2.86 |
| Size–zone variability | 0.0033 ± 0.0018 | 0.0034 ± 0.0016 | 0.0033 ± 0.0016 | 0.0036 ± 0.0009 |
ASM, angular second moment; cm3, cubic centimetres; contrastNGTDM, contrast acquired with neighbourhood intensity difference matrices; F, female; M, male; SUVcov, standardized uptake value of coefficient of variation; SUVmax, maximum SUV; SUVmean, mean SUV; SUVmin, minimum SUV; SUVsd, SUV standard deviation; TLG, total lesion glycolysis.
p-value < 0.05 in comparison with wild-type tumours.
The area under the curve (AUC) was obtained for the parameters that stood out in the univariate analysis. SUVcov was the feature with the highest AUC (0.65) and presented a sensitivity and specificity of 56% and 64%, respectively, for a cut-off of 0.25. SUVmax had an AUC of 0.65 and sensibility and specificity of 69% and 52%, respectively, for a cut-off of 15.5.
Correlations among positron emission tomography parameters
Several PET parameters were correlated with each other, and the Spearman's correlation coefficients are listed in thesupplementary data.
DISCUSSION
Determination of theRAS andBRAF mutational status of CRC is essential in the management of metastatic diseases, as the presence of one of these mutations forbids treatment with anti-EGFR monoclonal antibodies. There is currently no definite answer with regard to the relationship between18F-FDG uptake and mutational status of CRC. In a population of 121 patients with colon or rectal cancers, Chen et al25 showed that SUVmax and TW40% (PET tumour width determined on a segmented volume obtained using a threshold of 40% of the SUVmax) were predictors ofKRAS mutations in exon 2 (codons 12 and 13), while MTV and TLG were not. In a subgroup of rectal cancers, however, only TW40% was significantly higher in the mutant group, not SUVmax. Kawada et al26 studied a population of 51 patients with colon and rectal cancers independently of the location, looking for the mutation ofKRAS exon 2 andBRAF. They showed that SUVmax and tumour/liver activity ratio were significantly higher in the mutants group than in the wild-type group (SUVmax 17.3 ± 7.1vs 12.1 ± 5.7). Finally, Krikelis et al27 did not find any correlation between SUVmax andKRAS exon 2 mutation in a group of 44 patients with metastatic CRC. However, SUVmax was measured in the metastasis with the highest18F-FDG uptake, whereas theKRAS status was analyzed in the primary cancer. The mutational status between the primary tumour and the metastatic lesions may be different in up to 36%39 even though a recent meta-analysis showed a high concordance rate betweenKRAS status of primary and corresponding metastasis. 9% of patients with WT primary tumour who received treatment had mutantKRAS metastasis, whereas 11.3% of the patients with mutantKRAS primary tumour had WT metastasis and did not receive treatment.40 The development of non-invasive techniques that would allow determining the mutational status of metastasis is consequently all the more a real challenge to personalized cancer treatment. Recently, Kawada et al41 showed that metastatic lesions of CRC had a higher SUVmax when they wereKRAS mutated. However, this was observed in lesions >10 mm (p-value = 0.03), and there was no difference when the series included smaller lesions. They found that an SUVmax cut-off of 6.0 conferred the best accuracy with an area under the curve of 0.70 and sensitivity and specificity of 68% and 74%, respectively.
Our analysis was not limited to the SUVmax, as in most related studies. We also evaluated volume-based parameters (MTV and TLG) and several textural features, which have been shown to be robust and reproducible,42,43 and have provided encouraging results as predictive or prognostic factors for various types of cancers.34,35,44–46 Our results showed that mutated rectal cancers have a significantly higher glucose metabolism than WT. All the intensity features of the lesions were significantly higher when a mutation was present, regardless of the tumour size. All quantitative measurements are affected by the partial volume effect according to the lesion size and the system resolution. Furthermore, considering the textural features, a significant relationship exists with the volume, but it has been suggested that above the threshold of 10 cm3, they provide additional information compared with MTV.34 We thus analyzed the larger lesions,i.e. with an MTV superior to 10 cm3, and the results were quite similar except for SUVmin and SUVmean, which became irrelevant. Our results confirm the significant correlations and the trend observed in previously studies but in a larger and more homogeneous population of 151 patients with only rectal and rectosigmoid junction cancers. In particular, SUVcov, defined as SUVsd divided by SUVmean, was the only feature significantly associated with presence of the mutations from a multivariate standpoint. The coefficient of variation is one the most basic element of tumour heterogeneity expressing the variability of18F-FDG uptake in the entire tumour based on intensity histogram, which is quite simple to measure using routine software. SUVcov seems to have an important place in rectal cancers. Interestingly enough, Bundschuh et al47 showed that SUVcov at baseline was the best parameter to predict response to neoadjuvant radiochemotherapy in locally advanced rectal cancer. A recent publication confirms the significant association between the18F-FDG PET textural analysis and the tumoral regression of locally advanced rectal cancer after neoadjuvant treatment, but without studying the SUVcov specifically.48
Our analysis was limited to PET quantitative metrics. Image-derived features extracted from other modalities (CT, dynamic contrast-enhanced CT and MRI) may have clinical value in rectal cancer.49 In addition, it was recently shown that extracting and combining features from both PET and CT components of PET/CT could provide additional value50 and such an approach may be explored in the present cohort in a future work.
Compared with previous studies that focused only on the codons 12 and 13, we investigated the presence ofKRAS mutations on codons 61, 117 and 146, as well asNRAS mutations in codons 12, 13 and 61. Such mutations have recently been described in approximately 20% ofKRAS exon-2 WT and they influence the response to targeted therapy in a similar fashion.12,15,51 This explains the high rate ofKRAS-mutated tumours in this study as compared with similar previous studies, but our results are concordant with the overall rate ofRAS mutations recently described in a pooled analysis of randomized controlled trials in CRC.52 The rate ofRAS-mutated tumours was not significantly different between rectal and rectosigmoid locations, in agreement with the results of Yamauchi et al.53 On the other hand, noBRAF mutation was observed in our population, which is not surprising considering that the mutation rate is significantly lower in rectal than in colon cancers.11,53
The absence ofRAS mutation in metastatic CRC is associated with a better prognosis in association with the possibility to efficiently treat those patients with cetuximab.6,7,9,10 On the other hand, the prognostic value of baseline18F-FDG PET parameters in rectal cancer remains debated, whether it is for predicting the pathological response to neoadjuvant treatment or the survival.47,54–57 Our population is rather large but nonetheless too heterogeneous in terms of stages and treatments to meaningfully assess survival. In any case, the trend towards higher metabolism in mutated tumours warrants further investigation in order to test whether the combination of the two parameters could improve patient prognostication in a more homogeneous group.
These results, although statistically significant, have little clinical relevance however, as we could not define an optimal cut-off value that would reliably differentiate mutated and non-mutated cancers. For the SUVmax, the optimal cut-off was 15.5, with a sensitivity of 69% and a specificity of 52%, leading to major overlaps across groups. Consequently, we do not believe that18F-FDG PET/CT could play a clinical role for assessing the mutation status or even guide issue sampling.
CONCLUSION
Although rectal cancers withKRAS orNRAS mutations display a significantly higher glucose metabolism than wild-type cancers, the accuracy of currently available quantitative metrics extracted from18F-FDG PET/CT is not as such as the technique could play a meaningful clinical role.
Contributor Information
Pierre Lovinfosse, Email: pierre.lovinfosse@chu.ulg.ac.be.
Benjamin Koopmansch, Email: bkoopmansch@chu.ulg.ac.be.
Frederic Lambert, Email: frederic.lambert@chu.ulg.ac.be.
Sébastien Jodogne, Email: s.jodogne@chu.ulg.ac.be.
Gaelle Kustermans, Email: gaelle.kustermans@chu.ulg.ac.be.
Mathieu Hatt, Email: hatt@univ-brest.fr.
Dimitris Visvikis, Email: dimitris@univ-brest.fr.
Laurence Seidel, Email: laurence.seidel@chu.ulg.ac.be.
Marc Polus, Email: m.polus@chu.ulg.ac.be.
Adelin Albert, Email: aalbert@ulg.ac.be.
Philippe Delvenne, Email: p.delvenne@chu.ulg.ac.be.
Roland Hustinx, Email: rhustinx@chu.ulg.ac.be.
REFERENCES
- 1.Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012.Int J Cancer 2015; 136: E359–86. doi: 10.1002/ijc.29210 [DOI] [PubMed] [Google Scholar]
- 2.Siegel R, Desantis C, Jemal A. Colorectal cancer statistics, 2014.CA Cancer J Clin 2014; 64: 104–17. doi: 10.3322/caac.21220 [DOI] [PubMed] [Google Scholar]
- 3.Arnold CN, Goel A, Blum HE, Boland CR. Molecular pathogenesis of colorectal cancer: implications for molecular diagnosis.Cancer 2005; 104: 2035–47. doi: 10.1002/cncr.21462 [DOI] [PubMed] [Google Scholar]
- 4.Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA, Jr, Kinzler KW. Cancer genome landscapes.Science 2013; 339: 1546–58. doi: 10.1126/science.1235122 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Baselga J. The EGFR as a target for anticancer therapy—focus on cetuximab.Eur J Cancer 2001; 37(Suppl 4): S16–22. [DOI] [PubMed] [Google Scholar]
- 6.Karapetis CS, Khambata-Ford S, Jonker DJ, O'Callaghan CJ, Tu D, Tebbutt NC, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer.N Engl J Med 2008; 359: 1757–65. doi: 10.1056/NEJMoa0804385 [DOI] [PubMed] [Google Scholar]
- 7.Lievre A, Bachet JB, Boige V, Cayre A, Le Corre D, Buc E, et al. KRAS mutations as an independent prognostic factor in patients with advanced colorectal cancer treated with cetuximab.J Clin Oncol 2008; 26: 374–9. doi: 10.1200/JCO.2007.12.5906 [DOI] [PubMed] [Google Scholar]
- 8.Di Nicolantonio F, Martini M, Molinari F, Sartore-Bianchi A, Arena S, Saletti P, et al. Wild-type BRAF is required for response to panitumumab or cetuximab in metastatic colorectal cancer.J Clin Oncol 2008; 26: 5705–12. doi: 10.1200/JCO.2008.18.0786 [DOI] [PubMed] [Google Scholar]
- 9.Van Cutsem E, Kohne CH, Hitre E, Zaluski J, Chang Chien CR, Makhson A, et al. Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer.N Engl J Med 2009; 360: 1408–17. doi: 10.1056/NEJMoa0805019 [DOI] [PubMed] [Google Scholar]
- 10.Qiu LX, Mao C, Zhang J, Zhu XD, Liao RY, Xue K, et al. Predictive and prognostic value of KRAS mutations in metastatic colorectal cancer patients treated with cetuximab: a meta-analysis of 22 studies.Eur J Cancer 2010; 46: 2781–7. doi: 10.1016/j.ejca.2010.05.022 [DOI] [PubMed] [Google Scholar]
- 11.Russo AL, Borger DR, Szymonifka J, Ryan DP, Wo JY, Blaszkowsky LS, et al. Mutational analysis and clinical correlation of metastatic colorectal cancer.Cancer 2014; 120: 1482–90. doi: 10.1002/cncr.28599 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Sorich MJ, Wiese MD, Rowland A, Kichenadasse G, McKinnon RA, Karapetis CS. Extended RAS mutations and anti-EGFR monoclonal antibody survival benefit in metastatic colorectal cancer: a meta-analysis of randomized, controlled trials.Ann Oncol 2015; 26: 13–21. doi: 10.1093/annonc/mdu378 [DOI] [PubMed] [Google Scholar]
- 13.Amado RG, Wolf M, Peeters M, Van Cutsem E, Siena S, Freeman DJ, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer.J Clin Oncol 2008; 26: 1626–34. doi: 10.1200/JCO.2007.14.7116 [DOI] [PubMed] [Google Scholar]
- 14.Douillard JY, Oliner KS, Siena S, Tabernero J, Burkes R, Barugel M, et al. Panitumumab-FOLFOX4 treatment and RAS mutations in colorectal cancer.N Engl J Med 2013; 369: 1023–34. doi: 10.1056/NEJMoa1305275 [DOI] [PubMed] [Google Scholar]
- 15.National Comprehensive Cancer Network.NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) Rectal Cancer [v. 2 (2016)]. Available from:http://www.nccn.org/professionals/physician_gls/pdf/rectal.pdf [Google Scholar]
- 16.Rajagopalan H, Bardelli A, Lengauer C, Kinzler KW, Vogelstein B, Velculescu VE. Tumorigenesis: RAF/RAS oncogenes and mismatch-repair status.Nature 2002; 418: 934. doi: 10.1038/418934a [DOI] [PubMed] [Google Scholar]
- 17.Nagarajah J, Ho AL, Tuttle RM, Weber WA, Grewal RK. Correlation of BRAFV600E mutation and glucose metabolism in thyroid cancer patients: an18F-FDG PET study.J Nucl Med 2015; 56: 662–7. doi: 10.2967/jnumed.114.150607 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Yoon S, An YS, Lee SJ, So EY, Kim JH, Chung YS, et al. Relation between F-18 FDG uptake of PET/CT and BRAFV600E mutation in papillary thyroid cancer.Medicine (Baltimore) 2015; 94: e2063. doi: 10.1097/MD.0000000000002063 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Na II, Byun BH, Kim KM, Cheon GJ, Choe du H, Koh JS, et al. 18F-FDG uptake and EGFR mutations in patients with non-small cell lung cancer: a single-institution retrospective analysis.Lung Cancer 2010; 67: 76–80. doi: 10.1016/j.lungcan.2009.03.010 [DOI] [PubMed] [Google Scholar]
- 20.Huang CT, Yen RF, Cheng MF, Hsu YC, Wei PF, Tsai YJ, et al. Correlation of F-18 fluorodeoxyglucose-positron emission tomography maximal standardized uptake value and EGFR mutations in advanced lung adenocarcinoma.Med Oncol 2010; 27: 9–15. doi: 10.1007/s12032-008-9160-1 [DOI] [PubMed] [Google Scholar]
- 21.Caicedo C, Garcia-Velloso MJ, Lozano MD, Labiano T, Vigil Diaz C, Lopez-Picazo JM, et al. Role of [18F]FDG PET in prediction of KRAS and EGFR mutation status in patients with advanced non-small-cell lung cancer.Eur J Nucl Med Mol Imaging 2014; 41: 2058–65. doi: 10.1007/s00259-014-2833-4 [DOI] [PubMed] [Google Scholar]
- 22.Mak RH, Digumarthy SR, Muzikansky A, Engelman JA, Shepard J-AO, Choi NC, et al. Role of 18F-fluorodeoxyglucose positron emission tomography in predicting epidermal growth factor receptor mutations in non-small cell lung cancer.Oncologist 2011; 16: 319–26. doi: 10.1634/theoncologist.2010-0300 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lee EY, Khong PL, Lee VH, Qian W, Yu X, Wong MP. Metabolic phenotype of stage IV lung adenocarcinoma: relationship with epidermal growth factor receptor mutation.Clin Nucl Med 2015; 40: e190–5. doi: 10.1097/RLU.0000000000000684 [DOI] [PubMed] [Google Scholar]
- 24.Lee SM, Bae SK, Jung SJ, Kim CK. FDG uptake in non-small cell lung cancer is not an independent predictor of EGFR or KRAS mutation status: a retrospective analysis of 206 patients.Clin Nucl Med 2015; 40: 950–8. doi: 10.1097/RLU.0000000000000975 [DOI] [PubMed] [Google Scholar]
- 25.Chen SW, Chiang HC, Chen WT, Hsieh TC, Yen KY, Chiang SF, et al. Correlation between PET/CT parameters and KRAS expression in colorectal cancer.Clin Nucl Med 2014; 39: 685–9. doi: 10.1097/RLU.0000000000000481 [DOI] [PubMed] [Google Scholar]
- 26.Kawada K, Nakamoto Y, Kawada M, Hida K, Matsumoto T, Murakami T, et al. Relationship between 18F-fluorodeoxyglucose accumulation and KRAS/BRAF mutations in colorectal cancer.Clin Cancer Res 2012; 18: 1696–703. doi: 10.1158/1078-0432.CCR-11-1909 [DOI] [PubMed] [Google Scholar]
- 27.Krikelis D, Skoura E, Kotoula V, Rondogianni P, Pianou N, Samartzis A, et al. Lack of association between KRAS mutations and 18F-FDG PET/CT in caucasian metastatic colorectal cancer patients.Anticancer Res 2014; 34: 2571–9. [PubMed] [Google Scholar]
- 28.Ogino S, Kawasaki T, Brahmandam M, Yan L, Cantor M, Namgyal C, et al. Sensitive sequencing method for KRAS mutation detection by Pyrosequencing.J Mol Diagn 2005; 7: 413–21. doi: 10.1016/S1525-1578(10)60571-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Spittle C, Ward MR, Nathanson KL, Gimotty PA, Rappaport E, Brose MS, et al. Application of a BRAF pyrosequencing assay for mutation detection and copy number analysis in malignant melanoma.J Mol Diagn 2007; 9: 464–71. doi: 10.2353/jmoldx.2007.060191 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hatt M, Cheze-le Rest C, van Baardwijk A, Lambin P, Pradier O, Visvikis D. Impact of tumor size and tracer uptake heterogeneity in (18)F-FDG PET and CT non-small cell lung cancer tumor delineation.J Nucl Med 2011; 52: 1690–7. doi: 10.2967/jnumed.111.092767 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Hatt M, Cheze le Rest C, Descourt P, Dekker A, De Ruysscher D, Oellers M, et al. Accurate automatic delineation of heterogeneous functional volumes in positron emission tomography for oncology applications.Int J Radiat Oncol Biol Phys 2010; 77: 301–8. doi: 10.1016/j.ijrobp.2009.08.018 [DOI] [PubMed] [Google Scholar]
- 32.Hatt M, Cheze Le Rest C, Albarghach N, Pradier O, Visvikis D. PET functional volume delineation: a robustness and repeatability study.Eur J Nucl Med Mol Imaging 2011; 38: 663–72. doi: 10.1007/s00259-010-1688-6 [DOI] [PubMed] [Google Scholar]
- 33.Larson SM, Erdi Y, Akhurst T, Mazumdar M, Macapinlac HA, Finn RD, et al. Tumor treatment response based on visual and quantitative changes in global tumor glycolysis using PET-FDG imaging. The visual response score and the change in total lesion glycolysis.Clin Positron Imaging 1999; 2: 159–71. doi: 10.1016/S1095-0397(99)00016-3 [DOI] [PubMed] [Google Scholar]
- 34.Hatt M, Majdoub M, Valliéres M, Tixier F, Le Rest CC, Groheux D, et al. 18F-FDG PET uptake characterization through texture analysis: investigating the complementary nature of heterogeneity and functional tumor volume in a multi-cancer site patient cohort.J Nucl Med 2015; 56: 38–44. doi: 10.2967/jnumed.114.144055 [DOI] [PubMed] [Google Scholar]
- 35.Tixier F, Le Rest CC, Hatt M, Albarghach N, Pradier O, Metges JP, et al. Intratumor heterogeneity characterized by textural features on baseline 18F-FDG PET images predicts response to concomitant radiochemotherapy in esophageal cancer.J Nucl Med 2011; 52: 369–78. doi: 10.2967/jnumed.110.082404 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Haralick R, Shanmugam K, Dinstein I. Textural features for image classification.IEEE Trans Syst Man Cybern 1973; 3: 610–21. doi: 10.1109/TSMC.1973.4309314 [DOI] [Google Scholar]
- 37.Yu H, Caldwell C, Mah K, Mozeg D. Coregistered FDG PET/CT-based textural characterization of head and neck cancer for radiation treatment planning.IEEE Trans Med Imaging 2009; 28: 374–83. doi: 10.1109/TMI.2008.2004425 [DOI] [PubMed] [Google Scholar]
- 38.Amadasun M, King R. Textural features corresponding to textural properties.IEEE Trans Syst Man Cybern Syst 1989; 19: 1264–74. doi: 10.1109/21.44046 [DOI] [Google Scholar]
- 39.Albanese I, Scibetta AG, Migliavacca M, Russo A, Bazan V, Tomasino RM, et al. Heterogeneity within and between primary colorectal carcinomas and matched metastases as revealed by analysis of Ki-ras and p53 mutations.Biochem Biophys Res Commun 2004; 325: 784–91. doi: 10.1016/j.bbrc.2004.10.111 [DOI] [PubMed] [Google Scholar]
- 40.Mao C, Wu XY, Yang ZY, Threapleton DE, Yuan JQ, Yu YY, et al. Concordant analysis of KRAS, BRAF, PIK3CA mutations, and PTEN expression between primary colorectal cancer and matched metastases.Sci Rep 2015; 5: 8065. doi: 10.1038/srep08065 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Kawada K, Toda K, Nakamoto Y, Iwamoto M, Hatano E, Chen F, et al. Relationship between 18F-FDG PET/CT scans and KRAS mutations in metastatic colorectal cancer.J Nucl Med 2015; 56: 1322–7. doi: 10.2967/jnumed.115.160614 [DOI] [PubMed] [Google Scholar]
- 42.Hatt M, Tixier F, Cheze Le Rest C, Pradier O, Visvikis D. Robustness of intratumour18F-FDG PET uptake heterogeneity quantification for therapy response prediction in oesophageal carcinoma.Eur J Nucl Med Mol Imaging 2013; 40: 1662–71. doi: 10.1007/s00259-013-2486-8 [DOI] [PubMed] [Google Scholar]
- 43.Tixier F, Hatt M, Le Rest CC, Le Pogam A, Corcos L, Visvikis D. Reproducibility of tumor uptake heterogeneity characterization through textural feature analysis in 18F-FDG PET.J Nucl Med 2012; 53: 693–700. doi: 10.2967/jnumed.111.099127 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Cheng NM, Fang YH, Chang JT, Huang CG, Tsan DL, Ng SH, et al. Textural features of pretreatment 18F-FDG PET/CT images: prognostic significance in patients with advanced T-stage oropharyngeal squamous cell carcinoma.J Nucl Med 2013; 54: 1703–9. doi: 10.2967/jnumed.112.119289 [DOI] [PubMed] [Google Scholar]
- 45.Cook GJ, Yip C, Siddique M, Goh V, Chicklore S, Roy A, et al. Are pretreatment 18F-FDG PET tumor textural features in non-small cell lung cancer associated with response and survival after chemoradiotherapy?J Nucl Med 2013; 54: 19–26. doi: 10.2967/jnumed.112.107375 [DOI] [PubMed] [Google Scholar]
- 46.Lovinfosse P, Janvary ZL, Coucke P, Jodogne S, Bernard C, Hatt M, et al. FDG PET/CT texture analysis for predicting the outcome of lung cancer treated by stereotactic body radiation therapy.Eur J Nucl Med Mol Imaging Feb 2016. Epub ahead of print. doi: 10.1007/s00259-016-3314-8 [DOI] [PubMed] [Google Scholar]
- 47.Bundschuh RA, Dinges J, Neumann L, Seyfried M, Zsótér N, Papp L, et al. Textural parameters of tumor heterogeneity in18F-FDG PET/CT for therapy response assessment and prognosis in patients with locally advanced rectal cancer.J Nucl Med 2014; 55: 891–7. doi: 10.2967/jnumed.113.127340 [DOI] [PubMed] [Google Scholar]
- 48.Bang JI, Ha S, Kang SB, Lee KW, Lee HS, Kim JS, et al. Prediction of neoadjuvant radiation chemotherapy response and survival using pretreatment [(18)F]FDG PET/CT scans in locally advanced rectal cancer.Eur J Nucl Med Mol Imaging 2016; 43: 422–31. doi: 10.1007/s00259-015-3180-9 [DOI] [PubMed] [Google Scholar]
- 49.De Cecco CN, Ganeshan B, Ciolina M, Rengo M, Meinel FG, Musio D, et al. Texture analysis as imaging biomarker of tumoral response to neoadjuvant chemoradiotherapy in rectal cancer patients studied with 3-T magnetic resonance.Invest Radiol 2015; 50: 239–45. doi: 10.1097/RLI.0000000000000116 [DOI] [PubMed] [Google Scholar]
- 50.Desseroit MC, Visvikis D, Tixier F, Majdoub M, Perdrisot R, Guillevin R, et al. Development of a nomogram combining clinical staging with 18F-FDG PET/CT image features in non-small-cell lung cancer stage I–III.Eur J Nucl Med Mol Imaging Feb 2016. Epub ahead of print. doi: 10.1007/s00259-016-3325-5 [DOI] [PubMed] [Google Scholar]
- 51.Loupakis F, Ruzzo A, Cremolini C, Vincenzi B, Salvatore L, Santini D, et al. KRAS codon 61, 146 and BRAF mutations predict resistance to cetuximab plus irinotecan in KRAS codon 12 and 13 wild-type metastatic colorectal cancer.Br J Cancer 2009; 101: 715–21. doi: 10.1038/sj.bjc.6605177 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Peeters M, Kafatos G, Taylor A, Gastanaga VM, Oliner KS, Hechmati G, et al. Prevalence ofRAS mutations and individual variation patterns among patients with metastatic colorectal cancer: a pooled analysis of randomised controlled trials.Eur J Cancer 2015; 51: 1704–13. doi: 10.1016/j.ejca.2015.05.017 [DOI] [PubMed] [Google Scholar]
- 53.Yamauchi M, Morikawa T, Kuchiba A, Imamura Y, Qian ZR, Nishihara R, et al. Assessment of colorectal cancer molecular features along bowel subsites challenges the conception of distinct dichotomy of proximalversus distal colorectum.Gut 2012; 61: 847–54. doi: 10.1136/gutjnl-2011-300865 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Denecke T, Rau B, Hoffmann KT, Hildebrandt B, Ruf J, Gutberlet M, et al. Comparison of CT, MRI and FDG-PET in response prediction of patients with locally advanced rectal cancer after multimodal preoperative therapy: is there a benefit in using functional imaging?Eur Radiol 2005; 15: 1658–66. [DOI] [PubMed] [Google Scholar]
- 55.Martoni AA, Di Fabio F, Pinto C, Castellucci P, Pini S, Ceccarelli C, et al. Prospective study on the FDG-PET/CT predictive and prognostic values in patients treated with neoadjuvant chemoradiation therapy and radical surgery for locally advanced rectal cancer.Ann Oncol 2011; 22: 650–6. doi: 10.1093/annonc/mdq433 [DOI] [PubMed] [Google Scholar]
- 56.Lee SJ, Kim JG, Lee SW, Chae YS, Kang BW, Lee YJ, et al. Clinical implications of initial FDG-PET/CT in locally advanced rectal cancer treated with neoadjuvant chemoradiotherapy.Cancer Chemother Pharmacol 2013; 71: 1201–7. doi: 10.1007/s00280-013-2114-0 [DOI] [PubMed] [Google Scholar]
- 57.Sole CV, Calvo FA, Alvarez E, Peligros I, Garcia-Alfonso P, Ferrer C, et al. Clinical significance of VEGFR-2 and 18F-FDG PET/CT SUVmax pretreatment score in predicting the long-term outcome of patients with locally advanced rectal cancer treated with neoadjuvant therapy.Eur J Nucl Med Mol Imaging 2013; 40: 1635–44. doi: 10.1007/s00259-013-2479-7 [DOI] [PubMed] [Google Scholar]