Correlation of EGFR or KRAS mutation status with 18F-FDG uptake on PET-CT scan in lung adenocarcinoma

Background 18F-fluoro-2-deoxy-glucose (18F-FDG) positron emission tomography (PET) is a functional imaging modality based on glucose metabolism. The correlation between EGFR or KRAS mutation status and the standardized uptake value (SUV) of 18F-FDG PET scanning has not been fully elucidated. Methods Correlations between EGFR or KRAS mutation status and clinicopathological factors including SUVmax were statistically analyzed in 734 surgically resected lung adenocarcinoma patients. Molecular causal relationships between EGFR or KRAS mutation status and glucose metabolism were then elucidated in 62 lung adenocarcinomas using cap analysis of gene expression (CAGE), a method to determine and quantify the transcription initiation activities of mRNA across the genome. Results EGFR and KRAS mutations were detected in 334 (46%) and 83 (11%) of the 734 lung adenocarcinomas, respectively. The remaining 317 (43%) patients had wild-type tumors for both genes. EGFR mutations were more frequent in tumors with lower SUVmax. In contrast, no relationship was noted between KRAS mutation status and SUVmax. CAGE revealed that 4 genes associated with glucose metabolism (GPI, G6PD, PKM2, and GAPDH) and 5 associated with the cell cycle (ANLN, PTTG1, CIT, KPNA2, and CDC25A) were positively correlated with SUVmax, although expression levels were lower in EGFR-mutated than in wild-type tumors. No similar relationships were noted with KRAS mutations. Conclusions EGFR-mutated adenocarcinomas are biologically indolent with potentially lower levels of glucose metabolism than wild-type tumors. Several genes associated with glucose metabolism and the cell cycle were specifically down-regulated in EGFR-mutated adenocarcinomas.


Introduction
Recently, driver oncogene mutations are being discovered at a rapid pace. Therapeutic agents targeting some of these driver oncogenes have been successfully developed. The somatic mutations in epidermal growth factor receptor (EGFR) and v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog (KRAS) are the most frequently found in lung adenocarcinomas. The presence of an EGFR mutation is the most important predictor of the efficacy of EGFR tyrosine kinase inhibitors (TKIs) [1,2]. In contrast, KRAS mutations are a useful biomarker of EGFR-TKI resistance [3]. It is therefore important to understand the occurrence of EGFR and KRAS mutations when deciding the initial treatment for lung cancer. However, to obtain sufficient tumor tissue to perform the genetic analyses is frequently difficult in lung cancer patients, especially those with unresectable disease. Non-invasive methods to estimate the probability of the EGFR/KRAS mutation status are helpful in clinical practice. 18 F-fluoro-2-deoxy-glucose ( 18 F-FDG) positron emission tomography (PET), a functional imaging modality based on glucose metabolism, has become a standard tool for the diagnosis, initial staging, and evaluation of treatment efficacy in lung cancer [4]. High 18 F-FDG uptake reflects both the increased glucose metabolism and proliferative activity of tumor cells [5,6]. EGFR mutations activate the EGFR-signaling pathway, inhibit apoptosis, and increase cell proliferation, angiogenesis and metastatic potential [7]. KRAS plays a key role in the downstream signaling RAS/MAPK pathway of EGFR and other growth factor receptors [7]. Point mutations of KRAS also play a critical role in cancer cell growth. Therefore, we hypothesized that there is a causal relationship between increased glucose metabolism and EGFR or KRAS mutation.
The emergence of next-generation sequencing technologies has enabled a wide range of protocols for more comprehensive and accurate genome-wide analysis. Among these, cap analysis gene expression (CAGE) is a genome-wide approach forming a comprehensive profile of the transcriptome by sequencing only the 5'-ends of capped RNAs [8]. Profiles represent promoter activities based on the frequencies of transcription starting sites (TSSs). CAGE has been used in genome-wide studies such as the ENCODE project [9] and FANTOM5 project [10][11][12]. Given that the transcriptome represents the molecular basis underlying cellular characteristics, we recently applied CAGE to the study of biomarkers to discriminate distinct types of lung cancer [13].To date, however, CAGE has not been used to study glucose metabolism in tumor cells.
Using transcriptome data from lung adenocarcinomas that monitor expression levels of genes that play important and specific roles in glucose metabolism, we investigated possible correlations between the standardized uptake value (SUV) of 18 F-FDG PET and EGFR or KRAS mutation status in lung adenocarcinoma. Furthermore, we also investigated the specific molecular background of glucose metabolism in EGFR-or KRAS-mutated lung adenocarcinoma.

Patients
Between February 2009 and May 2014, 1414 patients with primary lung cancers, including 1062 with adenocarcinomas, underwent pulmonary resection at our institution. Among these, we retrospectively reviewed 734 adenocarcinoma patients who underwent 18 F-FDG PET-CT scanning within 2 months before surgery and whose surgically resected specimens were examined for EGFR and KRAS mutations. Patients who underwent induction chemotherapy and/or radiotherapy were excluded from this study. Patients were classified into three groups according to the mutation status of the tumors, namely EGFR mutation-positive (EGFR m + ), KRAS mutation-positive (KRAS m + ), and wild-type (WT) for both genes. Clinical characteristics such as age, gender, smoking status, preoperative serum carcinoembryonic antigen (CEA) level and SUV max and pathological findings such as tumor size, nodal status, lymphatic permeation and vascular invasion of EGFR m + and KRAS m + tumors were compared to those of WT tumors.
This study was performed using surgical specimens in the tissue bank at our department, which was established with the approval of the institutional review board (IRB) of Juntendo University School of Medicine. Written consent was obtained from all patients prior to surgery for the procurement of tissue for the research purposes. The IRB approved the use of specimens stored in the tissue bank without obtaining new informed consent and deemed that the contents of this study were ethically acceptable. 18

F-FDG PET-CT scanning
As detailed previously [14], PET-CT scan was carried out with a Discovery ST PET/CT scanner (GE Medical Systems; Waukesha, WI, USA) at the Yotsuya Medical Cube (Tokyo Japan). Two experienced nuclear medicine radiologists (W. K. and M. A.) evaluated the PET-CT images, side by side, and reached a consensus on the findings.

Mutation analyses for EGFR and KRAS
Genomic DNA was extracted from frozen lung cancer tissues sampled from surgically resected specimens. EGFR mutations were analyzed using the peptide nucleic acid-locked nucleic acid polymerase chain reaction (PCR) clamp method [15], and KRAS mutations using the peptide nucleic acid-mediated PCR clamping method [16].

Statistical analysis of the correlations between EGFR or KRAS mutation status and clinicopathological factors
The Steel-Dwass test was used to compare SUV max among multiple groups based on EGFR and KRAS mutation patterns. Receiver operating characteristic (ROC) curves were generated to obtain a cut-off for SUV max of the primary tumor which maximizes the sum of sensitivity and specificity for predicting EGFR or KRAS mutation status. Correlations between EGFR or KRAS mutation status and clinicopathological factors were evaluated. Univariate analyses between SUV max and each clinicopathological factor were performed by a logistic regression model. All of the variables identified to be significant in the univariate analyses were subsequently entered into the multivariate analyses using a bidirectional (i.e., forward and backward) step-wise logistic regression model. A P-value of < 0.05 was considered statistically significant. All statistical analyses were performed using the R statistical software package (version 3.0.2, http://www.r-project.org/). RNA with RNase I, and ligated with 3'-end and 5'-end adaptors specific to the samples. Double-stranded cDNAs were synthesized and mixed for sequencing in one lane of an Illumina HiSeq2500 sequencer (Illumina; San Diego, CA, USA). The CAGE reads were aligned to the reference genome (hg19) with high mapping quality of ! 20.

Differential and correlation analysis using the CAGE data
The aligned CAGE reads were counted in each region of the FANTOM5 robust peaks [11], a reference set of TSS regions, as raw signals for the promoter activities. Expression (activity) levels of individual promoters were quantified as counts per million (CPM) after normalization by the relative log expression method [18], and subjected to differential analysis using edgeR (version 3.2.4) [19] in R/Bioconductor [20]. Associations between expression levels and SUV max and their statistical significance were assessed by Spearman's rank correlation. Only results with a false discovery rate (FDR) less than 1% were considered statistically significant, in both the differential and correlation analyses.
Of the 734 lung adenocarcinomas, EGFR and KRAS mutations were detected in 334 (46%) and 83 (11%), respectively. The EGFR mutation spectra were distributed as follows. The point mutation L858R in exon 21 and deletions in exon 19 were detected in 194 and 120 tumors, respectively, which together accounted for 94% of all EGFR alterations. The remaining 6% of the minor EGFR mutations were exon 18 G719A in 8 tumors, exon 18 G719S in 5, exon 18 G719C in 2 and exon 21 L861Q in 3. Double mutations were found in 2 tumors; 1 harbored exon 21 L861Q and exon 20 T790M and the other had exon 18 G719A and exon 20 T790M, simultaneously. With regard to KRAS, a point mutation in codon 12 was found in 81 (98%) tumors, and a point mutation in codon 13 in 2 (2%). G to T, or G to C transversions were found in 60 (72%) tumors, and G to A transition in 23 (28%). EGFR and KRAS mutations were mutually exclusive.
The median SUV max of all primary tumors was 2.7 (range, 0-33.2). Median SUV max in the EGFR m + group, KRAS m + group, and WT group were 2.1 (range, 0-23), 3.0 (range, 0-23.5), and 3.9 (range, 0-33.2), respectively. SUV max of EGFR m + tumors was significantly lower than that of WT and KRAS m + tumors (Fig 1A). SUV max of tumors with exon 21 L858R or exon 19 deletions was significantly lower than that of WT tumors. However, no significant differences were noted in SUV max between tumors with minor mutations and WT tumors (Fig 1B). The SUV max of KRAS m + tumors did not significantly differ from that of WT tumors (Fig 1A). No significant differences were found in SUV max between tumors with any KRAS mutation spectrum (G to T/G to C transversions or G to A transition) and WT tumors (Fig 1C).
ROC curve analyses of the cut-off values of SUV max for the prediction of EGFR or KRAS mutations Next, we evaluated the prediction of EGFR or KRAS mutation using SUV max . A cut-off value of SUV max 2.69 provided the highest area under the curve (AUC; 0.610) for predicting EGFR mutation, while SUV max 3.40 provided the highest AUC (0.536) for KRAS mutation (Fig 2). Using these cut-off values, parameters for the prediction of EGFR mutations were sensitivity, 60%; specificity, 61%; accuracy, 60%; positive predictive value (PPV), 62%; and negative predictive value (NPV), 59%; and parameters for the prediction of KRAS mutations were sensitivity, 54%; specificity, 54%; accuracy, 54%; PPV, 23%; and NPV, 82%.   Univariate and multivariate analysis of the predictors of EGFR or KRAS mutations On univariate analysis, EGFR mutations were more frequent in females, non-smokers, patients with normal CEA levels, tumors without lymph node involvement or blood vessel invasion, and tumors with lower SUV max . On multivariate analysis, significant predictors of EGFR mutation were smoking status and SUV max ( Table 2). The probability of EGFR mutation was inversely correlated with SUV max. Univariate analyses showed that KRAS mutations were more frequent in males and smokers. On multivariate analysis, the only significant predictor of KRAS mutation was smoking history (Table 3). No relationship was found between the KRAS mutation status and SUV max . The predictability of EGFR mutation status was compared between combinations of well-established clinical factors with or without SUV max ( Table 4). PPV of EGFR mutation status was increased by adding SUV max to gender and smoking status.
We expanded this expression analysis to examine genes involved in the 2 pathways. Among genes whose promoters were more significantly down-regulated in EGFR m + tumors than WT tumors (FDR < 1%), we found that both glucose metabolism-related and cell cycle-related genes were enriched (P value < 5.2e-18 and 0.02, with GO term enrichment analysis with DAVID) [21,22]. Of these, 4 genes associated with glucose metabolism (GPI, G6PD, PKM2, and GAPDH) and 5 genes associated with the cell cycle (ANLN, PTTG1, CIT, KPNA2, and CDC25A) showed a positive correlation between expression and SUV max . (FDR < 1%; Fig 4). Notably, none of the genes down-regulated in KRAS m + tumors showed significant correlation with SUV max .

Discussion
In this study, we found that the probability of EGFR mutation in lung adenocarcinoma was inversely correlated with SUV max . In contrast, the probability of KRAS mutation was not correlated with SUV max . Further, several genes associated with glucose metabolism or the cell cycle were specifically down-regulated in EGFR m + adenocarcinomas. These findings suggest that EGFR m + adenocarcinomas are biologically indolent with potentially lower levels of glucose metabolism than wild-type tumors.
To our knowledge, this is the largest study to evaluate the correlations between 18 F-FDG uptake and EGFR mutation status in lung cancer, and the first to investigate the correlation between the 18 F-FDG uptake and KRAS mutation status. The 4 retrospective studies that previously investigated the correlation between the 18 F-FDG uptake and EGFR mutation status in lung cancer [23][24][25][26] reported contradictory findings (Table 4). In their multivariate analysis, Huang et al. [23] and Ko et al. [26] showed that a higher SUV max was a significant predictor of EGFR mutation, whereas Na et al. [25] and Mak et al. [24] reported that a lower SUV max of the primary tumor was predictive of EGFR mutation. Our findings are compatible with those of the latter groups [24,25]. These conflicting results may have resulted from differences in the ethnic background or the small size of the study populations (Table 5).
Consistent with numerous previous reports [27][28][29], EGFR mutations in the present study were more frequent in females and never-smokers. In addition, a higher probability of EGFR mutation was observed in tumors without lymph node involvement or blood vessel invasion and in those with a lower SUV max . Higashi et al. [30] reported that the prevalence rates of lymphatic permeation and lymph node involvement were lower in primary tumors with low 18 F-FDG uptake than those with a higher 18 F-FDG uptake. These findings suggest that EGFR m + adenocarcinomas are biologically indolent with potentially lower levels of glucose metabolism.
Although many factors have been reported to influence 18 F-FDG uptake, the precise biological mechanism by which 18 F-FDG accumulates in malignant cells remains to be clarified. In 1985, Mueckler et al. [31] initially reported that facilitative glucose transport across the plasma

Scatter plots of association of SUV max with expression levels of four genes associated with glucose metabolism (A-D) and two genes associated with cell proliferation (E and F): (A) GLUT-1, (B) HK-II, (C) CAIX, (D) HIF-1α, (E) VEGF, and (F) BCL2
. Y-axis represents SUV max and X-axis represents gene expression monitored by CAGE, in which the most correlated promoter activities are shown. Black and gray dots represent donors with EGFR mutation-positive (EGFR m + ) and wild-type, respectively.
https://doi.org/10.1371/journal.pone.0175622.g003 membrane was mediated by a family of structurally related proteins known as facilitated diffuse GLUTs. Among the 14 currently known GLUT isoforms [32], the overexpression of GLUT-1 has been shown to be most closely related to 18 F-FDG uptake in lung cancer [33][34][35]. Sasaki et al. [36] reported that GLUT-1 overexpression evaluated by immunohistochemistry was significantly correlated with EFGR or KRAS mutation status, with overexpression in 18 (24%) of 76 EGFR m + lung cancers and 20 (67%) of 30 KRAS m + lung cancers. In our present patients, we found that the expression level of GLUT-1 was positively correlated with SUV max , as were other genes related to glucose metabolism, namely HK-II, CAIX, and HIF-1α (Fig 3). This finding is consistent with previous reports [34,37]. GO term analysis revealed that the glucose metabolism-related and the cell cycle-related genes were enriched among the downregulated genes in EGFR m + adenocarcinomas, which supports our results for 18 F-FDG PET, with lower levels of SUV max . Notably, 4 of the glucose metabolism-related genes, GPI, G6PD, PKM2, and GAPDH and 5 of the cell cycle-related genes, ANLN, PTTG1, CIT, KPNA2, and CDC25A, were significantly down-regulated in EGFR m + adenocarcinomas, and showed a substantial correlation with SUV max (Fig 4). These likely comprise a common subset of the pathway underlying EGFR mutation and glucose metabolism.
Several limitations of our study warrant mention. First, it was conducted under a retrospective design in patients who required surgical resection, most for early stage disease. Accordingly, the selected cases might not have reflected the overall features of lung adenocarcinoma. Second, the sample size of KRAS m + tumors was too small to allow any firm conclusions. Although we found no significant relationship between 18 F-FDG uptake and KRAS mutation status in lung adenocarcinoma and did not identify any genes specifically correlated with glucose metabolism in KRAS m + tumors, a conclusive answer to this question would require a larger sample size.
In summary, the probability of EGFR mutation was inversely correlated with SUV max . In contrast, the probability of KRAS mutation was not correlated with SUV max . Several genes associated with glucose metabolism or the cell cycle were specifically down-regulated in EGFR m + adenocarcinomas. These findings confirm that EGFR m + adenocarcinomas are biologically indolent with potentially lower levels of glucose metabolism than wild-type tumors.