Introduction: The purpose of this study was to evaluate the predictive role of quantitative metabolic parameters using total lesion glycoly-sis (TLG) in 18F-2-fluoro-2-deoxyglucose-positron emission
tomog-raphy/computed tomography for developing gefitinib resistance in epidermal growth factor receptor (EGFR)-mutant non–small-cell lung cancer (NSCLC) patients treated with first-line gefitinib. Methods: A total of 75 NSCLC patients harboring activating EGFR mutation and receiving first-line gefitinib were analyzed. Whole-body 18F-2-fluoro-2-deoxyglucose positron emission tomography/
computed tomography scans were acquired before first-line gefitinib. The maximal standardized uptake value and TLG of all lesions were calculated. Maximal standardized uptake value and TLG were cat-egorized using the tertile cutoff. Treatment outcomes were compared between groups.
Results: Overall response rate of gefitinib was 69.4%, and median progression-free survival (PFS) of gefitinib was 11.5 months. Overall response rates were similar between low, intermediate, and high TLG groups (68.0% versus 76.0% versus 68.0%, p = 0.274). However, PFS significantly differed by TLG groups, and high TLG was asso-ciated with shorter PFS (7.2 months in high TLG; 11.9 months in intermediate TLG; and 24.2 months in low, p < 0.001). Multivariate models adjusted for disease status and response to gefitinib showed that TLG was an independent predictive factor for PFS. TLG was also significantly associated with overall survival (p = 0.005). Conclusion: TLG can predict PFS and development of gefitinib resistance in EGFR-mutant NSCLC patients treated with first-line gefitinib. Baseline metabolic tumor burdens measured with TLG
before first-line gefitinib will be of great help in predicting time to acquired resistance.
Key Words: gefitinib, resistance, EGFR mutation, PET/CT, total lesion glycolysis.
(J Thorac Oncol. 2015;10: 1189–1194)
E
pidermal growth factor receptor (EGFR) mutation is awell-known predictive and prognostic factor for EGFR tyrosine kinase inhibitors (TKIs) efficacy in non–small-cell lung can-cer (NSCLC),1–3 and first-line EGFR TKIs have become
stan-dard treatment for stage IIIB/IV NSCLC harboring activating
EGFR mutations.4,5 However, acquired TKI resistance is
inev-itable in almost all patients, and progression-free survival (PFS) of EGFR TKI has reached 9 to 11 months.4–6 Potential
mechanisms of resistance to gefitinib include EGFR T790M mutation in exon 20, c-Met amplification, and other bypass signaling activation.7 Although tumor burden and preexisting
EGFR T790M mutation at baseline decrease PFS in NSCLC
patients with activating EGFR mutation who received EGFR TKIs, it is difficult for clinicians to predict when resistance will occur during EGFR TKI treatment.8,9
18F-2-Fluoro-2-deoxyglucose (FDG)-positron
emis-sion tomography/computed tomography (PET/CT) is widely used for pretreatment staging, restaging, response monitoring, evaluation of tumor aggressiveness, and prognostication. In patients with NSCLC, FDG-PET/CT has an established role in monitoring treatment response10,11 and prognosis.12 With
FDG-PET/CT, the maximal standard uptake value (SUVmax) has been used as an index for metabolic activity. Recently, there is a growing recognition of volume-based metabolic parameters, such as total lesion glycolysis (TLG), as prom-ising quantitative PET indices.13,14 Compared with SUVmax,
TLG has an advantage of reflecting both the metabolic activity and the tumor burden. However, the potential role for TLG in predicting EGFR TKI resistance has not yet been elucidated. The purpose of this study was to evaluate the predictive role of quantitative metabolic parameters using TLG in FDG-PET/ CT for the development of gefitinib resistance in EGFR-mutant NSCLC patients treated with first-line gefitinib.
Total Lesion Glycolysis in Positron Emission Tomography
Can Predict Gefitinib Outcomes in Non–Small-Cell Lung
Cancer with Activating EGFR Mutation
Bhumsuk Keam, MD, PhD,*† Soo Jin Lee, MD, PhD,‡ Tae Min Kim, MD, PhD,*†
Jin Chul Paeng, MD, PhD,‡, Se-Hoon Lee, MD, PhD,*† Dong-Wan Kim, MD, PhD,*†
Yoon Kyung Jeon, MD, PhD,§ Doo Hyun Chung, MD, PhD,§ Keon Wook Kang, MD, PhD,‡,
June-Key Chung, MD, PhD,†‡ and Dae Seog Heo, MD, PhD*†
*Department of Internal Medicine, Seoul National University Hospital, Seoul, Republic of Korea; †Cancer Research Institute, Seoul National University College of Medicine, Seoul, Republic of Korea; ‡Department of Nuclear Medicine, and §Department of Pathology, Seoul National University Hospital, Seoul, Republic of Korea.
Disclosure: The authors declare no conflict of interest.
Address for correspondence: Tae Min Kim, MD, PhD, Department of Internal Medicine, Seoul National University Hospital, 101 Daehak-ro, Jongno-gu, Seoul 110-744, Republic of Korea. E-mail: gabriel9@snu. ac.kr; Jin Chul Paeng, MD, PhD, Department of Nuclear Medicine, Seoul National University Hospital, 101 Daehak-ro, Jongno-gu, Seoul 110-744, Republic of Korea. E-mail: paengjc@snu.ac.kr
DOI: 10.1097/JTO.0000000000000569
Copyright © 2015 by the International Association for the Study of Lung Cancer ISSN: 1556-0864/15/1008-1189
PATIENTS AND METHOD Study Patients and Gefitinib Treatment
We retrospectively analyzed a consecutive database15–17
of NSCLC patients who were treated with gefitinib at Seoul National University Hospital between May 2007 and January 2012. A total of 75 patients who met the following inclusion criteria were analyzed: pathologically confirmed stage IIIB/IV or recurred, chemotherapy-naive NSCLC; activating EGFR mutation including microdeletion in exon 19 or missense mutation in exon 21 (L858R); receiving first-line gefitinib as palliative chemotherapy; and FDG-PET/CT conducted before first-line gefitinib.
Clinicopathologic parameters were retrieved from electronic medical records, including age, sex, disease sta-tus, smoking history (never-smoker; smoked ≤100 cigarettes in lifetime; and current or ex-smoker), Eastern Cooperative Oncology Group performance status, and EGFR mutation sta-tus. Patients received gefitinib at a dose of 250 mg per day until disease progression or unacceptable toxicities. During gefitinib treatment, CT scans were routinely performed every 8 to 12 weeks for evaluation of response and progression. In case of clinical suspicion for disease progression, CT scans were performed earlier. Treatment response was evaluated based on the Response Evaluation Criteria in Solid Tumors criteria (RECIST) version 1.1.18 Development of gefitinib
resistance was defined by systemic progression of disease by RECIST version 1.1 during treatment with gefitinib. Mutation analyses of EGFR exon 18, 19, 20, and 21 including EGFR T790M mutation were performed by direct DNA sequencing as previously described.17,19
FDG-PET/CT and Image Analysis
Whole-body FDG-PET/CT scans were acquired before first-line gefitinib treatment. In all patients, FDG-PET/CT images were acquired according to our standard imaging protocol, using dedicated PET/CT scanners (Biograph 40 or Biograph 64, Siemens Healthcare, Germany). Patients fasted for at least 8 hours before FDG was injected intravenously. Patients were administered FDG (5.18 MBq/kg), and images were acquired 60 minutes after administration. CT scans were performed first for attenuation correction and lesion localiza-tion, and emission scans were obtained from the skull base to the proximal thigh for 1 minute per bed. PET data reconstruc-tion was done by an iterative algorithm. PET/CT images were reviewed by two experienced nuclear medicine physicians.
On a per-lesion basis, SUVmax and TLG values with three variable margin thresholds were acquired for analysis. TLG of a lesion was calculated as mean SUV (SUVmean) × metabolic tumor volume (MTV), which reflects both the metabolic activity and tumor burden. Details of the TLG-measuring method were described in our previous study.20
MTV for each threshold margin was measured by setting the margin thresholds as 50% of SUVmax for each lesion. In the image analysis, a spheroidal volume of interest (VOI) was drawn to include the entire lesion, and an isocontour VOI was automatically drawn by a dedicated software package (Syngo. via, Siemens Healthcare). Afterward, the values of SUVmax,
MTV, and SUVmean for the VOI were measured. If lesions were bulky and conglomerated so that discrimination of one lesion from another could not be achieved, the lesions were analyzed as a cluster. Measurement of these values was auto-matically performed using the dedicated software and thus had a reproducibility of 100%. Finally, on the patient level, the SUVmax of a patient was defined as the highest SUVmax among all lesions, and TLG of a patient was defined as the sum of all lesions.
Statistical Analysis
Statistical analyses of categorical variables were per-formed using Pearson’s χ2 test. PFS of gefitinib was
deter-mined as the interval between gefitinib administration, and the date when disease progression by RECIST version 1.1 or death was first documented. Overall survival (OS) was measured from the initiation date of gefitinib to the date of death or the last follow-up visit. Median PFS and OS were calculated using the Kaplan–Meier method. Comparisons of survival between the different groups were made using the log-rank test. Multivariate analysis was performed using a Cox proportional hazard regression model. Discrimination for survival data was evaluated using the C statistic with con-cordance index (C-index), which is similar in concept to the area under receiver-operating characteristic curve in the logis-tic model, but appropriate for censored data.21 The C-index
provides the probability that given two randomly selected patients, the patient with the worse outcome will in fact have a worse outcome. A two-sided p less than 0.05 was considered statistically significant. Analyses were completed with STATA software, version 12 (StataCorp LP, College Station, TX). This study was approved by the Institutional Review Board at Seoul National University Hospital (H-1204-021-403).The study was also conducted in accordance with the Principles of the Declaration of Helsinki.
RESULTS Clinical Features
A total of 75 NSCLC patients with activating EGFR mutation were analyzed in this study. All patients were stage IIIB or IV and received first-line gefitinib as palliative che-motherapy. The median follow-up duration was 35 months (range, 7–65 months). Median time between baseline PET/CT and initiation of gefitinib was 19 days (range 0–46 days). The clinicopathologic characteristics of the 75 patients are sum-marized in Table 1. Nearly 60% of patients harbored a deletion mutation in EGFR exon 19, and none had concomitant KRAS or EGFR T790M mutation. The median patient age was 68 years (range, 39–88 years). We used the tertile cutoff as a threshold for the TLG: low/intermediate, 102.8 and intermedi-ate/high, 455.1. We used the tertile cutoff as a threshold for the SUVmax: low/intermediate, 10.6 and intermediate/high, 13.4. Response to Gefitinib and
Metabolic Parameters
The overall response rate (ORR) was 69.4% (complete remission, 2.7% [N = 2]; partial response, 66.7% [N = 50]).
Thirteen patients (17.3%) had stable disease; 5 patients (6.7%) had progressive disease; and 5 patients (6.7%) were not evaluable. ORRs were similar between the SUVmax groups (p = 0.793) and TLG groups (p = 0.274; Table 2). Clinicopathologic factors were not significantly differed by TLG groups (Table 2). Gefitinib outcomes were not differed between adenocarcinoma and other subtypes (ORR, 72.9% versus 81.8%, p = 0.534; median PFS, 11.9 versus 8.5 months,
p = 0.297).
Survival Outcomes and Metabolic Parameters Median PFS was 11.5 months (95% confidence inter-val [CI]: 9.2–13.8 months), and median OS was 26.7 months (95% CI: 23.2–30.2 months). The correlation between TLG and survival was consistently associated with different cutoff levels of TLG in a dose-dependent manner (data not shown). Clinical characteristics and treatment outcomes by TLG are shown in Table 2. Clinical characteristics were comparable between the three TLG groups. PFS differed by TLG, and high TLG was associated with a shorter PFS (7.2 months in high TLG, 11.9 months in intermediate TLG, and 24.2 months in low; p < 0.001; Fig. 1A). Furthermore, TLG was significantly associated with OS proportionally (Fig. 1B). Univariate analy-sis found that TLG, disease status, and response to gefitinib were significantly associated with PFS. Multivariate models adjusted for disease status and response to gefitinib showed that TLG was an independent predictive factor for PFS
(Table 3). SUVmax was also associated with PFS but not OS (Fig. 2A and B). The C-index of SUVmax for PFS was lower than that of TLG (0.629 for SUVmax and 0.661 for TLG), which suggests that TLG was better index for predicting PFS.
DISCUSSION
In this study, we found that the metabolic tumor burden expressed as TLG can predict PFS and developing gefitinib resistance in EGFR-mutant NSCLC patients treated with first-line gefitinib. TLG was also an independent prognostic factor for PFS. To the best of our knowledge, this is the first report demonstrating that TLG can serve as a predictor of gefitinib resistance in EGFR-mutant NSCLC.
Several studies have addressed the clinical signifi-cance of TLG as determined by FDG-PET in NSCLC.13,22–26
TLG may be prognostic for OS in nonsurgical patients with NSCLC.22,23 In this study, TLG was significantly associated
with OS, suggesting that TLG shows both predictive and prog-nostic values in EGFR-mutant NSCLC. The response rate to cytotoxic chemotherapy was not statistically associated with TLG.23 The results of current study are consistent with prior
studies.13,22–26 Thus, taken together with our present results,
TLG might be one of the most useful FDG-PET parameters for prognostication in advanced NSCLC.
Several studies27,28 have shown that SUVmax was also
useful for predicting the clinical outcome of gefitinib in the patients with NSCLC. Low SUVmax was associated with higher response rate and prolonged PFS for gefitinib.27 Early
change of SUVmax after 2days of gefitinib was also useful to predict clinical outcome.28 In this study, SUVmax was
associated with prolonged PFS of gefitinib but not with OS. However, TLG that reflects both the metabolic activity and tumor burden showed significant association with OS and PFS. This suggested that TLC can be better index for predict-ing survival outcome of first-line gefitinib than SUVmax.
Even though the response rate to EGFR TKIs in EGFR-mutant NSCLC is approximately 60% to 80%,4–6 which is
much higher than conventional cytotoxic chemotherapy, the median PFS with EGFR TKIs has been consistently known to be 9 to 11 months,4–6 and all patients who initially respond
to EGFR TKIs eventually develop resistance. The range of PFS is wide, and clinicians have difficulty in predicting when resistance will occur during EGFR TKI therapy in individual patients. Hence, monitoring and predicting when EGFR TKI resistance will develop are clinically important issues. In our previous report,8 tumor burden, expressed as the number of
metastatic sites, is a predictor of poor survival in NSCLC patients with sensitive EGFR mutations who are treated with gefitinib. In this regard, TLG, which reflects both metabolic activity and tumor burden, can be a predictor for timing of EGFR TKI resistance.
In general, tumor cells under selective pressure from EGFR TKIs have undergone a genotypical evolution. In the evolutionary model of resistance during EGFR TKI therapy, the T790M mutation has been identified in approximately 1 of 3 million cells,29 which suggests that resistant mutant clones
increase in proportion to the tumor burden. In an animal model, a large tumor burden would be more likely to have
TABLE 1. Clinicopathologic Characteristics of Enrolled Patients
N = 75 No. of Patients (%)
Age (years) Median age (range) 68 (39–88)
Sex Male 25 (33.3)
Female 50 (66.7)
Status Recurred 9 (12.0)
Initial stage IIIB, IV 66 (88.0)
Smoking Never or light smoker 60 (80.0)
Current or ex-smoker 15 (20.0)
Pathology Adenocarcinoma 64 (85.3)
NSCC NOS 9 (12.0)
Squamous cell carcinoma 1 (1.3)
Pleomorphic carcinoma 1 (1.3)
ECOG PS 0 15 (20.0)
1 44 (58.7)
2 13 (17.3)
3 3 (4.0)
EGFR mutation Deletion 19 44 (58.7)
Exon 21, L858R 31 (41.3)
SUVmax Median (range) 10.6 (1.1–25.4)
Tertile cutoff 9.0/13.4
TLG Median (Range) 204.8 (3.1–8663.0)
Tertile cutoff 102.8/455.1
NSCC NOS, non–small-cell carcinoma not otherwise specified; ECOG PS, Eastern Cooperative Oncology Group Performance Status; EGFR, epidermal growth factor receptor; TLG, total lesion glycolysis; SUVmax, maximal standardized uptake value.
genetically complex tumors that would exhibit either primary or secondary resistance to erlotinib.30 Because the possibility
of tumor heterogeneity is higher in patients with a large tumor burden, the likelihood of developing EGFR TKI resistance may be higher. Hence, patients with a high TLG and higher tumor burden can be genetically more unstable in terms of intratumoral/intertumoral heterogeneity, and it is speculated that these patients show shorter PFS with EGFR TKIs, but nevertheless higher response rates. Tumors may exhibit differ-ent mechanisms of resistance at differdiffer-ent sites within a patidiffer-ent because of genetic heterogeneity.7
This study has several limitations, including a retro-spective design in a single institution and small number of patients that might weaken the statistical significance of TLG. In addition, the optimal cutoff of TLG for PET/CT needs to
be standardized. However, in our study, the patient population might be more homogeneous; all patients had EGFR muta-tions and received gefitinib as a first-line treatment. To our knowledge, this is the first study to have clarified the clinical significance of TLG in EGFR-mutant NSCLC patients treated with first-line EGFR TKIs.
In conclusion, our study indicates that metabolic tumor burden expressed as TLG is predictive of PFS, and a high TLG is significantly associated with reduced survival in
EGFR-mutant NSCLC patients treated with first-line gefitinib.
NSCLC patients with high TLG may experience early develop-ment of EGFR TKI resistance. Baseline metabolic tumor bur-dens measured with TLG before EGFR TKI therapy will be of value in predicting the development of EGFR TKI resistance. These results need to be validated in a prospective study.
TABLE 2. Treatment Outcomes According to SUV and TLG Groups
N = 75 Low TLG 1st Tertile 2nd TertileInt. TLG High TLG 3rd Tertile Pa
Low SUVmax
1st Tertile Int. SUVmax 2nd Tertile High SUVmax 3rd Tertile Pa
Age (yrs) Median (range) 67 (39–88) 72 (41–86) 67(43–81) 0.709 67 (41–86) 68 (39–88) 72 (43–82) 0.817
Sex Male 6 (24.0) 9 (36.0) 10 (40.0) 0.458 9 (36.0) 5 (20.0) 11 (44.0) 0.186
Female 19 (76.0) 16 (64.0) 15 (60.0) 16 (64.0) 20 (80.0) 14 (56.0)
Status Recurred 6 (24.0) 1 (4.0) 2 (8.0) 0.071 7 (28.0) 0 (0.0) 2 (8.0) 0.007
Initial stage IIIB, IV 19 (76.0) 24 (96.0) 23 (92.0) 18 (72.0) 25 (100.0) 23 (92.0)
Smoking Never or light smoker 20 (80.0) 21 (84.0) 19 (76.0) 0.779 20 (80.0) 22 (88.0) 18 (72.0) 0.368
Current or ex-smoker 5 (20.0) 4 (16.0) 6 (24.0) 5 (20.0) 3 (12.0) 7 (28.0)
Pathology Adenocarcinoma 22 (88.0) 23 (92.0) 19 (76.0) 0.487 23 (92.0) 23 (92.0) 16 (64.0) 0.153
NSCC NOS 3 (12.0) 1 (4.0) 5 (20.0) 1 (4.0) 2 (8.0) 6 (24.0)
Squamous cell carcinoma 0 (0.0) 0 (0.0) 1 (4.0) 0 (0.0) 0 (0.0) 1 (4.0)
Pleomorphic carcinoma 0 (0.0) 1 (4.0) 0 (0.0) 1 (4.0) 0 (0.0) 0 (0.0) ECOG PS 0–1 22 (88.0) 18 (72.0) 19 (76.0) 0.356 19 (76.0) 20 (80.0) 20 (80.0) 0.924 2–4 3 (12.0) 7 (28.0) 6 (24.0) 6 (24.0) 5 (20.0) 5 (20.0) Treatment response Responder (CR + PR) 17 (68.0) 19 (76.0) 17 (68.0) 0.274 18 (72.0) 16 (64.0) 18 (72.0) 0.793 Nonresponder (SD + PD) 6 (24.0) 4 (16.0) 8 (32.0) 5 (20.0) 7 (28.0) 6 (24.0) Not evaluable 2 (8.0) 3 (12.0) 0 (0.0) 2 (8.0) 2 (8.0) 1 (4.0) PFS Median PFS (95% CI) 24.2 (9.9–38.4) 11.9 (10.2–13.6) 7.2 (4.0–10.5) <0.001 24.2 (13.2–35.2) 11.4 (8.6–14.1) 8.5 (6.4–10.6) <0.001 OS Median OS (95% CI) NR 31.8 (20.7–42.9) 18.8 (9.7–27.9) 0.005 40.3 (20.2–60.4) 30.0 (16.7–43.3) 24.5 (20.9–28.1) 0.194
CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; PFS, progression-free survival; OS, overall survival; CI, confidence interval; NR, not reached; TLG, total lesion glycolysis; SUVmax, maximal standardized uptake value; ECOG PS, Eastern Cooperative Oncology Group performance status; NSCC NOS, non–small-cell carcinoma not otherwise specified.
aP-value based on analysis of variance, otherwise based on χ2 test.
FIGURE 1. Progression-free survival (A) and overall survival (B) of gefitinib by total lesion glycolysis (TLG).
ACKNOWLEDGMENTS
This study was supported by a grant from the Innovative Research Institute for Cell Therapy, Republic of Korea (A062260). The authors thank Mi Sun Park and Yunhee Choi for biostatistical advice.
REFERENCES
1. Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epider-mal growth factor receptor underlying responsiveness of non-sepider-mall-cell lung cancer to gefitinib. N Engl J Med 2004;350:2129–2139.
2. Paez JG, Jänne PA, Lee JC, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004;304:1497–1500. 3. Han SW, Kim TY, Hwang PG, et al. Predictive and prognostic impact of
epidermal growth factor receptor mutation in non-small-cell lung cancer patients treated with gefitinib. J Clin Oncol 2005;23:2493–2501. 4. Maemondo M, Inoue A, Kobayashi K, et al.; North-East Japan Study
Group. Gefitinib or chemotherapy for non-small-cell lung cancer with mutated EGFR. N Engl J Med 2010;362:2380–2388.
5. Mitsudomi T, Morita S, Yatabe Y, et al.; West Japan Oncology Group. Gefitinib versus cisplatin plus docetaxel in patients with non-small-cell lung cancer harbouring mutations of the epidermal growth factor recep-tor (WJTOG3405): an open label, randomised phase 3 trial. Lancet Oncol 2010;11:121–128.
6. Mok TS, Wu YL, Thongprasert S, et al. Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med 2009;361:947–957. 7. Chong CR, Jänne PA. The quest to overcome resistance to EGFR-targeted
therapies in cancer. Nat Med 2013;19:1389–1400.
8. Park JH, Kim TM, Keam B, et al. Tumor burden is predictive of survival in patients with non-small-cell lung cancer and with activating epidermal growth factor receptor mutations who receive gefitinib. Clin Lung Cancer 2013;14:383–389.
9. Su KY, Chen HY, Li KC, et al. Pretreatment epidermal growth factor receptor (EGFR) T790M mutation predicts shorter EGFR tyrosine kinase inhibitor response duration in patients with non-small-cell lung cancer. J
Clin Oncol 2012;30:433–440.
10. Hicks RJ. Role of 18F-FDG PET in assessment of response in non-small cell lung cancer. J Nucl Med 2009;50 (Suppl 1):31S–42S.
11. Nahmias C, Hanna WT, Wahl LM, Long MJ, Hubner KF, Townsend DW. Time course of early response to chemotherapy in non-small cell lung cancer patients with 18F-FDG PET/CT. J Nucl Med 2007;48:744–751. 12. Berghmans T, Dusart M, Paesmans M, et al. Primary tumor standardized
uptake value (SUVmax) measured on fluorodeoxyglucose positron emis-sion tomography (FDG-PET) is of prognostic value for survival in non-small cell lung cancer (NSCLC): a systematic review and meta-analysis (MA) by the European Lung Cancer Working Party for the IASLC Lung Cancer Staging Project. J Thorac Oncol. 2008;3:6–12.
13. Chung HW, Lee KY, Kim HJ, Kim WS, So Y. FDG PET/CT metabolic tumor volume and total lesion glycolysis predict prognosis in patients with advanced lung adenocarcinoma. J Cancer Res Clin Oncol 2014;140:89–98.
TABLE 3. Univariate and Multivariate Cox Proportional Hazard Regression Model for PFS
Variables
Univariate Analysis for PFS Multivariate Analysis for PFS
HR 95% CI P Value HR 95% CI P Value Age ≤60 yrs 1 0.99–3.02 0.055 1 0.58–1.98 0.837 >60 yrs 1.73 1.07
Disease status Recurred 1
1.20–12.56
0.023 1
0.85–10.65
0.087
Initial metastatic 3.90 3.01
EGFR mutation Deletion 19 1
0.61–1.83 0.832 — L858R 1.06 Smoking Never 1 0.38–1.44 0.370 — Current or ex-smoker 0.74 Response Responder 1 2.40–8.06 <0.001 1 3.43–15.78 <0.001 Nonresponder 4.40 7.36 TLG Continuous, In scale 1.35 1.14–1.61 <0.001 — TLG 1st tertile 1 <0.001 1 0.001 2nd tertile 1.95 0.97–3.94 0.062 2.56 1.15–5.70 0.021 3rd tertile 4.22 2.09–8.49 <0.001 4.46 2.10–9.47 <0.001
PFS, progression-free survival; HR, hazard ratio; CI, confidence interval; EGFR, epidermal growth factor receptor; TLG, total lesion glycolysis.
FIGURE 2. Progression-free survival (A) and overall survival (B) of gefitinib by maximal standardized uptake value (SUVmax).
14. Moon SH, Cho SH, Park LC, et al. Metabolic response evaluated by 18F-FDG PET/CT as a potential screening tool in identifying a subgroup of patients with advanced non-small cell lung cancer for immediate mainte-nance therapy after first-line chemotherapy. Eur J Nucl Med Mol Imaging 2013;40:1005–1013.
15. Lee E, Keam B, Kim DW, et al. Erlotinib versus gefitinib for control of leptomeningeal carcinomatosis in non-small-cell lung cancer. J Thorac
Oncol 2013;8:1069–1074.
16. Keam B, Kim DW, Park JH, et al. How molecular understanding affects to prescribing patterns and clinical outcome of gefitinib in non-small cell lung cancer? 10 Year experience of single institution. Cancer Res Treat 2013;45:178–185.
17. Keam B, Kim DW, Park JH, et al. Rare and complex mutations of epidermal growth factor receptor, and efficacy of tyrosine kinase inhibitor in patients with non-small cell lung cancer. Int J Clin Oncol 2014;19:594–600. 18. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation
criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J
Cancer 2009;45:228–247.
19. Kim YT, Kim TY, Lee DS, et al. Molecular changes of epidermal growth factor receptor (EGFR) and KRAS and their impact on the clinical out-comes in surgically resected adenocarcinoma of the lung. Lung Cancer 2008;59:111–118.
20. Kim TM, Paeng JC, Chun IK, et al. Total lesion glycolysis in posi-tron emission tomography is a better predictor of outcome than the International Prognostic Index for patients with diffuse large B cell lym-phoma. Cancer 2013;119:1195–1202.
21. Pencina MJ, D’Agostino RB. Overall C as a measure of discrimination in survival analysis: model specific population value and confidence interval estimation. Stat Med 2004;23:2109–2123.
22. Liao S, Penney BC, Wroblewski K, et al. Prognostic value of metabolic tumor burden on 18F-FDG PET in nonsurgical patients with non-small cell lung cancer. Eur J Nucl Med Mol Imaging 2012;39:27–38.
23. Zaizen Y, Azuma K, Kurata S, et al. Prognostic significance of total lesion glycolysis in patients with advanced non-small cell lung cancer receiving chemotherapy. Eur J Radiol 2012;81:4179–4184.
24. Zhang H, Wroblewski K, Liao S, et al. Prognostic value of metabolic tumor burden from (18)F-FDG PET in surgical patients with non-small-cell lung cancer. Acad Radiol 2013;20:32–40.
25. Bengtsson T, Hicks RJ, Peterson A, Port RE. 18F-FDG PET as a surro-gate biomarker in non-small cell lung cancer treated with erlotinib: newly identified lesions are more informative than standardized uptake value.
J Nucl Med 2012;53:530–537.
26. Kim K, Kim SJ, Kim IJ, Kim YS, Pak K, Kim H. Prognostic value of volu-metric parameters measured by F-18 FDG PET/CT in surgically resected non-small-cell lung cancer. Nucl Med Commun 2012;33:613–620. 27. Na II, Byun BH, Kang HJ, et al. 18F-Fluoro-2-deoxy-glucose uptake
predicts clinical outcome in patients with gefitinib-treated non-small cell lung cancer. Clin Cancer Res 2008;14:2036–2041.
28. Takahashi R, Hirata H, Tachibana I, et al. Early [18F]fluorodeoxyglucose positron emission tomography at two days of gefitinib treatment pre-dicts clinical outcome in patients with adenocarcinoma of the lung. Clin
Cancer Res 2012;18:220–228.
29. Chmielecki J, Foo J, Oxnard GR, et al. Optimization of dosing for EGFR-mutant non-small cell lung cancer with evolutionary cancer modeling. Sci
Transl Med 2011;3:90ra59.
30. Politi K, Fan PD, Shen R, Zakowski M, Varmus H. Erlotinib resistance in mouse models of epidermal growth factor receptor-induced lung adeno-carcinoma. Dis Model Mech 2010;3:111–119.
