Research ArticleClinical Investigations

Assessment of Simplified Methods to Measure 18F-FLT Uptake Changes in EGFR-Mutated Non–Small Cell Lung Cancer Patients Undergoing EGFR Tyrosine Kinase Inhibitor Treatment

Virginie Frings, Maqsood Yaqub, Lieke L. Hoyng, Sandeep S.V. Golla, Albert D. Windhorst, Robert C. Schuit, Adriaan A. Lammertsma, Otto S. Hoekstra, Egbert F. Smit, Ronald Boellaard and for the QuIC-ConCePT Consortium
Journal of Nuclear Medicine September 2014, 55 (9) 1417-1423; DOI: https://doi.org/10.2967/jnumed.114.140913

Abstract

3′-deoxy-3′-18F-fluorothymidine (18F-FLT) PET/CT provides a noninvasive assessment of proliferation and, as such, could be a valuable imaging biomarker in oncology. The aim of the present study was to assess the validity of simplified quantitative parameters of 18F-FLT uptake in non–small cell lung cancer (NSCLC) patients before and after the start of treatment with a tyrosine kinase inhibitor (TKI). Methods: Ten patients with metastatic NSCLC harboring an activating epidermal growth factor receptor mutation were included in this prospective observational study. Patients underwent 15O-H2O and 18F-FLT PET/CT scanning on 3 separate occasions: within 7 d before treatment, and 7 and 28 d after the first therapeutic dose of a TKI (gefitinib or erlotinib). Dynamic scans were acquired and venous blood samples were collected during the 18F-FLT scan to measure parent fraction and plasma and whole-blood radioactivity concentrations. Simplified measures (standardized uptake value [SUV] and tumor-to-blood ratio [TBR]) were correlated with fully quantitative measures derived from kinetic modeling. Results: Twenty-nine of thirty 18F-FLT PET/CT scans were evaluable. According to the Akaike criterion, a reversible 2-tissue model with 4 rate constants and blood volume parameter was preferred in 84% of cases. Relative therapy-induced changes in SUV and TBR correlated with those derived from kinetic analyses (r2 = 0.83–0.97, P < 0.001, slope = 0.72–1.12). 18F-FLT uptake significantly decreased at 7 and 28 d after the start of treatment compared with baseline (P < 0.01). Changes in 18F-FLT uptake were not correlated with changes in perfusion, as measured using 15O-H2O. Conclusion: SUV and TBR could both be used as surrogate simplified measures to assess changes in 18F-FLT uptake in NSCLC patients treated with a TKI, at the cost of a small underestimation in uptake changes or the need for a blood sample and metabolite measurement, respectively.

Lung cancer is the leading cause of cancer-related death worldwide, and 85% of all lung cancers are non–small cell lung cancer (NSCLC) (1). Moreover, most patients present at an advanced stage, when treatment options are limited (stages III and IV, based on the seventh edition of the TNM staging system for lung cancer (2)). Novel treatments with targeted drugs based on, among others, molecular alterations in the epidermal growth factor receptor (EGFR) have been developed for stage IV NSCLC (3). EGFR is a transmembrane receptor that is involved in cellular processes such as proliferation, angiogenesis, invasion, and resistance to apoptosis. Activating mutations in the EGFR domain result in continuous downstream effects. The intracellular tyrosine kinase part of EGFR can be inhibited reversibly by the EGFR tyrosine kinase inhibitors (TKI) gefitinib and erlotinib. Both drugs have shown efficacy in tumors harboring an activating mutation in the EGFR gene but limited efficacy in EGFR wild type (4–6). Compared with cytotoxic treatment, TKI toxicity profiles are mild and treatment results in a benefit with respect to progression-free survival. Consequently, in this subgroup of patients with an activating EGFR mutation, quality of life is improved with TKI treatment compared with cytotoxic chemotherapy (7).

To evaluate treatment response in patients, an objective noninvasive (imaging) biomarker that can be used early after the start of treatment would be useful, as it would provide a means to identify ineffective treatment at an early stage. Discontinuation of such treatment can prevent unnecessary toxicities and costs. Moreover, an imaging biomarker could provide an early readout of treatment efficacy in drug development (e.g., phase 2 and 3 trials) (8). PET/CT using 3′-deoxy-3′-18F-fluorothymidine (18F-FLT) may be a good candidate, as 18F-FLT is a proliferation marker and uptake of 18F-FLT correlates with immunohistochemistry for proliferation in lung, brain, and breast cancer (9). In addition, changes in maximum standardized uptake values (SUVs) 7 d after the start of treatment with erlotinib correlated with response measured on CT 6 wk after the start of treatment in NSCLC patients (10). 18F-FLT follows the salvage pathway of endogenous thymidine in the cell but is not incorporated into DNA (11). Published data on response evaluation using 18F-FLT PET, however, are contradictory (12–14), and it is not clear to what extent this heterogeneity is related to different pharmacokinetic characteristics, biologic changes, image resolutions, or PET quantification methods. The reference method for quantification of PET studies is full kinetic modeling, which requires arterial blood sampling and dynamic scanning (15). This procedure is not suited for daily clinical practice in which whole-body acquisitions are needed, and it limits the number of centers that can take part in multicenter studies. Therefore, accurate simplified protocols and analytic methods are needed. These methods should be validated against full kinetic modeling both before and after the start of treatment, as tumor blood flow, fractional blood volume, or plasma clearance of 18F-FLT may change because of treatment. Treatment-induced changes in kinetics are accounted for in kinetic modeling but not in simplified measures such as SUV or tumor-to-blood ratios (TBRs) (16), and these simplified measures should therefore be validated. The aim of this project was to perform such a technical validation study and to facilitate future biologic validation studies.

This prospective clinical study investigated whether simplified quantitative methods can be used as alternative measures to evaluate changes in 18F-FLT uptake in NSCLC patients after the start of treatment with TKIs.

MATERIALS AND METHODS

Patients

Patients with stage IV NSCLC and activating EGFR mutations were recruited at 6 medical centers in The Netherlands. The institutional review board of the VU University Medical Center approved this study, and all subjects gave written informed consent. The study was included in the Dutch trial register (trialregister.nl, identification number NTR3557). Patients were scheduled for 3 dynamic 18F-FLT PET/CT scans: before treatment, and 7 and 28 d after the first therapeutic dose of an EGFR TKI. Seven days was chosen as the first time point because erlotinib and gefitinib reach a steady state on day 7. The treating pulmonary physician chose the type of treatment (250 mg of gefitinib orally once a day or 150 mg of erlotinib orally once a day). All scanning was performed at the VU University Medical Center.

PET Imaging

PET scans were obtained using a Gemini TF-64 PET/CT scanner (Philips (17)) with an axial field of view of 18 cm. The scan field was determined by a nuclear physician such that the lung tumor was positioned centrally in the field of view based on a diagnostic CT scan of the thorax. Patients had been fasting for 4 h before the start of the scan to avoid possible food-induced thymidine changes. Tracer was injected and blood sampled through a venous cannula in the forearm. A 370-MBq bolus of 15O-H2O in 5 mL of saline was injected at a rate of 0.8 mL·s−1, followed by a 35-mL saline flush at 2.0 mL·s−1. At the start of the 15O-H2O injection, a dynamic emission scan was started with a total duration of 10 min, binned into 26 frames of the following lengths: 1 × 10, 8 × 5, 4 × 10, 2 × 15, 3 × 20, 2 × 30, and 6 × 60 s. Next, a low-dose CT scan with 120 kV and 50 mAs was performed to correct the former emission scan for attenuation. At least 15 min after injection of 15O-H2O to allow for decay of 15O, an 18F-FLT scan was obtained. This dynamic emission scan was started at the time of a bolus injection of 370 MBq of 18F-FLT in 5 mL of saline (18) at a rate of 0.8 mL·s−1 followed by a 35-mL flush of saline at 2.0 mL·s−1. Immediately after injection, the rest of the activity in the syringe was measured to calculate the net injected 18F-FLT dose. The 18F-FLT scan was binned into 36 frames (1 × 10, 8 × 5, 4 × 10, 3 × 20, 5 × 30, 5 × 60, 4 × 150, 4 × 300 and 2 × 600 s) with a total duration of 60 min. Afterward, a second low-dose CT scan was acquired to correct the 18F-FLT scan for attenuation. All dynamic scan data were corrected for dead time, decay, scatter, and randoms and were reconstructed using the 3-dimensional row-action maximum likelihood algorithm, resulting in a transaxial spatial resolution of approximately 5 mm in full width at half maximum in the center of the field of view (17).

Venous blood samples were drawn during the dynamic 18F-FLT PET scan at 5, 10, 20, 30, 40, and 60 min after injection. Three to five milliliters of blood were drawn before each sample, followed by drawing of a 7-mL sample and then flushing with 2.5 mL of saline. Whole-blood activity concentration, plasma activity concentration, and parent fraction of 18F-FLT were measured for all samples (19).

PET Data Analysis

The outer 5 planes (∼2 cm) of the field of view were not used for quantification. Tumors were delineated on an averaged image of the last 3 frames of the 18F-FLT scan using a 50% threshold of SUVpeak corrected for local contrast as described previously (20,21). In addition, healthy lung volumes of interest (VOIs) were drawn centrally in the contralateral lung using a cylindric VOI with a diameter of 1.2 cm and an axial length of 2.0 cm. A bone marrow VOI of 1.2-cm diameter and 2.4-cm axial length was drawn over a corpus vertebrae. Time–activity curves were generated by projecting VOIs onto all frames of the dynamic 15O-H2O and 18F-FLT scans.

An image-derived input function was extracted from the ascending aorta with 2 × 2 voxels in 5 planes of the early frames (frames 4–6) of both 15O-H2O and 18F-FLT scans, and corresponding time–activity curves were generated. Tails of the 18F-FLT image-derived input function time–activity curves (interval, 500–3,600 s) were calibrated with measured radioactivity concentrations in the venous blood samples, and the entire image-derived input function time–activity curve was rescaled with this calibration factor. In addition, image-derived input functions were corrected for both plasma-to-blood ratios and metabolites to obtain calibrated parent 18F-FLT plasma input functions (11).

Outcome measures of full kinetic modeling and simplified approaches were obtained using an in-house–developed software tool in MATLAB (version 7.10; The MathWorks Inc.). 15O-H2O data were analyzed using the standard single-tissue reversible plasma input model with additional blood volume parameter (22). For 18F-FLT, the compartmental model underlying the biology of 18F-FLT uptake has been described previously (11). Reversible and irreversible 2-tissue plasma input models were investigated for 18F-FLT uptake with blood volume parameter. Individual data points of time–activity curves were weighted on the basis of frame length and whole-scanner true counts per frame (23). These weighting factors were included in the figure of merit (cost function) during curve fitting. Dual-input models for both reversible and irreversible models with blood volume parameter were evaluated to assess the effects of possible cellular uptake of labeled metabolites (24). Net influx rate (Ki) and volume of distribution (VT) were derived from kinetic rate constants:Embedded ImageEq. 1Embedded ImageEq. 2In addition, nondisplaceable binding potential (BPND) was calculated for the reversible 2-tissue model:Embedded ImageEq. 3Simplified measures, SUV and TBR, were calculated for 2 time intervals; 40–60 min and 50–60 min:Embedded ImageEq. 4Embedded ImageEq. 5Normalization factors for SUV were body weight, lean body mass, and body surface area (25). For TBR, whole blood and parent plasma were used as the denominator. In addition, SUV and TBR were calculated per time frame to generate SUV and TBR curves over time.

Relative differences at 7 and 28 d after the start of treatment compared with baseline were calculated for all parameters:Embedded ImageEq. 6

Statistical Analysis

Data were tested for normality by evaluating histograms of all parameters. Mean and SD were used when the distribution was normal, and median and interquartile range (IQR) were used otherwise. The optimal pharmacokinetic model was selected on the basis of the Akaike criterion (26). Statistical analysis comprised linear regression to assess the correlation between simplified and full kinetic outcome measures. r2, slope, and intercept of linear regressions were derived, together with 95% confidence interval. Friedman and Wilcoxon signed-rank test were used to test for statistical differences between pre- and posttreatment scans. P values of less than 0.05 were considered significant. Statistical analyses were performed using SPSS Statistics 20 (IBM Corp.).

RESULTS

Between September 2012 and December 2013, 10 patients were included (demographics and EGFR mutation types are presented in Table 1). Baseline scans were obtained within a median of 1 d (range, 0–3 d) before the start of treatment with the exception of a single patient whose baseline scan was obtained 26 d before the start of treatment with gefitinib. This patient started erlotinib treatment after the baseline scan but stopped 5 d later because of gastrointestinal toxicity. Three weeks later, when symptoms had abated, gefitinib treatment was initiated and follow-up 18F-FLT scans were obtained after the start of treatment with gefitinib without performing a new baseline scan (as the patient had already been pretreated with erlotinib). Follow-up scans were obtained within a median of 7 d (IQR, 6–9) and 28 d (IQR, 27–29) after the start of treatment. Body weight of patients was not significantly different between scans: 67 kg (IQR, 67–89), 68 kg (IQR, 66–89), and 67 kg (IQR, 64–89) at baseline, 7 d, and 28 d after the start of treatment, respectively (Friedman test, P = 0.88).

View this table:
TABLE 1

Patient Demographics

All patients completed the 3 study visits. One 18F-FLT scan was not evaluable because of a scanner failure. Four 15O-H2O scans were not obtained because of technical or logistic problems. The median net injected 18F-FLT doses were 383 MBq (IQR, 352–394), 376 MBq (IQR, 364–390), and 376 MBq (IQR, 365–385) for scans 1, 2, and 3, respectively (Friedman test, P = 0.50). Injected 15O-H2O was 370 MBq for every scan. Median parent fractions of 18F-FLT at 60 min after injection were 76% (IQR, 73–84), 79% (IQR, 74–82), and 81% (IQR, 76–82) for scans 1, 2, and 3, respectively (Friedman test, P = 0.90). Median plasma-to-blood ratios at 60 min after injection were 1.19 (IQR, 1.13–1.27), 1.19 (IQR, 1.15–1.24), and 1.19 (IQR, 1.15–1.23) for scans 1, 2, and 3, respectively (Friedman test, P = 0.64). The median image-derived input function calibration factor was 0.86 (IQR, 0.80–0.94).

Twenty-four 18F-FLT–avid lesions were detected at baseline, with 22 and 17 lesions evaluable (visually detected above background) at 7 and 28 d, respectively, after the start of treatment (Fig. 1). Overall, based on the Akaike criterion, a 2-tissue reversible model was preferred over a 2-tissue irreversible model in 84% (88% at baseline and 82% and 82% at 7 and 28 d, respectively, after the start of treatment; Friedman test, P = 0.14). Dual-input pharmacokinetic modeling did not improve kinetic analyses according to the Akaike criterion. No differences in modeling of 18F-FLT kinetics between patients treated with erlotinib and patients treated with gefitinib were observed, and therefore data were pooled. The results of full kinetic modeling, and SUV and TBR analyses per scan, are shown in Table 2. VT decreased significantly relative to baseline at both 7 d and 28 d after the start of treatment (Friedman test, P < 0.001). Within the evaluable lesions, no significant difference between 7 and 28 d after the start of treatment was observed (Wilcoxon signed-rank test, P = 0.46). SUV and TBR showed the same absolute trend as VT. Relative changes in SUV and TBR correlated strongly with relative changes in VT, with an r2 of 0.83–0.97 (P < 0.001) and a slope of 0.72–1.12 (Fig. 2; Table 3). The intercept ranged from −5% to 12%, which was not significantly different from zero, except for TBR whole blood, which showed a positive bias of 7%–12% (Table 3). SUV reached equilibrium at 30 min after injection, whereas TBR was still increasing at 60 min after injection (Fig. 3).

FIGURE 1.
FIGURE 1.

Fused 18F-FLT PET/CT image of patient with stage IV NSCLC with primary tumor in right lung and contralateral bone metastasis, at baseline (A) and 7 d (B) and 26 d (C) after start of treatment with erlotinib.

View this table:
TABLE 2

18F-FLT Uptake at Baseline and 7 and 28 Days After Start of Treatment

FIGURE 2.
FIGURE 2.

Correlation of percentage change in SUV and TBR for time interval of 50–60 min and VT derived using 2-tissue reversible model with blood volume parameter at 7 d (A) and 28 d (B) after start of treatment. Solid line represents line of identity. LBM = lean body mass; PP = parent plasma; WB = whole blood.

View this table:
TABLE 3

Correlation of Relative Changes in Simplified Measures vs. VT

FIGURE 3.
FIGURE 3.

SUV lean body mass and TBR parent plasma as function of time at baseline for all patients, showing equilibrium of SUV lean body mass being reached at 30 min, whereas TBR parent plasma still increases at 60 min. Solid lines represent TBR parent plasma, and dashed lines represent SUV lean body mass.

18F-FLT kinetics in bone marrow and normal lung were best fitted using a 2-tissue irreversible model with blood volume parameter in 56%, 90%, and 80% (bone marrow) and in 78%, 80%, and 70% (lung) for scans 1, 2, and 3, respectively (Friedman test, P = 0.10 and 0.72 for bone marrow and lung, respectively). 18F-FLT influx in bone marrow and lung, obtained from Ki of the 2-tissue irreversible model, was not significantly different from baseline at 7 or 28 d after the start of treatment (Friedman test, P = 0.24 and 0.46 for bone marrow and lung, respectively).

Tumor perfusion did not significantly change after the start of treatment, with a median K1 of 15O-H2O of 0.41 (IQR, 0.31–0.61), 0.35 (IQR, 0.25–0.59), and 0.39 (IQR, 0.29–0.77) mL·cm−3·min−1 for scans 1, 2, and 3, respectively (Friedman test, P = 0.29). Relative changes in tumor perfusion did not correlate with relative changes in 18F-FLT uptake, with an r2 of 0.05 and 0.06 at 7 and 28 d, respectively after the start of treatment (P = 0.38 and 0.42; Fig. 4).

FIGURE 4.
FIGURE 4.

Correlation of percentage change in perfusion and proliferation, measured as K1 of 15O-H2O and VT of 18F-FLT at 7 d (○) and 28 d (•) after start of treatment.

DISCUSSION

This study investigated the correlation of relative changes in simplified and full kinetic 18F-FLT measures in NSCLC patients after the start of TKI treatment to assess whether simplified measures can be used as surrogate markers of treatment response. A 2-tissue reversible model with blood volume parameter was the preferred model for tumor 18F-FLT kinetics. Relative changes in SUV and TBR correlated well with relative changes in VT (r2 = 0.83–0.97), and therefore, these simplified measures can be used as alternative parameters to evaluate changes in 18F-FLT uptake due to TKI treatment.

Parent fractions and plasma-to-blood ratios were not significantly different between baseline and after the start of treatment, suggesting that the plasma kinetics of 18F-FLT are stable after the start of treatment with TKI, without inter- and intrasubject variability. Therefore, a population-based input function may be a valid alternative for kinetic analysis of response evaluation studies on patients who are treated with a TKI; however, this requires further evaluation in a larger patient group (27).

18F-FLT uptake showed reversible kinetics in most lesions. Although preference for the reversible model decreased after the start of treatment, this trend was not statistically significant. An extended scan duration might provide better fits and more robust estimates of k4 with a sustainable preference for a reversible model after the start of treatment (28). In addition to single-input-function models, dual-input-function models were investigated, to assess whether influx and efflux of metabolites play a role. However, these models did not provide better fits, and therefore results from the 2-tissue reversible model with blood volume parameter were used. From this model, the macroparameter VT represents the most robust result and was used as a reference to assess simplified uptake measures. VT decreased significantly after the start of treatment but was not significantly different between 7 and 28 d after the start of treatment (Table 2). Microparameter K1, which represents 18F-FLT influx, did not change after the start of treatment (P = 0.11), suggesting that expression of nucleoside transporters to the cell membrane are not up- or downregulated after the start of treatment, as has been described in mice (29). In addition, microparameter k3 decreased after the start of treatment (P = 0.05), indicating decreased thymidine kinase activity and thereby a decreased proliferation rate. Therefore, measured changes in 18F-FLT uptake could be attributed to changes in phosphorylation rate (k3) instead of cell membrane transport (K1).

Conceptually, TBR is a simplified model for reversible kinetics and SUV for irreversible models. However, changes in both parameters strongly correlated with changes in VT (r2 = 0.83–0.97) and could be used to evaluate 18F-FLT uptake responses. Relative differences in TBR whole blood had a significant but small positive bias, which might be caused by small differences in parent fraction per scan. Consequently, this may explain why differences in TBR parent plasma at 50–60 min after injection had a high correlation with differences in VT (0.89 and 0.95 after 7 and 28 d, respectively), a slope that was close to the line of identity (1.04 and 0.99), and no bias (intercept not significantly different from zero). Therefore, this parameter appears to be a sensitive measure for detecting relative changes. Unfortunately, however, TBR parent plasma was not stable over time, and differences in acquisition time could affect outcome (higher TBR parent plasma for longer uptake time intervals). In addition, both a blood VOI and a blood sample are needed to measure blood activity concentration, parent fraction, and plasma-to-blood ratio, limiting its feasibility in multicenter studies. Alternatively, SUV showed a plateau beyond 30 min after injection, and therefore measurements will be less dependent on uptake interval as long as the acquisition is performed somewhere between 30 and 60 min after injection, as reported previously (30). On the other hand, SUV underestimated relative therapy-induced changes with slopes significantly less than 1, as is in line with previous results in locally advanced breast cancer (16). Therefore, SUV may be preferred in multicenter 18F-FLT PET studies based on feasibility, but one should be aware of possible underestimation of response assessment (18%–28%) in patients treated with a TKI.

SUV normalization can be performed using body weight, lean body mass, or body surface area. For 18F-FLT, lean body mass might be the best normalization factor, as 18F-FLT has no specific uptake in fat or muscle. This is in agreement with the recommendation to use SUV lean body mass for response evaluation in 18F-FDG PET studies (31). In the present study, body weight was not significantly different between successive scans, and consequently, SUV body weight, lean body mass, and body surface area performed similarly and no definitive decision can be made with respect to the optimal SUV normalization.

Relative changes in tumor perfusion, measured using K1 of 15O-H2O, did not correlate with relative changes in 18F-FLT uptake after the start of treatment. This finding indicates that 18F-FLT uptake is independent of perfusion and that changes in 18F-FLT uptake are not caused by perfusion changes.

The treatment regimen with either gefitinib or erlotinib was determined by the treating physician. Both drugs are reversible TKI with a similar mechanism of action (32,33). In the present study, data were pooled, as no differences in 18F-FLT kinetics were observed between the 2 treatment regimens.

Our study described the technical validation of simplified 18F-FLT uptake measures in NSCLC patients treated with TKI. Biologic validation studies, in which relative changes in simplified measures of 18F-FLT uptake need to be correlated with pathology or clinical outcome, should be performed to determine the predictive value. Once this is confirmed, 18F-FLT PET could qualify as a predictive biomarker of response to TKI in NSCLC. Furthermore, it is tempting to extrapolate the present positive findings of the use of simplified measures to quantify 18F-FLT uptake to other tumor and treatment types. This should, however, be validated in each case separately, as systemic therapy may alter the correlation between simplified and fully quantitative measures (34).

CONCLUSION

In NSCLC patients treated with a TKI, relative changes in SUV and TBR correlated with those in VT of 18F-FLT. SUV lean body mass measured 30–60 min after injection or TBR parent plasma at 50–60 min after injection could be used for future response assessment studies in NSCLC patients with an activating EGFR mutation treated with a TKI, at the cost of a small underestimation in uptake changes or the need for a blood sample and metabolite measurement, respectively.

DISCLOSURE

The costs of publication of this article were defrayed in part by the payment of page charges. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734. The research leading to these results has received support from the Innovative Medicines Initiative Joint Undertaking (www.imi.europa.eu; grant agreement number 115151), whose resources are composed of a financial contribution from the European Union’s Seventh Framework Programme (FP7/2007-2013) and an in-kind contribution from the companies of the European Federation of Pharmaceutical Industries and Associations. No other potential conflict of interest relevant to this article was reported.

Acknowledgments

We thank the patients and their families for participating in this study. We also thank Arifa Moons-Pasic, Saskia Oedjaghir, Arthur Smit, Tineke Lammers, Martin Bard, Alle Welling, Roland Stallaert, and Roald Roeleveld for their help with patient recruitment for this study. In addition, we acknowledge the staff of the Department of Radiology and Nuclear Medicine of the VU University Medical Center, Amsterdam, The Netherlands, for their help with tracer production and data collection. The authors are members of the QuIC-ConCePT Consortium, whose participants include AstraZeneca, the European Organisation for Research and Treatment of Cancer (EORTC), Cancer Research U.K., the University of Manchester, Westfälische Wilhelms-Universität Münster, Radboud University Nijmegen Medical Center, Institut National de la Santé et de la Recherche Médical, Stichting Maastricht Radiation Oncology “Maastro Clinic,” VUmc Amsterdam, King’s College London, Universitair Ziekenhuis Antwerpen, Institute of Cancer Research–Royal Cancer Hospital, Erasmus Universitair Medisch Centrum Rotterdam, Imperial College of Science Technology and Medicine, Keosys S.A.S., Eidgenössische Technische Hochschule Zürich, Amgen NV, Eli Lilly and Company Ltd., GlaxoSmithKline Research & Development Limited, Merck KGa, Pfizer Limited, F. Hoffmann–La Roche Ltd., and Sanofi-Aventis Research and Development.

Footnotes

  • Published online Jun. 26, 2014.

REFERENCES

  1. 1.↵
    1. D’Addario G,
    2. Fruh M,
    3. Reck M,
    4. Baumann P,
    5. Klepetko W,
    6. Felip E
    . Metastatic non-small-cell lung cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2010;21(suppl 5):v116–v119.
    OpenUrlFREE Full Text
  2. 2.↵
    1. Rami-Porta R,
    2. Crowley JJ,
    3. Goldstraw P
    . The revised TNM staging system for lung cancer. Ann Thorac Cardiovasc Surg. 2009;15:4–9.
    OpenUrlPubMed
  3. 3.↵
    Pikor LA, Ramnarine VR, Lam S, Lam WL. Genetic alterations defining NSCLC subtypes and their therapeutic implications. Lung Cancer. 2013;82:179–189.
  4. 4.↵
    1. Kobayashi T,
    2. Koizumi T,
    3. Agatsuma T,
    4. et al
    . A phase II trial of erlotinib in patients with EGFR wild-type advanced non-small-cell lung cancer. Cancer Chemother Pharmacol. 2012;69:1241–1246.
    OpenUrlCrossRefPubMed
  5. 5.
    1. Rosell R,
    2. Carcereny E,
    3. Gervais R,
    4. et al
    . Erlotinib versus standard chemotherapy as first-line treatment for European patients with advanced EGFR mutation-positive non-small-cell lung cancer (EURTAC): a multicentre, open-label, randomised phase 3 trial. Lancet Oncol. 2012;13:239–246.
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. Mok TS,
    2. Wu YL,
    3. Thongprasert S,
    4. et al
    . Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med. 2009;361:947–957.
    OpenUrlCrossRefPubMed
  7. 7.↵
    1. Thongprasert S,
    2. Duffield E,
    3. Saijo N,
    4. et al
    . Health-related quality-of-life in a randomized phase III first-line study of gefitinib versus carboplatin/paclitaxel in clinically selected patients from Asia with advanced NSCLC (IPASS). J Thorac Oncol. 2011;6:1872–1880.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Tan DS,
    2. Thomas GV,
    3. Garrett MD,
    4. et al
    . Biomarker-driven early clinical trials in oncology: a paradigm shift in drug development. Cancer J. 2009;15:406–420.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Chalkidou A,
    2. Landau DB,
    3. Odell EW,
    4. Cornelius VR,
    5. O’Doherty MJ,
    6. Marsden PK
    . Correlation between Ki-67 immunohistochemistry and 18F-fluorothymidine uptake in patients with cancer: a systematic review and meta-analysis. Eur J Cancer. 2012;48:3499–3513.
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. Sohn HJ,
    2. Yang YJ,
    3. Ryu JS,
    4. et al
    . [18F]fluorothymidine positron emission tomography before and 7 days after gefitinib treatment predicts response in patients with advanced adenocarcinoma of the lung. Clin Cancer Res. 2008;14:7423–7429.
    OpenUrlAbstract/FREE Full Text
  11. 11.↵
    1. Frings V,
    2. de Langen AJ,
    3. Yaqub M,
    4. et al
    . Methodological considerations in quantification of 3′-deoxy-3′-[18F]fluorothymidine uptake measured with positron emission tomography in patients with non-small cell lung cancer. Mol Imaging Biol. 2014;16:136–145.
    OpenUrlPubMed
  12. 12.↵
    1. Frings V,
    2. van der Veldt AAM,
    3. Boellaard R,
    4. et al
    . Pemetrexed induced thymidylate synthase inhibition in non-small cell lung cancer patients: a pilot study with 3′-deoxy-3′-[18F]fluorothymidine positron emission tomography. PLoS ONE. 2013;8:e63705.
    OpenUrlCrossRefPubMed
  13. 13.
    1. Wieder HA,
    2. Geinitz H,
    3. Rosenberg R,
    4. et al
    . PET imaging with [18F]3′-deoxy-3′-fluorothymidine for prediction of response to neoadjuvant treatment in patients with rectal cancer. Eur J Nucl Med Mol Imaging. 2007;34:878–883.
    OpenUrlCrossRefPubMed
  14. 14.↵
    1. Zander T,
    2. Scheffler M,
    3. Nogova L,
    4. et al
    . Early prediction of nonprogression in advanced non-small-cell lung cancer treated with erlotinib by using [18F]fluorodeoxyglucose and [18F]fluorothymidine positron emission tomography. J Clin Oncol. 2011;29:1701–1708.
    OpenUrlAbstract/FREE Full Text
  15. 15.↵
    1. Gunn RN,
    2. Gunn SR,
    3. Cunningham VJ
    . Positron emission tomography compartmental models. J Cereb Blood Flow Metab. 2001;21:635–652.
    OpenUrlCrossRefPubMed
  16. 16.↵
    1. Lubberink M,
    2. Direcks W,
    3. Emmering J,
    4. et al
    . Validity of simplified 3′-deoxy-3′-[18F]fluorothymidine uptake measures for monitoring response to chemotherapy in locally advanced breast cancer. Mol Imaging Biol. 2012;14:777–782.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Surti S,
    2. Kuhn A,
    3. Werner ME,
    4. Perkins AE,
    5. Kolthammer J,
    6. Karp JS
    . Performance of Philips Gemini TF PET/CT scanner with special consideration for its time-of-flight imaging capabilities. J Nucl Med. 2007;48:471–480.
    OpenUrlAbstract/FREE Full Text
  18. 18.↵
    1. Machulla HJ,
    2. Blocher A,
    3. Kuntzsch M,
    4. Piert M,
    5. Wei R,
    6. Grierson JR
    . Simplified labeling approach for synthesizing 3-deoxy-3-[18F]fluorothymidine ([18F]FLT). J Radioanal Nucl Chem. 2000;243:843–846.
    OpenUrlCrossRef
  19. 19.↵
    1. de Langen AJ,
    2. Klabbers B,
    3. Lubberink M,
    4. et al
    . Reproducibility of quantitative 18F-3′-deoxy-3′-fluorothymidine measurements using positron emission tomography. Eur J Nucl Med Mol Imaging. 2009;36:389–395.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. Vanderhoek M,
    2. Perlman SB,
    3. Jeraj R
    . Impact of the definition of peak standardized uptake value on quantification of treatment response. J Nucl Med. 2012;53:4–11.
    OpenUrlAbstract/FREE Full Text
  21. 21.↵
    Frings V, Schouten R, Velasquez LM, et al. Repeatability of metabolically active tumor volume of [18F]FDG PET-CT in a multicenter setting [abstract]. J Nucl Med. 2013;54(suppl 2):4P.
  22. 22.↵
    1. van der Veldt AAM,
    2. Hendrikse NH,
    3. Harms HJ,
    4. et al
    . Quantitative parametric perfusion images using 15O-labeled water and a clinical PET/CT scanner: test-retest variability in lung cancer. J Nucl Med. 2010;51:1684–1690.
    OpenUrlAbstract/FREE Full Text
  23. 23.↵
    1. Yaqub M,
    2. Boellaard R,
    3. Kropholler MA,
    4. Lammertsma AA
    . Optimization algorithms and weighting factors for analysis of dynamic PET studies. Phys Med Biol. 2006;51:4217–4232.
    OpenUrlCrossRefPubMed
  24. 24.↵
    1. Tomasi G,
    2. Kimberley S,
    3. Rosso L,
    4. Aboagye E,
    5. Turkheimer F
    . Double-input compartmental modeling and spectral analysis for the quantification of positron emission tomography data in oncology. Phys Med Biol. 2012;57:1889–1906.
    OpenUrlPubMed
  25. 25.↵
    1. Sugawara Y,
    2. Zasadny KR,
    3. Neuhoff AW,
    4. Wahl RL
    . Reevaluation of the standardized uptake value for FDG: variations with body weight and methods for correction. Radiology. 1999;213:521–525.
    OpenUrlCrossRefPubMed
  26. 26.↵
    1. Akaike H
    . A new look at the statistical model identification. IEEE Trans Automat Contr. 1974;19:716–723.
    OpenUrlCrossRef
  27. 27.↵
    1. Contractor KB,
    2. Kenny LM,
    3. Coombes CR,
    4. Turkheimer FE,
    5. Aboagye EO,
    6. Rosso L
    . Evaluation of limited blood sampling population input approaches for kinetic quantification of [18F]fluorothymidine PET data. EJNMMI Res. 2012;2:11.
    OpenUrlPubMed
  28. 28.↵
    1. Muzi M,
    2. Vesselle H,
    3. Grierson JR,
    4. et al
    . Kinetic analysis of 3′-deoxy-3′-fluorothymidine PET studies: validation studies in patients with lung cancer. J Nucl Med. 2005;46:274–282.
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    1. Perumal M,
    2. Pillai RG,
    3. Barthel H,
    4. et al
    . Redistribution of nucleoside transporters to the cell membrane provides a novel approach for imaging thymidylate synthase inhibition by positron emission tomography. Cancer Res. 2006;66:8558–8564.
    OpenUrlAbstract/FREE Full Text
  30. 30.↵
    1. Kenny LM,
    2. Contractor KB,
    3. Stebbing J,
    4. et al
    . Altered tissue 3′-deoxy-3′-[18F]fluorothymidine pharmacokinetics in human breast cancer following capecitabine treatment detected by positron emission tomography. Clin Cancer Res. 2009;15:6649–6657.
    OpenUrlAbstract/FREE Full Text
  31. 31.↵
    1. Wahl RL,
    2. Jacene H,
    3. Kasamon Y,
    4. Lodge MA
    . From RECIST to PERCIST: evolving considerations for PET response criteria in solid tumors. J Nucl Med. 2009;50(suppl 1):122S–150S.
    OpenUrlAbstract/FREE Full Text
  32. 32.↵
    1. Galvani E,
    2. Alfieri R,
    3. Giovannetti E,
    4. et al
    . Epidermal growth factor receptor tyrosine kinase inhibitors: current status and future perspectives in the development of novel irreversible inhibitors for the treatment of mutant non-small cell lung cancer. Curr Pharm Des. 2013;19:818–832.
    OpenUrlPubMed
  33. 33.↵
    1. Wu JY,
    2. Wu SG,
    3. Yang CH,
    4. et al
    . Comparison of gefitinib and erlotinib in advanced NSCLC and the effect of EGFR mutations. Lung Cancer. 2011;72:205–212.
    OpenUrlCrossRefPubMed
  34. 34.↵
    1. Cheebsumon P,
    2. Velasquez LM,
    3. Hoekstra CJ,
    4. et al
    . Measuring response to therapy using FDG PET: semi-quantitative and full kinetic analysis. Eur J Nucl Med Mol Imaging. 2011;38:832–842.
    OpenUrlCrossRefPubMed
  • Received for publication April 1, 2014.
  • Accepted for publication June 3, 2014.
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Bookmark this article

Jump to section

Keywords