M Vanderhoek

Impact of the Definition of Peak Standardized Uptake Value on Quantification of Treatment Response

PET-based treatment response assessment typically measures the change in maximum standardized uptake value (SUVmax), which is adversely affected by noise. Peak SUV (SUVpeak) has been recommended as a more robust alternative, but its associated region of interest (ROIpeak) is not uniquely defined. We investigated the impact of different ROIpeak definitions on quantification of SUVpeak and tumor response. Methods: Seventeen patients with solid malignancies were treated with a multitargeted receptor tyrosine kinase inhibitor resulting in a variety of responses. Using the cellular proliferation marker 3′-deoxy-3′-18F-fluorothymidine (18F-FLT), whole-body PET/CT scans were acquired at baseline and during treatment. 18F-FLT–avid lesions (∼2/patient) were segmented on PET images, and tumor response was assessed via the relative change in SUVpeak. For each tumor, 24 different SUVpeaks were determined by changing ROIpeak shape (circles vs. spheres), size (7.5–20 mm), and location (centered on SUVmax vs. placed in highest-uptake region), encompassing different definitions from the literature. Within each tumor, variations in the 24 SUVpeaks and tumor responses were measured using coefficient of variation (CV), standardized deviation (SD), and range. For each ROIpeak definition, a population average SUVpeak and tumor response were determined over all tumors. Results: A substantial variation in both SUVpeak and tumor response resulted from changing the ROIpeak definition. The variable ROIpeak definition led to an intratumor SUVpeak variation ranging from 49% above to 46% below the mean (CV, 17%) and an intratumor SUVpeak response variation ranging from 49% above to 35% below the mean (SD, 9%). The variable ROIpeak definition led to a population average SUVpeak variation ranging from 24% above to 28% below the mean (CV, 14%) and a population average SUVpeak response variation ranging from only 3% above to 3% below the mean (SD, 2%). The size of ROIpeak caused more variation in intratumor response than did the location or shape of ROIpeak. Population average tumor response was independent of size, shape, and location of ROIpeak. Conclusion: Quantification of individual tumor response using SUVpeak is highly sensitive to the ROIpeak definition, which can significantly affect the use of SUVpeak for assessment of treatment response. Clinical trials are necessary to compare the efficacy of SUVpeak and SUVmax for quantification of response to therapy.

Pharmacodynamic Study Using FLT PET/CT in Patients with Renal Cell Cancer and Other Solid Malignancies Treated with Sunitinib Malate

Purpose: To characterize proliferative changes in tumors during the sunitinib malate exposure/withdrawal using 3′-deoxy-3′-[18F]fluorothymidine (FLT) positron emission tomography (PET)/computed tomography (CT) imaging. Patients and Methods: Patients with advanced solid malignancies and no prior anti-VEGF exposure were enrolled. All patients had metastatic lesions amenable to FLT PET/CT imaging. Sunitinib was initiated at the standard dose of 50 mg p.o. daily either on a 4/2 or 2/1 schedule. FLT PET/CT scans were obtained at baseline, during sunitinib exposure, and after sunitinib withdrawal within cycle #1 of therapy. VEGF levels and sunitinib pharmacokinetic (PK) data were assessed at the same time points. Results: Sixteen patients (8 patients on 4/2 schedule and 8 patients on 2/1 schedule) completed all three planned FLT PET/CT scans and were evaluable for pharmacodynamic imaging evaluation. During sunitinib withdrawal (change from scans 2 to 3), median FLT PET standardized uptake value (SUVmean) increased +15% (range: −14% to 277%; P = 0.047) for the 4/2 schedule and +19% (range: −5.3% to 200%; P = 0.047) for the 2/1 schedule. Sunitinib PK and VEGF ligand levels increased during sunitinib exposure and returned toward baseline during the treatment withdrawal. Conclusions: The increase of cellular proliferation during sunitinib withdrawal in patients with renal cell carcinoma and other solid malignancies is consistent with a VEGF receptor (VEGFR) tyrosine kinase inhibitor (TKI) withdrawal flare. Univariate and multivariate analysis suggest that plasma VEGF is associated with this flare, with an exploratory analysis implying that patients who experience less clinical benefit have a larger withdrawal flare. This might suggest that patients with a robust compensatory response to VEGFR TKI therapy experience early “angiogenic escape.” Clin Cancer Res; 17(24); 7634–44. ©2011 AACR.

Assessment of GS-9219 in a Pet Dog Model of Non-Hodgkin's Lymphoma

Purpose: To assess, in dogs with naturally occurring non-Hodgkins lymphoma, pharmacokinetics, safety, and activity of GS-9219, a prodrug of the nucleotide analogue 9-(2-phosphonylmethoxyethyl) guanine (PMEG), which delivers PMEG and its phosphorylated metabolites to lymphoid cells with preferential cytotoxicity in cells with a high proliferation index such as lymphoid malignancies. Experimental Design: To generate proof-of-concept, a phase I/II trial was conducted in pet dogs (n = 38) with naturally occurring non-Hodgkins lymphoma using different dose schedules of GS-9219. A subset of dogs was further evaluated with 3′-deoxy-3′-18F-fluorothymidine positron emission tomography/computed tomography imaging before and after treatment. Results: The prodrug had a short plasma half-life but yielded high and prolonged intracellular levels of the cytotoxic metabolite PMEG diphosphate in peripheral blood mononuclear cells in the absence of detectable plasma PMEG. Dose-limiting toxicities were generally manageable and reversible and included dermatopathy, neutropenia, and gastrointestinal signs. Antitumor responses were observed in 79% of dogs and occurred in previously untreated dogs and dogs with chemotherapy-refractory non-Hodgkins lymphoma. The median remission durations observed compare favorably with other monotherapies in dogs with non-Hodgkins lymphoma. High 3′-deoxy-3′-18F-fluorothymidine uptake noted in lymphoid tissues before treatment decreased significantly after treatment (P = 0.016). Conclusions: GS-9219 was generally well tolerated and showed significant activity against spontaneous non-Hodgkins lymphoma as modeled in pet dogs and, as such, supports clinical evaluation in humans.

Impact of Different Standardized Uptake Value Measures on PET-Based Quantification of Treatment Response

PET-based treatment response studies typically measure the change in the standardized uptake value (SUV) to quantify response. The relative changes of different SUV measures, such as maximum, peak, mean, or total SUVs (SUVmax, SUVpeak, SUVmean, or SUVtotal, respectively), are used across the literature to classify patients into response categories, with quantitative thresholds separating the different categories. We investigated the impact of different SUV measures on the quantification and classification of PET-based treatment response. Methods: Sixteen patients with solid malignancies were treated with a multitargeted receptor tyrosine kinase inhibitor, resulting in a variety of responses. Using the cellular proliferation marker 3′-deoxy-3′-18F-fluorothymidine (18F-FLT), we acquired whole-body PET/CT scans at baseline, during treatment, and after treatment. The highest 18F-FLT uptake lesions (∼2/patient) were segmented on PET images. Tumor PET response was assessed via the relative change in SUVmax, SUVpeak, SUVmean, and SUVtotal, thereby yielding 4 different responses for each tumor at mid- and posttreatment. For each SUV measure, a population average PET response was determined over all tumors. Standard deviation (SD) and range were used to quantify variation of PET response within individual tumors and population averages. Results: Different SUV measures resulted in substantial variation of individual tumor PET response assessments (average SD, 20%; average range, 40%). The most extreme variation between 4 PET response measures was 90% in individual tumors. Classification of tumor PET response depended strongly on the SUV measure, because different SUV measures resulted in conflicting categorizations of PET response (ambiguous treatment response assessment) in more than 80% of tumors. Variation of the population average PET response was considerably smaller (average SD, 7%; average range, 16%), and this variation was not statistically significant. Differences in tumor PET response were greatest between SUVmean and SUVtotal and smallest between SUVmax and SUVpeak. Variations of tumor PET response at midtreatment and posttreatment were similar. Conclusion: Quantification and classification of PET-based treatment response in individual patients were strongly affected by the SUV measure used to assess response. This substantial uncertainty in individual patient PET response was present despite the concurrent robustness of the population average PET response. Given the ambiguity of individual patient PET responses, selection of PET-based treatment response measures and their associated thresholds should be carefully optimized.

SU-E-I-176: Impact of Residual Activity and Clock Synchronization on Quantitative Treatment Response Assessment

Purpose: The accuracy of quantitative PET is influenced by many factors whose uncertainties need to be characterized for its reliable use in treatment response assessment. Among the factors are residual activity and lack of synchronization between PET‐scanner and dose‐calibrator clocks which are frequently overlooked. Therefore, we characterized their effects on PET‐ based treatment response assessment. Methods: Fifteen patients receiving targeted molecular therapy underwent whole body [18F]FLT PET/CT scans at multiple time points. Response was calculated as the change in SUV between subsequent scans. Residual‐activity left in syringe after injection effects were assessed by comparing response measures when SUV values were corrected for residual activity (range: 0.07mCi‐0.61mCi) to when they were not. Uncertainty in response was also compared between scans where the synchronization of the PET‐scanner and dose‐calibrator clocks were randomly offset (range: 1min‐60mins) to when they were synchronized. The effect of both factors on PERCIST classification of treatment response was assessed. Results: Residual‐activity and lack of clock synchronization had significant effects on treatment response (p=0.05). 3% change in clock offsets between subsequent scans resulted in 2.5% absolute difference in response. The mean difference in responses due to desynchronized clocks was 18% (maximum=35%). 3% change in residual activity resulted in 0.1% absolute difference in response. Not accounting for residual‐activity caused mean difference of 1% (maximum=6%) in the response. Changes in clock synchronization and not accounting for residual activity resulted in reclassification of the PERCIST response status of over 30% of the patients. Conclusions: Lack of synchronization between PET‐scanner and dose‐ calibrator clocks and not accounting for residual activity caused a difference in response. Administration procedures that minimize residual activity (e.g. flushing syringe with saline) have to be implemented. Also, all clocks used with dose‐calibrator have to be synchronized with PET‐scanner clock periodically. Otherwise patients can be wrongly classified as partial, stable, or progressive disease.

SU‐E‐J‐87: Imaging Biomarkers of Treatment Response: A Roadmap to Validation

Purpose: Quantitative imaging measures are essential for assessment of tumor response to therapy; however, systematic validation is required before imaging biomarkers can be successfully implemented. We performed preliminary validation of the optimal image reconstruction parameters to maximize the accuracy of PET‐based imaging biomarkers. Methods: Optimal PETreconstruction parameters were determined using four spheres of various sizes (10–22mm) containing 168 kBq/cc of [18F]FDG. Spheres were placed inside a NEMA IEC body phantom. Scans were acquired on a GE VCT PET/CT scanner in both 2D and 3D mode. Multiple images were reconstructed using OSEM by varying matrix dimensions, iteration number, and post‐filtration. For each sphere, the recovery coefficient (RC) was used to measure quantitative image accuracy. Robustness of reconstruction parameters to sphere size changes was measured as the coefficient of variation (CV) of the RCs of the individual spheres within an image. Optimal reconstruction parameters were those which resulted in the highest average RC and lowest CV. Results: RC averaged over all four spheres ranged from 0.72 – 0.95 for the image reconstructions investigated. Higher RCs were observed for 2D acquisition mode than 3D (0.91 vs. 0.80), and for 22mm spheres than 10mm spheres (0.99 vs. 0.74). No such trends were observed for CVs. CV values varied by a factor of two across the reconstructions investigated, ranging from 9% to 17%. Image accuracy and robustness to lesion size changes were highest for 2D acquisition, 256×256 matrix, 2 iterations, and 3mm post‐filtration, yielding an RC of 0.92 and CV of ±9%. Conclusions: We successfully optimized image reconstruction parameters to maximize quantitative imaging accuracy, which will increase reproducibility of biomarkers. Future work will include test‐retest imaging of biomarkers in patients to quantify remaining uncertainties. This study represents a critical first step towards validation of imaging biomarkers of treatment response.

SU‐E‐I‐167: Impact of Peak SUV Definition on Quantification of Treatment Response

Purpose: PET‐based treatment response assessment typically measures the change in SUVmax, which is adversely affected by noise. SUVpeak has been recommended as a more robust alternative, but currently its associated region‐of‐interest (ROIpeak) is not uniquely defined. We investigated the impact of different ROIpeak definitions on quantification of SUVpeak and tumor response. Methods: Seventeen patients with solid malignancies were treated with an anti‐proliferative, molecular targeted agent. Using the cellularproliferation marker [F‐18]FLT, whole‐body PET/CT scans were acquired at baseline and during treatment. Lesions with highest FLT uptake (∼2/patient) were segmented on PETimages and tumor response was assessed via the relative change in SUVpeak. For each tumor, 24 different SUVpeak and response values were determined by changing ROIpeak shape, size, and location. Within each tumor, variation of the 24 values was measured using range, coefficient of variation (CV) for SUVpeak, and standard deviation (SD) for response. For each ROIpeak definition, population average SUVpeak and response were determined over all tumors. Results: Substantial variation in both SUVpeak and response resulted from changing the ROIpeak definition. Intra‐tumor SUVpeak variation (CV: 17%) and response variation (SD: 9%) ranged as far as 50% from the mean. Population average SUVpeak variation (CV: 14%) ranged as far as 30% from the mean but population average tumor response variation (SD: 2%) ranged only 3% from the mean. Size of ROIpeak caused more variation in SUVpeak and response than location or shape of ROIpeak. Conclusion: Quantification of individual tumor response using SUVpeak is highly sensitive to ROIpeak definition, which can significantly impact the use of SUVpeak for treatment response assessment. However, population average response is robust to ROIpeak definition. Standardization of SUVpeak is crucial for consistent assessment of treatment response. Clinical trials are necessary to compare the efficacy of SUVpeak and SUVmax for quantification of response to therapy.

SU‐GG‐J‐151: Intra‐Patient Response Heterogeneity Using FLT PET during Chemotherapy in Canine Subjects with Lymphoma

Purpose:PET is routinely used for treatment response assessment in lymphoma after the completion of therapy. However, characterization of lymph node response heterogeneity during therapy could result in treatment modification to improve outcome. We used PET during chemotherapy to assess intra‐patient response heterogeneity in canine subjects with lymphoma. Method and Materials: Nine dogs with lymphoma were treated with GS‐9219, a novel antineoplastic agent preferentially targeting proliferative lymphoid cells. Using the cellularproliferation marker [18F]FLT, whole‐body PET/CT scans were acquired pre‐treatment and after 1, 3, and 5 chemotherapy cycles. Lymph nodes (∼16/dog) were segmented on CTimages and these contours were applied to PETimages to extract mean and maximum SUV for each node. Lymph node response was defined as relative change in SUV. For each dog, nodal responses and baseline values were normalized to their respective means. Heterogeneity was measured using the coefficient of variation (CV) and range of nodal responses and baseline values. Results: Intra‐patient heterogeneity of lymph node baseline values and responses were substantial. Within each dog, the average variation of mean SUV baseline values (CV: 30%) was slightly less than that of response (CV: 40%). Similar heterogeneities were measured using maximum SUV. There was no significant correlation between heterogeneities of baseline uptake and response (p=0.3). Individual dogs responded differently to treatment as some responses were very heterogeneous (most heterogeneous, CV: 115%) while others were quite uniform (least heterogeneous, CV: 5%). Response heterogeneity increased throughout therapy with CV of 35% after 1 chemotherapy cycle, 55% after 3 cycles, and 60% after 5 cycles. Heterogeneity trends measured using CV and ranges were consistent. Conclusions: During chemotherapy, there was substantial intra‐patient response heterogeneity (∼40%) in dogs with lymphoma. Early PET assessment of nodal response heterogeneity may identify poorly responding lymph nodes, permit early intervention, and improve outcomes in lymphoma patients.

MO‐E‐BRB‐05: A Computational Tumor Modeling Framework for the Optimization of Molecular Targeted Therapies

Purpose: Current treatment schedules of molecular targeted therapies are established through costly clinical trials. Although several dosing schemes are used clinically, optimal dosing regimens remain unknown. We developed a computational tumor modeling framework to compare dosing schedules based on simulated therapeutic response. Method and Materials: A pharmacokinetic/pharmacodynamic model was developed to simulate changes in tumorcell proliferation and vascular function. The model was applied to data from a clinical trial in which patients received sunitinib, a molecular targeted agent with anti‐angiogenic and anti‐proliferative effects, on a 4/2 (4 weeks on, 2 week break) or 2/1 schedule. Using [18F]FLT PET/CT imaging, levels of tumor proliferation and vascular function were assessed at baseline, peak drug exposure, and during treatment break. After testing the model on data from the 4/2 schedule, we compared simulated therapeutic responses for these dosing regimens: 4/2 cycle, two consecutive 2/1 cycles, and continuous dosing. Results: Trends in proliferative response were successfully simulated within one standard error of the population means. Two consecutive 2/1 cycles resulted in a 12% greater decrease in tumor proliferation as compared to one 4/2 cycle due to decreased drug washout during the off‐drug period. For iso‐response conditions, the dose for the 2/1 schedule could be reduced to 80% of the 4/2 schedule (from 50mg/kg/day to 40mg/kg/day). Continuous dosing using lower daily doses (32.5mg/kg/day) yielded the best growth inhibition after 6 weeks. Conclusion: The implemented model successfully reproduced trends in proliferative response observed in patients receiving sunitinib. Continuous dosing yielded the best growth inhibition, and outperformed two consecutive 2/1 cycles and the 4/2 regimen, indicating that this regimen might be favorable, especially for patients requiring lower daily doses to manage toxic side effects. Upon successful validation, the implemented model could serve as a cost‐effective tool to help identify improved drug regimens.

MO‐D‐204B‐05: Effect of PET Image Reconstruction Parameters on Quantitative Treatment Response Assessment

Purpose: Changes in SUV are typically used for PET‐based treatment response assessment. However, the absolute value of SUV varies with different reconstruction methods. We studied the effect of different PETimage reconstruction parameters on quantitative treatment response assessment. Method and Materials: Six patients (∼2 tumors/patient) were treated with molecular targeted therapy and received three whole body [18F]FLT (cellularproliferation marker) PET/CT scans at different time points of therapy. Images were reconstructed using filtered backprojection and the OSEM algorithm with varying grid size, number of iterations, and post filter. Each tumor was contoured on a reference reconstruction. This contour was applied to all reconstruction, and SUVmean, SUVmax, SUVpeak, and SUVtotal, were calculated. Treatment response was assessed by change in SUV between scans, normalized to the first scan. The effect of the reconstruction parameters on treatment response was evaluated via the percent difference from the reference reconstruction method which used the OSEM algorithm. Results: FBP yielded higher percent differences than the OSEM reconstructions for all measures. For the OSEM reconstructions, the range was 15%, 50%, 85%, and 25% for the change in SUVmean, SUVmax, SUVpeak, and SUVtotal respectively, while FBP yielded ranges of 60%, 100%, 150%, and 60%. Change in SUVmax, and SUVpeak had the greatest percent differences from the reference reconstruction. The percent difference was as high as 65% and 115% for the change in SUVmax and SUVpeak respectively when images were reconstructed with FBP compared to 25% and 65% when they were reconstructed with OSEM. Conclusion:Image reconstruction parameters affect treatment response assessment, particularly when SUVmax, or SUVpeak are used. Overall, the greatest variation was due to the image reconstruction algorithm. Consistency of reconstruction algorithm and parameters will not eliminate the inherent differences that exist with the use of SUV measures for treatment response assessment.