Tumor Voxels Research Articles

NI-55 * PARAMETRIC MAPS OF CONTRAST AGENT DIFFUSION USING A NOVEL DIFFUSION-PERFUSION MODEL APPLIED TO DCE-MRI CAN DIFFERENTIATE NON-ENHANCING TUMOR FROM EDEMA IN HUMAN GLIOBLASTOMA

INTRODUCTION: We propose a new method for estimating the diffusion of contrast agent in the perfusion scan, dynamic contrast enhanced (DCE)-MRI. We term the new parameteres “effective diffusion coefficients (Deff)”, due to the possibility of simultaneous diffusion and bulk flow in the brain. METHODS: Twenty-six glioblastoma patients were scanned after informed consent. All scans were acquired at the time of declared progression of the tumor. MRI was performed with a 3T Siemens scanner (Erlangen, Germany). A total of 130 images were acquired with five-second time resolution. The bolus of gadolinium-diethylenetriaminepentaacetate (Gd-DTPA, gadopenetic acid, Magnevist) was injected after 10-13 images. Diffusion was estimated via a perfusion-diffusion combination model. RESULTS: To establish whether the diffusion estimates provide new information or can simply be explained by perfusion, we first assess the correlation between diffusion and the traditional DCE parameter from the Tofts model, Ktrans. Within enhancing tumor lesions, where Ktrans is typically used to assess neovascularization, the average correlation is 0.28 ± 0.07 (mean ± standard error), which shows that diffusion estimates depend little on perfusion. We then compare interpatient variability of the same two parameters in normal appearing white matter, with Deff having mean to standard deviation ratio, a form of signal to noise ratio, of 1.55, while the same metric for Ktrans was 0.87. The mean of Deff in NAWM is 3.8x10−4 mm2/s and in tumor is 6.39x10−3 mm2/s. Finally, we determine thresholds that best separate tumor voxels from non-tumor voxels and assess classification accuracy over five trials with sensitivity/specificity of Deff being 91.6/75.3 and for Ktrans, 75.2/74.9, with sensitivity being significanly higher with Deff (p = 3.2x10−9). DISCUSSION: Our preliminary results suggest that Deff provides new information about glioblastomas not achievable by the Tofts model, has greater robustness across patients than Ktrans, and is more sensitive to tumors.

Validation of optimal DCE-MRI perfusion threshold to classify at-risk tumor imaging voxels in heterogeneous cervical cancer for outcome prediction

PurposeTo classify tumor imaging voxels at-risk for treatment failure within the heterogeneous cervical cancer using DCE MRI and determine optimal voxel's DCE threshold values at different treatment time points for early prediction of treatment failure. Material and MethodDCE-MRI from 102 patients with stage IB2–IVB cervical cancer was obtained at 3 different treatment time points: before (MRI 1) and during treatment (MRI 2 at 2–2.5weeks and MRI 3 at 4–5 weeks). For each tumor voxel, the plateau signal intensity (SI) was derived from its time-SI curve from the DCE MRI. The optimal SI thresholds to classify the at-risk tumor voxels was determined by the maximal area under the curve using ROC analysis when varies SI value from 1.0 to 3.0 and correlates with treatment outcome. ResultsThe optimal SI thresholds for MRI 1, 2 and 3 were 2.2, 2.2 and 2.1 for significant differentiation between local recurrence/control, respectively, and 1.8, 2.1 and 2.2 for death/survival, respectively. ConclusionOptimal SI thresholds are clinically validated to quantify at-risk tumor voxels which vary with time. A single universal threshold (SI=1.9) was identified for all 3 treatment time points and remained significant for the early prediction of treatment failure.

Tracer kinetic model selection for dynamic contrast-enhanced magnetic resonance imaging of locally advanced cervical cancer

Background. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) offers a unique capability to probe tumour microvasculature. Different analysis of the acquired data will possibly lead to different conclusions. Therefore, the objective of this study was to investigate under which conditions the Tofts (TM), extended Tofts (ETM), compartmental tissue uptake model (C-TU) and 2-compartment exchange model (2CXM) were the optimal tracer kinetic models (TKMs) for the analysis of DCE-MRI in patients with cervical cancer.Material and methods. Ten patients with locally advanced cervical cancer (FIGO: IIA/IIB/IIIB/IVA – 1/5/3/1) underwent DCE-MRI prior to radiotherapy. From the two-parameter TM it was possible to extract the forward volume transfer constant (Ktrans) and the extracellular-extravascular volume fraction (ve). From the three-parameter ETM, additionally the plasma volume fraction (vp) could be extracted. From the three-parameter C-TU it was possible to extract information about the blood flow (Fp), permeability-surface area product (PS) and vp. Finally, the four-parameter 2CXM extended the C-TU to include ve. For each voxel, corrected Akaike information criterion (AICc) values were calculated, taking into account both the goodness-of-fit and the number of model parameters. The optimal model was defined as the model with the lowest AICc.Results. All four TKMs were the optimal model in different contiguous regions of the cervical tumours. For the 24 999 analysed voxels, the TM was optimal in 17.0%, the ETM was optimal in 2.2%, the C-TU in 23.4% and the 2CXM was optimal in 57.3%. Throughout the tumour, a high correlation was found between Ktrans(TM) and Fp(2CXM), ρ = 0.91.Conclusion. The 2CXM was most often optimal in describing the contrast agent enhancement of pre-treatment cervical cancers, although this model broke down in a subset of the tumour voxels where overfitting resulted in non-physiological parameter estimates. Due to the possible overfitting of the 2CXM, the C-TU was found more robust and when 2CXM was excluded from comparison the C-TU was the preferred model.

WE-G-BRE-03: Dose Painting by Numbers Using Targeted Gold Nanoparticles

Purpose: Homogeneous dose enhancement in tumor cells of lung cancer patients treated with conventional dose of 60–66 Gy in five fractions is limited due to increased risk of toxicity to normal structures. Dose painting by numbers (DPBN) is the prescription of a non-uniform radiation dose distribution in the tumor for each voxel based on the intensity level of that voxel obtained from the tumor image. The purpose of this study is to show that DPBN using targeted gold nanoparticles (GNPs) could enhance conventional doses in the more resistant tumor areas. Methods: Cone beam computed tomography (CBCT) images of GNPs after intratumoral injection into human tumor were taken at 0, 48, 144 and 160 hours. The dose enhancement in the tumor voxels by secondary electrons from the GNPs was calculated based on analytical microdosimetry methods. The dose enhancement factor (DEF) is the ratio of the doses to the tumor with and without the presence of GNPs. The DEF was calculated for each voxel of the images based on the GNP concentration in the tumor sub-volumes using 6-MV photon spectra obtained using Monte Carlo simulations at 5 cm depth (10×10 cm2 field). Results: The results revealed DEF values of 1.05–2.38 for GNPsmore » concentrations of 1–30 mg/g which corresponds to 12.60 – 28.56 Gy per fraction for delivering 12 Gy per fraction homogenously to lung tumor region. Conclusion: Our preliminary results verify that DPBN could be achieved using GNPs to enhance conventional doses to high risk tumor sub-volumes. In practice, DPBN using GNPs could be achieved due to diffusion of targeted GNPs sustainably released in-situ from radiotherapy biomaterials (e.g. fiducials) coated with polymer film containing the GNPs.« less

MO‐G‐17A‐06: Kernel Based Dosimetry for 90Y Microsphere Liver Therapy Using 90Y Bremsstrahlung SPECT/CT

Purpose:90Y microsphere therapy in liver presents a situation where beta transport is dominant and the tissue is relatively homogenous. We compare voxel‐based absorbed doses from a 90Y kernel to Monte Carlo (MC) using quantitative 90Y bremsstrahlung SPECT/CT as source distribution.Methods:Liver, normal liver, and tumors were delineated by an interventional radiologist using contrast‐enhanced CT registered with 90Y SPECT/CT scans for 14 therapies. Right lung was segmented via region growing. The kernel was generated with 1.04 g/cc soft tissue for 4.8 mm voxel matching the SPECT. MC simulation materials included air, lung, soft tissue, and bone with varying densities. We report percent difference between kernel and MC (%Δ(K,MC)) for mean absorbed dose, D70, and V20Gy in total liver, normal liver, tumors, and right lung. We also report %Δ(K,MC) for heterogeneity metrics: coefficient of variation (COV) and D10/D90. The impact of spatial resolution (0, 10, 20 mm FWHM) and lung shunt fraction (LSF) (1,5,10,20%) on the accuracy of MC and kernel doses near the liver‐lung interface was modeled in 1D. We report the distance from the interface where errors become <10% of unblurred MC as d10(side of interface, dose calculation, FWHM blurring, LSF).Results:The %Δ(K,MC) for mean, D70, and V20Gy in tumor and liver was <7% while right lung differences varied from 60–90%. The %Δ(K,MC) for COV was <4.8% for tumor and liver and <54% for the right lung. The %Δ(K,MC) for D10/D90 was <5% for 22/23 tumors. d10(liver,MC,10,1–20) awere <9mm and d10(liver,MC,20,1–20) awere <15mm; both agreed within 3mm to the kernel. d10(lung,MC,10,20), d10(lung,MC,10,1), d10(lung,MC,20,20), and d10(lung,MC,20,1) awere 6, 25, 15, and 34mm, respectively. Kernel calculations on blurred distributions in lung had errors > 10%.Conclusions:Liver and tumor voxel doses with 90Y kernel and MC agree within 7%. Large differences exist between the two methods in right lung.Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number R01CA138986. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Effects of registration error on parametric response map analysis: a simulation study using liver CT-perfusion images

Purpose: To investigate the effects of registration error (RE) on parametric response map (PRM) analysis of pre and post-radiotherapy (RT) functional images.Methods: Arterial blood flow maps (ABF) were generated from the CT-perfusion scans of 5 patients with hepatocellular carcinoma. ABF values within each patient map were modified to produce seven new ABF maps simulating 7 distinct post-RT functional change scenarios. Ground truth PRMs were generated for each patient by comparing the simulated and original ABF maps. Each simulated ABF map was then deformed by different magnitudes of realistic respiratory motion in order to simulate RE. PRMs were generated for each of the deformed maps and then compared to the ground truth PRMs to produce estimates of RE-induced misclassification.Main findings: The percentage of voxels misclassified as decreasing, no change, and increasing, increased with RE For all patients, increasing RE was observed to increase the number of high post-RT ABF voxels associated with low pre-RT ABF voxels and vice versa. 3 mm of average tumour RE resulted in 18-45% tumour voxel misclassification rates.Conclusions: RE induced misclassification posed challenges for PRM analysis in the liver where registration accuracy tends to be lower. Quantitative understanding of the sensitivity of the PRM method to registration error is required if PRMs are to be used to guide radiation therapy dose painting techniques.

Combined measurement of tumor perfusion and glucose metabolism for improved tumor characterization in advanced cervical carcinoma

The aim of this pilot study was (1) to evaluate the combination of [(18)F]fluorodeoxyglucose (FDG) and [(15)O]water for detection of flow-metabolism mismatch in advanced cervical carcinomas, i.e., increased glycolysis at low blood flow, as a possible parameter for prediction of response to treatment, and (2) to propose a method for automated quantification of its spatial extent. The study retrospectively included 10women with advanced cervical carcinoma in whom PET with both FDG and [(15)O]water had been performed prior to therapy. The metabolically active tumor volume was delineated automatically in the FDG images. For computation of the regional blood flow in the tumor, a recovery corrected image-derived arterial input function was used. A tumor voxel was classified as mismatched when the voxel SUV of FDG was larger than the median tumor SUV and the voxel perfusion (K1) was smaller than the median perfusion. The absolute mismatch volume (aMMV) was defined as the volume of all mismatched voxels in ml, and the relative mismatch volume (rMMV) as the ratio of the aMMV to the metabolic tumor volume in percent. The tumors were quite heterogeneous with respect to both FDG uptake and perfusion. The aMMV clustered into 2 groups: "large aMMV" ≥ 10ml in 40 % of patients and "small aMMV" ≤ 5ml in 60 % of patients. The rMMV ranged from 12.7-24.9 %. There was no correlation between rMMV and metabolic tumor volume. There was a tendency (p = 0.126) for an association between rMMV and histological grading, rMMV being about 20 % higher in G3 than in G2 tumors. rMMV did not correlate with SUV or perfusion. These results suggest that combined PET with FDG and [(15)O]water allows detection and quantitative characterization of flow-metabolism mismatch in advanced cervical carcinomas.

Assessment of tumor necrotic fraction by dynamic contrast‐enhanced MRI: a preclinical study of human tumor xenografts with histopathologic correlation

Contrary to the common notion that tumor necrotic regions are non-enhancing after contrast administration, recent evidence has shown that necrotic regions exhibit delayed and slow uptake of gadolinium tracer on dynamic contrast-enhanced MRI (DCE MRI). The purpose of this study is to explore whether the mapping of tumor voxels with delayed and slow enhancement on DCE MRI can be used to derive estimates of tumor necrotic fraction. Patient-derived tumor xenograft lines of seven human cancers were implanted in 26 mice which were subjected to DCE MRI performed using a spoiled gradient recalled sequence. Gadolinium tracer concentration was estimated using the variable flip angle technique. To identify tumor voxels exhibiting delayed and slow uptake of contrast medium, clustering analysis was performed using a k-means clustering algorithm that classified tumor voxels according to their contrast enhancement patterns. Comparison of the percentage of tumor voxels exhibiting delayed and slow enhancement with the tumor necrotic fraction estimated on histology showed a strong correlation (r = 0.962, p < 0.001). The mapping of tumor regions with delayed and slow contrast uptake on DCE MRI correlated strongly with tumor necrotic fraction, and can potentially serve as a non-invasive imaging surrogate for the in vivo assessment of necrotic fraction.

Monitoring tumour response during chemo-radiotherapy: a parametric method using FDG-PET/CT images in patients with oesophageal cancer

BackgroundThe objective of this study is to investigate the feasibility and the additional interest of a parametric imaging (PI) method to monitor the early tumour metabolic response in a prospective series of oesophageal cancer patients who underwent positron emission tomography with fluoro-2-deoxy-d-glucose (FDG-PET/CT) before and during curative-intent chemo-radiotherapy.MethodsFifty-seven patients with squamous cell carcinoma (SCC) of the oesophagus prospectively underwent FDG-PET/CT before chemo-radiotherapy (CRT) (PET1) and at 21 ± 3 days after the beginning of CRT (PET2). The outcome was assessed at 3 months and 1 year after the completion of CRT (clinical examination, CT scan or FDG-PET/CT, biopsy). For each patient, PET1 and PET2 were registered using CT images. The 2 PET image sets were subtracted, so the voxels with significant changes in FDG uptake were identified. A model-based analysis of this graph was used to identify the tumour voxels in which significant changes occurred between the two scans and yielded indices characterising these changes (green and red clusters). Quantitative parameters were compared with clinical outcome at 3 months and at 1 year.ResultsThe baseline tumour FDG uptake decreased significantly at PET2 (p < 0.0001). The tumour volume significantly decreased between PET1 and PET2 (p < 0.02). The initial functional volume of the lesion (TV1) was significantly lower (p < 0.02) in patients in clinical response (CR) at 3 months and 1 year. The volume of the lesion during the treatment (TV2) was significantly lower in patients identified as in CR at 3 months (p < 0.03), but did not predict the outcome at 1 year. Multivariate analyses of outcome at 3 months showed that the risk of failure/death increased with younger age (p = 0.001), larger metabolic volume on PET1 (p = 0.009) and larger volume with decreased FDG uptake (p = 0.047). As for outcome at 1 year, the risk of failure/death increased with younger age (p = 0.006), nodal involvement (p = 0.08) and larger volumes with increased uptake (p = 0.03).ConclusionA parametric method to assess tumour response on serial FDG-PET performed during chemo-radiotherapy was evaluated. Early metabolic changes, i.e. variations in FDG uptake, provided additional prognostic information in multivariate analyses ClinicalTrials.gov NCT 00934505.Trial registrationCurrent Controlled Trials ISRCTN7824458

Accurate prostate tumour detection with multiparametric magnetic resonance imaging: Dependence on histological properties

Background. To benefit most of focal treatment of prostate tumours, detection with high precision of all tumour voxels is needed. Although diffusion-weighted imaging (DWI) and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) have good diagnostic performance, perfect tumour detection is challenging. In this study, we investigated the variation in prostate tissue characteristics Gleason score (GS), cell density (CD) and microvessel density (MVD) to explain the limitations in tumour voxel detection with a MRI-based logistic regression model. Material and methods. Twelve radical prostatectomy patients underwent a pre-operative 3.0T DWI and DCE-MRI exam. The MRI scans were used to calculate voxel-wise tumour probability with a logistic regression model for the peripheral zone (PZ) of the prostate. Tumour probability maps were correlated and validated with whole-mount histology. Additionally, from the whole-mount histological sections CD, MVD and GS were retrieved for every single voxel. GS, CD and MVD of true- and false-positive voxels and of true- and false-negative voxels were compared using Mann-Whitney U-tests. Results. False-negative tumour voxels had significantly lower CD and MVD (p < 0.0001) and were similar to non-tumour PZ. True-positive detected tumour voxels had high CDs and MVDs (p < 0.0001). In addition, tumour voxels with higher GS showed a trend towards more frequent detection (p = 0.06). Tumour voxels with GS ≥ 3 + 4 showed higher CD and MVD compared to tumour voxels with GS 3 + 3 (p < 0.0001). Conclusion. Tumour voxels with low CD and MVD resemble healthy tissue and are limiting tumour voxel detection using DWI and DCE-MRI. Nevertheless, the most aggressive tumour voxels, containing high CD, MVD and GS, are more likely to be detected and can therefore be treated with high dose using focal therapy or focal boosting.

EPR Oxygen Images Predict Tumor Control by a 50% Tumor Control Radiation Dose

Clinical trials to ameliorate hypoxia as a strategy to relieve the radiation resistance it causes have prompted a need to assay the precise extent and location of hypoxia in tumors. Electron paramagnetic resonance oxygen imaging (EPR O2 imaging) provides a noninvasive means to address this need. To obtain a preclinical proof-of-principle that EPR O2 images could predict radiation control, we treated mouse tumors at or near doses required to achieve 50% control (TCD50). Mice with FSa fibrosarcoma or MCa4 carcinoma were subjected to EPR O2 imaging and immediately radiated to a TCD50 or TCD50 ± 10 Gy. Statistical analysis was permitted by collection of approximately 1,300 tumor pO2 image voxels, including the fraction of tumor voxels with pO2 less than 10 mm Hg (HF10). Tumors were followed for 90 days (FSa) or 120 days (MCa4) to determine local control or failure. HF10 obtained from EPR images showed statistically significant differences between tumors that were controlled by the TCD50 and those that were not controlled for both FSa and MCa4. Kaplan-Meier analysis of both types of tumors showed that approximately 90% of mildly hypoxic tumors were controlled (HF10%< 10%), and only 37% (FSA) and 23% (MCa4) tumors controlled if hypoxic. EPR pO2 image voxel distributions in these approximately 0.5 mL tumors provide a prediction of radiation curability independent of radiation dose. These data confirm the significance of EPR pO2 hypoxic fractions. The 90% control of low HF10 tumors argue that 0.5 mL subvolumes of tumors may be more sensitive to radiation and may need less radiation for high tumor control rates. Cancer Res; 73(17); 5328-35. ©2013 AACR.

Improving the Therapeutic Ratio in Stereotactic Radiosurgery: Optimizing Treatment Protocols Based on Kinetics of Repair of Sublethal Radiation Damage

Sublethal damage after radiation exposure may become lethal or be repaired according to repair kinetics. This is a well-established concept in conventional radiotherapy. It also plays an important role in single-dose stereotactic radiotherapy treatments, often called stereotactic radiosurgery, when duration of treatment is extended due to source decay or treatment planning protocol. The purpose of this study is to look into the radiobiological characteristics of normal brain tissue and treatment protocols and find a way to optimize the time course of these protocols. The general problem is nonlinear and can be solved numerically. For numerical optimization of the time course of radiation protocol, a biexponential repair model with slow and fast components was considered. With the clinically imposed constraints of a fixed total dose and total treatment time, three parameters for each fraction (dose-rate, fraction duration, time of each fraction) were simultaneously optimized. A biological optimization can be performed by maximizing the therapeutic difference between tumor control probability and normal tissue complication probability. Specifically, for gamma knife radiosurgery, this approach can be implemented for normal brain tissue or tumor voxels separately in a treatment plan. Differences in repair kinetics of normal tissue and tumors can be used to find clinically optimized protocols. Thus, in addition to considering the physical dose in tumor and normal tissue, we also account for repair of sublethal damage in both these tissues.

A model to simulate the oxygen distribution in hypoxic tumors for different vascular architectures

As hypoxic cells are more resistant to photon radiation, it is desirable to obtain information about the oxygen distribution in tumors prior to the radiation treatment. Noninvasive techniques are currently not able to provide reliable oxygenation maps with sufficient spatial resolution; therefore mathematical models may help to simulate microvascular architectures and the resulting oxygen distributions in the surrounding tissue. Here, the authors present a new computer model, which uses the vascular fraction of tumor voxels, in principle measurable noninvasively in vivo, as input parameter for simulating realistic PO2 histograms in tumors, assuming certain 3D vascular architectures. Oxygen distributions were calculated by solving a reaction-diffusion equation in a reference volume using the particle strength exchange method. Different types of vessel architectures as well as different degrees of vascular heterogeneities are considered. Two types of acute hypoxia (ischemic and hypoxemic) occurring additionally to diffusion-limited (chronic) hypoxia were implemented as well. No statistically significant differences were observed when comparing 2D- and 3D-vessel architectures (p>0.79 in all cases) and highly heterogeneously distributed linear vessels show good agreement, when comparing with published experimental intervessel distance distributions and PO2 histograms. It could be shown that, if information about additional acute hypoxia is available, its contribution to the hypoxic fraction (HF) can be simulated as well. Increases of 128% and 168% in the HF were obtained when representative cases of ischemic and hypoxemic acute hypoxia, respectively, were considered in the simulations. The presented model is able to simulate realistic microscopic oxygen distributions in tumors assuming reasonable vessel architectures and using the vascular fraction as macroscopic input parameter. The model may be used to generate PO2 histograms, which are needed as input in models predicting the radiation response of hypoxic tumors.

SU-E-J-132: Lung Dose Reduction From Patient Specific 4D Motion Based Non-Uniform Dose Prescription in Lung IMRT Treatment

Purpose: Motion of the tumor target in lung can be significant in patients. Traditionally lung dose is uniformly prescribed to the entire Internal Tumor Volume (ITV) with certain setup margin. Such uniform dose prescription may deliver quite high dose to the normal tissue surrounding the target volume due to tumor motion. This study explores patient specific motion based non‐uniform dose prescription that is to deliver prescription dose to the target while maximizing the sparing of the surrounding normal tissue. Methods: Patient 4D CT images were collected during simulation. CT phases were deformably registered. The spatial tracks of the tumor and surrounding tissue voxels were extracted from the transform matrix. The voxel dose to the tumor target and the surrounding tissue was evaluated at each phase and then summed together. Simulated Annealing algorithm was used to search for the spatial dose prescription distribution that delivers sufficient tumor voxel dose while minimizing the surrounding normal tissue dose. Results: Simple phantom and patients studies were included. The optimization process generally converged within 60 seconds to obtain a solution for non‐uniform dose prescription. The Result has shown a reduction in the adjacent lung tissue as great as 18% while maintaining the tumor dose. Conclusion: This study proposed a methodology for a non‐uniform dose prescription that would deliver sufficient dose to the moving lung tumor target while maximizing the sparing of the surrounding lung tissue. The beam delivery and the interplay effect are being further studied.

Variance of SUVs for FDG-PET/CT is Greater in Clinical Practice Than Under Ideal Study Settings

Measurement variance affects the clinical effectiveness of PET-based measurement as a semiquantitative imaging biomarker for cancer response in individual patients and for planning clinical trials. In this study, we measured test-retest reproducibility of SUV measurements under clinical practice conditions and recorded recognized deviations from protocol compliance. Instrument performance calibration, display, and analyses conformed to manufacture recommendations. Baseline clinical (18)F-FDG PET/CT examinations were performed and then repeated at 1 to 7 days. Intended scan initiation uptake period was to repeat the examinations at the same time for each study after injection of 12 mCi FDG tracer. Avidity of uptake was measured in 62 tumors in 21 patients as SUV for maximum voxel (SUV(max)) and for a mean of sampled tumor voxels (SUV(mean)). The range of SUV(max) and SUV(mean) was 1.07 to 21.47 and 0.91 to 14.69, respectively. Intraclass correlation coefficient between log of SUV(max) and log of SUV(mean) was 0.93 (95% confidence interval [CI], 0.88-0.95) and 0.92 (95% CI, 0.87-0.95), respectively.Correlation analysis failed to show an effect on uptake period variation on SUV measurements between the 2 examinations, suggesting additional sources of noise.The threshold criteria for relative difference from baseline for the 95% CI were ± 49% or ± 44% for SUV(max) or SUV(mean), respectively. Variance of SUV for FDG-PET/CT in current clinical practice in a single institution was greater than expected when compared with benchmarks reported under stringent efficacy study settings. Under comparable clinical practice conditions, interpretation of changes in tumor avidity in individuals and assumptions in planning clinical trials may be affected.

Pharmacokinetic parameters derived from dynamic contrast enhanced MRI of cervical cancers predict chemoradiotherapy outcome

PurposeTo assess the prognostic value of pharmacokinetic parameters derived from pre-chemoradiotherapy dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) of cervical cancer patients. Materials and methodsSeventy-eight patients with locally advanced cervical cancer underwent DCE-MRI with Gd-DTPA before chemoradiotherapy. The pharmacokinetic Brix and Tofts models were fitted to contrast enhancement curves in all tumor voxels, providing histograms of several pharmacokinetic parameters (Brix: ABrix, kep, kel, Tofts: Ktrans, νe). A percentile screening approach including log-rank survival tests was undertaken to identify the clinically most relevant part of the intratumoral parameter distribution. Clinical endpoints were progression-free survival (PFS) and locoregional control (LRC). Multivariate analysis including FIGO stage and tumor volume was used to assess the prognostic significance of the imaging parameters. ResultsABrix, kel, and Ktrans were significantly (P<0.05) positively associated with both clinical LRC and PFS, while νe was significantly positively correlated with PFS only. kep showed no association with any endpoint. ABrix was positively correlated with Ktrans and νe, and showed the strongest association with endpoint in the log-rank testing. kel and Ktrans were independent prognostic factors in multivariate analysis with LRC as endpoint. ConclusionsParameters estimated by pharmacokinetic analysis of DCE-MR images obtained prior to chemoradiotherapy may be used for identifying patients at risk of treatment failure.

Abstract P4-01-04: MRI enhancement in stromal tissue surrounding breast tumors: Association with RFS following neoadjuvant chemotherapy

Abstract Objective: Stromal tissue surrounding breast tumors can harbor abnormalities that predict increased risk of recurrence. The purpose of this study was to evaluate low-level contrast enhancement patterns in normal appearing breast stroma surrounding tumors using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). A secondary objective was to investigate whether enhancement patterns varied with distance from tumor and whether proximity-dependent enhancement was associated with recurrence-free survival following neoadjuvant chemotherapy (NACT). Materials and Methods: Sixty-three patients with locally-advanced breast cancer were imaged with DCE-MRI before (V1) and after one cycle (V2) of adriamycin-cytoxan (AC) therapy. Normal-appearing stromal tissue on MR images was defined as fibroglandular tissue outside of tumor regions. Fibroglandular tissue was segmented from pre-contrast T1-weighted images using a fuzzy C-means clustering method1. Tumor regions were identified using a percent enhancement threshold of 70% in the first post-contrast image (PE1). Distance from tumor (proximity) was defined for each stromal voxel as the minimum three-dimensional distance from any tumor voxel. Stromal enhancement was characterized by calculating mean PE and mean signal enhancement ratio (SER = PE1/PE2) in distance shells of 5 mm, from 0 to 40 mm outside of tumor tissue. Global PE and SER were calculated by averaging all stromal voxels from 5 to 40 mm from tumor. Proximity-dependent PE and SER were analyzed using a linear mixed effects model and Cox proportional hazards model for recurrence-free survival. Results: The mixed effects model displayed a decreasing radial trend in PE at both V1(estimated mean = −0.39 per mm, 95% CI (−0.50, −0.29), p &amp;lt; 0.0001) and V2 (estimated mean = −0.34 per mm, 95% CI (−0.56, −0.13), p = 0.002). The trend was less pronounced in SER and did not achieve statistical significance (V1, p = 0.12; V2 p = 0.11). For the recurrent group, there was a marginal trend of lower global SER at V1 (estimated difference of −0.07, 95% CI (−0.13, 0.01) two-tail Wilcoxon-Mann-Whitney U-test, p = 0.11) but not at V2 (estimated difference in location of −0.04, 95% CI (−0.11, 0.26), p = 0.4). There was no evidence for a clear difference in PE for recurrent verses non-recurrent patients at either visit. Survival analysis with average PE or SER as predictors showed that although PE provided no detectable predictive power at V1 and V2, the SER hazard ratio estimates for each unit decrease in SER was marginally statistically significant at 11.5, 95% CI (0.85, 155), p = 0.07 for V1; but was not so for V2: estimate = 0.85, 95% CI (0.04, 20), p = 0.9. Conclusions: These findings show that stromal tissue outside the primary tumor can be quantitatively characterized by DCE-MRI, and suggest that stromal enhancement measurements may be worth exploring as a potential predictor of recurrence/disease-free survival following NACT. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P4-01-04.

FDG-PET/CT Metabolic Tumor Volume: A New Prognostic Marker in Hodgkin Lymphoma?

AbstractAbstract 3632 Introduction:Although Hodgkin Lymphoma (HL) is highly curable, about 15% of patients (pts) are refractory or relapsed after first line treatment. Classic prognostic scores (e.g. IPS) are useful for identifying high risk pts, who need intensive treatment, and low risk pts, who beneficiate de-escalation to minimize side effects. However, they are not enough suitable to predict outcomes. Consequently, finding new complementary tools for detecting refractory or relapsing pts, remains a challenge. Fluorodeoxyglucose (FDG)-PET/CT involvement in initial staging has been widely studied. Although clinical or CT tumor volume is an important prognostic factor, metabolic tumor volume (MTV) is not enough explored. We performed a study to 1) determine whether MTV and maximum standardized uptake value (SUV) max could be new prognostic markers and 2) compare metabolic tissue heterogeneity with CD68 expression, a promising new prognostic factor linked with inflammatory microenvironment. Patients and methods:Among 456 histologically proven HL pts registered in the Regional Lymphoma database of Limousin (SRRLL, France) since 1990’s, 158 have an available sample for CD68 staining. Among the 106 pts who have undergone FDG-PET, 43 pts have available quantitative initial and early response (post-C2) SUV FDG-PET/CT data. The median follow-up was 21 months (6–72.5). Pts were classified following Ann Arbor stages I: 4 pts, II: 17 pts, III: 9 pts, IV: 13 pts. FDG-PET/CT exams were performed with a biograph6 Siemens® device and analyzed with Siemens MI® application. MTV was computed for all pts with a 2D delineation technique and using a thresholding method. The threshold (T) corresponds to mean liver SUV (+3 sd) calculated into 50 cm3 of normal liver. All the tumor voxels (3D pixels) equal or greater than the T belong to MTV. MTV per pathological area (MTVa) was also analyzed. Pts with spleen lesions have an increased volume compared to the others. To minimize spleen’s impact, MTV was calculated without spleen (MTVws). Quantitative FDG uptake is routinely measured by SUVmax. The mean SUVmax was also worked out into 1 cm3 around the tumor SUVmax. ROC curves were plotted for continuous variables such as age, erythrocyte sedimentation rate (ESR), MTV, SUVmax and mean SUVmax. CD68, tumor associated macrophage expression, was tested with a 25% threshold for positivity. Significant factors allowed dividing pts into favorable and unfavorable groups. Event free survival (EFS) studies were carried out for all binary variables using COX model. A univariate regression analysis was performed. Variables with a p<0.20 were included in multivariate analysis. Results:Median age and sex ratio (n=43) were respectively 29 y-o (16–77) and 0.87. ROC and univariate analysis showed that MTV, MTVa and MTVws were significant predictors for absence of complete remission (CR) at post-C2 and EFS. Cut-off values for prognosis (Cp) were 310 cm3 for MTV or MTVws and 53 cm3 for MTVa (p<0.001). Cut-off values for post-C2 were 244 cm3 for MTV or MTVws and 62 cm3 for MTVa (p<0.007). For pts with a large tumor volume (MTV, MTVws or MTVa > Cp), 1 and 2 years EFS were shorter (figure1). Two years EFS for MTV, MTVa and MTVws are respectively: 40%, 50% and 21% for large tumor volume and 87%, 86% and 89% for small tumor volume (p= 0.004, p=0.01 and p=0.0003). The mean SUVmax and heterogeneity were significant in univariate analysis (p=0.01 and p=0.04) but not in ROC analysis. Heterogeneity level was not correlated with CD68 expression. Each new parameter was compared, in multivariate analysis, to ESR, stages, B symptoms, bulky, age. MTV was an independent factor for predicting outcomes and post-C2 results (p<0.01) and better than mean SUV max. Similar results were found for MTVa and MTVws (p<0.02 and p<0.01). Discussion:In spite of limited number of pts and short follow-up, prognostic value of MTV is very significant. Those data are in accordance with the few results previously reported (Hutchings et al. Int J Radiat Oncol Biol Phys, 2005). Song et al. (Cancer Science, 2012) also highlighted that MTV was better predictor than SUVmax in DLBCL. Conclusion:Large MTV seems to be a potential predictive marker for adverse post-C2 results and EFS. These data need further studies to confirm, in larger cohort, these first promising results to establish a better initial risk stratification of pts, leading to optimal adaptive therapy strategy. [Display omitted] Disclosures:No relevant conflicts of interest to declare.

Lung tumor segmentation in PET images using graph cuts

The aim of segmentation of tumor regions in positron emission tomography (PET) is to provide more accurate measurements of tumor size and extension into adjacent structures, than is possible with visual assessment alone and hence improve patient management decisions. We propose a segmentation energy function for the graph cuts technique to improve lung tumor segmentation with PET. Our segmentation energy is based on an analysis of the tumor voxels in PET images combined with a standardized uptake value (SUV) cost function and a monotonic downhill SUV feature. The monotonic downhill feature avoids segmentation leakage into surrounding tissues with similar or higher PET tracer uptake than the tumor and the SUV cost function improves the boundary definition and also addresses situations where the lung tumor is heterogeneous. We evaluated the method in 42 clinical PET volumes from patients with non-small cell lung cancer (NSCLC). Our method improves segmentation and performs better than region growing approaches, the watershed technique, fuzzy-c-means, region-based active contour and tumor customized downhill.

Spatially resolved regression analysis of pre-treatment FDG, FLT and Cu-ATSM PET from post-treatment FDG PET: An exploratory study

PurposeTo quantify associations between pre-radiotherapy and post-radiotherapy PET parameters via spatially resolved regression. Materials and methodsTen canine sinonasal cancer patients underwent PET/CT scans of [18F]FDG (FDGpre), [18F]FLT (FLTpre), and [61Cu]Cu-ATSM (Cu-ATSMpre). Following radiotherapy regimens of 50Gy in 10 fractions, veterinary patients underwent FDG PET/CT scans at 3months (FDGpost). Regression of standardized uptake values in baseline FDGpre, FLTpre and Cu-ATSMpre tumour voxels to those in FDGpost images was performed for linear, log-linear, generalized-linear and mixed-fit linear models. Goodness-of-fit in regression coefficients was assessed by R2. Hypothesis testing of coefficients over the patient population was performed. ResultsMultivariate linear model fits of FDGpre to FDGpost were significantly positive over the population (FDGpost∼0.17·FDGpre, p=0.03), and classified slopes of RECIST non-responders and responders to be different (0.37 vs. 0.07, p=0.01). Generalized-linear model fits related FDGpre to FDGpost by a linear power law (FDGpost∼FDGpre0.93,p<0.001). Univariate mixture model fits of FDGpre improved R2 from 0.17 to 0.52. Neither baseline FLT PET nor Cu-ATSM PET uptake contributed statistically significant multivariate regression coefficients. ConclusionsSpatially resolved regression analysis indicates that pre-treatment FDG PET uptake is most strongly associated with three-month post-treatment FDG PET uptake in this patient population, though associations are histopathology-dependent.