Assessment Of Tumor Treatment Response Research Articles

Automatic segmentation of hepatic metastases on DWI images based on a deep learning method: assessment of tumor treatment response according to the RECIST 1.1 criteria

BackgroundEvaluation of treated tumors according to Response Evaluation Criteria in Solid Tumors (RECIST) criteria is an important but time-consuming task in medical imaging. Deep learning methods are expected to automate the evaluation process and improve the efficiency of imaging interpretation.ObjectiveTo develop an automated algorithm for segmentation of liver metastases based on a deep learning method and assess its efficacy for treatment response assessment according to the RECIST 1.1 criteria.MethodsOne hundred and sixteen treated patients with clinically confirmed liver metastases were enrolled. All patients had baseline and post-treatment MR images. They were divided into an initial (n = 86) and validation cohort (n = 30) according to the examined time. The metastatic foci on DWI images were annotated by two researchers in consensus. Then the treatment responses were assessed by the two researchers according to RECIST 1.1 criteria. A 3D U-Net algorithm was trained for automated liver metastases segmentation using the initial cohort. Based on the segmentation of liver metastases, the treatment response was assessed automatically with a rule-based program according to the RECIST 1.1 criteria. The segmentation performance was evaluated using the Dice similarity coefficient (DSC), volumetric similarity (VS), and Hausdorff distance (HD). The area under the curve (AUC) and Kappa statistics were used to assess the accuracy and consistency of the treatment response assessment by the deep learning model and compared with two radiologists [attending radiologist (R1) and fellow radiologist (R2)] in the validation cohort.ResultsIn the validation cohort, the mean DSC, VS, and HD were 0.85 ± 0.08, 0.89 ± 0.09, and 25.53 ± 12.11 mm for the liver metastases segmentation. The accuracies of R1, R2 and automated segmentation-based assessment were 0.77, 0.65, and 0.74, respectively, and the AUC values were 0.81, 0.73, and 0.83, respectively. The consistency of treatment response assessment based on automated segmentation and manual annotation was moderate [K value: 0.60 (0.34–0.84)].ConclusionThe deep learning-based liver metastases segmentation was capable of evaluating treatment response according to RECIST 1.1 criteria, with comparable results to the junior radiologist and superior to that of the fellow radiologist.

18F-C2Am: a targeted imaging agent for detecting tumor cell death in vivo using positron emission tomography

IntroductionTrialing novel cancer therapies in the clinic would benefit from imaging agents that can detect early evidence of treatment response. The timing, extent and distribution of cell death in tumors following treatment can give an indication of outcome. We describe here an 18F-labeled derivative of a phosphatidylserine-binding protein, the C2A domain of Synaptotagmin-I (C2Am), for imaging tumor cell death in vivo using PET.MethodsA one-pot, two-step automated synthesis of N-(5-[18F]fluoropentyl)maleimide (60 min synthesis time, > 98% radiochemical purity) has been developed, which was used to label the single cysteine residue in C2Am within 30 min at room temperature. Binding of 18F-C2Am to apoptotic and necrotic tumor cells was assessed in vitro, and also in vivo, by dynamic PET and biodistribution measurements in mice bearing human tumor xenografts treated with a TRAILR2 agonist or with conventional chemotherapy. C2Am detection of tumor cell death was validated by correlation of probe binding with histological markers of cell death in tumor sections obtained immediately after imaging.Results18F-C2Am showed a favorable biodistribution profile, with predominantly renal clearance and minimal retention in spleen, liver, small intestine, bone and kidney, at 2 h following probe administration. 18F-C2Am generated tumor-to-muscle (T/m) ratios of 6.1 ± 2.1 and 10.7 ± 2.4 within 2 h of probe administration in colorectal and breast tumor models, respectively, following treatment with the TRAILR2 agonist. The levels of cell death (CC3 positivity) following treatment were 12.9–58.8% and 11.3–79.7% in the breast and colorectal xenografts, respectively. Overall, a 20% increase in CC3 positivity generated a one unit increase in the post/pre-treatment tumor contrast. Significant correlations were found between tracer uptake post-treatment, at 2 h post-probe administration, and histological markers of cell death (CC3: Pearson R = 0.733, P = 0.0005; TUNEL: Pearson R = 0.532, P = 0.023).ConclusionThe rapid clearance of 18F-C2Am from the blood pool and low kidney retention allowed the spatial distribution of cell death in a tumor to be imaged during the course of therapy, providing a rapid assessment of tumor treatment response. 18F-C2Am has the potential to be used in the clinic to assess early treatment response in tumors.

NIMG-73. NEUROLOGICAL ASSESSMENT IN NEURO-ONCOLOGY (NANO) CORRELATES WITH INTEGRATED PREDICTIVE RADIOGRAPHIC BIOMARKERS OF PROGRESSED HIGH GRADE ASTROCYTIC TUMORS (IPROMPT): A PROPOSED SUPPLEMENT TO RANO

Abstract This year nearly 86,000 persons in the US will be diagnosed with a primary brain tumor. Glioblastoma (GBM) is the deadliest and most common of these tumors in adults with overall survival around 15mos. Disease progression is inevitable, and effective second-line therapies are few. Delays in identifying progressed GBM risks neurologic decline, continuation of ineffective therapies, and increased tumor-mutational-burden, potentially rendering treatment less-effective. Brain tumor imaging can be ambiguous and is an active topic in neuro-oncology as there are currently no validated approaches to assessment of tumor treatment response. We present a single institution database including 622 gross-totally-resected GBMs (2006–2018) to evaluate changes in T2/FLAIR signal intensity (T2FSI), diffusion restriction, and contrast enhancement both in and around the resection cavity (RC) to determine earliest indicators of tumor progression. Analysis of NANO score was completed to correlate with predicted tumor progression. This is an early data release of an ongoing two-phased retrospective/prospective study with findings that support a novel integration of radiographic sequences and NANO to supplement established RANO criteria in earlier detection of GBM progression. Of 622 confirmed GBMs, 158 fit the rigorous criteria – included here are 58 cases. We report 70% (n=41) of patients show RC T2FSI correlating with restricted diffusion and increasing NANO score (0.5–1.4) as early as 3.8mos prior to and at time of radiographic progression, respectively. Previous studies have reported similar T2FSI but without associated NANO, less rigid inclusion criteria, and limited MRI sequences. We also report imaging phenomena, not previously described, such as FLAIR migration and FLAIR coring timed closely with progression. The proposed technique uses routine MRI, therefore readily standardized in clinical practice. With further validation, this novel integration of imaging and neurologic assessment might serve to supplement existing response criteria in GBM and improve clinician confidence in earlier detection of progression.

Metastatic renal cell carcinoma imaging evaluation in the era of anti-angiogenic therapies.

During the last decade, the arsenal of anti-angiogenic (AAG) agents used to treat metastatic renal cell carcinoma (RCC) has grown and revolutionized the treatment of metastatic RCC, leading to improved overall survival compared to conventional chemotherapy and traditional immunotherapy agents. AAG agents include inhibitors of vascular endothelial growth factor receptor signaling pathways and mammalian target of rapamycin inhibitors. Both of these classes of targeted agents are considered cytostatic rather than cytotoxic, inducing tumor stabilization rather than marked tumor shrinkage. As a result, decreases in tumor size alone are often minimal and/or occur late in the course of successful AAG therapy, while tumor devascularization is a distinct feature of AAG therapy. In successful AAG therapy, tumor devascularization manifests on computed tomography images as a composite of a decrease in tumor size, a decrease in tumor attenuation, and the development of tumor necrosis. In this article, we review Response Evaluation Criteria in Solid Tumors (RECIST)-the current standard of care for tumor treatment response assessment which is based merely on changes in tumor length-and its assessment of metastatic RCC tumor response in the era of AAG therapies. We then review the features of an ideal tumor imaging biomarker for predicting metastatic RCC response to a particular AAG agent and serving as a longitudinal tumor response assessment tool. Finally, a discussion of the more recently proposed imaging response criteria and new imaging trends in metastatic RCC response assessment will be reviewed.