BACKGROUND Microsatellite instability (MSI) of colorectal cancer (CRC) is an important biomarker for treatment response to immune checkpoint inhibitor therapy. Traditionally, this has been identified using histological methods, but more recently, radiomics techniques have been applied to computed tomography (CT) images showing ability to differentiate MSI tumors. However, the variable techniques and scanner noise may limit broad application of the technique. Identifying visible, macroscopic imaging features may present a more invariant technique to identify MSI tumors on imaging.AIM To identify CT-based imaging features without radiomics to differentiate between MSI status in CRC. METHODS Imaging features of 109 subjects (58 male, 51 female) from 2011 to 2018 were stratified based on MSI status using CT scan images and pathological records. These variables included primary tumor location, primary tumor size, initial cancer stage, metastatic locations, primary attenuation compared to the liver, primary tumor growth pattern, primary tumor content, tumor margin, tumor density, and the presence or absence of mesenteric infiltration for primary tumor. Statistical analysis was performed using chi 2, t tests, and Mann-Whitney U tests. RESULTS Four clinical characteristics were statistically significant for MSI-high (MSI-H). Although the MSI-H mass is larger (P < 0.001), the MSI-H tumors have a lower cancer stage at initial diagnosis (P = 0.012) as the MSI-H CRC tends to have regional metastasis than distant metastasis (P < 0.001). MSI-H CRC tends to be right-sided while MSI-low CRC tends to be left-sided (P < 0.001). The Hounsfield unit SD was lower in the MSI-H than the MSI-low CRC, P = 0.002. The remaining primary imaging characteristics (tumor attenuation relative to the liver, tumor content, tumor growth pattern, and tumor margin) were not statistically different. Both classes of tumors were more likely to be well-defined with mucosal growth and hypoattenuating relative to the liver on portovenous phase imaging. CONCLUSION Four clinical and one imaging characteristics were statistically significant for MSI-H CRC tumors. Larger and prospective trials would be needed to assess clinical applicability of these findings.

265 Background: This study aimed to identify correlations between CT imaging characteristics accessible to the radiologist with microsatellite instability (MSI) status among colorectal cancer. Noninvasive identification of MSI status can decrease time to treatment if pathology can be bypassed. Methods: 109 colorectal cancer patients were identified retrospectively from 2011 to 2018 with an average age of 61 years (58 male, 51 female) and had MSI pathological assessment. These subjects all had CT abdomen and pelvis obtained at the time of initial diagnosis. Imaging features of both the primary and metastatic lesions were assessed. Primary lesion features included: location, size, stage, attenuation compared to the liver, growth pattern, tumor margin, primary mass area, and the presence/absence of mesenteric infiltration. Metastatic lesion features included: size, attenuation relative to liver, and enhancement pattern. Patient age, sex, and tumor staging were obtained from patient medical records. Statistical analysis was performed using various methods, including chi-square, Mann-Whitney and Kruskal-Wallis tests. Results: 5 tumor characteristics displayed a statistically significant relationship with MSI status. MSI-H lesions were more likely to be clinical Stage 2, and MSI-L lesions were more likely to be Clinical Stage 3 (p=0.012). Tumors with distant metastasis were more likely MSI-L versus regional metastasis were more likely MSI-H (p<0.001). Primary mass area was larger in the MSI-H than MSI-L group (p<0.001). Tumors in the left colon are more likely to be MSI-L versus CRC tumors in the right colon are more likely to be MSI-H (p<0.001). MSI-H tumors had a lower primary tumor Hounsfield Unit standard deviation (SD) than MSI-L tumors (p=0.002). The remaining features were not statistically significant. For primary lesion, these include mesenteric infiltration (p=0.189), tumor margin (p=1.000), enhancement pattern (p=0.498), growth pattern (p=0.127), tumor CT density (p=0.162), and liver density (p=0.105). Enhancement of metastasis also had no difference (p=0.376). Conclusions: Multiple CT imaging characteristics were predictive of MSI classification in CRC tumors, holding potential clinical relevance that can help guide individualized therapy. While these results were statistically significant, more research must be done to validate these findings and to derive useful nomograms.

216 Background: Colorectal cancer (CRC) is the second leading cause of cancer-related deaths. We previously found that delayed diagnosis due to lack of radiological identification results in significantly worse outcome for patients. We had developed a rudimentary, AI second observer which demonstrated potential for detecting CRC on routine CT abdomen/pelvis (CTAP). However, the AI algorithm detected many false positives. In this study, we analyzed the data using TCIA as test cases and evaluated whether patient peritoneal fat content influenced the false positive rate. This could serve as a guide for future training of AI second observer to minimize false positive detection. Methods: 2D U-Net convolutional neural network (CNN) containing 31 million trainable parameters was trained with 58 CRC CT images from Banner MD Anderson (AZ) and MD Anderson Cancer Center (TX) (51 used for training and 7 for validation) and 59 normal CT scans from Banner MD Anderson Cancer Center. 18 of the 25 CRC cases from public domain data (The Cancer Genome Atlas) were used to evaluate the performance of the models (5 had no identifiable cancer and 2 were rejected for having no contrast). The CRC was segmented using ITK-SNAP open-source software (v. 3.8). To apply the deep ensemble approach, five CNN models were trained independently with random initialization using the same U-Net architect and the same training data. Given a testing CT scan, each of the five trained CNN models was applied to produce tumor segmentation for the testing CT scan. The tumor segmentation results produced by the trained CNN models were then fused using a simple majority voting rule (up to 2 voters) to produce consensus tumor segmentation results. The segmentation was analyzed for the number and location of false positives per case. The peritoneal fat content was classified at the level of aortic bifurcation by the distance of fat between adjacent small bowel loops (≤ or > 1 cm). Chi-square test was performed testing fat volume and number of voters as the intervention. Results: Our results showed that the higher volume of peritoneal fat (> 1 cm, N=6) decreases the rate of false positive compared with low volume (≤ 1 cm, N=12, p=0.013). When comparing between having one voter and two voter ensemble using low fat volume data, two voter ensemble also decreased the number of false positives but not statistically significant (p=0.286). Conclusions: Our results show that AI-based second observer generates more false positives when patients have lower peritoneal fat volume; this implies that future training may require higher percentage of cases with low peritoneal fat to improve second observer precision. Our analysis also showed that increasing the number of voter in the ensemble also decrease the number of false positives per case.

643 Background: Although modern cancer treatment has significantly improved overall survival, this comes with a high financial cost. Based on an NIH report on Financial Burden of Cancer Care, treating pancreatic adenocarcinoma (PDAC) costs $108K, $18K, and $125K for initial treatment, during follow-up, and in last year of life, respectively. During the course of treatment, patients will visit the hospital for acute symptoms for which cross-sectional imaging is obtained within 45 days of subsequent scheduled follow-up imaging. This presents an opportunity for cost saving if data could be collected to identify an algorithm to avoid unnecessary duplication of imaging in a short interval. The purpose of this pilot study is to evaluate tumor growth trajectory within this short time span of 45 days to guide future trials to clarify the circumstances for which a subsequent follow up study is necessary. Methods: We retrospectively identified patients from our tumor registry from 2016 to 2019 who were diagnosed with metastatic PDAC (mPDAC). Subjects were included for analysis if they met the following criteria: 1) diagnosed with mPDAC; 2) had at least 2 CT abdomen/pelvis (CTAP) within 45 days of each other with one of the studies being obtained in the emergency department; and 3) had at least 1-3 measurable lesions as defined by RECIST 1.1 criteria. Tumor dimension was measured on axial images with the largest cross-section. For our primary outcome we used paired t-tests to evaluate the difference in mean size of all lesions (primary and metastatic). We also performed sub-group analyses comparing mean size of primary and metastatic lesions separately. An alpha level of 0.05 was used to determine statistical significance. Results: We screened 383 subjects from the tumor registry and identified 28 subjects suitable for our study. 18 subjects were female and 10 were male. We identified 69 target lesions, 23 of which were primary and 46 were metastatic. The mean sum of the largest dimension of target lesions was 2.82 cm on the initial study and 3.23 on the latter study (p<0.001). The mean sum of the largest dimension of the primary lesion was 3.62 cm on the initial study and 3.89 on the latter study (p=0.01). For metastatic lesions, the initial sum was 2.38 cm while the latter was 2.85 (p<0.001). Conclusions: Our finding of increase in tumor size in 45 days for mPDAC suggests that mPDAC is aggressive and can grow even in short time span. This suggests that if progression is identified on a study obtained within 45 days of scheduled follow-up, the follow-up study could be cancelled to avoid added cost to the patient. However, if the earlier study is stable or shows response, the scheduled follow-up may still be necessary to demonstrate continued stability/response.

BACKGROUNDMissing occult cancer lesions accounts for the most diagnostic errors in retrospective radiology reviews as early cancer can be small or subtle, making the lesions difficult to detect. Second-observer is the most effective technique for reducing these events and can be economically implemented with the advent of artificial intelligence (AI). AIMTo achieve appropriate AI model training, a large annotated dataset is necessary to train the AI models. Our goal in this research is to compare two methods for decreasing the annotation time to establish ground truth: Skip-slice annotation and AI-initiated annotation. METHODSWe developed a 2D U-Net as an AI second observer for detecting colorectal cancer (CRC) and an ensemble of 5 differently initiated 2D U-Net for ensemble technique. Each model was trained with 51 cases of annotated CRC computed tomography of the abdomen and pelvis, tested with 7 cases, and validated with 20 cases from The Cancer Imaging Archive cases. The sensitivity, false positives per case, and estimated Dice coefficient were obtained for each method of training. We compared the two methods of annotations and the time reduction associated with the technique. The time differences were tested using Friedman's two-way analysis of variance. RESULTSSparse annotation significantly reduces the time for annotation particularly skipping 2 slices at a time (P < 0.001). Reduction of up to 2/3 of the annotation does not reduce AI model sensitivity or false positives per case. Although initializing human annotation with AI reduces the annotation time, the reduction is minimal, even when using an ensemble AI to decrease false positives. CONCLUSIONOur data support the sparse annotation technique as an efficient technique for reducing the time needed to establish the ground truth.

693 Background: Although modern cancer treatment has significantly prolonged patient’s life, it comes with a high financial cost. Based on an NIH report on Financial Burden of Cancer Care, treating PDAC costs $108K, $18K, and $125K for initial treatment, during follow up, and in last year of life, respectively. Of these, CA19-9 and CT studies are obtained concurrently to evaluate tumor burden whose cost can be reduced by decreasing imaging frequency when CA19-9 remains stable. As a pilot study, we evaluated whether changes in CA19-9, an inexpensive laboratory study commonly used for monitoring treatment response, correlates with tumor size change on imaging studies. A positive correlation may lead us to design future studies to evaluate whether stable CA19-9 can serve as an indicator for stable disease on CT scan such that imaging frequency (and cost) may be reduced. Methods: We retrospectively identified PDAC patients from our tumor registry from 2016 to 2019. Subjects were included if they met the following criteria: 1) diagnosed with PDAC 2) had at least 2+ CA19-9 and CT/MR follow up assessments with the laboratory and imaging studies within one month of each other, and 3) had measurable lesions as defined by RECIST 1.1 criteria. Due to limited resources, we limited the maximum number of lesions to 4 instead of the usual 5 for RECIST 1.1. Two-dimensional orthogonal tumor sizes were measured on axial images with the largest cross-section. Changes in the values were obtained serially (from the immediately prior study). Nonparametric correlations (Spearman’s rho) were used to evaluate monotonicity. The significance level was set at p < 0.05. Results: We screened 300 subjects from the tumor registry and identified 23 subjects suitable for our study. 13 subjects were female and 10 were male with ages ranging from 35 to 84. We identified 55 target lesions and 128 assessments (with both CA19-9 and imaging) from the 23 subjects. We analyzed the log absolute and log relative change in CA19-9 with change in tumor area. When absolute CA19-9 value, absolute change in CA19-9, and relative change in CA19-9 were tested against the absolute change in tumor size, the correlation was significant for the absolute value and absolute change of CA19-9 (p < 0.01), but not for relative change of CA19-9 (p < 0.11). When tested against relative change in tumor size, all three values were highly significant (p < 0.001). Conclusions: Our finding of direct correlations between changes of CA19-9 and relative change of tumor size on imaging suggests that CA19-9 can serve as a surrogate measure of relative change of tumor burden for patients undergoing treatment. This provides motivation for future studies evaluating the possibility that stable CA19-9 can represent stable tumor burden such that expensive imaging studies may be obtained at longer time intervals to reduce financial cost to patients.

141 Background: Colorectal cancer (CRC) is the second leading cause of cancer-related deaths, and survival can be improved if early, suspect imaging features on CT of the abdomen and pelvis (CTAP) can be routinely identified. At present, up to 40% of these features are undiagnosed on routine CTAP, but this can be improved with a second observer. In this study, we developed a deep ensemble learning method for detecting CRC on CTAP to determine if increasing agreement between ensemble models can decrease the false positives detected by artificial intelligence (AI) second-observer. Methods: 2D U-Net convolutional neural network (CNN) containing 31 million trainable parameters was trained with 58 CRC CT images from Banner MD Anderson (AZ) and MD Anderson Cancer Center (TX) (51 used for training and 7 for validation) and 59 normal CT scans from Banner MD Anderson Cancer Center. 20 of the 25 CRC cases from public domain data (The Cancer Genome Atlas) were used to evaluate the performance of the models. The CRC was segmented using ITK-SNAP open-source software (v. 3.8). To apply the deep ensemble approach, five CNN models were trained independently with random initialization using the same U-Net architect and the same training data. Given a testing CT scan, each of the five trained CNN models was applied to produce tumor segmentation for the testing CT scan. The tumor segmentation results produced by the trained CNN models were then fused using a simple majority voting rule to produce consensus tumor segmentation results. The segmentation was analyzed by the percentage of correct detection, the number of false positives per case, and the Dice similarity coefficient (DSC). If parts of the CRC were flagged by AI, then it was considered correct. A detection was considered false positive if the marked lesion did not overlap with any CRC; contiguous false positives across different slices of CT image were considered a single false positive. DSC measures the quality of the segmentation by measuring the overlap between the ground-truth and AI detected lesion. Results: Our results showed that increasing the agreement between the 5 models dramatically decreases the number of false positives per CT at the expense of slight decrease in accuracy and DSC. This is described in the table. Conclusions: Our results show that AI-based second observer can potentially detect CRC on routine CTAP. Although the initial result yields high false positives per case, ensemble voting is an effective method for decreasing the false positives with a slight decrease in accuracy. This technique can be further improved for eventual clinical application.[Table: see text]

142 Background: Colorectal cancer (CRC) is the second leading cause of cancer-related deaths, and its outcome can be improved with better detection of incidental early CRC on routine CT of the abdomen and pelvis (CTAP). AI-second observer (AI) has the potential as shown in our companion abstract. The bottleneck in training AI is the time required for radiologists to segment the CRC. We compared two techniques for accelerating the segmentation process: 1) Sparse annotation (annotating some of the CT slice containing CRC instead of every slice); 2) Allowing AI to perform initial segmentation followed by human adjustment. Methods: 2D U-Net convolutional neural network (CNN) containing 31 million trainable parameters was trained with 58 CRC CT images from Banner MD Anderson (AZ) and MD Anderson Cancer Center (TX) (51 used for training and 7 for validation) and 59 normal CT scans from Banner MD Anderson Cancer Center. Twenty of the 25 CRC cases from public domain data (The Cancer Genome Atlas) were used to evaluate the performance of the models. The CRC was segmented using ITK-SNAP open-source software (v. 3.8). For the first objective, 3 separate models were trained (fully annotated CRC, every other slice, and every third slice). The AI-annotation on the TCGA dataset was analyzed by the percentage of correct detection of CRC, the number of false positives, and the Dice similarity coefficient (DSC). If parts of the CRC were flagged by AI, then it was considered correct. A detection was considered false positive if the marked lesion did not overlap with CRC; contiguous false positives across different slices of CT image were considered a single false positive. DSC measures the quality of the segmentation by measuring the overlap between the ground-truth and AI detected lesion. For the second objective, the time required to adjust the AI-produced annotation was compared to the time required for annotating the entire CRC without AI assistance. The AI-models were trained using ensemble learning (see our companion abstract for details of the techniques). Results: Our results showed that skipping slices of tumor in training did not alter the accuracy, false positives, or DSC classification of the model. When adjusting the AI-observer segmentation, there was a trend toward decreasing the time required to adjust the annotation compared to full manual segmentation, but the difference was not statistically significant (Table; p=0.121). Conclusions: Our results show that both skipping slices of tumor as well as starting with AI-produced annotation can potentially decrease the effort required to produce high-quality ground truth without compromising the performance of AI. These techniques can help improve the throughput to obtain a large volume of cases to train AI for detecting CRC.[Table: see text]

Objective: Pancreatic ductal adenocarcinoma (PDAC) has the best survival when detected early with 5-year survival near 40% for small, resectable PDAC. We evaluate the undiagnosed PDAC imaging features on routine CT and their impact on resectability. Methods: 76 of the screened 134 CTs from 1/1/2012 to 12/31/2018 using our tumor registry were obtained prior to PDAC diagnosis for other indications at least one month before presentation. Each cross-sectional study was reviewed for features of early PDAC: pancreatic mass, pancreatic ductal dilatation, perivascular/peripancreatic soft-tissue infiltration, omental lesions/ascites, and lymphadenopathy. When such features were detectible by the reviewing radiologists, the original CT readings were classified as concordant/discrepant. Descriptive statistics are reported for discrepant reads, tumor resectability, and tumor size. Results: Of the 76 cases from 46 unique subjects (30 male/16 female), 25 CTs (33%) had undetected PDAC imaging features: masses (15/19 unreported), ductal dilatation (16/20 unreported), and peripancreatic/perivascular soft-tissue infiltration (20/36 unreported). 63% of early PDAC features were not identified initially. One year before clinical diagnosis, 75-80% of the PDAC cases were resectable; at < 6 months before clinical diagnosis, only 29% were resectable. Conclusion: Improving early detection of key PDAC features on routine CT examinations can potentially improve patient outcomes.

Background Skeletal metastases (SM) in advanced pancreatic ductal adenocarcinoma (PDAC) is an infrequent occurrence that has been previously reported in literature to occur in less than 2.5% of the cases. Complications such as pathological fractures can result in intractable pain, immobilization and a significant deterioration in quality of life. The purpose of this study is to improve the understanding of the increasing incidence of SM and the importance of surveillance and adequate management of SM in these patients. Methods A retrospective analysis was conducted using a clinical database at a single tertiary care institution for cancer patients; this included 207 patients with advanced PDAC diagnosed between December 2004 and March 2017 receiving palliative chemotherapy. SM were identified by computerized tomography (CT)/fluorodeoxyglucose positron emission tomography (FDG-PET)/magnetic resonance imaging (MRI). Information regarding demographics, clinical course and date of last follow-up/death were collected. After a median follow-up of 11 months, an analysis was conducted, including a Kaplan-Meier survival analysis. Results The study included 207 patients; 19 out of 207 patients (9.2%) developed SM; the primary tumor was located in the pancreatic body/tail in 12 out of 19 patients (63.2%). The thoracic and lumbar vertebrae were the most common sites of SM. Other common synchronous sites of metastases included the liver and lung. A majority of the lesions were osteolytic (63.2%). The median time of diagnosis from the initial diagnosis was 2 months (range, 0-60 months). Bone pain was observed as the initial symptom in 7 out of 19 patients (36.8%), 2 out of 19 patients (10.5%) had a pathological fracture and 1 out of 19 patients (5.3%) developed a para-spinal mass causing inferior vena cava compression. The median survival period for patients with SM was 11 months (range, 0-62 months) and for those without SM was 12 months (range, 0-147 months) [hazard ratio (HR) 1.24, 95% confidence interval (CI): 0.66-2.30, P=0.51]. Conclusions There has been a challenge with regards to management of the increasing number of patients with SM. Thoracic and lumbar vertebrae are the most common sites and pathological fractures in these sites can be catastrophic. Careful evaluation of skeletal signs and symptoms, early detection and intervention are essential to prevent morbidity and mortality from complications in patients with PDAC and SM.

380 Background: Pancreatic ductal adenocarcinoma (PDAC) is the third leading cause of cancer-related deaths. At present, the best 5-year survival is 25% for resectable PDAC. For small (1 cm) stage 1 PDAC, resection has resulted in much better survival. The goal of this study was to evaluate the appearance and location of early undiagnosed PDAC on computed tomography scans (CT) prior to diagnosis with the goal of minimizing missing early PDAC. We also categorize the errors as either perceptive or cognitive. Methods: PDAC cases were retrospectively reviewed from 1/1/2012 through 12/31/2018 from our tumor registry, identifying 81 cases with paired CT scans both at the time of and prior to diagnosis. Among these, 31 contained imaging features considered diagnostic or suspicious for early PDAC(38%). These “errors” were classified by radiologic features and as well as by location. In addition, errors were classified into “perceptive errors" when the first study was read as normal, and as “cognitive errors” when the report noted an abnormality but failed to note suspicion for malignancy. Results: Among the 31 undiagnosed PDAC, 18 had features of an identifiable mass (58%), 9 had pancreatic ductal dilatation (29%), and 4 had evidence of perivascular soft tissue (13%). 44% of undiagnosed tumors were located in the head-neck, 39% in the body, and 17% in the tail. Perceptive errors were found in 58% and 42% were cognitive. No significant differences were seen between perceptive and cognitive errors based on suspicious features. Conclusions: Radiologic findings of early PDAC was retrospectively evident in more than one third of cases in which prior imaging was performed. These findings are most often masses or ductal dilatation. Location of these undiagnosed tumors were distributed throughout the gland. This study identifies the radiologic features of undiagnosed PDAC which may provide an opportunity for future prospective studies and improved technology which may improve early detection of pancreatic cancer.

68 Background: Artificial intelligence (AI) can potentially improve patient care by assisting physicians as demonstrated with the recent approvals for technologies that detect intracranial hemorrhage on CT exams of the head. AI also has potential for assisting early detection of colorectal cancer (CRC) on routine CT abdomen and pelvis (CTAP). We assessed the difference in detection of abnormal colonic findings between an expert reader (JC) and amateur readers (ARs). Methods: ARs consisting of two third year medical students (RB - AR1, ZG - AR2) studied 20 CTAP for tracing the colon and identifying pathology. Their search pattern and assessment of the colon was then evaluated by an expert radiologist (JC). They then spent two hours reviewing abnormalities in 10 scans with JC, who highlighted suspicious neoplastic findings such as colonic wall thickening, fat stranding, edema, masses, and abnormal lymph nodes. The ARs then individually read 203 CTAP scans to assess for these suspicious findings. The studies were from a single institution and were reported in a prior study in 2019 GI-ASCO. The findings of the ARs were then compared to those of the expert reader and the initial reader for each study. Data was analyzed using t-test with 2 tails. Results: The incidence of suspicious neoplastic findings was 87% and 81% for AR1 and AR2, respectively, compared to 18% in the initial reads and 33% for expert reader (p=0.01). Greatest discordance were 94% and 87% between AR1 and AR2 to the initial reads. Additionally, the incidence of suspicious findings between the first and last 20 cases (p=0.03 and 0.17) examined by ARs declined from 79 to 40% for AR1 and 69 to 55% for AR2. Conclusions: ARs are capable of detecting CRC features on CTAP from ED, but with higher false-positive (FP) rate than trained experts. The FP rate decreases with increasing experience. ARs learning course simulates AI which will likely yield high FP rate with initial training, but with improving FP with deep training, especially with larger volume of normal variants.

Magnetic resonance spectroscopic imaging helps to determine abnormal brain tissue conditions by evaluating metabolite concentrations. Although a powerful technique, it is underutilized in routine clinical studies because of its long scan times. In this study, we evaluated the feasibility of scan time reduction in metabolic imaging using compressed-sensing-based MR spectroscopic imaging in pediatric patients undergoing routine brain exams. We retrospectively evaluated compressed-sensing reconstructions in MR spectroscopic imaging datasets from 20 pediatric patients (11 males, 9 females; average age: 5.4±4.5 years; age range: 3 days to 16 years). We performed retrospective under-sampling of the MR spectroscopic imaging datasets to simulate accelerations of 2-, 3-, 4-, 5-, 7- and 10-fold, with subsequent reconstructions in MATLAB. Metabolite maps of N-acetylaspartate, creatine, choline and lactate (where applicable) were quantitatively evaluated in terms of the root-mean-square error (RMSE), peak amplitudes and total scan time. We used the two-tailed paired t-test along with linear regression analysis to statistically compare the compressed-sensing reconstructions at each acceleration with the fully sampled reference dataset. High fidelity was maintained in the compressed-sensing MR spectroscopic imaging reconstructions from 50% to 80% under-sampling, with the RMSE not exceeding 3% in any dataset. Metabolite intensities and ratios evaluated on a voxel-by-voxel basis showed no statistically significant differences and mean metabolite intensities showed high correlation compared to the fully sampled reference dataset up to an acceleration factor of 5. Compressed-sensing MR spectroscopic imaging has the potential to reduce MR spectroscopic imaging scan times for pediatric patients, with negligible information loss.

Purpose: Colorectal cancer (CRC) is the third most common cancer in the United States, and prognosis is greatly influenced by stage at diagnosis.Early colorectal cancer can be subtle on CT scans showing only mild wall thickening, small polyps, or subtle lymph nodes in atypical draining location.Identifying these lesions on CT scan performed for nonspecific symptoms can help identify interval CRC and improve patient outcome.The purpose of the present study is to classify the undetected CRC on abdominal CT scan by their imaging features and whether early identification can downstage CRC patients. Materials and methods:A retrospective analysis was conducted of patients (pts) diagnosed with CRC and receiving treatment or sought second opinion at Banner MD Anderson Cancer Center.Data collection included age, gender, ECOG, KRAS mutation status, and overall survival (OS).CT imaging was obtained from the time of diagnosis, as well as any prior abdominal imaging available.Images were reviewed for multiple CT features including appearance of mass, mesenteric infiltration, abnormal draining lymph nodes, contrast enhancement relative to adjacent mucosa, and intralesional calcifications.Staging was evaluated using available clinical note and CT scan, based on the TNM staging system for CRC. Results:The 41 pts with 51 prediagnostic CTs from 1/1/2012 -12/31/2015 had mean age of 68 years (range:44-90) Mean ECOG status for the population was 1.46.41% of the prediagnostic CTs had undetected findings.52 and 43 % of the undetected findings were in the rectosigmoid and ascending colon respectively.Of the 15 undetected masses, 9 appeared as asymmetric wall thickening, 3 as concentric wall thickening, and 3 as polyps.Of the 14 undetected lymph node groups, 2 were excluded due to stability or nonrelated condition.The remaining lymph nodes were found in the associated draining station and averaged 3±1.2 mm in size.On average, the stage at prediagnostic CT was 3A and the diagnostic CT was 3C (p=0.0015).Average time lapse between prediagnostic and diagnostic CT was 21 months (3-64 months). ConclusionOur study demonstrated that high percentage of early-stage CRC findings are undetected on abdominal CT due to their subtle feature, with most undetected location in the rectosigmoid and ascending colon.In general, these subtle features predate the actual diagnosis by up to two years.Early detection of CRC can improve survival by lowering the stage from 3C to 3A, thus providing 36% improvement in 5-year survival.A dedicated search can be performed on the abdominal CT to improve detection by specifically looking for polyps, wall thickening, and small lymph nodes in the draining station.

510 Background: We have previously reported that up to 48% of the early features of colorectal cancer (subtle wall thickening, pericolonic stranding, and small lymph nodes in the draining nodal station) were not identified on the original CT abdomen and pelvis (CTAP) reports. This resulted in a 36% decrease in five-year survival based on historical data. In this report, we assessed whether dedicated assessment of the colon on routine CT scans could lead to early detection of colorectal cancer. Methods: 210 CTAPs over a three-month period were screened from the emergency room records at a tertiary care hospital. 194 scans met eligibility. Exclusion criteria included: cases known to the evaluating radiologist and age ≤ 19 or > 89 years. No study was excluded for suboptimal image quality. The original report was reviewed for abnormalities involving the colon, mesentery and bowel and was recorded. A blinded evaluation of the eligible case was then performed by a board-certified radiologist with attention specifically to the colon and the mesentery for the suspicious early features of CRC. The concordance and discordance was then tabulated. Discordant findings were re-evaluated to determine if the discordance was true. Results: 72/194 patients were male, median age 44.5 years (range 20 - 89). 55/194 patients (29.1%) included in the study were noted to have suspicious features. 26 had abnormal lymph nodes, 24 had abnormal colonic wall thickening and 16 had pericolonic stranding and/or wall edema. 45/55 studies were truly discordant from the original interpretation. These included one missed colorectal cancer (confirmed), one likely small bowel neuroendocrine tumor (no follow up), and one likely transitional cell carcinoma of the right renal pelvis (no follow up). Conclusions: Dedicated search of the colon and mesentery on CTAP can identify subtle findings, although their true relevance is being evaluated in a larger future study. Our observational data does indicate that there maybe a potential role for a focused evaluation of the colon and mesentery on routine CTAP in an attempt to potentially increase the rate of cancer detection especially in younger low-average risk patients.

233 Background: Pancreatic cancer has significant mortality at five years, even in resectable disease. Recent effort has been dedicated to identify a more specific tumor marker for screening and assessing tumor response, in part because 10% of the population does not produce CA19-9. Detection and quantification of circulating tumor DNA (ctDNA) in the bloodstream is a novel concept for screening and treatment response assessment. This study examined possible correlations between ctDNA levels and various aspects of pancreatic cancer in a total of 17 patients receiving treatment at Banner MD Anderson Cancer Center. Methods: Present study is approved by our local IRB. Research was conducted according HIPPA regulation. Subjects on the present study were obtained from the list of patients participating in our ctDNA trial. A total of 17 subjects were identified from the list. Their ctDNA index levels were obtained from the sponsor (Chronyx) at baseline and following each cycle of the treatment. For each CT scan and each ctDNA study, they are considered the same time if they are obtained within 4 weeks of each other. The sizes of the primary tumor and the largest metastatic lesion were on a transverse image at the largest extent of the lesion. Data was analyzed with Spearman's correlation. Results: Baseline ctDNA levels did not correlate with patient demographic data (N = 17; gender, p = 0.63; age, p = 0.82), baseline size of primary mass on CT scan (N = 16; p = 0.85), baseline vessel involvement on CT Scan (N = 17; p = 0.58), presence of metastasis on CT scan (N = 17; p = 0.78), size of largest metastasis on CT scan (N = 17; p = 0.85), presence of peripancreatic lymph nodes on CT scan (N = 17; p = 0.45), or overall survival (N = 8; p = 0.6). However, there is a trend toward correlating the change in ctDNA and change in size on CT scan following treatment (N = 7; p = 0.12). Conclusions: ctDNA at baseline appears to be secreted independent of the primary tumor size, location, or presence of metastasis. However, changes in ctDNA does seem to correlate with changes in tumor size following treatment. Although our data was not statistically significant, this may be related to the low sample size. With larger sample size, it is expected that changes in ctDNA may prove to correlate with changes in tumor size.

Magnetic resonance spectroscopic imaging (MRSI) is an important technique for assessing the spatial variation of metabolites in vivo. The long scan times in MRSI limit clinical applicability due to patient discomfort, increased costs, motion artifacts, and limited protocol flexibility. Faster acquisition strategies can address these limitations and could potentially facilitate increased adoption of MRSI into routine clinical protocols with minimal addition to the current anatomical and functional acquisition protocols in terms of imaging time. Not surprisingly, a lot of effort has been devoted to the development of faster MRSI techniques that aim to capture the same underlying metabolic information (relative metabolite peak areas and spatial distribution) as obtained by conventional MRSI, in greatly reduced time. The gain in imaging time results, in some cases, in a loss of signal-to-noise ratio and/or in spatial and spectral blurring. This review examines the current techniques and advances in fast MRSI in two and three spatial dimensions and their applications. This review categorizes the acceleration techniques according to their strategy for acquisition of the k-space. Techniques such as fast/turbo-spin echo MRSI, echo-planar spectroscopic imaging, and non-Cartesian MRSI effectively cover the full k-space in a more efficient manner per TR . On the other hand, techniques such as parallel imaging and compressed sensing acquire fewer k-space points and employ advanced reconstruction algorithms to recreate the spatial-spectral information, which maintains statistical fidelity in test conditions (ie no statistically significant differences on voxel-wise comparisions) with the fully sampled data. The advantages and limitations of each state-of-the-art technique are reviewed in detail, concluding with a note on future directions and challenges in the field of fast spectroscopic imaging.

261 Background: Cell-free tumor DNA (cfDNA) has potential to provide minimally invasive patient specific biomarkers to monitor tumor burden. Tumor-specific copy number instability (CNI) are used to quantify tumor-derived cfDNA in the plasma. We prospectively computed CNI Scores of cfDNA to compare with radiological and Ca 19-9 responses. Methods: In a laboratory blinded, prospective single-institution study, 119 plasma samples from 33 patients (pts) with PDAC were analyzed. Time-points were at baseline (C1), 2nd (C2) and 3rd (C3) cycle of systemic therapy. Tumor cfDNA was measured with a CNI scoring assay that quantifies cfDNA with somatic macro-alterations. CNI Score (CNIs) of 31 was defined as ref. range (97.5 % - control group; N = 135), pts below this threshold were censored. Progression of disease (PD) defined as C3 CNIs > 93 (3-fold threshold) and difference to the baseline > 31 (dispersion of reference population). Mutant KRAS in plasma was measured using ddPCR in a subset 22 pts. Pts with an increase of > 0.06% (critical difference) were classified PD. Radiologic imaging results were compared with CNIs and CA19-9 changes from baseline to C3, respectively. Results: By standard radiological imaging 33 pts were classified as: 14 PR/CR, 10 SD, 9 PD. 27/33 pts (81%) were evaluable by CNIs which ranged from decrease of 2017 - increase of 645. The C3 CNI classifier yielded a sensitivity of 86% for predicting PD and 95% for SD/PR/CR. KRAS classification yielded an accuracy of 72%, and only 3/9 PD were accurately predicted (33%). 26/33 pts were secretors evaluable by CA19-9. Only 2/8 PD pts showed increasing values in CA19-9 and 6/8 of evaluable by CNI; 5/6 showed increasing CNI scores. 3/20 deemed as SD on imaging with increasing values of CA19-9 were noted to have decreasing CNIs. CNI Score classification was significantly better than CA19-9 (P = 0.001 and 0.63, respectively). Conclusions: Our evaluation of a comparative study on cfDNA and CA19-9 versus imaging suggest that CNI quantification is potentially a more reliable blood-based marker for early assessment of efficacy to systemic therapy in PDAC. Furthermore, for patients not expressing CA19-9 it could serve as an alternative monitoring aid.

593 Background: Prognosis of colorectal cancer (CRC) is greatly influenced by stage at diagnosis. Early colorectal cancer can be subtle on CT scans showing only mild wall thickening, small polyps, or subtle lymph nodes. Identifying these lesions on CT performed for nonspecific symptoms can help identify interval CRC and improve patient outcome. The purpose of the present study is to classify missed CRC on abdominal CT by their imaging features and whether early identification can downstage CRC patients. Methods: A retrospective analysis was conducted of patients (pts) diagnosed with CRC. Data collection included age, gender, ECOG, KRAS mutation status, overall survival (OS). CT obtained prior to and at diagnosis were evaluated. Images were reviewed for multiple CT features including appearance of mass, mesenteric infiltration, abnormal draining lymph nodes, contrast enhancement relative to adjacent mucosa, and intralesional calcifications. Staging was evaluated using available CT scan and based on the TNM staging system for CRC. Results: The 41 pts with 51 prediagnostic CTs from 1/1/2012 - 12/31/2015 had mean age of 68 years (range:44-90 ) Mean ECOG status for the population was 1.46. 41% of the prediagnostic CTs had missed findings. 52 and 43 % of the missed findings were in the rectosigmoid and ascending colon respectively. Of the 15 missed masses, 9 appeared as asymmetric wall thickening, 3 as concentric wall thickening, and 3 as polyps. Of the 14 missed lymph node groups, 2 were excluded due to stability or nonrelated condition. The remaining lymph nodes were found in the associated draining station and averaged 3±1.2 mm in size. On average, the stage at prediagnostic CT was 3A and the diagnostic CT was 3C (p = 0.0015). Average time lapse between prediagnostic and diagnostic CT was 21 months (3-64 months). Conclusions: High percentage of CRC findings are missed on abdominal CT due to their subtle feature, with most misses in the rectosigmoid and ascending colon. A dedicated search can improve detection by specifically looking for polyps, wall thickening, and small lymph nodes in the draining station. Early detection of CRC can improve survival by lowering the stage from 3C to 3A, thus providing 36% improvement in 5-year survival.

TPS871 Background: Except for MSI-H tumors, CRC does not respond to immunotherapy. CRC does respond to the graft vs. tumor (GVT) immune effect that occurs after allogeneic stem cell transplantation. GVT is associated with GVHD toxicity which limits the clinical application. A bioengineered allograft (BAG) has been developed which can elicit host-mediated GVT-like effects without GVHD toxicity, chemotherapy conditioning or a HLA-matched donor. BAG are Th1 memory cells derived from blood of healthy donors with CD3/CD28 microbeads attached. These cells have immunomodulatory properties which enable modulation of Th1/Th2 balance and dysregulation of immunosuppressive circuits. We are evaluating the safety and efficacy of BAG in third-line mCRC. Methods: The study uses a standard 3+3 design followed by an expansion phase with the optimal dosing pattern. The protocol has four components: (A) priming; (B) in-situ vaccination; (C) extravasation and trafficking; and (D) counter immune suppression/avoidance. Priming involves intradermal injections of BAG cells which activates NK cells and develops allo-specific Th1/Tc1 immunity. In-Situ vaccination involves tumor cryoablation to release endogenous HSP which chaperone tumor neoantigens, followed immediately by the intralesional injection of BAG cells as adjuvant. Released HSP are engulfed and processed by immature dendritic cells (DC) attracted to the tissue damage. The inflammatory microenvironment created by the BAG and the subsequent allo-rejection response amplified by the priming induces DC maturation. These DC display processed tumor antigens on upregulated MHCI/II and express co-stimulatory CD80/86 enabling priming of a tumor-specific Th1/Tc1 response. CD40L and interferon-gamma expressed by BAG activates allo-specific and tumor-specific memory cells upon intravenous infusion, permitting trafficking to tumor. The host rejection of BAG releases endogenous danger signals creating a sustained systemic inflammatory cytokine release which serves to counter-regulate immunosuppressive mechanisms. Longitudinal CT scans biopsies, PBMC and serum samples are collected for analysis to verify immune events within each phase of the protocol. Clinical trial information: NCT02380443.