Multi-feature Multi-Scale CNN-Derived COVID-19 Classification from Lung Ultrasound Data.

The global pandemic of the novel coronavirus disease 2019 (COVID-19) has put tremendous pressure on the medical system. Imaging plays a complementary role in the management of patients with COVID-19. Computed tomography (CT) and chest X-ray (CXR) are the two dominant screening tools. However, difficulty in eliminating the risk of disease transmission, radiation exposure and not being cost-effective are some of the challenges for CT and CXR imaging. This fact induces the implementation of lung ultrasound (LUS) for evaluating COVID-19 due to its practical advantages of noninvasiveness, repeatability, and sensitive bedside property. In this paper, we utilize a deep learning model to perform the classification of COVID-19 from LUS data, which could produce objective diagnostic information for clinicians. Specifically, all LUS images are processed to obtain their corresponding local phase filtered images and radial symmetry transformed images before fed into the multi-scale residual convolutional neural network (CNN). Secondly, image combination as the input of the network is used to explore rich and reliable features. Feature fusion strategy at different levels is adopted to investigate the relationship between the depth of feature aggregation and the classification accuracy. Our proposed method is evaluated on the point-of-care US (POCUS) dataset together with the Italian COVID-19 Lung US database (ICLUS-DB) and shows promising performance for COVID-19 prediction.

Liver disease classification from ultrasound using multi-scale CNN.

PURPOSE: Ultrasound (US) is the preferred modality for fatty liver disease diagnosis due to its noninvasive, real-time, and cost-effective imaging capabilities. However, traditional B-mode US is qualitative, and therefore, the assessment is very subjective. Computer-aided diagnostic tools can improve the specificity and sensitivity of US and help clinicians to perform uniform diagnoses. METHODS: In this work, we propose a novel deep learning model for nonalcoholic fatty liver disease classification from US data. We design a multi-feature guided multi-scale residual convolutional neural network (CNN) architecture to capture features of different receptive fields. B-mode US images are combined with their corresponding local phase filtered images and radial symmetry transformed images as multi-feature inputs for the network. Various fusion strategies are studied to improve prediction accuracy. We evaluate the designed network architectures on B-mode in vivo liver US images collected from 55 subjects. We also provide quantitative results by comparing our proposed multi-feature CNN architecture against traditional CNN designs and machine learning methods. RESULTS: Quantitative results show an average classification accuracy above 90% over tenfold cross-validation. Our proposed method achieves a 97.8% area under the ROC curve (AUC) for the patient-specific leave-one-out cross-validation (LOOCV) evaluation. Comprehensive validation results further demonstrate that our proposed approaches achieve significant improvements compared to training mono-feature CNN architectures ([Formula: see text]). CONCLUSIONS: Feature combination is valuable for the traditional classification methods, and the use of multi-scale CNN can improve liver classification accuracy. Based on the promising performance, the proposed method has the potential in practical applications to help radiologists diagnose nonalcoholic fatty liver disease.

Plasma trial: Pilot randomized clinical trial to determine safety and efficacy of plasma transfusions.

BACKGROUND: Plasma is frequently administered to patients with prolonged INR prior to invasive procedures. However, there is limited evidence evaluating efficacy and safety. STUDY DESIGN AND METHODS: We performed a pilot trial in hospitalized patients with INR between 1.5 and 2.5 undergoing procedures conducted outside the operating room. We excluded patients undergoing procedures proximal to the central nervous system, platelet counts <40,000/µl, or congenital or acquired coagulation disorders unresponsive to plasma. We randomly allocated patients stratified by hospital and history of cirrhosis to receive plasma transfusion (10-15 cc/kg) or no transfusion. The primary outcome was change in hemoglobin concentration within 2 days of procedure. RESULTS: We enrolled 57 patients, mean age 56.0, 34 (59.6%) with cirrhosis, and mean INR 1.92 (SD = 0.27). In the intention to treat analysis, there were 10 of 27 (38.5%) participants in the plasma arm with a post procedure INR <1.5 and one of 30 (3.6%) in the no treatment arm (p < .01). The mean INR after receiving plasma transfusion was -0.24 (SD 0.26) lower than baseline. The change from pre-procedure hemoglobin level to lowest level within 2 days was -0.6 (SD = 1.0) in the plasma transfusion arm and -0.4 (SD = 0.6) in the no transfusion arm (p = .29). Adverse outcomes were uncommon. DISCUSSION: We found no differences in change in hemoglobin concentration in those treated with plasma compared to no treatment. The change in INR was small and corrected to less than 1.5 in minority of patients. Large trials are required to establish if plasma is safe and efficacious.

Time-aware deep neural networks for needle tip localization in 2D ultrasound.

PURPOSE: Accurate placement of the needle is critical in interventions like biopsies and regional anesthesia, during which incorrect needle insertion can lead to procedure failure and complications. Therefore, ultrasound guidance is widely used to improve needle placement accuracy. However, at steep and deep insertions, the visibility of the needle is lost. Computational methods for automatic needle tip localization could improve the clinical success rate in these scenarios. METHODS: We propose a novel algorithm for needle tip localization during challenging ultrasound-guided insertions when the shaft may be invisible, and the tip has a low intensity. There are two key steps in our approach. First, we enhance the needle tip features in consecutive ultrasound frames using a detection scheme which recognizes subtle intensity variations caused by needle tip movement. We then employ a hybrid deep neural network comprising a convolutional neural network and long short-term memory recurrent units. The input to the network is a consecutive plurality of fused enhanced frames and the corresponding original B-mode frames, and this spatiotemporal information is used to predict the needle tip location. RESULTS: We evaluate our approach on an ex vivo dataset collected with in-plane and out-of-plane insertion of 17G and 22G needles in bovine, porcine, and chicken tissue, acquired using two different ultrasound systems. We train the model with 5000 frames from 42 video sequences. Evaluation on 600 frames from 30 sequences yields a tip localization error of [Formula: see text] mm and an overall inference time of 0.064 s (15 fps). Comparison against prior art on challenging datasets reveals a 30% improvement in tip localization accuracy. CONCLUSION: The proposed method automatically models temporal dynamics associated with needle tip motion and is more accurate than state-of-the-art methods. Therefore, it has the potential for improving needle tip localization in challenging ultrasound-guided interventions.

Effect of cyclo-oxygenase inhibitor use during checkpoint blockade immunotherapy in patients with metastatic melanoma and non-small cell lung cancer.

BACKGROUND: Immune checkpoint inhibitors (ICIs) improve survival outcomes in metastatic melanoma and non-small cell lung cancer (NSCLC). Preclinical evidence suggests that overexpression of cyclo-oxygenase-2 (COX2) in tumors facilitates immune evasion through prostaglandin E2 production and that COX inhibition synergizes with ICIs to promote antitumor T-cell activation. This study investigates whether concurrent COX inhibitor (COXi) use during ICI treatment compared with ICI alone is associated with improved time-to-progression (TTP), objective response rate (ORR) and overall survival (OS) in patients with metastatic melanoma and NSCLC. METHODS: We retrospectively reviewed 90 metastatic melanoma and 37 metastatic NSCLC patients, treated with ICI between 2011 and 2019. Differences in TTP and OS by ICI+COXi versus ICI alone were compared using Kaplan-Meier and Cox regression. Interaction between ICI+COXi versus ICI alone and pretreatment neutrophil-lymphocyte ratio (NLR) was examined. Independent radiology review per Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 was performed. RESULTS: For patients with melanoma, median TTP was significantly prolonged in ICI+COXi versus ICI alone (245 vs 100.5 days, p=0.002). On multivariate analysis, ICI+COXi associated with increased TTP (HR 0.36, 95% CI 0.2 to 0.66, p=0.001), adjusted for age, pretreatment NLR, and gender. For NSCLC patients, ICI+COXi also associated with increased TTP compared with ICI alone on multivariate analysis (HR 0.45; 95% CI 0.21 to 0.97; p=0.042) adjusted for age. ORR at 6 months was significantly higher in patients who received ICI+COXi compared with ICI alone in both melanoma (58.6% vs 19.2%, p=0.0005) and NSCLC (73.7% vs 33.3%, p=0.036) cohorts. In the melanoma cohort, high pretreatment NLR (>5) associated with decreased TTP (HR 3.21, 95% CI 1.64 to 6.3; p=0.0007); however, ICI+COXi significantly associated with increased TTP in high NLR (>5) patients (HR 0.08, 95% CI 0.03 to 0.25), but not in low NLR (≤5) patients (HR 0.65, 95% CI 0.32 to 1.32). Similar outcomes were found in an adjusted melanoma cohort after RECIST review. CONCLUSIONS: Our study suggests that COXi use concurrently with ICI significantly associated with longer TTP and improved ORR at 6 months in patients with metastatic melanoma and NSCLC compared with ICI alone. Furthermore, COXi use appears to reverse the negative prognostic effect of a high NLR by prolonging TTP in patients with melanoma.

Conditional survival analysis of hepatocellular carcinoma.

BACKGROUND: Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality worldwide with an approximate 5-year survival of greater than 50% in patients after surgical resection. Survival estimates have limited utility for patients who have survived several years after initial treatment. We analyzed how conditional survival (CS) after curative-intent surgery for HCC predicts survival estimates over time. METHODS: NCDB (2004-2014) was queried for patients undergoing definitive surgical resection for HCC. Cumulative overall survival (OS) was calculated using the Kaplan-Meier method, and CS at x years after diagnosis was calculated as CS1 = OS (X+5) /OS(X) . RESULTS: The final analysis encompassed 11 357 patients. Age, negative margin status, grade severity and radiation before surgery were statistically significant predictors of cumulative overall conditional survival (P ≤ .0001). Overall unconditional 5-year survival was 65.7%, but CS estimates were higher. A patient who has already survived 3 years has an additional 2-year, or 5-year CS, estimate of 86.96%. CONCLUSION: Survival estimates following hepatic resection in HCC patients change according to survival time accrued since surgery. CS estimates are improved relative to unconditional OS. The impact of different variables influencing OS is likewise nonlinear over the course of time after surgery.

Rethinking the Role of Radiation Therapy in the Treatment of Unresectable Hepatocellular Carcinoma: A Data Driven Treatment Algorithm for Optimizing Outcomes.

Hepatocellular carcinoma (HCC) is the second most common cause of cancer death worldwide, with a majority of HCC patients not suitable for curative therapies. Approximately 70% of initially diagnosed patients cannot undergo surgical resection or transplantation due to locally advanced disease, poor liver function/underlying cirrhosis, or additional comorbidities. Local therapeutic options for patients with unresectable HCC, who are not suitable for thermal ablation, include transarterial embolization (bland, chemoembolization, radioembolization) and/or external beam radiation therapy (EBRT). Regarding EBRT specifically, technological advancements provide a means for safe and effective radiotherapy delivery in a wide spectrum of HCC patients. In multiple prospective studies, EBRT delivery in a variety of different fractionation schemes or in combination with transcatheter arterial chemoembolization (TACE) demonstrate improved outcomes, particularly with combination therapy. The Barcelona Clinic Liver Cancer classification provides a framework for treatment selection; however, given the growing complexity of treatment strategies, this classification system tends to simplify decision-making. In this review, we discuss the current literature regarding unresectable HCC and propose a modified treatment algorithm that emphasizes the role of radiation therapy for Child-Pugh score A or B patients with ≤3 nodules measuring >3 cm, multinodular disease or portal venous thrombosis.

Learning needle tip localization from digital subtraction in 2D ultrasound.

PURPOSE: This paper addresses localization of needles inserted both in-plane and out-of-plane in challenging ultrasound-guided interventions where the shaft and tip have low intensity. Our approach combines a novel digital subtraction scheme for enhancement of low-level intensity changes caused by tip movement in the ultrasound image and a state-of-the-art deep learning scheme for tip detection. METHODS: As the needle tip moves through tissue, it causes subtle spatiotemporal variations in intensity. Relying on these intensity changes, we formulate a foreground detection scheme for enhancing the tip from consecutive ultrasound frames. The tip is augmented by solving a spatial total variation regularization problem using the split Bregman method. Lastly, we filter irrelevant motion events with a deep learning-based end-to-end data-driven method that models the appearance of the needle tip in ultrasound images, resulting in needle tip detection. RESULTS: The detection model is trained and evaluated on an extensive ex vivo dataset collected with 17G and 22G needles inserted in-plane and out-of-plane in bovine, porcine and chicken phantoms. We use 5000 images extracted from 20 video sequences for training and 1000 images from 10 sequences for validation. The overall framework is evaluated on 700 images from 20 sequences not used in training and validation, and achieves a tip localization error of 0.72 ± 0.04 mm and an overall processing time of 0.094 s per frame (~ 10 frames per second). CONCLUSION: The proposed method is faster and more accurate than state of the art and is resilient to spatiotemporal redundancies. The promising results demonstrate its potential for accurate needle localization in challenging ultrasound-guided interventions.

Predictors of Survival after Yttrium-90 Radioembolization for Colorectal Cancer Liver Metastases.

PURPOSE: To identify clinical parameters that are prognostic for improved overall survival (OS) after yttrium-90 radioembolization (RE) in patients with liver metastases from colorectal cancer (CRC). MATERIALS AND METHODS: A total of 131 patients who underwent RE for liver metastases from CRC, treated at 2 academic centers, were reviewed. Twenty-one baseline pretreatment clinical factors were analyzed in relation to OS by the Kaplan-Meier method along with log-rank tests and univariate and multivariate Cox regression analyses. RESULTS: The median OS from first RE procedure was 10.7 months (95% confidence interval [CI], 9.4-12.7 months). Several pretreatment factors, including lower carcinoembryonic antigen (CEA; ≤20 ng/mL), lower aspartate transaminase (AST; ≤40 IU/L), neutrophil-lymphocyte ratio (NLR) <5, and absence of extrahepatic disease at baseline were associated with significantly improved OS after RE, compared with high CEA (>20 ng/mL), high AST (>40 IU/L), NLR ≥5, and extrahepatic metastases (P values of <.001, <.001, .0001, and .04, respectively). On multivariate analysis, higher CEA, higher AST, NLR ≥5, extrahepatic disease, and larger volume of liver metastases remained independently associated with risk of death (hazard ratios of 1.63, 2.06, 2.22, 1.48, and 1.02, respectively). CONCLUSIONS: The prognosis of patients with metastases from CRC is impacted by a complex set of clinical parameters. This analysis of pretreatment factors identified lower AST, lower CEA, lower NLR, and lower tumor burden (intra- or extrahepatic) to be independently associated with higher survival after hepatic RE. Optimal selection of patients with CRC liver metastases may improve survival rates after administration of yttrium-90.

Convolution neural networks for real-time needle detection and localization in 2D ultrasound.

PURPOSE: We propose a framework for automatic and accurate detection of steeply inserted needles in 2D ultrasound data using convolution neural networks. We demonstrate its application in needle trajectory estimation and tip localization. METHODS: Our approach consists of a unified network, comprising a fully convolutional network (FCN) and a fast region-based convolutional neural network (R-CNN). The FCN proposes candidate regions, which are then fed to a fast R-CNN for finer needle detection. We leverage a transfer learning paradigm, where the network weights are initialized by training with non-medical images, and fine-tuned with ex vivo ultrasound scans collected during insertion of a 17G epidural needle into freshly excised porcine and bovine tissue at depth settings up to 9 cm and [Formula: see text]-[Formula: see text] insertion angles. Needle detection results are used to accurately estimate needle trajectory from intensity invariant needle features and perform needle tip localization from an intensity search along the needle trajectory. RESULTS: Our needle detection model was trained and validated on 2500 ex vivo ultrasound scans. The detection system has a frame rate of 25 fps on a GPU and achieves 99.6% precision, 99.78% recall rate and an [Formula: see text] score of 0.99. Validation for needle localization was performed on 400 scans collected using a different imaging platform, over a bovine/porcine lumbosacral spine phantom. Shaft localization error of [Formula: see text], tip localization error of [Formula: see text] mm, and a total processing time of 0.58 s were achieved. CONCLUSION: The proposed method is fully automatic and provides robust needle localization results in challenging scanning conditions. The accurate and robust results coupled with real-time detection and sub-second total processing make the proposed method promising in applications for needle detection and localization during challenging minimally invasive ultrasound-guided procedures.

A novel combined interventional radiologic and hepatobiliary surgical approach to a complex traumatic hilar biliary stricture.

INTRODUCTION: Benign strictures of the biliary system are challenging and uncommon conditions requiring a multidisciplinary team for appropriate management. PRESENTATION OF CASE: The patient is a 32-year-old male that developed a hilar stricture as sequelae of a gunshot wound. Due to the complex nature of the stricture and scarring at the porta hepatis a combined interventional radiologic and surgical approach was carried out to approach the hilum of the right and left hepatic ducts. The location of this stricture was found by ultrasound guidance intraoperatively using a balloon tipped catheter placed under fluoroscopy in the interventional radiology suite prior to surgery. This allowed the surgeons to select the line of parenchymal transection for best visualization of the stricture. A left hepatectomy was performed, the internal stent located and the right hepatic duct opened tangentially to allow a side-to-side Roux-en-Y hepaticojejunostomy (a Puestow-like anastomosis). DISCUSSION: Injury to the intrahepatic biliary ductal confluence is rarely fatal, however, the associated injuries lead to severe morbidity as seen in this example. Management of these injuries poses a considerable challenge to the surgeon and treating physicians. CONCLUSION: Here we describe an innovative multi-disciplinary approach to the repair of this rare injury.

Signal attenuation maps for needle enhancement and localization in 2D ultrasound.

PURPOSE: We propose a novel framework for enhancement and localization of steeply inserted hand-held needles under in-plane 2D ultrasound guidance. METHODS: Depth-dependent attenuation and non-axial specular reflection hinder visibility of steeply inserted needles. Here, we model signal transmission maps representative of the attenuation probability within the image domain. The maps are employed in a contextual regularization framework to recover needle shaft and tip information. The needle tip is automatically localized by line-fitting along the local-phase-directed trajectory, followed by statistical optimization. RESULTS: The proposed method was tested on 300 ex vivo ultrasound scans collected during insertion of an epidural needle into freshly excised porcine and bovine tissue. A tip localization accuracy of [Formula: see text] was achieved. CONCLUSION: The proposed method could be useful in challenging procedures where needle shaft and tip are inconspicuous. Improved needle localization results compared to previously proposed methods suggest that the proposed method is promising for further clinical evaluation.

Endovascular treatment of pulmonary embolism: Selective review of available techniques.

Acute pulmonary embolism (PE) is the third most common cause of death in hospitalized patients. The development of sophisticated diagnostic and therapeutic modalities for PE, including endovascular therapy, affords a certain level of complexity to the treatment of patients with this important clinical entity. Furthermore, the lack of level I evidence for the safety and effectiveness of catheter directed therapy brings controversy to a promising treatment approach. In this review paper, we discuss the pathophysiology and clinical presentation of PE, review the medical and surgical treatment of the condition, and describe in detail the tools that are available for the endovascular therapy of PE, including mechanical thrombectomy, suction thrombectomy, and fibrinolytic therapy. We also review the literature available to date on these methods, and describe the function of the Pulmonary Embolism Response Team.

Gastrointestinal bleeding in a patient with a continuous-flow biventricular assist device.

The association between continuous-flow left ventricular assist devices (CF-LVADs) and gastrointestinal (GI) bleeding from angiodysplasia is well recognized. However, the association between continuous-flow biventricular assist devices (CF-BIVADs) and bleeding angiodysplasia is less understood. We report a case of GI bleeding from a patient with a CF-BIVAD. The location of GI bleeding was identified by nuclear red blood cell bleeding scan. The vascular malformation leading to the bleed was identified and localized on angiography and then by pathology. The intensity of bleeding, reflected by number of units of packed red blood cells needed for normalization of hemoglobin, as well as the time to onset of bleeding after transplantation, are similar to that seen in the literature for CF-LVADs and pulsatile BIVADs. While angiography only detected a dilated late draining vein, pathology demonstrated the presence of both arterial and venous dilation in the submucosa, vascular abnormalities characteristic of a late arteriovenous malformation.

Non-operative therapies for colorectal liver metastases.

Locoregional therapies for colorectal liver metastases complement systemic therapy by providing an opportunity for local control of hepatic spread. The armamentarium for liver-directed therapy includes ablative therapies, embolization, and stereotactic body radiation therapy. At this time, prospective studies comparing these modalities are limited and decision-making relies on a multidisciplinary approach for optimal patient management. Herein, we describe multiple therapeutic non-surgical procedures and an overview of the results of these treatments.

Decreased vascularization of retroperitoneal kaposiform hemangioendothelioma induced by treatment with sirolimus explains relief of symptoms.

Kaposiform hemangioendothelioma (KHE) is a vascular tumor with poor prognosis. We present a child with progressive disability, extreme pain, and autonomic dysfunction due to a retroperitoneal KHE where radiologic characteristics were essential for diagnosis and monitoring of response to therapy. He received sirolimus, and the symptomatology resolved completely. Repeat MRIs revealed fast marked decrease in vascularity of the tumor, although the volume was not significantly affected. We suggest that the sirolimus-induced tumor de-vascularization may explain the clinical and coagulopathy improvement.

Vascular anomalies: A pictorial review of nomenclature, diagnosis and treatment.

Vascular anomalies, including vascular malformations and tumors, are frequently straightforward to detect; however, accurate diagnosis and appropriate treatment are often challenging. Misdiagnosis of these lesions can lead clinicians in the wrong direction when treating these patients, which can have unfavorable results. This review presents an overview of the classification systems that have been developed for the diagnosis of vascular lesions with a focus on the imaging characteristics. Pictorial examples of each lesion on physical examination, as well as non-invasive and minimally invasive imaging are presented. An overview of the endovascular treatment of these lesions is also given. In some cases, vascular anomalies may be associated with an underlying syndrome and several of the most commonly encountered syndromes are discussed. Understanding of the classification systems, familiarity with the treatment options and knowledge of the associated syndromes are essential for all physicians working with this patient population. The approach to the described entities necessitates an organized multi-disciplinary team effort, with diagnostic imaging playing an increasingly important role in the proper diagnosis and a combined interventional radiologic and surgical treatment method showing promising results.

The power of genes: a case of unusually severe systemic toxicity after localized hepatic chemoembolization with irinotecan-eluted microspheres for metastatic colon cancer.

OBJECTIVE: To report a case of systemic irinotecan toxicity following regional transarterial chemoembolization with drug-eluting beads loaded with irinotecan (DEBIRI-TACE) in a patient later found to have a homozygous mutation for UGT1A1*28. CASE SUMMARY: An 80-year-old woman presented with a cecal colon cancer with synchronous metastases to the liver. After resection of the primary tumor, the patient underwent DEBIRI-TACE with 100 mg of irinotecan to treat the residual disease in the liver. A week after this procedure, the patient developed grade 4 neutropenia, and later, alopecia. Eventually, it was found that the patient had a mutation of UDP glucuronosyltransferase 1 family polypeptide A1 (UGT1A1), which provided a reasonable explanation for the observed reaction. DISCUSSION: The toxic effects of irinotecan are well understood. Patients with genetic polymorphisms of the genes encoding for the enzyme UGT1A1 may have increased incidence of irinotecan-associated toxicities because of decreased clearance of the active metabolite SN38 via the glucuronidation pathway. To date, there have been limited publications describing systemic adverse events following TACE or DEBIRI-TACE and, based on a thorough literature search, none following these procedures in patients with UGT1A1 polymorphisms. Based on the scoring results of the Naranjo algorithm (7), we are confident in attributing the observed reaction to the patient's genetic polymorphism. CONCLUSION: Although genetic testing prior to the initiation of irinotecan therapy is not currently recommended, assessment of UGT1A1 polymorphism is warranted when severe adverse events typical of systemic therapy manifest following DEBIRI-TACE.

Patterns of failure in colorectal patients with liver metastases after yttrium-90 radioembolization.

OBJECTIVES: To assess patterns of failure and factors affecting recurrence patterns in colorectal cancer patients treated with Y-labeled resin microspheres for metastatic liver disease. METHODS: We retrospectively reviewed 30 colorectal patients treated with Yttrium-90 radioembolization and assessed follow-up computed tomography scans and positron emission tomography scans to determine disease outcomes. All patients were included in survival analysis. Twenty-six patients with hepatic metastases were assessed for patterns of failure after radioembolization treatment and grouped into 3 patterns: (1) hepatic; (2) extrahepatic; and (3) intrahepatic and extrahepatic. RESULTS: The median overall survival and progression-free survival for all colorectal patients treated with radioembolization was 9.4 and 3.2 months, respectively. Overall survival and progression-free survival were not significantly different between patterns of failure (P=0.43 and 0.26, respectively). Logistic regression analysis demonstrated a trend toward the predictive value of tumor volume in determining patterns of failure. Smaller tumor volumes had a higher predictive probability for extrahepatic failure than larger tumor volumes (P=0.057). Tumor volumes <300 mL were predictive for extrahepatic failure patterns compared with hepatic recurrence (P=0.046). CONCLUSIONS: Radioembolization with Y-labeled resin microspheres continues to be an effective salvage treatment for colorectal liver metastases. Analysis of patterns of radiologic failure demonstrated that patients treated by radioembolization develop a greater proportion of extrahepatic failure. Tumor volumes >300 mL were predictive for hepatic recurrence, suggesting that increased dosing or retreatment of these lesions may lead to improved hepatic control of disease and better patient outcomes.

Pilot study of angiogenic response to yttrium-90 radioembolization with resin microspheres.

PURPOSE: To investigate the impact of radioembolization with yttrium-90 resin microspheres on the regulation of angiogenesis through observation of serial changes in a spectrum of angiogenic markers and other cytokines after therapy. MATERIALS AND METHODS: This prospective pilot study enrolled 22 patients with liver-dominant disease deriving from biopsy-proven hepatocellular carcinoma (HCC) (n = 7) or metastatic colorectal carcinoma (mCRC) (n = 15). Circulating angiogenic markers were measured from serum samples drawn at baseline and at time points after therapy ranging from 6 hours to 120 days. Using multiplex enzyme-linked immunosorbent assay, several classic angiogenesis factors (vascular endothelial growth factor [VEGF], angiopoietin-2 [Ang-2], basic fibroblast growth factor [bFGF], platelet-derived growth factor subunit BB [PDGF-BB], thrombospondin-1 [Tsp-1]) and nonclassic factors (follistatin, leptin, interleukin [IL]-8) were evaluated. RESULTS: Increases in cytokine levels ≥ 50% over baseline were observed in more than half of all patients studied for many cytokines, including classic angiogenic factors such as VEGF, Ang-2, and Tsp-1 as well as nonclassic factors IL-8 and follistatin (range, 36%-82% for all cytokines). Baseline cytokine levels in patients with overall survival (OS) < 6 months differed significantly from patients with longer survival for Ang-2 (P = .033) and IL-8 (P = .041). Patients with OS ≤ 6 months exhibited transient increases in VEGF and PDGF-BB after therapy compared with patients with OS > 6 months. CONCLUSIONS: Radioembolization is associated with early transient increases in many angiogenic cytokines. In this small sample size, some of these changes were associated with worse OS. This research has important implications for future studies of radioembolization with antiangiogenic therapy performed during and after the procedure.