Automated graded prognostic assessment for patients with hepatocellular carcinoma using machine learning.

BACKGROUND: Accurate mortality risk quantification is crucial for the management of hepatocellular carcinoma (HCC); however, most scoring systems are subjective. PURPOSE: To develop and independently validate a machine learning mortality risk quantification method for HCC patients using standard-of-care clinical data and liver radiomics on baseline magnetic resonance imaging (MRI). METHODS: This retrospective study included all patients with multiphasic contrast-enhanced MRI at the time of diagnosis treated at our institution. Patients were censored at their last date of follow-up, end-of-observation, or liver transplantation date. The data were randomly sampled into independent cohorts, with 85% for development and 15% for independent validation. An automated liver segmentation framework was adopted for radiomic feature extraction. A random survival forest combined clinical and radiomic variables to predict overall survival (OS), and performance was evaluated using Harrell's C-index. RESULTS: A total of 555 treatment-naïve HCC patients (mean age, 63.8 years ± 8.9 [standard deviation]; 118 females) with MRI at the time of diagnosis were included, of which 287 (51.7%) died after a median time of 14.40 (interquartile range, 22.23) months, and had median followed up of 32.47 (interquartile range, 61.5) months. The developed risk prediction framework required 1.11 min on average and yielded C-indices of 0.8503 and 0.8234 in the development and independent validation cohorts, respectively, outperforming conventional clinical staging systems. Predicted risk scores were significantly associated with OS (p < .00001 in both cohorts). CONCLUSIONS: Machine learning reliably, rapidly, and reproducibly predicts mortality risk in patients with hepatocellular carcinoma from data routinely acquired in clinical practice. CLINICAL RELEVANCE STATEMENT: Precision mortality risk prediction using routinely available standard-of-care clinical data and automated MRI radiomic features could enable personalized follow-up strategies, guide management decisions, and improve clinical workflow efficiency in tumor boards. KEY POINTS: • Machine learning enables hepatocellular carcinoma mortality risk prediction using standard-of-care clinical data and automated radiomic features from multiphasic contrast-enhanced MRI. • Automated mortality risk prediction achieved state-of-the-art performances for mortality risk quantification and outperformed conventional clinical staging systems. • Patients were stratified into low, intermediate, and high-risk groups with significantly different survival times, generalizable to an independent evaluation cohort.

Automated MRI liver segmentation for anatomical segmentation, liver volumetry, and the extraction of radiomics.

OBJECTIVES: To develop and evaluate a deep convolutional neural network (DCNN) for automated liver segmentation, volumetry, and radiomic feature extraction on contrast-enhanced portal venous phase magnetic resonance imaging (MRI). MATERIALS AND METHODS: This retrospective study included hepatocellular carcinoma patients from an institutional database with portal venous MRI. After manual segmentation, the data was randomly split into independent training, validation, and internal testing sets. From a collaborating institution, de-identified scans were used for external testing. The public LiverHccSeg dataset was used for further external validation. A 3D DCNN was trained to automatically segment the liver. Segmentation accuracy was quantified by the Dice similarity coefficient (DSC) with respect to manual segmentation. A Mann-Whitney U test was used to compare the internal and external test sets. Agreement of volumetry and radiomic features was assessed using the intraclass correlation coefficient (ICC). RESULTS: In total, 470 patients met the inclusion criteria (63.9±8.2 years; 376 males) and 20 patients were used for external validation (41±12 years; 13 males). DSC segmentation accuracy of the DCNN was similarly high between the internal (0.97±0.01) and external (0.96±0.03) test sets (p=0.28) and demonstrated robust segmentation performance on public testing (0.93±0.03). Agreement of liver volumetry was satisfactory in the internal (ICC, 0.99), external (ICC, 0.97), and public (ICC, 0.85) test sets. Radiomic features demonstrated excellent agreement in the internal (mean ICC, 0.98±0.04), external (mean ICC, 0.94±0.10), and public (mean ICC, 0.91±0.09) datasets. CONCLUSION: Automated liver segmentation yields robust and generalizable segmentation performance on MRI data and can be used for volumetry and radiomic feature extraction. CLINICAL RELEVANCE STATEMENT: Liver volumetry, anatomic localization, and extraction of quantitative imaging biomarkers require accurate segmentation, but manual segmentation is time-consuming. A deep convolutional neural network demonstrates fast and accurate segmentation performance on T1-weighted portal venous MRI. KEY POINTS: • This deep convolutional neural network yields robust and generalizable liver segmentation performance on internal, external, and public testing data. • Automated liver volumetry demonstrated excellent agreement with manual volumetry. • Automated liver segmentations can be used for robust and reproducible radiomic feature extraction.

Detection of differentially abundant cell subpopulations in scRNA-seq data.

Comprehensive and accurate comparisons of transcriptomic distributions of cells from samples taken from two different biological states, such as healthy versus diseased individuals, are an emerging challenge in single-cell RNA sequencing (scRNA-seq) analysis. Current methods for detecting differentially abundant (DA) subpopulations between samples rely heavily on initial clustering of all cells in both samples. Often, this clustering step is inadequate since the DA subpopulations may not align with a clear cluster structure, and important differences between the two biological states can be missed. Here, we introduce DA-seq, a targeted approach for identifying DA subpopulations not restricted to clusters. DA-seq is a multiscale method that quantifies a local DA measure for each cell, which is computed from its k nearest neighboring cells across a range of k values. Based on this measure, DA-seq delineates contiguous significant DA subpopulations in the transcriptomic space. We apply DA-seq to several scRNA-seq datasets and highlight its improved ability to detect differences between distinct phenotypes in severe versus mildly ill COVID-19 patients, melanomas subjected to immune checkpoint therapy comparing responders to nonresponders, embryonic development at two time points, and young versus aging brain tissue. DA-seq enabled us to detect differences between these phenotypes. Importantly, we find that DA-seq not only recovers the DA cell types as discovered in the original studies but also reveals additional DA subpopulations that were not described before. Analysis of these subpopulations yields biological insights that would otherwise be undetected using conventional computational approaches.

Diabetes medications and risk of HCC.

Type 2 diabetes mellitus is a recognized risk factor for HCC in patients with liver disease, independent from the etiology of their liver disease. Hence, prevention and treatment of type 2 diabetes mellitus and its underlying cause, insulin resistance, should be considered a treatment target for patients with liver disease. The drug armamentarium for diabetes is wide and consists of agents with insulin-sensitizing activity, agents that stimulate insulin secretion, insulin itself, and agents that reduce gastrointestinal and urinary glucose absorption. From an endocrinology perspective, the main goal of treatment is the achievement of euglycemia; however, in patients at risk of, or with known underlying liver disease, the choice of diabetic medication as it relates to potential hepatic carcinogenesis remains complex and should be carefully considered. In the last decade, increasing evidence has suggested that metformin may reduce the risk of HCC, whereas evidence for other classes of diabetic medications, particularly some of the newer agents including the sodium glucose cotransporter-2 inhibitors and glucagon-like peptide-1 receptor agonists, is fewer and often inconsistent. In this review, we aim to summarize the current evidence on the potential effects of the most widely used diabetic agents on liver cancer tumorigenesis.

Immune checkpoint inhibitors in malignancies after liver transplantation: A systematic review and pooled analysis.

BACKGROUND AND AIMS: Treatment of de novo malignancies and recurrent hepatocellular carcinoma with immune checkpoint inhibitors (ICI) in liver transplant recipients (LT) is an attractive strategy that is infrequently pursued because of the lack of strong evidence regarding their safety and efficacy. In this systematic review with pooled analysis, we aimed to assess safety and efficacy of ICI therapy following LT. METHODS: We performed a systematic search of case reports and series published until January 2022. We included 31 publications reporting a total of 52 patients treated with ICIs after LT and assessed in a pooled analysis the risk of graft rejection and the outcome of ICI therapy. RESULTS: Acute graft rejection occurred in 15 patients (28.8%) and 7 patients (13.4% of the total cohort) died because of graft loss. Rejection was associated with shorter overall survival (OS) (17.2 months, confidence interval [CI] 12.1-22.2 vs. 3.5 months, CI 1.6-5.4, p < 0.001). Disease control rate was 44.2% (n = 23), and in these patients, OS was longer than in non-responders (26.4 months, CI 20.8-32.0 vs. 3.4 months, CI 2.1-4.7, p < 0.001). CONCLUSIONS: Observational, off-label experience suggests that treatment with ICI for advanced malignancies in LT recipients might not be discarded a priori. This notwithstanding, ICI treatment in these patients is associated with a substantial risk of graft rejection and mortality. Prospective studies are needed to provide adequate safety and efficacy figures of ICI treatment in this fragile population.

Optimization of the BCLC Staging System for Locoregional Therapy for Hepatocellular Carcinoma by Using Quantitative Tumor Burden Imaging Biomarkers at MRI.

Background Patients with intermediate- and advanced-stage hepatocellular carcinoma (HCC) represent a highly heterogeneous patient collective with substantial differences in overall survival. Purpose To evaluate enhancing tumor volume (ETV) and enhancing tumor burden (ETB) as new criteria within the Barcelona Clinic Liver Cancer (BCLC) staging system for optimized allocation of patients with intermediate- and advanced-stage HCC to undergo transarterial chemoembolization (TACE). Materials and Methods In this retrospective study, 682 patients with HCC who underwent conventional TACE or TACE with drug-eluting beads from January 2000 to December 2014 were evaluated. Quantitative three-dimensional analysis of contrast-enhanced MRI was performed to determine thresholds of ETV and ETB (ratio of ETV to normal liver volume). Patients with ETV below 65 cm3 or ETB below 4% were reassigned to BCLC Bn, whereas patients with ETV or ETB above the determined cutoffs were restratified or remained in BCLC Cn by means of stepwise verification of the median overall survival (mOS). Results This study included 494 patients (median age, 62 years [IQR, 56-71 years]; 401 men). Originally, 123 patients were classified as BCLC B with mOS of 24.3 months (95% CI: 21.4, 32.9) and 371 patients as BCLC C with mOS of 11.9 months (95% CI: 10.5, 14.8). The mOS of all included patients (including the BCLC B and C groups) was 15 months (95% CI: 12.3, 17.2). A total of 152 patients with BCLC C tumors were restratified into a new BCLC Bn class, in which the mOS was then 25.1 months (95% CI: 21.8, 29.7; P < .001). The mOS of the remaining patients (ie, BCLC Cn group) (n = 222; ETV ≥65 cm3 or ETB ≥4%) was 8.4 months (95% CI: 6.1, 11.2). Conclusion Substratification of patients with intermediate- and advanced-stage hepatocellular carcinoma according to three-dimensional quantitative tumor burden identified patients with a survival benefit from transarterial chemoembolization before therapy. © RSNA, 2022 Online supplemental material is available for this article.

Holistic management of hepatocellular carcinoma: The hepatologist's comprehensive playbook.

Hepatocellular carcinoma (HCC) is a common complication in patients with chronic liver disease and leads to significant morbidity and mortality. Liver disease and liver cancer are preventable by mitigating and managing common risk factors, including chronic hepatitis B and C infection, alcohol use, diabetes, obesity and other components of the metabolic syndrome. The management of patients with HCC requires treatment of the malignancy and adequate control of the underlying liver disease, as preserving liver function is critical for successful cancer treatment and may have a relevant prognostic role independent of HCC management. Hepatologists are the ideal providers to guide the care of patients with HCC as they are trained to identify patients at risk, apply appropriate surveillance strategies, assess and improve residual liver function, evaluate candidacy for transplant, provide longitudinal care to optimize and preserve liver function during and after HCC treatment, survey for cancer recurrence and manage its risk factors, and prevent and treat decompensating events. We highlight the need for a team-based holistic approach to the patient with liver disease and HCC and identify necessary gaps in current care and knowledge.

Reliable prediction of survival in advanced-stage hepatocellular carcinoma treated with sorafenib: comparing 1D and 3D quantitative tumor response criteria on MRI.

OBJECTIVES: To compare 1D and 3D quantitative tumor response criteria applied to DCE-MRI in patients with advanced-stage HCC undergoing sorafenib therapy to predict overall survival (OS) early during treatment. METHODS: This retrospective analysis included 29 patients with advanced-stage HCC who received sorafenib for at least 60 days. All patients underwent baseline and follow-up DCE-MRI at 81.5 ± 29.3 days (range 35-140 days). Response to sorafenib was assessed in 46 target lesions using 1D criteria RECIST1.1 and mRECIST. In addition, a segmentation-based 3D quantification of absolute enhancing lesion volume (vqEASL) was performed on the arterial phase MRI, and the enhancement fraction of total tumor volume (%qEASL) was calculated. Accordingly, patients were stratified into groups of disease control (DC) and disease progression (DP). OS was evaluated using Kaplan-Meier curves with log-rank test and Cox proportional hazards regression model. RESULTS: The Kaplan-Meier analysis revealed that stratification of patients in DC vs. DP according to mRECIST (p = 0.0371) and vqEASL (p = 0.0118) successfully captured response and stratified OS, while stratification according to RECIST and %qEASL did not correlate with OS (p = 0.6273 and p = 0.7474, respectively). Multivariable Cox regression identified tumor progression according to mRECIST and qEASL as independent risk factors of decreased OS (p = 0.039 and p = 0.006, respectively). CONCLUSIONS: The study identified enhancement-based vqEASL and mRECIST as reliable predictors of patient survival early after initiation of treatment with sorafenib. This data provides evidence for potential advantages 3D quantitative, enhancement-based tumor response analysis over conventional techniques regarding early identification of treatment success or failure. KEY POINTS: • Tumor response criteria on MRI can be used to predict survival benefit of sorafenib therapy in patients with advanced HCC. • Stratification into DC and DP using mRECIST and vqEASL significantly correlates with OS (p = 0.0371 and p = 0.0118, respectively) early after initiation of sorafenib, while stratification according to RECIST and %qEASL did not correlate with OS (p = 0.6273 and p = 0.7474, respectively). • mRECIST (HR = 0.325, p = 0.039. 95%CI 0.112-0.946) and qEASL (HR = 0.183, p = 0.006, 95%CI 0.055-0.613) are independent prognostic factors of survival in HCC patients undergoing sorafenib therapy.

Bile-Derived Organoids From Patients With Primary Sclerosing Cholangitis Recapitulate Their Inflammatory Immune Profile.

Primary sclerosing cholangitis (PSC) is a heterogeneous and progressive fibroinflammatory cholangiopathy with no known etiology or effective treatment. Studies of PSC are limited due to difficulty in accessing the cholangiocyte, the small percentage of these cells in the liver, instability of in vitro culture systems, and reliance on samples from end-stage disease. Here, we demonstrate that stem cells can be isolated from the bile of PSC patients undergoing endoscopic retrograde cholangiopancreatography earlier in their clinical course and maintained long term in vitro as three-dimensional (3D) organoids that express a biliary genetic phenotype. Additionally, bile-derived organoids (BDOs) can be biobanked and samples obtained longitudinally over the course of the disease. These BDOs express known cholangiocyte markers including gamma glutamyl transferase, cytokeratin 19, epithelial cellular adhesion molecule, cystic fibrosis transmembrane conductance regulator, and anion exchanger 2. RNA sequence analysis identified 39 genes whose expression differed in organoids from PSC patients compared to non-PSC controls, including human leukocyte antigen DM alpha chain and chemokine (C-C motif) ligand 20 (CCL20), immune-related genes previously described in genome-wide association studies of PSC. Incubation of these BDOs with interleukin 17A or tumor necrosis factor alpha led to an immune-reactive phenotype with a significant increase in secretion of proinflammatory mediators, including CCL20, a T-cell chemoattractant. Conclusion: This study demonstrates that bile can be used as a source of biliary-like cells that can be maintained long term in vitro as 3D organoids; these BDOs retain features of cholangiopathies, including the ability to react to inflammatory stimuli by secreting chemokines and propagating an immune-reactive phenotype reflective of the pathogenesis of these diseases; thus, BDOs represent a platform for the study of the pathogenesis and therapy of cholangiopathies, particularly PSC.

Health related quality of life in chronic liver diseases.

BACKGROUND & AIMS: The impact of chronic liver diseases (CLDs) on health-related quality of life (HRQoL) is relevant to understand the burden of these conditions and inform decision-making processes related to their care. Studies simultaneously comparing the HRQoL of patients affected by the major CLDs to that of the general population are still lacking and are the subject of this study. METHODS: Using the EQ-5D-3L questionnaire, we analysed and compared HRQoL data from 2962 Italian patients affected by CLDs and forming a representative sample of the general Italian population (6800 individuals). Exploratory analyses were conducted to investigate the effects of each CLD on HRQoL, using the general population as reference and adjusting for possible confounders. RESULTS: Patients with CLDs (HBV, HCV, PSC, PBC, AIH, NAFLD/NASH) in the chronic hepatitis stage and with compensated cirrhosis (CC) showed HRQoL similar to the general population. However, AIH were more likely to report problems in self-care and lower EQ-5D VAS score, while NAFLD/NASH and HCV showed an increased risk of anxiety/depression. On the other hand, with progression to more advanced stages of liver disease (DC or HCC), HRQoL decreased significantly with higher risk of reporting problems in the physical domains, and significant reductions in the VAS and utility index scores. CONCLUSIONS: Different subtypes of CLD affected different QoL domains. This study therefore provides a real estimate of the impact of CLDs on patients' HRQoL, and represents a much needed tool to inform decision-making while assessing the effectiveness and cost-effectiveness of the care of these patients.

Randomized near-neighbor graphs, giant components and applications in data science.

If we pick n random points uniformly in [0, 1] d and connect each point to its c d log n-nearest neighbors, where d ≥ 2 is the dimension and c d is a constant depending on the dimension, then it is well known that the graph is connected with high probability. We prove that it suffices to connect every point to c d,1 log log n points chosen randomly among its c d,2 log n-nearest neighbors to ensure a giant component of size n - o(n) with high probability. This construction yields a much sparser random graph with ~ n log log n instead of ~ n log n edges that has comparable connectivity properties. This result has nontrivial implications for problems in data science where an affinity matrix is constructed: instead of connecting each point to its k nearest neighbors, one can often pick k' ≪ k random points out of the k nearest neighbors and only connect to those without sacrificing quality of results. This approach can simplify and accelerate computation; we illustrate this with experimental results in spectral clustering of large-scale datasets.

Determinants of Follow-up Care for Patients With Celiac Disease.

GOALS: This study aimed to investigate follow-up patterns among celiac disease (CD) patients. BACKGROUND: Gender factors are important in CD with women diagnosed more frequently than men despite equal seropositivity in screening studies. To determine if gender influences postdiagnosis care, we performed a retrospective cohort study investigating the impact of gender and mode of presentation on follow-up patterns after diagnosis. STUDY: The study included adults with biopsy-proven CD presenting to a single tertiary care center between 2005 and 2014. The primary exposure was at least 1 visit with a CD specialist. The primary outcome was ≥2 follow-up visits, including office visits and endoscopic procedures. Data extracted included whether patients had tissue transglutaminase antibodies performed by our laboratory. RESULTS: We analyzed 708 patients of which 70.5% were female. Follow-up was good with a majority of patients (69%) having at least 1 follow-up visit. On bivariate analysis, patients least likely to follow-up were ages 18 to 29 (P=0.03) and women with atypical presentations (P=0.003). After adjusting for potential confounders, individuals over age 65 were significantly more likely to attend at least 2 follow-up visits (odds ratio, 2.07; 95% confidence interval, 1.21-3.55; P=0.0079). Individuals with an abnormal baseline tissue transglutaminase antibody value in our laboratory were significantly more likely to follow-up (odds ratio, 1.99; 95% confidence interval, 1.39-2.85; P=0.0002). CONCLUSIONS: Gender had no impact on follow-up patterns despite prior studies demonstrating an impact on diagnosis rates. Future attention should focus on retaining young patients and those with atypical modes of presentation.

Community-Level Factors Associated with Hepatocellular Carcinoma Incidence and Mortality: An Observational Registry Study.

BACKGROUND: Hepatocellular carcinoma (HCC) incidence and outcomes vary across populations in the United States, but few studies evaluate local drivers of observed disparities. We measured HCC incidence at the community level and assessed community-level HCC risk factors with the goal of informing resource allocation to improve early case detection, which is associated with improved outcomes. METHODS: Clinical and demographic data including census tract of residence for all adults diagnosed with HCC in the Connecticut Tumor Registry between 2008 and 2019 were combined with publicly available U.S. Census and Centers for Disease Control and Prevention (CDC) data at the ZIP Code tabulation area (ZCTA) level. The average annual incidence of HCC was calculated for each ZCTA and associations between community-level characteristics, HCC incidence, stage at diagnosis, and survival were evaluated. RESULTS: Average annual HCC incidence during the study period was 8.9/100,000 adults and varied from 0 to 97.7 per 100,000 adults by ZCTA. At the community level, lower rates of high school graduation, higher rates of poverty, and rural community type were associated with higher HCC incidence. Persons with HCC living in the highest incidence ZCTAs were diagnosed at a younger age and were less likely to be alive at 1, 2, and 5 years after diagnosis. CONCLUSIONS: Community-level socioeconomic factors are strongly associated with HCC incidence and survival in Connecticut. IMPACT: This reproducible geo-localization approach using cancer registry, Census, and CDC data can be used to identify communities most likely to benefit from health system investments to reduce disparities in HCC.

Artificial intelligence in liver cancer - new tools for research and patient management.

Liver cancer has high incidence and mortality globally. Artificial intelligence (AI) has advanced rapidly, influencing cancer care. AI systems are already approved for clinical use in some tumour types (for example, colorectal cancer screening). Crucially, research demonstrates that AI can analyse histopathology, radiology and natural language in liver cancer, and can replace manual tasks and access hidden information in routinely available clinical data. However, for liver cancer, few of these applications have translated into large-scale clinical trials or clinically approved products. Here, we advocate for the incorporation of AI in all stages of liver cancer management. We present a taxonomy of AI approaches in liver cancer, highlighting areas with academic and commercial potential, and outline a policy for AI-based liver cancer management, including interdisciplinary training of researchers, clinicians and patients. The potential of AI in liver cancer is immense, but effort is required to ensure that AI can fulfil expectations.

Safety and Efficacy of Atezolizumab/Bevacizumab in Patients with Hepatocellular Carcinoma and Impaired Liver Function: A Systematic Review and Meta-Analysis.

Background: Safety and outcome of atezolizumab/bevacizumab in Child-Pugh B patients with hepatocellular carcinoma (HCC) have not been completely characterized. Objectives: In this study, we aimed at addressing safety and efficacy of atezolizumab/bevacizumab in Child-Pugh B patients by reviewing the available data and analyzing them by meta-analysis. Methods: We compared the safety and efficacy of atezolizumab/becavizumab treatment in patients with unresectable HCC and various degrees of liver dysfunction. A total of 8 retrospective, non-randomized, cohort studies were included in this meta-analysis, for a total of 1,071 Child-Pugh A and 225 Child-Pugh B patients. The albumin-bilirubin (ALBI) grade was also used to assess liver function, when available. Results: Grade ≥3 adverse events were observed in 11.8% of Child-Pugh class A and 26.8% class B patients (p = 0.0001), with an odds ratio (OR) of 0.43 (confidence interval [CI] 0.21-0.90; p = 0.02). Progression-free survival (PFS) at both 6 months (4.90 ± 2.08 vs. 4.75 ± 2.08 months; p = 0.0004) and 12 months (8.83 ± 2.32 vs. 7.26 ± 2.33 months; p = 0.002) was lower in Child-Pugh class B patients. A trend toward a higher objective response rate (ORR) was observed in Child-Pugh class A patients (219/856, 25.6%) as compared to Child-Pugh class B patients (25/138, 18.1%; p = 0.070), while the probability of obtaining an ORR was significantly greater in Child-Pugh A patients (OR 1.79, CI 1.12-2.86; p = 0.02). Median overall survival (OS) was 16.8 ± 2.0 and 6.8 ± 3.2 months in Child-Pugh A and B patients, respectively (mean difference 9.06 months, CI 7.01-11.1, p < 0.0001). Lastly, OS was longer in patients with ALBI grades 1-2 than in those with grade 3 (8.3 ± 11.4 vs. 3.3 ± 5.0 months, p = 0.0008). Conclusions: Oncological efficacy of atezolizumab/bevacizumab is moderate in Child-Pugh class B patients, and the shorter PFS and OS associated with the greater likelihood of experiencing treatment-related adverse events observed in these patients suggest great caution and individualization of treatment, possibly with the support of the ALBI grade.

Adherence to Tumor Board Recommendations in the Treatment of Patients with Hepatocellular Carcinoma.

Background: Hepatocellular carcinoma (HCC) is a heterogeneous disease that typically arises in the setting of chronic liver disease, making treatment selection complex. Multidisciplinary liver tumor boards (MDLTB) have been shown to improve outcomes in patients with HCC. However, in many cases, patients evaluated by MDLTBs ultimately do not receive the board's recommended treatment. Purpose: This study aims to assess adherence to MDLTB recommendations for the treatment of HCC, the reasons for non-adherence, and the survival of Barcelona Clinic Liver Cancer (BCLC) Stage A patients treated with curative treatment compared to palliative locoregional therapy. Patients and Methods: A single-site, retrospective cohort study was conducted of all patients with treatment-naïve HCC who were evaluated by an MDLTB at a tertiary care center in Connecticut between 2013 and 2016, of which 225 patients met inclusion criteria. Investigators conducted a chart review and recorded adherence to the MDLTB's recommendations, and in cases of discordance, evaluated and recorded the underlying cause; investigators assessed MDLTB recommendations' compliance with BCLC guidelines. Survival data was accrued through February 1st of 2022 and analyzed via Kaplan-Meier analysis and multivariate Cox regression. Results: Treatment adherent to MDLTB recommendations occurred in 85.3% of patients (n=192). The majority of non-adherence occurred in the management of BCLC Stage A disease. In cases where adherence was possible but the recommendation was not followed, most discrepancies were whether to treat with curative or palliative intent (20/24), with almost all discrepancies occurring in patients (19/20) with BCLC Stage A disease. For patients with Stage A unifocal HCC, those who received curative therapy lived significantly longer than patients who received palliative locoregional therapy (5.55 years vs 4.26 years, p=0.037). Conclusion: Most forms of non-adherence to MDLTB recommendations were unavoidable; however, treatment discordance in the management of patients with BCLC Stage A unifocal disease may present an opportunity for clinically significant quality improvement.

Spectral top-down recovery of latent tree models.

Modeling the distribution of high-dimensional data by a latent tree graphical model is a prevalent approach in multiple scientific domains. A common task is to infer the underlying tree structure, given only observations of its terminal nodes. Many algorithms for tree recovery are computationally intensive, which limits their applicability to trees of moderate size. For large trees, a common approach, termed divide-and-conquer, is to recover the tree structure in two steps. First, separately recover the structure of multiple, possibly random subsets of the terminal nodes. Second, merge the resulting subtrees to form a full tree. Here, we develop spectral top-down recovery (STDR), a deterministic divide-and-conquer approach to infer large latent tree models. Unlike previous methods, STDR partitions the terminal nodes in a non random way, based on the Fiedler vector of a suitable Laplacian matrix related to the observed nodes. We prove that under certain conditions, this partitioning is consistent with the tree structure. This, in turn, leads to a significantly simpler merging procedure of the small subtrees. We prove that STDR is statistically consistent and bound the number of samples required to accurately recover the tree with high probability. Using simulated data from several common tree models in phylogenetics, we demonstrate that STDR has a significant advantage in terms of runtime, with improved or similar accuracy.

Advancing diagnosis and management of liver disease in adults through exome sequencing.

BACKGROUND: Whole-exome sequencing (WES) is an effective tool for diagnosis in patients who remain undiagnosed despite a comprehensive clinical work-up. While WES is being used increasingly in pediatrics and oncology, it remains underutilized in non-oncological adult medicine, including in patients with liver disease, in part based on the faulty premise that adults are unlikely to harbor rare genetic variants with large effect size. Here, we aim to assess the burden of rare genetic variants underlying liver disease in adults at two major tertiary referral academic medical centers. METHODS: WES analysis paired with comprehensive clinical evaluation was performed in fifty-two adult patients with liver disease of unknown etiology evaluated at two US tertiary academic health care centers. FINDINGS: Exome analysis uncovered a definitive or presumed diagnosis in 33% of patients (17/52) providing insight into their disease pathogenesis, with most of these patients (12/17) not having a known family history of liver disease. Our data shows that over two-thirds of undiagnosed liver disease patients attaining a genetic diagnosis were being evaluated for cholestasis or hepatic steatosis of unknown etiology. INTERPRETATION: This study reveals an underappreciated incidence and spectrum of genetic diseases presenting in adulthood and underscores the clinical value of incorporating exome sequencing in the evaluation and management of adults with liver disease of unknown etiology. FUNDING: S.V. is supported by the NIH/NIDDK (K08 DK113109 and R01 DK131033-01A1) and the Doris Duke Charitable Foundation Grant #2019081. This work was supported in part by NIH-funded Yale Liver Center, P30 DK34989.