SURVIVAL DISPARITIES FOLLOWING LIVER TRANSPLANTATION VS. HEPATECTOMY FOR EARLY-STAGE HEPATOCELLULAR CARCINOMA

Background: The majority of HCC patients within Milan criteria, compete with cirrhosis patients with high MELD for liver transplantation based on an exceptional score which is the calculated MELD at registration, 28 after six months of waiting, 31 after 1 year and thereafter capped at 34 after one and a half years. In competitive UNOS regions, HCC patients may have to wait for more than a year before they receive a transplant. This prolonged waiting can lead to progression of their cancer and/or cirrhosis and loss of candidature for transplantation while may have been acceptable resection candidates at the beginning of their waiting period. In such regions, survival may be better with hepatectomy as it obviates the waiting.

Aim: To compare the survival of HCC patients with single tumor and low Child’s class after liver transplantation or hepatectomy.

Material and Methods: We used the UNOS and NCDB databases to make the comparison. Between 2004 to 2013, 13, 519 patients with HCC were registered for liver transplantation in the US. Of them 5, 338 were of Child’s class A; sufficient tumor details were available for 2,726 cases allowing us to select cases with single tumor of up to 5 cm size and within Milan criteria were included in the study. The NCDB database does not record preoperative staging information. Therefore, we selected patients with single tumor of up to 5 cm in size at the time of diagnosis and without major vascular invasion at pathologic staging after partial hepatectomy. 2, 507 patients were included. We assumed that these patients were Child’s class A or early B as the NCDB does not record CPT score . AKaplan-Meier survival analysis was performed. We used intention to treat principles and included all patients registered for transplantation whether or not they were transplanted. Survival analysis was was initiated at the time of listing for transplant. Reasons for removal from waiting list were summarized.

Results: The mean age was 58.8 years (std. dev. 7.3) and 77% were male in the transplant (UNOS) arm. The mean age was 62.4 years (std. dev. 11) and 71.8% were male in the hepatectomy (NCDB) arm. Kaplan-Meier survival estimates show that for the first 700 days patients have similar chances of survival. However, as time goes by, patients in the UNOS group experience better survival rates. The Log-rank test for equality of survivor functions, suggests that UNOS patients have better survival rates compared with NCBD (Chi-sq(1) = 83.8, p < 0.001).

Conclusion: In cirrhotic patients with HCC, survival after partial hepatectomy was non-inferior to that after transplantation for the first two years for low Child’s class liver disease. Resection should be strongly considered in eligible patients and this strategy could improve organ allocation.

Introduction
Unplanned care (readmission and emergency room (ER) visits) is a Center for Medicare Services measure of quality of care. Patients who undergo major gastrointestinal (GI) surgery are at high risk for postoperative complications which lead to unplanned care. We implemented a pilot program of postoperative outpatient monitoring using an FDA-approved wearable biometric monitoring device (BMD) to identify patients who suffered setbacks to mitigate severity of significant complications and to facilitate intervention for minor setbacks so as to prevent unplanned care.

Methods
Adult patients who underwent high-risk GI operations from the colorectal and hepatopancreatobiliary services were prospectively enrolled in a quality improvement remote monitoring project from discharge until 30 days after surgery. The BMD relayed biometric data (temperature, heart rate, respiratory rate, and activity) to a cloud-based monitoring dashboard. Patients with abnormal data (alerts) were contacted and, if necessary, care was escalated to their clinical team. In-hospital death and a home-monitoring period of less than 7 days were exclusion criteria. The primary outcome was rate of avoidance of unplanned care. Secondary outcomes were 30-day ER visit rates, readmission rates, and their associated monitoring device-detection rate.

Results
A total of 129 patients were enrolled but 25 (19.4%) patients were excluded from analysis due to death (n=2) and monitoring <7 days (n=23). Of the 104 patients included in analysis, the average age was 60±13.4 year and 40 (38.5%) patients were female. We contacted 78 (75.0%) patients at least once in response to 192 alerts. In 17 (8.9%) of 192 cases, the setback was managed in the outpatient setting successfully avoiding unplanned care. In 145 (75.5%) cases, escalation was not required and in 37 (19.3%) cases, the patients could not be reached. There were a total of 22 ER visits by 18 (17.3%) patients. Nineteen patients (18.3%) had a total of 24 readmissions of which 5 (20.8%) were direct and not via the ER. BMD alerts resulted in 8 patient (7.7%) escalations that translated to ER visits and/or readmission. In 7 instances (6.7%) gaps in biometric data interfered with our ability to detect setbacks that resulted in unplanned care.

Conclusion
Patients who undergo major gastrointestinal operations can be safely monitored remotely. The technological and logistic complexity of monitoring outpatients are significant and require further optimization. Biometric findings that merit clinical alerts in the postoperative setting require refinement. However, this pilot data suggests remote monitoring has the potential to both reduce unplanned care and assist in detection of clinical deterioration.

Introduction. Chronic inflammation leads to liver fibrosis and, potentially, cirrhosis. Hepatic stellate cell (HSC) activation represents the initial step of liver fibrogenesis since the HSC is the major producer of extracellular matrix (ECM). Understanding the mechanism of inflammation and fibrogenesis is critically important to developing treatments for liver fibrosis. NF-ΚB is a key inflammatory signaling pathway, and the survival of activated HSC was found to be NF-KB dependent. Using a fragment-based drug design approach, our team previously designed and synthesized a group of small molecule compounds for cancer therapy. After screening these molecules in cancer cells and activated HSC In vivo and In vitro, HJC0416 was identified to be a novel, orally bioavailable molecules with potent anti-cancer, anti-inflammatory and anti-fibrotic effects. For example, HJC0416 induced activated HSC cell cycle arrest and apoptosis, inhibited endogenous and TGFβ-stimulated ECM expression. However, the molecular mechanisms of HJC0416’s anti-fibrogenetic effects in HSC remain largely unknown. In this study, we examined the effects of HJC0416 on NF-KB and its associate factor HSP90 in HSC in vitro.
Methods. Activated human HSC line LX-2 was treated with either JHC0416 or 17-AAG, then exposed to TNFα or TGFβ as indicated. Nuclear and cytosolic proteins were isolated for Western blots or immunofluorescence assay.
Results. HJC0416 treatment significantly attenuated TNFα–induced IκBα phosphorylation, NF-KBp65 nuclear translocation and DNA binding activity. Endogenous and TNFα–induced p65 phosphorylation of Ser536 was suppressed by HJC0416. Notably, HJC0416 treatment dose-dependently down-regulated the expression of IKKβ, RIP1 and AKT, FAK, CDK9, all of which are HSP90 interacting proteins, suggesting that HSP90 may be involved in HJC0416 regulated-NF-KB signaling and fibrogenesis in HSC. Our results confirmed that HSP90 specific inhibitor 17-AAG prevented TNFα-induced IκBα phosphorylation and degradation, p65 nuclear translocation and DNA binding. Similar with previous HJC0416 data, 17-AAG inhibited endogenous and TGFβ-stimulated fibrosis markers collagen type I and fibronectin. Conclusion. The anti-fibrogenetic effect of orally bioavailable compound HJC0416 is through the HSP90/NF-KB pathway. HJC0416 may be a promising drug candidate for liver fibrosis treatment.

Background
Intra-abdominal infections have a high morbidity and mortality rate and require early diagnosis and management. The COVID-19 pandemic makes care disruptions and delayed presentation an important concern. Little is known about the disparity in excess mortality and racial/ethnic disparities in intra-abdominal infections-related mortality before and during the pandemic through September 2022.
Methods
Death records among Americans aged ≥25 years were obtained from the National Vital Statistics System (NVSS), which reports over 99% of deaths in the United States. Using ICD-10 codes, intra-abdominal infections were defined as acute cholecystitis, acute appendicitis, C. difficile colitis, diverticulitis, and pyelonephritis. We calculated age-standardized mortality rates (ASMR) (per 100,000 population) and years of potential life lost (YPLL) from 1999 to 2021. Predicted ASMRs and YPLLs during the pandemic based on the observed trend from 1999 to 2019 were estimated. Excess death was defined as the percentage difference between the observed and the predicted values. Joinpoint analysis was used to calculate annual percentage change (APC). Monthly ASMR was updated through September 2022.
Results
From 1999 to 2021, there were 469,655 intra-abdominal infections-related deaths. Of these, 82.4% were Whites, followed by Blacks (8.6%), Hispanics (6.32%), Asians/Pacific Islanders (2.06%), and American Indians/Alaska Natives (0.59%). The overall intra-abdominal infections-related ASMR increased from 2001 to 2006 (APC 6.9, P<0.001), followed by a decreasing trend from 2006 to 2015 (APC -0.8, P<0.05), and sharply decreased to 8.38 in 2019 (APC -5.3, P<0.01). During the pandemic, ASMR rose to 8.64 in 2020 and 9.23 in 2021, leading to excess mortality of 10.6% in 2020 and 29.4% in 2021. (Table 1) Excess mortality was observed across all racial/ethnic groups in 2020, with Hispanics having the highest excess mortality rates, followed by Whites, while Blacks had the lowest. Focusing on deaths below the age of 65, significant racial/ethnic disparities in excess YPLLs, with Hispanics showing the greatest excess YPLL and Whites experiencing the lowest. Furthermore, the monthly trend of ASMR updated through 9/2022 showed persistent excess death, with peaked ASMR occurring in 1/2022 and no significant difference in the decreasing trends between 1-9/2021 and 1-9/2022. Notably, the gaps between predicted and observed ASMR in both Blacks and Whites widened in 2022 when compared to 2021.
Conclusions
We observed a large increase in intra-abdominal infections-related death during the pandemic in the U.S. Both ASMR and YPLL increased across all racial/ethnic groups but also widened existing disparities during the COVID-19 pandemic. Hispanics with intra-abdominal infections experienced a disproportionate excess death compared to Whites and Blacks.

Background: Postoperative complications are associated with mortality in geriatric patients following major abdominal operations. Predicting which geriatric patients are at the highest risk for mortality after a complication may aid in preventing deaths. Failure to rescue (FTR) is a publicly reported metric of quality and is broadly defined as mortality after a potentially preventable complication following surgery. Typically, a 30-day follow-up period is used for reporting FTR, but there is concern that this inadequately captures postoperative deaths; instead, a 90-day follow-up period has been advocated. Our objective was to examine the association of a validated frailty metric, the Risk Analysis Index (RAI-A), with 90-day FTR (FTR-90).

Methods: Patients 65 years or older who underwent a major abdominal operation between 2014-2020 at a quaternary care center were abstracted. Institutional data, the ACS NSQIP Procedure Targeted Participant Use Data, and Geriatric Surgery Research File were merged and 90-day clinical outcomes were collected. Univariable analysis compared baseline characteristics. Multivariable logistic regression was used to evaluate the association between RAI-A and FTR-90.

Results: Of the 1565 patients included in the institutional data abstraction, 398 patients had a postoperative complication and were included. Fifty-two (13.0%) of the patients with a postoperative complication died during 90-day follow-up. Of the 398 patients, 178 (44.7%) underwent colorectal surgery, 135 (33.9%) underwent hepatopancreatobiliary surgery, and 85 (21.4%) underwent other types of major abdominal operations. Patients who experienced FTR-90 were older (median age 76 vs 73, p = 0.002), had a greater preoperative ASA classification (p < 0.001), and had a higher NSQIP estimated risk of morbidity (0.33% vs 0.20%, p < 0.001) and mortality (0.067% vs 0.012%, p < 0.001). The FTR-90 group had a greater median RAI-A score (23 vs 19, p = 0.002). A greater proportion of patients with FTR-90 experienced a major complication after surgery (pneumonia, re-intubation, prolonged ventilation, myocardial infarction, cardiac arrest, pulmonary embolism, renal failure, sepsis, septic shock, deep wound or organ space infection, and stroke) (86.5% vs 47.7%, p < 0.0001). The RAI-A score was independently associated with FTR-90 (OR 1.04, 95% CI 1.0042 – 1.077, p = 0.028), but was not independently associated with FTR-30 (p = 0.13).

Conclusion: Pre-operative frailty, as defined by RAI-A, is independently associated with failure to rescue at 90 days, but not 30 days, after major abdominal surgery. The use of a 90-day failure to rescue definition captured nearly 57% more deaths compared to 30-day failure to rescue. Frailty has major implications beyond the typical 30-day follow-up period and a longer follow-up period must be considered in this patient population.

OBJECTIVES: Neoadjuvant therapy (NT) is increasingly used for many localized gastrointestinal (GI) cancers. However, there is a paucity of data assessing patient experience during NT and its impact on health-related quality of life (HRQOL). Given the limitations of traditional surveys administered at regularly scheduled physician appointments, we customized a smartphone app for prospective real-time assessment of patient experience during NT.

METHODS: Patients with locally GI cancers receiving NT were prospectively enrolled if age>18, English-speaking, owned a smartphone, and were treatment naive. A customized version of the SeamlessMD mobile application was developed with engagement from patient and physician stakeholders. HRQOL, measured using the Functional Assessment of Cancer Therapy-General (FACT-G) questionnaire, was pushed directly through the smartphone app at baseline, every 30 days, and at completion of NT. Participants were also encouraged to use the mood tracker, symptom tracker, and free-text journaling as often as desired during NT. Mean total overall and sub-section HRQOL scores were calculated at each time point during NT and differences assessed using one way repeated measured ANOVA. Statistical significance was achieved if p<0.05.

RESULTS: Among 104 enrolled patients, mean age was 60.4±11.5 years and 55.2% were male. The most common cancer diagnoses were rectal (38.5%), pancreatic (37.5%), and esophageal (15.4%). After a mean 3.0 months of NT, 71 patients (68.3%) underwent surgical resection whereas 10 (9.6%) had a complete clinical response resulting in active surveillance, and 25 (24.0%) were unable to have surgery because of disease progression, unresectable disease, or poor performance status. Main side effects of NT were nausea, neutropenia, and fatigue (11.5%, 9.6%, 6.7%, respectively). Mean overall FACT-G scores did not significantly change during NT (Figure 1, p=0.73). Functional wellbeing (FWB) scores were consistently the lowest (mean score 15.0) whereas social wellbeing (SWB) scores were the highest (mean score 21.5) at all time points. FACT subscores did not change significantly during NT. The most common symptoms reported during NT were fatigue, trouble sleeping, and anxiety (39.3%, 34.5%, and 28.3% of patient entries, respectively). Qualitative analysis of free-text journaling entries identified anxiety, fear, and frustration as the most common themes, but also the importance of social support systems and confidence in their health care providers.

CONCLUSIONS: While patient symptom burden remains high, results of this prospective cohort study suggest HRQOL is maintained during NT for localized GI cancers. Future research will evaluate whether patient-reported measures are associated with outcomes of NT and/or surgical resection.

Background

We examined the disparities in undergoing liver transplantation at minority-serving hospitals (MSH) versus non-MSH among patients with early-stage hepatocellular carcinoma (HCC). In addition, we investigated associations between liver transplantation and overall survival (OS), stratified by MSH status.

Methods

Patients with early HCC, defined as cT1, were identified within the National Cancer Database (2004-2018). The primary outcome was undergoing liver transplantation. The MSH status for each hospital was determined based on the proportion of minority (non-Hispanic Black or Hispanic) patients, where hospitals were ranked based on the proportion of minority patients and those in the top decile were considered MSH.

Results

A total of 46,703 patients with early HCC were identified, of whom 4214 (9.0%) were treated at MSH. Patients treated at MSH were more likely than those treated at non-MSH to be younger, live in a metropolitan area, and travel a shorter distance to treating facility. However, patients treated at MSH were less likely to live in areas with higher education and median income, have private insurance, and receive care at high-volume hospitals.
Throughout the study period, there was a declining trend for undergoing liver transplantation to patients with early HCC at both MSH and non-MSH. However, patients treated at MSH were less likely to undergo liver transplantation than patients treated at non-MSH (11.6% vs. 19.3%, OR=0.70, 95% CI: 0.59-0.83). Minority patients who were treated at non-MSH were less likely to undergo liver transplantation than White patients (OR=0.82, 95% CI: 0.75-0.90). However, minority patients had a further associated decrease in the likelihood of undergoing liver transplantation when treated at MSH (OR=0.65, 95% CI: 0.46-0.93) as compared to White patients.
Patients who underwent liver transplantation had an associated improvement in OS compared to those who did not when treated either at non-MSH (median OS: 147.4 vs. 24.6; HR=0.30, 95% CI: 0.28-0.33; Figure) or MSH (median OS: not reached vs. 26.5 months; HR=0.25, 95% CI: 0.19-0.34). Regardless of MSH status, there were no clinically meaningful differences in OS between White and minority patients who underwent liver transplantation.

Conclusions

Patients with early HCC had an associated decrease in the likelihood of undergoing liver transplantation when treated at MSH. Compared to White patients, minority patients treated at non-MSH had an associated decrease in the likelihood of undergoing liver transplantation, but to a lesser extent when treated at non-MSH. Liver transplantation was associated with improved survival outcomes regardless of race or MSH status.

Introduction: The impact of Medicaid Expansion (ME) on complex gastrointestinal cancers such as hepatocellular carcinoma (HCC) remains controversial. Despite improving overall health insurance coverage, the heterogeneous impact of ME on access to cancer care may be related to sociodemographic factors. We sought to evaluate the association between ME and receipt of surgical treatment among patients with early-stage HCC.

Methods: Patients diagnosed with early-stage HCC between the ages of 40 and 64 were identified from the National Cancer Database; patients were divided into pre- (2004-2012) versus post- (2015-2017) ME cohorts. Multivariable logistic regression was performed to assess predictors of receipt of surgical treatment. A Difference-in-Difference (DID) analysis was used to evaluate changes in surgical treatment among patients living in ME versus non-ME states.

Results: Among 19,745 patients, 12,220 (61.9%) individuals were diagnosed before the implementation of ME and 7,525 (38.1%) after. Although overall rates of receipt of surgical treatment decreased in the post-expansion era among ME and non-ME states (ME state, pre-ME era: 62.2% vs post-ME era: 51.6%; non-ME state, pre-ME era: 62.1% vs post-ME era: 50.8%; both p<0.001), this trend varied relative to insurance status. Of note, among uninsured/Medicaid patients living in ME states (n=2,249, 47.7%), receipt of surgical treatment was more likely to occur in the post-ME era (pre-ME era: 48.1%; post-ME era: 52.3%, p<0.001). Moreover, among these patients, treatment at an academic (OR 1.35, 95%CI 1.10-1.65) or high-volume facility (HVF) (OR 1.45, 95%CI 1.18-1.78) was associated with an increased likelihood of undergoing surgical treatment in the pre-expansion period (both p<0.01). In contrast, in the post-expansion era, treatment at an academic facility (OR 1.83, 95%CI 1.47-2.30) and living in an ME state at the time of diagnosis (OR 1.28, 95%CI 1.07-1.54) were predictors of receipt of surgical treatment (both p<0.01). DID analysis demonstrated that uninsured/Medicaid patients living in ME states had increased utilization of surgical treatment relative to individuals in non-ME states (uninsured/Medicaid: 6.4%, p<0.05), although no differences were noted relative to other insurance types (private: -2.0%, other government: 0.3%, overall: 0.7%, all p>0.05)(Figure).

Conclusion: While the implementation of ME did not homogeneously impact receipt of surgery, uninsured/Medicaid patients with early-stage HCC residing in ME states had an increased utilization of surgical treatment after expansion. The data demontrate that ME had a beneficial effect on surgical utilization among the most vulnerable patient populations with HCC.

Background:
Acute liver failure affects patients of every demographic. Etiology of liver failure varies and has a wide array of treatment options. Orthotopic liver transplantation (OLT) remains the treatment of choice for patients who fail medical management. We sought to identify if racial disparities affect the survival of patients with acute liver failure.

Methods:
The National Inpatient Sample (NIS) database (2016-2020) was analyzed to identify adult (age > 18 years) patients admitted with ICD-10 CM/PCS codes for acute liver failure and OLT respectively. Multivariate logistic regression analysis was used to estimate the odds ratios of in-hospital mortality. Multivariate linear regression was used to determine the average length of hospital stay (LOS) and average total hospitalization charges. Weighted analysis using Stata 17 MP was performed. Patient age, race, and gender were controlled during regression analyses.

Results:
A total of 668 (n=668) patients underwent OLT with diagnoses of acute liver failure between 2016-2020. Of these 81 (12.1%) were black. As demonstrated in Table 1, multivariate logistic regression analysis was performed. Black patients were found to have a higher mortality rate (OR 2.55, CI 1.18-5.51, p<0.05) when compared to white patients. There was no statistically significant difference in mortality for Hispanic, Asian and Pacific Islander, or Native American patients.

Asians and Pacific islanders were found to have a decreased average LOS (-8.98 days, CI -15.2 - -2.71, p<0.01) when compared to white patients. There was no statistically significant variation in LOS and total hospital charges among different races.

Conclusions:
Black patients have worse in-hospital mortality when undergo OLT secondary to acute liver failure, even in the modern era. Black patients have historically worse outcomes secondary to lower socioeconomic status and access to care. Despite improvement in critical care, fairer organ allocation policy and access to care, Black patients still have worse outcomes. Additional resources should be made available to rectify these findings.

BACKGROUND: Hepatocellular carcinoma (HCC) with macroscopic vascular invasion is frequently addressed with non-curative treatments, especially systemic chemotherapy. Otherwise, some centers advocate liver resection (LR) for HCC with non-main trunk portal vein invasion. In fact, the best approach for this subgroup of patients is controversial. The aim of this study was to evaluate the outcomes of different treatments for patients with HCC and macroscopic vascular invasion.
METHODS: A systematic review and meta-analysis following the PRISMA recommendations were performed. The search was performed in PubMed, Embase, Cochrane (CENTRAL), and LILACS/BVS. We included retrospective or prospective studies that compared LR with other types of treatment, including transarterial chemoembolization (TACE) and systemic chemotherapy, in patients with HCC and macrovascular invasion. Risk of bias was performed with Robins-I and certainty assessment with GRADEPro. Results were expressed as risk difference (RD) with the corresponding 95% confidence interval (95% CI). The I² statistics were applied to investigate statistical heterogeneity, and a random model was used. A network analysis was used to compare LR with TACE and chemotherapy.
RESULTS: The initial search found 890 articles, and after applying eligibility criteria, 13 studies were finally included. All included studies were observational, with the mean age across the studies ranging from 47 to 65 years. There was a male predominance in all studies (range 75 to 95%). All studies comprised mainly Child-Pugh A patients (range 73 to 100%). LR had similar mortality to non-LR alternatives (RD= 0.00; 95% CI -0.00 to 0.00; I²=1.5%). LR had a higher rate of complications than non-LR therapies (RD= 0.06; 95% CI 0.00 to 0.12; I²=62.5%). LR showed a higher 3-year overall survival rate (RD= 0.12; 95% CI 0.05 to 0.20; I² 85%). In the network meta-analysis, the risk of death in 3-year follow-up was lower in the LR group than in the TACE group (network RD: -0.149; 95% CI -0.227 to -0.071) and than the chemotherapy group (network RD: -0.117; 95% CI -0.006 to -0.227). Chemotherapy and TACE had similar survival outcomes (network RD: -0.033; 95% CI -0.148 to 0.083). There was no significant inconsistency. See Figure 1. The main risk of bias was related to the risk of selection bias, and certainty was very low.
CONCLUSION: LR was associated with a higher risk of procedure-related complications; however, offers a higher chance for long-term survival than non-LR alternatives in patients with HCC and macroscopic vascular invasion.

BACKGROUND
Systemic inflammation is important in the development of intrahepatic cholangiocarcinoma (iCCA). However, controversial results exist on the role of inflammatory scores in patients undergoing resection for iCCA. The aim of this study was to explore the value of different blood inflammatory indexes for predicting the prognosis of resected iCCA patients.
METHODS
A retrospective cohort of patients who underwent hepatic resection for iCCA between January 2010 and January 2021 was analyzed. Neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), and platelet-to-lymphocyte ratio (PLR) together with all the traditional clinical-pathological prognostic factors were recorded. Cutoff values for NLR, LMR and PLR were calculated by using the Yale X-tile software. Kaplan-Meier and Cox regression analyses were conducted to identify independent prognostic factors.
RESULTS
A total of 101 patients were considered. NLR>3.83 and LMR<2.28 were found to be associated with worse overall survival (OS), while PLR was not. Patients were divided into two groups: 68 (67.3%) patients in the low-risk group (NLR<3.83 and LMR>2.28) and 33 (32.7%) patients in the high-risk group (having at least one or both worse prognostic ratios). After a median follow-up of 44.6 months (IQR 29.6-95.7), the 5-year OS was 22.2% for the low-risk group and 9.0% for high-risk group (P=0.008; Fig.1). At multivariate analysis being more than 65 year-old (HR=2.149; 95%CI=1.086-4.255), elevated Ca19.9 (HR=2.102; 95%CI=1.165-3.793) and being in the high-risk group (HR=1.92; 95%CI=1.045-3.532) were found to be independently associated with worse OS. A patient with elevated Ca19.9 in the high-risk group had 2.063 HR of mortality (P=0.042; Fig.2a) and 2.216 HR of disease recurrence (P=0.007; Fig.2b).
CONCLUSIONS
The combination of blood inflammatory indexes resulted in the identification of two patients’ groups with different risk of survival after resection for iCCA. Adding Ca19-9 allows also the identification of patients at increased risk of recurrence.

Introduction
Hepatocellular carcinoma (HCC) is a leading cause of cancer-related deaths worldwide. In the United States, the incidence and mortality of HCC are expected to continue increasing over the decade. Furthermore, disparities have been described in access to different treatment modalities. Using a nationwide database, we sought to identify survival disparities following liver transplantation (LT) or hepatectomy.

Methods
We conducted a retrospective analysis of the National Cancer Database. Black and White patients with Stage I-II HCC were identified, and two separate analyses were conducted for patients who underwent LT or hepatectomy. Groups were compared using univariable analyses. Survival analyses were conducted through the Kaplan-Meier method, log-rank test, and multivariable Cox regression.

Results
We identified 8714 patients who underwent LT of whom 10.5% were Black, and 89.5% were White. Compared to White patients, Black patients were more likely to be younger (58 years, IQR 54-63 vs. 59 years, IQR 55-64, p<0.001), female (29% vs. 22%, p<0.001), government-insured (48% vs. 44%, p=0.002), had lower income (44% vs. 21%, p<0.001), were closer to their treatment center (47% vs. 21%, p<0.001), and were more likely to receive chemotherapy (60% vs. 52%,p<0.001). Median survival was shorter in Black patients (143 months, 95% CI 132-167 vs. 167 months, 95% CI 156-NA, p=0.02) compared to White patients. In the multivariable analysis, Black race was an independent predictor of decreased survival (HR 1.18, 95% CI 1.04-1.35, p=0.013).
In patients who underwent hepatectomy, there were a total of 11,006 patients identified, of whom 81.2% were White, and 18.8% were White. Compared to White patients, Black patients were more likely to be younger (62 years, IQR 57-67 vs. 66 years IQR 59-73, p<0.001), female (31% vs 28%, p=0.010), treated at academic facilities (67 vs. 58%), uninsured (4% vs 2%, p<0.001), had lower median income (50% vs 22%, p<0.001), more likely to have ≥ 2 comorbidities (26% vs 23%, p=0.035) to be stage II (30% vs 27%, p=0.022), and were closer to their treatment center (64% vs 36%, p<0.001). Median survival was similar between both groups (69 months, 95% CI 62-75 vs. 66 months, 95% CI 64-69%, p=0.73). Race was not an independent predictor of survival in patients who underwent hepatectomy.

Conclusion
We identified survival disparities in Black patients who underwent LT for HCC. Conversely, there were no survival disparities identified following hepatectomy. While the cause is likely multifactorial, the complex management surrounding LT might provide barriers related to care. Prompt investigation is needed to address ongoing racial disparities.