Alpha-fetoprotein: Past, present, and future.

Alpha-fetoprotein (AFP) is a glycoprotein that plays an important role in immune regulation with critical involvement in early human development and maintaining the immune balance during pregnancy. Postfetal development, the regulatory mechanisms controlling AFP undergo a shift and AFP gene transcription is suppressed. Instead, these enhancers refocus their activity to maintain albumin gene transcription throughout adulthood. During the postnatal period, AFP expression can increase in the setting of hepatocyte injury, regeneration, and malignant transformation. It is the first oncoprotein discovered and is routinely used as part of a screening strategy for HCC. AFP has been shown to be a powerful prognostic biomarker, and multiple HCC prognosis models confirmed the independent prognostic utility of AFP. AFP is also a useful predictive biomarker for monitoring the treatment response of HCC. In addition to its role as a biomarker, AFP plays important roles in immune modulation to promote tumorigenesis and thus has been investigated as a therapeutic target in HCC. In this review article, we aim to provide an overview of AFP, encompassing the discovery, biological role, and utility as an HCC biomarker in combination with other biomarkers and how it impacts clinical practice and future direction.

Association of Residency Applicants' Hobbies and Subsequent Technical Proficiency Ratings as General Surgery Chief Residents: A Multi-Institutional Study.

OBJECTIVE: To determine whether participation in certain hobbies (e.g., participation in sports, playing musical instruments, or other hobbies requiring fine motor skills), preresidency, are associated with higher technical skills ratings at the time of residency graduation. DESIGN: Faculty members from 14 general surgery residency programs scored individual graduates from 2017 to 2020 on their technical skills using a 5-point Likert scale. Hobbies for these residents were collected from their Electronic Residency Application Service (ERAS) data. A single reviewer classified each ERAS hobby into predefined categories including musical instruments, sports requiring hand-eye coordination, team sports, and activities necessitating hand-eye coordination. Spearman correlation coefficients were calculated for the relationship between each category of hobby-as well as the total number of hobbies in each category-and the outcome of surgical faculty ratings of residents' technical surgical skills during their last year of residency. A proportional odds model including the above predictive variables was also fit to the data. SETTING: Fourteen general surgery residency programs. PARTICIPANTS: General surgery residency graduates from 14 different programs from 2017 to 2020. RESULTS: There were 296 residents across 14 institutions. The average ranking of residents' technical skills was 3.24 (SD 1.1). A total of 40% of residents played sports involving hand-eye coordination, 31% played team sports, 28% participated in nonsport hobbies that require eye-hand coordination, and 20% played musical instruments. Correlation coefficients were not statistically significant for any of the categories. In the proportional odds model, none of the variables were associated with statistically significant increased odds of a higher technical skills rating. CONCLUSIONS: There was no correlation between general surgery chief residents' technical skills as rated by faculty, and self-reported pre-residency hobbies on the ERAS application. These findings suggest such hobbies prior to residency are unlikely to predict future technical skills prowess.

Portable hypothermic oxygenated machine perfusion for organ preservation in liver transplantation: A randomized, open-label, clinical trial.

BACKGROUND AND AIMS: In liver transplantation, cold preservation induces ischemia, resulting in significant reperfusion injury. Hypothermic oxygenated machine perfusion (HMP-O 2 ) has shown benefits compared to static cold storage (SCS) by limiting ischemia-reperfusion injury. This study reports outcomes using a novel portable HMP-O 2 device in the first US randomized control trial. APPROACH AND RESULTS: The PILOT trial (NCT03484455) was a multicenter, randomized, open-label, noninferiority trial, with participants randomized to HMP-O 2 or SCS. HMP-O 2 livers were preserved using the Lifeport Liver Transporter and Vasosol perfusion solution. The primary outcome was early allograft dysfunction. Noninferiority margin was 7.5%. From April 3, 2019, to July 12, 2022, 179 patients were randomized to HMP-O 2 (n=90) or SCS (n=89). The per-protocol cohort included 63 HMP-O 2 and 73 SCS. Early allograft dysfunction occurred in 11.1% HMP-O 2 (N=7) and 16.4% SCS (N=12). The risk difference between HMP-O 2 and SCS was -5.33% (one-sided 95% upper confidence limit of 5.81%), establishing noninferiority. The risk of graft failure as predicted by Liver Graft Assessment Following Transplant score at seven days (L-GrAFT 7 ) was lower with HMP-O 2 [median (IQR) 3.4% (2.4-6.5) vs. 4.5% (2.9-9.4), p =0.024]. Primary nonfunction occurred in 2.2% of all SCS (n=3, p =0.10). Biliary strictures occurred in 16.4% SCS (n=12) and 6.3% (n=4) HMP-O 2 ( p =0.18). Nonanastomotic biliary strictures occurred only in SCS (n=4). CONCLUSIONS: HMP-O 2 demonstrates safety and noninferior efficacy for liver graft preservation in comparison to SCS. Early allograft failure by L-GrAFT 7 was lower in HMP-O 2 , suggesting improved early clinical function. Recipients of HMP-O 2 livers also demonstrated a lower incidence of primary nonfunction and biliary strictures, although this difference did not reach significance.

Treating rare tumors with liver transplantation.

PURPOSE OF REVIEW: The success of liver transplantation (LT) in treating unresectable hepatocellular carcinoma (HCC) has resulted in interest in LT for other oncologic conditions. Here, we discuss the role of LT for rare oncologic indications including metastatic gastroenteropancreatic neuroendocrine tumors (GEP-NETs), hepatic epitheliod hemangioendothelioma (HEHE), fibrolamellar hepatocellular carcinoma (FLC), and hepatic angiosarcoma (HAS). RECENT FINDINGS: Conditions reviewed have been documented indications for LT in the available literature. We summarize the experience of LT for these indications and proposed management guidelines. SUMMARY: GEP-NETs with isolated metastases to the liver can be treated with LT with excellent long-term outcomes (10-year survival 88%) if strict selection criteria are used (low-intermediate grade, Ki-67% < 20%, complete resection of primary tumor, stable disease for 6 months, <50% hepatic involvement). HEHE is a rare hepatic tumor for which LT can be performed with reported 10-year survival around 70%. FLC is a distinct clinical entity to HCC and is optimally treated with surgical resection though experience with LT is described in observational series (5-year survival 50%, recurrence in 10%). HAS is a rapidly progressive tumor with a dismal prognosis with or without treatment, including LT.

Single center experience using stereotactic body radiation therapy (SBRT) on orthotopic liver transplant protocol for unresectable cholangiocarcinoma.

PURPOSE: To evaluate tolerability, pathologic response, and disease outcomes utilizing pre-operative stereotactic body radiation therapy (SBRT) followed by consolidation chemotherapy (CHT) prior to orthotopic liver transplant (OLT) in unresectable cholangiocarcinoma (CCA). METHODS: This was a retrospective chart review of patients treated on OLT protocol at a single tertiary center from 2012 to 2019. Patients received pre-operative SBRT (40-50 Gy in 5 fractions) followed by CHT until progression or OLT. Progression-free survival (PFS) and overall survival (OS) were compared via log-rank test and Cox proportional hazards regression. RESULTS: 26 patients (84.6% hilar, 15.4% intrahepatic) were identified for analysis. Eight patients (30.8%) patients developed acute toxicity after SBRT, mostly grade 1 nausea. Nine (34.6%) patients underwent OLT of which 4 (44.4%) achieved a pathologic complete response (pCR). Five (55.6%) OLT patients, including 2 pCR, developed recurrence at a median time of 49.9 weeks after OLT. 3-year OS for the OLT and dropout cohort was 75% and 9%, respectively (p < 0.0001). OS in hilar tumors only was statistically different for those that achieved a pCR (p = 0.014). CONCLUSIONS: Pre-operative SBRT is a well-tolerated and effective radiation technique as part of OLT protocol for unresectable CCA and conferred in a pCR rate of 44% within our cohort.

Comprehensive tissue deconvolution of cell-free DNA by deep learning for disease diagnosis and monitoring.

Plasma cell-free DNA (cfDNA) is a noninvasive biomarker for cell death of all organs. Deciphering the tissue origin of cfDNA can reveal abnormal cell death because of diseases, which has great clinical potential in disease detection and monitoring. Despite the great promise, the sensitive and accurate quantification of tissue-derived cfDNA remains challenging to existing methods due to the limited characterization of tissue methylation and the reliance on unsupervised methods. To fully exploit the clinical potential of tissue-derived cfDNA, here we present one of the largest comprehensive and high-resolution methylation atlas based on 521 noncancer tissue samples spanning 29 major types of human tissues. We systematically identified fragment-level tissue-specific methylation patterns and extensively validated them in orthogonal datasets. Based on the rich tissue methylation atlas, we develop the first supervised tissue deconvolution approach, a deep-learning-powered model, cfSort, for sensitive and accurate tissue deconvolution in cfDNA. On the benchmarking data, cfSort showed superior sensitivity and accuracy compared to the existing methods. We further demonstrated the clinical utilities of cfSort with two potential applications: aiding disease diagnosis and monitoring treatment side effects. The tissue-derived cfDNA fraction estimated from cfSort reflected the clinical outcomes of the patients. In summary, the tissue methylation atlas and cfSort enhanced the performance of tissue deconvolution in cfDNA, thus facilitating cfDNA-based disease detection and longitudinal treatment monitoring.

Super-massive transfusion during liver transplantation.

BACKGROUND: Massive hemorrhage and transfusion during liver transplantation (LT) present great challenges. We aimed to investigate the incidence and risk factors for super-massive transfusion (SMT) and survival outcome and factors that negatively affect survival in patients who received SMT during LT. STUDY DESIGN AND METHODS: We included adult patients undergoing LT from 2004 to 2019. SMT was defined as transfusion of ≥50 units of red blood cells (RBC) during LT. Independent risk factors were identified by multivariable logistic regression. Ninety-day survival was recorded and factors that negatively affected survival were analyzed by the Cox survival test. RESULTS: Of 2772 patients, 158 (5.6%) received SMT during LT. Mean RBC transfusion was 72.6 (±23.4) units with a maximum of 168 units. Four variables (MELD-Na score, previous upper abdominal surgery, portal vein thrombosis, and remote retransplant) were independent risk factors for SMT (odds ratio 1.800-8.274, 95% CI 1.008-16.685, all p < .005). The 90-day survival rate in SMT patients was 81.6%. Preoperative pulmonary hypertension and massive postreperfusion transfusion negatively affected 90-day survival (hazard ratio 2.658-4.633, 95% CI 1.144-10.130, and all p < .05). CONCLUSIONS: In this large retrospective study, we found that SMT occurred in a small percentage of patients and was associated with relatively satisfactory short-term survival. Identification of preoperative risk factors for SMT and factors that negatively affect survival improve our understanding of this unique LT patient population.

The Evolution of Redo Liver Transplantation Over 35 Years: Analysis of 654 Consecutive Adult Liver Retransplants at a Single Center.

OBJECTIVE: To examine liver retransplantation (ReLT) over 35 years at a single center. BACKGROUND: Despite the durability of liver transplantation (LT), graft failure affects up to 40% of LT recipients. METHODS: All adult ReLTs from 1984 to 2021 were analyzed. Comparisons were made between ReLTs in the pre versus post-model for end-stage liver disease (MELD) eras and between ReLTs and primary-LTs in the modern era. Multivariate analysis was used for prognostic modeling. RESULTS: Six hundred fifty-four ReLTs were performed in 590 recipients. There were 372 pre-MELD ReLTs and 282 post-MELD ReLTs. Of the ReLT recipients, 89% had one previous LT, whereas 11% had ≥2. Primary nonfunction was the most common indication in the pre-MELD era (33%) versus recurrent disease (24%) in the post-MELD era. Post-MELD ReLT recipients were older (53 vs 48, P = 0.001), had higher MELD scores (35 vs 31, P = 0.01), and had more comorbidities. However, post-MELD ReLT patients had superior 1, 5, and 10-year survival compared with pre-MELD ReLT (75%, 60%, and 43% vs 53%, 43%, and 35%, respectively, P < 0.001) and lower in-hospital mortality and rejection rates. Notably, in the post-MELD era, the MELD score did not affect survival. We identified the following risk factors for early mortality (≤12 months after ReLT): coronary artery disease, obesity, ventilatory support, older recipient age, and longer pre-ReLT hospital stay. CONCLUSIONS: This represents the largest single-center ReLT report to date. Despite the increased acuity and complexity of ReLT patients, post-MELD era outcomes have improved. With careful patient selection, these results support the efficacy and survival benefit of ReLT in an acuity-based allocation environment.

Development and validation of a REcurrent Liver cAncer Prediction ScorE (RELAPSE) following liver transplantation in patients with hepatocellular carcinoma: Analysis of the US Multicenter HCC Transplant Consortium.

HCC recurrence following liver transplantation (LT) is highly morbid and occurs despite strict patient selection criteria. Individualized prediction of post-LT HCC recurrence risk remains an important need. Clinico-radiologic and pathologic data of 4981 patients with HCC undergoing LT from the US Multicenter HCC Transplant Consortium (UMHTC) were analyzed to develop a REcurrent Liver cAncer Prediction ScorE (RELAPSE). Multivariable Fine and Gray competing risk analysis and machine learning algorithms (Random Survival Forest and Classification and Regression Tree models) identified variables to model HCC recurrence. RELAPSE was externally validated in 1160 HCC LT recipients from the European Hepatocellular Cancer Liver Transplant study group. Of 4981 UMHTC patients with HCC undergoing LT, 71.9% were within Milan criteria, 16.1% were initially beyond Milan criteria with 9.4% downstaged before LT, and 12.0% had incidental HCC on explant pathology. Overall and recurrence-free survival at 1, 3, and 5 years was 89.7%, 78.6%, and 69.8% and 86.8%, 74.9%, and 66.7%, respectively, with a 5-year incidence of HCC recurrence of 12.5% (median 16 months) and non-HCC mortality of 20.8%. A multivariable model identified maximum alpha-fetoprotein (HR = 1.35 per-log SD, 95% CI,1.22-1.50, p < 0.001), neutrophil-lymphocyte ratio (HR = 1.16 per-log SD, 95% CI,1.04-1.28, p < 0.006), pathologic maximum tumor diameter (HR = 1.53 per-log SD, 95% CI, 1.35-1.73, p < 0.001), microvascular (HR = 2.37, 95%-CI, 1.87-2.99, p < 0.001) and macrovascular (HR = 3.38, 95% CI, 2.41-4.75, p < 0.001) invasion, and tumor differentiation (moderate HR = 1.75, 95% CI, 1.29-2.37, p < 0.001; poor HR = 2.62, 95% CI, 1.54-3.32, p < 0.001) as independent variables predicting post-LT HCC recurrence (C-statistic = 0.78). Machine learning algorithms incorporating additional covariates improved prediction of recurrence (Random Survival Forest C-statistic = 0.81). Despite significant differences in European Hepatocellular Cancer Liver Transplant recipient radiologic, treatment, and pathologic characteristics, external validation of RELAPSE demonstrated consistent 2- and 5-year recurrence risk discrimination (AUCs 0.77 and 0.75, respectively). We developed and externally validated a RELAPSE score that accurately discriminates post-LT HCC recurrence risk and may allow for individualized post-LT surveillance, immunosuppression modification, and selection of high-risk patients for adjuvant therapies.

Implications of Pleural Fluid Composition in Persistent Pleural Effusion following Orthotopic Liver Transplant.

Persistent pleural effusions (PPEf) represent a known complication of orthotopic liver transplant (OLT). However, their clinical relevance is not well described. We evaluated the clinical, biochemical, and cellular characteristics of post-OLT PPEf and assessed their relationship with longitudinal outcomes. We performed a retrospective cohort study of OLT recipients between 2006 and 2015. Included patients had post-OLT PPEf, defined by effusion persisting >30 days after OLT and available pleural fluid analysis. PPEf were classified as transudates or exudates (ExudLight) by Light's criteria. Exudates were subclassified as those with elevated lactate dehydrogenase (ExudLDH) or elevated protein (ExudProt). Cellular composition was classified as neutrophil- or lymphocyte-predominant. Of 1602 OLT patients, 124 (7.7%) had PPEf, of which 90.2% were ExudLight. Compared to all OLT recipients, PPEf patients had lower two-year survival (HR 1.63; p = 0.002). Among PPEf patients, one-year mortality was associated with pleural fluid RBC count (p = 0.03). While ExudLight and ExudProt showed no association with outcomes, ExudLDH were associated with increased ventilator dependence (p = 0.03) and postoperative length of stay (p = 0.03). Neutrophil-predominant effusions were associated with increased postoperative ventilator dependence (p = 0.03), vasopressor dependence (p = 0.02), and surgical pleural intervention (p = 0.02). In summary, post-OLT PPEf were associated with increased mortality. Ninety percent of these effusions were exudates by Light's criteria. Defining exudates using LDH only and incorporating cellular analysis, including neutrophils and RBCs, was useful in predicting morbidity.

Outcomes in liver transplant recipients with nonalcoholic fatty liver disease-related HCC: results from the US multicenter HCC transplant consortium.

NAFLD will soon be the most common indication for liver transplantation (LT). In NAFLD, HCC may occur at earlier stages of fibrosis and present with more advanced tumor stage, raising concern for aggressive disease. Thus, adult LT recipients with HCC from 20 US centers transplanted between 2002 and 2013 were analyzed to determine whether NAFLD impacts recurrence-free post-LT survival. Five hundred and thirty-eight (10.8%) of 4981 total patients had NAFLD. Patients with NAFLD were significantly older (63 vs. 58, p<0.001), had higher body mass index (30.5 vs. 27.4, p<0.001), and were more likely to have diabetes (57.3% vs. 28.8%, p<0.001). Patients with NAFLD were less likely to receive pre-LT locoregional therapy (63.6% vs. 72.9%, p<0.001), had higher median lab MELD (15 vs. 13, p<0.001) and neutrophil-lymphocyte ratio (3.8 vs. 2.9, p<0.001), and were more likely to have their maximum pre-LT alpha fetoprotein at time of LT (44.1% vs. 36.1%, p<0.001). NAFLD patients were more likely to have an incidental HCC on explant (19.4% vs. 10.4%, p<0.001); however, explant characteristics including tumor differentiation and vascular invasion were not different between groups. Comparing NAFLD and non-NAFLD patients, the 1, 3, and 5-year cumulative incidence of recurrence (3.1%, 9.1%, 11.5% vs. 4.9%, 10.1%, 12.6%, p=0.36) and recurrence-free survival rates (87%, 76%, and 67% vs. 87%, 75%, and 67%, p=0.97) were not different. In competing risks analysis, NAFLD did not significantly impact recurrence in univariable (HR: 0.88, p=0.36) nor in adjusted analysis (HR: 0.91, p=0.49). With NAFLD among the most common causes of HCC and poised to become the leading indication for LT, a better understanding of disease-specific models to predict recurrence is needed. In this NAFLD cohort, incidental HCCs were common, raising concerns about early detection. However, despite less locoregional therapy and high neutrophil-lymphocyte ratio, explant tumor characteristics and post-transplant recurrence-free survival were not different compared to non-NAFLD patients.

American Society of Transplant Surgeons recommendations on best practices in donation after circulatory death organ procurement.

The American Society of Transplant Surgeons supports efforts to increase the number of organs that are critically needed for patients desperately awaiting transplantation. In the United States, transplantation using organs procured from donation after circulatory death (DCD) donors has continued to increase in number. Despite these increases, substantial variability in the utilization and practices of DCD transplantation still exists. To improve DCD organ utilization, it is important to create a set of best practices for DCD recovery. The following recommendations aim to provide guidance on contemporary issues surrounding DCD organ procurement in the United States. A work group was composed of members of the American Society of Transplant Surgeon Scientific Studies Committee and the Thoracic Organ Transplantation Committee. The following topics were identified by the group either as controversial or lacking standardization: prewithdrawal preparation, definition of donor warm ischemia time, DCD surgical technique, combined thoracic and abdominal procurements, and normothermic regional perfusion. The proposed recommendations were classified on the basis of the grade of available evidence and the strength of the recommendation. This information should be valuable for transplant programs as well as for organ procurement organizations and donor hospitals as they develop robust DCD donor procurement protocols.

MR imaging biomarkers in HCC: outcomes correlation in liver transplant listed patients.

PURPOSE: To determine hepatocellular carcinoma (HCC) magnetic resonance imaging (MRI) biomarkers that enable the prediction of delisting from tumor progression versus successful transplantation in patients listed for orthotopic liver transplantation (OLT). METHODS: With IRB approval and HIPPA compliance, patients with HCC awaiting OLT who were delisted due to HCC progression from 2006 to 2015 were identified. Patients with adequate MR images for review were subsequently matched with a cohort of patients successfully bridged to OLT in the same time period. Matching considered the tumor stage and the dominant treatment strategy adopted to bridge the patient to OLT. Potential MRI features were evaluated by univariable and multivariable analysis using a conditional logistic model. RESULTS: There were 53 patients included in each cohort. On uni-variable analysis, significant unfavorable MR imaging features included T2 hyperintensity (odds ratio [OR], 19.0), infiltrative border (OR, 7.50), lobulated shape (OR, 4.5), T1 hypointensity (OR, 3.0), heterogeneous arterial enhancement (OR, 7.0), and corona venous enhancement (OR, 4.0). A significant favorable MR imaging feature was the presence of intralesional fat (OR = .36). The best multivariable logistic prediction model derived from the above notable features included only T1 and T2 signal intensity, border definition, and absence of intra-lesional fat as significant variables, with an area under the receiver operating characteristic curve (AUC) of .86 in the prediction of delisting. CONCLUSION: Select MR imaging features of HCC at presentation before any treatment are significantly associated with the risk of tumor progression regardless of tumor stage and treatment strategy in patients awaiting liver transplantation.

Perioperative Nephrectomy Outcomes for Patients with Liver Disease: Implications for Liver Transplant Candidates.

OBJECTIVE: To improve the management of cirrhotic patients diagnosed with new renal masses, we used a nationally representative cohort to assess the perioperative outcomes of nephrectomy in the setting of liver disease. The incidences of liver disease and renal masses are both rising in the US. Delaying liver transplantation to address other health concerns may have life changing consequences in these patients, thus these results help to guide treatment decisions at this critical junction in care. METHODS: A retrospective study of the 2016-2019 Nationwide Readmissions Database was performed in adults undergoing nephrectomy for non-emergent indications. Outcomes were compared between 3 cohorts: no chronic liver disease (no CLD), chronic liver disease (CLD), and decompensated cirrhosis (DC). Mixed regression models were used to evaluate the association between CLD and DC with outcomes of interest including morbidity, mortality, readmission rates, non-home discharges, length of stay, and costs. RESULTS: A total of 183,362 patients were evaluated. The mortality rate in the DC cohort (7%) was higher than with CLD (0.4%) and no CLD (0.3%), (P <.001). DC was associated with higher mortality (OR 8.29, 95% CI 4.07 - 16.88), postoperative bleeding requiring transfusion (OR 5.55, 95% CI 3.72 - 8.26), non-home discharge (OR 5.12, 95% CI 3.16 - 8.30) and readmission (OR 1.79, 95% CI 1.09 - 2.94) compared to no CLD. The DC cohort had the greatest length of stay and costs. CONCLUSION: Patients undergoing nephrectomy with DC have increased morbidity, mortality, readmission rates, non-home discharges, LOS and costs. Alternative management strategies may be considered in these patients.

Hepatocellular carcinoma surveillance: current practice and future directions.

Hepatocellular carcinoma (HCC) is among the leading causes of cancer incidence and mortality worldwide. Surveillance of individuals with cirrhosis or other conditions that confer a high risk of HCC development is essential for early detection and improved overall survival. Biannual ultrasonography with or without alpha-fetoprotein is widely recommended as the standard method for HCC surveillance, but it has limited sensitivity in early disease and may be inadequate in certain individuals. This review article will provide a comprehensive overview of the current landscape of HCC surveillance, including the rationale and indications for HCC surveillance, standard methods for HCC surveillance, and their strengths/limitations. Alternative surveillance methods such as the role of cross-sectional imaging, emerging circulating biomarkers, as well as the problem of under-utilization of HCC surveillance and surveillance-related harms will also be discussed in this review.

Cost-effective methylome sequencing of cell-free DNA for accurately detecting and locating cancer.

Early cancer detection by cell-free DNA faces multiple challenges: low fraction of tumor cell-free DNA, molecular heterogeneity of cancer, and sample sizes that are not sufficient to reflect diverse patient populations. Here, we develop a cancer detection approach to address these challenges. It consists of an assay, cfMethyl-Seq, for cost-effective sequencing of the cell-free DNA methylome (with > 12-fold enrichment over whole genome bisulfite sequencing in CpG islands), and a computational method to extract methylation information and diagnose patients. Applying our approach to 408 colon, liver, lung, and stomach cancer patients and controls, at 97.9% specificity we achieve 80.7% and 74.5% sensitivity in detecting all-stage and early-stage cancer, and 89.1% and 85.0% accuracy for locating tissue-of-origin of all-stage and early-stage cancer, respectively. Our approach cost-effectively retains methylome profiles of cancer abnormalities, allowing us to learn new features and expand to other cancer types as training cohorts grow.