Abstract Background: Colorectal cancer (CRC) metastases frequently recur due to minimal residual disease (MRD) and persisting micrometastases after therapy. However, the spatial and molecular features underlying micrometastatic persistence and CRC recurrence remain poorly defined. Study design and methods: We performed integrative spatial multi-omics profiling—including Visium spatial transcriptomics (ST), Visium HD ST, laser-capture microdissection with whole-genome sequencing (LCM-WGS), and PhenoCycler-Fusion multiplex imaging—across 49 tumors from 19 patients with paired primary CRC, liver (CLiM), and lung (CLuM) metastases. The analysis encompassed 341,328 Visium spots and approximately 3.8 million Visium HD bins. Non-negative matrix factorization (NMF) was applied to identify conserved and distinct spatial metaprograms across CLiM, CLuM, and primary CRC using Visium ST datasets. For Visium HD ST data, StarDist-SMURF segmentation was used to transform subcellular bins into single-cell-level data. Cross-modality alignment and Jaccard similarity analyses integrated spatially resolved DNA, RNA, and protein profiles across both corresponding and independent tissue blocks, enabling multi-layer characterization of tumor evolution and microenvironmental organization. Results: Spatial phylogenetic and molecular analyses delineated distinct evolutionary trajectories of primary and metastatic CRC, revealing early clonal divergence and stem-like phenotypes in liver micrometastases (CLiMi) across DNA, RNA, and protein levels. Spatial profiling uncovered stromal interactions in both CLiM and CLuM, with macrophages enriched in CLiM and lymphocytes predominating in CLuM. Micrometastases exhibited pronounced immunosuppression and T cell exhaustion, potentially mediated by PGE2/PTGES2-PTGER4 and NECTIN2/3-TIGIT signaling interactions. A CLiMi-specific six-gene signature predictive of micrometastasis was identified and validated, correlating with disease-free survival (DFS) and MRD-DFS in the MDACC cohort (n = 117), and with DFS and overall survival (OS) in TCGA (n = 610) and GSE17538 (n = 232). Conclusions: Our integrative spatial multi-omics analysis provides a comprehensive atlas of CRC micrometastases, revealing their evolutionary and immune landscapes. These findings illuminate the molecular and spatial determinants of micrometastatic persistence and identify potential therapeutic vulnerabilities for preventing CRC recurrence. Citation Format: Yang Liu, Akshaya S. Jadhav, Yuwen Pan, Jianlong Liao, Isha Khanduri, Yunhe Liu, Riham Katkhuda, Wei Lu, Kyung Serk Cho, Tieling Zhou, Baohua Sun, Mei Jiang, Sharia D. Hernandez, Idania Carolina Julio, Patrick Brennan, Guangsheng Pei, Kai Yu, Yibo Dai, Tian Chu, Fuduan Peng, Khaja Khan, Saxon Rodriguez, Ling Xia, Youming Guo, Alicia Mejia, Zhiming Tong, Sean W. Barnes, Ou Shi, Shreeya Indulkar, Alaa Mohamed, Natalie Wall Fowlkes, Timothy Newhook, Yun Shin Chun, Van K. Morris, David G. Menter, Dadi Jiang, Jean-Nicolas Vauthey, Ruoyan Li, Humam Kadara, Luisa M. Solis Soto, Scott Kopetz, Linghua Wang, Dipen M. Maru. Spatial multi-omics dissection of colorectal cancer micrometastasis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 6116.

PURPOSE:To determine whether wash-in (WI) and wash-out (WO) rates of contrast involving the future liver remnant (FLR) based on 2-dimensional venograms following portal vein embolization (PVE) can predict liver hypertrophy. METHODS:This is a single-center, retrospective, observational, cohort study of consecutive patients who underwent PVE from August 2019 to March 2024 during which the post-PVE venogram was obtained for more than 15 s. Enhancement of the FLR on post-PVE venograms was plotted as a function of time. Piecewise linear fits were applied to the datasets to derive WI and WO rates. Rates were compared to degree of hypertrophy (DH) and kinetic growth rate (KGR). RESULTS:16 patients who underwent PVE were included in the analysis (RPVE, n = 10; RPVE+4, n = 6). Median standardized FLR (sFLR) prior to PVE increased from 23% (median, range 15-49%) to 34% (median, range 24-54%), P = 0.0008. DH was 11.9% (median, range 2.0-19.9%) and KGR was 2.5% (median, range 0.5-4.7%). WI rates measured 0.0878 (median, range 0.0669-0.1760). WO rates measured 0.0315 (median, range 0.0039-0.0473). WO:WI ratios measured 0.2886 (median, range 0.0538-0.5760). Spearman's rank order correlations were calculated between WO:WI and DH (ρ = 0.6265) as well as WO:WI and KGR (ρ = 0.6529). Multivariate linear regression analysis of WO:WI with DH and KGR yielded P-values of 0.0470 and 0.0980, respectively. CONCLUSIONS:WO:WI ratios of the FLR calculated from 2-dimensional venograms following PVE may correlate with DH, providing immediate post-PVE assessment of the regenerative capacity of the FLR.

Colorectal cancer (CRC) metastases frequently recur due to minimal residual disease (MRD) and persistent micrometastases after therapy. Here, we performed spatial multimodal profiling using spot-level and high-resolution spatial transcriptomics, multi-regional whole-genome sequencing following laser-capture microdissection, and high-plex protein imaging to map 49 tumors from 19 patients, encompassing paired primary CRC and matched liver (CLiM) and lung (CLuM) metastases. Phylogenetic reconstruction revealed that liver micrometastases (CLiMi) arose from early clonal divergences and harbored a stem-like, quiescent state consistent with metastatic dormancy. Spatially, we uncovered distinct stromal barriers: macrometastases were encapsulated by myofibroblasts, whereas micrometastases were surrounded by immunosuppressive niches characterized by T cell exhaustion and distinct ligand-receptor signaling networks. Notably, we identified a CLiMi-specific six-gene signature associated with MRD status, disease-free survival, and chemotherapy resistance across multiple independent cohorts. These findings elucidate the spatial evolutionary landscape of CRC metastases and provide tissue-based spatially validated biomarkers for surveillance and therapeutic targeting.

Surgical and systemic management of neuroendocrine liver metastases (NELM) has evolved over the past decades. This study aimed to evaluate temporal changes in prognosis and identify factors associated with improved survival after hepatic cytoreduction for NELM. Patients who underwent hepatic cytoreduction for NELM during 1998–2023 were retrospectively analyzed and divided into earlier period (EP, 1998–2008) and later period (LP, 2009–2023) cohorts. Clinicopathological features, perioperative outcomes, and survival were compared. Covariate-adjusted survival curves and propensity score matching were used to compare overall survival (OS) between major hepatectomy and liver-parenchyma-sparing cytoreduction (LPSC). The study included 293 patients, 132 in the EP cohort and 161 in the LP cohort. The LP cohort had a lower rate of major hepatectomy (32.9

BACKGROUND:Total neoadjuvant therapy for localized rectal cancer entails delivery of all systemic chemotherapy and pelvic radiation before proctectomy to improve pathologic response, disease-free survival, and potentially organ preservation. In patients who develop liver metastases during or after total neoadjuvant therapy, the effects of total neoadjuvant therapy on patient outcomes and chemotherapy-associated liver injury are unknown. METHODS:A single-institution prospectively maintained database was queried for patients who underwent hepatectomy for metastatic rectal cancer, with or without total neoadjuvant therapy between 2014 and 2024. Surgical pathology reports were reviewed for histologic changes in the background liver. RESULTS:Among 286 patients undergoing hepatectomy for metastatic rectal cancer, 30 patients received total neoadjuvant therapy, and 256 patients did not receive total neoadjuvant therapy. Overall survival was significantly lower among patients in the total neoadjuvant therapy group (median overall survival 48.7 months, total neoadjuvant therapy vs 99.5 months, non-total neoadjuvant therapy, P = .01). On multivariable analysis, total neoadjuvant therapy remained an independent predictor of worse overall survival (hazard ratio 0.41, 95% confidence interval 0.22-0.79, P = .008). Sinusoidal injury rates were not significantly different with or without total neoadjuvant therapy (26.9% total neoadjuvant therapy vs 32.9% non-total neoadjuvant therapy, P = .69). Patients who received total neoadjuvant therapy had significantly higher rates of steatosis or steatohepatitis (30.8% total neoadjuvant therapy vs 8.7% non-total neoadjuvant therapy, P = .002). CONCLUSION:In this preliminary analysis, patients who developed liver metastases during or after total neoadjuvant therapy for rectal cancer had significantly worse overall survival after hepatectomy than patients who did not receive total neoadjuvant therapy. Steatosis and steatohepatitis rates were higher with total neoadjuvant therapy. Further study is needed to identify factors associated with worse outcomes in patients who undergo hepatectomy after total neoadjuvant therapy.

Background: Hepatocellular carcinoma (HCC) is a leading cause of cancer-related deaths globally. Although the hypoxia-inducible factor 1A (HIF1A) pathway is crucial in HCC progression, its regulatory mechanisms remain unclear as mutations in its primary regulator, von Hippel–Lindau tumor suppressor (VHL), are rare in HCC. We aimed to elucidate the role of proliferation and apoptosis adaptor protein 15 (PEA15), identified through proteomic analysis, as a regulator of the VHL/HIF1A pathway and a therapeutic target in HCC. Methods: Proteomic and genomic analyses of over 1,000 HCC samples were conducted, identifying PEA15 amplification. Functional validation involved in vitro and in vivo assays, including gene knockdown, ectopic expression, and antisense oligonucleotide (ASO) therapy in xenograft models. Protein interactions were assessed using immunoprecipitation and ubiquitination assays. Results: We identified 3 clinically distinct HCC subtypes and found that PEA15 was selectively amplified and highly expressed in the mesenchymal (MES) subtype, which exhibited the poorest prognosis. PEA15 acted as a regulator of the VHL/HIF1A pathway and a key oncogene in HCC. The amplification of PEA15 was significantly associated with the poor survival of HCC patients. Moreover, by interacting with the β-domain of VHL, PEA15 promoted HCC cell proliferation and migration by inhibiting VHL’s interaction with the VHL/elongin C (ELOC)/elongin B (ELOB)/cullin 2 (CUL2) E3 ligase complex, destabilizing the complex and consequently activating HIF1A. Importantly, pharmacologically inhibiting PEA15 using PEA15 ASO drugs attenuated tumor burden and restored VHL function in a xenograft mouse model. Conclusions: This study identified PEA15 as a potential oncogene in HCC, regulating the VHL/HIF1A axis and driving tumor progression. Targeting PEA15 using ASOs offers a promising therapeutic strategy for HCC, particularly in the MES subtype. These findings provide a basis for further exploration of PEA15-targeted therapies to improve HCC outcomes.

BACKGROUND:The AJCC staging system has assigned variable prognostic weight to multifocal disease across versions for intrahepatic cholangiocarcinoma (iCCA). We aimed to better delineate how multifocality influences survival in patients with resectable iCCA to refine clinical staging and better inform treatment sequencing. PATIENTS AND METHODS:We retrospectively identified patients undergoing curative-intent hepatectomy for iCCA between 2000 and 2024 from 6 international, high-volume hepatobiliary centers. Patients were stratified by multifocality, T classification, and N classification. Overall survival (OS) was evaluated using Kaplan-Meier and Cox regression analyses, and a novel modified system was proposed based on current AJCC groupings. RESULTS:A total of 731 patients met the study inclusion criteria. Multifocal disease was present in 36% of patients with T2-T4 tumors and was independently associated with worse OS (hazard ratio, 1.64; 95% CI, 1.21-2.22; P=.002) compared with solitary tumors on multivariable analysis. This survival disadvantage persisted across T2 classification (28.8 vs 44.8 months; P=.001) and T3 classification (21.2 vs 36.5 months; P=.003), and remained significant after adjustment for nodal status (P=.003 and P=.017, respectively). Compared with solitary disease, multifocality also conferred worse OS in patients with node-negative (36.0 vs 58.9 months; P=.003) and regionally node-positive disease (19.2 vs 23.8 months; P=.016). Current AJCC staging failed to distinguish prognostic differences between stage II and stage IIIA disease (48.3 vs 46.9 months; P=.190). However, our proposed modifications, which classify multifocal tumors as T3 and upstage multifocality from stage II to stage IIIA in node-negative disease and from stage IIIB to a newly defined stage IIIC in node-positive disease, better stratified overall risk (P<.001). CONCLUSIONS:Multifocality independently predicts poor prognosis in resectable iCCA. Refining the current staging system by reclassifying multifocal tumors as T3 may improve prognostic stratification and better inform treatment sequencing.

Reliable prognostic models of death or liver recurrence following resection of colorectal liver metastases are critical to stratify patients for treatment. The study aimed to develop models incorporating clinical and imaging data into multimodal preoperative prediction models of hepatic disease-free survival and overall survival. We conducted a retrospective cohort study with 1,301 consecutive patients from Memorial Sloan Kettering Cancer Center and The University of Texas MD Anderson Cancer Center. Clinical and computed tomography (CT) data were included in radiomic and deep learning models and compared. Our findings suggest that a deep learning model that utilizes the tumor region from CT imaging is the highest performing individual model with C-index of 0.61 for both hepatic disease-free (HDFS) and overall survival. Combining radiomic and clinical models into a multimodal 'clinicoradiomic' model (C-index = 0.63 [0.58-0.69]) outperformed unimodal models (C-index = 0.61 [0.55-0.66]) and the current clinical risk score (C-index = 0.55 [0.49-0.60]) for HDFS. Our study successfully developed preoperative imaging models that integrate clinical data to predict patients at risk of hepatic recurrence, outperforming all currently available models.

Background and Objective: This study introduces the liver cancer segmentator (LCS), a deep learning model designed for automatic and robust segmentation of liver parenchyma and tumors in abdominal contrast-enhanced computed tomography images from patients with colorectal liver metastases. The primary aim was to enhance confidence scoring for more reliable clinical segmentation assessment. Methods: In this retrospective study, 446 abdominal contrast-enhanced computed tomography examinations were collected; 355 (80%) were used for training and 91 for testing. Data originated from routine clinical cases at two institutions, representing diverse disease stages and treatment settings. A state-of-the-art neural network segmentation framework was trained on these cases, with performance evaluated using the Dice score and the normalized surface distance. An iterative training process, supported by an integrated annotation workflow, was employed to refine the training set. The final model was applied to the 91 test examinations to assess the impact of tumor volume and slice thickness on confidence scoring. Reliability was quantified through pairwise Dice score for failure detection and the area under the risk coverage curve. Results: The LCS achieved a Dice score of 0.9707 (95% CI: 0.9663-0.9751) for liver parenchyma and 0.7695 (95% CI: 0.7166-0.8224) for tumors. Normalized surface distance values at a 3-millimeter tolerance were 0.9605 (95% CI: 0.9539-0.9671) for parenchyma and 0.8412 (95% CI: 0.7928-0.8896) for tumors. Confidence scoring analysis demonstrated strong correlations between tumor volume, slice thickness, and segmentation reliability, reducing the area under the risk coverage curve from 16.7 to 10.3. Conclusions: The LCS achieved high segmentation accuracy in patients with colorectal liver metastases. Incorporating tumor volume and slice thickness into the confidence scoring process improved failure detection, enhanced reliability, and provided valuable insights for refining clinical deployment of automated segmentation algorithms.

PURPOSE:Biliary tract cancer (BTC) is the leading cause of death in patients with primary sclerosing cholangitis (PSC). PSC-related BTC is poorly understood, and the risks and benefits of conventional and immunotherapy treatments are unknown. We aimed to characterize clinical outcomes and genomes of PSC-related BTCs. EXPERIMENTAL DESIGN:This was a retrospective cohort study of patients with BTC with underlying PSC treated at MD Anderson Cancer Center (N = 46) and Princess Margaret Cancer Centre (N = 16), which were contrasted to patients with non-PSC-related BTC (N = 146). We compared outcomes between PSC and non-PSC, and PSC treated with and without immunotherapy. A combination of targeted sequencing (N = 139), whole-genome sequencing (WGS; N = 27), and WGS with paired RNA sequencing (N = 33) delineated the genomic and transcriptomic landscape of PSC-associated BTCs. RESULTS:In PSC-related BTC, the addition of immunotherapy to chemotherapy was associated with improved first-line progression-free survival (PFS; N = 22 vs. 11; median PFS, 12.2 vs. 4.7 months; P = 0.01). Immune-related adverse events were rare (N = 2, 12.5%) and improved after treatment discontinuation. Classic actionable genomic alterations, including IDH1 mutations and FGFR2 fusions, were absent in PSC-related BTCs. PSC tumors had a 2.6-fold higher tumor mutational burden (P = 3.28e-05) compared with non-PSC tumors. Transcriptomic profiling revealed a subset of PSC tumors displaying RNA signatures of immunotherapy response. CONCLUSIONS:Immunotherapy in PSC-associated BTCs seemed safe, with a potential signal of effectiveness. Given the sample size and retrospective design, these results are hypothesis-generating. Together, these results demonstrate the unique biology underlying PSC-associated BTCs, highlighting the need for prospective trials and the development of specialized treatment strategies.

BACKGROUND:Acute kidney injury after abdominal surgery is associated with increased morbidity, but little is known about acute kidney injury after hepatectomy, which was the focus of our study. METHODS:Patients undergoing curative-intent hepatectomy during 2017-2023 were included. Creatinine levels were collected within 2 weeks after hepatectomy. Acute kidney injury was defined as a creatinine level increase by ≥0.3 mg/dL within 48 hours or ≥1.5 times the baseline within 7 days; early acute kidney injury was defined as acute kidney injury by postoperative day 2. RESULTS:The study included 1,658 patients, of whom 180 (11%) had early acute kidney injury. Age >65 years, cardiovascular disease, diabetes mellitus, liver disease, kidney disease, longer operative time, and higher estimated blood loss were independently associated with an increased risk of early acute kidney injury. Median length of hospital stay was longer among patients with early acute kidney injury than among those without, for both low-risk (5 vs 3 days; P < .001) and high-risk hepatectomy (5 vs 4 days; P = .002). Patients with early acute kidney injury had a higher rate of major (Accordion ≥3) complications (22% vs 10%, P < .001). Multivariable analysis revealed 3 predictors of major complications: early acute kidney injury (odds ratio, 1.81; 95% confidence interval, 1.18-2.76), high-risk hepatectomy (odds ratio, 1.57; 95% confidence interval, 1.12-2.19), and operative time >7 hours (odds ratio, 2.66; 95% confidence interval, 1.90-3.73). The major complication rate was 6% for patients without the aforementioned factors, 13% for patients with 1 factor, and 33% for patients with all 3 factors. CONCLUSION:Early postoperative acute kidney injury is associated with extended hospital stay and predicts major complications, especially for patients with high-risk hepatectomy and extended operative time.

Accurate segmentation of colorectal liver metastases (CRLM) in contrast-enhanced computed tomography (CT) is important for response assessment, surgical planning, and follow-up. We propose two parameter-efficient spectral adapters for the Segment Anything Model (SAM): the Directional Spectral Adapter (DiSECT) and Spectral Instance-Guided Adapter (SiGA). DiSECT uses singular value decomposition of frozen weights to constrain residual updates to leading spectral directions, while SiGA adds global and input-conditioned gating through a multilayer perceptron. We evaluate these methods on 446 contrast-enhanced CT volumes (355 training, 91 testing) and compare them with LoRA, QLoRA, convolutional adapters (CAD), and a 3D nnU-Net baseline. Experiments consider single-point, three-point, bounding-box, and no-prompt regimes. SiGA achieves the best single-point performance with a Dice score of 0.77, IoU of 0.69, and HD95 of 35.39 mm. Under no-prompt inference, SiGA reaches 0.76 Dice, 0.68 IoU, and 46.76 mm HD95, comparable to the nnU-Net baseline (0.758 Dice). DiSECT uses only 0.14 million trainable parameters. These results show that spectral adapters can efficiently adapt SAM for CRLM segmentation while retaining strong accuracy with limited trainable parameters.

552 Background: Fibrolamellar carcinoma (FLC) is a rare liver malignancy primarily in adolescents / young adults with limited therapeutic options. We analyzed clinical outcomes and genomic data and investigated tumor microenvironment (TME) to identify immunotherapy response markers. Methods: This is a retrospective review of 165 patients (pts) treated between 1992 and 2024. mRNA-seq was performed in 8 patients treated with immunotherapy. 15 immune gene signatures, through unsupervised clustering, revealed 2 dominant phenotypes: T-cell/Myeloid-Infiltrated (TMI) or Immune-Desert (ID). PFS/OS was correlated with TME via Kaplan-Meier and Log-rank test. Results: Out of 165 pts (median age 23, 48% F), 75% presented with stage III-IV. Adverse prognosticators included advanced stage, macrovascular invasion, and nodal metastasis. Surgery (liver resection and/or metastasectomy) is associated with improved survival (mOS 65.2 vs 19.7 m, p<0.001). Our institutional genomic analysis revealed low TMB (median 1.2 mut/Mb) and MSS. A separate genomic cohort (n=68) from FMI (no survival outcome data) showed mutations beyond DNAJB1-PRKACA fusion as Table1. In the immunotherapy cohort with available mRNA data, TMI phenotype (n=5) was associated with significantly longer PFS than ID phenotype (n=3), with mPFS of not reached vs. 4.3 m (p=0.008). Disease control rate was 100% for TMI group (2 PR, 3 SD) versus 0% for ID group on immunotherapy. Analysis of TME evolution in paired pre- and post-treatment biopsies from 2 exceptional responders revealed a marked on-treatment increase in key effector and helper immune cell signatures including T-cells (+34%), cytotoxic cells (+33%), Th1 cells (+37%), and B-cells (+44%). Compared to a pan-cancer cohort (TCGA), FLC tumors showed elevated expression of angiogenesis genes (VEGFA/B/C, KDR, FLT1/4) and ADC targets with GPC3 and CEACAM5 expression being over 3-fold higher than the pan-cancer average; no pts met criteria for high ERBB2 expression (log2(TPM+1) > 8.5). Conclusions: While surgical management including metastasectomy remains the cornerstone of FLC treatment, TME is a predictive biomarker for immunotherapy. The unique molecular landscape with high expression of angiogenesis factors and ADC targets suggests that rational combination therapies are needed to overcome resistance in Immune-Desert tumors. Immune gene signatures B-cells, CD45, CD8 T cells, Cytotoxic cells, DC (Dendritic Cells), Exhausted CD8, Macrophages, Mast cells, NK cells, Neutrophils, T-cells, TFH (T follicular helper cells), Th1 cells, Th2 cells, Treg (Regulatory T cells) Genomic alterations (n=68) TERT 17.7% (MD Anderson 11%), CDKN2A/B 11.8%, MYC 5.9%, LYN 4.4%, CTNNB1 4.4%, TP53 4.4% Immunotherapy Atezolizumab + bevacizumab (n=1); 5-fluorouracil + Interferon-a2 + nivolumab (n=7) ADC, antibody-drug conjugate.