BACKGROUND:Hepatocellular carcinoma (HCC) remains a major cause of cancer-related mortality. Optimal treatment decisions are challenging, and existing treatment allocation frameworks may not fully reflect the evolving therapeutic landscape or contemporary clinical practices in North America. METHODS:Using a modified Delphi process, a multidisciplinary panel of 20 North American experts in hepatology, medical oncology, surgery, radiology, and radiation oncology developed a consensus-based treatment framework for HCC, termed the BEACON-HCC system. The framework was informed by current evidence and expert opinion through iterative discussion and voting. In a pilot study using 29 real-world patient cases, we assessed concordance between external expert recommendations and BEACON-HCC recommendations and Barcelona Clinic Liver Cancer (BCLC) 2025 treatment recommendations. RESULTS:The BEACON-HCC treatment allocation framework diverges from prior frameworks by incorporating nuanced clinical features such as degree of intrahepatic tumor burden and vascular invasion, and adverse tumor prognostic markers, to align treatment allocation with tumor biology and therapeutic potential. Key innovations include the incorporation of emerging modalities such as external beam radiation therapy (EBRT), transarterial radioembolization (TARE), and systemic-locoregional combination therapies. In the pilot exercise, expert treatment decisions showed 96.6% concordance with BEACON-HCC recommendations and 72.4% concordance with BCLC recommendations. CONCLUSIONS:BEACON-HCC is a consensus-based framework for treatment allocation that incorporates the expanding range of therapeutic options available to patients in the North American HCC population. Expert treatment decisions showed a high concordance with the BEACON-HCC system; further validation using empiric clinical data is planned through the HCC-Live Consortium.

ABSTRACT Background Vascular invasion by hepatocellular carcinoma (HCC), particularly involving the main portal vein or inferior vena cava (IVC), is associated with poor prognosis. However, contemporary real‐world outcomes and the clinical benefit of liver‐directed therapies (LDT) in this setting are poorly defined. We evaluated outcomes of patients with vascular invasion managed in a multidisciplinary liver cancer clinic. Methods We retrospectively analyzed outcomes for patients with HCC who presented to the Johns Hopkins Liver Multidisciplinary Clinic between 2020 and 2024, with radiographic evidence of the tumor thrombus in the main portal vein and/or IVC. Overall (OS) and progression‐free survival and gastrointestinal variceal bleeding incidence were evaluated. Associations between clinical variables and outcomes were analyzed using the Cox proportional hazards regression. Results Fifty‐four patients met our study criteria. Forty‐eight patients (89%) had portal vein main trunk invasion, and 13 (24%) had IVC tumor thrombus. Treatment included systemic therapy alone in 30 patients (56%), systemic therapy and LDT in 15 patients (28%), best supportive care in six patients (11%), and LDT alone in three patients (6%). Median OS was 5.7 months (95% CI: 2.6–12 months). Median survival was shorter among patients with IVC involvement (4.1 months, 95% CI: 0.82–5.7 months) than those without (7.6 months, 95% CI: 2.3–17 months), which was independently associated with worse prognosis in multivariable analysis (HR: 4.0, 95% CI: 1.5–11, p = 0.0051). Other significant findings of multivariable analysis include performance status (HR: 7.2, 95% CI: 2.7–19, p < 0.0001) and viral hepatitis (HR: 0.46, 95% CI: 0.23–0.79, p = 0.027). Four patients (7%) experienced bleeding esophageal varices requiring intervention. Conclusions HCC with tumor thrombus involving the main portal vein and/or IVC is associated with poor survival in real‐world clinical practice. Prospective studies are needed to define the optimal management of this high‐risk population.

e16301 Background: Surgical resection is the mainstay for localized, resectable biliary tract cancers (BTCs). Although genomic clusters have been associated with outcomes in BTCs, data linking specific molecular subtypes to post-surgical outcomes remain limited. We aimed to evaluate the prognostic role of tumor genetic signatures in patients with BTC undergoing curative-intent surgery. Methods: We retrospectively analyzed Liver Multi-Disciplinary Clinic (LMDC) patients who underwent curative-intent surgery between 2015 and 2025 with pathology proven BTC. We included patients with detailed pathological records and next generation sequencing (NGS) results. Our endpoints of disease-free survival (DFS) and overall survival (OS) were defined as the time from surgery to their first documented disease recurrence and death from any cause, respectively. Results: There were 97 patients (52.6% male, 47.4% female) in our cohort with a mean age of 65.1 years. Most patients had AJCC Stage II (28.8%) or III disease (27.8%). Median DFS was 20.1 months (CI 16.0–28.7) and median OS was 40.4 months (CI 37.3-52.5). Recurrence occurred in 64.2% of patients, and the median survival post-recurrence was 15.4 months. In gene-level analyses, alterations in SMAD4 and BRAF were significantly associated with shorter DFS (SMAD4: HR 4.2, CI 1.61-11.00, p 0.003; BRAF HR 3.89, CI 1.17-12.92, p 0.03). Mutations in SMAD4 and NTRK1/2/3 were also associated with worse OS (SMAD4: HR 3.53, CI 1.22-10.19, p 0.02; NTRK1/2/3: HR 7.72, CI 2.23 - 26.70, p 0.001), while mutations in FGFR2 significantly improved OS (HR 0.22, CI 0.05-0.92, p 0.03). Subsequently, cluster-based analysis was performed to prevent distortion from low-frequency alterations. Cluster 1 (TP53/KRAS/ATM) and cluster 2 (CDKN2A/2B) were not associated with OS or DFS. In contrast, cluster 3 (ARID1A/PBRM1/IDH1) alterations were associated with decreased OS (HR 2.36, CI 1.23-4.56, p 0.01). Cluster 4 (FGFR2/BAP1) mutations were associated with improved OS (HR 0.33, CI 0.12, 0.92, p 0.03) relative to patients without these alterations. Conclusions: In the post-surgical setting, patients harboring mutations in chromatin remodeling pathways (ARID1A/PBRM1/IDH1), as well as SMAD4 and NTRK1/2/3, experienced worse outcomes, whereas those with BAP1 and FGFR2 alterations demonstrate improved outcomes. Although limited by sample size, our study lays out the foundation for future investigations evaluating the impact of molecular profiling in BTC patients undergoing surgical resection.

Objectives: This study aims to identify clinical characteristics and biomarkers influencing survival outcomes in colorectal cancer (CRC) patients with spinal metastases. Methods: We conducted a retrospective cohort study involving 27 patients treated for CRC-derived spinal metastases at Johns Hopkins Hospital. Data on demographics, biomarker profiles of the primary colorectal tumor site, surgical outcomes, and survival were collected. Neurological function was assessed pre- and postoperatively using Frankel scores. Survival outcomes included overall survival (OS) and post-metastasis survival. Results: The median age of the patients was 58 years, with 63% being women. The sacral spine was the most frequently involved site (59.3%), followed by the thoracic and lumbar regions. Most patients (89%) already had extraspinal metastases, predominantly in the lungs. Biomarker analysis showed microsatellite stability in 63% of patients and CDX2 (Caudal-type homeobox 2) expression in 37%. Laminectomy was performed in 85% of cases and sacrectomy in 55.6%, leading to postoperative improvements in ambulatory function and neurological status. The main indications included local recurrence of the tumor and neurological deficits attributed to the impinging tumor. The median overall survival was 4.9 years, while the median post-metastasis survival was 3.0 years. Univariable analysis revealed that CK20 expression (p = 0.041) and spinal tumor recurrence (p = 0.045) were significantly associated with shorter post-metastasis survival. Conclusions: This study highlights the prognostic importance of CK20 expression and spinal tumor recurrence in CRC patients diagnosed with spinal metastases. Surgical intervention significantly improved neurological outcomes, enhancing patient quality of life. Further research with larger cohorts is needed to confirm these findings and optimize treatment strategies for this challenging patient population.

Background: Bile tract cancer (BTC) is a heterogeneous and aggressive malignancy with a poor prognosis. Surgical resection is the main potentially curative treatment, but only about 20-30 % of patients present with resectable disease. The benefit of neoadjuvant chemotherapy (NAC) in BTC remains controversial. We explored outcomes of patients with BTCs receiving NAC compared to those who underwent upfront surgical resection using a propensity score weighting approach. Methods: We identified patients who underwent surgical resection for stage I-III BTC between 2019 and 2023 and were evaluated at the Johns Hopkins Liver Multidisciplinary Cancer Clinic (Liver MDC). Propensity score weighting (PSW) was used to balance groups. Groups were assessed for differences in pathological response, recurrence-free survival (RFS), and overall survival (OS). Hazard ratios (HRs) were estimated using the PSW Cox proportional hazard regression model. Results: Among 56 BTC patients that fit the inclusion criteria, 34 underwent upfront surgery, and 22 received NAC. Patients receiving NAC were more likely to have higher risk disease compared to patients receiving upfront resection (77.3 % vs 23.5 %, p < 0.001), including tumor size greater than 5 cm (45.5 % vs 11.8 %, p = 0.0221), lymph node involvement on staging imaging (50.0 % vs 23.5 %, p = 0.0495), and a trend towards higher mean CA19-9 at diagnosis (258.7 vs 87.7 U/mL, p = 0.179). In the univariate analysis, the NAC group showed a trend toward shorter RFS compared to the upfront surgery group (15.1 months; 95 % CI 10.3-not available (NA) versus 23.5 months; 95 % CI 15.5-NA, HR= 2.24; 95 % CI: 0.88-5.69, p = 0.082). No significant differences were seen for OS (36.5 months; 95 % CI 26.5-NA versus 48.2 months; 95 % CI 44.5-NA, HR= 1.53; 95 % CI: 0.51-4.64-5.69, p = 0.448). Following PSW multivariable Cox analyses, receipt of NAC was associated with a trend toward reduced risk of recurrence (HR 0.83; 95 % CI 0.24-2.87 p = 0.7651) and death (HR 0.66; 95 % CI 0.2-2.16 p = 0.4890), although no statistically significant differences were observed. Conclusion: These findings suggest that NAC may provide benefit for patients with higher-risk BTCs, achieving outcomes comparable to upfront surgical resection. The similar pathological results between the two groups further support the potential efficacy of neoadjuvant approaches. However, prospective trials are essential to definitively establish the optimal sequencing of chemotherapy and surgery in this patient population.

BackgroundNonoperative management in patients with rectal cancer with complete response to radiation therapy and chemotherapy is of increasing interest. Most of the data on nonoperative management have centered around patients treated with long-course chemoradiotherapy. The ability of short-course radiation-based treatment courses to achieve durable complete responses with sustained organ preservation is less defined. This study updates our institution's long-term experience with nonoperative management following upfront short-course radiation therapy and sequential/consolidation chemotherapy.MethodsWe retrospectively reviewed patients with nonmetastatic rectal cancer treated with sequential short-course radiation therapy and chemotherapy who reached complete response and were subsequently followed with nonoperative management. We report on disease control outcomes, including rates of regrowth and results of salvage surgery. We investigated characteristics associated with local tumor regrowth.ResultsOur study included 52 patients. The 2-year freedom from local regrowth for the entire cohort was 75%. Notably, patients with high-risk disease characteristics at diagnosis exhibited a trend toward a higher rate of local tumor regrowth. No patient with sustained clinical complete response developed metastatic disease. Of the twelve patients undergoing surgical salvage for regrowth of disease, ten were resected with complete/near-complete total mesorectal surgical specimens with negative margins.ConclusionsThe optimal approach to achieving sustained organ preservation through the use of radiation therapy and chemotherapy is not well defined. Our findings show the viability of neoadjuvant therapy incorporating short-course radiation therapy to achieve durable complete responses.

Abstract Surgical resection for localized hepatocellular carcinoma (HCC) is typically reserved for a minority of patients with favorable tumor features and anatomy. Neoadjuvant immunotherapy can expand the number of patients who are candidates for surgical resection and potentially reduce the chance for recurrence, but its role in HCC not defined. We retrospectively examined the outcomes of patients who underwent surgical resection for HCC at the Johns Hopkins Hospital and compared the clinical outcomes of patients who received neoadjuvant immunotherapy with those who underwent upfront resection. The clinical cohort included a total of 92 patients, 36 of whom received neoadjuvant immune checkpoint inhibitor (ICI)-based treatment. A majority of patients (61.1%) who received neoadjuvant ICI–based therapy were outside of standard resectability criteria and were more likely to have features known to confer risk of disease recurrence, including α-fetoprotein ≥ 400 ng/mL (P = 0.02), tumor diameter ≥ 5 cm (P = 0.001), portal vein invasion (P < 0.001), and multifocality (P < 0.001). Patients who received neoadjuvant immunotherapy had similar rates of margin-negative resection (P = 0.47) and recurrence-free survival (RFS) as those who underwent upfront surgical resection (median RFS 44.8 months compared with 49.3 months, respectively, log-rank P = 0.66). There was a nonsignificant trend toward superior RFS in the subset of patients with a pathologic response (tumor necrosis ≥ 70%) with neoadjuvant immunotherapy. Neoadjuvant ICI-based therapy may allow high-risk patients, including those who are outside traditional resectability criteria, to achieve comparable clinical outcomes with those who undergo upfront resection. Significance: Surgical resection for localized HCC is typically only reserved for those with solitary tumors without vascular invasion. In this retrospective analysis, we show that neoadjuvant immunotherapy may allow high-risk patients, including those who are outside of standard resection criteria, to undergo successful margin-negative resection and achieve comparable long-term clinical outcomes compared with upfront resection. These findings highlight need for prospective studies on neoadjuvant immunotherapy in HCC.

Background: Total neoadjuvant therapy (TNT) is an accepted approach for the management of locally advanced rectal cancer (LARC) and is associated with a decreased risk of development of metastatic disease compared to standard neoadjuvant therapy. However, questions remain regarding surgical outcomes and local control in patients who proceed to surgery, particularly when radiation is given first in the neoadjuvant sequence. We report on our institution's experience with patients who underwent short-course radiation therapy, consolidation chemotherapy, and surgery. Methods: We retrospectively reviewed surgical specimen outcomes, postoperative complications, and local/pelvic control in a large cohort of patients with LARC who underwent neoadjuvant therapy incorporating upfront short-course radiation therapy followed by consolidation chemotherapy. Results: In our cohort of 83 patients who proceeded to surgery, a complete/near-complete mesorectal specimen was achieved in 90 % of patients. This outcome was not associated with the time interval from completion of radiation to surgery. Postoperative complications were acceptably low. Local control at two years was 93.4 % for all patients- 97.6 % for those with low-risk disease and 90.4 % for high-risk disease. Conclusion: Upfront short-course radiation therapy and consolidation chemotherapy is an effective treatment course. Extended interval from completion of short-course radiation therapy did not impact surgical specimen quality.

Purpose/Objective(s) Proximity of organs at risk (OAR) has led to interest in protracted-fractionation (PF: 15 to 25 fractions) courses employing moderate hypofractionation (MHF: 3-4 Gy/fraction) for the treatment of pancreatic adenocarcinoma, as an alternative to the limitations of classical SBRT. However, underdosing of target volumes near their interfaces with OARs is still seen even with PF-MHF courses. The impact of compromised (relative to the prescription dose) target coverage and global target dose heterogeneity on tumor control probability is an active area of investigation. We report our institution’s initial treatment planning experience with PF-MHF in pancreatic cancer. Materials/Methods We retrospectively reviewed radiation courses for pancreatic cancer planned with a PF-MHF approach: 45 Gy in 25 fractions (1.8 Gy/fraction) to PTV with 75 Gy (3 Gy/fraction) as an integrated boost to the GTV. We included patients treated with unresected disease as well as patients treated for local recurrence following surgery. We then collected describing dosimetric parameters for the GTV: D99.9%, D0.1cc, Dmean, V75, and V60Gy. To assess global DVH characteristics, we also calculated the generalized equivalent uniform dose (gEUD) value, using two different a values (-5 and -15). For the intact pancreas plans, we evaluated whether the gEUD differed depending on the location of tumor (head/uncinate versus body/tail). Results There was a total of 27 plans included in our analysis: 15 and 12 intact and resected pancreas plans, respectively. The median and range of collected dosimetric parameters are as follows: D99.9% 50.1 Gy (35.0 – 77.2 Gy), D0.1cc 81.3 Gy (77.9 – 91.6Gy), Dmean 74.7 Gy (70.8 – 78.6 Gy), V75 Gy 71.2% (50.99 - 100%), and V60 Gy 92.48% (79.28 – 100%). Assuming an a value of -5 and -15, the median (range) gEUD values were 71.0 Gy (61.5 – 78.6 Gy) and 63.2 Gy (49.4 – 78.6 Gy), respectively. For the intact pancreas plans, there were no statistically significant differences in the mean gEUDs between head/uncinate (n = 7) and body/tail (n = 8) plans even if factoring in the GTV volumes. Conclusion Although the dose was escalated within the GTV, the relatively low coverage by the prescription dose, and correspondingly low gEUD values, especially for highly negative a values, illustrate that the PF-MHF approach is associated with numerically significant dosimetric compromise. One consideration is to prioritize gEUD optimization for further refinement. Continued investigation of how dose heterogeneity impacts local control is also warranted.

BACKGROUND:Radiation oncologists closely monitor patients during weekly on-treatment visits (OTVs). This study examines whether routine patient-reported outcome measures (PROMs) during OTVs change physicians' perceptions of treatment-toxicity and inform symptom-management. PATIENT AND METHODS:IMPROVE is a single-arm prospective multicenter trial, conducted from 2020 to 2023. Patients with locally-advanced or oligometastatic thoracic or gastrointestinal cancers receiving definitive-intent radiation, with or without chemotherapy, and their physicians enrolled. Patients completed a 14-question disease-specific PROM in clinic prior to OTVs. Physicians rated their patient's global toxicity-burden based on clinical data/assessments, then re-rated their patient's toxicity-burden and reported management-changes after PROM review. At radiotherapy end, physicians completed a Feedback Form. PROMs and outcome-data collection used electronic or paper forms. We report any change in physician-assessed burden-score and symptom-management due to PROMs. RESULTS:The 100 patients enrolled (49 academic, 51 community-based) were 70 years old (median), 51% female, 81% Caucasian, 95% ECOG 0-1, and 94% received concurrent chemotherapy. The median radiation dose was 60 Gy, delivered over 6 weeks. PROMs were available for review for 607/629 (97%) OTVs: full 433/629 (69%), partial 174/629 (28%). For 75/100 patients (75%; 95% CI:65%-83%), PROM review resulted in any change in physician-reported burden-score, and for 50/100 patients (50%; 95% CI:40%-60%) any change in patients' on-treatment management. Rates of burden-score and management-changes were similar between academic and community-based practices (78% vs. 73%; 53% vs. 47%, respectively). For 78/100 patients with Feedback Forms, physicians agreed/strongly agreed that PROMs improved patients' quality-of-care (91%). CONCLUSIONS:PROM review changes radiation oncologists' on-treatment toxicity assessment in 75% and care delivery in 50% of their patients.

Despite advances in treatment and response assessment in locally advanced rectal cancer (LARC), it is unclear which patients should undergo nonoperative management (NOM). We performed a single-center, retrospective study to evaluate post-total neoadjuvant therapy (TNT) circulating tumor DNA (ctDNA) in predicting treatment response. We found that post-TNT ctDNA had a sensitivity of 23% and specificity of 100% for predicting residual disease upon resection, with a positive predictive value (PPV) of 100% and a negative predictive value (NPV) of 47%. For predicting poor tumor regression on MRI, ctDNA had a sensitivity of 16% and specificity of 96%, with a PPV of 75% and NPV of 60%. A commercially available ctDNA assay was insufficient to predict residual disease after TNT and should not be used alone to select patients for NOM in LARC.

Baseline characteristics of patients who underwent ICI-based neoadjuvant therapy or upfront resection for HCC

Introduction Proximity of organs at risk (OAR) hinders radiation dose escalation for the treatment of pancreatic cancer. To address this limitation, there is interest in protracted-fractionation (PF: 15 to 25 fractions) courses employing moderate hypofractionation (MHF: 3-4 Gy/fraction). However, there persists underdosing where tumor interfaces with OAR. The significance of compromised tumor coverage and dose heterogeneity on tumor control remains unknown. Here, we report our initial planning experience with PF-MHF in pancreatic cancer. Methods We retrospectively reviewed radiation courses for locally advanced or recurrent pancreatic cancer with a PFMHF approach: 45 Gy in 25 fractions (1.8 Gy/fraction) to PTV with 75 Gy (3 Gy/fraction) as an integrated boost to the GTV. We reviewed dosimetric parameters for the GTV: percentage overlap with planning OAR volume (PRV-GTV overlap), D99.9%, D0.1cc, Dmean, V75Gy, and V60Gy. We also calculated the GTV's generalized equivalent uniform dose (gEUD) value using two different a values (-5 and-15). Lastly, we reoptimized two plans with two approaches: increasing gEUD or relaxing the maximum dose constraint. Results A total of 26 plans were included in our analysis: 14 locally advanced and 12 locally recurrent pancreatic cancer cases. While the D0.1cc median value was 81.7 Gy, target volume coverage was relatively low (V75Gy median 71%). Median gEUD were 71 Gy (a a =-5) and 62.8 Gy (a a =-15) and inversely correlated with PRV-GTV overlap. On reoptimized plans, both approaches yielded similar results, but an increase in target coverage and gEUD were seen only when there was limited PRV-GTV overlap. Conclusion Although radiation dose can be escalated within the GTV, there continues to be low coverage by the prescription dose, especially with high PRV-GTV overlap. Relaxing the maximum dose constraint in planning allows for meaningful improvement in tumor coverage in limited PRV overlap scenarios. Continued refinement of the PF-MHF approach is needed.

S1. Flow diagram depicting inclusion and exclusion criteria for current study

Detailed baseline characteristics in the ICI-treated cohort by whether tumor was resectable at baseline according to BCLC criteria