255 - Baseline Hemoglobin Predicts Oncologic Outcomes in Intrahepatic Cholangiocarcinoma Treated with Yttrium-90 Radioembolization

Tuesday, April 14, 2026

4:21 PM - 4:30 PM ET

Presenting Author(s)

AG

Andrew C Gordon, MD, PhD

Interventional Radiologist
Northwestern University, United States

Author/Co-author(s)

SA

Saad Abu Zahra, MD

Post-Doctoral Scholar
Northwestern University, United States
Muhamad Serhal, MD

PGY1
Northwestern University, United States
Aakash N. Gupta, MD

Resident Physician
Stanford University, United States
AG

Ahmed Gabr, MD

Resident
Northwestern University, United States
Arsalan Nadeem, MBBS (he/him/his)

Postdoctoral Research Fellow
Northwestern University, United States
AK

Ashima Kundu, BS

Medical Student
Loyola Stritch School of Medicine, United States
DB

Daniel Borja-Cacho, M.D.

Attending Physician
Northwestern Memorial Hospital, United States
AK

Aparna Kalyan, MD

Hemato-oncologist
Northwestern University, United States
LK

Laura Kulik, MD

Attending Physician
Northwestern University, United States
AB

Amir Borhani, MD

Attending Physician
Northwestern University, United States
Robert J. Lewandowski, MD, FSIR

Professor of Radiology, Medicine and Surgery
Northwestern University, Dept. of Radiology, IR section, United States

Purpose: To evaluate whether baseline hemoglobin (HGB) predicts treatment response, disease progression, and survival in patients with unresectable intrahepatic cholangiocarcinoma (ICC) undergoing Yttrium-90 (Y90) transarterial radioembolization (TARE).

Materials and Methods: A retrospective cohort of 136 ICC patients treated with glass microspheres [Boston Scientific, MA, USA] Y90-TARE (2004–2020) was analyzed. Patients were stratified by baseline HGB: ≥12 g/dL (n=40), 10–11.9 g/dL (n=56), and < 10 g/dL (n=40). Tumor response was assessed by RECIST 1.1; outcomes included duration of response (DoR), time to progression (TTP), and overall survival (OS). Kaplan–Meier and log-rank testing compared OS; Cox regression identified independent OS predictors.

Results: Median age was 64 years; 49% were female. RECIST response differed across HGB strata: 38% (15/39) for ≥12 g/dL, 42% (22/53) for 10–11.9 g/dL, and 18% (7/39) for < 10 g/dL (p=0.041). Median DoR was 15.0 (95% CI 7.0–24.9), 6.5(95% CI 3.9–11.1), and 6.2 months (95% CI 1.1–9.6) (p=0.04). Median TTP was 7.1 months (95% CI 3.7–13.1), 5.8 months (95% CI 2.9–8.3), and 1.7 months (95% CI 1.1–3.4) (p=0.0075). Median liver TTP was 13.7 (95% CI 6.0–39.5), 6.3 (95% CI 4.1–8.6), and 3.3 months (95% CI 1.2–6.9) (p=0.0051), and extrahepatic TTP was 24.5, 14.6, and 5.9 months (p=0.0284). Median OS was 17.1 months (95% CI 11.8–31.3), 13.8 months (95% CI 8.6–19.3), and 6.0 months (95% CI 4.2–12.4) (p=0.0007). On multivariate analysis, HGB < 10 g/dL independently predicted worse OS (HR 1.79, 95% CI 1.03–3.10, p=0.037), along with ascites (HR 2.06, p=0.008) and portal vein thrombosis (HR 1.81, p=0.024).

Conclusion: Baseline HGB < 10 g/dL was associated with reduced response, shorter progression-free intervals, and decreased survival after Y90-TARE. As a low-cost biomarker, HGB may improve risk stratification and guide treatment planning. Prospective studies should investigate whether correction of anemia prior to TARE improves clinical outcomes.