Prospective study on stereotactic body radiation therapy before liver transplantation for hepatocellular carcinoma

Abstract

Background

Stereotactic body radiation therapy (SBRT) is effective for hepatocellular carcinoma (HCC). Its role as a bridging therapy before deceased-donor liver transplantation (DDLT) remains unclear.

Methods

Our institution conducted a prospective study on SBRT before DDLT for 32 HCC patients within University of California, San Francisco (UCSF) criteria as the first cohort, staged with gadoxetate disodium-enhanced MRI and 11C-acetate (ACC) and 18F-fluorodeoxyglucose (FDG) (dual-tracer) positron-emission tomography with computed tomography (PET-CT), followed by SBRT of 35–50 Gy in 5 fractions. They were followed up with the same imaging modalities afterwards while awaiting DDLT. The results were published in June 2024. In view of the encouraging treatment outcomes, SBRT before DDLT has become a routine clinical practice in our institution and the study has been carried on without dual-tracer PET-CT imaging for all subsequent patients recruited prospectively for another 39 patients following revised sample size estimation. Co-primary end points were progression-free survival (PFS) and objective response rates (ORRs) by RECIST 1.1 and mRECIST. Secondary end points were overall survival (OS), safety, and others. The trial is registered with clinicaltrials.gov, NCT04186234.

Findings

Between July 1, 2015 until October 31, 2023, 71 patients with 118 HCC lesions were recruited with a median follow-up duration of 62.9 months (IQR 36.5–95.0). The median dose of SBRT was 45Gy (IQR 37.5–50). Both the median PFS and OS were not reached. The 5-year PFS and OS were 57.2% (95% CI 44.3%–70.1%) and 62.3% (95% CI 49.8%–74.8%), respectively. In terms of number of patients, ORRs were in 73.2% (53 patients) by RECIST 1.1 and 77.5% (55 patients) by mRECIST. In terms of number of lesions, ORRs were 78.0% (92 lesions) by RECIST 1.1 and 82.2% (97 lesions) by mRECIST. Thirty-four patients (47.9%) with 54 lesions (45.8%) received DDLT, of whom 14 patients (41.1%) with 31 lesions (57.4%) exhibited pathologic complete response. Univariable analyses revealed ≤2 lesions (vs 3) (p=0.046), partial response (PR) by either RECIST 1.1 or mRECIST after SBRT (p=0.001), and DDLT (p=0.001) were prognostic of PFS, while tumours within Milan’s criteria (p=0.004), PR after SBRT (p=0.013) and DDLT (p<0.001) were prognostic of OS. Multivariable analyses revealed that PR after SBRT (HR 0.26, 95% CI 0.11–0.55; p=0.002) and DDLT (HR 0.27, 95% CI 0.12–0.64; p=0.001) was prognostic of PFS, while tumours within Milan’s criteria (HR 0.33, 95% 0.17–0.71; p=0.011), PR after SBRT (HR 0.35, 95% CI 0.17–0.80; p=0.011), and DDLT (HR 0.15, 95% CI 0.06–0.41; p<0.001) were prognostic of OS. One patient developed grade 3 peptic ulcer treated with distal gastrectomy during DDLT and one patient had grade 3 ascites after SBRT resolved after drainage. Eight and four patients died of progressive HCC before and after DDLT, respectively. Another one died of postoperative complications and liver graft failure soon after DDLT.

Interpretation

Our study represents the world’s largest prospective cohort demonstrating a high objective and pathological response, promising survival and safety outcomes of SBRT before DDLT for HCC. Prognostic subgroups were identified which may guide future patient selection.