Institutional experience with multiple radiation fractionation regimens for the treatment of portal vein tumor associated thrombosis

Purpose: The optimal radiation dose and fractionation for treatment of portal vein tumor thrombosis (PVTT) remains an area of active investigation. Here, we report on our institutional experience with multiple fractionation regimens for the treatment of PVTT.

Methodology: We retrospectively assessed 36 patients with malignant PVTT treated from 2004 to 2021 at our institution, reviewing baseline characteristics, toxicity, and outcomes. The Kaplan-Meier (KM) method was utilized to estimate progression-free survival (PFS), local control (LC) and overall survival (OS). Chi-Square test and ordinal regression analyses were used to compare baseline patient and tumor characteristics, radiation details, and treatment outcomes. Toxicity was graded using CTCAE v5.0 criteria.

Results: The median biologically effective dose (BED) was 62 Gray (Gy) (Range 33.60-105.60) and fractionation schemes ranged from 3 to 25 fractions ,including conventional fractionation, hypofractionation ( < 10 fractions), and stereotactic body radiation treatments (SBRT) (≤ 5 fractions), with median treatment volumes of 174 cc (Range 16.5-1685). Treatment was well tolerated with no grade 2+ acute or late toxicities. Hypofractionation was associated with lower acute toxicity (p < 0.05). Hypofractionation was also associated with smaller treatment volumes (median 133 cc) compared to conventional fractionation (median 798 cc, p< 0.001). The 1-year LC, PFS and OS was 50%, 32%, and 44%, respectively, and did not differ among treatment regimens, PVTT etiology, or location of the PVTT within the portal system. While baseline bilirubin levels did not correlate with response, lower post-treatment bilirubin levels assessed at a median of 4 weeks after radiation were associated with a favorable PVTT response (p < 0.05). Larger radiation treatment volumes were associated with an increase in neutrophil to lymphocyte ratio (NLR) values post-treatment (p < 0.05). Patients with high baseline neutrophil to lymphocyte ratio (NLR), a negative prognostic sign in multiple cancers, trended toward worse survival (p=0.051) with a cutoff value of 4.0.

Conclusions: Taken together, our data suggests hypofractionated smaller treatment volumes had lower acute toxicity and improvement in the immune biomarker neutrophil-to-lymphocyte ratio, with similar outcomes to more protracted regimens. The convenience of a short treatment course coupled with a 4 week post-treatment bilirubin assessment may be predictive of response and warrants validation in larger prospective studies.

References: