# Visit‑to‑visit variability of inflammation–immunity indices and prognosis in hepatocellular carcinoma

## TL;DR

This study shows that changes in blood inflammation-immunity indices over time predict survival in liver cancer patients, beyond standard measures.

## Contribution

The study introduces visit-to-visit variability of the systemic immune-inflammation index (SII-VIM) as a novel prognostic factor in hepatocellular carcinoma.

## Key findings

- Higher SII-VIM variability was independently linked to worse survival in HCC patients.
- Adding SII-VIM improved survival prediction beyond standard factors and mean index levels.
- Results were consistent across multiple analytical approaches and time windows.

## Abstract

Inflammatory–immune blood indices such as the systemic immune‑inflammation index (SII) relate to outcomes in hepatocellular carcinoma (HCC). Visit‑to‑visit variability (VVV) may capture instability of host inflammatory homeostasis that is not reflected by average levels. We examined whether early VVV of these indices predicts overall survival (OS) and adds discriminatory value beyond standard prognostic factors and mean levels.

We conducted a single‑center retrospective cohort of adults with HCC from June 2020 to December 2024. All laboratory “visits” within days 0–90 informed per‑patient VVV; the primary exposure was SII variability independent of the mean (SII‑VIM), with average real variability (ARV) and SD/CV% as secondary metrics. The primary outcome was OS from the day‑90 landmark. Predictive increment was assessed by change in Harrell’s C‑index, time‑dependent AUC (1 and 2 years), and NRI/IDI with bootstrap internal validation; sensitivity analyses included a 0–180‑day window, time‑varying rolling VVV, and multiple imputation (m = 20).

Of 780 patients, 528 survived to the landmark and were analyzed. Median follow-up was 25.1 months with 216 deaths (41%). Each 1-SD increase in SII-VIM was associated with higher mortality (adjusted HR 1.26, 95% CI 1.12–1.41); the highest versus lowest quartile yielded HR 1.72 (95% CI 1.25–2.36). Splines showed a monotonic dose–response. Adding SII-VIM improved discrimination from C-index 0.68 to 0.71 and increased AUC from 0.70 to 0.73 at 1 year and from 0.67 to 0.70 at 2 years; continuous NRI was 0.14 (95% CI 0.06–0.22) and IDI 0.018 (95% CI 0.008–0.030). Results were consistent using ARV (HR 1.20), a 0–180-day window (HR 1.22), and time-varying analyses (HR 1.18).

Early visit-to-visit variability—particularly SII-VIM—was independently associated with worse OS in HCC and added modest, clinically meaningful discrimination beyond standard factors and mean levels, and prospective validation is warranted.

The online version contains supplementary material available at 10.1186/s12876-025-04506-6.