# Liver metastases in advanced urothelial carcinoma (ARON-2): do pembrolizumab and avelumab make a difference in a poor-prognosis scenario?

**Authors:** Linda Cerbone, Giandomenico Roviello, Fabio Calabrò, Tarek Taha, Enrique Grande, Kirstin Binz, Ravindran Kanesvaran, Alina Pirstuk, Ondřej Fiala, Javier Molina-Cerrillo, Teresa Alonso-Gordoa, Alessandro Rizzo, Renate Pichler, Zin W. Myint, Alexandr Poprach, Gaetano Facchini, Melichar Bohuslav, Hana Studentova, Franco Morelli, Tomas Buchler, Alessia Mennitto, Deniz Tural, Luigi Formisano, Inmaculada Orejana, Martina Catalano, Sebastiano Buti, Fernando Sabino Marques Monteiro, Andrey Soares, Shilpa Gupta, Francesco Massari, Matteo Santoni

PMC · DOI: 10.3389/fimmu.2026.1667155 · Frontiers in Immunology · 2026-02-24

## TL;DR

Liver metastases in advanced urothelial carcinoma are linked to poor survival, but immunotherapies like pembrolizumab and avelumab show better outcomes than traditional chemotherapy.

## Contribution

This study evaluates the effectiveness of pembrolizumab and avelumab in mUC patients with liver metastases, a high-risk group with historically poor prognosis.

## Key findings

- Patients with liver metastases had significantly shorter survival on pembrolizumab (9.4 months) compared to those without (20.1 months).
- Avelumab as maintenance therapy showed improved survival in patients without liver metastases (27.0 months) compared to those with (16.4 months).
- ECOG performance status was a strong predictor of survival in both pembrolizumab and avelumab cohorts.

## Abstract

Over the past decade, the treatment landscape for metastatic urothelial carcinoma (mUC) has improved significantly with the introduction of immunotherapy, targeted agents, and antibody–drug conjugates. The median overall survival (mOS) reached 36.7 months in cisplatin-eligible and 25.6 months in cisplatin-ineligible patients in the first-line setting and over 10 months post-platinum failure. However, liver metastases remain a known poor prognostic factor.

We conducted a retrospective analysis of mUC patients treated at 79 global institutions. Two cohorts were defined: cohort 1 included patients who progressed after platinum-based therapy and received pembrolizumab, and cohort 2 included patients who received avelumab as maintenance therapy. Treatments were administered between 1 January 2016 and 31 October 2024.

Cohort 1 (n = 1,341) had an mOS of 17.5 months. Patients without liver metastases had significantly longer OS than those with liver involvement (20.1 vs. 9.4 months, p <0.001). Among patients with liver metastases, OS was 11.8 months in males vs. 5.8 months in females (p = 0.066). OS was longer in those with BMI ≥25 kg/m² (14.1 vs. 8.1 months, p = 0.028) and better ECOG-PS (ECOG 0: 17.0 months; ECOG 1: 9.8; ECOG ≥2: 3.1; p <0.001). Cohort 2 (n = 291) had an mOS of 25.8 months. Again, OS was longer in patients without liver metastases (27.0 vs. 16.4 months, p <0.001). Among those with liver involvement, OS was 14.7 months in males and 20.0 months in females (p = 0.310). Patients with BMI ≥25 had non-reached OS versus 17.1 months in those with lower BMI (p <0.001). ECOG-PS remained a strong prognostic factor (NR for ECOG 0; 14.7 months for ECOG 1; 4.6 months for ECOG ≥2, p <0.001).

Liver metastases are associated with significantly reduced survival in patients with mUC receiving immunotherapy. However, both pembrolizumab and avelumab demonstrated improved outcomes compared with historical chemotherapy data. These findings underscore the need for personalized treatment strategies in high-risk subgroups.

## Linked entities

- **Diseases:** urothelial carcinoma (MONDO:0040679)

## Full-text entities

- **Diseases:** urothelial carcinoma (MESH:D014523), Liver metastases (MESH:D009362), liver (MESH:D017093), mUC (MESH:C538445), ARON-2 (MESH:D020803)
- **Chemicals:** avelumab (MESH:C000609138), platinum (MESH:D010984), cisplatin (MESH:D002945), pembrolizumab (MESH:C582435)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12971714/full.md

## Figures

4 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12971714/full.md

## References

45 references — full list in the complete paper: https://tomesphere.com/paper/PMC12971714/full.md

---
Source: https://tomesphere.com/paper/PMC12971714