# Addressing clinical uncertainties in ATMP reimbursement: a review of methodological guidelines and European practice

## TL;DR

This paper reviews how to handle clinical uncertainties in evaluating advanced therapies for reimbursement, focusing on methods like indirect comparisons and surrogate endpoints.

## Contribution

The paper systematically reviews European guidelines and practices for addressing clinical uncertainties in ATMP reimbursement, highlighting gaps and potential solutions.

## Key findings

- Single-arm trials are common for ATMPs, limiting direct comparisons and requiring alternative methods like indirect treatment comparisons.
- Surrogate endpoints are widely used but require careful validation, with varying acceptance criteria across guidelines.
- Innovative trial designs may help generate more robust evidence for ATMPs in the absence of traditional comparative data.

## Abstract

Advanced Therapy Medicinal Products (ATMPs) often present substantial clinical uncertainties at the time of reimbursement evaluation, particularly due to the lack of appropriate comparators and the absence of long-term clinical endpoints. This study primarily examined two methodological areas relevant to these challenges: indirect treatment comparisons (ITCs) and surrogate endpoints. In addition, the review was supplemented with an assessment of innovative trial designs to explore how emerging approaches may contribute to evidence generation for ATMPs.

A structured literature review was conducted using PubMed and Embase, combining keywords such as “Health Technology Assessment,” “ATMP,” “Indirect Treatment Comparison,” and “Surrogate Endpoint,” supplemented with grey literature searches. European pharmacoeconomic guidelines were analyzed using the ISPOR database and referenced national documents.

By April 2025, 27 ATMPs had received European Medicines Agency approval; 20 relied on single-arm trials, and 19 used surrogate endpoints. Single-arm trials limit direct comparative effectiveness assessment, requiring alternative approaches such as ITCs. Conventional ITCs need a common comparator, which is unavailable in single-arm evidence, while population-adjusted indirect comparisons (PAICs) offer a potential solution but depend on strong assumptions and are sensitive to unmeasured confounding. Guidelines generally accept the use of ITCs when direct comparative evidence is unavailable. However, they provide limited guidance on the separate question of when RCTs should be considered infeasible. Surrogate endpoints are widely used due to curative aims of ATMPs, but their validation typically requires RCT evidence and is highly context-specific. Guideline acceptance criteria for surrogate endpoints vary. Some require robust validation demonstrating a causal relationship, while others allow correlation-based or context-specific justification, particularly when direct evidence is limited or the technology addresses rare or serious conditions. Innovative trial designs may help generate more robust evidence in these challenging contexts.

The study highlights the need to refine methods for handling single-arm trials, projecting long-term outcomes, and using surrogate endpoints in ATMP evaluations. While some clinical uncertainty is inevitable, healthcare systems must manage it effectively to ensure timely, equitable access without creating additional barriers.