Advanced therapy medicinal products (ATMPs) include gene therapies, substantially modified somatic cell therapies, tissue-engineered products and combined ATMPs [1]. Their medical indications cover a wide spectrum of human pathologies, such as cancer, neurodegenerative diseases, cardiovascular or tissue damage [2]. By the end of 2021, 19 ATMPs had been authorised by the European Medicines Agency (EMA), 6 of which with conditional approval (CMA) and 2 under exceptional circumstances (ECMA) [3] (See supplementary Table S1: ATMP with marketing authorization and withdrawn). It has been estimated that 1100 biotech and pharma companies are developing these types of products [4] and forecasts suggest that between 10 and 20 new ATMPs could be approved per year by 2025 [4]. Over 1000 clinical trials on ATMPs were underway in 2019, of which 152 were in Phase 3 [5], and in 2020 249 clinical trials were in development or completed on ATMPs in the clinicaltrails.gov database [6].

However, regulatory approval does not guarantee commercial success, and 6 ATMPs have been withdrawn from the market in Europe by 2022, for reasons that are not always clear [3]. Industry analysts suggest various factors, including unfavourable interim data from Phase III trials (Zalmoxis) [7], failure to obtain reimbursement in key European health systems (Glybera) [8] multiplicity of criteria for health technology assessment (HTA) for adoption in different European countries [8], complex administration and manufacturing (Provenge) [9], and competition from cheaper alternatives (MACI, Provenge) [10, 11].

The path from basic research to healthcare delivery is often termed “translation” but this term is used in diverse ways and hence often leads to confusion. Woolf et al [12] distinguishes between T1 translation, the enterprise of translating pre-clinical (animal) research into clinical (first-in-human) studies, and T2 translation, which refers to moving from clinical trials to healthcare practice.

ATMPs encounter barriers to translation in both T1 and T2 [13] (Figure 1). With this in mind, EMA created the Committee for Advanced Therapies (CAT). The CAT works together with other EMA committees in the regulatory evaluation and surveillance of ATMPs, as well as initiatives for unmet need, such as PRIME [14]. There are differences in evidence requirements between regulatory and HTA agencies. EMA marketing authorisation is based on the benefit-risk ratio, manufacturing quality and level of unmet medical need, whereas HTA agencies are tasked with assessing comparative efficacy, budget impact, and/or the cost-effectiveness of a therapy [7]. However, HTA agencies in each country apply different criteria and it has been suggested that such assessments do not always take account of the special character of ATMPs [5].

In the case of ATMPs, European regulation establishes an alternative route to patient access without EMA marketing authorization [15]- the hospital exemption (HE). Under the supervision of the national competent authorities in each country, HE allows the non-routine use of an ATMP for patients who lack therapeutic alternatives without evaluation by the usual EMA and national HTA agencies. These authorisation rules differ from country to country, which can lead to inequalities in access to therapies between patients in different countries [16].

There have been several previous reviews of the challenges and incentives faced by stakeholders of ATMPs at the preclinical, clinical and clinical practice phases in the development pathway and from distinct perspectives: discontinuation of clinical trials [17], accelerated regulatory assessment [14], use of non-randomised studies in regulatory approval [18], horizon scanning [19], methods for HTA [20], methods for economic evaluation in HTA [21,22], ethics in HTA [11], and financing ATMP [23].

This paper adds to this literature an umbrella review, that is, a review of the previous systematic reviews (SR). We aim to collect all the evidence from existing reviews to give a high-level overview. This approach is considered especially useful and pragmatic when there are multiple interventions of interest [24]. Our aims are to summarise, according to the literature, the main barriers for the evaluation of ATMPs and their translation into practice across all the points in the development pathway and all the different pathologies.

We define "a barrier" as a statement made by the authors of the SR that represents an obstacle to granting a positive recommendation by regulators or HTA agencies and/or to facilitating its translation into clinical practice. If a barrier exists, it does not necessarily mean that it should be removed. Control mechanisms protect patients from unnecessary harm and ensure that therapies are effective, but they should not prevent or slow down access to medicines with a favourable benefit-risk and cost-effectiveness profile.

The article is structured as follows. The methodology section sets out the search strategy and the umbrella review methodology. The results section describes the characteristics of the included studies and barriers to evaluation or translation. In the discussion part, we comment on the implications of the results for the regulation, evaluation and translation of these medicines.