On Selection and Analytical Transparency in the <scp>FRAME</scp> Framework for <scp>RWE</scp> Evaluation

We read with great interest the paper by Condore et al. (2025) on the FRAME framework for real-world evidence assessment and commend the authors for addressing this important topic.1 However, certain methodological choices may constrain the validity of the resulting conclusions. Only 15 of the 87 identified indications were analyzed, yet neither the complete list nor the selection rationale was provided. While prioritization criteria are outlined, the purposive sampling approach—designed to ensure diversity across agencies and outcomes—may limit generalizability. The authors aimed to “prioritize products assessed by multiple authorities,” yet table 1 shows palbociclib, tenecteplase, catridecacog, and tacrolimus were each assessed by only one regulatory authority (Food and Drug Administration (FDA) or European Medicines Agency (EMA)). Another criterion was to include both positive and negative opinions. However, given the research question “which characteristics might be associated with an increasing role of RWE in decision making” the selection rationale remains unclear. Regarding positive opinions, why was only atidarsagene autotemcel included when other products (cerliponase alfa, asfotase alfa, sebelipase alfa, risdiplam, and eladocagene exuparvovec) had Real World Evidence (RWE) (including natural history studies) accepted (e.g., by G-BA) as primary evidence?2 Some of these were also assessed by CDA-AMC or PBAC, whereas atidarsagene autotemcel reportedly was not.1 The selection appears questionable, as several included products (131I-omburtamab, alpelisib, palbociclib, tenecteplase, catridecacog, abatacept, and tacrolimus) received positive regulatory assessments based on RWE as primary or supportive evidence. However, since no RWE was submitted to Health Technology Assessment (HTA) bodies for these products, no corresponding HTA assessment occurred, limiting cross-agency comparability. More informative examples would have been products such as ciltacabtagene autoleucel, amivantamab, tepotinib, and capmatinib, for which RWE was submitted and evaluated by both regulatory and HTA authorities. In contrast, the study highlights entrectinib (the product of the article’s sponsor) as the only case with critical RWE assessments by both, although other products also received negative RWE evaluations from agencies. Moreover, a positive correlation between favorable regulatory outcomes and subsequent HTA decisions has already been demonstrated by Tomeczkowski et al. (2022).3 To achieve the intended cross-agency comparison, it would have been more appropriate to select those indications among the 87 identified for which the largest number of assessments were available across both regulatory and HTA authorities. Furthermore, the convergence/divergence analysis does not sufficiently account for differences in HTA mandates. For instance, the authors do not specify whether Haute Autorité de Santé considered RWE studies for reimbursement decisions or for added benefit ratings. Since all products are reimbursed in Germany but not in France, meaningful convergence or divergence should be analyzed only for comparable assessment types, specifically added benefit evaluations. Collectively, these clarifications would strengthen the paper’s contribution to understanding RWE’s role in regulatory and HTA decision-making. No funding was received for this work. The authors declared no competing interest for this work.