SSn Shipyard

Medicines, Aifa criteria on therapeutic innovativeness still to be clarified

From the promotion of research in Italy to the very identikit of innovative medicine: the document on the new criteria published by the Italian Medicines Agency is promising, but outlines a still-open construction site

by Patrizio Armeni *, Ludovico Cavallaro *, Francesco Costa *, Monica Otto *

24 July 2025

6' min read

6' min read

The publication by the Italian Medicines Agency (Aifa) of the new criteria for the attribution of therapeutic innovativeness represents an important step, which finally allows us to draw some considerations and launch some stimuli for discussion. We would like to state at the outset that these new criteria should, in our opinion, be interpreted as an open work site on which there is still much to be done.

A starting point

From this perspective, it is certainly a promising document, but there are so many areas of vagueness that a prior categorisation of some measures as 'positive' or 'negative' is sometimes impossible. This vagueness may reflect a desire to leave room for the concrete application of the negotiation process, from which useful elements for a future clarification of the criteria may emerge. In this sense, therefore, it is useful to consider this document as a starting point and not as a point of arrival. Some observations, which could prove useful in this evolutionary perspective, can already be made now. Among the many possible ones, we have extrapolated four that have captured a large part of the debate among experts in the field since the first hours after the document was released.

Research Promotion

In the text of the determination (not in the annex on the criteria) it is said that the CSE, in the evaluation of the criteria of innovativeness "may take into account that the drug has been developed and conducted predominantly in Italy" in the presence of certain conditions. This is an example of the concept of vagueness, since it is made explicit that the localisation of at least some stages of research and development in Italy is a condition to be rewarded in the assessment of the quality of the evidence, but it is not made clear how this is to be taken into account in practice. The absence of this clarification risks making it abstract. In this way, the Technical and Scientific Commission (CSE) would find itself having to declare that it has 'taken it into account', but without having clear parameters for doing so. Moreover, the criteria currently used to assess therapeutic need, added therapeutic benefit and the quality of scientific evidence are not easily adapted to an assessment that takes into account the location of the research.

Purely by way of example, it seems difficult to imagine that the quality of the trials would go from low to high just because these trials were, at least in part, produced in Italy, because this would be tantamount to distorting an evidence-based judgement with a qualitative consideration that is entirely logically unrelated to the object of the judgement itself. If anything, a critical reflection on the consistency between the characteristics of the trial population and the Italian population potentially subjected to treatment is desirable, but such an assessment is common sense and certainly already included in the past in the assessment phase.

The fate of conditional innovativeness

Another important issue concerns the category of drugs for which a conditional innovativeness designation is in force. The new criteria and the determination hardly mention conditional innovativeness, which only appears as a retrospective reference in the annexes and in the transitional mechanisms, and no current criteria provide for 'conditional' recognition ex novo. It is inferred (since it is not made explicit) that this category is to be understood as exhausted, and that the new evaluations will simply give a dichotomous outcome (innovative/non-innovative).

On this point there is a divergence between what is vaguely suggested by the Aifa document and the latest Budget Law, which establishes, as of 2025, an annual sub-fund dedicated to these drugs and amounting to a maximum of 300 million euros. The drugs included in this list, to date, according to the data attached to the Aifa determination, will not generate more than EUR 126 million in expenditure by 2024. This estimate, if the category is indeed exhausted, raises the question as to why such conspicuous resources have been allocated without a fixed deadline, subtracting them from other areas of pharmaceutical expenditure (specifically the 'full' innovative drugs). In fact, even assuming a doubling in consumption (a hypothesis that is hardly credible), the fund remains more than capacious, especially in 2026 and 2027, when the deadlines for the permanence of the conditional innovativeness requirement will gradually accumulate, thus again leading to a non-utilisation of resources available for pharmaceuticals.

In short, there is a mismatch between what is provided for in the budget law and what can be deduced from the Aifa determination on the new criteria for innovativeness. The most reasonable expectation is that, in the event of the disappearance of conditional innovativeness, the resources dedicated to it will also be reallocated to the other areas of pharmaceutical expenditure (and not, as at present, making the unused resources revert to the general disposition of the NHS).

Which scope

This is perhaps the most controversial and, at the same time, the most vague point. In fact, on the one hand, the new criteria seem to restrict the scope of the concept of innovativeness 'to those specific indications relating to serious diseases or pathological conditions with a medium to low epidemiological impact'. On the other hand, however, they define seriousness broadly (a disease or pathological condition capable of leading to life-threatening or life-threatening complications, inducing repeated hospitalisation, leading to disease progression or causing disability that impairs patients' quality of life).

Thus defined, severity could be discretely recognised in virtually any condition for which a new drug is launched. For example, an early state of a disease is defined as such because it could lead to progression but, by definition, does not generally carry an immediate risk of fatality. Similarly, the degree of 'impairment' of quality of life is an abstract concept, which can be discretionarily adapted to any starting level, even if measured correctly. The problem is that the definition of criteria, for any evaluation process, should serve to reduce and not to increase the space for discretion.

The same reasoning is partly applicable to the concept of medium-low epidemiological impact concerning the prevalence of the condition being indicated. The lack of numerical prevalence thresholds (e.g. <10/100,000, <5/10,000) makes the concept of medium-low impact extremely interpretable. This shortcoming is even more relevant if one considers that at European level, with the launch of Regulation (EU) 2021/2282 on the Joint Health Technology Assessment, there is a move towards greater methodological standardisation.

In this sense, the disappearance from the preliminary document of the partial list of excluded conditions with a medium to low epidemiological impact ('diseases with a high prevalence such as, by way of example but not limited to, arterial hypertension, arthrosis, hyperlipidaemia, chronic obstructive bronchopathy, diabetes mellitus, dementia, etc.') is a positive development. However, at the same time this deletion, which is not compensated for by objective indications, further increases the vagueness of the criterion and raises the doubt as to whether this list is still valid in the intentions of the evaluating party, who could simply exclude the main chronic conditions, with the exception of cancer, from the assessment of innovativeness. Regardless of the degree of vagueness of this criterion, this restriction is highly questionable, especially in light of the fact that, to date, competition for economic resources for innovations has not been high, even without excluding major chronic conditions from the potential perimeter of innovativeness.

Innovativeness according to Aifa

These considerations lead one to wonder what the identity of the concept of a 'drug with an innovative indication' emerges from this document. The incipit of the new criteria is broad, inclusive and promising in stating: "The innovativeness of a medicine is assessed on the basis of the production technology of its active ingredient, its mechanism of action, the way it is administered to the patient, its clinical efficacy and safety, its effects on quality of life as well as its implications for the organisation of healthcare." However, the set of subsequent indications does not fully restore this theoretical framework.

The sketch that emerges is still ill-defined and leaves ample room for interpretative discretion, both at the assessment stage and at the negotiation stage. In particular, the treatment of quality of life is partial and uneven: for example, the dimensions relating to pain and the performance of work activities are only considered for rare and ultra-rare diseases, while the anxiety-depression component is not mentioned. The three core criteria (therapeutic need, added therapeutic value, quality of evidence) remain similar to the previous ones, except for a qualitative revision of the thresholds. However, the introduction of new elements that can be assessed in an unstructured manner tends to make these criteria less central and, consequently, less readable in the overall logic of the process.

If, as we consider desirable, this document represents a starting point, it will be important to proceed as soon as possible with the operational definition of certain key concepts (such as the severity of the condition, epidemiological thresholds or levels of clinical utility deemed significant). Overall, the document certainly opens up margins for flexibility, but it also entrusts significant responsibility to the CSE, which will have to manage these margins consistently and transparently. The monitoring of the first applications will be crucial to understand whether this flexibility will be able to generate value or risk translating into further uncertainty.

* Osfar, Cergas Sda Bocconi School of Management.