Pricing for Multi-Indication Drugs in the Italian Regulatory Context

Different clinical conditions may involve common pathways, mechanisms, or mediators, thus some drugs, especially those used in oncology and immunology, have multiple therapeutic indications [1‐9]. Multiple indications of medicines have led to a significant challenge in current drug pricing and reimbursement systems in many countries. Several studies have investigated how prices are negotiated following an extension of indications (EoIs), but it is not possible to define a standardized method for evaluating each indication because of the heterogeneity of drug samples and the different negotiation methods used in pricing and reimbursement systems.

Traditionally, a single price has been applied to drugs regardless of the number of indications they treat. However, this approach often fails to capture the varying clinical value across indications, delays access for patients, and reduces incentives for pharmaceutical companies to expand research and development into new indications [10]. The need for a more nuanced system of drug pricing that accounts for the differential benefits across indications has given rise to concepts such as indication-based pricing (IBP) [11].

There are three primary models of IBP: distinct brands with separate pricing for each indication, an average price across all indications, and a single price with differential discounts for various indications [12]. Many countries have adopted mechanisms to align drug prices with outcomes, such as monitoring drug use by indication and linking reimbursement to clinical outcomes [13]. Systems such as managed entry agreements (MEAs), clinical restrictions, and risk-sharing arrangements are frequently used [14]. However, implementing IBP requires substantial changes in how medicines are priced, procured, and monitored, as the administrative burden of tracking drug use by indication remains a significant concern [15, 16].

Recent studies indicate that drug pricing is often not commensurate with the clinical benefits provided [17, 18]. This raises concerns about the quality of evidence used, particularly for initial approvals that often rely on single-arm studies or open-label randomized controlled trials. These trials receive regulatory advantages despite being based on weaker evidence [19, 20]. In oncology, a “narrow-first” strategy is common, where the first indication launched demonstrates the highest clinical benefit in a smaller, well-defined population. This approach allows for differentiated pricing and builds a favorable reference point for later indications. Conversely, a “broad-first” strategy, often seen in targeted autoimmune therapies (e.g., anti-TNF, IL-inhibitors, JAK inhibitors), seeks to maximize early revenue by targeting broader patient populations [21].

A consensus of 16 experts from 11 countries assessed current payment models, focusing on the challenges posed by drugs with multiple indications. They identified opportunities to optimize existing systems to improve patient access and support pharmaceutical innovation. Differentiating pricing based on value at the indication level emerged as a key solution, with approaches such as blended pricing, single prices with differential adjustments, and outcome-based payments discussed [22]. These discussions highlight the complexity of developing pricing and reimbursement models that balance access, value, and health system sustainability.

In Italy, the reimbursement and price-setting procedure for EoIs is subject to a structured decision-making process, often leading to a price reduction. After European approval by the European Medicines Agency (EMA), the Italian Medicines Agency (AIFA) decides on the reimbursement of an EoI by the Italian National Health Service (INHS). This process involves evaluating the therapeutic positioning of the drug, the risk–benefit ratio compared with alternatives, added therapeutic value, and cost-effectiveness. These steps enable value-based pricing for each indication, ensuring patient access and controlling pharmaceutical expenditure.

Italy used IBP in the past [16] on the basis of the presence of MEAs, but, as Rossini et al. showed [23], this approach prolonged negotiation times and reduced incremental discounts. Consequently, Italy transitioned to a blended pricing model (BPM), prioritizing budget impact over clinical value during negotiations for new indications. This shift has significant pharmacoeconomic implications, including longer negotiation durations, uniform discounts, and less differentiation based on specific indication value. A recent report [24] highlights that while IBP aligns better with value-based healthcare, its complexity and administrative burdens have driven a preference for the BPM. This simpler model, however, inadequately differentiates between indications.

Currently, the most common pricing model in Italy for new indications is the blended price approach. A single, weighted average price is negotiated for all indications, with the additional discount often weighted by the revenue impact of the new indication. Applying a single discount across the molecule simplifies administrative management at the regional level, as it is often unclear for which indication the drug will be prescribed and used (except when regulated by specific registries).

However, the continuous introduction of new therapeutic indications hinders determining the economic value of drugs, particularly as NHS expenditure rises; it is thus crucial to evaluate how the frequency of negotiated EoIs influences discount rates over time. Rossini et al. [23] focused on analyzing the impact of price and reimbursement negotiations for new drugs, particularly regarding the discounts achieved and the length of the negotiation process. Conversely, the aim of our study is to determine the average discount negotiated during extensions of indication and identify the specific factors (determinants) that affect the additional discount agreed upon. Although other studies [25, 26] have sought to explore how multi-indication medicines are handled in the Italian context, there is still limited evidence on the key determinants influencing negotiations. The negotiation timeline is not taken into account, as it is not a direct factor in the discount itself and depends on various elements, including subjective ones [27].

The study included a brief description of the medicines that obtained approval for at least four therapeutic indications (including the first approved indication) and were subjected to full negotiation to obtain a subset of medicines with at least three approved EoIs. This cutoff was considered necessary to conduct the analyses for evaluating drugs that, owing to transversal mechanisms of action, can have many clinical areas of interest, have an important expense impact, and for which the negotiation process is complex since the molecule is renegotiated each time. Drugs were selected from those that had the first indication authorized in the period 2003–2017. We chose December 2021 as data cutoff to outline a period in which the process of negotiating the latest indications extensions had concluded with publication in the Official Gazette of the Italian Republic of the pricing and reimbursement determination; those with expired patents have been excluded. Thus, in the analyses, we included only those medicines and related EoIs that completed the reimbursement phase.

For the selected medicines, the following information was collected: the first therapeutic indication approved and reimbursed, the EoIs reimbursed, the number of negotiation events in which multiple indication extensions were approved simultaneously, the additional discount negotiated following reimbursement for a new EoI, and the total cumulative discount for the medicinal product following all negotiation events.

Trends in the number of EoIs submitted over the years and the average additional discount observed during negotiation for the introduction of new EoIs were also analyzed. A linear regression model was employed to evaluate the correlation between the expenditure increase observed for each drug at individual negotiation events and the total discount obtained for the drug, cumulative across all negotiation events.

The associations between specific determinants that may explain the extent of the discount negotiated by each EoI were investigated. The following determinants were identified: type of drug (oncology or non-oncology), orphan drug designation, innovativeness status (full, conditional, or non-innovative) [28], progressive number of EoIs, disease incidence and prevalence, estimated number of patients, estimated revenue (starting from the second year of reimbursement), availability of therapeutic alternatives in the same clinical setting, overall efficacy outcomes (effect size), and expenditure variation (observed period from 2016–2021). Specifically, the effect size for each EoI was assessed using a score that assigns a numerical value to the primary endpoint of the clinical trial brought to support the reimbursement request, on the basis of the clinical relevance of the endpoints in the disease-specific guidelines, as suggested by the GRADE approach for selecting and rating the importance of outcomes [29]. The upper end of the scale, 9 to 7, identifies outcomes of critical importance to decision-making; scores 6–4 represent important but not critical outcomes; and scores 3–1 represent items of limited importance. After the descriptive observations of the determinants of the selected discounts were made, the independent variables (determinants) and the dependent response variable (additional discount) were categorized into dichotomous variables for complete interpretation of the results. To assess the impact of determinants on the negotiated discount, the sample was divided into those that reached an additional discount above or below the median value. The median value for the additional discount was chosen because the median is a position indicator that is not sensitive to extreme values and divides the data into two equally sized groups. This allowed us to compare two groups of equal size in relation to other factors, enabling us to see which of these characteristics are generally more present in the high additional discount group and which in the low discount group. Wilcoxon’s nonparametric test for non-independent samples was applied to observe whether differences existed between pairs of independent variables and the dependent variable of the additional discount. The chosen alpha probability level was 0.05. The null hypothesis of the test predicts that there is no tendency for either population to have an additional discount greater than that of the other population. The alternative hypothesis is that this trend is present. Regarding the orphan/non-orphan drug variable, the previously mentioned test could not be applied because no discounts above the median of 11% were found for orphan drugs.

The data for the analyses related to single medicines were retrieved from the Price and Reimbursement Negotiation system (the so-called NPR system), which collects drug dossiers submitted by pharmaceutical companies for each drug (and for each EoI) and from assessments issued during the negotiation process.

Overall, we found nine medicines that obtained initial approval between 2003 and 2017 and with at least three EoIs reimbursed until 2021 (Table 1). In total, five of these nine medicines had multiple indications in onco-hematologic therapeutic areas, while the remaining four had multiple indications in other therapeutic areas (immunosuppressants with indications in dermatologic and/or rheumatologic areas). These 9 active substances corresponded to 65 approved therapeutic indications, 50 of which were reimbursed by the NHS (including the first indication). The analysis included the 40 reimbursed EoIs negotiated in 29 negotiation events (i.e., 1.4 EoIs per negotiation event) since one drug (ibrutinib) has two first indications reimbursed.

To calculate the average additional discount obtained from the analysis of the 40 total EoIs reimbursed by the INHS, and for which the negotiation process ended in an agreement resulting in an additional discount; we have included the indications reimbursed at unchanged prices because they affect the total number of EoIs examined and reduce the impact of the discount, as we have considered an additional 0%. Therefore, the additional discounts for EoIs averaged 12.5% (95% CI 9.4–16.6%), with a median value of 11% (95% CI 6.3–13%) (Fig. 1).

All nine medicines negotiated the first EoI, obtaining reimbursement with an average additional discount of 9.4%; seven medicines that submitted NPR dossiers for the second EoI obtained reimbursement with an average additional discount of 12.7%. Although the number of medicines with more than three EoIs reimbursed declined sharply, from the 5th to the 9th EoI in reimbursement, the average discount achieved was higher (from 25.8 to 14%) and then remained stable within this range until the 13th EoI negotiated.

The relationship between the discount obtained during the negotiation and the subsequent increase in drug expenditure was explored for a subset of six active substances (four oncology and two non-oncology) out of nine because complete expenditure data were only available for the 2016–2021 time window (Fig. 2).

The 6 active substances corresponded to 23 reimbursed EoIs (23/40, 57.5% of total reimbursed EoIs) which were negotiated during 15 negotiation events.

The coefficient of determination (R² = 0.44) demonstrates a moderate correlation between the variables, suggesting

the trend; however, the magnitude of the increase in the expenditure delta is greater than the corresponding magnitude of overall discount (in percentage), as can be seen from the length of the arrows. This approach captures how percentage discounts influence absolute expenditure, even if the variables differ in units of measurement.

To identify the determinants involved in the negotiation and assess their impact on the extent of the discount obtained, we used the information collected from the 40 EoIs reimbursed.

The descriptive analysis of all determinants analyzed for the 40 EoIs is shown in Fig. 3.

Figure 3 describes all the selected variables (as pairs of independent variables divided by the median value) that may influence the extent of the additional discount granted during the negotiation phase of an EoI. The threshold for the additional discount was set at the median value of 11%.

An additional discount greater than 11% was associated with extensions beyond the fifth (statistically significant, p = 0.0004 according to Wilcoxon’s test) and for oncology drugs (statistically significant, p = 0.004 according to Wilcoxon’s test). By contrast, for variables that are not statistically significant, a minimum of 20% difference can be seen between the pairs of independent variables related to the additional discount (variables in yellow). It can also be seen that the variable comparators above four are at the limit of significance (p = 0.08 according to Wilcoxon’s test).

This study, conducted in Italy, investigated the trend of negotiations for supplemental indications in a sample of drugs with at least four authorized indications (at least three EoIs) and highlighted an average additional discount of 12.5% obtained for each EoI (median additional discount 11%).

Although this selection criterion limits the sample size, it allows us to define a subset of drugs characterized by complex multiple negotiation procedures and frequent EoIs during the considered life cycle. The sample should be considered sufficient to evaluate the effect of negotiations on multiple therapeutic indications of a single medicine and thus better suited for the objective of the analysis.

The issue of multiple EoIs is a relevant topic for the sustainability of pharmaceutical care, especially in countries such as Italy, where there is a universalistic service financed by public taxation; therefore, each new EoI for an already reimbursed medicine can be considered as a new drug, involving new patient populations, different diseases, and diverse outcomes. In the subset of drugs considered, it is observed that in most cases, the increase in expenditure over the years is not offset by the increase in the negotiated discount. While the negotiation process succeeds in securing discounts for new therapeutic indications (EoIs), these savings might not be sufficient to counteract the overall increase in drug costs. As more indications are approved for a single medicine, the use of that drug expands, leading to higher total spending. Essentially, even though discounts are being negotiated, the rising costs due to broader drug use may outweigh the savings achieved. This creates a potential imbalance, where the efforts to control costs through discounts do not fully keep pace with the increasing expenditure driven by the approval of new indications for the same medicine.

Given the limited evidence available in literature, this analysis focused on identifying potential factors that influence negotiated discounts. By examining those with the greatest impact on price setting, and correlating them with the median additional discount value, we offer an initial understanding of the relative weight each factor carries in the negotiation process. Particularly, we observed that if negotiation events involved the use of medicines with more than five EoIs or oncological medicinal products, there was a statistically significant trend toward an additional discount greater than the median of 11%. These data are in line with the expected data, as medicines with more than five authorized indications are oncology drugs; therefore, an increase in the discount does not imply low therapeutic value for these drugs but is affected by an increase in the treatable population and expenditure.

The availability of therapeutic alternatives in the analyzed clinical setting, whether more or fewer than four, appears to influence the additional discount, as it reflects the level of medical need in the disease area under consideration. Furthermore, the presence of alternatives allows for a more consistent definition of the clinical benefit provided and strengthens negotiating power when seeking discounts. This may explain why some studies in other contexts have shown that drugs targeting rare diseases often exhibit higher price increases despite weaker clinical evidence, likely owing to the lack of alternatives [30].

In our sample, the clinical outcome achieved in the pivotal study, compared with the median value of eight (very close to the maximum clinical benefit of nine), appears to influence the negotiation process, with additional discounts exceeding 11% for drugs with a clinical outcome score below eight; this reflects an appropriate valuation of the drug relative to the quality of the supporting evidence. This result contrasts with findings from other studies focused on the US context [17‐20] which operates under a very different healthcare system than the Italian one (private versus public) and highlights how the Health Technology Assessment (HTA) process in Italy aims to adequately reward therapeutic value while controlling public spending. This is also highlighted in our sample, where drugs that experienced a median consumption increase of 211% during the observation period (2016–2021) secured an additional discount of more than 11%, compared with those with a lower consumption increase. Similar trends were observed in the analysis of second-year revenue estimates, whether above or below €23,293,882, and in expenditure variations greater or less than 258%. However, it should be noted that these variables did not show a statistically significant correlation with an additional discount greater than 11%, making these observations specific to our sample and not generalizable.

The estimated number of patients to be treated (with a median threshold of 643) appears to be a key factor in securing a discount greater than 11%, as it directly influences the budgetary impact. However, in our sample, neither the prevalence nor the incidence of the disease seems to affect the likelihood of obtaining an additional discount above 11%. Similarly, the designation of innovativeness, based on national stability criteria, does not appear to impact the final discounts negotiated. Only 2 of the extensions in our sample refer to orphan drugs, while the remaining 38 pertain to non-orphan drugs, which prevents us from drawing any conclusions regarding differences between these determinants.

The main limitation of this study is the limited sample of medicines with multiple EoIs, which did not allow us to draw firm conclusions on the determinants’ analysis; the results of univariate analyses were not statistically significant for all variables, even when considering pairwise comparisons using the Wilcoxon test. Therefore, it was not possible to further investigate them in multivariate analyses to generalize the obtained results.

The relationship between the discount obtained during the negotiation and the subsequent increase in drug expenditure was explored for a subset of six active substances out of nine because it was not possible for us to access pre-2016 expenditure data through our internal databases.

This analysis represents an initial step in monitoring extensions of indications and provides insights into the levels of discounting achieved and the factors having a greater or lesser impact on the extent of the negotiated discount in the Italian context.

For a subset of medicines with a high frequency of EoIs, we observed a median additional discount of 11%. We also highlighted that the magnitude of the additional discount achieved did not correlate with the increase in spending on the medicine at the time of the last negotiation of an EoI (comprehensive of all therapeutic indications). However, it is necessary to further study the prospective association between the additional discount achieved at the time of EoI negotiation and future spending to evaluate the “effectiveness” of the agreed upon discount.

Overall, our analysis highlights the many determinants that come into play during EoI negotiations, often overlapping, and underscores the complexity of the discount and pricing process with the goal of ensuring the availability of an increasing number of therapeutic options for complex diseases, such as rheumatological and oncological conditions, while simultaneously striving to control public spending. This study was based on a small sample size, therefore, the future goal will be to integrate and expand the dataset with the most recent negotiations of indication extensions to obtain more robust results. It will be interesting to see how the introduction of the new European HTA regulation [31], which aims to harmonize HTA methodologies across the EU by ensuring consistent evaluation of clinical evidence and promoting equitable access to innovations, will consider the influence of these determinants and integrate them with national evaluations, including organizational, economic, legal, and socio-ethical aspects, when determining reimbursement and setting drug prices in Italy.

owing to the presence of confidential data with the pharmaceutical companies holding the mentioned medicines.