In the following sections, we provide a more detailed description of each of the above-mentioned framework pillars.

Moreover, we aim to emphasize the advantages identified in our approach to R&D digital transformation at Chiesi.

4.1 Alignment with Company and R&D strategy and creation of the R&D Digital Coordination Team

The overarching framework for defining opportunities, a high-level scope, and the mandate for R&D digital transformation was defined within the broader company and R&D strategic plan.

This starting point is considered of paramount importance to ensure alignment with company and R&D strategic objectives, and to ensure that patients, caregivers, and other stakeholders ultimately benefit from the transformation. This mindset helps to avoid investment in misaligned and resource-intensive R&D digital initiatives for the sake of digital hype.

One of the first steps in this process was the establishment of an R&D Digital Coordination Team. The team comprises representatives from various R&D functions with an aim to identify a single reference point for all the digital initiatives across the Chiesi R&D value chain. This ensures harmonization with other R&D functions rather than creating a new organizational silo.

It should be noted that the R&D Digital Coordination Team is not a dedicated digital working group. Following a thoughtful and precise strategic decision, it comprises digitally savvy individuals with specific operational or management roles, and who are therefore fully aware of business processes within the R&D value chain and any related issues and opportunities.

The team includes members from main R&D areas such as Clinical, CMC, Preclinical, Regulatory, Quality, Pharmacovigilance and Project & Portfolio Management and is supported by a third-party consultancy agency.

The three main objectives of the R&D Digital Coordination Team are to:

1.

Build and refresh R&D digital strategy and data governance;

2.

Set a relevant strategic roadmap and priorities; and

3.

Enable new digital capabilities and skills in R&D.

To achieve these goals, the R&D Digital Coordination Team is continuously engaged in a series of initiatives to:

Enhance cross-functional collaboration on digital transformation among different functions and disciplines across the R&D value chain;

Promote the creation of a suitable mindset supportive of digital innovation and relevant change in ways of working across the company by interacting with leadership and all other functions and involved stakeholders;

Manage the identification of any missing technical expertise or assets within R&D departments, cover any short-term gaps, and satisfy longer-term needs for internal digital competences and new skills;

Gain access to external expertise and assets, including facilitating participation in consortia and networks focused on digital transformation and collaboration with third parties able to provide useful technologies; and

Identify and refresh long-term objectives in the digital area, then define the roadmap to achieve them.

4.2 Identification of Key Strategically Ownable Digital Focus Areas Across R&D Value Chain

This process, starting from an initial identification of >50 data sources, >25 methods and tools, >10 platforms, and >400 potential value-generating opportunities, allowed us to identify the most promising value generation areas along the R&D value chain. The main result has been the identification of six key focus areas to concentrate efforts on in the mid-term, as well as 10 satellite focus areas to complement these efforts in the longer-term (Fig. 1).

Fig. 1

Key digital focus areas along R&D value chain. RWD real-world data

The definition of priorities for the medium- and long-term, outlining the business areas where new technologies will add the most value throughout the development chain, is a fundamental step to enable focus and the efficient use of resources. This is especially important for a mid-sized company.

4.3 Definition of the Digital Strategic Roadmap (until 2030)

Having identified the main areas of action, all ongoing projects and new digital opportunities within those areas were mapped together with their own specific digital enablers (including tools, platforms, and methods). Any possible points of contact between projects were then investigated to find synergies and common processes, tools, and/or data sources to optimize the use of related resources.

The analysis resulted in the identification of three main clusters of data needed to support R&D digital transformation, which deserve dedicated attention, them being either proprietary or public, regulated, involved in good practice guidelines for pharmaceuticals framework (GxP):

1.

GxP/Regulated proprietary data and documents;

2.

Non-GxP/Non-regulated proprietary data; and

3.

Publicly available data.

Another important result of the mapping exercise was the acknowledgement that, in order to ensure connection between the three data clusters and satisfy cross-functional data needs, it would be necessary to create a common, unique Chiesi digital R&D data model and improve data governance within the company (Fig. 2).

Fig. 2

Main clusters of data across Chiesi R&D value chain. GxP good practices

We believe that by identifying data clusters deserving of dedicated treatment, establishing a unique common data model to connect them, and enacting robust data governance practices, we form a foundation upon which diverse digital initiatives can build synergies and optimize data management.

4.4 Creation of a Common Language within the R&D Value Chain Through a Semantic Data Model, Identifying and Implementing Specific Use Cases

The semantic data model was identified as the most appropriate to describe and interrogate Chiesi R&D data. These models act as a semantic structure tying together the organization’s information and data, therefore improving the interoperability between machines and humans, and enabling access to the right data despite any physical boundaries of underlying data silos.

In the pharmaceutical sector, this could represent a fundamental competitive advantage. Enabling quick access and navigation of data generated across different areas of expertise, irrespective of possible underlying data silos, enables stronger and quicker interdepartmental communication and cooperation. In practice, this means efficiently answering business questions that require cross-functional data analysis, reducing redundancies, and supporting decision makers in taking informed action to improve business outcomes.

Ontologies are the most popular semantic data models, enabling the provision of data representations that are fully compliant with FAIR principles (Findable, Accessible, Interoperable and Reusable).

In light of the above considerations, the Chiesi R&D Digital Coordination Team agreed to create an ontology-based data model to enhance collaboration and better support interoperability of data originated within different areas of the R&D value chain.

The first step of this journey was the selection of the methodological approach used to support ontology development. After a careful analysis of the most popular approaches proposed in the scientific literature [32,33,34], the team chose the bottom-up iterative approach illustrated in Fig. 3.

Fig. 3

Methodological approach adopted for the development of Chiesi Digital R&D ontology-based data model

In alignment with this approach, the ontology has been built through the following phases:

Knowledge acquisition: primary terms belonging to each knowledge area along the R&D value chain were identified through dedicated meetings involving the main relevant R&D subject matter experts.

Conceptualization: the team reached an agreement on terms and definitions through a series of plenary team workshops. Afterwards, they identified main ontology concepts and enriched them by specifying the hierarchical relationships and associations between them.

Integration of existing ontologies: the ontology under development was integrated with public-domain ontologies and standard dictionaries to exploit previously established conceptualizations. Ontology mediation was performed following the ontology mapping approach that consists in a (declarative) specification of the semantic overlap between different entities of two ontologies [35].

Formalization: all gathered knowledge was then formalized in the W3C Web Ontology Language (OWL) with the support of Protégé, one of the most popular free, open-source ontology editors.

Evaluation, documentation and maintenance: the ontology was evaluated to assess the extent to which ontology specification requirements have been satisfied.

The most valuable result of these activities is the Chiesi Digital R&D OWL Ontology, which includes over 400 classes interconnected by nearly 500 relationships (Fig. 4).

Fig. 4

Chiesi Digital R&D OWL Ontology

The ontology includes concepts belonging to the R&D value chain in its entirety, semantically linked to the most well-known and widely accepted public ontologies in the biomedical domain (e.g., MeSH, ChEBI, NCIt, MedDRA, SNOMED CT). Cross-mapping with public ontologies is crucial for achieving a more connected, interoperable, and standardized information ecosystem, as well as enabling easier and faster updates to the overall ontology based on revisions to public ontologies.

The Chiesi Digital R&D OWL Ontology will facilitate data integration, promote semantic consistency, and support knowledge sharing across various domains and applications of the R&D value chain. Moreover, it will support the semantic analysis of data belonging to both Chiesi internal and external data sources, enabling the identification of actionable insights to drive business decisions.

To test the ability of the newly created Chiesi Digital R&D OWL Ontology to support data integration across the R&D value chain and to boost data-driven decision-making, the team decided to apply the ontology model to specific business cases. This approach follows the “lean startup” validated learning approach articulated by Ries [36], which is based on quickly testing hypotheses in the field.

The following process was followed to identify use cases that could help validate the ontology-based approach:

Step 1: Identify domain stakeholders’ main strategic interests. Here, they are represented by the key focus areas previously identified.

Step 2: Identify goals/tasks which must be achieved to satisfy the higher-level strategic interests. In our analysis, more than 40 use cases were initially identified by the full Chiesi R&D digital team and R&D subject matter experts in co-creation workshops.

Step 3: Define a priority list of the goals/tasks to be implemented first. Subject matter experts and data scientists worked together to select the first two use cases to be implemented according to two criteria: (1) business relevance and (2) technical feasibility (Fig. 5).

Fig. 5

Use cases evaluation criteria. FAIR Findability, Accessibility, Interoperability, Reusability

The evaluation criteria identified to prioritize use cases are the result of an attempt to balance long-term goals with quick-win use cases. While maintaining alignment with the long-term digital R&D strategy, the selected use cases can provide immediate and tangible results, boosting morale and engagement within the organization. This early success also helps to gain support from stakeholders and leadership, who may be more willing to invest in long-term goals once presented with tangible, short-term benefits. Moreover, implementing quick-win use cases allows organizations to test ideas and technologies on a smaller scale, minimizing potential risks and failures before committing to broader initiatives and larger related investments.

The use cases that have been selected will support the early stages of the R&D process (identifying potential correlated biomarkers between animal models and humans for a given disease) and pharmacovigilance (predicting possible side effects of drugs before they reach the market, starting from adverse events reporting). The implementation of the two use cases is ongoing with the support of cutting-edge semantic analytics technologies provided by a market leadership provider.

4.5 Implementation of a Monitoring Tool to Oversee and Control Planned and Ongoing Digital Initiatives

Recognizing the need for a monitoring tool, the R&D Digital Coordination Team deployed a Business Intelligence dashboard to oversee the progression of digital transformation and measure the value generated. The dashboard provides a series of business insights, including the number of ongoing digital initiatives per R&D digital strategic focus area. For each initiative, it offers a concise description of the main objectives, deliverables, timelines, and costs. This dashboard enables iterative monitoring of the R&D digital roadmap, thereby facilitating informed decision making on R&D digital project prioritization in the event of strategic adjustments

4.6 Discussion of Overall Process and Future Aims

In the context of pharma digital transformation, the literature presents a spectrum of frameworks. These range from comprehensive approaches to those focusing on either organizational cultural change, specific R&D value chain segments, or purely technical aspects [37,38,39,40]. The P.O.L.A.R. Star framework stands out as a comprehensive approach. It captures the extensive scope of holistic models while also addressing the strategic imperatives, stages, and technical considerations necessary for a detailed focus. Its adaptability positions the P.O.L.A.R. Star framework as a practical guide, adapt at navigating the theoretical breadth and the tangible demands of R&D in the pharmaceutical sector, thus ensuring a relevant and effective digital progression.

Although the Chiesi R&D digital transformation is ongoing and far from over, in summary we believe that the definition of a common framework to develop a unified R&D digital strategy enabled a solid foundation for digital transformation in R&D.

The implementation of the P.O.L.A.R. Star framework has already yielded tangible outcomes:

Enhanced communication effectiveness in sharing the R&D digital strategy and roadmap, facilitating closer collaboration with key stakeholders involved in the digital transformation process both internally (e.g., R&D Subject Matter Experts, Global Transformation Office, and Global Information & Communication Technologies) and externally (e.g., Consortia and 3rd party technology providers). This has also led to the establishment of a common data language across the R&D value chain.

Increased efficiency in the prioritization of digital initiatives targeting specific areas of the R&D value chain, accompanied by improved resource monitoring.

The nurture of an internal mindset that embraces innovation and encourages active participation in key external networks and consortia.

The opportunity of quickly deploying use cases of interest to evaluate the ontology-based approach.

These results represent a solid backbone, enabling us to drive digitalization synergistically and systematically across departments and functions of Chiesi R&D. In this, we can maximize the value of any data generated across the R&D value chain, regardless of whether the data originated in research, early development, late development, or product life-cycle management – including in real-world setting, which are essential for the discovery and development of new medical treatments by biopharmaceutical companies.

Future R&D digital transformation efforts at Chiesi will concentrate on enabling broader access to R&D data interrogation and analysis among digitally savvy individuals in the R&D organization with a deep knowledge of cross-functional internal processes. This will ensure the development of digital solutions that are not only technically sound, but also closely aligned with the organization’s goals, processes, and needs.

It will be similarly important to implement new use cases that validate and streamline the P.O.L.A.R. Star digitalization framework and R&D digital ontology to identify new and more innovative ways to bring value to health care.

As R&D digital transformation requires a significant commitment from the R&D Digital Coordination Team, especially in the first years, and a non-negligible investment of time and economic resources, it is important that the efforts generate a positive impact for the company. At Chiesi, the construction of P.O.L.A.R. Star and the definition of strategic directives for digital transformation will lead to a long-term, positive impact on the company in the following areas:

Investment optimization: Having a clear strategic plan centered on cross-functionally agreed upon focus areas will allow us to concentrate future efforts and resources on the most promising digital initiatives. This avoids the dispersion of capital to many initiatives that may not generate value. Moreover, the semantic-based technologies and approaches that foster knowledge sharing and data interoperability are scalable and reusable to use cases across organizational units, with minimal customization effort needed. This results in additional cost savings.

Reduced time: From an organizational point of view, the definition of a clear R&D digital strategy will enable teams to find synergies between projects, share insights, and accelerate new findings. On the technological side, the implementation of a common R&D data model across the R&D value chain will speed access to information, enable connections of data generated across different departments and by different systems, significantly reducing time wasted when processing and sharing data manually.

Improved quality: All pillars of the P.O.L.A.R. Star digitalization framework – from the dedicated team to the adoption of semantic technologies for data modelling – will improve ongoing internal processes and benefit R&D outcomes.

The P.O.L.A.R. Star framework has been designed to ensure its long-term sustainability and scalability. It is adaptable to companies of any size and provides a robust approach, which focuses on strategic drivers. The framework supports iterative updates of digital initiatives, helping to address risks arising from the rapidly changing technological environment. This ensures that digital transformation efforts are aligned with the company’s strategic goals and can adapt to future challenges.