Dr. Zoheb Hassan, a Principal consultant at Frost & Sullivan’s Healthcare and Life Science Practice based in London, UK, spoke with the CAS team to share his insights on the complexities that define the current drug discovery landscape. His perspectives highlight the challenges in managing and packaging the increasing flood of data into actionable insights and the primary reason drug discovery is being held back—a lack of data prioritization.    

CAS: Dr. Hassan, would you start by telling us more about your role at Frost & Sullivan?  

Zoheb: I focus on helping clients navigate and overcome their growth challenges in the drug discovery and life science sectors. Clients typically come to us with specific problems related to expanding their market presence or improving their operational efficiencies. My role involves analyzing these issues to offer actionable insights and strategic advice that can help drive their business forward. We aim to convert complex market data into understandable and executable strategies for our clients, ensuring they can achieve sustainable growth despite the industry's dynamic nature.  

CAS: I imagine this provides you with a comprehensive understanding of the industry as a whole. What do you see as the single biggest issue holding back drug discovery today?  

Zoheb: In my opinion, the core issue is that most drug discovery teams overlook the significance of data, making it a secondary consideration in their decision-making process. Everyone wants to focus on building a solution that does everything. It does the molecular modeling, it does the prediction, and it will help you reach your endpoints. But is it optimal for the drug developer and, most importantly, the patient? You must focus on the data that's working behind the scenes, which is a critical oversight. I think companies need to go back and reassess the quality of the underlying data as a first step toward solidifying their data infrastructure, which requires thorough evaluation and curation of all their datasets.  

Moreover, there is a lack of sufficient dialogue within the industry about best practices for data standardization and removing data from silos to create novel connections between datasets. I believe that increased collaboration within the industry is essential to overcome this. While every company has its own solutions and competes for market share, I see the adage 'when the sea rises, all boats will rise together' applying here. Without addressing the foundational issues of data standards and standardization methods, the limitations on what can be achieved in this space will persist.  

‍

CAS: What is the significance of establishing a solid data foundation for drug discovery?  

Zoheb: Recent shifts in our research focus underscore how crucial high-quality data is in the drug discovery process. Previously, the emphasis was predominantly on the drug target itself. However, the industry has realized that ‘garbage in, garbage out’ holds particularly true; the quality of your data fundamentally determines the output's quality. The challenge is effectively integrating diverse and often siloed data sets while simultaneously verifying the quality of that data to enhance drug discovery outcomes.   

The priority is establishing a solid data foundation or knowledge graph from which we can develop highly differentiated outputs. This strong data foundation is not just about ensuring success in getting drug assets through to the approval stage; it also plays a critical role in avoiding intellectual property issues down the line. For example, in the early stages of drug discovery, how you manage and utilize data can lead to uniquely differentiated chemistry, which maximizes success and minimizes the risk of IP litigation.  

Thus, the key challenge for drug discovery organizations is to develop robust data foundations that enable the creation of truly unique chemical entities. This involves not only building but also continuously enhancing these data foundations to support innovative drug discovery. Ensuring data integrity should involve several layers, from the initial data collection to its final analysis. Organizations must adopt stringent data validation and curation practices to maintain data quality, which is especially crucial given the complex data sets involved in drug discovery.   

Moreover, collaboration plays a critical role within teams and across the industry. Sharing best practices and standardizing data handling procedures can significantly enhance data integrity across the board. There’s also a growing trend toward creating alliances or platforms where companies can come together to discuss and overcome common data challenges. 

‍

CAS: Are you hopeful that organizations can collaborate to solve some of the industry's biggest data challenges?  

Zoheb: I am very hopeful! But, I think this starts with individual organizations increasing investment in building thought leadership around data management. Many organizations do not sufficiently focus on the data they collect, particularly data sourced from public domains, which is a critical area for improvement.   

The next step would be enhancing cross-collaboration between organizations, as this could greatly benefit the industry as a whole. Establishing a platform for sharing ideas and discussing how to integrate data from early drug discovery into regulatory submission formats could streamline the approval process.  

There is a significant discussion within the regulatory sphere about improving these processes, but such discussions must extend to data management. An alliance or platform where companies could regularly meet to address common bottlenecks and challenges in establishing a strong data foundation would be invaluable. While a competitive edge will always be sought, the overarching goal should be to benefit patients, who deserve nothing less than access to best in class, effective and safe treatments.   

By improving data management practices and fostering collaboration, organizations can speed up the development and delivery of therapies to patients, ultimately performing a great service to society by enhancing the overall data foundation landscape.  

CAS: You mentioned data management as a key component of success. Are there any common challenges your clients face regarding data management?  

Zoheb: The overall drug discovery process faces multifaceted challenges due to the increasing complexity and array of data types. Integrating biological data such as bioactivity, proteomic, and pharmacodynamic data is crucial for making informed decisions rapidly, a necessity underscored by recent global health crises such as the COVID-19 pandemic. These situations have heightened the urgency for quickly developing and approving new drugs to get them from the laboratory to the bedside without delay.  

Furthermore, the drug discovery sector is contending with not only the technical difficulties of managing these vast data streams but also procedural challenges. These include adapting to rising costs and stringent regulatory demands, which amplify the need for robust data management strategies. As the volume and diversity of data continues to expand, teams must efficiently analyze and utilize this information to drive critical decision-making processes.

‍

CAS: Do traditional intra-organizational structures impact the efficiency of information sharing and data management?  

Zoheb: Traditional organizational structures in pharmaceutical companies, often characterized by hierarchical and siloed departments, can significantly impede the flow of information and collaboration. These structures are particularly problematic in the context of modern drug discovery, where a multidisciplinary approach is essential. The isolation of departments leads to duplicated efforts and prevents effective sharing of insights and data across different teams, resulting in missed opportunities for synergy and increased inefficiencies.  

There is a noticeable shift in the industry towards adopting more integrated team structures to address these challenges, bringing together computational chemists, medicinal chemists, structural/research biologists, antibody engineering teams, pharmacometricians, pharmacists, and quantitative biologists as needed. These newer models promote cross-disciplinary interactions and leverage the collective expertise of diverse teams, which is crucial for fostering a collaborative environment.   

These integrated structures enable a more holistic approach to drug development by facilitating better communication and removing barriers to information flow. This transformation allows for faster problem-solving and greater innovation, ultimately streamlining the drug development process and enhancing the ability to respond swiftly to new research findings or regulatory changes.  

Such organizational adaptations are becoming increasingly important as the pace of scientific advancement accelerates, demanding more agile and responsive operational models to keep up with the rapid evolution of drug discovery technologies and methodologies.  

CAS: Are there broader implications for data management inefficiencies within drug discovery teams?  

Zoheb: These inefficiencies can drastically inhibit a team's ability to respond to new information and adapt strategies accordingly. This can delay the development cycle and increase the time-to-market for new drugs. Beyond just slowing down processes, these inefficiencies can lead to significant financial losses, as delayed projects mean missed market opportunities and diminished returns on investment. Additionally, when data is not utilized to its fullest potential, viable drug candidates may be prematurely discarded or overlooked entirely, resulting in significant setbacks in therapeutic innovation.   

Data curation and cleanup are currently major challenges for companies, often proving to be a burdensome process. Integrating various data sets can introduce anomalies, raising crucial questions about whether certain data segments should be removed and the potential downstream consequences of such removals—akin to a butterfly effect in scientific research.   

This is where artificial intelligence (AI) can be extremely valuable. AI has the potential to provide deeper insights and enhance foresight, particularly in predicting and managing the impact of data integration. By supporting data cleanup, validation, and curation, AI can significantly streamline these processes, mitigating risks and improving the quality of the data handled.  

CAS: Speaking of AI, where do you think it can have the most impact, and what are some of the potential pitfalls?  

Zoheb: Adopting and adapting to innovative computational techniques, particularly platforms incorporating AI and machine learning, is becoming increasingly critical. These technologies can handle large datasets with greater efficiency and enhance the accuracy of predicting outcomes, streamlining the entire drug development pipeline. AI, for instance, significantly boosts early-stage drug discovery by improving the prediction of drug-target interactions and optimizing the selection of promising leads.  

The main issue in the current landscape of AI deployment within the drug discovery industry is the rush to integrate without a clear strategy or understanding of its potential benefits. Many organizations attempt to leverage AI across various applications without pinpointing where they excel and where AI can amplify their strengths, such as in data aggregation, molecular simulation, or virtual screening.  

A crucial step for successful AI integration is clearly defining how AI will interact and enhance working with multiple data sets. Understanding and linking these data sets effectively within a client's operational environment is essential. This involves integrating AI solutions smoothly with existing systems and ensuring that AI-generated insights are effectively aligned with established data points.  

AI's real power lies in its ability to facilitate communication between different data sets, enabling deeper data analysis that goes beyond human capability. This could significantly accelerate the development of new drug assets by rapidly identifying more effective candidates. However, despite the potential of AI to revolutionize prediction modeling and simulations, the drug discovery field still faces challenges in maximizing the value extracted from data sets and improving data interoperability. The quality of AI outcomes/interpretation will be determined by the quality of your data foundation (quality data sources, well-structured data, data set interoperability/cross communication or linkages, data diversity,)

The focus should be on enhancing how data sets interact and how AI can derive and interpret complex data layers to deliver meaningful insights that can be directly applied to drug development. This approach will improve the discovery process's efficiency and ensure that new therapies are developed and delivered to patients more effectively.  

CAS: In such a complex environment, do you have any general solutions for how companies can overcome these challenges?  

Zoheb: The rapid changes in the drug discovery landscape necessitate that companies adopt new technologies and continuously evolve their business models and operational strategies to maintain competitiveness. Implementing advanced data sourcing, analytics, and management systems is crucial in this context. Equally important is fostering a culture that embraces change and promotes data literacy within organizations. Such a culture empowers teams to make informed decisions swiftly and supports the development of dynamic and responsive operational models.   

Additionally, flexible data governance frameworks are essential for ensuring data integrity and accessibility, which are key components that support regulatory compliance, facilitate innovation, and ensure development of safe drug therapies for patients. These comprehensive adaptations help companies effectively navigate the complexities of today's drug discovery environment, ensuring they are well-equipped to respond to new challenges and opportunities.  

CAS: Lastly, if you had a magic wand to change anything about drug discovery, what would you change?  

Zoheb: If I could introduce one major improvement in the drug discovery field, it would be the enhanced integration of software tools. Many laboratories and research facilities currently use multiple platforms that do not effectively communicate with one another, leading to inefficiencies and delays in the discovery process. By improving interoperability between these tools, we could significantly accelerate the development of new therapies and enhance the overall effectiveness of the drug discovery process, ultimately leading to better patient care outcomes. We're in this game together, and we all want to get therapies to patients to ultimately extend life and alleviate disease.

‍

‍