Insights from ISCR 2025
Integrating innovation, technology, and patient-centric approaches to advance clinical research in India
ICON, an Accredited Member Organisation of the Indian Society for Clinical Research (ISCR), recently participated in the 18th annual ISCR conference, held 30th January to 1st February 2025 in Navi Mumbai, India. This conference serves as a pivotal platform for professionals engaged in clinical research activities across India, fostering knowledge exchange and learning opportunities. With over 2,000 delegates and nearly 30 booths representing major national and international pharma companies, global CROs, SMOs and technology companies, the event showcased the extensive capabilities in all clinical research domains in India.
Innovation, technology and patient centricity for successful clinical research in India
The central theme of 2025 conference revolved around "innovations and technology implementation in clinical research for better patient centricity." Spanning six major tracks, including Clinical Operations, Regulatory Affairs, Biostatistics and Statistical Programming, Clinical Data Management, Pharmacovigilance, Real-World Evidence (RWE), and Health Outcome Economic Research (HOER), the event explored various facets of this transformation. Discussions ranged from artificial intelligence (AI) and machine learning (ML) implementations in clinical data management (CDM), biostats, site selection, decentralised clinical trials and various other aspects where technology innovations are enhancing patient centricity.
Focus on clinical data management and biostatistics
Of particular interest were the sessions within the Clinical Data Management and Biostatistics and Statistical Programming tracks. These sessions explored aspects including:
- The role of clinical data standards in an evolving technology landscape, implementing data standards in AI/ML frameworks
- Case studies and best practices on implementing Risk-Based Quality Management (RBQM) in clinical trials
- DCTs: Addressing barriers and prioritising patient-centricity
- Convergence of generative AI and LLM technologies with CDM processes
- Turning data into actionable insights: Practical applications of data analytics for informed decision-making
- Patient centricity from statistical perspectives
- Visualisation helping clinical decision making
ICON’s industry-leading insights
ICON, with our extensive experience in clinical research acceleration, shared valuable insights on Clinical Data Science and Biostatistics & Programming during the conference. Seven of our distinguished experts and thought leaders presented the following topics, helping to drive the collective conversation toward innovation and efficiency.
Harnessing the power of DCTs: Addressing barriers and prioritising patient-centricity
Sessions chair: Kedar Deshmukh, Director, Clinical Data Science
ICON chaired this session exploring how sponsors and CROs can focus in on DCT solutions like telemedicine, home healthcare, direct to patient services/products and the Internet of healthcare things to deliver more pragmatic solutions to clinical trial challenges.
Analysis done in late 2022 revealed that 81% DCT studies out 543 DCT studies registered on clinicaltrials.gov had Internet of Healthcare Things as a key component of patient centricity, thus making device-driven digital health technology (DHT) a critical component for patient centricity.
Within patient-facing technologies making an impact on patient centricity, eCOAs and ePROs are a prime focus. Choice of eCOA administration approach depends on trial design, patient demographics, resource and technology availability. There is growing empirical evidence that ePROs provide greater opportunities in collecting continuous, right frequency, granular endpoint data to provide a holistic view for enhanced clinical evaluation. Outcome assessments can utilise different device implementation approaches, including bring-your-own-device (BYoD), hybrid and dedicated devices depending on trial design. This session highlighted the need for clinical data management to partner with clinical data systems and tech vendors to create sophisticated systems ensuring right patient authentication and the ability to flag data patterns which may indicate incorrect use, artificial conditions influencing false positive or false negative data generation, or missing data due to synchronisation issues.
Signal detection and data trending in clinical trials: Enhancing oversight with statistical monitoring
Speaker: Madhumathi Raghunathan, Group Lead
ICON’s Madhumathi Raghunathan spoke on this panel focused on enhancing oversight with statistical monitoring. Risk-based data monitoring is a powerful tool that ensures patient safety and the integrity of a trial. Recent developments in RBDM include Risk-Based Monitoring and Risk-Based Quality Management. According to ICH E6 R2, with the increase in scale, complexity and cost of clinical trials, it is necessary to implement more improved, advanced and efficient approaches. In an analysis by Medidata, it was observed that, on average, only 3% of eCRF data entered by sites are updated after their initial entry. Therefore, traditional on-site monitoring, which is time-consuming and costly and has only a marginal impact on data quality, should be shifted to an approach that is cost- effective and focuses only on critical data.
Early risk indicators and AI-driven signal detection enhance safety monitoring, enabling real-time interventions. Digital innovation helps drive efficient signal detection through data collection and integration, followed by pattern recognition, predictive analytics, real-time monitoring, risk stratification and signal validation. Advanced tools improve sensitivity, helping stakeholders make informed decisions. ICON also discussed cases based on a visual analytics tool that empowers safety data review, allowing stakeholders to make informed decisions.
Given the evolving trial landscape, RBQM provides a flexible framework to optimise quality while balancing patient safety, trial design and technology to achieve endpoint goals.
Pre-conference: Harmonising data management and biostatistics
Speaker: Rekha Mohan, Principal Clinical Data Scientist Lead
Data managers and biostatisticians work towards the same goal — collecting high quality data. However, their approaches to data management and specialist technical expertise are different. Without biostatisticians oversight, eCRF design may fail to capture key variables required for endpoint evaluation.
Data managers ensure the CRFs meet technical and regulatory standards while biostatisticians oversee the alignment in variable formats to prevent delays or errors during analysis. During this pre-conference workshop, ICON’s Rekha Mohan highlighted the importance of collaboration between both these vital functions to ensure trials are setup for operational efficiency, analytical validity and regulatory success.
Pre-conference: Data science & biostatistics: Relevance to clinical research
Speaker: Modak, Dimpy, Principal Clinical Data Scientist Lead
This discussion explored the ways integration of data science and biostatistics are fundamentally reshaping clinical research, optimising everything from trial design to real-world applications. While most are familiar with AI’s role in healthcare, few realise the depth of its impact — such as how predictive analytics can now anticipate clinical trial dropouts with up to 87% accuracy, allowing for proactive retention strategies and cost savings of millions per study.
Over the years, the industry has transitioned from basic eCRFs and risk-based monitoring to cutting-edge applications like natural language processing (NLP) automated adverse event detection, which reduces manual safety reporting efforts by 65%. Similarly, Bayesian methodologies have accelerated decision-making in trials, slashing recruitment times by 40% in adaptive oncology studies.
One of the most underappreciated innovations is AI-driven study data tabulation model (SDTM) automation, which has cut dataset preparation time by 60%, significantly reducing human error and expediting regulatory submissions. Another little-known yet powerful trend is risk-based quality management (RBQM), where just 15% of collected clinical trial data determines nearly 80% of final outcomes, shifting the industry toward more efficient, focused data collection.
Looking forward, the clinical research space is moving toward ethical AI, decentralised trials and interdisciplinary collaboration, ensuring trials are faster, more accurate, and truly patient-centric. The fusion of data science and biostatistics is not just about technological advancement, it is about redefining the future of drug development with smarter, data-driven strategies that were once deemed impossible.
Development in programming techniques
Session Chair: Smitha Nalam, Director, Statistical Programming
The field of statistical analysis in clinical research has witnessed remarkable advancements over the years, and at the heart of this evolution lies the integration of robust programming techniques that drive efficiency, accuracy and compliance in statistical analysis.
In this session, ICON’s Smitha Nalam chaired a discussion on how modern programming methodologies using R are reshaping the way data is analysed in clinical trials. From traditional SAS-based analytics to the growing influence of open-source tools like R and Python, programming in clinical statistics is undergoing a transformation, enabling faster and more transparent decision-making.
Presenters explored how an excess of Tables, Figures and Listings (TFLs) for a Clinical Study Report (CSR) pulls time and resources for programmers in terms of development and validation and described how an app, Rshiny, could assess the utilization of TFLs in CSRs. Rshiny compares the list of generated TFLs with the ones used in the CSR and identifies the ones that are not being referenced to help avoid unnecessary outputs that will make team processes leaner and more efficient.
Another presentation explored the concept of multiple imputation to handle missing data effectively using different techniques of SAS MI, R MICE and R Amelia packages. Presenters explained these packages, differences in assumptions and approaches, and provided a case study to illustrate how packages can be adapted with data.
Advancing innovative adaptive designs in clinical trials using Bayesian approaches: A different perspective of patient centricity
Speakers: Dr. Sabyasachi Mukherjee, Senior Biostatistician & Supratik Chakraborty, Principal Biostatistician
Adaptive designs in clinical trials are increasingly popular due to the flexibility to modify an ongoing trial, including the ability to conclude the trial based upon interim analyses, sample size re-estimation (SSRE), deleting or adding a treatment arm or extending an existing one, and stopping the trial due to evidence of efficacy, futility or safety.
Adaptive trials can also be used as an innovative approach to significantly enhance patient centricity in clinical trials. They improve patient safety over traditional trials by decreasing the likelihood that patients will be exposed to ineffective treatments or placebo any longer than necessary. The combination of Bayesian methods and the flexible structure of adaptive trials enable adaptations to occur following prespecified updating as the data accrue and are analysed in conjunction with prior evidence.
Futility assessment has the potential to increase efficiency, save patients and decrease costs in clinical trials. However, stopping trials early may lead to abandonment of the final analysis which could potentially demonstrate a positive result or a trend towards a positive result. One may save a trial that is not on the right trajectory by using SSRE approach. Researchers can plan for a lower effect size or sample size to begin with but could still stop the trial early and change this.
This presentation emphasised the importance of selecting appropriate sample sizes and appropriately timing the declaration of study futility based on interim results. We discussed the futility analysis and SSRE procedure using posterior predictive power approach illustrating through some suitable study examples in the context of patient centricity.
Aligning expertise and innovation to forward India’s clinical research arena
The 2025 ISCR Conference was a powerful testament to the impact of collective expertise and innovation in clinical research. As experts, thought leaders and industry pioneers came together, the discussions went beyond blue-sky innovation and dove deep into the real-world implications for patients, policies and the future of healthcare.
Contact us
For further insights or to discover how ICON can support your clinical data management and biostatistics needs, please contact us. We look forward to partnering with you on your journey toward better patient outcomes.
Authors:
Kedar Deshmukh
Smitha Nalam