Better analytics leads to fewer regulator questions
Two of India's biologics manufacturers say stronger data analytics is cutting the number of questions regulators ask about their intensified processes.
Biocon and Ipca both intensify to bring down cost of goods sold (COGS), and both put the analytics on the same footing as the process itself, with the payoff landing in the regulatory file. Kumaran Sivalinga Vasu, functional leader for upstream at Biocon Biologics, and Sanjeev Gupta, senior vice president and head of biosimilar and advanced biotech lab at Ipca Laboratories, spoke in separate interviews at Sartorius's Process Intensification Forum in Hyderabad, held from 2nd to 4th June 2026.
Fewer questions on the file
“There are a lot of instances where we have got questions from the process characterisation, but in some of the recent programs we do have [fewer] questions,” said Sivalinga Vasu. “That means our data analytics, and how we show the data, has improved a lot.”
The mechanism runs through the chemistry, manufacturing, and controls (CMC) dossier. What makes for a good CMC package is the analytics behind it. Regulators are more readily convinced when the argument is made statistically.
Gupta described the same advantage from the inspection side, pulling process parameters, batch records, and manufacturing records into a single online system so that the evidence is fully assembled. “When the regulator comes, they ask for the data,” he said. “If you have everything in the record, everything online developed, you can show then and there the basis of scale up, the basis of changes.”
Predicting the batch, not watching it
Gupta, who holds a PhD in molecular biology and cell line development and has worked on more than 15 biosimilars launched in markets including the US and Europe, framed the problem as one of scale. “In research and development, we are working at a small scale. It works fine,” he said. “But the moment you talk about the larger scale – 1,000 L, 2,000 L, or 5,000 L scale – you have a lot of challenges.”
Ipca started employing the SIMCA Online software for real-time batch monitoring around five years ago, added design-of-experiments software, and has since brought in cell culture analytics, integrating the three. The goal, Gupta said, is producing a batch that works the first time at commercial scale. “These tools together can help in predicting the batches in a better way, and you can succeed even from the very first batch.”
Real-time monitoring should, he continued, go beyond the dashboard and be based on “sample-less” processes. “When I say sample-less, it means you don't have any sample. You have PAT tools, you have SIMCA online monitoring – everything is online.”
Sivalinga Vasu, meanwhile, focused on visibility. Bioprocessing already forces developers to hold physics, chemistry, and biology in view at once, and moving to N-1 or production-stage perfusion adds variables, rather than removing them. “You tend to increase the darkness of the black box more and more, so somewhere you need to throw some light.”
Biocon runs a separate data analytics team building models across its processes. The tools Sivalinga Vasu named include multivariate data analysis and Raman spectroscopy, which he said can be modelled to return concentrations of anything from amino acids through to vitamins. The goal is to increase the complexity of the bioprocess by intensifying it while reducing the entropy of the system, he said, though warned that by getting the analytics wrong, the development turns into “intensified chaos.”
While both Gupta and Sivalinga Vasu referenced Sartorius’ SIMCA in these examples, suppliers in general were given significant praise for industry’s PAT progress. Indeed, Sivalinga Vasu credited the users explicitly, describing roughly half the impetus behind analytics packages as coming from manufacturers needing to see their own data more cleanly.
Must have or nice to have
As to how far regulators have embraced new analytical technologies, the two interviewees were not fully on the same page.
“Right now, it's like, if you have it, it's good. It has not come to a point of a must have," Sivalinga Vasu said. “But if you have it, it's more advantageous for your file to be considered, and as well as to be reviewed faster.”
Gupta, however, said there was a turning point around 2023/2024, citing the International Council for Harmonisation's Q13 guideline on continuous manufacturing, adopted in November 2022, which was taken up by the US Food and Drug Administration (FDA) and the European Medicines Agency. He also stressed how industry led the way ahead of both agencies. “We developed the processes, we developed the data, and we submitted our dossier, and we forced them to think about the continuous processing.” Other regulators, he said, should now follow.
Regulators have had one eye on analytics far before the Q13 guidance and their guidance continues to evolve. The FDA published its process analytical technology (PAT) framework in September 2004, and issued draft guidance in January 2025 on the use of artificial intelligence to support regulatory decisions about the safety, effectiveness, and quality of drugs and biologics. Sivalinga Vasu argued this leaves agencies better briefed than industry tends to assume, since they no longer need to visit every manufacturer to work out what is being done and can read the direction of travel across the wider technology sector.
What is still missing?
Asked where the gaps in the analytical toolbox remain, Sivalinga Vasu pointed at packaging, rather than capability. Cloud computing has removed the constraint on running simulations, he said, and what is still wanted is an interface that lets scientists feed in data, run the analysis, and carry the interpretation through to a report without an intermediary. However, such gaps will close rapidly, he expects, as “there will be a humongous improvement [in the next two/three years] because of the AI boom.”
Gupta said the remaining obstacle is not a tool, but the general attitude to new tech. His talk at the forum was titled 'Biopharmaceuticals, process intensification, hypes or hopes,' and he landed on hopes, with mindset as the obstacle. Whatever legacy processes, legacy clones, and legacy facilities a manufacturer is carrying, the question is how to get the most out of them, he said, and a company that cannot bring its COGS down will not sustain itself in the market.
“Do not be scared about the process intensification,” said Gupta. “Continuous manufacturing is a big, big, big thing and challenging. Just jump into it, and then slowly and gradually start implementing one by one.”
About the author
Dan Stanton has been a B2B journalist for nearly 15 years, with expertise in the pharmaceutical and healthcare sectors, in particular CROs, CDMOs, M&A, IPOs, biotech, bioprocessing methods and equipment, drug delivery, regulatory affairs, and business development. Formerly managing editor at BioProcess International, he is currently editorial expert at Sartorius.
