Rivia pushes AI deeper into clinical trial data cleaning as Scancell signs on early
Swiss data company Rivia has moved its clinical trial platform further into automated review with the launch of DataFindings, an AI-powered module built to speed data cleaning and reduce the lag between database lock, analysis, and regulatory submission. The new add-on sits on top of Rivia Core, the company’s platform for unifying fragmented trial data into a single structured foundation drawn from multiple systems and vendors.
The commercial logic is straightforward. Clinical development slows when data issues remain unresolved at the end of a study, and every extra week spent reconciling mismatches delays the handoff from trial completion to interpretation. In a setting where timelines are already compressed by protocol complexity, vendor sprawl, and rising data volume, a platform that can continuously inspect incoming data rather than waiting for a late-stage cleanup sprint offers a different operating model.
Clinical trial data cleaning has long been one of the least visible bottlenecks in development. The task is not just clerical; it directly affects whether a sponsor can trust the dataset enough to analyze outcomes and submit to regulators. Rivia’s pitch is that AI agents can help teams define data review rules, then keep watch over the dataset as it evolves, flagging changes that need human attention before they accumulate into a backlog.
Continuous review replaces the end-of-study scramble
The module begins with AI-assisted rule creation. Once those rules are in place, DataFindings monitors trial data continuously, looking for changes that trigger review. When an issue appears, users can assign tasks to team members inside the platform, and every action is preserved in a complete audit trail.
That workflow matters because end-of-study cleanup is usually where complexity becomes most visible. By the time a trial closes, multiple sources may need to be reconciled, queries may have piled up, and the burden of documenting each decision can be substantial. A system that keeps the review process active during the study may reduce the risk that problems surface all at once after the last patient visit.
One sentence captures the operational aim of the launch: Scancell hopes to cut end-of-study data cleaning from eight weeks to two.
The U.K.-based cancer immunotherapy biotech is serving as a design partner and the first adopter of DataFindings. Joe Thornton, vice president of clinical operations at Scancell, said the tool has enabled the company to translate phase 2 clinical trial data with greater granularity, confidence, and efficiency. Those words point to the central promise of the platform: not merely faster cleanup, but cleaner interpretation of the data already collected.
Why trial infrastructure now carries more weight
Rivia’s chief executive, Erik Scalfaro, framed the launch around a broader shift in drug development. As trial programs become more complex, the underlying data infrastructure has to do more than store information; it has to manage it intelligently. His argument reflects a reality familiar to clinical operations teams: when protocols grow more intricate and data streams multiply, fragmented systems can turn routine review into a slow, manual process.
The problem is especially acute when trial data are spread across separate vendors and operational systems. Each layer adds opportunities for inconsistencies, delayed queries, and repeated reconciliation. Rivia Core is designed to bring those inputs together, and DataFindings extends that function by adding continuous review logic on top of the unified dataset. In practical terms, the system is trying to shift data oversight from a retrospective exercise into a live process.
That approach has clear appeal for sponsors that want to compress the time between the last data point and the final analysis package. Faster review can shorten the interval between trial completion and decision-making, but only if the automation remains transparent enough for clinical teams, auditors, and regulators to follow every correction. Rivia’s emphasis on an audit trail suggests the company understands that traceability is not optional in this environment.
Scancell’s role signals where the first demand is coming from
Scancell’s adoption is instructive because it comes from a cancer immunotherapy biotech, a field where trial datasets can be operationally demanding and where development timelines are closely tied to proof of biological effect. In that setting, delayed data reconciliation can slow the interpretation of early efficacy and safety signals, especially when the study design requires detailed review across multiple endpoints and time points.
The launch also lands after Rivia secured $15 million in Series A financing earlier this year to scale the platform and embed more intelligence into clinical development workflows. That funding round provided the company with the capital to expand beyond core infrastructure into a more active layer of automation, and DataFindings is the first visible expression of that strategy.
For sponsors, the appeal is not only speed. A system that continuously reviews data may also reduce the chance that inconsistencies are discovered late, when fixing them is more disruptive and more likely to affect submission timing. In a regulated setting, however, automation has to prove that it can assist without obscuring judgment. Human review remains essential when a flagged issue requires interpretation, correction, or escalation.
The launch leaves one important operational question unresolved: how broadly the module can deliver the same efficiency gains outside a design-partner setting, across trials with different structures, vendors, and data burdens. The answer will depend on how well the platform performs when it is asked to handle more than a single early adopter’s workflow, and whether the promised reduction in cleaning time holds when used at scale.