Synexis Life Sciences
Discover the signals others miss.
Reveal what they mean.
Detection is not investigation. Reveal.IQ is an early signal discovery and governed investigation platform for clinical trials, built for the people who have to know what is happening inside a study and answer for it.
Rules detect. Discovery uncovers. Playbooks investigate. AI explains. You decide.
Where we started
Drug-induced liver injury has been the single most frequent cause of safety-related drug withdrawals for the last fifty years.
FDA, Drug-Induced Liver Injury guidance
Fifty years. The signals were in the data.
Hepatic safety is our first playbook, not the product. We started with the highest-consequence investigation we could pick. The same failure repeats in renal, cardiac and hematologic safety, in adverse event and concomitant medication review, and in operational and data quality review: the evidence is in the data before anyone recognizes what it means.
What inconsistent investigation costs
Every sponsor has as many investigation methods as it has reviewers.
Three reviewers or three hundred, the arithmetic is the same. One signal gets a different depth of investigation, and a different conclusion, depending on who picked it up. That is not anyone's fault. The method was never written down.
Investigations vary
Quality depends on reviewer habits. Methods are rebuilt from scratch each time, and traceability is often incomplete.
The manual burden
Reviewers spend significant time reconstructing context around every alert. Evidence is fragmented across systems, and the work is close to impossible to audit.
The amendments
Cross-domain patterns accumulate undetected until they force a mid-study correction. 76 percent of trials require an amendment, at $141K to $535K each.
Tufts CSDD. Figures are pre-inflation.
The inspection finding
Regulators scrutinize how a signal was investigated, not only what was found.
A signal that was flagged but poorly followed up is worse than one that was never flagged at all.
The gap
Every tool in a trial today looks for what someone already knew to look for.
Between them is what the industry covers today.
What nothing looks for
Unknown unknowns.
Patterns no one predefined, because they only emerge when several signals interact across different domains at once, none of them conclusive on its own.
A faint liver lab trend.
An unusual dosing gap.
One site's specimen handling quirk.
Each, easy to dismiss alone. Consistent together, across participants. Found late, by accident, or not at all, and that is where the costly safety surprises actually live.
And every tool evaluates a signal as a single reading, not as a trajectory. An alert tells you a value crossed a line today. It does not tell you whether that value has been climbing for six weeks or spiked once and settled, or whether it moved before the dose changed or after.
The reading is what raises the question. The trajectory is what answers it.
Today's tools tell you what changed. Reveal.IQ is the investigation that follows.
What it does
Four things, in order.
Detect
Data quality rules, KRIs, QTLs and medical alerts run consistently and auditably across every study, on one harmonized data layer. This is the part the industry already does; Reveal.IQ does it reproducibly, so everything after it starts from the same floor. It can also run on the rules and alerts already in your systems.
Discover
Reveal.IQ surfaces patterns no rule was written to catch. Statistical analysis runs across clinical and operational data together, not one domain at a time, which is what makes a pattern visible that no single domain would have raised. Findings are statistically validated before anyone is asked to look at them.
Investigate
Every detected and discovered signal gets investigated, not only the ones someone had time for. A playbook loads the study context, evaluates the competing explanations, and produces a result with a complete evidence package.
Explain
A study insights report, and a chat interface to interrogate the finding and its evidence, both strictly bounded to the facts in the package.
AI explains governed results. It does not generate them. Every finding traces to the evidence it came from, and the reasoning is the method your experts approved, not a path a model chose that morning.
What earlier understanding changes
Protect participants
Surface developing safety patterns for earlier, better-informed medical review.
Improve execution
Recognize cross-domain issues affecting adherence, quality, sites and protocol compliance.
Reduce disruption
Support corrective action before issues contribute to delay, rework, or costly study changes.
Preserve expert capacity
Reuse expert-designed methods instead of repeatedly reconstructing evidence by hand.
Any number of reviewers. One method. That is the whole difference.
Not a set of features.
One idea.
Whether a signal is found at all, and the quality of the investigation that follows, should not depend on who happens to be looking.
We call it Reveal.IQ.
Who it is for
Decided by clinical leadership. Used by the people who investigate. Inherited by everyone who has to defend the call later.
The decision maker
Head of Clinical Operations or Clinical Development. A clinical leadership decision, not an IT purchase, owned by the leaders accountable for study quality and speed.
The daily users
Medical monitors, clinical scientists, data managers, biostatisticians and study monitors: the people investigating signals today across spreadsheets and siloed tools, who sign off still wondering what the data has not shown them.
The beneficiaries
Sponsors, CROs, sites, regulators and patients. One evidence trail, in one format, defensible at inspection, held to the same standard for every signal.
Behind every one of them is a patient who is still being dosed while the question is open.
Where this goes
The playbook model scales in two directions.
The same governed architecture runs medical, operational, data quality and efficacy investigations. Hepatic safety is where we proved it, not the boundary of it.
Expand by domain
- Safety signals
- Clinical operations
- Data integrity and quality
- Efficacy signals
Scale by reach
- Study
- Program
- Enterprise
- Therapeutic area
- Industry
The playbook library compounds
Anyone can demo AI. Nobody can shortcut a governed investigation library built study by study, approved by the experts who own the review. The long ambition is a shared standard for investigation logic, the way CDISC became a shared standard for clinical data.
Why this matters
Real-time clinical development means deciding in real time. Discovery and investigation have to keep pace, or the decision is just faster guesswork.
- Detection scales. Investigation does not.
- Every signal deserves the same rigorous, governed investigation. Most never get one.
- The missing capability is explaining what the evidence supports, what it weakens, and what it cannot yet answer.
Shared methods. Sponsor-controlled data. Traceable decisions.
Why now
On April 28, 2026 the FDA announced major steps toward real-time clinical trials, with proof-of-concept trials already running at AstraZeneca and Amgen. As visibility accelerates, the bottleneck moves decisively from seeing signals to investigating them responsibly.
What has never existed is a safety intelligence watching a drug class during active development. That is the open white space.
Why us
Sixty combined years building the systems this problem outlived.
We built the first generation of clinical data infrastructure and watched the same investigation gap survive every technical upgrade. Every person on our leadership team has either been a patient or cared for one, which is why a delayed investigation is not an abstraction to us.