Loading...
Every Burna grade ships with the source sentence, the CTCAE criterion, and the WHO-UMC or Kramer attribution attached. EDC write-back to Medidata Rave, Veeva Vault EDC, and Oracle Clinical One on day one. Confirmation queries close before they open. Database lock arrives ahead of timeline.
SMART on FHIR inside Epic, Oracle Health (Cerner), and regional EHRs. Sanctioned EDC API surfaces only, no credentials stored. HIPAA aligned. 21 CFR Part 11 aligned. SOC 2 Type II audit kicked off Q2 2026. BAA on day one.
Per query, oncology. Industry benchmark per resolved query (Oracle ClearTrial, 2021), against a $44 cross-trial average.
Are confirmation requests. Pronker et al., British Journal of Clinical Pharmacology, 2011. 924 queries across three Phase 1 oncology trials. Structurally closeable.
Annual portfolio query labor. Conservative to stretch CRO oncology portfolios at 100 to 300 active trials. Approximately 85% closeable.
After thirty years in clinical trials operations, the queries are still the work. The grade is downstream; the rationale should be in the record at the moment of grading, not reconstructed weeks later when a monitor asks. Citation-bound grading is the structural change.
Scott SchliebnerCOO at P1 Pratia Oncology; formerly Global Drug Development Consulting at Novotech; ex-PRA Health Sciences
Most adverse event grades in oncology trials are reviewed by sponsor monitors and CRO data managers. Most of those reviews end in a query. The query is the bottleneck. The query is the work.
Composed of coordinator response time, principal investigator review, CRA verification, data manager routing, and sponsor monitor closure. Against a $44 average across all therapeutic areas.
Oracle ClearTrial industry benchmark, 2021.
71.9% produce no data change at all. The monitor asks the site to confirm what the site already documented.
Pronker et al., British Journal of Clinical Pharmacology, 2011. 924 queries across three Phase 1 oncology trials.
21% of queries require more than one round. Database lock slips at the pace of correspondence, not the pace of the science.
Tolmie et al., Trials, 2011. 1,595 endpoint packages across a multinational adjudication study.
Most queries today produce no data change. They confirm what was already documented. The labor cost is real. The data quality lift is not.
Phase 1 oncology trials carry the highest AE volume per patient, the densest documentation, the most frequent dose modifications, and the most monitoring intensity in any oncology phase.
Safety is the primary endpoint. Forty percent of all queries in a Phase 1 oncology trial land on adverse event and safety data (Pronker et al., 2011). Each AE field requires defensible rationale at the moment of entry, or it becomes a query downstream.
The work honors itself. Phase 1 oncology monitoring teams are the most experienced operators in clinical research. The grading is intricate, the regimens are layered, the timelines are unforgiving. Citation-bound architecture meets that work at the same standard: every grade arrives at the monitor’s desk with the source sentence quoted, the CTCAE criterion named, and the per-drug attribution path visible. The monitor’s expertise reviews the rationale, not the absence of one.
The Phase 1 oncology trial is where citation-bound architecture is most operationally consequential. It is also where the per-trial economics are the largest.
Three structural changes the monitoring team feels by week two of a pilot.
See it on a real adverse event grade →First-read closure
The source clinical sentence and the CTCAE criterion are inside the audit envelope when the grade is signed. The monitor opens the record, finds the rationale already attached, and closes it. The query the monitor would have asked is answered before it gets opened.
Single-round
When the rationale is missing from the original submission, monitors come back with a second, third, or fourth question. Citation-attached grades end the loop on the first review. Twenty-one percent of queries that previously took two or three rounds resolve on the first response.
Database lock
Median 23-day query resolution drops to days when the answer ships with the question’s source data. Database lock arrives ahead of timeline. The trial moves forward at the pace of the science, not the pace of correspondence.
The per-trial labor anchor is $192K to $710K. At portfolio scale, the math compounds across active trials and across the calendar year. Three scenarios, calibrated to oncology-leaning CRO portfolios in the top 100.
$192K per-trial labor anchor. Approximately $16.5M closeable structurally with citation-bound architecture.
$450K per-trial labor anchor (midpoint). Approximately $77M closeable structurally.
$710K per-trial labor anchor. Approximately $183M closeable structurally.
The numbers are projections from peer-reviewed and industry-benchmark inputs. The order of magnitude holds across oncology-leaning portfolios in the top 100 CROs. The Query Economics whitepaper walks the methodology.
Burna’s grading engine is twelve specialized agents in a patented cascading constraint pipeline. Each agent’s output bounds the valid output space of every downstream agent. The system structurally cannot produce a grade that contradicts its own prior findings, cannot skip citation, and cannot return a grade outside the defined CTCAE set. Named attribution algorithms (WHO-UMC and Kramer) with per-drug probability scores. Visible math. Two patents filed.
Read the full engine architecture →AI suggests, clinicians decide. The clinician is the signatory of record on every grade.
Multi-tenant SaaS under BAA. Fastest deployment. Standard for design partners and community oncology sites. PHI processed under BAA; data residency configurable.
Single-tenant inside your VPC or customer-controlled cloud. PHI stays in your boundary. Used for CROs and large sponsors with specific data residency requirements. Architecture and audit trail format identical to cloud deployment; only the hosting changes.
Isolated AI environment deployed inside the sponsor’s own cloud infrastructure. Per-trial agreements for confidential Phase 1 protocols where the protocol IP cannot leave sponsor systems. Only structured probability scores leave the sponsor’s boundary.
Sanctioned API surfaces only. No EHR or EDC credentials stored on Burna infrastructure. Multi-EHR support standard. EDC write-back on day one.
Aria deployments use a headless browser running in the coordinator’s authenticated session. No PHI passes through Burna infrastructure. No credentials stored.
Burna operates upstream of where downstream PV case management can help. The differentiation is structural, not feature-level.
| Capability | Burna | Veeva · Argus · ArisGlobal | Offshore vendors | Internal CRO build |
|---|---|---|---|---|
| Helps investigators make the attribution decision | Yes | No | No | Variable |
| Named WHO-UMC + Kramer attribution | Yes | Limited | No | Variable |
| Multi-drug oncology regimen profiles (42+) | Yes | No | No | No |
| Citation-bound CTCAE grading | Yes | Partial | No | Variable |
| Real-time cross-site consistency | Yes | No | No | No |
| Sponsor tenant isolation | Yes | N/A | No | Variable |
| Peer-reviewed validation path | Yes | Vendor-claim | No | No |
| Time to deployment | 12 weeks | 18 to 36 months | Variable | 24 to 48 months |
The market math: a CRO that builds internally is paying $5M to $20M over 24 to 48 months and arriving at parity with where Burna ships in week 12. The math does not support an internal build for any CRO outside the top-3 globally.
A CRO partnership is structurally interesting because the partnership reaches both sides of the market.
Citation-bound grading and cross-site consistency arrive at every site running the trial.
Citation-bound grading and cross-site consistency arrive at every site running the trial. The CRO inherits the audit trail discipline at every monitor desk.
Coordinators and PIs experience the architectural close on queries directly. Their grading workflow compresses, their query burden drops, the monitoring rhythm changes.
Once a site has lived inside the architecture for a CRO-led trial, the procurement conversation for investigator-initiated and other sponsor trials becomes much shorter.
Sponsors notice the cleaner safety data on the CRO trial. Pull from sponsors creates Protocol Safe conversations across the sponsor’s full portfolio.
Three revenue streams, one platform. Cancer centers buy SaaS directly. CROs deploy for operational efficiency. Pharma sponsors pull for data quality and postmarket pharmacovigilance. The CRO partnership is the distribution flywheel.
The standard CRO motion is per-trial pricing. The top-6 CRO motion is enterprise white-label, scoped separately below.
$25K to $50K, scaled to trial size. First 30 days at no charge to remove procurement friction. Days 31 to 90 paid after baseline query metrics confirm reduction. Conversion to a 12-month annual contract at $75K to $150K per trial when pilot success metrics are met.
Volume pricing available for CROs deploying across multiple concurrent trials. Standard pricing applies for the first three trials; volume pricing kicks in at trial four.
Burna offers white-label enterprise deployment for top-6 CROs and select mid-major CROs. Sponsors and sites see the CRO. Burna operates underneath.
Per-sponsor
Each sponsor engagement runs in a logically isolated tenant with dedicated encryption keys. Cross-sponsor data leakage is architecturally prevented.
Citation-bound
Every grade carries its source sentence and CTCAE criterion. The CRO inherits Burna’s defensibility, which means CRO monitors and CRO medical reviewers carry the same audit trail discipline into every sponsor conversation.
Real-time
A multi-site trial run by the CRO surfaces grading inconsistency and attribution drift in real time, while the trial is running.
Annually. Single-region delivery footprint, up to 25 active sponsor engagements.
Annually. Multi-region delivery footprint, up to 100 active sponsor engagements.
Annually. Global delivery footprint, unlimited sponsor engagements, named tenant architecture, custom validation work.
The first three top-10 CRO white-label partnerships include first-mover commercial economics. Active conversations underway with three of the top-10 global CROs.
Burna does not run on vendor claims. The platform is reviewed, validated, and co-authored with the people who carry the operational and regulatory weight.
“After thirty years in clinical trials operations, the queries are still the work. The grade is downstream; the rationale should be in the record at the moment of grading, not reconstructed weeks later when a monitor asks. Citation-bound grading is the structural change.”
Scott SchliebnerCOO at P1 Pratia Oncology; formerly Global Drug Development Consulting at Novotech; ex-PRA Health Sciences
June 2026 cohort. 1,200+ patient charts. Co-authored with Dr. Andrea Pirzkall, Oncology Clinical Development Consultant at Precision For Medicine (formerly CMO at Replimune and Asher Biotherapeutics; ED Clinical Development at BeiGene; Principal Medical Director at Genentech).
Member of the HHS Cancer Moonshot public-private partnership. FDA EDSTP application planned for June 2026.
Spanning clinical oncology, pharma development, federal health IT, regulatory strategy, and AI infrastructure. Commercial Council includes Scott Schliebner. Regulatory & Standards Council co-led by Dr. Howard Fingert.
What Monday looks like for your monitoring team.
Monday morning at the CRO. The lead CRA on a Phase 1 oncology trial opens her queue. Twenty-two adverse event grades posted from sites over the weekend. Burna has shipped each grade inside an audit envelope: source sentence quoted from the clinical note, CTCAE criterion attached, WHO-UMC or Kramer attribution path visible. She reviews twelve in under twenty minutes and clears them. She flags one for clarification; the system already shows her the prior cycles, the regimen, and the comorbidity context her question would have requested. The follow-up resolves on the first response.
The data manager dashboard shows confirmation query volume on the trial down meaningfully from baseline. Multi-round queries on adverse event forms are approaching zero. The sponsor’s PV team has not opened a query in eight days. The Director of Data Management for the program sees query closure cycle compressed from a 23-day median to under five.
The COO’s Monday dashboard shows the same compression across the oncology portfolio. Database lock for the lead Phase 1 trial is ahead of timeline. The sponsor RFP that arrived last week now has a citation-bound architecture differentiator the proposal team can write to. The senior CRA who was carrying three trials at saturation now has bandwidth for the fourth.
The trial moves forward at the pace of the science, not the pace of correspondence.
The ten we hear most often from CRO COOs and procurement leads.
Read the full FAQ →The pre-registered pilot endpoint targets a 50% reduction in confirmation queries on adverse event forms. At industry-standard rates of $192K to $710K per trial in query labor, with 85.7% of those queries being confirmation requests (Pronker et al., 2011), the structurally closeable labor per trial is meaningful at portfolio scale. The exact figure depends on your current baseline.
Constraint architecture, not statistical guarantee. The output space is restricted to valid CTCAE grades. The engine cannot return a grade without a source citation. Constraint enforcement is structural, not a transparency layer added after generation. Two patents filed.
Fail-closed. Low-confidence outputs are surfaced to the clinician with the constraint failure flagged. The system does not produce an ambiguous grade.
Veeva Vault Safety AI Agents (April 2026 launch) handle ICSR lifecycle management downstream of the attribution decision. Burna operates upstream of the attribution decision, helping the investigator make the original grading and attribution call at the point of clinical encounter. The two are complementary in a sponsor PV stack. Burna integrates with Veeva Vault EDC.
Yes. Headless browser deployed in the coordinator’s authenticated session. No PHI passes through Burna infrastructure. The Aria integration is one of the integration patterns in production.
Standard CRO IT security review takes 4 to 8 weeks. Burna has worked with timelines from 4 to 16 weeks. Deployment phasing accommodates extended review without breaking the integration architecture.
No. Customer data is processed for inference only. No model training on customer data without explicit written consent.
90-day paid pilots at $25K to $50K, scaled to trial size. First 30 days at no charge to remove procurement friction. Days 31 to 90 paid after baseline query metrics confirm reduction. Conversion to a 12-month annual contract at $75K to $150K per trial when pilot success metrics are met. Top-6 CRO white-label engagements scoped separately at enterprise pricing.
Burna deploys inside the CRO’s branded customer-facing environment. Sponsors and sites see the CRO; Burna operates underneath. Tenant isolation per sponsor with dedicated encryption keys. Real-time cross-site consistency. The architectural commitments are the same as in standard CRO deployments; the surface is white-labeled.
By default, white-label deployments surface the CRO’s brand. A small "Powered by Burna" attribution can be added at the CRO’s option. Decision lives with the CRO.
A 15-minute walkthrough covers the audit envelope on a real adverse event grade, the integration pattern for your EHR and EDC stack, and the query economics calculator built on industry benchmarks. If the math holds for your portfolio, Burna scopes a 90-day paid pilot on a single oncology trial. Clean exit if it does not convert.
The 2026 design partner cohort accepts pilot scoping conversations through the end of Q3 2026. Top-6 CRO white-label conversations are scoped on a rolling basis.
Scope, endpoints, and integration timeline are defined before kickoff. The Master Service Agreement and Business Associate Agreement are pre-negotiated single-document downloads.
For procurement and integration leads tracking the category. Subscribe to the quarterly safety and data quality briefing. Validation readouts, design partner cohort updates, and conference timing only.
No tracking beacons, no marketing email cadence.