Agent

TLFs generated & validated

Horizon 1 biostats_agent

ICH E3 Section 14 tables from live data (disposition, demographics, exposure, AEs, Kaplan-Meier efficacy), pending double-programming validation.

Escalates: Every generated TLF batch, for double-programming-equivalent validation before release — to the validating biostatistician.

Agents

Runs

Escalations

Failures

Avg duration

4.0 s

Literature & benchmarks

ICH E3 §14 — the standard tables, listings and figures set
Kaplan–Meier estimates for time-to-event endpoints
Double-programming gate — nothing is released without independent validation

Run history (3 across all studies)

Newest first, 50 per page. Expand a row for the recorded reasoning and evidence.

When

Study

Outcome

Model

Duration

Reasoning

2026-09-29 23:43 UTC

A Multicenter, Randomized, Open-Label Phase 2b Study to Investigate the Preliminary Efficacy and Safety of INNO-206 (Doxorubicin-EMCH) Compared to Doxorubicin in Subjects With Metastatic, Locally Advanced, or Unresectable Soft Tissue Sarcoma

escalated human

claude-sonnet-5

4.5 s

12 TLF outputs were generated per ICH E3 Section 14 for this Phase II soft-tissue sarcoma study (126 randomised, 140 screened). Safety shows a clear differential: any AE 80.0% (Doxorubicin 75 mg/m2) vs 95.2% (Aldoxorubicin 350 mg/m2), grade ≥3 57.5% vs 79.5%, serious AEs 20.0% vs 39.8%, and 66 total deaths (26 vs 40). Efficacy favors Aldoxorubicin: OS median 439 vs 479 days (log-rank p=0.3535, not significant), PFSINV 133 vs 254 days (p=0.0003), PFSIRC 87 vs 168 days (p=0.0298), and ORR 0.0% vs 25.0% (BOR) and 5.0% vs 22.9% (BORINV). The batch (hash 8b11fd774ddf26c5) is parked for double-programming validation, and review_opened is true. A human reviewer is asked to decide whether the safety and efficacy TLF outputs, given these AE and survival differences, should be accepted following double-programming validation.

Input: A Multicenter, Randomized, Open-Label Phase 2b Study to Investigate the Preliminary Efficacy and Safety of INNO-206 (Doxorubicin-EMCH) Compared to Doxorubicin in Subjects With Metastatic, Locally Advanced, or Unresectable Soft Tissue Sarcoma — ae_overview=%{"Aldoxorubicin (INNO-206) 350 mg/m2" => %{

Evidence

Blinded: no
Citation: ICH E3 Structure and Content of Clinical Study Reports, Section 14 (14.1 disposition and demographics, 14.2 efficacy, 14.3 safety)
Deterministic summary: Generated 12 TLF outputs (ICH E3 Section 14). 126 randomised subjects in 2 arm(s) (Doxorubicin 75 mg/m2 40, Aldoxorubicin (INNO-206) 350 mg/m2 86); 140 screened. Any AE 80.0% / 95.2%, grade >=3 57.5% / 79.5%, serious 20.0% / 39.8%; deaths 66. OS median 439 days (14.4 mo) vs 479 days (15.7 mo), log-rank p=0.3535. PFSINV median 133 days (4.4 mo) vs 254 days (8.3 mo), log-rank p=0.0003. PFSIRC median 87 days (2.9 mo) vs 168 days (5.5 mo), log-rank p=0.0298. ORR 0.0% vs 25.0%. ORR 5.0% vs 22.9%. Batch 8b11fd774ddf26c5 parked for double-programming validation.
Documents: 12
Efficacy included: yes
Hash: 8b11fd774ddf26c5
Review opened: yes
Screen failures: 14
Screened: 140
Superseded batch: —
Unchanged: no

Ae overview

Aldoxorubicin (inno-206) 350 mg/m2

Any ae pct: 95.20
Deaths: 40
Grade3 pct: 79.50
Related pct: 90.40
Safety n: 83
Serious pct: 39.80

Doxorubicin 75 mg/m2

Any ae pct: 80.00
Deaths: 26
Grade3 pct: 57.50
Related pct: 72.50
Safety n: 40
Serious pct: 20.00

Total

Any ae pct: 90.20
Deaths: 66
Grade3 pct: 72.40
Related pct: 84.60
Safety n: 123
Serious pct: 33.30

Arms

Doxorubicin 75 mg/m2
Aldoxorubicin (INNO-206) 350 mg/m2

Unit: days

Events

Aldoxorubicin (inno-206) 350 mg/m2: 40
Doxorubicin 75 mg/m2: 26

Median

Aldoxorubicin (inno-206) 350 mg/m2: 479.00
Doxorubicin 75 mg/m2: 439.00

Median ci

Aldoxorubicin (inno-206) 350 mg/m2

391.00
—

Doxorubicin 75 mg/m2

311.00
664.00

Aldoxorubicin (inno-206) 350 mg/m2: 86
Doxorubicin 75 mg/m2: 40

Pfsinv

Unit: days

Events

Aldoxorubicin (inno-206) 350 mg/m2: 67
Doxorubicin 75 mg/m2: 39

Median

Aldoxorubicin (inno-206) 350 mg/m2: 254.00
Doxorubicin 75 mg/m2: 133.00

Median ci

Aldoxorubicin (inno-206) 350 mg/m2

211.00
295.00

Doxorubicin 75 mg/m2

88.00
171.00

Aldoxorubicin (inno-206) 350 mg/m2: 86
Doxorubicin 75 mg/m2: 40

Pfsirc

Unit: days

Events

Aldoxorubicin (inno-206) 350 mg/m2: 74
Doxorubicin 75 mg/m2: 40

Median

Aldoxorubicin (inno-206) 350 mg/m2: 168.00
Doxorubicin 75 mg/m2: 87.00

Median ci

Aldoxorubicin (inno-206) 350 mg/m2

87.00
208.00

Doxorubicin 75 mg/m2

85.00
127.00

Aldoxorubicin (inno-206) 350 mg/m2: 86
Doxorubicin 75 mg/m2: 40

Log rank

Overall

Chi square: 0.86
Df: 1
P value: 0.35

Pairwise p

Aldoxorubicin (inno-206) 350 mg/m2: 0.35

Pfsinv

Overall

Chi square: 12.86
Df: 1
P value: 0.00

Pairwise p

Aldoxorubicin (inno-206) 350 mg/m2: 0.00

Pfsirc

Overall

Chi square: 4.72
Df: 1
P value: 0.03

Pairwise p

Aldoxorubicin (inno-206) 350 mg/m2: 0.03

Orr

Bor

Aldoxorubicin (inno-206) 350 mg/m2

N: 80
Orr pct: 25.00
Responders: 20

Ci pct

16.80
35.50

Doxorubicin 75 mg/m2

N: 38
Orr pct: 0.00
Responders: 0

Ci pct

0.00
9.20

Borinv

Aldoxorubicin (inno-206) 350 mg/m2

N: 83
Orr pct: 22.90
Responders: 19

Ci pct

15.20
33.00

Doxorubicin 75 mg/m2

N: 40
Orr pct: 5.00
Responders: 2

Ci pct

1.40
16.50

Population

Aldoxorubicin (inno-206) 350 mg/m2: 86
Doxorubicin 75 mg/m2: 40
Total: 126

Safety population

Aldoxorubicin (inno-206) 350 mg/m2: 83
Doxorubicin 75 mg/m2: 40
Total: 123

Tables

Number	Rows	Title
14.1.1	11	Table 14.1.1 Subject Disposition
14.1.2	15	Table 14.1.2 Demographic and Baseline Characteristics
14.1.3	9	Table 14.1.3 Extent of Exposure to Study Drug
14.3.1	15	Table 14.3.1 Overview of Adverse Events
14.3.2	37	Table 14.3.2 Adverse Events by System Organ Class and Preferred Term
14.3.3	42	Table 14.3.3 Serious Adverse Events by Preferred Term
14.3.4	66	Listing 14.3.4 Deaths
14.2.1	9	Table 14.2.1 Overall survival (OS) - Kaplan-Meier Analysis
14.2.2	9	Table 14.2.2 Progression-free survival, investigator assessment (RECIST 1.1) (PFSINV) - Kaplan-Meier Analysis
14.2.3	9	Table 14.2.3 Progression-free survival, independent central review (RECIST 1.1) (PFSIRC) - Kaplan-Meier Analysis
14.2.4	7	Table 14.2.4 Best overall response, independent central review (RECIST 1.1) (BOR) - Response Rates
14.2.5	8	Table 14.2.5 Best overall response, investigator assessment (RECIST 1.1) (BORINV) - Response Rates

Open study

2026-09-29 21:14 UTC

A Phase III, Randomized, Double-Blind Study of Torivumab Plus Chemotherapy Versus Placebo Plus Chemotherapy in Previously Untreated Advanced Non-Small Cell Lung Cancer

escalated human

claude-sonnet-5

3.4 s

This TLF batch (hash 476ab4f205a31270, 7 tables) is unchanged since the last run and is awaiting double-programming validation, per the deterministic summary. The study is blinded and active, with 317 randomised subjects reported pooled (317 screened, 0 screen failures) and a safety population of 307. Pooled safety data show any AE 84.4%, grade ≥3 41.4%, serious 26.4%, related 30.0%, and 47 deaths, per the ae_overview evidence. By-arm and efficacy outputs (KM, log-rank, ORR) are suppressed until database lock/unblinding, consistent with efficacy_included:false. Tables follow ICH E3 Section 14 structure (14.1 disposition/demographics, 14.2 efficacy, 14.3 safety). A review task has been opened. A human reviewer is asked to decide whether to proceed with double-programming validation and sign-off of this unchanged TLF batch.

Input: A Phase III, Randomized, Double-Blind Study of Torivumab Plus Chemotherapy Versus Placebo Plus Chemotherapy in Previously Untreated Advanced Non-Small Cell Lung Cancer — ae_overview=%{"All patients (blinded)" => %{"any_ae_pct" => 84.4, "deaths" => 47, "grade3_pct" => 41.4, "related_pct" => 30.0, "sa

Evidence

Blinded: yes
Citation: ICH E3 Structure and Content of Clinical Study Reports, Section 14 (14.1 disposition and demographics, 14.2 efficacy, 14.3 safety)
Deterministic summary: TLF batch 476ab4f205a31270 unchanged since the last run (7 tables); awaiting double-programming validation. Blinded study in progress: 317 randomised subjects reported pooled (317 screened). Any AE 84.4%, grade >=3 41.4%, serious 26.4%; deaths 47. By-arm and efficacy outputs suppressed until database lock/unblinding.
Documents: 7
Efficacy included: no
Hash: 476ab4f205a31270
Review opened: yes
Screen failures: 0
Screened: 317
Unchanged: yes

Ae overview

All patients (blinded)

Any ae pct: 84.40
Deaths: 47
Grade3 pct: 41.40
Related pct: 30.00
Safety n: 307
Serious pct: 26.40

Arms

Placebo + Chemotherapy
Torivumab + Chemotherapy

No evidence recorded for this run.

Log rank

No evidence recorded for this run.

Orr

No evidence recorded for this run.

Population

All patients (blinded): 317

Safety population

All patients (blinded): 307

Tables

Number	Rows	Title
14.1.1	8	Table 14.1.1 Subject Disposition
14.1.2	19	Table 14.1.2 Demographic and Baseline Characteristics
14.1.3	9	Table 14.1.3 Extent of Exposure to Study Drug
14.3.1	15	Table 14.3.1 Overview of Adverse Events
14.3.2	21	Table 14.3.2 Adverse Events by System Organ Class and Preferred Term
14.3.3	23	Table 14.3.3 Serious Adverse Events by Preferred Term
14.3.4	47	Listing 14.3.4 Deaths

Open study

2026-09-19 08:59 UTC

A Phase III, Randomized, Double-Blind Study of Torivumab Plus Chemotherapy Versus Placebo Plus Chemotherapy in Previously Untreated Advanced Non-Small Cell Lung Cancer

escalated human

claude-sonnet-5

4.3 s

This is a blinded interim TLF run for this Phase III NSCLC study, per ICH E3 Section 14 (14.1 disposition/demographics, 14.2 efficacy, 14.3 safety). 7 tables/listings were generated covering disposition (14.1.1, 8 rows), demographics (14.1.2, 19 rows), exposure (14.1.3, 9 rows), AE overview (14.3.1, 15 rows), AEs by SOC/PT (14.3.2, 21 rows), SAEs (14.3.3, 23 rows), and a deaths listing (14.3.4, 47 rows). Pooled across 317 screened/randomised subjects (307 in safety population), any AE occurred in 84.4%, grade ≥3 in 41.4%, serious in 26.4%, with 47 deaths reported. Because the study remains blinded, by-arm and efficacy (KM, log-rank, ORR) outputs are suppressed until database lock/unblinding. Batch 476ab4f205a31270 is parked for double-programming validation, and a review task has been opened. A human should decide whether to release this batch for double-programming sign-off.

Evidence

Blinded: yes
Citation: ICH E3 Structure and Content of Clinical Study Reports, Section 14 (14.1 disposition and demographics, 14.2 efficacy, 14.3 safety)
Deterministic summary: Generated 7 TLF outputs (ICH E3 Section 14). Blinded study in progress: 317 randomised subjects reported pooled (317 screened). Any AE 84.4%, grade >=3 41.4%, serious 26.4%; deaths 47. By-arm and efficacy outputs suppressed until database lock/unblinding. Batch 476ab4f205a31270 parked for double-programming validation.
Documents: 7
Efficacy included: no
Hash: 476ab4f205a31270
Review opened: yes
Screen failures: 0
Screened: 317
Superseded batch: —
Unchanged: no

Ae overview

All patients (blinded)

Any ae pct: 84.40
Deaths: 47
Grade3 pct: 41.40
Related pct: 30.00
Safety n: 307
Serious pct: 26.40

Arms

Placebo + Chemotherapy
Torivumab + Chemotherapy

No evidence recorded for this run.

Log rank

No evidence recorded for this run.

Orr

No evidence recorded for this run.

Population

All patients (blinded): 317

Safety population

All patients (blinded): 307

Tables

Number	Rows	Title
14.1.1	8	Table 14.1.1 Subject Disposition
14.1.2	19	Table 14.1.2 Demographic and Baseline Characteristics
14.1.3	9	Table 14.1.3 Extent of Exposure to Study Drug
14.3.1	15	Table 14.3.1 Overview of Adverse Events
14.3.2	21	Table 14.3.2 Adverse Events by System Organ Class and Preferred Term
14.3.3	23	Table 14.3.3 Serious Adverse Events by Preferred Term
14.3.4	47	Listing 14.3.4 Deaths

Open study