GalenOps prod · 0.1.0

Agent

Recruitment funnel and screening plan

Horizon 2 recruitment_agent

Enrollment funnel, accrual pace and eligibility yield computed; candidates to screen proposed per site (import is human-gated).

Escalates: Protocol-matched candidates, batched for import; nobody is contacted or enrolled without a coordinator's approval — to the study coordinator.

Agents

Runs

3

Escalations

1

Failures

0

Avg duration

4.4 s

Literature & benchmarks

  • Screen-failure funnel — referred → pre-screened → consented → eligible, with the eligibility pass fraction as the tracked indicator
  • Eligibility pass fraction applied against the study's target enrollment and enrollment window

Run history (3 across all studies)

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

Expand When Study Outcome Model Duration Reasoning
2026-09-29 23:42 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 completed claude-sonnet-5 4.5 s This is a completed Phase II sarcoma trial reviewed retrospectively. The funnel shows 140 entered screening, 14 screened-only, 126 accrued, 106 completed, 20 withdrawn, with a screen-fail rate of 10.0% (14/140). Pace ratio is 1.20 (126 accrued vs 104.9 expected pro-rata), so enrollment exceeded the plan; projected last-patient-in is not applicable since the target of 105 was already reached. Eligibility yield across the 8 detected protocol criteria is Y=0.2732, well above the Aldox feasibility memo's "narrow funnel" threshold of Y<0.01, so narrow_funnel is marked false. Per-site consents ranged widely (e.g., Sarcoma Oncology Center 16, several sites at 1), with 11 sites below median accrual of 2.0. No screening plan was proposed and review_opened is false, so no human decision is required here.
2026-09-29 21:13 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 completed claude-sonnet-5 3.9 s Overall enrollment is on pace: pace ratio 1.00 (317/450 accrued vs ~317.1 expected), with projected last-patient-in 2027-04-13 at the trailing 3-month rate of 20.6 consents/month, against a plan of 23.68/month. All 15/15 sites are open, median site accrual is 21.0, with 6 sites below median (SITE006 GBR, SITE007 JPN, SITE008 KOR, SITE009 AUS, SITE010 ESP, SITE014 NLD) and 0 sites activated over 60 days with zero consents. Eligibility yield Y=0.4850 from 2 detected criteria, well above the narrow-funnel threshold of 0.01 (Aldox feasibility memo), so narrow_funnel is false. The screening plan calls for 60 candidates across 15 open sites (partner:ehr-matching estimate), unchanged/updated this run, sized from remaining target 133 ÷ Y 0.485 ÷ 5.61 months remaining. A review task is pending: the human must decide whether to approve import of this screening roster.
2026-09-19 08:58 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 4.9 s Enrollment overall looks on track: pace ratio is 1.02 versus the pro-rata plan (317 of 450 accrued vs 309.3 expected to date), with trailing 3-month pace at 22.97/month and a projected last-patient-in date of 2027-03-14. Eligibility yield Y is 0.4850 from 2 protocol criteria, well above the "narrow funnel" threshold of 0.01 (Aldox feasibility memo), so narrow_funnel is false. Sites: 15/15 open, median accrual 21.0, with 6 sites (SITE006 GBR, SITE007 JPN, SITE008 KOR, SITE009 AUS, SITE010 ESP, SITE014 NLD) below median, though none idle over 60 days with zero consents. The screening plan proposes 60 candidates (4 per site, no caps hit) via partner:ehr-matching. Because import of matched candidates is human-gated, a review task was opened: the human must decide whether to approve importing these 60 candidate records into the screening funnel.