{"generated_at":"2026-05-28T12:03:18.098Z","verdicts":[{"id":"biomarker-quality-nsclc-2026-05","signal":"Biomarker quality","context":"NSCLC · Phase II/III","claim":"Genomic-grade biomarker selection raises a program's probability of success.","verdict":"validated","shipped":true,"shippedNote":"Scored in the production engine (2.6.0) for oncology solid tumor at Phase II/III.","effect":"+5.2pp held-out AUC over the structural baseline, stable across cohort sizes. The genomic_validated cohort odds ratio was 5.59 vs the Schwaederle (2016) literature anchor of 1.35 — we ship the lower anchor with the overshoot disclosed.","method":"70/30 stratified held-out backtest, N=85 NSCLC (approvals + failures), paired DeLong.","decision":"Scored: genomic_validated 1.35x / protein_only 0.85x / unselected 1.00x (log-odds, Phase II/III).","evidence":{"json":"/evaluations/verdicts/biomarker-quality-nsclc-2026-05.json"},"related":["/research/backtest-nsclc","/methodology/pos-calibration"],"provenance":{"engine_version":"2.6.0","methodology_version":"methodology@2026-05-28"},"dataset_version":"drug-specific-signals@2026-Q3","as_of":"2026-05-28","status":"current"},{"id":"phase1-orr-oncology-solid-2026-05","signal":"Phase 1 objective response rate (ORR magnitude)","context":"Oncology solid tumor · Phase II/III","claim":"A strong early-phase tumor response rate predicts later-stage success (a signal several incumbents market).","verdict":"not_predictive","shipped":false,"shippedNote":"Never scored in the engine. Tested as a candidate; surfaced as a non-scored informational flag only.","effect":"A joint biomarker x ORR-bucket model beat the biomarker-only baseline by only +0.5pp held-out AUC (paired DeLong p=0.48), and the comparison is statistically unpowerable — detecting a +3pp gain at this baseline needs ~830 drugs; the cohort is 85. This followed the Phase 1 finding that the two signals combined fell below baseline (-0.3pp at 43-drug coverage) — the double-counting that motivated the joint-table test.","method":"Gate-0 joint-cell backtest on N=85, 70/30 held-out, paired DeLong + minimum-detectable-difference power analysis.","decision":"Not scored. Remains a non-scored, surfaced flag in engine 2.6.0. Published as a transparent negative result.","evidence":{"json":"/evaluations/verdicts/phase1-orr-oncology-solid-2026-05.json"},"related":["/research/backtest-nsclc","/methodology/pos-calibration"],"provenance":{"engine_version":"2.6.0","methodology_version":"methodology@2026-05-28"},"dataset_version":"drug-specific-signals@2026-Q3","as_of":"2026-05-28","status":"current"}]}