BioinvestGPT ApS · Research use only
BVCT solutions
19 case studies and newsletters from BioinvestGPT, each featuring a unique BVCT capability, from atom-level clinical benefit prediction, patient-data-free calls made before the first patient enrols, first-in-class mechanism calls that look past the biomarker, any-phase calls spanning Phase 1/2 and Phase 3, pipeline-level GO/NO-GO calls, portfolio-wide NO-GO calls shared before the readouts, clinical benefit prediction from mRNA sequence, Phase 3 outcome calls made well before the readout, PFS/OS discordance calls, overall-survival calls in HER2-positive cancers, sealed calls on the year’s most-watched readouts, variant-level clinical derisking before wet-lab spend, verification of AI co-scientist hypotheses, patient-data-free AI with no jailbreak surface, in vivo CAR-T design simulation, ADC design and biomarker discovery, clinical prediction from the genome alone, mathematics-based hazard-ratio prediction, to a self-service GO/NO-GO web app. Each cited prediction links to its trial page, with the prediction date, the recorded calls and the readout classification.
Pipeline and asset case studies
- Case study · ZiltivekimabZiltivekimab: ZEUS, HERMES, ATHENA. Three BVCT NO-GOs, already on record.BVCT locked NO-GO calls on ziltivekimab for ZEUS, HERMES and ATHENA on 24 Mar 2024. ZEUS missed its primary endpoint on 31 Jul 2026.5 dashboard predictions cited
- Case study · ATHENA · ZiltivekimabBefore the first patient was enrolled, BVCT called NO-GO on ATHENA — from Drug MoA and the protocol alone.BVCT locked its NO-GO on ATHENA (ziltivekimab) on 24 Mar 2024, 8 days before the first participant was enrolled, from the mechanism and public protocol.1 dashboard prediction cited
- Case study · AOC 1001 (del-desiran) in DM1 · 23 Sep 2026HARBOR missed on 8 September 2026. The NO-GO was sealed 1,235 days earlier.BVCT sealed a FAILURE call on AOC 1001 on 22 Apr 2023, at Phase 1/2, and again at Phase 3; HARBOR missed its primary endpoint on 8 Sep 2026.2 dashboard predictions cited
- Case study · Sanofi pipeline · 30 Jul 2026Sanofi’s 10+ post-readout decisions mirror BVCT’s pre-readout calls sent to Sanofi well in advanceThirteen BVCT calls on Sanofi trials, sealed 18 Apr 2025 and sent to Sanofi: eleven adjudicable, all classified TP or TN; one pending, one not adjudicable.12 dashboard predictions cited
- Case study · Aficamten vs mavacamten in non-obstructive HCM · 14 May 2026Two Verdicts. Same Indication. Both Right.From molecular differences, BVCT called mavacamten (No. 1959) 12 days and aficamten (No. 2094) months before their non-obstructive HCM readouts.4 dashboard predictions cited
- Case study · Novo Nordisk pipeline · 18 Aug 2026Ziltivekimab. CagriSema. Monlunabant. BVCT Shared those NO-GOs with Novo Leadership Before the Readouts.BVCT sealed NO-GO calls on ziltivekimab, CagriSema and monlunabant in 2024 and sent them to Novo Nordisk R&D before each readout. All three: TN.5 dashboard predictions cited
- Case study · Moderna pipeline · 28 Aug 2026BVCT predicted clinical benefit from mRNA sequence in 7 days with 4-for-4 ex-ante GO/NO-GO accuracy on Moderna readouts.BVCT called four Moderna mRNA readouts ex ante: GO on intismeran autogene, NO-GO on mRNA-1010, mRNA-1647 and VX-522. All four classified TP or TN.4 dashboard predictions cited
- Case study · Eplontersen in ATTR-CM · 12 Aug 2026AstraZeneca Could Have Avoided a £19 Billion Loss with BVCT.BVCT sealed its eplontersen ATTR-CM call on 12 Dec 2024, 574 days before CARDIO-TTRansform missed its primary endpoint on 9 Jul 2026.1 dashboard prediction cited
- Case study · Tovecimig · PFS/OS discordance · 4 May 2026Patient-Data-Free BVCTs Eliminate PFS-OS Discordance Before DecisionsBVCT called PFS superior and DLL4-dependent OS for tovecimig in COMPANION-002 (No. 1195) on 3 Nov 2024. The 27 Apr 2026 readout: PFS superior, OS not.11 dashboard predictions cited
- Case study · ZanidatamabWe called the Zanidatamab triplet’s OS benefit, locked 19 Jan 2025, 302 days before the readoutBVCT locked its call on the zanidatamab triplet in HERIZON-GEA-01 (No. 590) on 19 Jan 2025, 302 days before the readout, after three earlier HER2+ calls.4 dashboard predictions cited
Calls ahead of readouts
- Calls ahead of readouts · BioPharma Dive’s 2H 2026 watch list · 28 Jul 2026All Ten Most-Anticipated Readouts of 2H 2026. Called Before the Data.Fourteen sealed BVCT calls on the ten entries BioPharma Dive listed for 2H 2026: seven GO, seven NO-GO. Three are scored on the dashboard so far.3 dashboard predictions cited
Platform and perspective
- Platform · Genetic variants · 8 Jul 2026Clinically Derisk Every Genetic Variant Before Preclinical Spend.BVCT simulates a Phase-III effect size vs standard of care for each candidate variant, patient-data-free. VERVE-101 was called FAILURE 12 days before data.3 dashboard predictions cited
- Platform · BVCT and pharma AI co-scientists · 2 Jun 2026The Clinical Derisking Layer for Every Pharma AI Co-Scientist.How BVCT scores AI co-scientist hypotheses by simulated Phase III effect size vs standard of care, with mavacamten (No. 1959) and aficamten (No. 2094).2 dashboard predictions cited
- Perspective · Trusted AI for pharma · 17 Jun 2026BVCT vs Anthropic — Better. Safer. Jailbreaking-Free. National-Security Aligned.Semafor's Reed Albergotti wrote that no AI model has ever avoided being jailbroken. Why patient-data-free BVCT has no jailbreak surface: two HCM calls.2 dashboard predictions cited
- Decisive clinical advantage · In vivo CAR-T design · 23 Apr 2026The Clinical GPS of in vivo CAR-T DesignBVCT simulates in vivo CAR-T design features, from fusogens to payload decay, to forecast effect size (HR, ORR) against a competitor in vivo CAR-T or SoC.
- Decisive clinical advantage · ADC designThe Clinical GPS of BeatSoC ADC Drug Design and Biomarker DiscoveryBVCT simulates how each ADC design and patient stratification parameter changes the PFS/OS hazard ratio vs standard of care, without retrospective RWD.
- Patient-Data-Free · TRL-9 · Deployed · 3 May 2026The Genome Is Enough — A TRL-9 Existence ProofBVCT predicts the clinical-benefit hazard ratio of any candidate drug versus standard of care from the genome, with no patient data ingested.
- Perspective · Mathematics, not data · 4 May 2026Mathematics is EnoughWhy BVCT predicts Phase-3 hazard ratios from mathematics and the genome, with no patient data, where statistical AI is structurally blind.
- BioinvestGPT · BVCT web app · Introduction · 22 Apr 2026Gain clinical foresight before every decision.The BVCT web app: predicted clinical effect sizes vs standard of care, a causal knowledge graph and virtual clinical trials for every GO/NO-GO decision.