BVCT predicted clinical benefit from mRNA sequence in 7 days with 4-for-4 ex-ante GO/NO-GO accuracy on Moderna readouts.
BVCT sealed commercial FAILURE calls on mRNA-1010's first Phase 3 efficacy trial (), mRNA-1647 () and VX-522 (), and a SUCCESS call on intismeran autogene (). The dashboard classifies the NO-GO readouts True Negative (TN), True Negative (TN) and True Negative (TN), and the INTerpath-001 readout True Positive (TP).
1 correct GO for intismeran autogene. 3 correct NO-GO for mRNA-1010 (first Phase 3 efficacy trial), mRNA-1647, and VX-522.
In our view, in the absence of a reliable data-independent clinical derisking capability before GO decisions, Moderna took a trial-and-error approach to mRNA medicine that witnessed the monumental Phase 3 success of intismeran autogene for high-risk melanoma after a series of failures in mRNA-1010 for influenza (in its first Phase 3 efficacy trial), mRNA-1647 for cytomegalovirus, and VX-522 for cystic fibrosis.
Before each of those major Moderna readouts, BeatSoC Virtual Clinical Trial (BVCT) issued a corresponding prior-data-independent clinical benefit prediction. All four have now read out. All four matched the BVCT call on the recorded readout — one GO that succeeded, three NO-GOs that failed or were discontinued. Lead times run from 24 days to 670 days.
BVCT simulates the clinical consequence of any mRNA sequence design for any clinical setting without requiring any prior preclinical or clinical data. Across these four Moderna readouts the ex-ante GO / NO-GO accuracy is four for four.
Each BVCT prediction was timestamped before its readout; predictions were sent to Moderna stakeholders on and . The intismeran call was sent directly to Stéphane Bancel and the Moderna board on , 22 days before the INTerpath-001 readout.
Every prediction was generated prior-data-independently — no patient-level data from the trial being predicted, none from any prior trial. The full scorecard is on the dashboard: see No. 2161, 887, 1480 and 843.
In our view, BVCT has outperformed Moderna by 4x in clinical derisking mRNA medicine of any target, delivery or indication.
The GO Call.
Correct GO calls prove that BVCT can identify reward as well as risk.
SUCCESS (statistically significant additive clinical benefit in RFS and OS compared with placebo)
Commercial Prediction
SUCCESS (commercially sufficient additive clinical benefit in RFS weakly-to-moderately superior to pembrolizumab with RR < 0.7 and in OS weakly-to-moderately superior to pembrolizumab with HR < 0.7)
GO · NCT05933577 · Moderna / Merck · Intismeran autogene (V940) + pembrolizumab · Phase 3 · 1,089 patients (registry estimate; the release reports 1,137 enrolled) · Adjuvant High-Risk Stage IIB-IV Melanoma. Delivered: Sent to Stéphane Bancel and the Moderna board on — 22 days before the readout — as part of the Top 10 Anticipated 2H 2026 Readouts letter. Outcome: Met the primary endpoint (RFS) and the secondary endpoint (DMFS) versus pembrolizumab at a pre-specified interim analysis. OS data pending. CORRECT. The dashboard classifies the readout as True Positive (TP). Source: Merck / Moderna press release, . BVCT PandaDoc Ref: UF97Q-NMVZE-JUBGE-DF3AK.
The NO-GO Calls Moderna’s Own Data Confirmed.
Three programmes BVCT called NO-GO have since missed their endpoint in the recorded trial or been terminated (VX-522 on tolerability), matching the predicted lack of sufficient clinical benefit in those readouts.
FAILURE (inferior anti-ILI efficacy for influenza A)
NO-GO · Moderna · mRNA-1010 seasonal influenza vaccine · Phase 3 · 22,502 participants · Seasonal Influenza · versus licensed quadrivalent comparator. Outcome: Did not achieve non-inferiority in anti-ILI efficacy versus active comparator Fluarix Quadrivalent — p = 0.1715, CI −24.1% to 22.2%, against a 10% non-inferiority margin. A later Phase 3 trial of mRNA-1010 (NCT06602024) met its primary endpoint against Fluarix; this call and its classification concern NCT05566639 only. CORRECT. The dashboard classifies the readout as True Negative (TN). Source: ClinicalTrials.gov results, submitted , first posted . BVCT PandaDoc Ref: JMNHN-XAD5W-8WLWM-3HJR5.
NO-GO · Moderna, Inc. · mRNA-1647 cytomegalovirus vaccine · CMVictory · NCT05085366 · Phase 3 · 7,454 participants · CMV Infection. Outcome: Did not meet the primary efficacy endpoint of preventing CMV infection versus placebo; further development in congenital CMV discontinued (Moderna said a Phase 2 trial in bone marrow transplant patients would continue). CORRECT. The dashboard classifies the readout as True Negative (TN). Source: Moderna press release, . BVCT PandaDoc Ref: SNEJB-3T35I-ZVHEZ-VH6ZR.
Every mRNA Sequence Design Can Be Clinically Derisked.
The four records above are not four calls on one modality. They are four calls on four distinct mRNA constructs.
Four Sequence Designs · Four Correct Calls
Asset
mRNA encodes
Delivery
Call
Lead time
mRNA-1010
haemagglutinin antigens, seasonal influenza A and B
Intramuscular prophylactic vaccine
NO-GO
670 days early
mRNA-1647
glycoprotein B plus the pentameric complex, CMV
Intramuscular prophylactic vaccine
NO-GO
526 days early
VX-522
full-length CFTR protein
Inhaled LNP to airway epithelium
NO-GO
621 days early
Intismeran autogene
individualised patient-specific neoantigens
Intramuscular therapeutic cancer immunotherapy
GO
24 days early
Different antigens. Different encoded proteins. Different delivery routes. Different therapeutic areas. The one thing they share is that BVCT has correctly simulated the clinical consequence for each mRNA medicine in its recorded readout, without prior patient data.
Why This Matters to Your Board.
An mRNA programme is a sequence-design decision before it is anything else: which antigen, which construct, which delivery. BVCT resolves the clinical consequence of that decision in seven days. It simulates causal, deterministic clinical effect sizes in proprietary virtual patients, without patient-level data from the trial being predicted or from any prior trial.
Each simulation functions as a head-to-head phase-3-scale quantification against a named comparator, regardless of the programme’s actual stage of development. The default GO / NO-GO boundary is HR = 0.60 as a clear threshold for blockbuster FIC/BIC assets.
The value of a BVCT NO-GO is the saved opportunity cost of capital, manpower, manufacturing slots and patients that can still be redirected to programmes that will deliver: mRNA-1010's first Phase 3 efficacy trial (NCT05566639) enrolled 22,502 participants, CMVictory enrolled 7,454.
At full scale, BVCT NO-GOs can help spare tens of thousands of patients each year from investing time and hope in futile trials, while helping sponsors avoid enrollment costs when prevailing approaches cannot see the futility in advance.
Seven days per asset. Before the commit.
Name any first-in-class or best-in-class mRNA medicine asset, the indication, the patient stratum, and the comparator standard of care. Receive a simulated Phase 3 effect size, a GO / NO-GO call, and the mechanistic root cause behind it — within one week. No patient data required. Sealed and timestamped, exactly as the four records above.
Clinical Foresight Before Decisions. Earlier is Kinder for Patients.