Pipeline

Three programs. Three provisionals.

Patent-defensible multi-specifics across oncology, neurology and senolytics — each on the same eight-to-sixteen-week path from sequence to in-vitro PoC; in-vivo schedule partner-lab dependent.

NOVA-3 · P1

CLDN18.2 × CD3 × 4-1BB
trispecific TCE

Gastric / GEJ adenocarcinoma. Tumor-restricted T-cell engagement with 4-1BB costimulation to address CD3-bispecific exhaustion. Provisional filed Q2 '26.

CLDN18.2 ECL1 — tumor restriction
CD3ε — TCR engagement
4-1BB — costimulation gate
In silico PoC PROV-2026-001
NOVA-3 · P2

TfR1 / IGF-1R BBB-VHH shuttle

Blood-brain-barrier penetration shuttle for CNS-restricted biologics. Bispecific VHH design optimized for transcytosis kinetics over receptor sequestration.

TfR1 — primary transferrin route
IGF-1R — capacity-additive
Cargo-agnostic format
In vitro PoC PROV-2026-002
NOVA-3 · P3

GD3 × IL-7R
senolytic bispecific

Carbohydrate-binding active space — the flagship moat. Senescence-cell targeting via GD3 ganglioside (Iltis et al., Nat Aging 2024). Second arm provides survival-pathway gating; specific co-target under wet-lab evaluation.

GD3 ganglioside — senescence marker
IL-7R — survival pathway
Quantum-corrected glycan binding
Lead candidate ranked PROV-2026-003
Workflow · the DBTL loop

Sequence to PoC in 8–16 weeks.

The same design–build–test–learn path runs under every program. Wet-lab results feed back into the Module Library, so each cycle starts from a stronger corpus. In-vivo schedule is partner-lab dependent.

Eight-to-sixteen-week DBTL timeline: Design, Build, Test, Learn, with a flywheel feeding the Learn stage back to the Module Library. DBTL FLYWHEEL · LEARN → MODULE LIBRARY wk 0 wk 4 wk 6 wk 12 wk 16 DESIGN Generate & rank top-24 · NOVA-Compute BUILD Gene synthesis & expression TEST SPR · BLI · ADCC dose-response LEARN In-vitro PoC · evidence bundle
Figure — DBTL cycle. Indicative durations from target intake to in-vitro proof-of-concept; in-vivo efficacy work runs on the partner lab’s schedule. Each completed cycle re-trains the Module Library.
The DBTL flywheel Design · Build · Test · Learn — each cycle compounds the corpus continuous loop · clockwise schematic — not measured data D DESIGN generate candidates B BUILD express / assemble T TEST wet-lab assays · SPR / BLI / ADCC L LEARN assay results re-train corpus Module Library compounds every cycle +~0.8% accuracy / program / cycle (internal target — not a measured result) arc thickness grows toward LEARN→DESIGN — momentum carried back into the next design cycle
Figure — The DBTL flywheel. Each program turns the same loop: DESIGN generates candidates, BUILD expresses and assembles them, TEST runs wet-lab SPR / BLI / ADCC assays, and LEARN folds the results back into the Module Library so the next cycle starts from a stronger corpus. The compounding rate shown (+~0.8% accuracy / program / cycle) is an internal target, not measured data.
Program detail · numbers

Where each program stands.

The pipeline today — all figures below are in-silico predictions and design targets; wet-lab validation is scheduled, not yet complete, and is labelled as such. P2 is the most advanced (in-vitro); P1 leads on clinical-relevance scope; P3 is the flagship on quantum-corrected glycan binding.

Program pipeline stage chart: three internal programs (P1, P2, P3) plotted across five sequential development stages. Solid copper, verdigris and sodium bars mark completed stages up to each program's honest current stage; dashed grey segments mark planned future stages not yet started. A vertical TODAY marker separates completed from planned work; no program shows in-vivo or clinical data. PROGRAM PIPELINE · HONEST STAGE OF EVIDENCE internal status · pre-partnered · no clinical data DESIGN IN-SILICO PoC IN-VITRO PoC IN-VIVO PoC IND-ENABLING TODAY P1 CLDN18.2 × CD3 × 4-1BB trispecific TCE · PROV-2026-001 current: In-silico PoC (computational only) P2 TfR1 / IGF-1R BBB-VHH shuttle brain-shuttle · PROV-2026-002 current: In-vitro PoC (binding / cell assays) P3 GD3 × IL-7R senolytic bispecific ★ flagship · PROV-2026-003 current: lead candidate ranked · wet-lab not yet started solid = completed dashed = planned / not yet TODAY marker no in-vivo or clinical data claimed
Figure — Program stages. Honest current stage of evidence for three internal programs across five sequential development stages. Solid bars mark completed stages; dashed grey segments are planned, not-yet-started work, and the TODAY line marks the furthest any program has reached — no program shows in-vivo or clinical data.
P3 · Flagship
Lead candidate ranked

P3 · GD3 × IL-7R senolytic bispecific

CVaR-VQE active-space correction over the glycan binding pose. Lead candidate CAND-P3-LEAD-08 ranked at a predicted 42 pM Kd; SPR and ADCC validation are scheduled, not yet run. Quantum-corrected glycan binding is the moat — no surveyed competitor ships this.

Predicted Kd
42 pM
SPR Kd
Scheduled
ADCC EC50
Pending
Design bundle
Anchored
Evidence · EVB-P3-2026-0517-3F2A
Verify the bundle →
P1
In silico PoC

P1 · Trispecific TCE with conditional costimulation

CLDN18.2 tumor restriction × attenuated CD3 × 4-1BB conditional agonism. The third arm is a costimulation gate that only triggers in the tumor microenvironment — designed to address the exhaustion phenotype of classical CD3 bispecifics. Lead candidate CAND-P1-LEAD-04 assembled from MOD-0492 + MOD-0617 + MOD-0744.

Predicted Kd
72 pM
Structural conf
0.91 pTM
Wet-lab status
Scheduled Q3 '26
Provisional
PROV-2026-001
P2
In vitro PoC

P2 · BBB shuttle tuned for kinetics

TfR1 (low-affinity, transcytosis-optimized) × IGF-1R (capacity-additive). Deliberately weak TfR1 binding avoids receptor sequestration; IGF-1R adds a parallel BBB-crossing route. Target brain exposure >3% — roughly 30× industry standard of 0.1% — to be confirmed in the Q3 '26 in-vivo study. Cargo-agnostic scaffold.

Predicted Kd (TfR1)
2.4 nM
Brain exposure
>3% target
In vivo
Q3 '26
Provisional
PROV-2026-002
Full case studies → Partner with us