Deep Dive · rNPV Rank 43Partnership candidate

Entirely GRAS Material-based Large Brain-Penetrating Nanoparticles for Widespread Therapeutic Distribution in Healthy and Tumor-bearing Brain Tissues

Generated by an autonomous AI research agent — Anthropic Claude Opus 4.7 or OpenAI GPT-5.5, max reasoning effort. Sources cited inline. Full disclosure at /methodology/jhtv-deep-dive.

Indication

Glioblastoma (brain tumor)

Modality

Gene Therapy

Mechanism

GRAS-material brain-penetrating nanoparticle gene delivery

Target

rNPV Envelope

Low

-$60.9M

costs +25% · peak −25%

Base

-$47.9M

cumulative PoS 2.0%

High

-$34.8M

costs −25% · peak +25%

Costs and durations follow the cell/gene-therapy CNS archetype scaled down for an orphan GBM population with biomarker-free, surgically-delivered intervention (MDNA55 CED single-administration Phase 2b ran without large patient counts; DELYTACT approved on a single-arm Phase 2). PoS values are deliberately conservative and FINAL (fully calibrated, no downstream lift): the platform has strong preclinical distribution data from the same lab lineage but ZERO clinical validation of any payload-loaded version, and the closest gene-therapy GBM precedent (Toca 511) failed its Phase 3 — so Phase 2 PoS is held well below generic oncology benchmarks to reflect the field's repeated late-stage GBM failures. Cumulative LoA ≈ 0.40×0.55×0.28×0.40×0.80 ≈ 1.97%, consistent with a preclinical CNS gene-therapy-archetype asset and the top-10 cohort's 6–11% discipline applied to a riskier delivery-platform-plus-payload combination.

01

Composite score breakdown

Locked rubric — 40/30/30 weights

Clinical relevance · 40%

0.60

Modality fit · 30%

0.74

Whitespace · 30%

0.50

Composite 0.612 — composite-score rank #27 of 50 top-tier inventions in the jhtv-portfolio@2026-Q2 cohort. The page header uses rNPV rank (#43) to match the index ordering.

02

Comparators

Real programs anchoring the engine inputs

Toca 511 / Toca FC (Tocagene / NCI) — vocimagene amiretrorepvec retroviral replicating vector

Closest archetype anchor for a locally-administered gene-therapy modality in glioblastoma: a retroviral vector delivered into the tumor/resection cavity, exactly the development lane a CED-delivered nanoparticle gene-therapy payload would occupy. CAUTIONARY PRECEDENT, NOT A LIVE ANCHOR — the pivotal Toca 5 Phase 3 in recurrent high-grade glioma missed its overall-survival endpoint and the program was effectively ended; the follow-on newly-diagnosed NCI study NCT04105374 was withdrawn ('NCI approval withdrawn', verified withdrawn 2020). Anchors the harsh clinical reality that delivery elegance has not converted to a GBM survival benefit.

Indication: Recurrent / newly diagnosed glioblastoma (high-grade glioma)
Modality: Gene Therapy (replicating retroviral vector)
Approval:
Peak revenue:

Criteria 1 and 2: same modality bucket (gene therapy) and same indication (GBM), locally administered into the tumor field. Used as a cautionary regulatory/clinical precedent, not a revenue anchor.

DELYTACT / teserpaturev (G47Δ, Daiichi Sankyo) — third-generation oncolytic HSV-1

The only gene-therapy-modality product to reach approval in malignant glioma (conditional/provisional approval in Japan, 2021, on a single-arm Phase 2 with ~1-year survival 84.2% and median OS 20.2 months). Administered by repeated intratumoral injection — not a nanoparticle carrier, so it is a modality/indication anchor for the achievable economic ceiling of a locally-delivered GBM gene-therapy product, not a delivery-mechanism match. Establishes that the gene-therapy bucket in GBM has at most a niche, single-geography commercial precedent, not a blockbuster one.

Indication: Malignant glioma (including glioblastoma)
Modality: Gene Therapy (oncolytic HSV-1 virus)
Approval: 2021
Peak revenue: $50.0M

Criteria 1 and 3: same modality bucket and same indication with an actual (regional, conditional) approval; anchors the realistic upper bound on GBM gene-therapy product economics and the orphan/expedited regulatory pathway.

MDNA55 (Medicenna Therapeutics) — IL4R-targeted immunotoxin delivered by convection-enhanced delivery

Best available same-route anchor: an investigational agent delivered into recurrent GBM by CED with multi-catheter stereotactic placement and real-time imaging — the identical administration paradigm this nanoparticle platform requires. Completed Phase 2b (median OS ~11.6 months all-comers, ~15 months high-dose) with a positive signal but no Phase 3 yet; the program is partnership-seeking. It is an immunotoxin, not a gene therapy, so it anchors CED feasibility, trial design, and catheter logistics — not the molecular mechanism.

Indication: Recurrent glioblastoma
Modality: Recombinant immunotoxin (IL4-Pseudomonas exotoxin fusion), CED-administered
Approval:
Peak revenue:

Criteria 3 and 4: same regulatory/administration pathway (CED into GBM) and same patient population; calibrates Phase 1/2 cost, duration, catheter logistics and the partnership-rather-than-standalone funding reality.

Hanes/Suk Johns Hopkins brain-penetrating-nanoparticle (BPN) lineage — dense-corona non-adhesive PLGA/PEG nanoparticles, CED-delivered (cisplatin- and paclitaxel-loaded preclinical programs)

Direct technology-platform predecessor from the same laboratory lineage as this invention (Nance 2012; cisplatin-loaded BPN 'cures malignant glioma in rats'; non-adhesive biodegradable nanoparticles giving widespread orthotopic-tumor transgene expression, PMID 32323162). Establishes the platform is real and reproducible across multiple payloads but that every prior generation has remained preclinical — the defining risk for this asset. Archetype anchor for a CNS nanoparticle delivery PLATFORM, explicitly not a product comparator.

Indication: Malignant glioma (preclinical, multiple payloads: chemotherapeutic and nucleic-acid)
Modality: Non-viral nanoparticle delivery platform (PLGA + poloxamer/PEG non-adhesive corona)
Approval:
Peak revenue:

Criteria 1: same mechanism/technology class (non-adhesive brain-penetrating nanoparticle delivered by CED) from the same lab lineage; used as a development-archetype anchor for platform value, not a marketed product.

03

Stage profile

Asset-specific cost, duration, and PoS by stage

StageCostDurationPoSCitations
Preclinical$22.0M30 mo40.0%[0] [1] [4]
Phase I$60.0M18 mo55.0%[2] [3]
Phase II$110.0M30 mo28.0%[2] [3] [5]
Phase III$230.0M36 mo40.0%[3] [5]
NDA/BLA Review$15.0M12 mo80.0%[3] [5]

Multiplier handling: Eligible multipliers (orphan_designation_eligible, expedited_pathway_eligible) are already reflected in Day-1 comparator-calibrated PoS. Re-applying them via log-odds stacking would double-count, so per-stage PoS is taken as final. See methodology for the rule.

04

Peak revenue and discount rate

$120.0M peak · WACC 15.0%

Peak revenue. This is a delivery PLATFORM, not a product, so no genuine product peak exists — the figure is an explicitly hypothetical illustrative envelope anchored to the nearest real archetype (DELYTACT, the only approved GBM gene-therapy-modality product, a regional conditional approval with a niche, geography-limited commercial footprint we anchor near $50M, scaled up modestly for a hypothetical broader-geography orphan GBM gene-therapy product). Realistically, JHU's economic capture from an enabling nanoparticle platform is upfront + milestone + low-single-to-mid-single-digit royalty on a licensee's GBM gene-therapy product, NOT this product-peak number; the $120M figure exists only so the pipeline can compute a cohort-consistent rNPV envelope and must not be read as a standalone-equity revenue forecast.

WACC. A preclinical CNS gene-therapy-archetype delivery platform with no clinical validation and a field littered with Phase 3 GBM failures sits at the high end of biotech discount rates; 15% reflects CNS gene-therapy risk above a standard 12% biologic and the additional platform-plus-payload integration risk.

05

Sensitivity (tornado)

Top drivers of rNPV variance

PoS: Preclinical
32%48%
-$42.0M
-$53.8M
$11.8M
Cost: Preclinical
$15M$29M
-$42.4M
-$53.4M
$11.1M
Cost: Phase I
$42M$78M
-$43.3M
-$52.5M
$9.1M
WACC
12%18%
-$51.5M
-$44.5M
+$7.1M
Cost: Phase II
$77M$143M
-$44.4M
-$51.4M
$7.0M
PoS: Phase I
44%66%
-$45.1M
-$50.7M
$5.7M

Drivers ranked by absolute rNPV swing. The vertical tick inside each bar marks the base rNPV (-$47.9M); each bar spans the rNPV range produced by flexing one input between its low and high values. Gold = the input pushes rNPV up when increased; red = the input pushes rNPV down when increased.

06

Monte Carlo distribution

1,000 trials · rpNPV mode

Failure cluster · 99.9% of paths
$0 ↓
Success tail · 0.1% of paths
$0P50 medianBase rNPV (mean)-$336.2MeNPV outcome bin (sqrt-scaled height)$19.1M

This is a bimodal distribution by construction, not a Gaussian. Most paths terminate in clinical failure (red cluster — accumulated cost only); a minority succeed and capture full peak revenue (green tail). Bar heights are square-root-scaled so the success tail stays visible alongside the much taller failure cluster; exact counts are preserved in the percentiles below. Gold line = median (P50). Navy dashed = base rNPV (mean) — the probability-weighted expected value, which can sit above the median when the upper tail is strong enough to outweigh the failure cluster (and close to the median when it isn’t).

P5

-$146.7M

P25

-$71.1M

P50 (median)

-$25.8M

P75

-$13.8M

P95

-$7.9M

Prob ≥ 0

0.1%

07

Comparable launch curves

Revenue trajectories of named comparators

DELYTACT / teserpaturev (G47Δ, Daiichi Sankyo) — third-generation oncolytic HSV-1

Launched 2021 · peak $47.5M (estimated)

Y0Y10
08

Evidence register

6 per-assumption citations

AssumptionSourceDateConfidence
Asset is a GRAS PLGA/poloxamer non-adhesive brain-penetrating nanoparticle DELIVERY platform (not a therapeutic), CED-administered
cmo_findings.asset_class_reality
A Highly Translatable Dual-arm Local Delivery Strategy To Achieve Widespread Therapeutic Coverage in Healthy and Tumor-bearing Brain Tissues (Negron et al., Small 2023)
peer_review
2023-01-17high
Same-lab BPN lineage is reproducible across payloads but has remained preclinical (platform maturity + key risk)
comparators[3]
Non-adhesive and highly stable biodegradable nanoparticles that provide widespread and safe transgene expression in orthotopic brain tumors
peer_review
2020-04-22high
Brain-penetrating nanoparticle + CED can produce preclinical efficacy in malignant glioma (platform plausibility)
stage_profile.preclinical.pos
Convection enhanced delivery of cisplatin-loaded brain penetrating nanoparticles cures malignant glioma in rats
peer_review
2017-09-01high
CED into recurrent GBM is a clinically feasible administration route with established multi-catheter trial design (Phase 1/2 calibration)
stage_profile.phase_2.pos
Targeting the IL4 receptor with MDNA55 in patients with recurrent glioblastoma: Results of a phase IIb trial
trial_disclosure
2023-01-16high
Closest gene-therapy GBM precedent FAILED late-stage / was withdrawn (cautionary, downward Phase 2/3 PoS calibration)
stage_profile.phase_3.pos
Toca 511/Toca FC added to standard treatment for newly diagnosed glioblastoma (NCT04105374) — trial record
trial_disclosure
2020-03-01high
GBM is an orphan-eligible, uniformly fatal indication (orphan/expedited pathway + envelope ceiling anchoring)
peak_revenue_usd
Epidemiology and Outcome of Glioblastoma (NCBI Bookshelf, Glioblastoma)
peer_review
2017-09-27high
09

Thesis

Why this asset earns its rank

JHU 36712 is not a therapeutic candidate — it is a delivery platform. The invention is a family of GRAS-material nanoparticles (PLGA cores with a non-adhesive poloxamer/PEG corona, 60–200 nm) administered by convection-enhanced delivery so that a co-formulated payload distributes widely and deeply through healthy and tumor-bearing brain parenchyma rather than staying trapped near the catheter. The biological insight is real and reproducible across this Hopkins lab lineage: a sufficiently dense, non-adhesive corona lets large particles diffuse through brain extracellular matrix, and prior generations carried chemotherapeutic and nucleic-acid payloads to preclinical efficacy (cisplatin-loaded BPN 'cured' rat glioma; non-adhesive nanoparticles gave widespread orthotopic-tumor transgene expression). Because it is an enabling carrier with no target, no mechanism of its own, and no biomarker (target = None), the rNPV of a product does not apply here; the value path is licensing the platform to a glioblastoma drug developer for upfront, milestone, and royalty economics. The pinned 'gene_therapy' bucket is retained per methodology, but the platform is payload-agnostic and its lead data are not gene-therapy-specific — disclosed below as a caveat and a HARD finding rather than a re-rank.

The comparators frame the economics honestly through archetype anchors, not product analogues. DELYTACT (teserpaturev), the only approved gene-therapy-modality GBM product, reached only a regional conditional approval on a single-arm Phase 2 — evidence the bucket has a niche, geography-limited ceiling, not a blockbuster one. Toca 511 is the cautionary precedent: an elegant locally-delivered GBM gene therapy whose pivotal Phase 3 missed overall survival and whose follow-on study was withdrawn. MDNA55 anchors the identical CED administration paradigm (multi-catheter, single-dose, stereotactic) and shows the route is feasible but partnership-bound. The engine result is -$60.9M to -$34.8M, with a base rNPV of -$47.9M and cumulative PoS of 2.0%; that envelope is shown only for cohort consistency — the rNPV is not the decision criterion here, which is why the asset is classified as a partnership candidate rather than a standalone VC-fundable product, and the realistic JHU capture is a low-single-digit royalty on a licensee's GBM program, not the illustrative product peak.

Verdict: a credible, well-published CNS delivery platform whose value is as licensable infrastructure for someone else's glioblastoma payload, not as an equity-fundable drug. It earns its rubric rank on a high gene-therapy modality_pos, strong IRA exposure for a CNS indication, and a neutral whitespace score for an 'other'-mapped indication — not on being a fundable standalone product; the partnership_candidate archetype follows directly because the platform must be paired with a validated payload and a strategic developer before any clinical or commercial value is realized.

10

Key risks

Asset-specific, not generic biotech risks

  • Asset-class reality: this is a delivery vehicle, not a drug. There is no therapeutic value without a separately-validated payload, a target, and a developer partner; every economic figure here is an illustrative cohort-consistency envelope, not a product forecast — a CMO will (correctly) refuse to value it as a standalone equity asset.
  • The entire field is a clinical graveyard: the closest gene-therapy GBM precedent (Toca 511) failed its pivotal Phase 3 and was withdrawn; CED-delivered GBM agents (MDNA55) have positive signals but no Phase 3; the only approved gene-therapy-modality GBM product (DELYTACT) is a single-geography conditional approval. Delivery elegance has repeatedly failed to convert into a survival benefit.
  • Platform maturity gap: every prior generation of this Hopkins BPN lineage has remained preclinical/rodent. CED dosimetry, catheter reflux, payload stability in the non-adhesive corona, and human brain distribution (orders of magnitude larger than rat) are unproven — the translational risk sits in the carrier-plus-payload integration, which no comparator de-risks.
  • Modality-bucket vs reality: the pinned 'gene_therapy' classification is retained for methodology consistency, but the lead Negron 2023 data and the cisplatin precedent are chemotherapeutic/nucleic-acid carriers — the platform is payload-agnostic, so any gene-therapy-specific economic read (e.g. DELYTACT analogy) is an archetype anchor only, not a like-for-like comparator.
  • Funding-path specific: as a platform, monetization requires a strategic/pharma licensee who already owns a GBM payload and is willing to adopt an invasive CED administration model; JHU's realistic capture is upfront + milestones + low-single-digit royalty, and a wrong partner choice (or no partner) leaves the technology stranded regardless of preclinical strength.