PhaseFolio
Deep Dive · rNPV Rank 02VC-fundable

Gene therapy for neurofibromatosis type 1

Generated by a Claude Opus 4.7 agent (max thinking effort, 1M-context). Sources cited inline. Full disclosure at /methodology/jhtv-deep-dive.

Indication

Neurofibromatosis type 1 (NF1)

Modality

Gene Therapy

Mechanism

AAV gene replacement

Target

NF1

rNPV Envelope

Low

-$43.5M

costs +25% · peak −25%

Base

$27.0M

cumulative PoS 10.8%

High

$97.4M

costs −25% · peak +25%

NF1 AAV is costed like a rare pediatric gene therapy with higher early CMC spend but smaller pivotal studies. Koselugo anchors NF1 clinical/regulatory precedent; Zolgensma and UX111 anchor AAV pricing, trial size, and BLA timing.

01

Composite score breakdown

Locked rubric — 40/30/30 weights

Clinical relevance · 40%

0.80

Modality fit · 30%

0.74

Whitespace · 30%

0.50

Composite 0.692 — composite-score rank #3 of 10 top-tier inventions in the jhtv-portfolio@2026-Q2 cohort. The page header uses rNPV rank (#2) to match the index ordering.

02

Comparators

Real programs anchoring the engine inputs

Koselugo (AstraZeneca/Merck) — selumetinib

Approved NF1 therapy establishing regulatory feasibility and disease-market precedent.

Indication: NF1 plexiform neurofibromas
Modality: Small Molecule
Approval: 2020
Peak revenue: $500.0M

Criteria 2 and 3: same indication and pediatric orphan NF1 pathway, though symptom-modifying rather than gene replacement.

Zolgensma (Novartis) — onasemnogene abeparvovec

One-time AAV economics benchmark for pediatric monogenic disease.

Indication: Spinal muscular atrophy
Modality: Gene Therapy
Approval: 2019
Peak revenue: $1.40B

Criteria 3 and 4: same rare pediatric AAV gene-therapy modality and premium-pricing model.

UX111 (Ultragenyx) — rebisufligene etisparvovec

Late-stage AAV CNS orphan program anchoring small pivotal trial and BLA timing.

Indication: Sanfilippo syndrome type A
Modality: Gene Therapy
Approval:
Peak revenue:

Criteria 3 and 4: same rare pediatric CNS AAV pathway with expedited regulatory features.

03

Stage profile

Asset-specific cost, duration, and PoS by stage

StageCostDurationPoSCitations
Preclinical$28.0M24 mo56.0%[1] [2] [5]
Phase I$85.0M18 mo72.0%[1] [2] [5]
Phase II$190.0M30 mo48.0%[0] [2] [5]
Phase III$260.0M30 mo62.0%[0] [2] [5]
NDA/BLA Review$18.0M12 mo90.0%[0] [2] [5]

Multiplier handling: Eligible multipliers (genetic_validation_2.6x, biomarker_1.7x, orphan_1.4x, gene_therapy_1.41x, fast_track_or_rmat, pediatric_voucher) 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

$900.0M peak · WACC 12.0%

Peak revenue. Revenue is modeled above Koselugo because root-cause AAV could command one-time gene-therapy pricing, but below Zolgensma because NF1 penetrance and lesion heterogeneity complicate addressability. The $900M peak assumes pediatric plexiform neurofibroma first use with later expansion optionality.

WACC. Genetically defined orphan AAV development is risky but squarely within venture-backed rare-disease playbooks.

05

Sensitivity (tornado)

Top drivers of rNPV variance

WACC
9%15%
$96.0M
-$13.1M
$109.1M
Peak Revenue
$630M$1.17B
-$19.4M
$73.3M
+$92.7M
PoS: Phase III
50%74%
-$3.8M
$57.7M
+$61.5M
PoS: Phase II
38%58%
$647K
$53.3M
+$52.6M
Exclusivity Years
9 yr15 yr
-$1.7M
$47.4M
+$49.1M
PoS: NDA/BLA Review
72%100%
-$3.9M
$44.1M
+$48.0M

Drivers ranked by absolute rNPV swing. The vertical tick inside each bar marks the base rNPV ($27.0M); 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 · 89.8% of paths
$0 ↓
Success tail · 10.2% of paths
$0P50 medianBase rNPV (mean)-$492.5MeNPV outcome bin (sqrt-scaled height)$1.88B

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

-$306.3M

P25

-$147.9M

P50 (median)

-$37.3M

P75

-$15.7M

P95

$1.16B

Prob ≥ 0

10.2%

07

Comparable launch curves

Revenue trajectories of named comparators

Koselugo (AstraZeneca/Merck) — selumetinib

Launched 2020 · peak $475.0M (estimated)

Y0Y10
08

Evidence register

6 per-assumption citations

AssumptionSourceDateConfidence
NF1 approved-therapy comparator
comparators[0]
NCI Cancer Currents: Selumetinib approved for NF1 plexiform neurofibromas
regulatory
2020-05-08high
Koselugo commercial/regulatory precedent
peak_revenue_usd
Merck press release: Koselugo approved by FDA for pediatric NF1 plexiform neurofibromas
news
2020-04-10high
Zolgensma AAV economics benchmark
comparators[1]
Novartis Annual Report 2024
company_filing
2025-01-31high
UX111 late-stage AAV orphan comparator
comparators[2]
ClinicalTrials.gov NCT02716246: AAV gene transfer for MPS IIIA
trial_disclosure
2016-03-25high
UX111 follow-up and pivotal timing
stage_profile.phase_3.duration_months
ClinicalTrials.gov NCT04088734: long-term follow-up after ABO-102 gene transfer
trial_disclosure
2019-09-12high
Rare gene-therapy PoS adjustment
stage_profile.phase_2.pos
BIO/QLS/Informa Clinical Development Success Rates 2011-2020
peer_review
2021-02-17medium
09

Thesis

Why this asset earns its top-10 rank

The NF1 AAV asset is a root-cause gene-replacement concept engineered around AAV's ~4.7 kb packaging limit: full-length NF1 (~8.5 kb CDS) cannot fit in a single AAV, so the asset uses a mini-NF1 construct — a truncated GAP-related domain (GRD) fused to the KRAS4B C-terminal targeting domain — paired with an in-vivo-evolved capsid (AAV-NF K55) developed via sequential capsid shuffling and peptide library screening. The 2025 Nature Communications paper from the JHU group demonstrates tumor suppression in xenograft mouse models of NF1-related cancers. The asset's top-10 position comes from strong genetic validation, pediatric orphan fit, and a credible disease-modification thesis that addresses the gene-size constraint head-on rather than papering over it.

Koselugo proves that NF1 plexiform neurofibroma is an approvable commercial indication, while Zolgensma and UX111 anchor rare pediatric AAV economics and development cost. The engine result is -$43.5M to $97.4M, with a base rNPV of $27.0M and cumulative PoS of 10.8%; that supports the VC-fundable archetype if the initial program focuses on a high-need, genetically confirmed NF1 subgroup with tractable delivery and measurable lesion endpoints.

The verdict is a credible rare-disease gene-therapy deep dive that earns its ranking on engineered-vector merit, not on naive gene-replacement framing. The main remaining questions are whether GRD-only neurofibromin restores enough RAS-GAP activity in vivo to drive durable lesion regression, and whether the AAV-NF K55 capsid achieves the tissue distribution NF1's multisystem pathology demands.

10

Key risks

Asset-specific, not generic biotech risks

  • NF1 disease is multisystem; even an evolved capsid (AAV-NF K55) may not reach all relevant lesion compartments — plexiform neurofibroma is the lead but cutaneous neurofibromas, optic pathway gliomas, and MPNST may need separate delivery strategies.
  • GRD-only neurofibromin retains catalytic RAS-GAP activity in xenografts but lacks domains that may be required for full physiologic regulation; whether catalytic-only restoration delivers durable clinical benefit at scale is unproven.
  • Koselugo sets a symptom-modifying standard that may raise expectations for measurable tumor shrinkage rather than growth inhibition.
  • Pediatric gene-therapy safety monitoring could slow development even with orphan incentives, and engineered-capsid redosing carries vector-immunity risk specific to the AAV-NF K55 sequence.