Deep Dive · rNPV Rank 16VC-fundable

Therapeutic KCNK9 Antibody

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

Breast cancer, gastric cancer, lung cancer (KCNK9-overexpressing tumors)

Modality

Monoclonal Antibody

Mechanism

KCNK9 ion channel inhibitor (antibody)

Target

KCNK9

rNPV Envelope

Low

-$50.7M

costs +25% · peak −25%

Base

-$23.8M

cumulative PoS 3.8%

High

$3.2M

costs −25% · peak +25%

Costs follow naked-biologic CMC norms with a biomarker-stratified (KCNK9-overexpression-selected) Phase 2/3 that is smaller than an all-comers oncology program, anchored to the Elahere FRα-stratified ovarian pathway and Herceptin's over-expression-defined breast development. Phase 1 PoS uses oncology-mAb baselines; Phase 2 is held below BIO/QLS oncology-biologic levels with an explicit downward adjustment because targeting an ion-channel cap-domain epitope for tumor cytotoxicity is mechanistically unprecedented in an approved drug and proof-of-concept in patients carries elevated translational risk beyond the Nature Communications 2016 xenograft/metastasis data. Phase 3 PoS reflects biomarker-stratified oncology-antibody precedent, kept under the buoyant Phase-1 number so cumulative loss-of-approval lands in the single-digit-percent band.

01

Composite score breakdown

Locked rubric — 40/30/30 weights

Clinical relevance · 40%

0.70

Modality fit · 30%

0.51

Whitespace · 30%

0.50

Composite 0.584 — composite-score rank #37 of 50 top-tier inventions in the jhtv-portfolio@2026-Q2 cohort. The page header uses rNPV rank (#16) to match the index ordering.

02

Comparators

Real programs anchoring the engine inputs

Herceptin (Genentech/Roche) — trastuzumab

Closest mechanism analogue: a naked humanized IgG1 directed against a cell-surface protein over-expressed on ~15-20% of breast tumors, acting through receptor engagement plus ADCC/immune-dependent cytotoxicity — the exact therapeutic logic the KCNK9 mAb proposes (target an over-expressed surface epitope in a marker-defined breast/gastric subset). Establishes that an over-expression-stratified naked anti-surface-protein mAb can reach blockbuster scale in breast and gastric cancer.

Indication: HER2-positive breast cancer; HER2-positive gastric cancer
Modality: Monoclonal Antibody
Approval: 1998
Peak revenue: $7.00B

Criteria 2 and 4: launched naked-mAb breast/gastric comparator that is the canonical over-expression-stratified surface-target precedent; pre-biosimilar peak anchors the naked-mAb breast/gastric revenue ceiling.

Elahere (AbbVie/ImmunoGen) — mirvetuximab soravtansine-gynx

Cleanest biomarker-defined, biologic-based oncology comparator: FRα-directed antibody therapeutic approved (accelerated Nov-2022, full Mar-2024) only in the FRα-expressing platinum-resistant ovarian subset on a companion-diagnostic. Mirrors this asset's strongest indication — ovarian cancer, where KCNK9 is over-expressed in >90% of tumors — and frames the marker-stratified ovarian regulatory/commercial pathway and a realistic biomarker-restricted launch trajectory.

Indication: Folate-receptor-α-positive platinum-resistant ovarian cancer
Modality: Monoclonal Antibody
Approval: 2022
Peak revenue: $1.00B

Criteria 2 and 3: launched biomarker-stratified antibody-based ovarian comparator anchoring companion-diagnostic regulatory design and a marker-restricted ramp; ~$0.5B+ annualized run-rate within first full year sets a conservative biomarker-defined-mAb peak band.

Enhertu (Daiichi Sankyo/AstraZeneca) — trastuzumab deruxtecan

Upper-bound benchmark for an antibody-based agent directed at an over-expressed antigen in breast and gastric cancer; ~$2.9B 2024 sales (Daiichi-reported) defines the realistic ceiling for a successful breast/gastric surface-target antibody franchise. Used only as a ceiling — it is an antibody-drug conjugate with a cytotoxic payload, not a naked mAb, so it overstates what an unconjugated KCNK9 mAb would earn.

Indication: HER2-positive / HER2-low breast cancer; HER2-positive gastric cancer
Modality: Monoclonal Antibody
Approval: 2019
Peak revenue: $5.00B

Criteria 2 and 4: launched breast/gastric antibody-based comparator setting the franchise revenue ceiling; flagged as an ADC, not a naked mAb, so it bounds rather than predicts this asset's peak.

ML308 (NIH Molecular Libraries KCNK9 probe)

Same-target tool comparator, not a clinical competitor: ML308 is a selective small-molecule KCNK9 inhibitor (IC50 ~130 nM) developed under the NIH Molecular Libraries Program as a research probe. It confirms KCNK9 is a chemically tractable but clinically undeveloped target — no KCNK9-directed therapeutic has entered the clinic — supporting the first-in-clinic framing while showing the target has been probed for over a decade without a drug.

Indication: Preclinical / target-validation only (no clinical indication)
Modality: Small Molecule
Approval:
Peak revenue:

Criteria 1: same molecular target (KCNK9) but a non-clinical chemical probe; provides target-tractability and competitive-vacuum context, not a revenue anchor.

03

Stage profile

Asset-specific cost, duration, and PoS by stage

StageCostDurationPoSCitations
Preclinical$16.0M24 mo45.0%[0] [1] [5]
Phase I$55.0M18 mo60.0%[1] [2] [5]
Phase II$150.0M30 mo32.0%[2] [3] [4]
Phase III$330.0M42 mo52.0%[3] [4]
NDA/BLA Review$16.0M12 mo85.0%[3] [4]

Multiplier handling: Eligible multipliers (genetic_validation_2.6x, biomarker_1.7x) 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 13.0%

Peak revenue. KCNK9 is over-expressed in ~40% of breast, ~30% of lung, and >90% of ovarian tumors, so a companion-diagnostic-selected addressable population across a lead solid-tumor indication plus expansion is on the order of low-hundreds-of-thousands of patients globally. Benchmarked between Elahere's biomarker-restricted ovarian trajectory (~$0.5B+ first-full-year run-rate, conservative near-term ceiling for a marker-defined launch) and Herceptin's over-expression-defined breast scale, a successful naked anti-KCNK9 mAb launching into one biomarker-defined subset with measured expansion supports a ~$0.9B risk-unadjusted peak — deliberately below Enhertu's ~$2.9B because that is a payload-bearing ADC across multiple settled indications whereas this is an unconjugated single-mechanism antibody whose monotherapy effect size in patients is unproven.

WACC. An oncology antibody in a biomarker-stratified solid-tumor subset is a mainstream pharma category (12%), but engaging an ion-channel cap-domain epitope for tumor-cell cytotoxicity is an unprecedented therapeutic mechanism with no clinical precedent for any KCNK9-directed agent, so 13% sits one notch above standard oncology-mAb risk.

05

Sensitivity (tornado)

Top drivers of rNPV variance

Peak Revenue
$630M$1.17B
-$36.4M
-$11.2M
+$25.2M
WACC
10%16%
-$7.1M
-$32.3M
$25.1M
PoS: Phase III
42%62%
-$32.1M
-$15.4M
+$16.7M
PoS: NDA/BLA Review
68%100%
-$32.2M
-$16.4M
+$15.8M
Cost: Phase II
$105M$195M
-$17.0M
-$30.6M
$13.6M
Exclusivity Years
9 yr15 yr
-$31.3M
-$18.6M
+$12.7M

Drivers ranked by absolute rNPV swing. The vertical tick inside each bar marks the base rNPV (-$23.8M); 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 · 97.2% of paths
$0 ↓
Success tail · 2.8% of paths
$0P50 median-$559.5MeNPV outcome bin (sqrt-scaled height)$1.33B

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

-$207.3M

P25

-$70.6M

P50 (median)

-$19.0M

P75

-$10.5M

P95

-$5.3M

Prob ≥ 0

2.8%

07

Comparable launch curves

Revenue trajectories of named comparators

Herceptin (Genentech/Roche) — trastuzumab

Launched 1998 · peak $6.65B (estimated)

Y0Y10
08

Evidence register

7 per-assumption citations

AssumptionSourceDateConfidence
KCNK9 is the oncogene at the 8q24.3 breast-cancer amplicon (target genetic validation)
multipliers_rationale
Genomic amplification and oncogenic properties of the KCNK9 potassium channel gene
peer_review
2003-04-01high
Anti-KCNK9 extracellular-domain mAb (Y4) inhibits tumor growth and metastasis — mechanism feasibility
stage_profile.preclinical.pos
A monoclonal antibody against KCNK9 K+ channel extracellular domain inhibits tumour growth and metastasis
peer_review
2016-02-08high
K2P channels have a large structured extracellular cap domain (antibody-accessible epitope)
stage_profile.phase_2.pos
Structures of TASK-1 and TASK-3 K2P channels provide insight into their gating and dysfunction in disease
peer_review
2024-08-05medium
KCNK9 is a clinically undeveloped target — only research-probe small molecules exist (first-in-clinic context)
comparators[3]
Development of a Selective Chemical Inhibitor for the Two-Pore Potassium Channel, KCNK9 (ML308) — NIH Molecular Libraries Probe Report
peer_review
2013-03-07high
Elahere biomarker-stratified ovarian antibody-therapeutic regulatory + early-revenue anchor
comparators[1].peak_revenue_usd
FDA grants accelerated approval to mirvetuximab soravtansine-gynx for FRα-positive, platinum-resistant epithelial ovarian cancer
regulatory
2022-11-14high
KCNK9 imprinting dysregulation in triple-negative breast cancer (indication breadth + biomarker)
peak_revenue_usd
Epigenetic Dysregulation of KCNK9 Imprinting and Triple-Negative Breast Cancer
peer_review
2021-11-30medium
KCNK gene over-expression correlates with poor breast-cancer survival (unmet-need / peak framing)
peak_revenue_usd
Integrative Analysis of KCNK Genes and Establishment of a Specific Prognostic Signature for Breast Cancer
peer_review
2022-05-12medium
09

Thesis

Why this asset earns its rank

This is a genuine therapeutic asset: a humanized monoclonal antibody (patent US-2023-0220067, Hopkins/Maryland/Kennedy Krieger, JHTV case C16217) directed against the extracellular domain of KCNK9 (TASK-3), a two-pore-domain potassium channel that is the sole over-expressed gene at the recurrently amplified 8q24.3 amplicon and a validated oncogene conferring hypoxia and serum-deprivation resistance (Mu 2003). The mechanistic question a CMO will raise first — can a conventional IgG functionally engage an ion channel, whose extracellular loops are usually tiny — has an asset-specific answer: K2P channels are atypical in carrying a large structured extracellular cap domain over the ion pathway, and the founding Nature Communications 2016 study showed an anti-extracellular-domain mAb (Y4) induces channel internalization, kills KCNK9-expressing carcinoma cells, and suppresses lung-xenograft growth and breast-cancer metastasis via cell-autonomous and immune-dependent cytotoxicity. This is mechanistically unprecedented for an approved drug and is the central translational risk, not a settled fact.

Economically the asset is anchored to surface-target oncology antibodies. Herceptin establishes that a naked IgG against an over-expressed surface protein in a marker-defined breast/gastric population can reach blockbuster scale, and Elahere frames the realistic biomarker-stratified path — a companion-diagnostic-restricted antibody therapeutic in the KCNK9-rich ovarian setting (>90% over-expression) with a measured first-year ramp — while Enhertu bounds the franchise ceiling as an ADC, not a naked mAb. The engine result is -$50.7M to $3.2M, with a base rNPV of -$23.8M and cumulative PoS of 3.8%; that spread says the case is real but mechanism-defensibility-heavy, hinging on whether cap-domain engagement produces a monotherapy effect size in patients rather than only in 2016 mouse models.

Verdict: a credible, mechanism-risk-loaded oncology antibody worth a conversation. It earns its rubric rank on clinical relevance (validated oncogenic target across breast/lung/ovarian), high IRA exposure for an oncology biologic, and a neutral whitespace score — not on de-risked clinical data. It is vc_fundable because the target genetics and the published proof-of-concept antibody support a standard biotech equity path, conditional on a biomarker-tight first indication.

10

Key risks

Asset-specific, not generic biotech risks

  • Mechanism-feasibility risk: targeting an ion-channel cap-domain epitope for tumor-cell cytotoxicity has no approved-drug precedent; the effect depends on antibody-induced channel internalization and immune-dependent killing demonstrated only in Nature Communications 2016 xenograft/metastasis models, and human monotherapy effect size is unproven.
  • First-in-clinic / undrugged-target risk: no KCNK9-directed therapeutic has ever entered the clinic; the only same-target agents are research-grade small-molecule probes (e.g. NIH ML308), so there is no clinical precedent, no validated biomarker assay, and no regulatory template specific to KCNK9.
  • Selectivity and on-target safety: KCNK9 (TASK-3) is expressed in neurons (KCNK9 loss causes Birk-Barel intellectual-disability syndrome); a systemically dosed channel-engaging antibody must demonstrate tumor-selective activity and an acceptable CNS/cardio-respiratory safety margin given TASK-channel physiology.
  • Companion-diagnostic dependence: the value case rests on a KCNK9-over-expression selection assay (IHC/CNV) that does not yet exist as a validated companion diagnostic; Elahere's FRα-assay history shows marker-assay development can gate both trial enrollment and label.
  • Stage risk: the asset is preclinical and 'ready for corporate partnership' (RoI 2020, web-published 2025) with data only available on request; the rubric rank reflects target validation and modality, not de-risked clinical evidence, and timelines to a first-in-human study are still front-loaded.