The Market Panorama市场全景 · what we organized so you didn't have to

§0 · What this page is — and is not本页是什么 — 不是什么

On 2026-05-15 at the Antibody Conference, you raised three questions worth answering well. This page is a second attempt — aligned to what you actually asked, not to what we wanted to tell you. The earlier version at /strategic-overview framed two of those three questions wrong; this one is the correction.

§1 · Q1 — What can AI actually do in mRNA delivery / LNP / cell therapy today?问题一 — AI 在 mRNA 递送 / LNP / 细胞治疗领域今天真正能做什么?

You asked the panel this question, and you answered it for SunVax with empirical screening. We are not going to contradict that answer. What we can add is an honest external read of where AI is actually shipping value in this field — versus where the marketing copy says it is.

Where AI is shipping real value today

• Cargo design (mRNA sequence optimization · codon optimization · UTR engineering) — Moderna, BioNTech, Generate Biomedicines all use ML here; results are measurable in mRNA half-life and translation efficiency. Real value, well-established.
• Neoantigen prediction (HLA-presentation likelihood · T-cell receptor affinity) — Moderna V940 / Merck personalized vaccine pipeline depends on it. Phase 3 readout pending.
• Literature surveillance + claim verification (what LIGHT HOPE does on this site) — incremental utility · not a frontier.

Where AI is over-promised today

• De novo ionizable lipid design — Generate Biomedicines, various university labs claim this; the wet-lab validation rate from ML-predicted candidates is order-of-magnitude lower than the marketing decks suggest. Your panel answer (empirical wins for now) matches the actual evidence base, not the hype.
• Cell-type-specific tropism prediction — even thinner. The chemical space is ~10¹⁰; public training data ~10³-10⁴. The arithmetic does not support extrapolation at the level vendors imply.
• Outcome forecasting in clinical settings — over-promised across the industry, not specific to mRNA delivery.

What this means for SunVax positioning

When pharma BD asks SunVax "what about AI?", the defensible answer is the one you gave at the panel: empirical screening wins in this field today, and we are tracking ML developments without depending on them. That is more credible than competitors whose entire deck is "AI-designed everything" — because the evidence base is on your side.

§2 · Q2 — Where does SunVax sit in the market landscape?问题二 — SunVax 在市场版图里坐标在哪?

You know your own strengths better than we do. What we can usefully add is an outside-in map of where each peer sits, so that in any pharma BD conversation you have the landscape context they bring to the meeting. Below is the 10-company landscape for the delivery layer in EGFR+ NSCLC immunotherapy.

Three modality lines · what delivery each needs

Where SunVax sits in this map is a conversation you can have with pharma BD partners directly — we are not in a position to tell you what your defensible position is. What we have organized is the surrounding landscape, so the conversation starts from a shared map.

§3 · Q3 — The R&D kit + pharma partnership tradeoff问题三 — R&D kit 直销 + pharma partnership 两条路径的权衡

You operate two channels in parallel: research-grade kit sales (single-digit listed prices on order.sunvaxmrna.com) and partnership engagements (custom formulation work for pharma cargo). Most LNP platform companies pick one and lose the other. The dual-channel choice is unusual, and the open question is how to allocate scarce founder-time between them. Below is the outside view; the inside view is yours.

R&D kit channel — outside view

• Strength: predictable monthly revenue · low customer-acquisition cost · builds reference data each sale
• Constraint: TAM bounded by academic + early-biotech R&D budgets · ceiling ~$2-5M ARR without major channel scaling
• 2026-2027 tailwind: cell-type-specific demand growing as immunotherapy moves cell-type-specific (vaccine APC · in-vivo CAR-T · NK reprogramming)
• 2026-2027 headwind: competitive entrants (Capstan kit-adjacent products · Mittal GM3-LNP if commercialized · ML-driven competitors at lower wet-lab cost)

Pharma partnership channel — outside view

• Strength: ceiling is 100× larger than R&D kit channel · single deal can fund 18-36 months of runway · IP-licensing precedent strong (Acuitas / Pfizer)
• Constraint: long sales cycle (12-24 months) · founder-attention-intensive · pharma BD prefers IP they can absorb cleanly (FTO clarity matters)
• 2026-2027 timing: bispecific (LiTE-RNA · Moderna V940 · BioNTech BNT116) + in-vivo CAR-T programs entering Phase 1-3 · LNP-supply partnerships getting signed now
• Risk: a pharma partner that absorbs your IP wholesale can compete with your kit channel

Allocation question (you decide, not us)

The question is not which one; the question is how much founder-time to weight to each in the next 6-12 months. Our outside view: with the LiTE-RNA + V940 + BNT116 + GM3-LNP programs hitting Phase milestones in 2026-2027, the pharma BD window is unusually open right now, and the cell-type-specific delivery niche is exactly what those programs need. But this is your call — we can refresh the market coordinates each quarter; the allocation decision is yours.

§4 · What we offer — in one paragraph我们提供什么 — 一段话

§5 · The discipline纪律

• ✗Will not contradict the founder's stated methodology · NCNU 5-禁 P_5
• ✗Will not claim to audit SunVax's wet-lab work · SunVax owns its own data
• ✗Will not produce clinical-decision-style content · NCNU 5-禁 P_1 / P_2 / P_3
• ✗Will not claim regulatory endorsements that do not exist · P_4
• ✗Will not publish anything SunVax asks us to remove · 1-business-day takedown SLA
• ✗Will not forward SunVax intel to other customers · confidentiality protocol §7
• ✗Will not push commercial conversations SunVax has not asked for

§6 · Further reading延伸阅读

• The 30-page strategic brief — delivered separately under NDA-INTERNAL data class.
• /insights/empirical-vs-ml-lnp — your on-the-record empirical methodology, written up.
• /insights/how-this-site-is-built — the discipline behind every claim on this site.
• /strategic-overview — the earlier version of this page, preserved for transparency. This page (industry-map) supersedes it on framing.
• /about · /contact — methodology, takedown commitment, and how to engage / correct / end.