Vaccines, oncology, and rare disease programs each carry distinct cost structures, attrition profiles, and funding dynamics. Understanding the differences serves as a foundation for sound investment and program strategy.
CAMBASE Research Series: Biotech Startup Financing and Clinical Development Risk • Sources: Gouglas et al. (Lancet, 2018) • Wong et al. (Biostatistics, 2019) • Jayasundara et al. (Orphanet, 2019) • Getz et al. (TIRS, 2023) • ASPE • BioPharma Dive
Vaccine: No single source publishes individual phase costs. Total to end of Phase 2a: $31–68M (Gouglas et al., Lancet GH, 2018). Total to market: $886.8M avg (ASPE, Dec 2024). Phase 3 ~70% of total costs (npj Vaccines, 2025). Individual phase estimates are modelled distributions, not directly published. Oncology Phase 1: $4.5M (Sertkaya et al., ASPE 2016, citing Medidata). Phase 2: $11.2M (Sertkaya 2016). Phase 3: ~$100M for first-line metastatic trial (Chen & Zhang, Front. Oncol., 2023 — higher than Sertkaya $22M all-oncology average). Oncology licensure is estimated. Rare disease Phase 3: median $63–100M (Health Advances, 2019); average $103M (Tufts CSDD cited in Applied Clinical Trials, 2025) — corrected upward from earlier $55M figure. Other rare disease phase costs estimated. ✓ = confirmed from a named primary source.
6.7%
Circle size is proportional to probability of success. The 2019 Biostatistics paper’s phase-transition tables (Table 1 & 2) were corrected in a corrigendum; overall PoS figures confirmed from the abstract are unaffected.
| Driver | Vaccine | Oncology | Rare disease |
|---|---|---|---|
| Phase 3 trial size | Large efficacy cohorts required Immunogenicity & infection endpoints |
~800–1,000 patients Chen & Zhang 2023, first-line metastatic |
Small-N accepted under orphan pathway Tufts CSDD; as few as 12–60 patients in some trials |
| Per-patient cost (Phase 3) | Moderate; assay & follow-up driven | >$100K per patient PhRMA/Battelle: ~$69K oncology vs ~$42K cross-area avg (Health Advances 2019) |
$137K–$5M+ per patient Health Advances 2019 (ERT, gene therapy, large-molecule) |
| Trial operations | Standard; cold-chain adds cost for some platforms | Complex endpoints (OS, PFS); multi-year follow-up | 30% more visits; 23% longer start-up; 19% longer treatment Getz, Smith & Kravet (2023), Ther. Innov. Reg. Sci. 57(1):49–56 — PubMed |
| Manufacturing | Platform-dependent; mRNA/viral vector complex | Moderate–high; ADCs & cell therapies highest | Very high for gene/cell therapy; small-batch cost dominant |
| Public funding | High — CEPI, BARDA, Innovate UK, Wellcome | Moderate — NCI, cancer charities (mainly early-stage) | High — NIH NCATS, EU rare disease funds, patient advocacy, orphan grants |
| Main attrition risk | Phase 3 efficacy failure; platform-specific immunogenicity | Phase 2→3 translation failure; tumour heterogeneity | Recruitment failure; statistical underpowering in small-N trials |
Ask most biotech investors which area commands the most capital, and oncology is the inevitable answer. Ask which area has the worst odds of actually producing an approved drug, and the answer is the same. That tension — between capital concentration and probability of success — is the starting point for a more honest conversation about where development economics actually sit across the three most capital-intensive areas in medicine: vaccines, oncology, and rare disease.
The short version: rare disease is cheaper per program than most people assume, oncology is more expensive per approved drug than most models reflect, and vaccines present a risk profile that makes them uniquely dependent on public co-funding. Each area rewards a different capital structure and a different approach to risk. This article unpacks the data.
| The attrition-adjusted cost of one oncology approval is estimated at $1–4 billion or more. The equivalent figure for a rare disease program is $166–291 million. That difference is not about the difficulty of the science — it is about probability. |
The core counterintuition: smaller is not always more expensive
The instinct in early-stage biotech is to treat rare disease as a premium category — niche patient populations, high per-patient costs, specialist manufacturing. That instinct is not entirely wrong, but it points to the wrong metric. Per-patient costs in rare disease advanced trials are substantially higher than in conventional drug development; Health Advances (2019) found ranges from $137,000 to over $5 million per patient in advanced orphan drug trials, with gene and cell therapies at the upper end. A ratio of approximately 2.5× versus non-orphan programs is sometimes cited in secondary literature, but this requires verification against the full text of Jayasundara et al. (Orphanet Journal, 2019) before formal use.
And yet, across published data, the out-of-pocket cost to achieve one regulatory approval is lower for rare disease programs than for non-orphan ones: approximately $166 million for orphan drugs versus $291 million for non-orphan drugs (both in 2013 USD — Jayasundara et al., 2019), after accounting for the probability of clinical success. The reason is structural. Smaller trials mean lower absolute Phase 3 costs even when per-patient costs are elevated. Orphan Drug Designation provides a 50% US R&D tax credit, fee waivers, and regulatory acceptance of smaller datasets. And phase-transition success rates are significantly higher than in oncology.
| $166–291M Rare disease attrition-adjusted cost to approval (2013 USD, Jayasundara et al. 2019) | Higher Rare disease per-patient costs substantially higher than non-orphan (verify ratio vs Jayasundara 2019) | 3.4–6.7% Oncology overall PoS (Wong et al. 2019, confirmed) | 33.4% / 39.6% Vaccine PoS: all sponsors / industry-sponsored (Wong 2019; Lo/Siah/Wong 2020) |
Sources: Jayasundara et al. (Orphanet Journal of Rare Diseases, 2019 — 2013 USD); Health Advances (2019) for per-patient cost ranges; Wong et al. (Biostatistics, 2019, corrected) for oncology PoS; Lo, Siah & Wong (HDSR/NBER, 2020) for vaccine PoS and phase-transition rates. All figures require independent verification before use in investment or program planning.
Phase-by-phase: where the money actually goes
The table below presents midpoint cost estimates for each development stage across the three areas. These are derived from multiple published sources and should be treated as illustrative reference points, not precise forecasts. Individual programs vary considerably based on indication, geography, trial design, and platform technology.
| Stage | Vaccine | Oncology | Rare disease |
| Preclinical | ~$12M | ~$8M | ~$8M |
| Phase 1 | ~$10M | ~$4.5M | ~$8M |
| Phase 2 / 2a | ~$25M | ~$10M | ~$15M |
| Phase 3 / Pivotal | ~$130M | ~$70M | ~$55M |
| Licensure / Approval | ~$150M | ~$80M | ~$40M |
| Total (no failure, single program) | ~$327M | ~$172M | ~$126M |
Midpoint estimates only. Sources: Vaccine — Gouglas et al. (Lancet Global Health, 2018) and HHS/ASPE. Oncology — Pharmaphorum / ASPE Clinical Trial Costs report; Chen & Zhang (Frontiers in Oncology, 2023). Rare disease — Jayasundara et al. (Orphanet Journal, 2019); Health Advances (2019). Stage-level figures for Phases 1–2 are inferred from reported totals and should be treated as indicative.
The most important observation from this table is the Phase 3 column. Vaccine Phase 3 costs are the highest of the three areas in absolute terms — driven by the need for large immunogenicity and efficacy cohorts. A vaccine Phase 3 efficacy trial can require thousands of participants followed over multiple years, with outcomes that depend on actual infection events in the study population. This makes vaccine Phase 3 trials expensive in a way that is structurally different from both oncology and rare disease.
Oncology Phase 3 costs are moderate in absolute terms but carry a specific design problem: first-line metastatic trials typically enrol 800–1,000 patients, take three years to complete, and can cost around $100 million for a single trial — without any guarantee of translating Phase 2 signals into Phase 3 success. The standard-of-care bar has risen substantially in the past decade, driven by checkpoint inhibitors, and this has increased both trial complexity and the risk of failure at the confirmatory stage.
Rare disease Phase 3 costs are the lowest in absolute terms, but the comparison is somewhat misleading. Small-N trials accepted under orphan pathways reduce total cost, but they also introduce a different risk: statistical underpowering. A trial with 60 patients has limited ability to detect modest treatment effects, and this contributes to the higher-than-expected completion failures and delayed approvals that characterise some rare disease programs.
| Vaccine Phase 3 costs are the highest of the three — but vaccines have the best overall probability of success. Oncology costs less per trial, but requires many more attempts to produce one approval. |
The metric that actually matters: attrition-adjusted cost
Per-program cost is a useful starting point, but it is not the metric that should drive capital allocation decisions. The relevant question is: given realistic attrition rates, how much capital does it take to reliably produce one regulatory approval? The answer requires combining per-program cost with overall probability of success.
The data on probability of success are drawn primarily from Wong et al. (2019), published in Biostatistics, which analyzed over 406,000 clinical trial records from 2000 to 2015. Their findings are stark:
| Area | Phase 1 → 2 | Phase 2 → 3 | Phase 3 → Approval | Overall PoS |
| Vaccine (industry-sponsored) | ~82.5% | ~65.4% | ~80.1% | 39.6% (industry) 33.4% (all sponsors) |
| Oncology | Indicative only—see note | Indicative only—see note | Indicative only—see note | ~3.4–6.7% (confirmed) |
| Rare disease (indicative) | ~70% | ~55% | ~78% | ~25% |
Vaccine and oncology data: Wong et al. (Biostatistics, 2019), analysis of 406,038 trials. Rare disease rates are indicative estimates from multiple sources including Nature Reviews Drug Discovery (2025) supplementary data. Rare disease PoS varies substantially by modality (gene therapy, small molecule, biologic) and should not be treated as a single fixed benchmark.
When these success rates are applied to per-program costs, the attrition-adjusted picture looks like this: vaccines require roughly $319–469 million to achieve one successful candidate from preclinical, accounting for portfolio attrition (Gouglas et al., 2018). Rare disease programs require approximately $166–291 million per approved drug. Oncology, with a 3–7% overall PoS, requires an illustrative $1–4 billion or more — a figure derived by applying that PoS to published per-program costs rather than from a single published source, and one that should be treated with appropriate caution.
The practical implication is significant. Oncology returns need to be disproportionately large to compensate for this PoS profile, and they frequently are — which is why oncology continues to attract the most VC capital despite the worst attrition numbers in any therapeutic area. But for founders and investors who are not operating at the scale required to build a diversified oncology portfolio, the risk-adjusted economics of rare disease or epidemic vaccine programs may be substantially more favorable.
Why oncology PoS is so low — and what it means
The 3.4–6.7% overall probability of success for oncology is not simply a reflection of difficult biology, though the biology is genuinely hard. It reflects several structural features of how oncology drug development works in practice.
- Tumour heterogeneity means that Phase 2 signals in a selected or biomarker-enriched patient population frequently fail to translate to Phase 3 outcomes in a broader population. The history of oncology is littered with programs that looked transformative at Phase 2 and failed at Phase 3 confirmation.
- The standard-of-care bar has risen sharply over the past decade. Checkpoint inhibitors have transformed survival outcomes in melanoma, lung cancer, and other indications, meaning that new entrants must demonstrate improvement over a much higher baseline than programs designed a decade ago.
- Accelerated approval pathways, while valuable for getting drugs to patients faster, have introduced a lag between early approval on surrogate endpoints and the post-marketing confirmatory trials that determine whether drugs remain on the market. Several high-profile oncology withdrawals in recent years reflect this structural tension.
- The financial incentive to run large late-stage trials has pushed program timelines to 6–7 years across phases for a single approval, with per-patient costs exceeding $100,000 in Phase 3 (PhRMA/Battelle data, cited in Health Advances 2019 as ~$69,000 per patient for oncology vs ~$42,000 cross-area median). Trial costs have risen meaningfully over the past decade, though a specific percentage figure requires a primary peer-reviewed source and has been removed from this version pending verification.
For investors, the implication is that oncology investing requires either deep portfolio diversification — multiple shots on goal from the same scientific platform — or extraordinary program selection capability. Single-asset oncology bets are high-variance by structural design, not by accident.
The rare disease advantage: regulatory strategy as financial strategy
One of the least well understood aspects of rare disease development economics is that Orphan Drug Designation is not simply a regulatory label. It is a financial instrument, and a meaningful one.
In the United States, orphan designation provides a 50% tax credit on qualifying clinical research expenses, a waiver of FDA application fees (which can exceed $3 million for a standard new drug application), and seven years of market exclusivity post-approval. In the EU, the equivalent is ten years of exclusivity. Regulatory agencies also accept smaller, shorter trials for orphan indications — which directly reduces Phase 3 absolute cost — and provide enhanced pre-submission support through mechanisms like the FDA’s Office of Orphan Products Development.
Taken together, these advantages meaningfully alter the effective cost structure of a rare disease program relative to its headline figures. A program that costs $55 million across all three phases in out-of-pocket terms may have an effective net cost significantly lower after tax credits and fee waivers are applied. For early-stage startups operating with limited capital, this distinction matters enormously.
| Orphan Drug Designation is among the most powerful financial instruments available to a biotech founder. The 50% US R&D tax credit alone can be transformative for a capital-constrained early-stage program. |
The counterweight is execution risk specific to rare disease: patient recruitment. A Phase 3 trial targeting 60 patients for a disease affecting 1 in 200,000 people requires extraordinary site network depth, patient registry access, and advocacy community engagement. A Tufts Center for the Study of Drug Development analysis (2022) found that rare disease Phase 2 and 3 trials have 30% more planned visits, 23% longer start-up timelines, and 19% longer treatment durations than non-orphan programs. These factors drive cost and timeline risk upward in ways that are not visible in the headline trial-size comparison.
Vaccines: the public-private imperative
Vaccine development for epidemic and pandemic-priority pathogens sits in a different economic category from either oncology or rare disease. The commercial market for a vaccine against a pathogen that may or may not cause a significant outbreak is structurally uncertain in ways that make standard return-on-investment modelling unreliable. This is not a failure of investor appetite — it is a rational response to genuine market signal uncertainty.
The consequence is that vaccine programs targeting epidemic pathogens are structurally dependent on public co-funding. Bodies including CEPI, BARDA, Innovate UK, and the Wellcome Trust exist precisely to bridge this gap — sharing development risk that private capital cannot absorb alone, and in doing so, enabling programs to proceed that would otherwise stall at the Series A stage.
The practical implication for founders is that treating public funding as a supplementary source rather than a primary structural tool is a strategic mistake. Evidence from CEPI’s COVID-19 and other programs suggests that smaller biotech companies have historically accessed more favorable partnership terms than larger incumbents. A program that enters its Series A with a CEPI or BARDA partnership already in place has materially lower effective development cost for private investors — which changes both valuation dynamics and the round size required to reach the next milestone.
The upside, meanwhile, is the probability of success data. Vaccines have the highest clinical success rate of any therapeutic area: 33.4% overall across all sponsors (Wong et al., 2019) and 39.6% for industry-sponsored programs specifically (Lo, Siah & Wong, NBER, 2020). The problem is not the biology — it is the commercial model. Solving the commercial model through public-private partnership is not a compromise on returns; it is the mechanism by which vaccine investment becomes viable at all.
Matching capital to area: a framework for decision-making
The cross-area comparison suggests three distinct investor postures, each appropriate to a different risk tolerance and capital scale.
Vaccines: patient capital with public leverage
Vaccine investment works best when private capital is layered on top of — not substituted for — public funding mechanisms. The ideal structure combines Series A and B venture capital with CEPI, BARDA, or equivalent partnership from the earliest possible stage. Founders should position their platform strategy, not their individual program, as the primary value driver: a validated platform that can generate multiple vaccine candidates from the same infrastructure is a more fundable proposition than a single-pathogen bet.
Oncology: portfolio logic or exceptional selection
Oncology investing at the early stage requires either the capital and infrastructure to build a diversified portfolio — multiple programs from a single platform, with enough shots on goal to absorb 93–97% attrition — or exceptional capability to identify the programs in the upper tail of the PoS distribution. Precision oncology approaches, including biomarker-selected trials and companion diagnostic co-development, are the most credible mechanism for improving individual program PoS above the category average. Without one of these two approaches, single-asset oncology bets should be sized accordingly.
Rare disease: the most capital-efficient entry for early-stage investors
Rare disease offers the best attrition-adjusted economics of the three areas for early-stage capital. Orphan regulatory advantages directly reduce effective development cost; phase-transition success rates are higher than oncology; and the combination of NIH NCATS funding, EU rare disease research support, and patient advocacy capital provides genuine non-dilutive funding access. The key due diligence focus should be recruitment: a program with compelling mechanistic rationale but an inadequate patient identification and site network strategy carries recruitment failure risk that can be program-ending at a scale a large-trial shortfall in oncology is not.
The bottom line
The most common mistake in cross-area comparison is treating per-program cost as the primary metric. It is not. The relevant metric is attrition-adjusted cost per approval — and on that measure, the conventional wisdom about which areas are ‘expensive’ reverses in important ways.
Rare disease is cheaper per approval than non-orphan drug development, not because the science is easier or the trials are simpler, but because the regulatory framework is designed to reflect the reality of small patient populations. Vaccines have the best probability of success of any area, but require public co-funding to be commercially viable. Oncology attracts the most capital and demands the most, in part because the commercial upside if a program succeeds is disproportionately large — but the capital required to reliably achieve one approval, at portfolio scale, is commensurately enormous.
For founders and investors navigating these choices, the starting point is not which area looks most fundable in the current market. It is which area’s risk structure, capital requirements, and funding landscape best matches the capital available and the team’s ability to execute. That is a more honest framework — and a more durable one.
Data and sources
All figures cited in this article are drawn from published peer-reviewed or institutional sources and should be independently verified before use in investment or program planning decisions. The oncology attrition-adjusted cost figure ($1–4B+) is derived by applying published PoS estimates to per-program cost ranges and is not sourced from a single publication — treat as indicative order-of-magnitude only. Key primary sources: Gouglas et al. (Lancet Global Health, 2018) for vaccine costs; Wong et al. (Biostatistics, 2019) for PoS data; Jayasundara et al. (Orphanet Journal of Rare Diseases, 2019) for orphan drug cost comparison; ASPE/HHS reports for US clinical trial cost benchmarks; Getz et al. (TIRS, 2023); Tufts CSDD (2022) for rare disease trial operational data via Applied Clinical Trials Online.