A Framework for Democratizing Personalized Cancer Immunotherapy Through Blockchain-Verified Patient Sovereignty
Personalized mRNA neoantigen vaccines represent one of the most promising frontiers in cancer immunotherapy, with Phase 1 data demonstrating vaccine-induced T-cell responses persisting up to four years and correlating with prolonged recurrence-free survival in pancreatic cancer patients. However, the current paradigm treats patients as raw material in a value chain where they surrender their tumor tissue, neoantigen sequences, and all resulting intellectual property to pharmaceutical companies.
This whitepaper presents a patient co-ownership model enabled by Biosample NFT (BioNFT) technology—blockchain-based digital tokens that establish verifiable chain of custody, encode revocable consent, and preserve patient sovereignty from biopsy through vaccine delivery. We demonstrate the technical feasibility and regulatory pathways for enabling individual patients to retain ownership of their neoantigen sequences while accessing GMP-grade mRNA manufacturing through CDMO partnerships.
Through a detailed clinical case study of a treatment-naive patient with resectable pancreatic ductal adenocarcinoma (PDAC) harboring KRAS G12D and TP53 G266R mutations, we illustrate how this framework could provide an alternative pathway to personalized cancer immunotherapy—one where the patient is not merely a subject, but a co-owner of the therapeutic derived from their own biology.
In 2023, BioNTech and Genentech published landmark Phase 1 results demonstrating that individualized mRNA neoantigen vaccines could induce durable T-cell responses in patients with resected pancreatic ductal adenocarcinoma (PDAC)—one of the deadliest cancers with an 88% mortality rate.1 The vaccine, autogene cevumeran, showed remarkable efficacy: eight of sixteen patients developed vaccine-induced immune responses, with six of those eight remaining cancer-free at three-year follow-up.2
This represents a genuine breakthrough. However, the trial also illuminates a fundamental asymmetry in the patient-pharma relationship:
The patient is transformed from a human being with cancer into raw material in a value chain where they have zero ownership stake. Their unique genetic signature—the very mutations that make their cancer both deadly and potentially targetable—becomes the intellectual property of corporations.
The global market for mRNA cancer vaccines is projected to exceed $5-7 billion by 2030, with compound annual growth rates surpassing 30%.3 First regulatory approvals are anticipated by 2027-2029. Yet patients—the source of every neoantigen, every tissue sample, every immune response that validates these therapies—are positioned to share in none of this value.
This whitepaper proposes a fundamentally different model: one where patients retain ownership of their neoantigen sequences and chain of custody through manufacturing, enabled by blockchain-verified Biosample NFTs (BioNFTs). This model does not require pharmaceutical companies to change their practices—it creates a parallel pathway for patients who choose to exercise sovereignty over their own biology.
"The question is not whether personalized cancer vaccines work—the science increasingly suggests they do. The question is: who owns the vaccine derived from your own tumor?"
Personalized mRNA neoantigen vaccines work by teaching the patient's immune system to recognize and attack cancer cells expressing tumor-specific mutations. The process involves several key steps:
Complete workflow from tumor collection to anti-tumor immune response. The entire process takes 6-10 weeks, with manufacturing being the rate-limiting step (4-6 weeks).
Molecular mechanism: (1) LNP delivers mRNA to antigen-presenting cells; (2) Endosomal uptake and mRNA release; (3) Ribosomal translation of neoantigen peptides; (4) MHC Class I presentation on cell surface; (5) T-cell recognition, activation, tumor killing, and memory formation.
Neoantigens are peptide sequences derived from somatic mutations in tumor cells that are presented on the cell surface by major histocompatibility complex (MHC) molecules. Unlike tumor-associated antigens, which are also expressed on normal cells, neoantigens are truly tumor-specific and therefore represent ideal targets for immunotherapy.4
The success of a personalized neoantigen vaccine depends entirely on obtaining high-quality tumor tissue from the patient's surgical specimen. There is no substitute. The tumor tissue provides:
Without the patient's own tumor tissue, no personalized neoantigen vaccine can be manufactured. This is why the Whipple surgical specimen represents an irreplaceable asset that the patient must retain rights to.
The identification process requires:
Multiple computational tools are available for neoantigen identification and prioritization:
| Tool/Pipeline | Function | Key Capability |
|---|---|---|
| pVACtools | End-to-end neoantigen pipeline | Integrates multiple predictors; includes pVACseq, pVACfuse, pVACview |
| NetMHCpan 4.1 | MHC Class I binding prediction | Gold standard; pan-allele prediction |
| NetMHCIIpan 4.0 | MHC Class II binding prediction | CD4+ helper T-cell epitopes |
| MHCflurry 2.0 | ML-based binding prediction | 7,000+ predictions/sec; includes antigen processing |
| nextNEOpi | Nextflow comprehensive pipeline | SNVs, indels, gene fusions; Docker/Singularity support |
| IEDB Tools | Epitope database + prediction | Immunogenicity prediction; validated epitopes |
| OptiType | HLA typing from NGS | 4-digit HLA-A/B/C typing from WES/RNA-seq |
These tools are publicly available and can be deployed by qualified bioinformatics teams, enabling patient-directed neoantigen discovery independent of pharmaceutical company pipelines.
Once neoantigens are identified, mRNA sequences encoding these peptides are designed and synthesized. Key technical considerations include:
| Component | Function | Optimization Strategy |
|---|---|---|
| 5' Cap | Protects mRNA, enables ribosome binding | CleanCap technology (Cap 1 structure) |
| 5' UTR | Translation initiation | Optimized Kozak sequence |
| Coding Sequence | Encodes neoantigen(s) | Codon optimization, N1-methylpseudouridine |
| 3' UTR | mRNA stability | AES and mtRNR1 sequences |
| Poly(A) Tail | Stability and translation | ~120 nucleotides |
| Delivery Vehicle | Cellular uptake | Lipid nanoparticles (LNPs) |
Upon injection, mRNA is taken up by antigen-presenting cells, translated into neoantigen peptides, and presented on MHC molecules. This triggers both CD8+ cytotoxic T-cell responses (which directly kill tumor cells) and CD4+ helper T-cell responses (which coordinate the immune attack).
The most compelling evidence for personalized mRNA neoantigen vaccines comes from the Phase 1 trial of autogene cevumeran in resected pancreatic cancer:1,2
Critically, 98% of the T cells targeting individual neoantigens were de novo—meaning they were not detectable before vaccination. The vaccine created entirely new immune responses against the patient's specific tumor mutations.
While fully personalized vaccines target patient-specific neoantigens, "off-the-shelf" vaccines targeting common oncogenic mutations represent a parallel development track. KRAS mutations, present in >90% of pancreatic cancers, are particularly attractive targets:
| Vaccine Program | Target | Phase | Key Results |
|---|---|---|---|
| ELI-002 (Elicio) | KRAS G12D, G12R | Phase 1/2 | 100% T-cell response at highest dose; CA19-9 reduction in 84% |
| mKRAS-VAX (MSK) | 6 KRAS mutations | Phase 1 | Immune response to additional KRAS mutations beyond vaccine targets |
| mRNA-5671 (Moderna) | KRAS G12D, G12V, G13D, G12C | Phase 1 | Combined with pembrolizumab; results pending |
The patient case study presented below represents an ideal candidate for both approaches: personalized neoantigen vaccination and KRAS G12D-targeted therapy.
This case study is based on a composite of clinical data and represents a typical patient who would be eligible for personalized neoantigen vaccine therapy.
| Parameter | Value |
|---|---|
| Age | 49 years |
| Sex | Male |
| Diagnosis | Adenocarcinoma of the pancreatic head |
| Stage | Borderline resectable (pending final staging) |
| Treatment Status | Treatment-naive |
| Prior Therapy | None |
Fine-needle biopsy of the pancreatic head lesion confirmed adenocarcinoma. The specimen was reviewed at intradepartmental conference with concurrence on the diagnosis.
| Gene | Variant | Protein Alteration | Exon | Variant Frequency | Interpretation |
|---|---|---|---|---|---|
| KRAS | c.35G>A | p.G12D | 2 | 34% | Pathogenic |
| TP53 | c.796G>A | p.G266R | 8 | 37% | Pathogenic |
| Biomarker | Result | Clinical Significance |
|---|---|---|
| Microsatellite Instability (MSI) | Stable | Not eligible for pembrolizumab monotherapy |
| Tumor Mutational Burden (TMB) | Low (5 mut/Mb) | Below threshold for immunotherapy benefit |
| PD-L1 (22C3) | TPS: 1% | Low expression |
| Genomic LOH | Equivocal (12%) | Below threshold for HRD |
| MHC Class I | Allele 1 | Allele 2 |
|---|---|---|
| HLA-A | A*02:01 | A*29:02 |
| HLA-B | B*38:01 | B*51:08 |
| HLA-C | C*12:03 | C*16:02 |
HLA-A*02:01 is one of the most common HLA alleles globally and is well-characterized for neoantigen prediction. Multiple KRAS G12D-derived peptides have been validated to bind this allele with high affinity, making this patient an excellent candidate for both personalized and off-the-shelf KRAS-targeted vaccines.
Result: NEGATIVE
No pathogenic or likely pathogenic germline variants were identified in 62 genes associated with hereditary cancer syndromes, including BRCA1, BRCA2, PALB2, ATM, and MLH1/MSH2/MSH6/PMS2.
This confirms the somatic (tumor-only) origin of the KRAS and TP53 mutations—they arose in the tumor and are not inherited.
| Finding | Detail |
|---|---|
| Primary Tumor | 1.7 x 1.5 cm hypermetabolic mass in pancreatic head; SUVmax 3.7 |
| Vascular Involvement | None (celiac axis, SMA, CHA, MPV, SMV all uninvolved) |
| Distant Metastases | None (lungs, bones, abdominal/pelvic organs clear) |
| Lymph Nodes | Non-FDG avid subcentimeter peripancreatic nodes (likely reactive) |
| Liver | Indeterminate hypodensity in segment 8 (non-FDG avid; recommend MRI) |
Surgically Resectable: The absence of vascular involvement (arterial or venous) and distant metastases indicates this tumor is potentially resectable with curative intent. This patient would be eligible for upfront surgery followed by adjuvant therapy—the same treatment paradigm used in the BioNTech Phase 1 trial.
Based on the comprehensive molecular and clinical profile, this patient demonstrates multiple features that make them an ideal candidate for personalized neoantigen vaccination:
| Criterion | Status | Significance |
|---|---|---|
| Resectable disease | Yes | Enables adjuvant vaccine approach post-surgery |
| Identified driver mutations | KRAS G12D, TP53 G266R | Validated neoantigen targets |
| Favorable HLA type | HLA-A*02:01 | Well-characterized for neoantigen binding |
| No germline mutations | Confirmed | Somatic mutations are tumor-specific targets |
| Good performance status | Treatment-naive | Intact immune system for vaccine response |
| Adequate tissue available | Yes (surgical specimen anticipated) | Enables comprehensive sequencing |
The pancreatic head tumor removed during the Whipple procedure is the sole source of material for creating a personalized neoantigen vaccine. This tissue provides:
| Material | Purpose | Why It's Critical |
|---|---|---|
| Tumor DNA | Identify somatic mutations (KRAS G12D, TP53 G266R, etc.) | Reveals all mutations present in the tumor |
| Tumor RNA | Confirm which mutations are EXPRESSED | ESSENTIAL: Only expressed mutations produce targetable neoantigens. A silent mutation cannot be attacked by T-cells. |
| Normal tissue (margin) | Distinguish somatic vs. germline variants | Ensures vaccine targets tumor-specific mutations only |
Consider this critical distinction:
Without RNA from the patient's own tumor, you cannot determine which mutations are expressed. Without expression data, you cannot design an effective vaccine.
This is why tissue handling at the time of surgery is critical. The tumor specimen must be:
The Whipple surgical specimen is not just tissue—it is the irreplaceable biological source code for the patient's personalized cancer vaccine.
Following surgical resection (Whipple procedure), there is typically a 6-12 week recovery period before adjuvant chemotherapy begins. This window represents the critical opportunity for personalized vaccine manufacturing and initial dosing—the same approach used in the BioNTech trial.
Timeline assumes 4-6 week CDMO manufacturing turnaround. BioNTech achieved similar timelines in their Phase 1 trial.
Under current practices, patients are disconnected from their donated biospecimen with little to no visibility into how their contributions are being used. Once tissue leaves the patient's body, it enters a chain of custody controlled entirely by healthcare institutions and their commercial partners. The patient loses all rights to:
BioNFTs (Biosample Non-Fungible Tokens) are blockchain-based digital tokens that establish verifiable, patient-controlled chain of custody for biospecimens and derived data. The technology is protected by two granted U.S. patents:5,6
| Patent Number | Title | Key Innovation |
|---|---|---|
| US-11,984,203-B1 | Family Vault for Biological Data | Hierarchical consent management for family genomic data |
| US-11,915,808-B1 | BioNFT Technology | Non-fungible tokens for biosample ownership and consent |
The BioNFT architecture integrates with Story Protocol for IP management and uses BioFS (GenoBank's GDPR-compliant file system built on AWS S3) for secure, deletable storage. Unlike IPFS, BioFS supports the "right to erasure" required by GDPR Article 17.
IPFS is immutable—once data is pinned, it cannot be deleted. This violates GDPR Article 17 (Right to Erasure). If a patient revokes consent, their genomic data MUST be deletable.
BioFS uses AWS S3 with AES-256 encryption, providing:
| Component | Technology | Function |
|---|---|---|
| Blockchain | Avalanche C-Chain / Story Protocol | Immutable record of ownership and consent |
| Smart Contracts | Solidity (ERC-721) | Automated consent enforcement, royalty distribution |
| Data Storage | AWS S3 (AES-256 encrypted) | GDPR-compliant deletable storage (not IPFS) |
| Access Control | Bloom Filters | Privacy-preserving permission checks |
| Patient Interface | BioWallet (MetaMask-compatible) | Non-custodial wallet for managing BioNFTs |
Unlike traditional informed consent (a static document signed once), BioNFT enables "Metamorphic Consent"—consent that transforms from a static permission into an ongoing economic relationship. Key features:
In the context of patient co-owned mRNA neoantigen vaccines, BioNFTs serve as the legal and technical infrastructure for maintaining patient sovereignty throughout the manufacturing process:
The patient maintains ownership of the BioNFT throughout the process. The CDMO provides manufacturing services but does not acquire IP rights to the neoantigen sequences.
The BioNFT approach operates within existing legal frameworks while establishing clearer patient rights:
Several contract development and manufacturing organizations (CDMOs) have established capabilities for GMP-grade mRNA synthesis. Leading candidates for patient co-owned vaccine manufacturing include:
| CDMO | Key Technology | GMP Capacity | Estimated Timeline |
|---|---|---|---|
| TriLink BioTechnologies | CleanCap mRNA capping | 1g to >100g per batch | 4-6 weeks |
| Aldevron | Plasmid DNA, mRNA | Clinical through commercial | 6-8 weeks |
| Catalent | LNP formulation | Large-scale fill/finish | 8-12 weeks |
TriLink BioTechnologies, part of Maravai LifeSciences (NASDAQ: MRVI), has emerged as a leading CDMO for mRNA therapeutics. Key capabilities relevant to patient co-owned vaccines:
TriLink opened a 32,000 square foot cGMP manufacturing facility in Sorrento Valley, San Diego, specifically designed for mRNA manufacturing. Features include:
The patient co-owned model proposes a service relationship with CDMOs where:
| Component | Estimated Cost | Timeline |
|---|---|---|
| Comprehensive tumor sequencing (WES + RNA-seq) | $5,000 - $10,000 | 2-3 weeks |
| Neoantigen identification and prioritization | $5,000 - $15,000 | 1-2 weeks |
| mRNA sequence design | $5,000 - $10,000 | 1 week |
| GMP mRNA synthesis (single patient batch) | $50,000 - $100,000 | 4-6 weeks |
| LNP formulation (if separate) | $20,000 - $40,000 | 2-3 weeks |
| Quality control and release testing | $10,000 - $20,000 | 1-2 weeks |
| Total Estimated Range | $95,000 - $195,000 | 8-14 weeks |
For context, BioNTech's autogene cevumeran is estimated to cost over $100,000 per patient in clinical trial settings. The patient co-owned model achieves comparable costs while preserving patient IP rights.
Patient co-owned personalized vaccines can access treatment through several regulatory mechanisms:
| Pathway | Key Requirements | Timeline | Best Suited For |
|---|---|---|---|
| Clinical Trial | FDA IND approval; IRB oversight | 6-12+ months | Systematic data collection |
| Physician-Sponsored IND | Individual IND; FDA 30-day review | 1-2 months | Individual patient access with FDA oversight |
| Right to Try | No FDA review; drug must have completed Phase 1 | Days to weeks | Urgent cases; limited liability protection |
| Expanded Access | FDA and IRB approval; manufacturer agreement | Weeks | Serious conditions outside trials |
The federal Right to Try Act (2018) provides a pathway for terminally ill patients to access investigational drugs that have completed Phase 1 clinical trials. Key requirements:
A critical limitation: Right to Try requires the investigational drug to have completed a Phase 1 clinical trial. A truly patient-specific vaccine (unique neoantigen sequences) has, by definition, never been tested in a trial. This pathway may be more suitable for "off-the-shelf" KRAS-targeted vaccines that have completed Phase 1 testing.
For patient co-owned personalized vaccines, a physician-sponsored IND (Investigational New Drug application) represents the most robust regulatory approach:
For the physician-sponsored IND, the CDMO would need to provide:
The FDA has shown flexibility in regulating personalized cancer therapies:
The agency has indicated willingness to work with sponsors on novel personalized medicine approaches that do not fit traditional drug development paradigms.
| Stakeholder | Current Model | Patient Co-Owned Model |
|---|---|---|
| Pharmaceutical Company | 100% of neoantigen IP; 100% of commercial value | 0% of neoantigen IP; manufacturing service revenue |
| CDMO | Manufacturing fee (contracted by pharma) | Manufacturing fee (contracted by patient/provider) |
| Healthcare Provider | Clinical care fees; trial participation fees | Clinical care fees; IND sponsorship fees |
| Patient | $0 ownership; experimental subject | 100% of neoantigen IP; treatment recipient and owner |
If personalized neoantigen vaccines prove effective and become standard of care, patient-owned neoantigen sequences could have significant economic value:
Pharmaceutical companies developing next-generation cancer vaccines may seek access to validated neoantigen-response pairs. Patients with documented vaccine responses could license their sequence data for research.
Machine learning models for neoantigen prediction require training data linking sequences to clinical outcomes. Patient-owned data could be licensed for AI development with appropriate compensation.
Longitudinal biospecimen collections (blood draws, tumor samples) from vaccine-treated patients represent valuable research resources that patients could contribute on their own terms.
The patient co-ownership model addresses these concerns through:
Personalized mRNA neoantigen vaccines represent a transformative opportunity in cancer treatment. Phase 1 data demonstrates that these vaccines can induce durable, tumor-specific immune responses that correlate with prolonged survival in pancreatic cancer—one of the deadliest malignancies.
The technology works. The question is: who owns the vaccine derived from your own tumor?
The BioNFT-enabled patient co-ownership model offers an alternative paradigm where:
For patients with resectable pancreatic cancer who wish to pursue a co-owned neoantigen vaccine approach:
"The broader vision is a platform where patients retain ownership of their neoantigen sequences and chain of custody through manufacturing—essentially democratizing what pharmaceutical companies charge $100K+ to deliver through traditional channels."
This is not an anti-pharma position. Pharmaceutical companies have developed remarkable science that is saving lives. But patients deserve the choice to participate in that value creation as owners, not merely as raw material.
The technology exists. The regulatory pathways exist. The manufacturing capabilities exist. What remains is the will to build a system that respects patient sovereignty over their own biology.
Based on the clinical case study molecular profile, the following represent potential neoantigen targets:
| Mutation | Peptide Sequence (Example) | HLA Restriction | Predicted Binding Affinity |
|---|---|---|---|
| KRAS G12D | KLVVVGADGV | HLA-A*02:01 | Strong (IC50 < 50 nM) |
| KRAS G12D (long) | MTEYKLVVVGADGVGKSALTI | Multiple Class II | Validated immunogen |
| TP53 G266R | To be predicted | HLA-A*02:01, A*29:02 | Requires validation |
Note: Actual peptide sequences would be determined through comprehensive neoantigen prediction pipeline including NetMHCpan, MHCflurry, and immunogenicity filtering.
| Attribute | Specification | Test Method |
|---|---|---|
| Identity | Sequence confirmed | RT-PCR sequencing |
| Purity (% full-length) | >90% | Capillary electrophoresis |
| Capping efficiency | >95% | LC-MS |
| dsRNA content | <1% | ELISA or dot blot |
| Endotoxin | <10 EU/mL | LAL assay |
| Sterility | No growth | USP <71> |
| Residual DNA | <10 ng/mg mRNA | qPCR |
Document Classification: Confidential - For Discussion Purposes
Version: 1.0 | Date: January 2026
© 2026. All rights reserved.
This document is intended for discussion with potential CDMO partners and does not constitute medical advice. All treatment decisions should be made in consultation with qualified healthcare providers.