By Wolfgang Hackl, CEO, OncoGenomX Inc., Allschwil, Switzerland
In precision oncology, nothing is more sensitive – or more valuable – than patient data.
Genomic profiles, longitudinal symptom logs, treatment responses, tumour evolution patterns, imaging signatures, real-world toxicity data, and behavioural markers power a new generation of AI-driven oncology tools.
They promise earlier detection, better prognostic models, and personalised treatments that can save lives.
But precision oncology tools – especially those positioned as “free” – sit at the crossroads of enormous benefit and enormous risk.
Cancer patients routinely provide the richest datasets in medicine.
When those data are repurposed, ambiguously shared, or monetised without patient benefit, the power imbalance becomes ethically untenable. The question is no longer abstract:
When does data use in precision oncology serve patients, and when does it exploit their vulnerability?
Cancer patient data: uniquely valuable and uniquely exposed
Cancer datasets are among the highest-value assets in the entire health sector because they combine:
- Genomic information that is uniquely identifying
- Multimodal clinical data across imaging, pathology, labs, EHR notes
- Treatment, toxicity, and survivorship trajectories
- Longitudinal biomarker evolution
- Behavioural, lifestyle and wearable signals
- Real-world evidence capturing outcomes outside controlled trials
These datasets fuel diagnostics, drug discovery, clinical decision support, and AI-enabled stratification models.
But their sensitivity means misuse carries high stakes:
- Re-identification risks even from de-identified omics data
- Insurance discrimination (e.g., inferred hereditary risk, cost-based triage)
- Targeted marketing of unproven or predatory therapies
- Profiling based on prognosis or treatment cost
- Misuse in non-medical contexts (employment, financial profiling)
Cancer patients often share data during moments of uncertainty and distress.
That context creates a moral obligation strong enough to shape policy: collection must be proportional, purpose-bound, and patient-centric.
Exploitation in “free” oncology tools: where it emerges
The precision oncology ecosystem is evolving rapidly, and with it, new forms of data exploitation:
1. Value extraction from large-scale genomics
Many free or low-cost tumour sequencing programs offset operational costs by monetising aggregated data – licensing them to pharmaceutical companies, device makers, or AI developers.
When patients are not informed that their tumour genomes may be used for future commercial applications, trust fractures.
2. AI model training without reciprocity
Oncology apps, symptom trackers, and molecular-informatics platforms often build proprietary AI models on patient-permitted data – yet the tools developed may later be sold back to hospitals, insurers, or governments at high cost.
Patients subsidise the innovation but receive little direct benefit.
3. Ambiguous consent for secondary research
Catch-all language (“data may be used for research and commercial purposes”) is common, but unacceptable in oncology, where the stakes are existential and the data nearly impossible to anonymise.
4. Ecosystem sharing with unclear boundaries
Data may pass through labs, cloud providers, analytics companies, CROs, registries, device manufacturers, and pharma partners.
Without transparency, patients cannot meaningfully consent.
5. Engagement-driven tracking
Free symptom or medication apps sometimes collect additional details beyond therapeutic relevance—location, device identifiers, behavioural patterns – feeding wider commercial profiling engines.
What cancer patients expect from data sharing
Across global surveys, cancer patients consistently express high willingness to share data – but only under conditions that protect dignity, autonomy, and fairness.
Patients expect:
Clear and specific purpose
They will share genomic or clinical data for research, drug development, or improved treatment – not for targeted advertising, actuarial scoring, corporate acquisition strategies, or indefinite storage.
Granular, ongoing control
Cancer trajectories shift. Patients want dynamic control over which datasets are shared, with whom, and for how long – especially for genomic or hereditary risk data affecting relatives.
Real-world benefits
Patients expect that insights derived from their data meaningfully return to the cancer community – more effective treatments, lower prices, transparent findings, or enhanced access.
Protection from downstream harm
No data use should increase the risk of denial, delay, or affordability barriers in treatment.
This is not a preference – it is a moral imperative.
Where precision oncology must draw the line
The line between acceptable and exploitative data use becomes clearer when we consider six essential standards:
1. Rigorous transparency
Cancer patients must understand, in plain language:
- which datasets are being collected
- what they power (AI training, drug development, risk models, etc.)
- which parties will access them
- potential commercial outcomes
Opaque or generic disclosures are ethically insufficient.
2. Data minimisation and purpose limitation
Collect only what is necessary for clinical or scientifically valid oncology applications – not broad behavioural or commercial data.
If a precision oncology app needs genomic and clinical data, it does not need:
- location history
- advertising IDs
- photo library access
- continuous device metadata
3. High-security, zero-trust architecture
Given the irreversible nature of genomic exposure, oncology requires:
- end-to-end encryption
- strict segregation of identifiable and research datasets
- independent algorithm audits
- governance committees with patient advocates
- routine permission reviews and real-time logs
4. Explicit bans on harmful secondary uses
Policies – and code – must prohibit uses that could:
- influence insurance underwriting
- enable discriminatory pricing
- support surveillance or legal exposure
- create exploitative marketing funnels
5. Fair value sharing and reciprocity
If patient-derived oncology data help develop a therapy, diagnostic, or Clinical Decision Support algorithm, the originating population should benefit.
That might include:
- affordability commitments
- reinvestment in oncology services
- open publication of findings
- patient-access programs
- discounting for systems that contributed data
This transforms the ecosystem from extractive to collaborative.
6. Demonstrated patient benefit
Precision oncology data ecosystems should be able to answer: Does this data flow improve treatment, accelerate research, or strengthen care pathways? If not, the rationale for collecting the data is weak—and likely exploitative.
The future of precision oncology depends on trust
As datasets grow more complex and AI models more influential, precision oncology stands at an ethical crossroads.
Companies that treat patient data as a co-owned asset—not a commodity – will define the field’s trajectory.
The next generation of oncology tools must:
- demonstrate patient benefit clearly
- secure genomic and clinical data at the highest level
- align incentives around outcomes rather than extraction
- return value to the cancer community
- make consent dynamic, specific, and comprehensible
Ultimately, the distinction between acceptable and exploitative data use hinges on one question:
Is this data being used to extend and improve life – or simply to extract value from those fighting to preserve it?
The precision oncology ecosystem will be judged by how it answers that question – and by how faithfully it honors the patients whose data make breakthroughs possible.
