HERCULES · Project

Predicting the Right Drug Combination for Ovarian Cancer Patients Using Single-Cell Analysis

healthPrototypeTRL 4

Ovarian cancer is deadly because tumors aren't made of just one type of cell — they're a messy mix, and different cells respond to different drugs. HERCULES took tumor samples from patients, sorted them cell by cell, and tested which drug cocktails actually kill each cell type. Think of it like figuring out that a weed in your garden has five different root systems, each needing a different herbicide — so you mix the right combination instead of guessing. The end goal was a diagnostic kit that tells doctors which drug mix will work best for each individual patient.

By the numbers

40,000

Women dying from ovarian cancer in Europe every year

Consortium partners

Countries in consortium (FI, FR, IT, SE, UK)

Total project deliverables

The business problem

What needed solving

Ovarian cancer kills more than 40,000 European women annually, largely because doctors lack reliable ways to predict which drug combinations will work for each patient. Current treatment relies on trial-and-error, wasting time and money on ineffective therapies while the disease progresses. The diagnostics and pharmaceutical industries need better tools to stratify patients and predict treatment response before starting expensive regimens.

The solution

What was built

The project built and evaluated flow cytometry and imaging-based clonal drug response assays for analyzing tumor subpopulations at single-cell level. It also developed computational network models that predict effective drug combinations based on single-cell genetic and transcriptomic data from ovarian cancer patients, with 10 deliverables produced in total.

Audience

Who needs this

Companion diagnostics companies developing cancer treatment selection testsPharmaceutical companies running ovarian cancer combination therapy clinical trialsPrecision medicine platforms building oncology decision support toolsClinical laboratories offering advanced tumor profiling servicesBiotech startups commercializing single-cell analysis technologies

Business applications

Who can put this to work

Diagnostics & Companion Diagnostics

mid-size

Target: Diagnostic kit manufacturers or companion diagnostics companies

If you are a diagnostics company struggling to develop tests that predict which cancer treatments will actually work — this project developed flow cytometry and imaging-based clonal drug response assays that can identify tumor subpopulations and match them to effective drug combinations. The approach was validated on high-grade serous ovarian cancer, which kills more than 40,000 women in Europe every year, representing a significant unmet diagnostic need.

Pharmaceutical & Drug Development

enterprise

Target: Pharma companies with oncology pipelines focused on combination therapies

If you are a pharmaceutical company running expensive clinical trials for ovarian cancer drug combinations without knowing which patient subgroups will respond — this project built computational models that predict effective combinatorial treatments based on single-cell genetic and transcriptomic data. This could help you design better stratified clinical trials and reduce the cost of testing inefficient drug combinations.

Digital Health & Precision Medicine

SME

Target: Precision medicine platforms or health-tech companies building treatment decision tools

If you are a health-tech company building clinical decision support tools and need better algorithms for matching cancer patients to treatments — this project generated single-cell datasets and network models that predict optimal drug combinations for individual patients. The approach moves beyond trial-and-error clinical assessment toward systematic prediction, which could power your treatment recommendation engine.

Frequently asked

Quick answers

What would it cost to license or adopt this technology?

The project aimed to commercialize a predictive biomarker kit, but specific pricing or licensing terms are not available in the project data. The consortium includes 1 industry partner and 1 SME, suggesting some commercial pathway was explored. Interested companies should contact the coordinator at University of Helsinki to discuss licensing.

Can this scale to industrial or clinical use?

The project used prospectively and longitudinally collected fresh samples from multiple anatomic sites, indicating clinical-grade sample handling. However, the primary output is a predictive model and assay evaluation — scaling to routine clinical diagnostics would require regulatory approval and manufacturing partnerships beyond what the project delivered.

What is the IP situation?

Based on available project data, the consortium of 10 partners across 5 countries would share IP according to their grant agreement. The objective explicitly mentions commercializing a predictive biomarker kit, so IP protection was likely pursued. Specific patent filings are not disclosed in the available data.

How close is this to being used in hospitals?

The project ran from 2016 to 2021 and produced evaluations of flow cytometry and imaging-based drug response assays. Key results were validated with prospective and retrospective cohorts and in vivo models. However, routine clinical deployment would still require regulatory clearance and clinical trial validation beyond the project scope.

Does this work only for ovarian cancer or can it be applied elsewhere?

The project specifically targeted high-grade serous ovarian cancer, but the single-cell analysis approach and combinatorial drug prediction models could in principle be adapted to other solid tumor types. The methodology — sorting cells, testing drugs, building network models — is not inherently limited to one cancer type.

What data and tools came out of this project?

The project produced 10 deliverables including evaluation of flow cytometry and imaging-based clonal drug response assays. It also generated single-cell genetic and transcriptomic datasets, ex vivo drug screening results, and network models for predicting effective drug combinations. These represent valuable research assets for companies in precision oncology.

Consortium

Who built it

The HERCULES consortium brings together 10 partners from 5 European countries (Finland, France, Italy, Sweden, UK), led by the University of Helsinki. The group is heavily research-oriented: 5 universities and 4 research organizations, with only 1 industry partner and 1 SME (10% industry ratio). This composition is strong for producing high-quality science but signals that commercial translation would need external business partners. A diagnostics or pharma company looking to adopt these results would likely need to bring its own manufacturing, regulatory, and distribution capabilities to the table.

HELSINGIN YLIOPISTOCoordinator · FI
AB ANALITICA SRLparticipant · IT
THE CHANCELLOR MASTERS AND SCHOLARS OF THE UNIVERSITY OF CAMBRIDGEparticipant · UK
UNIVERSITA DEGLI STUDI DI TRIESTEparticipant · IT
VARSINAIS-SUOMEN SAIRAANHOITOPIIRIN KUNTAYHTYMAparticipant · FI
ISTITUTO SUPERIORE DI SANITAparticipant · IT
CENTRO DI RIFERIMENTO ONCOLOGICO DI AVIANOthirdparty · IT
INSTITUT PASTEURparticipant · FR
KAROLINSKA INSTITUTETparticipant · SE
TURUN YLIOPISTOparticipant · FI

How to reach the team

The coordinator is University of Helsinki (Finland). Use SciTransfer to get a warm introduction to the research team.

Order a report on this consortium See HELSINGIN YLIOPISTO profile →

Next steps

Talk to the team behind this work.

Want to explore licensing the biomarker kit technology or accessing the drug combination prediction models? SciTransfer can connect you directly with the HERCULES research team and help structure a collaboration.

Order a report on this topic More in Health & Biomedical