At OncoInv we believe that people deserve affordable, accessible, and available early detection of cancer.
When cancer is found early, mortality and patient burden during treatment will be reduced. We strive to detect cancer early at low cost so it can be used everywhere in the world. OncoInv offers such a test with OncoSeek®.
How OncoSeek® works
OncoSeek® is an Machine Learning algorithm that uses a simple blood test which can detect 9 high-mortality cancer types by analysing the concentration of 7 protein tumour markers.
The algorithm analyses the specific relations between markers and clinical factors, drastically reducing false positives when compared to the standard analysis of multiple, individually judged, protein tumour markers.
The average sensitivity is 51.7% (37.1% to 77.6%) with a specificity of 92.9% (1). The test also gives an indication for the tissue of origin which guides the following diagnostic steps.
Straightforward
The lab analysis process is straightforward and can be performed on a widely available analysis platform using off the shelf reagents. The Machine Learning based technology, combined with an off the shelf analysis process, results in a very cost-effective and implementable solution.
OncoSeek® has been validated with several validation cohorts with a total of 15.000 study participants*.
OncoSeek® is a CE marked registered test.
Use cases
OncoSeek® targets symptomatic patients with suspicions for cancer in primary and ambulatory care settings.
OncoSeek® also targets at-risk populations for cancer through screening, either stand alone or as a complementary tool to existing screening.
The OncoSeek® process
The OncoSeek process optimally leverages the existing infrastructure. Blood draw can be done by any health care worker. This capacity and capability is available anywhere within reasonable travel distance for the patient. Next the blood has to be transported to a lab; this can be done using locally available transport options. When using specific tubes, no specific treatment is required during transport.
A central laboratory uses a commonly available electrochemiluminescence platform to measure the protein tumor marker values. These values are input for the Machine Learning-algorithm which runs in the cloud and is operated by OncoInv. Results are electronically returned to the treating health care worker.
Impact of multi-cancer early detection
Reduces patient mortality
Reduces treatment burden
Improves quality of life
Reduces healthcare cost
The facts
Clinical
Sample collection: 8ml of venous blood collected by medical professional.
Cancer types: Breast, Colorectal, Liver, Lung, Lymphoma, Oesophagus, Ovary, Pancreas, Stomach.
- Collectively account for ~60% of cancers deaths worldwide
- Detects aggressive cancers, e.g. pancreatic cancer
- Detects cancer that are currently not screened for
Use cases:
- Early detection for patients with elevated risk profile and symptoms
- Primary care and ambulatory care settings
Screening for populations at risk
Protein Tumour markers: AFP, CA125, CA15-3, CA19-9, CA72-4, CEA, and CYFRA 21-1
Test results: The test results are communicated via a low-, medium- or high-risk score. In case of a high-risk score, a Tissue Of Origin is indicated. The results should be interpreted by a physician.
Performance*:
- Overall sensitivity 51.7%
- Overall specificity 92.9%
Concentration values are analysed by an Machine Learning powered algorithm. Uniquely, the algorithm measures the interrelations between the different markers which generate superior results.
Operational
Lab analysis in EU on Roche Cobas. Rest of world on Roche Cobas, Abbott, GBI, BIO-Rad or equivalent. Analysis with off the shelf reagents, widely available globally.
Algorithm analysis is cloud based. Data is processed in the EU (Germany) in compliance with the EU General Data Protection Regulation.
Service delivery Specific patient and lab data is communicated to our platform; the platform will produce and share the report to a specified recipient.
Lab analysis on Roche Cobas or equivalent. Analysis with off the shelf reagents, widely available globally.
Algorithm analysis is cloud based. Data is processed in the EU (AWS in Frankfurt) in compliance with the EU General Data Protection Regulation.
Service delivery OncoSeek® can be purchased as full service including lab analysis or algorithm only, based on local context.
Costs
- Low-cost test
- First research into cost effectiveness of MCED’s indicates positive outcomes*
- Not for profit in low- & middle-income countries
Factsheet
You can download our factsheet here:
Please note that our factsheet is for professional use for resellers and distributors only. We have an EU version and a global version for the rest of the world.
Case control study
A panel of seven protein tumour markers for effective and affordable multi-cancer early detection by machine learning. Click here for a large-scale and multicentre case–control study:
Test results
The test results are communicated via a low-, medium- or high-risk score. In case of a high-risk score, a tissue of origin is indicated. The results should be interpreted by a physician.
