OwkinHepatocellular Carcinoma (HCC)
Discover Owkin’s commitment to advancing research in HCC
Hepatocellular carcinoma (HCC) is the most common form of primary liver cancer, and the fourth cause of cancer-related death worldwide.
Characterized by an aggressive clinical course, approximately two-thirds of patients are diagnosed in advanced stages of disease, at which point only limited therapeutic options exist. Furthermore, even those who receive seemingly curative treatments at earlier stages (i.e., surgical resection or ablation) face high tumor recurrence rates (up to 70-80%).
With identification and greater understanding of prognostic biomarkers, available therapies can be better directed and novel approaches can be efficiently explored, providing hope to patients.
Owkin HCC Loop
The challenge: how can we effectively learn from the millions of siloed patient data points without compromising data privacy and protection?
Owkin is federating a unique Research Ecosystem to bring academic and pharmaceutical industry researchers together. Here, the company’s proprietary infrastructure and artificial intelligence (AI) technologies enable researchers to train machine learning (ML) models on distributed data at scale across multiple medical institutions without centralizing the data. Owkin safeguards patient data and privacy. All data remains protected and adheres to data governance frameworks: data does not move, only algorithms travel through remote access and execution.
In these institutions, we focus our work on accessing fit-for-AI, high-quality research-grade HCC cohorts, with access to rich raw data: radiology, digital pathology, genetics, and longitudinal clinical follow-ups, curated by the best medical experts in their fields. Knowing the hidden secrets and biases within a dataset, asking the right question, providing annotations and interpretations of the results: these skills are as important as the data itself.
Preview Owkin data access in HCC:
32 datasets – 8.7k patients
Owkin Research in HCC
Owkin has developed a strong expertise in HCC through collaborations with leading academic institutions.
- Prediction of prognosis from whole slide images – read about HCCInterpret and HCCExpert models in Hepatology
- Liver segmentation & lesion detection from ultrasounds – see Owkin’s paper following its winning entry in the JFR 2018 competition
- Prediction of response to treatment – ongoing projects leveraging radiomics on MRI & CT-Scan data
Focus on HCCInterpret and HCCExpert: To train the models we used a dataset from the Henri-Mondor Hospital of 390 histology slides from 194 patients with resected HCC. The models were validated using an independent dataset of 328 histology slides from The Cancer Genome Atlas (TCGA). Both models significantly outperformed existing survival scores that incorporate clinical, biological, and pathological prognostic markers.GET IN TOUCH Explore HCCExpert live in Studio
Owkin Drug Development Applications in HCC
Owkin’s data access, models, and know-how built within our HCC Loop can bring value to pharmaceutical companies by enhancing all stages of their drug development.
HCCInterpret leads to the identification of biomarkers of survival. This can enable pharmaceutical companies to narrow down druggable targets for HCC and enhance the HCC drug development process.
HCCExpert improves clinical trial efficiency. We simulated the impact of the model applied to the patient selection process of phase 3 clinical trials. By adjusting for the model’s prognostic scores, as covariates, the number of patients required to meet the trial endpoints and determine drug efficacy is materially reduced. This leads to cost savings of ~15%, and to a 1-year gain in the development process, accelerating the drug approval process.
Hear from our Key Opinion Leaders
“There is a real need for these models to improve therapeutic strategies and to develop adjuvant therapies for HCC. These models performed better than any known clinical, biological or pathological prognostic markers and provided novel insights on the biological features related to the aggressiveness of the tumor.”
Dr. Julien Calderaro – Clinical Pathologist at Henri Mondor Hospital, AP-HP/Greater Paris University Hospitals
“Prognostic markers of response to trans-arterial chemoembolization (TACE) in liver cancer are not well known. We are collaborating with Owkin to apply machine learning techniques to multimodal data from a large curated research cohort. We hope that these novel approaches will improve our understanding of the disease and we expect to use this knowledge to personalize & improve TACE treatments.”
Pr. Valerie Vilgrain – Chief of Radiology at Beaujon Hospital, AP-HP/Greater Paris University Hospitals