Title: GRAMMY – InteGRAtive analysis of tuMor, Microenvironment, immunitY and patient expectation for personalized response prediction in Gastric Cancer

Duration: 01.12.2019 – 31.05.2024

Research Area: Bioinformatics – Cancer Research

Gastric cancer is a complex disease that represents the fifth most common malignancy in the world and the third leading cause of cancer death in both sexes. It shows a high level of heterogeneity as well as a marked gender difference in incidence, with gastric cancer affecting twice as many men as women. Chemotherapy combined with surgery represents the standard of care for stages II-III gastric cancer, but the efficacy of such treatments is still limited for many patients. It is mandatory to develop novel therapeutic strategies aimed at identifying predictive markers, as well as deciphering the impact of the psychological-social and cultural environment of each patient on the outcome. ScaDS.AI Dresden/Leipzig as data analysis and machine learning partner supports the project by developing and applying methods to integrate the multi-modal data and finding predictive biomarkers for a personalized therapy of gastric cancer patients.

Aims

The EU-wide consortium aims to use various methodologies to characterize and stratify patients based on specific cellular and molecular characteristics relating to tumor biology and immune response. In addition, GRAMMY aims to evaluate the physician-patient interaction, as distress and negative coping styles can lower therapeutic compliance and negatively influence patient’s response to therapy. We aim to integrate these various data modalities and identify relevant prognostic biomarkers for therapy response.

Technology

A plethora of data will be generated and used in the study by various experimental methods (“Omics”), e.g.:

• genomic variants (NGS sequencing panel)
• immune cell population profiling (FACS, ELISA)
• pyscho-social scoring (physican interviews)

Further, GRAMMY is supported by hosting an OpenClinica server instance for secure sharing of clinical data between project partners.

For the identification of biomarkers we develop and apply appropriate statistical and machine learning approaches (clustering, survival analysis, feature selection) in combination with publicly available large data sets.

Outlook

Integration of multi-omic data is challenging but extremely promising approach towards identification of the putative links between disease-specific cellular and molecular characteristics, patient perception and prognosis. The psycho-social component of cancer treatment has previously been largely disregarded but is now increasingly recognized as an important part of successful therapy. GRAMMY will therefore provide a thus far unique and holistic description of gastric cancer.

Team

Lead

Team Members

Partners

• Dr. Chiara Molinari (IRST IRCCS – Meldola, Italy)
• Dr. Laurent Corcos (INSERM – Brest, France)
• Dr. Ghislaine Rolland-Lozachmeur (INSERM – Brest, France)
• Prof. Manousos Konstadoulakis (NKUA – Athens, Greece)