Презентация английский

1.

Institute of personalized oncology World-Class Research Center
"Digital biodesign and personalized healthcare", Sechenov First
Moscow State Medical University
Development of a predictive model of response to
immunotherapy in non-small cell lung cancer
Varvara D. Sanikovich
MD, Medical Oncology Department
Junior Research Assistant
Institute of Personalized Oncology
Sechenov University

2.

RELEVANCE
Lung cancer remains the leading cause of cancer deaths in 2023 and
represents a major public health problem.
With the advent of immunotherapy, incredible advances have been
made in the treatment of patients with a wide range of malignancies,
particularly in patients with non-small cell lung cancer.
However, only 17 to 48% of patients responding to therapy
Existing predictors of tumor response to immunotherapy (tissue
biomarkers) are not accurate enough to predict the efficacy
Some patients experience immune-mediated side effects
The unique mechanisms of action of immunotherapy drugs have led
to the emergence of new, atypical response patterns, that do not
meet the standard criteria for assessing tumor response

3.

RADIOMICS
a new approach of analysing imaging study data that uses quantitative
features extracted from medical images to predict treatment response
and clinical outcomes of patients. These radiomic features are further
used to create statistical models to individualise the diagnosis and
treatment of diseases of different organs and systems.

4.

RADIOMICS
Radiomic features extracted from medical images may provide a non-invasive
way to obtain prognostic information and identify patients who are more likely
to respond to immunotherapy.
Molecular genetic profile
of the tumour
Predicting response to
therapy
Predicting the
development of adverse
events
Visual biomarkers

5.

RADIOMICS predictive model
The development of a predictive model of response to
immunotherapy in non-small cell lung cancer using artificial
intelligence represents significant scientific and practical
interest and demonstrates the relevance of the chosen area of
research.
The development of personalized medicine is one of the main
objectives by the Strategy for the development of healthcare in
the Russian Federation, approved by the Decree of the President
of the Russian Federation in 2019.

6.

RESEARCH OBJECTIVES AND GOALS
GOAL: Development of an artificial intelligence-based model that predicts the
response to immunotherapy in non-small cell lung cancer based on tumor
radiomic characteristics and patient clinical data.
OBJECTIVES:
1.
2.
To analyse the radiomic characteristics of CT images of lung cancer, to highlight relevant
characteristics
To create and train a statistical model based on radiomic features and their relationship
with immunotherapy response patterns (complete response, partial response/stabilisation,
progression, pseudoprogression, hyperprogression)
3.
4.
5.
6.
Evaluate the predictive value of the model when adding clinical characteristics, tissue
biomarkers
Test the model on a control group
Statistical analysis
Evaluate the likely advantage of the radiomic model over existing predictive biomarkers

7.

CONCLUSION
Enhanced Predictive Accuracy
AI-driven radiomics significantly improves the accuracy of predicting treatment responses by
identifying subtle imaging patterns
Personalized Treatment Strategies
The model enables personalized treatment plans tailored to individual tumor characteristics,
optimizing therapeutic efficacy
Non-Invasive Assessment
Radiomics allows for non-invasive evaluations, reducing the need for invasive procedures like
biopsies and enhancing patient comfort
Improved Patient Stratification
AI models enhance patient stratification, identifying those most likely to benefit from
immunotherapy and guiding clinical decisions
Addressing Clinical Challenges
By providing a quantitative analysis framework, radiomics reduces variability in imaging
interpretation and enhances the reliability of prognostic assessments