Markers predicting immune-related adverse events of atezolizumab in lung cancer

Immune-related side effects, also known as immune-related adverse events (irAEs), occur when treatments called immune checkpoint inhibitors accidentally trigger the immune system to attack healthy parts of the body. These treatments are widely used to treat non-small cell lung cancer (NSCLC), a disease where abnormal cells in the lungs grow uncontrollably.
Generally, malignant tumor cells are generated in healthy individuals every day. However, immune cells effectively maintain a healthy balance in the human body by identifying immune checkpoints, a type of receptor on the cell surface that regulate programmed cell death of tumor cells. Programmed cell death is a common and orderly cell death process determined by genes during biological development. However, this balance is disrupted in cancer patients.
Immune checkpoints mainly include programmed cell death protein 1 (PD-1)/programmed cell death ligand 1 (PD-L1) and cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) and work by boosting the body’s immune system to fight cancer and have been applied clinically. However, this immune boost can sometimes cause harmful side effects in organs like the skin, gastrointestinal tract, liver, and lungs, making it challenging to balance treatment effectiveness with patient safety.
NSCLC affects many people worldwide, and while immune checkpoint inhibitors have proven effective in improving outcomes for some patients, these immune-related side effects can be severe and unpredictable. Understanding why some patients develop irAEs and others do not is a critical part of improving cancer treatments. It has been observed that patients who experience these side effects might respond better to treatment, but it’s not clear why this happens. Research into finding reliable markers—signs that could predict who will experience these side effects—is necessary. By doing this, doctors can better tailor treatments, reducing harm to patients while still fighting the cancer effectively.
Considering that there may be numerous indicators or markers related to irAEs, it is necessary to integrate and comprehensively analyze them. Therefore, machine learning, a technique that enables computers to automatically learn and make predictions or decisions from data through algorithms and models, will be applied in our research to provide more reliable prediction accuracy and stability.
This research aims to explore the relationship between irAEs and treatment response in NSCLC. The study will involve analyzing patients who are receiving immune checkpoint inhibitors to understand which factors might predict the occurrence of these side effects. This will include looking at their medical history, immune system activity, and genetic markers. By studying these factors, the research hopes to develop better ways to identify patients who might be at risk for severe side effects, helping doctors choose the safest and most effective treatments for each individual.
The researchers will analyze this data to identify any patterns or markers and attempt to establish a clinical prediction model for irAEs through machine learning using the indicators we have discovered that could be used to predict who is likely to experience irAEs and who may have the best response to treatment.