Research Report on the Analysis of Immunological Characteristics of Peripheral Blood in Cancer Patients

Cancer is a major and pervasive health problem globally. Despite significant advances in cancer treatment in recent years, many challenges remain, including how to accurately predict patients’ responses to various treatments. Immunotherapy, particularly immune checkpoint inhibitors (immune checkpoint blockade, ICB), has made remarkable progress over the past decade, yet most patients’ response rates remain unpredictable, often accompanied by severe immune-related side effects. Thus, a comprehensive diagnostic and consistent analytical model to assess the status of patients’ immune systems, monitor treatment response, and predict outcomes is urgently needed.

Source of This Paper

This paper, titled “comprehensive peripheral blood immunoprofiling reveals five immunotypes with immunotherapy response characteristics in patients with cancer,” was published in the journal “Cancer Cell” on May 13, 2024. The team of authors includes Daniiar Dyikanov, Aleksandr Zaitsev, Tatiana Vasileva, among others, primarily from Bostongene, Corp. (Waltham, MA, USA), as well as Thomas Jefferson University (Philadelphia, PA, USA) and The Parker Institute for Cancer Immunotherapy (San Francisco, CA, USA).

Background Introduction

The human immune system exhibits high diversity in responding to immune challenges like aging, microbial exposure, metabolic changes, and chronic diseases such as cancer. The composition of the immune system determines an individual’s ability to respond to different immunological stimuli, including anti-cancer treatment. Analysis of peripheral blood leukocytes has shown potential value in cancer treatment selection. However, systematic immune analysis is not yet routinely applied in the evaluation of cancer patients, mainly due to the lack of standardized analytical methods. This study aims to develop a clinical immunoprofiling platform using high-parameter flow cytometry to evaluate the different immune cell compositions in cancer patients and their relationship with various treatment responses.

Research Process

Research Workflow

To achieve this goal, the authors developed an immunoprofiling system combining multiparametric flow cytometry and machine learning (ML) platforms. The research involved the following major steps:

1. Sample Collection and Processing: Peripheral blood samples were collected from 408 healthy volunteers and 442 cancer patients, and processed using red blood cell removal techniques.
2. Antibody Panel Design: Designed 10 dedicated antibody panels (9 specific cell type panels and one general framework panel) to cover different immune cell subsets.
3. Flow Cytometry Analysis: Conducted multicolor flow cytometry on leukocytes isolated from peripheral blood, using antibody panels to quantitatively analyze all CD45+ cells.
4. Machine Learning Model Training: Manually labeled the flow cytometry data and used these labeled data to train gradient boosting machine learning models to automatically identify different immune cell subsets.
5. Data Validation and Analysis: Used the trained models for systematic analysis of samples from cancer patients and healthy volunteers to identify significant diagnostic or prognostic differences in immune cell composition.

Main Experimental Results

Initial comparisons between healthy volunteers and cancer patients revealed significant differences in monocytes and T cells (CD4+ and CD8+) and B cells. In further large-scale analyses, the authors identified five types of immunotypes, each with distinct cell type distributions and gene expression profiles. These immunotypes include:

1. G1 Type: Rich in naïve CD4+ T cells, CD8+ T cells, and B cells.
2. G2 Type: Exhibits a higher proportion of differentiated CD4+ central and transitional memory T cells, and CD39+ regulatory T cells.
3. G3 Type: Increased mature NK cells and PD-1+, TIGIT+ CD8+ T cells.
4. G4 Type: Includes NKT cells, terminally differentiated effector memory CD45RA+ and CD45RA- CD4+ and CD8+ T cells.
5. G5 Type: Rich in classical monocytes, HLA-DRlow monocytes, and neutrophils.

Additionally, to validate the clinical significance and practicality of these immunotypes, the authors created an immunotype-based signature score system using spectral clustering, gene expression profiling, and systematic immune response evaluations. This system was applied to a series of studies on cancer patients’ treatment responses.

Research Significance

This study developed a new clinical immunoprofiling platform by combining high-parameter flow cytometry and machine learning technologies. The research indicates that peripheral blood immunotypes not only reflect the state of the patient’s immune system but can also predict cancer patients’ responses to different treatments, such as immunotherapy. This platform can be widely applied among cancer patients, providing stratified prognosis and treatment response predictions through simple blood tests, thereby enhancing the specificity and effectiveness of clinical treatments.

Research Highlights

1. Novel Approach: Developed an efficient clinical immunoprofiling platform by combining flow cytometry with machine learning models.
2. Multiparametric Analysis: Conducted systematic immune cell typing through ten extensive antibody panels, achieving comprehensive analysis of 650 cell types and activation states.
3. Immunotype Scoring System: Developed a continuous immunotype-based signature scoring system that effectively explains the systematic immune circumstances of cancer patients’ different treatment responses.
4. Wide Applicability: Verified the universal presence and applicability of five immunotypes across various cancer types and treatment responses, demonstrating the platform’s broad clinical value.

Summary

This research systematically analyzed and identified five conserved immunotypes in the peripheral blood of cancer patients and evaluated their relationships with immunotherapy responses. The findings not only provide new perspectives and tools for personalized immunotherapy but also offer clinicians a simple and feasible method for immune detection, improving the predictive and monitoring capabilities of cancer treatment. In the future, further research and application may simplify the diagnostic and therapeutic process for cancer patients and promote the optimization and development of immunotherapy.