Bioinformatics-based analysis of the role of immune-related genes in acute rejection after kidney transplantation and renal cancer development

Medicine (Baltimore) 2024 AI 7 Explanations View Original
Original Paper (PDF)

Unable to display PDF. Download it here or view on PMC.

Plain-English Explanations
Pages 1-2
When the Immune System Attacks a Transplanted Kidney

Acute rejection (AR) is one of the most common and serious complications after a kidney transplant. It happens when the body's immune system recognizes the donor kidney as foreign and attacks it. Although modern medicines have reduced rejection rates from nearly 100% down to around 10%, rejection still threatens the long-term survival of the transplanted kidney.

Currently, the gold standard for diagnosing acute rejection is a biopsy, where a needle is inserted into the kidney to collect a tissue sample. This is an invasive procedure that can cause bleeding or other complications, creating a stressful experience for patients who are already managing a difficult health situation.

Researchers noticed that immune rejection and cancer both involve the same types of immune cells and genes. Immune-related genes (IRGs) help regulate how immune cells grow, change, and respond to threats. Understanding which genes are active during rejection could also reveal important information about how kidney cancer develops in transplant patients.

TL;DR: Acute rejection after kidney transplantation and renal cancer share common immune-related genes, and finding those genes could improve early detection of both conditions.
Pages 2-3
Using Bioinformatics to Find Key Immune Genes

The team used two publicly available patient gene datasets from the GEO database: one (GSE15296) to train and test their model, and another (GSE14346) to validate it externally. They began by downloading 2,483 known immune-related genes from the ImmPort database.

First, they looked for genes with different activity levels between acute rejection patients and healthy controls, finding 683 such differentially expressed genes (DEGs). They then used WGCNA (Weighted Gene Co-expression Network Analysis) to identify groups of genes that tend to work together, narrowing down to 4,568 genes that were strongly linked to acute rejection.

After overlapping these gene lists with the immune-related gene database, 21 candidate genes were found. These were then filtered using LASSO regression (a statistical shrinkage method) and the Boruta algorithm (a random forest-based feature selection tool). Single-variable and multi-variable logistic regression analyses ultimately identified 4 final feature genes: CD1D, FPR2, FAM3C, and HMOX1.

TL;DR: Researchers combined three powerful computational tools to narrow thousands of genes down to just four that are most important for diagnosing acute rejection.
Pages 6-7
Building the Best Diagnostic Model from 10 Machine Learning Approaches

Using the 4 feature genes, the researchers built and compared 10 different machine learning models including support vector machines, random forest, decision trees, neural networks, and logistic regression. Each model was tested with 10-fold cross-validation to ensure results were not due to chance.

Logistic regression stood out as the best overall approach. On the independent test dataset, it achieved an accuracy of 91.7%, an AUC (area under the curve) of 0.969, and a Matthews correlation coefficient of 0.837, all indicating excellent diagnostic performance. Its calibration curve was also the closest to ideal, meaning its predicted probabilities closely matched actual outcomes.

To help clinicians use the model in practice, the team created an interactive online tool (a Shiny web application) that allows a user to input gene expression levels and receive a prediction for whether acute rejection is likely. SHAP analysis showed that CD1D made the largest contribution to each prediction, followed closely by FAM3C.

TL;DR: Logistic regression outperformed nine other machine learning methods and achieved over 96% accuracy in detecting acute rejection from blood gene data.
Pages 7-9
FAM3C: A Shared Link Between Rejection and Kidney Cancer

Among the four genes, FAM3C attracted special interest because it had been shown to play a role in cancer progression in other organs, but its role in kidney cancer was previously unknown. Using the GSCA cancer analysis platform, the researchers examined FAM3C across the three main kidney cancer subtypes: KICH, KIRC, and KIRP.

FAM3C expression was significantly elevated in KICH (chromophobe renal cell carcinoma) tumor tissue compared to normal tissue, with a fold-change of 2.10. High FAM3C expression was also linked to poorer overall survival in KICH and KIRP patients. This suggests FAM3C may not just indicate rejection risk but also help predict kidney cancer outcomes.

On the immune side, FAM3C showed strong positive links to tumor-infiltrating lymphocytes (TILs), B-cells, and NK-cells, and a strong negative link to dendritic cells. These connections suggest FAM3C influences how the immune system responds inside the kidney, which could be important for both rejection biology and anti-tumor immunity.

TL;DR: FAM3C, one of the four key genes, is significantly different in kidney cancer tissue compared to normal tissue and may act as a biomarker for both rejection and cancer.
Pages 9-10
What the Four Genes Tell Us About Immune Biology

CD1D is a protein related to the major histocompatibility complex that helps present antigens to a specialized immune cell type called iNKT cells. Studies have shown that CD1D expression in kidney cancer is linked to more aggressive tumors and worse survival. It also plays a role during ischemia-reperfusion injury, which occurs when a transplanted kidney is reconnected to blood flow.

FPR2 is a receptor found on many cell types including immune cells and kidney cells. It plays a role in sterile inflammation and was shown to help neutrophils migrate into injured kidney tissue during ischemia-reperfusion. HMOX1 encodes an enzyme that breaks down heme, offering cell protection. It may serve as a biomarker for kidney injury and has been linked to how well immunotherapy works in advanced kidney cancer.

FAM3C's protein product, known as ILEI, promotes a process called epithelial-to-mesenchymal transition (EMT), where cells change shape and become more mobile, which is a key step in cancer spread. FAM3C has been implicated in breast, prostate, and liver cancers, and this study provides early evidence that it also matters in kidney cancer.

TL;DR: Each of the four identified genes has established roles in immune responses and kidney injury, supporting their relevance as markers of both rejection and cancer risk.
Pages 1, 9, 11
A Noninvasive Path to Early Detection

For kidney transplant patients, today's diagnostic gold standard is a biopsy, which carries real risks. A noninvasive blood test based on gene expression patterns would be a much safer and more convenient option for regular monitoring. The logistic regression model developed in this study, applied to peripheral blood samples, shows strong promise as a foundation for such a test.

The overlap between acute rejection biology and kidney cancer biology is meaningful for transplant recipients, who face an elevated risk of developing renal cell carcinoma after transplantation. Monitoring the same immune genes for both conditions could simplify follow-up care.

The researchers acknowledge that more validation is needed, including laboratory experiments and larger clinical studies, before this model could be used routinely. Future work should focus on understanding exactly how FAM3C drives both rejection and cancer, and whether targeting this gene could offer therapeutic benefit for patients facing both risks.

TL;DR: This work lays the groundwork for blood-based tests that could detect kidney transplant rejection or cancer risk without invasive biopsies.
Page 11
Key Takeaways: Immune Genes at the Crossroads of Rejection and Cancer

This study identified CD1D, FPR2, FAM3C, and HMOX1 as a powerful gene panel for diagnosing acute rejection after kidney transplantation. Among machine learning methods tested, logistic regression built from these four genes performed best, achieving over 96% AUC on held-out data.

FAM3C stands out as a particularly important gene because it appears to influence both the immune rejection process and the development of kidney cancer. It may serve as an independent prognostic factor and a potential therapeutic target. A drug or strategy that modulates FAM3C could theoretically reduce both rejection and cancer risk in transplant patients.

Ultimately, this research bridges two areas that are often studied separately: transplant immunology and cancer biology. By finding shared molecular pathways, it opens the door to more integrated approaches to caring for kidney transplant recipients.

TL;DR: Four immune-related genes, especially FAM3C, offer new diagnostic and therapeutic possibilities for kidney transplant patients at risk of rejection or cancer.
Citation: Open Access, 2024. Available at: PMC11596506.