This 2025 study investigated the connection between macrophages (a type of immune cell) and programmed cell death (PCD) in kidney renal clear cell carcinoma (KIRC), the most common and deadly form of kidney cancer, representing about 70% of all kidney cancer cases.
Kidney cancer is particularly dangerous because it rarely causes symptoms in early stages. Up to 30% of patients already have cancer that has spread to other organs by the time they first see a doctor. This makes accurate prognostic tools essential for guiding treatment decisions.
The researchers aimed to create a new scoring system, called the MacPCD (Macrophage-associated Programmed Cell Death) model, that could predict individual patient survival by analyzing how macrophage behavior and cell death gene patterns interact within kidney tumors.
The study analyzed data from 693 kidney cancer samples collected from four databases: TCGA (522 samples), E-MTAB-1980 (101 samples), GSE22541 (39 samples), and GSE29609 (68 samples). Using multiple independent datasets is a rigorous approach that tests whether findings hold up beyond a single institution.
Researchers identified 1,548 PCD-related genes across 18 different types of programmed cell death, including apoptosis, autophagy, necroptosis, ferroptosis, pyroptosis, and others. They also gathered 863 macrophage-associated genes from a cell marker database. This provided a rich molecular landscape to search for meaningful patterns.
To build the MacPCD model, the team tested 10 machine learning algorithms and 101 unique combinations, including support vector machines (SVM), LASSO, gradient boosting (GBM), random forests, and CoxBoost, among others. The best-performing combination was selected based on its consistency across all four patient cohorts.
Programmed cell death (PCD) is the body's built-in system for eliminating cells that are damaged, infected, or no longer needed. Unlike accidental cell death caused by injury, PCD is an orderly, controlled process governed by specific genes and molecular signals.
In healthy tissue, PCD maintains the balance between cell birth and cell removal. When PCD malfunctions in cancer, tumor cells can evade their natural death signals, surviving and multiplying when they should be eliminated. This is a core reason why cancers are so difficult to treat.
Researchers now recognize 18 distinct forms of PCD, including apoptosis (classical cell suicide), ferroptosis (iron-dependent cell death), and pyroptosis (inflammatory cell death), among others. Different cancer types exploit different PCD pathways, and understanding which pathways are disrupted in kidney cancer opens new avenues for targeted treatment.
The final MacPCD model contains 6 genes. Three of them, BID, SLC25A37, and BNIP3L, were found to be highly expressed in tumor tissue compared to normal kidney tissue. Three others, ACSL1, SDHB, and ALDH2, were more highly expressed in normal tissue, meaning they appear to be suppressed or lost in cancer cells.
BID and BNIP3L are both involved in apoptosis (programmed cell death). Their overexpression in tumors may represent a failed or altered cell death response. SLC25A37 is a mitochondrial iron transporter linked to ferroptosis, a form of cell death dependent on iron and oxidative stress.
ACSL1 and ALDH2 are involved in fatty acid metabolism and alcohol/aldehyde metabolism respectively, while SDHB is a mitochondrial enzyme crucial for energy production. The loss of these three genes in kidney cancer tissue reflects the broad metabolic disruptions that characterize this disease.
The MacPCD model achieved an area under the curve (AUC) of 0.920, meaning it correctly predicted survival outcomes in 92% of cases. This is one of the highest reported values for a kidney cancer prognostic model based on molecular data.
Crucially, MacPCD outperformed all standard clinical variables tested, including patient age, cancer stage (Stage), and distant metastasis status (M stage), which are the factors doctors currently rely on most. A molecular model beating these established clinical variables is a strong indication of its added value.
Patients were divided into high and low MacPCD groups based on their score. Those in the high MacPCD group showed significantly worse overall survival, higher tumor mutation burden (more genetic changes in the cancer), and greater immune cell infiltration with high expression of immune checkpoint genes that help tumors evade immune attack.
Macrophages are among the most abundant immune cells found within kidney tumors. In the tumor microenvironment, macrophages can either attack cancer cells or, when reprogrammed by the tumor, switch to a mode that actually supports cancer growth and suppresses other immune responses.
The high MacPCD group showed elevated immune cell infiltration alongside high expression of immunomodulators, molecules that regulate immune activity. This paradox, more immune cells but worse outcomes, suggests that macrophages in these tumors are functioning in a tumor-supporting rather than tumor-fighting mode.
This finding has practical implications: patients in the high MacPCD group may be candidates for immunotherapy treatments that can redirect macrophages and other immune cells back into a cancer-fighting mode, such as checkpoint inhibitors or macrophage-reprogramming therapies currently in clinical trials.
The MacPCD model represents a significant advance because it integrates two distinct aspects of tumor biology, immune cell behavior and cell death pathway disruption, into a single predictive score. This combined view is more informative than either perspective alone.
For patients, this research means that analyzing the expression of just 6 genes from a tumor biopsy or surgical specimen could provide a personalized risk score that is more accurate than what doctors currently have available. This could support decisions about whether to pursue more aggressive treatment or enroll in specific clinical trials.
The study also opens new doors for drug development. The 6 genes identified, along with the macrophage-PCD connection, point to specific molecular targets that future treatments could exploit, particularly for patients with advanced kidney cancer who have run out of effective standard options.