Kidney renal papillary cell carcinoma (KIRP) is the second most common type of kidney cancer, accounting for 10-20% of all kidney cancers. Unlike the more extensively studied clear cell type, KIRP has fewer established treatment options, and outcomes for advanced or metastatic KIRP remain poor.
In recent years, immunotherapy - treatments that help the body's own immune system fight cancer - has transformed care for many cancers, including the more common clear cell kidney cancer. These therapies work by releasing 'brakes' on immune cells so they can attack tumors more effectively. However, whether immunotherapy can work as well in papillary kidney cancer has been much less studied.
A critical part of understanding immunotherapy potential is mapping the tumor immune microenvironment - the collection of immune cells, signals, and structures that exist within and around a tumor. Some tumors are 'hot' (rich in immune cells that could potentially attack the cancer) while others are 'cold' (relatively devoid of immune activity). Hot tumors generally respond better to immunotherapy.
This study aimed to comprehensively characterize the immune microenvironment of KIRP using data from 289 patients in The Cancer Genome Atlas (TCGA) database, ultimately identifying which patients and tumor subtypes are most likely to benefit from immunotherapy approaches.
The research team used computational biology methods to analyze gene expression data from 289 KIRP tumor samples and 32 healthy kidney tissue samples from TCGA. First, they identified 5,132 genes whose expression levels were significantly different between tumor and normal tissue.
To understand how these genes work together, the team applied weighted gene co-expression network analysis (WGCNA) - a method that groups genes with similar expression patterns into modules. Each module represents a cluster of genes likely working together in the same biological pathway. This approach identified 32 such modules in the KIRP data.
The relative abundance of 22 different immune cell types in each tumor sample was estimated using the CIBERSORT method - a computational tool that infers immune cell composition from gene expression data. This allowed the team to understand which immune cells were present in KIRP tumors without physically examining each sample.
To predict which patients would respond to immunotherapy, the team used a computational tool called TIDE (Tumor Immune Dysfunction and Exclusion), which calculates scores reflecting how well T cells can function within a tumor and whether they are being blocked from entering it. Lower TIDE scores indicate a higher likelihood of responding to immunotherapy.
Among the 32 gene modules identified, the purple module showed the strongest correlation with CD8+ T cells (r=0.67) - the immune cells primarily responsible for killing cancer cells. This module contained 169 genes enriched in pathways related to T-cell activation, response to immune signaling molecules (interferons), and natural killer cell activity.
From this module, researchers used two complementary methods - protein-protein interaction network analysis and co-expression network analysis - to identify the 12 most central and important genes. These 12 hub genes were: GZMA, FASLG, CD2, PDCD1, CCL5, CD8A, CD3D, EOMES, NKG7, CD3E, CD8B, and CTLA4.
All 12 hub genes were significantly more highly expressed in KIRP tumor tissue compared to normal kidney tissue. In laboratory tissue samples from 30 patients, four of these genes - CCL5, FASLG, EOMES, and PDCD1 - were directly confirmed to be elevated in KIRP tumors. Importantly, high expression of 8 of the 12 genes was associated with worse patient survival.
The genes cluster into two functional categories: those that stimulate immune activity (CD2, CD8A, CD8B, CD3D, CD3E, EOMES, GZMA) and those that suppress immune activity (PDCD1/PD-1, CTLA4, FASLG, CCL5). The simultaneous presence of both suggests that KIRP tumors may have active immune cells that are being suppressed - a state called T-cell exhaustion.
Using the CIBERSORT analysis, researchers found that KIRP tumors contained significant numbers of immune cells - a characteristic of 'hot' tumors that are generally considered good candidates for immunotherapy. Specifically, CD8+ T cells (cancer-killing immune cells) were present in substantial numbers.
However, a paradox emerged: despite the presence of many CD8+ T cells, their levels actually increased as the cancer advanced to higher stages. Normally, more CD8+ T cells would be expected to control tumor growth. This finding suggests the T cells are present but functionally exhausted - they exist in the tumor but cannot kill cancer cells effectively.
The levels of regulatory T cells (Tregs) - immune cells that suppress the immune response - also increased with tumor progression. This rising tide of immunosuppressive cells likely contributes to the silencing of the CD8+ T cells, creating an environment where immune killing is blocked despite abundant immune cells being present.
High infiltration levels of B cells and CD8+ T cells were paradoxically associated with worse prognosis in KIRP patients, further supporting the idea that the immune cells present are exhausted or suppressed rather than actively fighting the cancer. This creates a clear rationale for immunotherapy approaches that re-activate these exhausted cells.
Using the 12 hub genes as inputs, researchers applied consensus clustering to categorize the 289 KIRP patients into five distinct immune subtypes: C1 (71 patients), C2 (52 patients), C3 (29 patients), C4 (87 patients), and C5 (44 patients). These subtypes showed significantly different survival outcomes and immune compositions.
Subtypes C1 and C3 stood out as having the highest overall immune infiltration - the most abundant levels of all six immune cell types measured, and the highest scores for immune activity, stromal activity, and immune gene expression. These represent the most immunologically active tumor environments.
Subtype C2 was at the opposite end of the spectrum - the lowest immune cell infiltration, the lowest immune scores, and the highest tumor purity (meaning the tumor had the fewest immune cells mixed in). C4 and C5 fell in the middle of the immune activity spectrum.
Using the TIDE computational tool to predict immunotherapy response, subtype C3 showed the highest predicted response rate (96.6%), followed by C1 (81.7%), C4 (44.8%), C5 (29.5%), and lastly C2 (only 7.7%). This identified C1 and C3 as the high immunotherapy response group, making them prime candidates for immune checkpoint inhibitor treatment.
Beyond the well-known immune checkpoint targets like PD-1 and CTLA4, this study highlights two less-studied genes - CCL5 and FASLG - as potential contributors to the immunosuppressive environment in KIRP tumors. Both genes showed the same pattern across all five subtypes: highest in C3 and C1, lowest in C2.
CCL5 is a signaling molecule (chemokine) that can attract regulatory T cells (Tregs) to the tumor, enhancing immune suppression. It can also promote tumor growth by stimulating macrophages to produce enzymes that help cancer spread. Drugs targeting the CCL5 signaling axis are already being explored in other cancer types.
FASLG (Fas Ligand) is a protein that can kill immune cells when they arrive at the tumor, essentially defending the tumor against immune attack. High FASLG expression helps cancer cells create a barrier against immune killing, contributing to T-cell exhaustion. Strategies to block FASLG could enhance the effectiveness of immunotherapy.
The identification of CCL5 and FASLG as key drivers of immune suppression in KIRP - separate from the more commonly targeted PD-1/PD-L1 axis - opens the door to novel combination treatment strategies. Targeting multiple immunosuppressive pathways simultaneously may be more effective than single-checkpoint inhibition alone.
Comparing the genetic mutation profiles of patients predicted to respond well versus poorly to immunotherapy, researchers found that the genes mutated most often differed between the two groups. Five genes (MUC16, KMT2C, MET, TTN, MUC4) were mutated in both groups, but each group had genes mutated exclusively within it.
The most striking difference was in the mutation frequency of the gene TTN: it was mutated in 17% of low-to-medium immunotherapy responders but only 7% of high responders. TTN mutations may influence the immune microenvironment in ways that reduce immunotherapy benefit.
Pathway analysis (GSEA) showed that the high immunotherapy response group was enriched in multiple immune-related biological processes, including antigen rejection pathways, interleukin-12 secretion, and MHC class II protein complex activity - all processes related to stronger immune recognition and attack of cancer cells.
These mutation and pathway differences provide additional layers of biological context for why some KIRP patients are more likely to respond to immunotherapy. In the future, genetic profiling could be combined with immune subtype classification to give an even more accurate prediction of who will benefit from treatment.
This study provides the most comprehensive map of the immune landscape in papillary kidney cancer to date. The key finding is that KIRP tumors have immune cells present (are 'hot') but those cells are blocked from working - a state called CD8+ T-cell exhaustion. Treatments that can reverse this exhaustion may be particularly effective in KIRP.
Among the five immune subtypes identified, patients in C1 and C3 are most likely to benefit from immunotherapy targeting CD8+ T-cell exhaustion. These subtypes already have the most immune activity and the highest predicted response rates to checkpoint inhibitors, making them the ideal group for clinical trial enrollment.
The identification of CCL5 and FASLG as likely contributors to immunosuppression in KIRP suggests that combination therapies targeting these molecules alongside traditional checkpoints like PD-1 could potentially improve outcomes even in currently low-responding subtypes.
These findings lay a foundation for future clinical trials specifically designed for KIRP patients, with patient selection guided by immune subtype. Validating these findings in prospective clinical studies and developing companion diagnostics to identify the C1 and C3 subtypes in real patients will be important next steps toward personalized immunotherapy for papillary kidney cancer.