Systemic and tumor intrinsic expansion of FCRL5 expressing B cells associates with poor response to Bacillus Calmette-Guerin immunotherapy in patients with non-muscle invasive bladder cancer

This study identifies that the systemic and tumor-intrinsic expansion of FCRL5-expressing atypical B cells, which form immunosuppressive niches and drive IgG-skewed humoral responses, is a key mechanism underlying poor response to BCG immunotherapy in patients with high-risk non-muscle invasive bladder cancer.

The Gist

The Big Picture: A Broken Alarm System

Imagine your bladder is a fortress, and Bladder Cancer is a group of intruders trying to sneak in. The standard treatment for this cancer (called BCG immunotherapy) is like hiring a specialized security team. You pour a harmless bacteria (BCG) into the bladder to wake up the immune system, hoping it will spot the cancer and destroy it.

For many patients, this works like a charm. But for about half of them, the security team shows up, gets confused, and the cancer comes back.

This study asks: Why does the security team fail for some people? The researchers discovered that the problem isn't that the security team is lazy; it's that the "base camp" is already occupied by a group of confused, exhausted lookouts who are actually making things worse.

---

The Key Characters: The "Atypical" B-Cells

To understand the failure, we need to meet the main characters: B-Cells. Think of these as the immune system's intelligence officers. Their job is to learn what the enemy looks like and send out the right weapons (antibodies) to fight them.

Usually, when you get a vaccine or treatment, your body creates Memory B-Cells. These are like elite, well-trained snipers who remember the enemy perfectly and can strike fast.

However, this study found that in patients who don't respond to BCG, their bodies are flooded with a different type of cell called Atypical B-Cells (ABCs).

• The Analogy: If Memory B-Cells are elite snipers, ABCs are like exhausted, confused guards who have been standing watch for too long. They are tired, they don't listen to orders well, and instead of fighting the cancer, they start building walls that block the good guys from getting through.

The Discovery: The "Bad Neighborhood"

The researchers looked at patients before and after BCG treatment. Here is what they found:

1. The Pre-Existing Problem: Even before the treatment started, patients who would eventually fail had higher levels of these "exhausted guards" (ABCs) in their blood. It was a pre-existing condition.
2. The Treatment Makes it Worse: When they gave these patients the BCG treatment, it didn't wake up the elite snipers. Instead, it accidentally gave the "exhausted guards" a caffeine shot. The number of ABCs exploded.
3. The "Tertiary Lymphoid Structures" (TLS): Inside the tumor, there are little neighborhoods called TLS. Think of these as community centers where immune cells meet to plan their attack.
   • In Responders (Success): The community centers are full of elite snipers and helpful helpers. They are organized and ready to fight.
   • In Non-Responders (Failure): The community centers are overrun by the "exhausted guards" (ABCs). These guards are sitting in the middle of the room, blocking the door, and talking to the cancer cells instead of fighting them. They are essentially hijacking the community center.

The "Wrong Kind of Ammo"

The study also looked at the "weapons" (antibodies) being produced.

• The Good Response: A healthy immune response produces IgA, which is like a specialized shield that protects the mucous membranes (the lining of the bladder).
• The Bad Response: The non-responders were producing mostly IgG. While IgG is a powerful weapon, in this specific location, it was like bringing a tank to a neighborhood street fight. It wasn't the right tool for the job. Furthermore, these IgG weapons were accidentally attacking the body's own healthy cells (autoantibodies), causing more confusion and inflammation without killing the cancer.

The Smoking Gun: FCRL5

The researchers found a specific "ID badge" on these exhausted guards called FCRL5.

• If a patient has a high number of cells wearing this badge, they are almost guaranteed to fail BCG treatment.
• The study validated this in hundreds of patients from different countries. High FCRL5 = High risk of cancer returning.

The Conclusion: Why This Matters

This paper changes how we view bladder cancer treatment. It suggests that for some patients, the immune system is already "burned out" before they even start treatment. Pouring more BCG on top of this exhausted system just makes the "bad guards" (ABCs) stronger, not the "good guards."

The Takeaway:
Instead of just giving everyone the same BCG treatment, doctors might soon be able to test a patient's blood first.

• If they have low levels of "exhausted guards" (ABCs): Give them BCG. It will likely work.
• If they have high levels: They need a different strategy. Maybe they need a drug to clear out the exhausted guards first, or a combination therapy that wakes up the elite snipers while silencing the bad ones.

In short, the study found that you can't win a war if your own troops are too tired and confused to fight. Identifying these tired troops early could save lives by guiding doctors to the right treatment sooner.

Technical Summary

1. Problem Statement

Non-muscle invasive bladder cancer (NMIBC) is the most common urological malignancy. The standard of care for intermediate- and high-risk patients is transurethral resection followed by intravesical Bacillus Calmette-Guérin (BCG) immunotherapy. However, over 50% of patients fail to respond, experiencing early recurrence or disease progression, often necessitating radical cystectomy. While the role of T-cell-mediated cytotoxicity and innate immunity in BCG response is well-studied, the contribution of B cells and humoral immunity remains poorly defined. Previous work suggested that "atypical B cells" (ABCs), often associated with aging and chronic inflammation, might drive immune exhaustion, but their specific role in BCG failure in human NMIBC was unknown.

2. Methodology

The study employed a multi-modal, longitudinal, and spatially resolved approach integrating systemic and local immune profiling:

• Cohorts:
  • Primary Cohort (n=45): High-risk NMIBC patients treated with BCG. Peripheral blood was collected at three timepoints: pre-BCG, post-1st instillation, and post-4th instillation. Tumor tissues (pre-BCG and post-recurrence) were analyzed.
  • Validation Cohorts (n=409): Two independent cohorts (Strangaard et al., n=126; de Jong et al., n=283) used for bulk RNA-seq validation of FCRL5 expression.
• Systemic Profiling:
  • Flow Cytometry: Multiparametric analysis of circulating B cells to quantify subsets (Naïve, Memory, ABCs defined as CD19⁺CD21⁻/lowCD11c⁺, and FCRL5⁺).
  • scRNA-seq: Single-cell transcriptomics of enriched peripheral B cells (pre- and post-4th BCG) to define transcriptional states and differentiation trajectories.
  • Proteomics & Serology: Olink proteomics (96 proteins) for systemic immune activation; HuProt™ microarrays (>21,000 proteins) for autoantibody profiling; ELISA for anti-BCG IgG titers; and LegendPlex for Ig isotyping.
• Local (Tumor) Profiling:
  • Multiplex Immunofluorescence (mIF): 8-plex and 25-plex panels (PhenoCycler Fusion) to map spatial distribution of immune cells (ABCs, T cells, Macrophages, DCs) within Tumor-Associated (TA) stroma, Tertiary Lymphoid Structures (TLS), and Lamina Propria (LP).
  • Spatial Transcriptomics: Xenium In Situ platform (~5,000 genes) on FFPE sections to resolve cell-cell interactions and epithelial states at single-cell resolution.
• Analysis: Pseudotime trajectory inference (Slingshot), cell-cell communication (CellChat), spatial neighborhood analysis (k-nearest neighbor, niche modeling), and survival analysis (Kaplan-Meier).

3. Key Contributions

• Identification of ABCs as a Biomarker: The study identifies the expansion of Atypical B cells (ABCs), specifically those expressing FCRL5, as a defining feature of BCG non-responders.
• Mechanism of Failure: It demonstrates that BCG failure is driven by a pre-existing, ABC-dominated immune landscape that is further amplified by repeated BCG instillations, leading to an immunosuppressive microenvironment rather than effective anti-tumor immunity.
• Spatial Reorganization: The paper provides high-resolution spatial evidence that ABCs in non-responders disrupt normal TLS architecture, accumulating in TLS cores and epithelial niches, unlike in responders where they are peripheral.
• Humoral Dysregulation: It characterizes a distinct "IgG-skewed" humoral response in non-responders, marked by expanded autoantibody repertoires and progressive anti-BCG IgG accumulation, contrasting with the protective mucosal immunity seen in responders.

4. Key Results

A. Systemic Expansion of ABCs in Non-Responders

• Longitudinal Dynamics: Non-responders showed a significant increase in circulating ABCs (CD19⁺CD21⁻CD11c⁺) and FCRL5⁺ B cells after the 4th BCG dose, whereas responders did not.
• Differentiation Trajectory: scRNA-seq revealed that non-responders' B cells preferentially differentiated toward the ABC lineage (driven by TBX21, ZEB2, ITGAX), while responders maintained naïve B cell pools and class-switched memory cells.
• Transcriptional Signature: Non-responders exhibited elevated expression of exhaustion and chronic activation genes prior to treatment, suggesting a pre-configured immune state.

B. Dysregulated Humoral Immunity

• Isotype Skewing: Non-responders displayed an IgG1-skewed antibody profile (dominant IGHG1) and reduced IgA, whereas responders showed higher IGHG3 and IGHA1.
• Autoimmunity: Non-responders developed a broad repertoire of autoantibodies (829 differentially abundant autoantigens between pre/post-BCG) compared to responders (72), indicating a loss of self-tolerance.
• Anti-BCG Response: Non-responders showed a progressive increase in IgG reactivity against BCG antigens, suggesting immunological misdirection rather than protective immunity.

C. Spatial Architecture and Immunosuppressive Niches

• ABC Localization: In non-responders, ABCs (CD79a⁺CD21⁻CD11c⁺) were enriched in TA-stroma and mature TA-TLS cores, often co-localizing with PD-1⁺ B cells, regulatory T cells (Tregs), and CD163⁺ M2 macrophages.
• TLS Disruption: In responders, ABCs were located at the TLS periphery (compatible with Germinal Center activity). In non-responders, they occupied the TLS center, disrupting organized lymphoid architecture.
• Epithelial Interaction: ABCs in non-responders were in close proximity to tumor epithelial cells (TE), particularly in basal epithelial niches exhibiting inflammatory signaling (TNF/IFN-γ). This proximity was associated with high expression of immune checkpoints (PD-1/PD-L1) on both ABCs and epithelial cells.
• Cellular Neighborhoods: Non-responders lacked the "helper T cell/myeloid" supportive niches seen in responders; instead, ABCs formed niches with cytotoxic T cells and Tregs, creating a tumor-permissive environment.

D. Clinical Validation

• FCRL5 as a Predictor: High expression of FCRL5 (an ABC-specific transcript) in tumor bulk RNA-seq was significantly associated with shorter Recurrence-Free Survival (RFS) and Progression-Free Survival (PFS) in two independent validation cohorts (n=409).
• Systemic Score: High circulating ABC scores pre-treatment also correlated with shorter RFS.

5. Significance and Implications

• Predictive Biomarker: The study proposes FCRL5 expression and circulating ABC frequency as robust biomarkers to predict BCG failure, potentially allowing for early identification of patients who need alternative therapies (e.g., radical cystectomy or combination immunotherapy) before recurrence occurs.
• Therapeutic Strategy: The findings suggest that the "exhausted" ABC-dominated state in non-responders creates an immunosuppressive barrier. This supports the clinical rationale for combining BCG with immune checkpoint inhibitors (e.g., anti-PD-1) to reverse this exhaustion, a strategy already showing promise in trials like CREST.
• Mechanistic Insight: The paper redefines the understanding of BCG failure, moving beyond T-cell exhaustion to include B-cell driven immune dysregulation, where chronic antigen stimulation drives ABC expansion, autoimmunity, and the formation of immunosuppressive niches that protect the tumor.

In summary, this study establishes that a pre-existing, ABC-biased immune landscape, exacerbated by BCG therapy, drives early recurrence in NMIBC by disrupting local immune architecture and promoting an IgG-skewed, auto-reactive, and immunosuppressive microenvironment.