Mechanisms of irAEs in Barrier Organs

Introduction

The paper "When checkpoint inhibitors break barriers: Mechanisms and challenges of irAEs of the skin, gastrointestinal tract, and lung"¹ reviews why barrier organs—skin, gastrointestinal (GI) tract, and lung—are the most frequent sites of immune-related adverse events (irAEs) during immune checkpoint inhibitor (ICI) therapy. These tissues are in constant contact with the external environment, harbor microbiota, and contain tissue-resident immune populations that must balance tolerance with protection. The review synthesizes shared immunologic features and highlights the roles of T cells, myeloid cells, interferons, interleukins, autoantibodies, oxygenation, and dysbiosis in tissue-specific irAEs, while noting key knowledge gaps.

1. Why barrier organs are frequent irAE sites

Barrier organs (skin, GI tract, lung) are commonly affected—skin up to 60%, GI tract up to 40%, and lung up to 9.5% of patients. Compared to non-barrier organs (heart, pituitary, kidneys), these tissues:

• Continuously face environmental antigens and must maintain active tolerance.
• Harbor commensal microbiota, which shape local immune tone.
• Contain resident immune populations (e.g., TRM cells) that can rapidly respond to perturbations.

2. Dermatologic irAEs and mechanisms

The skin is a highly immunologically active barrier with keratinocytes, fibroblasts, resident myeloid cells, and TRM cells maintaining tolerance. Checkpoint inhibition can disrupt this balance.

Common subtypes and clinical notes

• ICI-lichen planus (up to 17% of anti–PD-(L)1 patients) with a band-like lymphocytic infiltrate and high IFN-γ expression.
• ICI-bullous pemphigoid (ICI-BP) with autoantibodies to basement membrane antigens (BP180, BP230) and infiltrating lymphocytes/eosinophils; can involve ocular and upper respiratory mucosae.
• ICI-SJS/TEN is rare but severe, with mortality rates up to 48%.

Mechanistic themes

• TRM expansion and activation: scRNA-seq studies implicate expanded cytotoxic TRM cells (notably Th1-like) in cutaneous irAEs.
• Checkpoint dependence of tolerance: PD-1 is required for CD8+ T-cell tolerance to cutaneous neoantigens, suggesting a direct role for PD-1 blockade in skin toxicity.
• Autoantibodies: In ICI-BP, autoantibodies to DEJ antigens are central; tumors in barrier organs (melanoma, NSCLC, urothelial carcinoma) may express these antigens, potentially driving antibody generation.
• Tertiary lymphoid structure (TLS) signals: TLS-promoting chemokines (e.g., CXCL13, CCL21) are upregulated after checkpoint inhibition, though their exact role in cutaneous irAEs is unclear.

3. Gastrointestinal irAEs and mechanisms

The GI tract is the largest mucosal surface and is protected by mucus layers, epithelial cells, and specialized secretory cells. Immune tolerance here is active and involves DC–Treg interactions and B-cell class-switching.

Clinical patterns and incidence

• Diarrhea incidence: anti–CTLA-4 up to 34%, anti–PD-1 21%, dual therapy 45%.
• Biopsy-proven colitis: 3–45%; enteritis 1–15%; gastritis <5%.
• Histology matters: Up to 35% of patients with new GI symptoms lack histologic inflammation on biopsy.
• Microscopic lymphocytic colitis: ~35% of anti–PD-1 cases, <10% of anti–CTLA-4 cases.
• Upper GI involvement: ~10% have isolated upper GI disease, sometimes resembling celiac-like enteropathy.

Mechanistic themes

• Loss of active tolerance: CD103+ DC–Treg interactions and FOXP3+ CD8+ T cells help maintain tolerance; disruption can promote inflammation.
• TRM populations: Effector responses may contract into TRM cells that persist in the epithelium.
• B-cell shifts: IBD shows skewing from IgA to IgG; B-cell depletion can precipitate de novo IBD.
• Shared-but-distinct features: ICI-colitis overlaps with spontaneous IBD but has distinct histologic and molecular features.

4. Lung irAEs (ICI-pneumonitis) and mechanisms

The lung continuously samples environmental antigens while preserving gas exchange, relying on epithelial defenses and resident immune cells.

Clinical patterns and incidence

• Incidence: up to 9.5% overall; higher with PD-(L)1 regimens than anti–CTLA-4.
• Mortality: 12–23%, making it one of the most lethal irAEs.
• Cancer-type differences: NSCLC 4.1% vs melanoma 1.6%; real-world NSCLC rates may be higher (up to 19% in a retrospective study).
• Risk factors: age, male gender, prior thoracic radiation, pre-existing interstitial lung disease, and smoking history.
• Diagnosis: primarily by imaging and exclusion of infection/hemorrhage/progression. Typical patterns include cryptogenic organizing pneumonia, hypersensitivity pneumonitis, acute interstitial pneumonia, and nonspecific interstitial pneumonia; often bilateral.

Mechanistic themes

• BAL scRNA-seq findings: expanded Tregs, CXCL13-expressing T cells, proinflammatory monocytes, and depletion of alveolar macrophages.
• Histology: nonspecific, with interstitial thickening, lymphocytic inflammation, pneumocyte desquamation, intra-alveolar fibrin, and foamy macrophages.

5. Barrier organs rarely impacted by irAEs

Despite similar features, genitourinary, nasal, and ocular mucosae are rarely involved:

• ICI ureteritis/cystitis: <1% (urinary urgency, sterile pyuria, hematuria).
• Ophthalmic irAEs: ~1% (decreased visual acuity, diplopia, tearing).

Why these sites are spared remains an important open question.

6. Shared immune mechanisms across barrier irAEs (Table 1 highlights)

The review emphasizes shared immunologic perturbations across skin, GI tract, and lung:

• TRM activation/expansion: prominent in skin and GI; lung TRM role remains less defined in the review.
• Autoantibodies: anti-BP180/BP230 (skin), TTG-IgA (ICI-celiac-like enteropathy), anti-CD74 (pneumonitis).
• Interleukins: IL-6/IL-17/IL-23 implicated across organs (with organ-specific patterns in Table 1).
• Interferons: IFN-γ and type I IFNs appear in skin and GI; IFN-γ also in lung.
• Microbiota associations:
  • Skin: murine models colonized with S. epidermidis show increased IL-17/IFN-γ.
  • Gut: higher baseline Faecalibacterium linked to higher colitis risk; Bacteroidetes associated with lower enterocolitis risk in anti–CTLA-4 therapy.
  • Lung: increased Proteobacteria and Firmicutes associated with pneumonitis risk.
• Treg changes: decreased Tregs in ICI-lichen planus vs spontaneous disease; increased but potentially altered Treg function in colitis; reduced PD-1/CTLA-4 expression in Tregs in pneumonitis cases.
• Androgen/sex effects: higher dermatologic and colitis risk reported in females; sex differences in pneumonitis incidence reported but vary by study.
• Oxygen tension: HIF1A/VEGFA upregulated in ICI-colitis; hypoxemia reported in ICI-pneumonitis.

Concluding remarks

The review argues that understanding shared barrier-organ mechanisms is essential for better prediction, diagnosis, and management of irAEs without compromising antitumor immunity. It highlights the need for interdisciplinary approaches and points to key gaps—especially why some barrier organs are rarely affected and how common mechanistic pathways could guide safer immunotherapy.

1. Sinha A, Seervai RNH, Vlastelica KM, Thomas MF, Hornick NI. Clinical Cancer Research. Author manuscript available in PMC, 2026. DOI: 10.1158/1078-0432.CCR-25-3352. PMCID: PMC12758644. ↩