🔬 PD-1/PD-L1 Checkpoint Evasion

T-Cell Recognition — Immune Surveillance of Tumor Cells

Every day, the adaptive immune system performs continuous surveillance of the body's tissues. Cytotoxic CD8+ T-cells circulate through blood and lymph, patrolling for cells displaying abnormal peptides on MHC class I molecules. Tumor cells, carrying mutated or overexpressed proteins, can present neoantigens that a matching T-cell receptor (TCR) recognizes as foreign — triggering the potential for an anti-tumor immune attack, provided no downstream signal interferes.

• ~10¹¹: CD8+ T-cells in circulation (patrolling at any given time)
• ~10⁷–10⁸: TCR diversity (repertoire) (unique receptor clonotypes)
• 8–10: MHC-I peptide length (amino acids presented)
• variable: Neoantigens per tumor (depends on tumor mutational burden)

The TCR–peptide–MHC recognition event

Every nucleated cell continuously samples its intracellular proteome, degrading a fraction of proteins into short peptides that are loaded onto MHC class I molecules and displayed at the cell surface. This is a constant, cell-intrinsic "reporting" system.

A cytotoxic T-cell expresses a single TCR clonotype, generated by random V(D)J recombination during thymic development. When a TCR encounters its cognate peptide-MHC complex with sufficient affinity, a cascade of intracellular signaling (CD3 complex, ZAP-70, LAT) is triggered, priming the T-cell for activation.

Tumor cells frequently display neoantigens — peptides derived from somatic mutations, aberrant splicing, or overexpressed self-proteins — that are absent from healthy tissue. A T-cell bearing a matching TCR can, in principle, recognize and eliminate the tumor cell. This immunosurveillance process operates continuously but is only the opening move: recognition alone does not guarantee an effective attack, since numerous regulatory checkpoints — including PD-1/PD-L1 — can intervene downstream.

From recognition to a coordinated attack

Successful recognition sets off a multi-step activation program: costimulatory signals (CD28–CD80/86), cytokine support (IL-2, IL-12), and formation of an immunological synapse between the T-cell and its target. Only once fully activated does the T-cell acquire cytotoxic effector function — release of perforin and granzymes, and Fas–FasL-mediated apoptosis induction in the target cell.

This multi-checkpoint requirement exists by design: the immune system must distinguish a genuine threat from healthy self-tissue to avoid autoimmunity. Tumors exploit exactly this built-in caution, most notably by co-opting the PD-1/PD-L1 checkpoint axis — the subject of the following stages.

Tumor PD-L1 Expression — Co-opting a Physiological Checkpoint

PD-1 (programmed cell death protein 1) and its ligand PD-L1 form a physiological "brake" that normally prevents excessive T-cell activity and protects healthy tissue from autoimmune damage. Many tumors upregulate PD-L1 on their surface, hijacking this self-tolerance mechanism to present themselves as immunologically "off-limits" — actively suppressing the very T-cells that recognize them as abnormal.

• many: PD-L1+ tumor types (melanoma, NSCLC, RCC, urothelial, more)
• CD274: PD-L1 gene (chromosome 9p24.1)
• IFN-γ, hypoxia: PD-L1 upregulation drivers (and oncogenic signaling)
• self-tolerance: Normal PD-L1 role (limits autoimmune tissue damage)

Why PD-L1 exists — physiological immune tolerance

The PD-1/PD-L1 axis is not inherently pathological. In healthy tissue, PD-L1 is expressed on many cell types — particularly in response to local inflammation and interferon-gamma (IFN-γ) release — to limit collateral damage from an ongoing immune response. This "adaptive immune resistance" prevents T-cells from over-reacting and destroying healthy bystander tissue once a threat has been addressed. Placental trophoblasts, corneal cells, and cells in immune-privileged sites also rely on PD-L1 to avoid unwanted immune attack.

Tumor cells frequently commandeer this same mechanism. Two broad routes drive tumor PD-L1 expression: intrinsic (oncogenic pathway activation, e.g., PI3K/AKT, MYC, EGFR signaling driving constitutive PD-L1 transcription) and extrinsic (adaptive upregulation in response to IFN-γ secreted by infiltrating T-cells themselves — the tumor senses it is being attacked and responds by raising its checkpoint shield).

PD-L1 as a marker of an active but suppressed immune response

Counterintuitively, high tumor PD-L1 expression is often a sign that an anti-tumor immune response was already underway — IFN-γ from infiltrating T-cells is one of the strongest inducers of PD-L1. This is the basis for "adaptive immune resistance": the tumor is not hiding from the immune system, it is actively defending against a real attack by co-opting a checkpoint pathway.

This has direct clinical relevance: tumors with higher baseline PD-L1 expression and evidence of pre-existing T-cell infiltration tend to be more responsive to checkpoint blockade therapy (see Stage 4), since the machinery for an anti-tumor attack is already present — it merely needs to be un-suppressed.

PD-L1 expression is measured clinically by immunohistochemistry as the Tumor Proportion Score (TPS) or Combined Positive Score (CPS), and is used — together with tumor mutational burden and other biomarkers — to help estimate the likelihood of benefit from checkpoint inhibitor therapy.

PD-1 Engagement — Delivering the Inhibitory Signal

Binding of tumor PD-L1 to PD-1 on the T-cell surface is not a passive occlusion — it actively transmits an inhibitory signal into the T-cell. This signal counteracts TCR and costimulatory signaling, effectively disabling the T-cell's capacity to proliferate, secrete cytokines, and exert cytotoxic effector function against the tumor.

• PDCD1: PD-1 gene (expressed on activated T-cells)
• ITIM / ITSM: Inhibitory motif (cytoplasmic tail of PD-1)
• SHP-2: Key phosphatase recruited (dephosphorylates TCR signaling nodes)
• T-cell exhaustion: Functional outcome (reduced proliferation & cytotoxicity)

Molecular mechanism of PD-1 inhibitory signaling

PD-1 is a type I transmembrane receptor upregulated on T-cells following activation. Its cytoplasmic tail contains an immunoreceptor tyrosine-based inhibitory motif (ITIM) and an immunoreceptor tyrosine-based switch motif (ITSM). Upon PD-L1 engagement, these motifs become phosphorylated and recruit the phosphatase SHP-2 (and to a lesser extent SHP-1) to the immunological synapse.

SHP-2 dephosphorylates key signaling intermediates downstream of the TCR and CD28 costimulatory receptor — including ZAP-70, PI3K, and CD28 itself — effectively short-circuiting the activation cascade at multiple nodes simultaneously. The net result is dampened calcium flux, reduced transcription factor activity (NFAT, AP-1, NF-κB), and a T-cell that fails to fully activate despite productive TCR engagement with its target antigen.

From acute suppression to T-cell exhaustion

A single PD-1/PD-L1 engagement event dampens T-cell signaling transiently. But chronic, repeated checkpoint engagement — as occurs continuously within a PD-L1-high tumor microenvironment — drives T-cells into a distinct, more stable dysfunctional state known as "T-cell exhaustion."

Exhausted T-cells progressively lose effector functions in a hierarchical fashion: first IL-2 production, then proliferative capacity, then TNF-α production, and finally cytotoxic granzyme/perforin release and IFN-γ secretion. Exhausted T-cells also upregulate additional inhibitory receptors (LAG-3, TIM-3, CTLA-4, TIGIT), compounding the suppression through parallel checkpoint pathways. This is the immunological end state that tumor PD-L1 expression is driving toward — a T-cell population physically present within the tumor but functionally incapacitated.

Because exhaustion is layered and partially reversible, blocking a single checkpoint (PD-1/PD-L1) can restore substantial function in less severely exhausted T-cells, but deeply exhausted populations may require combination blockade of multiple inhibitory receptors to fully reinvigorate.

Checkpoint Blockade Antibodies — Disrupting PD-1/PD-L1 Engagement

Checkpoint inhibitor antibodies — one of the most significant advances in oncology of the past two decades — are engineered monoclonal antibodies that bind either PD-1 (on the T-cell) or PD-L1 (on the tumor cell), physically occupying the interaction surface. With the binding interface sterically blocked, the inhibitory signal can no longer be transmitted, freeing the T-cell to resume its anti-tumor program.

• nivolumab, pembrolizumab: Anti-PD-1 antibodies (and others)
• atezolizumab, durvalumab: Anti-PD-L1 antibodies (and others)
• IgG4: Antibody isotype (typical) (reduced effector-cell engagement)
• 2011–2014: First FDA approval (ipilimumab (CTLA-4), then PD-1/PD-L1)

Two blockade strategies — targeting the receptor or the ligand

Checkpoint inhibitor antibodies can target either side of the PD-1/PD-L1 interaction:

• Anti-PD-1 antibodies (e.g., nivolumab, pembrolizumab) bind the PD-1 receptor on the T-cell surface, blocking engagement with both of its ligands, PD-L1 and PD-L2.

• Anti-PD-L1 antibodies (e.g., atezolizumab, durvalumab, avelumab) bind PD-L1 on the tumor (or other) cell surface, blocking its interaction with PD-1 while leaving the PD-1/PD-L2 axis intact.

Both strategies achieve the same functional outcome — preventing delivery of the inhibitory signal — but differ in target-cell distribution, pharmacokinetics, and the specific side effect profiles observed clinically. The choice between them, and between blocking PD-1 versus PD-L1, is guided by tumor biology, biomarker status, and accumulated clinical trial evidence for each indication.

Structural basis of blockade

Structural studies of antibody–PD-1 and antibody–PD-L1 complexes show that these therapeutic antibodies bind epitopes that overlap directly with, or sterically occlude, the natural PD-1/PD-L1 binding interface. Once bound, the antibody's bulk physically prevents the two proteins from docking in the geometry required for productive signaling — even if some residual low-affinity contact remains possible, it is insufficient to cluster the receptor and trigger ITIM/ITSM phosphorylation.

Because this is a competitive, reversible blockade at the molecular level, checkpoint inhibitor antibodies are typically dosed on a repeating schedule (every 2–6 weeks depending on the agent) to maintain sufficiently high receptor occupancy over the treatment course.

Checkpoint blockade does not add a new anti-tumor mechanism — it removes a brake on a pre-existing one. Its efficacy therefore depends heavily on whether tumor-reactive T-cells are already present and capable of being reinvigorated, which is why biomarkers like PD-L1 expression and T-cell infiltration help predict response.

Restored T-Cell Anti-Tumor Activity After Checkpoint Blockade

Once the PD-1/PD-L1 interaction is blocked, T-cells that had been functionally suppressed regain their capacity for activation, proliferation, cytokine secretion, and cytotoxic killing. In patients who respond, this reinvigoration of pre-existing tumor-reactive T-cells can translate into durable tumor regression — sometimes lasting years after treatment ends.

• ~15–45%: Objective response rate (varies by tumor) (across approved indications)
• possible: Durable responses (in a subset of long-term responders)
• occur in a subset: Immune-related adverse events (from broader immune activation)
• common: Combination approaches (chemo, CTLA-4, targeted therapy)

From reinvigoration to tumor clearance

Upon checkpoint blockade, T-cells that were signaling-suppressed but not irreversibly exhausted resume TCR and costimulatory signaling. This restores proliferative capacity (clonal expansion of tumor-reactive clones), cytokine secretion (IFN-γ, TNF-α), and cytotoxic effector function — granzyme/perforin-mediated lysis and Fas–FasL-induced apoptosis of tumor cells.

Reinvigorated T-cells can also recruit and activate additional immune populations at the tumor site (a phenomenon sometimes called "epitope spreading"), broadening the anti-tumor response beyond the original recognized antigens as tumor cell death releases further antigenic material for presentation.

What determines whether treatment response occurs

Response to checkpoint blockade is not universal, and several factors are associated with the likelihood of benefit:

• Baseline PD-L1 expression: higher expression often correlates with a greater probability of response, though it is an imperfect predictor used alongside other markers. • Tumor mutational burden / neoantigen load: more mutations can mean more potential targets for T-cell recognition. • Pre-existing T-cell infiltration: an "inflamed" tumor microenvironment with T-cells already present responds better than an immunologically "cold" tumor lacking infiltration. • Degree of T-cell exhaustion: deeply exhausted T-cell states may not be fully reversible by single-agent blockade, motivating combination checkpoint strategies.

Successful checkpoint disruption is necessary but not sufficient on its own — durable response depends on the interplay between baseline tumor immunogenicity, the state of tumor-infiltrating T-cells, and how completely the inhibitory signal has been removed.

The 2018 Nobel Prize in Physiology or Medicine was awarded to James P. Allison and Tasuku Honjo for the discovery of cancer therapy by inhibition of negative immune regulation — the foundational work behind CTLA-4 and PD-1/PD-L1 checkpoint blockade.