Back to BlogClinical AI

Immunotherapy Checkpoint Inhibitors: A Clinical Guide for Oncologists

Sam AndersonSam Anderson
8 min read
All claims reviewed against primary literature by Director of Research, Sam Anderson
Oncology infusion center with immunotherapy IV and physician reviewing protocols

Checkpoint Inhibitor Landscape in 2026

Immune checkpoint inhibitors (ICIs) have become the backbone of treatment for numerous solid tumors, including melanoma. With over 80 FDA-approved indications across pembrolizumab, nivolumab, atezolizumab, durvalumab, and ipilimumab, the clinical landscape has become increasingly complex. This guide provides a practical framework for selecting, monitoring, and managing checkpoint inhibitor therapy.

The challenge for the practicing oncologist is no longer whether immunotherapy works — that question was answered decisively years ago. The challenge is navigating the complexity. Which biomarker drives the decision in this specific tumor type? When do you choose monotherapy versus combination? How do you counsel the patient who develops thyroiditis at week six? And when a patient on pembrolizumab presents with new-onset diarrhea, how do you distinguish immune colitis from an unrelated infectious cause before it becomes a grade 3 emergency? These are the clinical decisions that define daily practice in immuno-oncology, and this guide is structured around them.

Biomarker-Driven Patient Selection

PD-L1 expression (measured by TPS or CPS depending on tumor type), microsatellite instability (MSI-H/dMMR, relevant to colorectal cancer treatment), and tumor mutational burden (TMB) remain the primary biomarkers guiding ICI selection. In NSCLC (see also lung cancer screening criteria), PD-L1 TPS above 50% predicts robust response to pembrolizumab monotherapy, while patients with TPS 1-49% generally require combination chemoimmunotherapy.

Understanding TPS vs. CPS

The distinction between Tumor Proportion Score and Combined Positive Score matters more than many clinicians realize, because it directly determines which patients are eligible for specific agents and regimens. TPS measures PD-L1 expression only on tumor cells and is the primary scoring method for NSCLC. CPS includes PD-L1 staining on tumor cells, lymphocytes, and macrophages, expressed as a percentage of total viable tumor cells, and is used for gastric, esophageal, cervical, head and neck, and urothelial cancers among others. A patient with a TPS of 5% might have a CPS of 20 or higher if there is robust immune infiltration in the tumor microenvironment. Using the wrong scoring method for the tumor type can lead to incorrect treatment selection — an easily avoidable error that nonetheless occurs when pathology reports are not reviewed carefully in the context of the specific indication being considered.

When TMB and MSI-H Drive the Decision

MSI-H/dMMR status has become the most important tumor-agnostic biomarker in oncology. Pembrolizumab carries a tissue-agnostic approval for MSI-H/dMMR solid tumors that have progressed on prior treatment, meaning that the molecular finding overrides the anatomic site of origin in determining treatment eligibility. In practice, this means that routine MSI testing — or at minimum, immunohistochemistry for mismatch repair proteins — should be considered in any patient with a solid tumor who might be a candidate for immunotherapy, particularly in colorectal, endometrial, and gastric cancers where MSI-H prevalence is highest. TMB testing is more nuanced. A TMB of 10 or more mutations per megabase supports pembrolizumab use in a tissue-agnostic setting, but the clinical utility is more variable than MSI-H status, and TMB thresholds remain a subject of ongoing refinement.

Immune-related adverse events (irAEs) affect 60-80% of patients receiving ICI therapy, with grade 3-4 events occurring in 10-30%. The most common irAEs include dermatologic toxicity (rash, pruritus), endocrinopathies (thyroiditis, adrenal insufficiency), hepatitis, colitis, and pneumonitis. Early recognition and grading-appropriate management with corticosteroids is essential.

A Grading-Based Management Approach

The management of irAEs follows a grading framework that clinicians should internalize. Grade 1 events (mild symptoms, no interference with daily activities) generally allow continuation of immunotherapy with close monitoring and symptomatic treatment. Grade 2 events (moderate symptoms, some limitation of daily activities) typically warrant holding the checkpoint inhibitor and initiating moderate-dose corticosteroids, with resumption once symptoms resolve to grade 1 or less. Grade 3 events (severe symptoms requiring hospitalization or significantly limiting self-care) mandate discontinuation of immunotherapy, high-dose corticosteroids, and consideration of additional immunosuppression if steroids are insufficient. Grade 4 events (life-threatening) require permanent discontinuation and aggressive immunosuppressive management in an inpatient setting.

Timing and Pattern Recognition

Knowing when to expect specific irAEs improves early detection. Dermatologic toxicity tends to appear earliest, often within the first 2-4 weeks of treatment. Colitis typically presents at 6-8 weeks. Hepatitis and pneumonitis cluster around 8-12 weeks. Endocrinopathies — particularly thyroiditis — may present at any point but most commonly appear at 6-14 weeks. The critical point is that irAEs can develop at any time during treatment and even months after discontinuation, so surveillance must continue well beyond the last infusion. Patients should be educated to report new symptoms promptly, and every clinic visit should include a targeted review of systems covering the common irAE organ systems.

Endocrinopathies Require Special Attention

Unlike most irAEs, immune-mediated endocrinopathies are frequently permanent. Thyroiditis often progresses through a transient thyrotoxic phase before settling into permanent hypothyroidism requiring lifelong levothyroxine. Immune-mediated hypophysitis can cause panhypopituitarism with permanent cortisol, thyroid hormone, and gonadotropin deficiency. The clinical implication is that endocrine irAEs do not necessarily require discontinuation of checkpoint inhibitor therapy — the endocrinopathy is managed with hormone replacement, and immunotherapy can often continue. This distinguishes endocrine irAEs from hepatitis, colitis, and pneumonitis, where the irAE itself mandates treatment interruption or permanent discontinuation.

Combination Strategies: IO-IO and Chemo-IO

The CheckMate 227 and CheckMate 9LA trials established nivolumab plus ipilimumab as an effective IO-IO combination in NSCLC. KEYNOTE-789 and IMpower150 demonstrated chemo-IO benefit. The choice between IO-IO and chemo-IO depends on PD-L1 status, histology, patient performance status, and tolerance for immune toxicity.

Choosing Between IO-IO and Chemo-IO

This decision is one of the most common clinical dilemmas in thoracic oncology. IO-IO combinations (nivolumab plus ipilimumab) offer the advantage of a chemotherapy-free regimen, which is appealing to patients concerned about traditional chemotherapy side effects and to clinicians managing frail patients who may not tolerate platinum doublets. However, IO-IO combinations carry a higher risk of immune toxicity — the addition of ipilimumab (anti-CTLA-4) approximately doubles the rate of grade 3-4 irAEs compared to PD-1 monotherapy. Chemo-IO combinations offer a more predictable toxicity profile and potentially faster tumor response (the chemotherapy component provides early cytoreduction while the immunotherapy builds a durable response). For patients with high PD-L1 expression and good performance status, IO-IO may be preferred. For patients with PD-L1-low tumors, non-squamous histology, or those who need rapid symptomatic improvement from tumor shrinkage, chemo-IO is typically the stronger choice.

Communicating with Patients About Immunotherapy

Patients often arrive at the oncology visit with preconceptions about immunotherapy — some view it as a miracle cure, others are terrified by what they have read about autoimmune side effects. Effective counseling requires addressing both extremes. Immunotherapy works differently from chemotherapy: it does not directly kill cancer cells but rather releases the brakes on the immune system, allowing the body to mount its own antitumor response. This means that responses may take longer to become apparent (pseudoprogression, where tumors appear larger on early imaging before shrinking, is a real phenomenon though relatively uncommon). It also means that the side effects are fundamentally different — not hair loss and nausea, but inflammation in any organ where the activated immune system targets normal tissue. Framing immunotherapy as harnessing the patient's own immune system helps set appropriate expectations for both efficacy timeline and side effect profile.

Limitations and Evolving Questions

Despite the remarkable success of checkpoint inhibitors, fundamental questions remain. Primary resistance — patients who never respond to immunotherapy — is the norm rather than the exception, with objective response rates typically in the 20-40% range for most solid tumors. Acquired resistance, where initial responders eventually progress, affects the majority of responding patients over time. Biomarker selection, while improved, remains imperfect: some patients with PD-L1-negative tumors respond dramatically, while some PD-L1-high patients do not respond at all. The field is actively pursuing next-generation biomarkers, novel combination strategies, and approaches to overcome resistance, but the practicing oncologist should counsel patients honestly about these limitations while still conveying the genuine advances that checkpoint inhibitors represent.

Frequently Asked Questions

What PD-L1 TPS threshold predicts pembrolizumab monotherapy response in NSCLC?
PD-L1 TPS above 50% predicts robust response to pembrolizumab monotherapy in NSCLC. Patients with TPS 1-49% generally require combination chemoimmunotherapy for optimal outcomes.
What is the incidence of grade 3-4 immune-related adverse events with checkpoint inhibitors?
Grade 3-4 irAEs occur in 10-30% of patients receiving ICI therapy, while any-grade irAEs affect 60-80%. The most common include dermatologic toxicity, endocrinopathies, hepatitis, colitis, and pneumonitis.
When should I use IO-IO versus chemo-IO combinations in NSCLC?
CheckMate 227 and 9LA established nivolumab plus ipilimumab for IO-IO in NSCLC. The choice between IO-IO and chemo-IO depends on PD-L1 status, histology, performance status, and tolerance for immune toxicity.
Which biomarkers guide checkpoint inhibitor selection?
PD-L1 expression (TPS or CPS by tumor type), microsatellite instability (MSI-H/dMMR), and tumor mutational burden (TMB) are the primary biomarkers. Testing methodology varies by tumor type and specific ICI indication.
How should immune-related adverse events be managed?
Early recognition and grading-appropriate corticosteroid management is essential for irAEs. The most common events include dermatologic toxicity, thyroiditis, adrenal insufficiency, hepatitis, colitis, and pneumonitis, each requiring specific monitoring and intervention protocols.

Explore This Topic in Ailva

Ailva is a free clinical intelligence platform for NPI-verified US physicians. Get evidence-based answers with verified citations from 16M+ indexed papers — plus free CME credits.

Sam Anderson
Sam Anderson

Founder of Ailva.ai | Former Director of Research and Author of 200+ Medically Reviewed Articles | Editor-in-Chief of EudaLife Magazine