Immune checkpoint blockade (ICB) has transformed cancer treatment, gaining approval as first- or second-line therapy across an expanding range of metastatic cancers and increasingly being explored for early-stage tumors. However, clinical responses remain limited, with only 10% to 40% of cancer patients achieving potentially long-lasting responses, depending on the malignancy subtype.
The search for reliable predictive biomarkers has become critical as researchers work to identify patients most likely to benefit from these therapies while avoiding unnecessary exposure to potential side effects in non-responders. Currently, only PD-L1 expression in tumor tissues is used as a predictive biomarker in clinical practice, but its limitations are significant - some patients with absent PD-L1 tumor expression may still respond to PD-1 blockade, making treatment decisions challenging.
Dynamic Biomarkers Offer New Approach
A promising emerging strategy focuses on dynamic biomarkers that can be tested early after treatment initiation to identify patients experiencing immune response failure. These biomarkers monitor changes in the immune system and tumor environment during checkpoint inhibitor treatment, providing insights into treatment mechanisms and patient outcomes.
Recent research has compiled advances in four complementary areas of dynamic biomarker development, offering new perspectives on predicting checkpoint inhibitor efficacy.
Monitoring Tumor-Specific Immune Responses
Studies have demonstrated that tumor mutational burden (TMB) serves as a useful biomarker for predicting immunotherapy response. Research indicates that high TMB combined with high PD-L1 expression can predict favorable outcomes in lung cancer patients. In elderly melanoma patients, increased TMB profiles may restore age-related immune dysfunction, leading to favorable immune responses comparable between patients younger or older than 75 years.
Researchers have also identified specific genetic signatures associated with treatment response. An integrative analysis showed that CTNNB1 (catenin beta-1) gene mutations, associated with better prognosis in multiple tumors, could help improve clinical outcomes in hepatocellular carcinoma patients receiving immunotherapy. This mutation was linked to increased tumor-infiltrated NK cells and downregulation of immunoinhibitory genes.
The NY-ESO-1 cancer testis antigen has emerged as both a dynamic biomarker and potential immunotherapy target. Studies show that melanoma patients treated with ipilimumab had increased rates of NY-ESO-1-specific immunity associated with improved clinical benefit, particularly in patients developing both NY-ESO-1-specific antibodies and CD8+ T cells.
Immune Cell Phenotypic Markers
Phenotypic characteristics of immune cells play important roles in predicting patient prognosis and treatment response. Research has identified specific immune cell populations that correlate with treatment outcomes.
In patients responding to checkpoint inhibitors, studies have identified a "favorable immune periphery" characterized by high levels of circulating CD8+ PD-1+ T cells, effector-memory CD8+ T cells, and abundant dendritic cells, combined with low levels of myeloid-derived suppressor cells and monocytes at baseline.
Advanced technologies including multi-omics, single-cell sequencing, flow cytometry, and mass cytometry are accelerating the characterization of immune cell functional phenotypes before and during treatment. These approaches enable exploration of immune cell dynamics and discovery of novel biomarkers.
Cytokine and Chemokine Profiles
Cytokines and chemokines serve as pivotal biomarkers in immune cell activity, with differential expression levels observed during checkpoint blockade that can help monitor clinical outcomes.
Studies have shown that several cytokines and chemokines involved in immune activation are upregulated after ICB treatment. In head and neck squamous cell carcinoma patients, two biomarkers were reduced at progression: interleukin (IL)-10 and CX3CL1 (both p < 0.0001). Conversely, cytokines contributing to immune inhibition were downregulated after treatment, with IL-6 and IL-8 increased at progression (both p < 0.0001).
Research has also demonstrated that downregulation of IFN-γ, tumor necrosis factor-α, and specific interleukins, combined with upregulation of perforin and other immune markers, correlated with disease resolution in metastatic gastric cancer patients treated with combined radioimmunotherapy.
Treatment-Related Toxicity Biomarkers
Immune checkpoint inhibitors can cause immune-related adverse events (irAEs) that may lead to severe outcomes and treatment discontinuation. Identifying biomarkers for early recognition and management of these toxicities is critical for clinical practice.
Studies have identified combinations of biomarkers useful for predicting ICB toxicity, including blood-based markers such as neutrophil-to-lymphocyte ratio, absolute lymphocyte count, autoantibodies, and various serum cytokines and chemokines. Immunogenetic factors including single nucleotide polymorphisms and human leukocyte antigen subtypes, as well as microbial biomarkers related to microbiome composition, also show predictive potential.
Research suggests that demographic parameters such as female gender and obesity, along with preexisting clinical factors including high eosinophil or white blood cell counts and pre-existing autoimmune disease, could help identify patients at higher risk of developing irAEs.
Future Directions
The study of immune system and tumor microenvironment dynamics during checkpoint blockade is crucial for predicting favorable responses and managing adverse events. Current dynamic predictive biomarkers remain in the exploration phase with limited reliability on an individual patient basis.
Development of models combining multiple variable dynamics using novel approaches such as multi-omics, single-cell analysis, and rigorous biostatistics and bioinformatics tools represents a key strategy for identifying reliable predictive dynamic biomarkers. These advances could pave the way toward more personalized, beneficial, and safer checkpoint inhibitor therapy.
The research emphasizes that while individual biomarkers show promise, the complex nature of immune responses suggests that comprehensive approaches incorporating multiple dynamic markers will likely provide the most accurate predictions of treatment success.
