Study on Real-world Outcomes of First-line Immune Checkpoint Inhibitor Therapy in Metastatic Melanoma Predicted by Tumor Mutational Burden

The treatment of metastatic melanoma has significantly benefited from immune checkpoint inhibitors (ICIs) in recent years. However, despite the remarkable improvement in survival rates among melanoma patients, the transient responses and immune-related toxicities of ICIs remain clinical limitations. Therefore, understanding and identifying biomarkers that can predict the efficacy of ICI treatment has become a crucial research direction. Among various predictive biomarkers, Tumor Mutational Burden (TMB) has been verified and approved as a predictive molecular marker for ICI therapy in metastatic solid cancers. TMB is defined as the number of somatic mutations per million base pairs and shows a correlation with the patient’s response to ICI inhibitors. This concept hypothesizes that the number of immunogenic neoantigens is proportional to TMB. ICI therapy can enhance or restore potential immune responses to these immunogenic neoantigens.

Research Background and Purpose

This study, authored by Dr. Miles C. Andrews, Dr. Gerald Li, Dr. Ryon P. Graf, Dr. Virginia A. Fisher, Dr. Jerry Mitchell, Dr. Ali Aboosaiedi, Dr. Harriet O’Rourke, Dr. Mark Shackleton, Dr. Mahesh Iddawela, Dr. Geoffrey R. Oxnard, and Dr. Richard S.P. Huang, was published in the JCO Precision Oncology journal by the American Society of Clinical Oncology (ASCO), DOI https://doi.org/10.1200/po.23.00640. The primary objective of the research was to evaluate the predictive value of TMB in advanced melanoma patients receiving first-line ICI therapy and to validate the practical effectiveness of TMB as a predictive marker through real-world data.

Research Methodology

Patient Cohort

The study utilized data from approximately 280 cancer clinics in the United States, encompassing around 800 treatment points. The data was sourced from the Flatiron Health - Foundation Medicine Inc (FMI) melanoma clinical genomic database (CGDB). With Institutional Review Board (IRB) approval allowing a waiver of informed consent, the study included 497 metastatic melanoma patients (257 receiving combined ICI therapy and 240 receiving single-agent ICI therapy) who had tissue-based TMB scores.

Clinical Genomic Analysis

The clinical genomic analysis (CGP) of patients was conducted using FoundationOne and FoundationOne CDx sequencing tests by Foundation Medicine, covering 404 and 324 genes, respectively. TMB was classified into TMB-L (<10 muts/mb), TMB-H (≥10 muts/mb), and TMB-VH (≥20 muts/mb). Clinical and pathological variables were extracted from electronic health records (EHR) and applied in multivariate analysis.

Research Process

Data Preprocessing and Analysis

The study initially performed data cleaning, excluding non-compliant samples such as those lacking systemic therapy lines, those without continuous 90-day EHR activity, and patients who did not receive first-line treatment after a metastatic diagnosis. Ultimately, complete data from 497 patients was included for analysis. A comparative description analyzed the characteristics of patients with TMB-L, TMB-H, and TMB-VH, examining the association between different TMB levels and clinical features like gender, BRAFV600E/K mutation status, metastasis sites, etc.

Statistical Analysis Methods

Multivariable Cox proportional hazards models and Kaplan-Meier survival estimates were used to evaluate the predictive value of TMB for first-line ICI treatment outcomes. The primary analysis considered the effects of TMB at multiple mutation burden thresholds (10 and 20 muts/mb) and further explored the interaction between BRAFV600E/K mutation status and TMB.

Main Research Results

Association of TMB with ICI Therapy Outcomes

1. Correlation Between TMB Levels and Survival Rates: The study found that TMB-H (≥10 muts/mb) predicted better Real-World Progression-Free Survival (RWPFS) and Overall Survival (OS) in both single-agent ICI and combined ICI patients. For instance, in the single-agent ICI treatment group, the progression-free survival rate of TMB-H patients was significantly higher than that of TMB-L patients (HR of 0.45), and this correlation remained significant in multivariate analysis (Figures 2a-2d).

2. Predictive Effects of Different TMB Thresholds: Further exploration found that TMB-VH (≥20 muts/mb) patients had the best prognosis in single-agent ICI treatment, whereas in combined ICI treatment, TMB-VH patient survival outcomes were inferior to those of TMB-H (10-19 muts/mb) patients. This inverse relationship suggests that TMB-VH patients may not be suitable for combined ICI treatment but are more appropriate for single-agent ICI therapy (Figures 3a-3d).

3. Unique Characteristics of TMB-VH Patients: TMB-VH patients more frequently presented with brain metastases and were more commonly found in males but were less likely to carry the BRAFV600E/K mutation. Additionally, TMB-VH patients’ responses to combined ICI were similar to those of single-agent ICI, further illustrating the necessity of evaluating both BRAFV600E/K mutation status and TMB together.

TMB and ICI Therapy Side Effects

While TMB is considered a predictor of ICI efficacy, the study found no significant relationship between TMB levels and immune-related adverse events (AEs). Consistent with expectations, patients undergoing combined ICI treatment more frequently required steroid medications to manage immune-related AEs compared to single-agent ICI patients.

Prognostic Significance of Combined Evaluation of BRAF Status and TMB

The combined evaluation of BRAF mutation status and TMB levels showed stronger predictive effects in both TMB-H (≥10 muts/mb) and TMB-VH (≥20 muts/mb) patients. Particularly for BRAFV600E/K mutant and TMB-VH patients, single-agent ICI might be more ideal. This underscores the importance of integrating multiple biomarkers and clinical pathologic characteristics in making first-line treatment decisions (Figure 4).

Conclusion and Research Value

This study validated the unique value of TMB in predicting ICI efficacy through real-world data and proposed the theoretical foundation for personalized therapy choices by combining BRAF mutation status and TMB levels. These results strongly support the implementation of comprehensive genomic testing, including TMB, in clinical practice, particularly for optimizing first-line ICI treatment management decisions. In summary, TMB as a predictive marker shows broad promise in metastatic melanoma, laying the foundation for precision medicine in immunotherapy.