{
  "id": "tumor-mutational-burden",
  "title": "Tumor Mutational Burden",
  "lang": "en",
  "source_lang": "en",
  "summary": "Tumor mutational burden measures the density of mutations in a cancer’s DNA and is used in defined settings as a biomarker for response to immune checkpoint inhibitors.",
  "category": "health",
  "subcategory": "medicine",
  "type": "concept",
  "aliases": [
    "Tumour mutational burden",
    "TMB",
    "Tumor mutation burden",
    "Tumour mutation burden"
  ],
  "generated_at": "2026-11-03",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/tumor-mutational-burden/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/tumor-mutational-burden/",
      "json": "https://aiwiki.page/zh/tumor-mutational-burden.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Tumor mutational burden (TMB) is a measure of the number of somatic mutations in a cancer’s DNA, usually expressed as mutations per megabase of genomic sequence examined. Somatic mutations arise during a person’s lifetime, rather than being inherited through the germ line. TMB is studied and used as a biomarker for response to certain forms of cancer immunotherapy, particularly immune checkpoint inhibitors. It is not a measure of tumor size, cancer stage, or the amount of cancer in the body."
    },
    {
      "heading": "Biological basis",
      "text": "Cancer genomes acquire mutations through processes including errors during DNA replication, exposure to mutagens such as tobacco smoke or ultraviolet radiation, and defects in DNA repair. Mutation burdens therefore vary substantially between cancer types and between individual tumors of the same type.\n\nSome mutations alter the amino acid sequence of a protein. Fragments of these altered proteins can become neoantigens: molecular targets that the immune system may recognize as different from normal tissue. When presented on the surface of cancer cells, some neoantigens can be recognized by T cells.\n\nA higher number of mutations can increase the probability that a tumor produces recognizable neoantigens. This provides a biological rationale for an association between high TMB and response to checkpoint inhibitors, which block inhibitory signals that restrain immune activity.\n\nHowever, TMB is only an indirect measure of potential tumor immunogenicity. Many mutations do not produce an altered protein, are not expressed, or fail to generate a peptide that can be effectively presented and recognized. Immune responses also depend on antigen presentation, immune-cell access to the tumor, and local immune-suppressive mechanisms. A high-TMB tumor can therefore resist immunotherapy, while a low-TMB tumor can respond."
    },
    {
      "heading": "Measurement",
      "text": "Definition and calculation\n\nTMB is commonly represented as:\n\n\\[\\mathrm{TMB} = \\frac{\\text{number of eligible somatic mutations detected}} {\\text{megabases of sequence assessed}}\\]\n\nOne megabase is one million DNA bases. For illustration, 24 eligible mutations detected across 1.2 megabases would yield a TMB of 20 mutations per megabase.\n\nThe word eligible is important: there is no single counting procedure shared by all assays. Methods differ in whether they include synonymous substitutions, which do not change a protein’s amino acid sequence, and in how they count small insertions and deletions. Structural rearrangements and changes in gene copy number are generally not included in conventional TMB calculations.\n\nWhole-exome and targeted sequencing\n\nWhole-exome sequencing surveys most protein-coding regions of the genome and has been widely used to investigate TMB. Clinical tests often use next-generation sequencing panels that examine a selected set of cancer-associated genes. These panels estimate a broader mutation density from a smaller genomic region.\n\nPanel-based estimates depend on the amount and composition of DNA examined. Small panels are more susceptible to sampling variability, and panels enriched for frequently mutated cancer genes may require correction or calibration. Consequently, equal numerical results from different assays are not necessarily interchangeable.\n\nTesting also requires distinguishing tumor-acquired mutations from inherited variants. Sequencing matched non-tumor DNA can help make this distinction. Tumor-only methods instead use population databases and computational filters, which can leave uncertainty about the origin of some variants.\n\nTissue and blood measurements\n\nTissue TMB is measured using DNA extracted from a tumor specimen, often obtained by biopsy or surgery. Results can be affected by the proportion of tumor cells, DNA degradation, sequencing depth, and technical artifacts introduced during specimen processing.\n\nBlood-based TMB is estimated from circulating tumor DNA in plasma, an application of liquid biopsy. Its measurement depends partly on how much tumor DNA reaches the bloodstream. Low tumor DNA abundance can limit reliable estimation. Mutations arising in blood-cell populations can also complicate interpretation.\n\nBlood and tissue TMB are related measurements, but their sampling characteristics and assay thresholds differ. A cutoff established for one cannot automatically be applied to the other."
    },
    {
      "heading": "Clinical significance",
      "text": "Predicting response to checkpoint inhibitors\n\nStudies have found associations between higher TMB and response to checkpoint inhibition in several cancer settings, including melanoma and lung cancer. The strength and clinical usefulness of this association vary by tumor type, treatment, assay, and study population.\n\nTMB is generally investigated as a predictive biomarker: a characteristic associated with the likelihood of benefit from a particular treatment. That differs from a prognostic biomarker, which is associated with disease outcome independently of a specific treatment. An association between TMB and outcomes among treated patients does not, by itself, establish that TMB predicts greater benefit from that treatment than from an alternative.\n\nRegulatory use\n\nIn June 2020, the United States Food and Drug Administration granted accelerated approval to pembrolizumab for adults and children with unresectable or metastatic TMB-high solid tumors, defined as at least 10 mutations per megabase by an FDA-approved test, that had progressed following prior treatment and had no satisfactory alternative treatment options. FoundationOne CDx was approved as a companion diagnostic for this indication.\n\nThe approval was based on response rate and durability of response in a prospectively planned retrospective analysis of the KEYNOTE-158 clinical trial. It established a specific tissue-agnostic regulatory use: eligibility was defined by a molecular feature rather than a single organ of origin.\n\nThe threshold of 10 mutations per megabase is not a universal biological boundary. It belongs to a particular assay and clinical context, and does not imply that every tumor above it responds or that tumors below it cannot respond."
    },
    {
      "heading": "Relationship to other biomarkers",
      "text": "Microsatellite instability and mismatch repair deficiency\n\nMicrosatellite instability and DNA mismatch repair deficiency reflect abnormalities in a DNA repair system. Mismatch repair–deficient tumors often accumulate many mutations and can therefore have high TMB.\n\nThese measurements nevertheless describe different properties. TMB quantifies mutation density; microsatellite instability measures instability at repetitive DNA sequences; mismatch repair testing assesses the repair pathway. High TMB can arise without mismatch repair deficiency, including through ultraviolet damage, tobacco-associated mutagenesis, or certain defects in DNA replication proofreading.\n\nLynch syndrome is an inherited cancer predisposition syndrome involving mismatch repair genes. High TMB alone does not establish Lynch syndrome or show that a tumor’s mutations have an inherited cause.\n\nPD-L1 expression\n\nPD-L1 expression measures a protein involved in immune checkpoint signaling, rather than mutation density. TMB and PD-L1 expression can provide different information, and either can be high or low independently of the other. Their clinical uses depend on the specific cancer and treatment setting."
    },
    {
      "heading": "Limitations and standardization",
      "text": "Several factors constrain interpretation:\n\n- Assay dependence: Sequenced regions, mutation filters, technical sensitivity, and calibration methods affect the reported value.\n- Threshold uncertainty: TMB is continuous. Converting it into “high” and “low” categories can obscure differences, particularly near a cutoff.\n- Tumor heterogeneity: Different regions or metastases may contain different mutations. Mutations present throughout a tumor may have different immune implications from mutations restricted to small subpopulations.\n- Changes over time: Tumor evolution and treatment can alter the mutation profile, so samples collected at different times may produce different results.\n- Biological incompleteness: TMB does not directly measure neoantigen quality, immune-cell activity, or the ability of cancer cells to present antigens.\n- Evidence limitations: Findings from selected cohorts or single-arm trials do not automatically establish predictive value across all cancers or treatment combinations.\n\nStandardization efforts address reference materials, panel size, mutation-counting rules, and agreement between assays. Reporting the numerical result together with the testing method and its validated clinical context is essential to distinguishing a measured mutation burden from the interpretation attached to it."
    }
  ],
  "links": [
    "mutation",
    "cancer",
    "dna",
    "biomarker",
    "cancer-immunotherapy",
    "immune-checkpoint-inhibitor",
    "protein",
    "immune-system",
    "t-cell",
    "whole-exome-sequencing",
    "next-generation-sequencing",
    "biopsy",
    "liquid-biopsy",
    "lung-cancer",
    "clinical-trial",
    "microsatellite-instability",
    "dna-mismatch-repair",
    "lynch-syndrome"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}