Full Approval for Keytruda in TMB-H: Why Was the Threshold for Tumor Mutational Burden Raised to 13?

It Is the Mutation Count, Not the Tumor Site: Pembrolizumab Now Has Full Approval in High TMB, but the Threshold Has Risen

On October 16, 2026, the US Food and Drug Administration (FDA) converted the accelerated approval of pembrolizumab (Keytruda) and its subcutaneous formulation (Keytruda Qlex) for advanced solid tumors with high tumor mutational burden (TMB-H) to traditional (full) approval. With one important change: the "high TMB" threshold was raised from 10 to 13 mutations per megabase, because benefit could not be confirmed in patients with values between 10 and 13. In 220 patients with TMB ≥13, the response rate was 36%. In this article, we look at what the approval changes, what TMB is, and for whom it matters. We have previously covered pembrolizumab (Keytruda) and tissue-agnostic cancer therapies.

Why does this matter?

For many years, cancer treatment was planned according to the organ in which the tumor arose. Over the past decade, however, some drugs have begun to be approved based on a specific molecular feature of the tumor, regardless of the organ. The 2020 TMB-H approval of pembrolizumab was one of the first instances of an immunotherapy being approved across all solid tumors based solely on "mutation count," and it became one of the most debated approvals: critics argued that high TMB does not predict response to immunotherapy in every cancer type and that the cutoff of 10 mutations/megabase was arbitrary.

This full approval matters for two reasons. First, it shows that the review of accelerated approvals against confirmatory data is working: the FDA removed the patient group in whom benefit could not be confirmed. Second, for patients who have run out of options and those with rare tumors, it confirms that an evidence-based immunotherapy option exists when the appropriate biomarker is present.

What is tumor mutational burden (TMB)?

Tumor mutational burden (TMB) is the number of mutations per megabase (one million base pairs) in the DNA of tumor cells. As the number of mutations increases, the tumor produces more abnormal proteins that the immune system can recognize as "foreign." For this reason, tumors with high TMB may respond better to immunotherapy. Under the FDA's current definition, high TMB means at least 13 mutations per megabase.

TMB is calculated with comprehensive next-generation sequencing panels performed on tumor tissue. FoundationOne CDx was used as the companion diagnostic in the 2020 approval. The main causes of high TMB are mutagenic exposures such as smoking (lung cancer) and sunlight (melanoma), DNA repair defects (MSI-H/dMMR, POLE mutations), and the intrinsic biology of certain tumor types. As we discussed in our article on the proteogenomic map, the effect of a mutation can also depend on where in the gene it occurs.

What data is the approval based on?

Response rates were assessed by blinded independent central review. The safety profile was consistent with the known adverse event profile of pembrolizumab; the main risks are immune-mediated adverse events (such as inflammation of the lungs, bowel, liver, thyroid and kidneys), infusion reactions, and fetal harm during pregnancy. Early recognition and management of immunotherapy side effects (including kidney and persistent skin side effects) is important.

Why was the threshold raised to 13?

According to the FDA, efficacy sufficient to support full approval could not be confirmed in patients with TMB between 10 and 13. This is a rare example of a biomarker threshold being recalibrated on the basis of confirmatory data. The practical consequence: a patient whose TMB has been reported as 10–12 at any time since 2020 is not covered by this indication under the new approval.

Does high TMB mean the same thing in every cancer?

No. A study published in the Annals of Oncology in 2021, which analyzed more than 10,000 tumors from 31 cancer types, showed that high TMB predicts response to immunotherapy only in some cancers. In cancers where the number of tumor-infiltrating immune cells (CD8+ T cells) rises along with mutation count (such as melanoma, lung and bladder cancer), patients with high TMB had markedly higher response rates. In cancers where this relationship was not seen (such as breast, prostate and brain tumors), high TMB did not predict response.

So although the approval applies to "all solid tumors," the meaning of TMB should be interpreted according to tumor type. In addition, most MSI-H/dMMR tumors already have high TMB, and these patients have a separate approval supported by stronger evidence; Cohort M in the new approval specifically includes non-MSI-H, TMB-H tumors.

How should the results be interpreted?

Strengths

• The findings of the accelerated approval were confirmed in a separate cohort (M) that prospectively enrolled only TMB-H patients; the number of patients doubled.
• Responses were assessed by independent central review, and in more than half of responders the response lasted at least one year.
• The group in whom benefit was not confirmed (TMB 10–13) was removed from the approval, an evidence-based correction.

Points to keep in mind

• No comparator arm: This is a single-arm study; the approval, based on response rate and duration of response, did not demonstrate an overall survival or progression-free survival benefit.
• Differences by tumor type: The FDA announcement did not report the distribution of the 220 patients by tumor type or the response rate in each type. There are data indicating that high TMB does not predict response in breast, prostate and brain tumors.
• Assay variability: TMB values can vary with the panel, calculation method and laboratory used; results from different assays may not be interchangeable. The approval relies on TMB determined by "an FDA-approved test." Blood-based TMB (liquid biopsy) is not the same as tissue TMB; the data supporting the approval are based on tissue testing.
• Retrospective component: One third of the patients were identified as TMB-H after the fact, in cohorts set up for other purposes.

What does it mean in practice?

• Who should be considered: In patients who have exhausted standard therapies or have no effective standard treatment, particularly those with rare tumors (such as cervical, anal, salivary gland, thyroid and vulvar cancer, and neuroendocrine tumors), the TMB value should be checked if comprehensive panel testing has been done.
• The threshold is now 13: Even if TMB was flagged as "high" on an older report, a patient with a value of 10–12 falls outside the new approval.
• Other biomarkers first: If targets with stronger evidence are present, such as MSI-H/dMMR, NTRK, BRAF, RET or HER2 (trastuzumab deruxtecan), they take priority. In cancers that have their own organ-specific immunotherapy approval (such as lung cancer, melanoma and bladder cancer), those approvals apply.
• Tumor type matters: Evidence that high TMB predicts response in breast, prostate and brain tumors is weak; in these patients, the decision should be made carefully and in a tumor board.
• Expectations should be realistic: Roughly one in three patients responds; in more than half of responders, the response lasts longer than a year. In patients who do not respond, a change of treatment should be planned in a timely manner.
• Subcutaneous option: The subcutaneous formulation (Keytruda Qlex), approved in the US in September 2025, can also be used in this indication; administration time is shorter.

Practical tips for patients and families

• If you have previously had comprehensive genomic panel testing, look for the "TMB" or "tumor mutational burden" value and its unit (mutations/megabase) in your report, and interpret it together with your oncologist.
• If your TMB is between 10 and 13, be aware that this approval no longer covers you; ask about other options and clinical trials.
• During immunotherapy, report any new symptoms such as diarrhea, shortness of breath, cough, extreme fatigue, skin rash or jaundice without delay.
• It is important that the treatment decision is made by an experienced team that also takes your tumor type and other biomarkers into account; if needed, seek a second opinion.

Frequently asked questions

What does high tumor mutational burden (TMB) mean?

It means that the tumor DNA carries a large number of mutations per megabase. Under the FDA's current definition from October 2026, high TMB is at least 13 mutations per megabase. Tumors with high TMB may be more "visible" to the immune system and may respond better to immunotherapy.

If my TMB is high, will pembrolizumab definitely work?

No. In the study supporting the approval, about 36% of patients with TMB ≥13 responded to treatment. The likelihood of response also varies by tumor type; there are data indicating that high TMB does not predict response in breast, prostate and brain tumors.

My TMB came back as 11. Does this approval cover me?

No. The threshold in the 2020 approval was 10; however, because benefit could not be confirmed in patients between 10 and 13, the FDA raised the threshold to 13. In this situation, it is advisable to discuss other treatment options and clinical trials with your oncologist.

Is pembrolizumab for TMB-H reimbursed by SGK in Türkiye?

No. The TMB-H indication is not reimbursed in Türkiye; use is generally possible only with patient-specific off-label approval from TİTCK, with the cost of the drug borne by the patient.

Glossary

Tumor mutational burden (TMB)

The number of mutations per megabase in tumor DNA.

Megabase (Mb)

One million DNA base pairs.

Tissue-agnostic (tumor-agnostic) approval

Approval of a drug based on a specific molecular feature, regardless of the organ in which the tumor arose.

Accelerated approval

An FDA approval based on early data that must be verified in confirmatory studies.

Traditional (full) approval

A permanent FDA approval granted once benefit has been confirmed with adequate data.

MSI-H / dMMR

Biomarkers indicating a defect in the system that repairs mismatch errors in DNA; TMB is usually high in these tumors.

Companion diagnostic

A test approved alongside a drug to identify the patients likely to benefit from it.