Tumor mutational burden (TMB) and Microsatellite Instability (MSI) are now established biomarkers guiding immunotherapy decisions across a growing list of solid tumors. This article outlines how TMB and MSI are defined and assessed in VarSome Clinical and how they are related to regulatory approvals and clinical utility across different cancer types. By summarising the latest guidelines, FDA indications, and landmark trials, we highlight where these biomarkers are already shaping treatment and where their role is still emerging.
TMB is the count of somatic, coding mutations per megabase of interrogated DNA (mut/Mb). A common clinical definition for TMB-high is ≥10mut/Mb. Mechanisms associated with TMB include: loss or weakening of DNA repair pathways (e.g., mismatch repair loss, POLE mutations); exogenous mutations (UV, tobacco smoke, some chemotherapies); novel mutations may result in neoantigen production, increasing the likelihood that T cells recognise the tumor once immune checkpoints are blocked.
MSI describes somatic insertions or deletions that change the length of microsatellites (short tandem repeats) when tumor DNA is compared with matched normal DNA. These repetitive tracks are prone to slippage during replication. When mismatch repair (MMR) activity is lost or inactive, the slippage loops are not repaired, and the repeat units expand or contract, producing an MSI signature. Loss of MMR also allows replication errors to persist in SNVs and short indels throughout the genome. Consequently, most MSI high (MSI-H) tumors also meet TMB-high thresholds.
VarSome Clinical estimates TMB in tumor-only, targeted analyses of WES data (≥ 31 Mb capture, non-WGS) or Agilent SureSelect Cancer CGP panels, provided the mean coverage reaches 50x. We report two metrics:
To keep noise to a minimum, the pipeline calls variants with Sentieon TNHaplotyper2, flags likely germline sites with a panel of normals and public germline databases, then applies additional filters to remove any residual germline variants, resulting in a TMB value more reflective of the true somatic load. Samples are classified to reflect the following thresholds:
TMB scores are capped at 40 mut/Mb for reporting purposes, even if the calculated value exceeds this limit.
For more information on how we calculate TMB status, please visit: http://docs.varsome.com/en/tmb
Unlike traditional MSI detection methods that require matched normal samples, VarSome Clinical enables accurate MSI evaluation from tumor-only sequencing data. It supports somatic analyses starting from FASTQ with various assays, currently including:
The tool outputs the MSI score (High or Not Detected) as the fraction of unstable sites (sites with evidence of indels or misalignment in microsatellite repeats) to qualifying sites (sites with adequate coverage analyzed for instability).
For more information on how we calculate MSI status, please visit: https://docs.varsome.com/en/msi
In 2022, colorectal cancer (CRC) was the third most common cancer globally, with more than 1.9 million cases based on estimates from the International Agency for Research on Cancer (IARC)1.
MSI-H CRCs make up around 15% of early-stage CRCs; approximately 12% are sporadic cases, and 3% are linked to Lynch syndrome, which carries an estimated 80% lifetime risk of CRC2. However, MSI-H prevalence drops to around 5% in metastatic CRC3.
Immunotherapy, specifically Pembrolizumab (PD-1 inhibitor), has demonstrated favorable outcomes in MSI-H CRCs with an objective response rate (ORR) of 40% and an immune-related progression-free survival rate of 78%4.
While MSI-H CRCs commonly exhibit high TMB, subsets of microsatellite stable (MSS) CRCs may present with elevated TMB due to alterations in KRAS signalling, DNA damage repair (DDR) genes, or epigenetic regulators5. Evidence suggests TMB may have prognostic and therapeutic relevance even in MSS metastatic colon cancer6.
By contrast, MSS CRCs typically do not benefit from PD-1 blockade alone, reinforcing the importance of MSI status as a predictive biomarker. Nonetheless, subsets of MSS CRCs with high TMB, including those with POLE mutations, may still respond to immunotherapy7,8.
These molecular features have translated into regulatory approvals for immune checkpoint inhibitors (ICIs), particularly in MSI-H CRC. The table below outlines current biomarker-driven approvals relevant to colorectal cancer.
|
Biomarker |
FDA-Approved Therapy |
Indication / Notes |
|
MSI-H / dMMR |
Pembrolizumab (Keytruda) – 2017 (accelerated) Any unresectable/metastatic MSI-H solid tumor after prior therapy (first tissue-agnostic approval; included CRC)9. Pembrolizumab – 2020 First-line unresectable/metastatic MSI-H CRC (improved PFS vs chemo)10. Nivolumab (Opdivo) – 2017 (accelerated) Refractory MSI-H CRC (≥2 prior lines of chemo)11. Nivolumab + Ipilimumab (Yervoy) – 2018 (accelerated) Metastatic MSI-H CRC (combination post-chemo)12. Nivolumab + Ipilimumab – 2025 (full approval) Unresectable/metastatic MSI-H CRC (confirmed benefit; combo vs single-agent)13. |
Immune checkpoint inhibitors targeting PD-1 (± CTLA-4) show high and durable response rates in MSI-H CRC. The 2017 accelerated approvals were based on multi-cohort trials showing ~40% response in MSI-H CRC9. Combination nivolumab/ipilimumab yields a higher response (~46%) than nivolumab alone in MSI-H CRC12. |
|
TMB-High |
No CRC-specific label (implicit via pan-tumor TMB approval) – Pembrolizumab 2020 for TMB ≥10 mut/Mb14. |
Nearly all MSI-H CRCs are TMB-high. Additionally, a POLE or POLD1 mutation can cause ultra–high TMB in CRC even if MSS. Such tumors (~1% of mCRC) have shown exceptional responses to immunotherapy (ORR ~89%)7, even exceeding typical MSI-H CRC outcomes (ORR ~54%). |
Endometrial cancers (ECs), the sixth most common cancer in women15, often exhibit molecular features with therapeutic implications. Approximately 26%-31% of endometrial carcinomas are classed as MSI-H or deficient MMR (dMMR)16,17. Additionally, POLE exonuclease domain mutations (POLEmut) occur in roughly 8% of endometrial carcinomas, based on pooled estimates from metanalyses18,19.
Both MSI-H and POLEmut tumors are associated with high TMB and are known to be responsive to immunotherapy. Therefore, MSI status can guide treatment decisions: MSI-H tumors can receive immunotherapy alone, whereas MSS tumors may require combination therapy.
ESMO guidelines recommend dMMR testing for all recurrent or advanced ECs20. For MSI-H tumors, dostarlimab (PD-1 inhibitor) has shown efficacy in EC, with an ORR around 42%21, leading to accelerated FDA approval in April 2021 for dMMR recurrent/advanced EC after platinum therapy. ECs also fall under the tumor-agnostic pembrolizumab approval for MSI-H/dMMR tumors.
For the majority of ECs, which are MSS, combining immunotherapy with a targeted agent has proven beneficial. In the KEYNOTE-775 trial, pembrolizumab + lenvatinib significantly improved progression-free survival (PFS) and overall survival (OS) compared with chemotherapy alone22.
|
Biomarker |
FDA-Approved Therapy |
Indication / Notes |
|
MSI-H/dMMR |
Pembrolizumab (Keytruda) – 2017 Pan-tumor MSI-H approval (includes recurrent endometrial cancer after prior treatment)9. Dostarlimab (Jemperli) – 2021 (accelerated) dMMR recurrent or advanced endometrial cancer after platinum; converted to full approval in 2023 based on confirmatory data23. |
PD-1 inhibitors for dMMR endometrial cancer have shown high efficacy after chemotherapy. Dostarlimab demonstrated a 45% ORR in dMMR endometrial cancer, leading to approval. Pembrolizumab is likewise used under the tumor-agnostic MSI-H indication23. |
|
MSS |
Pembrolizumab + Lenvatinib – 2019 (accelerated); 2021 (full)24. |
Approved for advanced endometrial carcinoma that is NOT MSI-H/dMMR after prior systemic therapy. This approval explicitly excludes MSI-H cases (which should get immunotherapy alone). The combo (anti–PD–1 + anti-angiogenic) improved survival over chemotherapy in MSS disease (KEYNOTE-775)24. |
|
TMB-High |
No specific label beyond tissue-agnostic. Pembrolizumab 2020 (TMB ≥10)14. |
Many MSI-H or POLE-mutated endometrial tumors qualify as TMB-high. POLE-ultramutated tumors, though rare (~7% of certain subtypes), have TMB levels often >100 mut/Mb and excellent outcomes with immunotherapy7. These are managed similarly to MSI-H cases. |
Non-Small Cell Lung Cancer (NSCLC): While MSI-H is exceedingly rare in lung tumors (~0.4%)25, high TMB has been shown to indicate improved PFS and ORR with checkpoint inhibitors. In CheckMate-227, first-line nivolumab + ipilimumab doubled 1-year progression-free survival versus chemotherapy among patients with TMB ≥ 10 mut/Mb, regardless of PD-L1 status26. High TMB is often associated with smoking-related NSCLCs, which tend to have high somatic loads, potentially leading to increased neoantigen load, enhanced immunogenicity27. PD-L1 expression is the primary biomarker for selecting patients for ICI therapy in NSCLS, with high TMB yet to be officially recognised as a standalone biomarker for immunotherapy selection. However, TMB has emerged as an additional factor influencing outcomes, and NSCLC falls under the tumor-agnostic remit of pembrolizumab.
Small Cell Lung Cancer (SCLC): Like NSCLC, SCLC tumors may exhibit high TMB, often associated with tobacco carcinogenesis, which may contribute to their initial responsiveness to immunotherapy28. However, unlike NSCLC, SCLC does not have established biomarkers, such as PD-L1 expression or MSI-H, to indicate immunotherapy efficacy; dMMR is rare in lung cancer, affecting around 1.9% of SCLC patients25. TMB has been explored as a predictive biomarker in SCLC clinical trials, with some evidence that high TMB may be associated with improved responses to checkpoint inhibitors29. Nevertheless, the clinical utility of TMB in SCLC remains under investigation, and no standardized thresholds have been established.
|
Biomarker |
Therapy |
Indication / Notes |
|
TMB-High (NSCLC) |
No NSCLC-specific TMB requirement - (PD-1/L1 inhibitors are approved irrespective of TMB, with PD-L1 as the main marker). The pan-cancer pembrolizumab TMB ≥10 approval may apply in difficult-to-manage cases14. |
In the CheckMate-227 trial, first-line nivolumab + ipilimumab showed significantly improved PFS in TMB-high NSCLC (≥10 mut/Mb). One-year PFS was 43% vs 13% on chemo (HR 0.58). This benefit was seen regardless of PD-L1 status26. TMB-high also correlated with better OS in multiple studies. However, FDA approvals for NSCLC focus on PD-L130. |
|
TMB-High (SCLC) |
No formal biomarker approval. |
Retrospective analyses suggest that SCLC patients with higher TMB respond better to combined PD-1/CTLA-4 blockade. In the CheckMate-032 trial, the TMB-high subgroup showed an objective response rate of ≈ 21% versus ≈ 5% in TMB-low and experienced longer survival on nivolumab ± ipilimumab. These findings prompted exploratory use of TMB enrichment in later SCLC immunotherapy studies29. The FDA granted nivolumab accelerated approval for third-line SCLC in 2018 without any biomarker requirement. This indication was voluntarily withdrawn in 2021 after phase-3 confirmatory trials (CheckMate 451 and 331) failed to improve overall survival31. Researchers are now testing whether TMB or other tumour features can better identify SCLC subsets most likely to benefit from immunotherapy. |
Stomach and gastroesophageal junction (GEJ) adenocarcinoma: Estimates of MSI-H incidence in gastric cancers (GCs) range from 6-25%32–34. These cancers qualify for pembrolizumab monotherapy under the FDA’s tumor-agnostic approval for MSI-H/dMMR solid tumors. Front-line nivolumab + chemotherapy gained approval in April 2021 after CheckMate-649 showed that MSI-H cases derived a marked OS benefit (HR 0.34) compared to chemotherapy alone35. Reflecting this and other data, NCCN recommends universal MSI/MMR testing in all newly diagnosed gastric and GEJ cancers to inform immunotherapy eligibility36.
High TMB in gastric and GEJ adenocarcinomas is uncommon but clinically meaningful. In CheckMate-649, 8% of evaluable tumors were TMB-high35. These patients displayed improved OS rates when treated with nivolumab + chemotherapy versus chemotherapy alone (HR 0.48). The benefit was smaller in TMB-low cases, suggesting that mutation load enhances, but is not essential for, successful PD-1 blockade. Nivolumab + chemotherapy has FDA approval in gastric/GEJ cancers regardless of TMB status. However, gastric/GEJ cancers that meet the ≥10 mut/Mub cut-off also qualify for pembrolizumab single-agent under the tumor-agnostic TMB-high indication.
Esophageal adenocarcinoma (EAC): MSI-H is uncommon. A 2021 study places incidence at around 1.4% in a 1965 patient German cohort37. However, NCCN guidelines recommend MSI/MMR testing at diagnosis in all new cases to determine candidates for immunotherapy36. Nivolumab + chemotherapy is an approved treatment for EAC irrespective of MSI status. Nevertheless, MSI-H patients show marked increases in relative survival gain35.
High TMB is more common in EAC; one review estimates TMB-high (>10 mut/Mb) in 9% of EAC38. Patients with high TMB show the greatest improvement in OS following nivolumab + chemotherapy therapy (HR 0.48), but CheckMate-649 only included a small esophageal cohort (<10%)35.
Pancreatic adenocarcinoma: dMMR incidence in pancreatic ductal adenocarcinoma (PDAC) is estimated to be around 1-2%39 with a similar Incidence of TMB-high40. Where MMR/TMB-High does occur, these pancreatic tumors are eligible for pembrolizumab under the tissue-agnostic label, and NCCN advises routine MSI/MMR/TMB testing in metastatic disease41.
Cholangiocarcinoma (Bile Duct cancer): MSI-H/dMMR is identified in approximately 1.3% of cholangiocarcinomas42. The KEYNOTE-185 and -028 cohorts displayed meaningful responses to pembrolizumab in these rare cases, supporting the use of PD-1 blockade when MSI-H is detected43. One study of Korean patients suggests the incidence of TMB-high in advanced cholangiocarcinomas to be as high as 18.5% (≥10Mut/Mb)44. These patients likewise fall under the tissue-agnostic FDA approval of pembrolizumab.
Melanoma: Melanoma is associated with a high mutational burden, driven by pervasive ultraviolet DNA damage30. Melanoma patients with TMB-high have shown significantly improved OS (HR 0.48) following anti-CTLA-4 or anti-PD-(L)-1 inhibitor monotherapy compared with TMB-low patients45. MSI-H disease is exceedingly rare, therefore, dMMR testing is uncommon in clinical practice.
Cervical Cancer: The incidence of MSI-H in cervical cancers is around 1.3%46. High TMB is more common, estimated to be 15-28%46,47. NCCN guidance recommends PD-L1, TMB, and MSI testing for all patients, and pembrolizumab as a second-line therapy48.
Prostate Cancer (PC): A 2024 PC cohort showed MSI-H/dMMR and TMB-high/MSS in 2.5% and 1.5% of tumors, respectively. MSI-H patients responded better to immune checkpoint blockade, suggesting MSI, not TMB alone, is the key predictor of benefit49. In response to KEYNOTE-199, a NCCN panel upgraded pembrolizumab to a category 2A option for MSI-H/dMMR metastatic castration-resistant PC after docetaxel and/or novel hormone therapy. Additionally, this panel recommends that PC patients with MSI-H/dMMR be referred for genetic counselling to evaluate Lynch syndrome risk50.
TMB and MSI have already proven to be clinically actionable biomarkers that influence treatment decisions and help predict immunotherapy response. As testing becomes increasingly routine, guidelines will advance and their relevance will grow. Understanding how and where to apply these biomarkers will be key to helping deliver personalized, evidence-based care.
The VarSome Clinical TMB/MSI workflow is for Research Use Only. Not for diagnostic procedures.