Microsatellite Instability and Tumor Mutational Burden Testing AHS - M2178

Description of Procedure or Service

Tumor mutational burden (TMB) is the genetic characteristic of non-inherited mutations within tumor tissue, often reported as the total number of DNA mutations per one million bases (megabase). Original studies calculated TMB based on whole-exome sequencing and reported TMB as the number of mutations that exist within the exome. However, TMB testing has expanded to targeted gene sequencing panels that do not cover the entire exome. TMB may serve as a biomarker to identify patients likely to have a favorable response to immunotherapy, as high TMB levels correlate with objective response rates to immunotherapy in several different cancer types (Ritterhouse, 2019; C.Willis et al., 2019).

Microsatellites are short, repetitive segments of DNA that are highly prone to mutation. Microsatellite instability (MSI) in tumor DNA is defined as the presence of alternate sized repetitive DNA sequences that are not present in the corresponding germline DNA (Nojadeh, et al., 2018). Tumors with high microsatellite instability (MSI-H) are more immunogenic and may therefore respond to drugs that activate the immune system.

Related Policies:

AHS-G2054 Liquid Biopsy

AHS-M2004 Lynch Syndrome

AHS-M2026 Testing for Colorectal Cancer Management

AHS-M2030 Testing for Targeted Therapy of Non-Small Cell Lung Cancer

AHS-M2065 Molecular Profiling for Cancers of Unknown Primary Origin

AHS-M2146 General Genetic Testing, Somatic Disorders

AHS-M2168 Proteogenomic Testing of Individuals with Cancer

AHS-M2171 Esophageal Pathology Testing

***Note: This Medical Policy is complex and technical. For questions concerning the technical language and/or specific clinical indications for its use, please consult your physician.

Policy

BCBSNC will provide coverage for microsatellite instability and tumor mutational burden when it is determined to be medically necessary because the medical criteria and guidelines shown below are met.

Benefits Application

This medical policy relates only to the services or supplies described herein. Please refer to the Member's Benefit Booklet for availability of benefits. Member's benefits may vary according to benefit design; therefore, member benefit language should be reviewed before applying the terms of this medical policy.

When Microsatellite Instability and Tumor Mutational Burden is covered

1. For individuals being considered for immune checkpoint inhibitor (ICI) targeted therapy, tissue -based tumor mutational burden (TMB) and/or microsatellite instability (MSI) testing (Note 1) for the solid tumors outlined in Note 2, is considered medically necessary.
2. When tissue-based testing is infeasible (i.e., quantity not sufficient for tissue-based test or invasive biopsy is medically contraindicated) for an individual being considered for ICI targeted therapy, liquid biopsy based TMB and/or MSI testing for the solid tumors outlined in Note 2 is considered medically necessary.

When Microsatellite Instability and Tumor Mutational Burden is not covered

TMB and/or MSI testing for all other tumors and in any other situations not listed above, is considered not medically necessary.

Note 1: NGS panels designed to provide a TMB score are allowed with no restrictions on the number of genes being tested, so long as the panel is designed for TMB assessment.

Note 2: Table of all solid tumors that will be allowed for TMB or MSI testing.

MSI-H (MSI-HIGH; MSI-POSITIVE); NCCN DEFINES TMB-H AS A TMB VALUE OF >10 MUTATIONS/MEGABASE

Policy Guidelines

Tumor mutational burden (TMB) is an emerging biomarker associated with predicting the response to immune checkpoint inhibitors (ICIs)– therapies that have made significant progress in helping to treat certain advanced cancers. ICIs work by releasing the brakes on the immune system’s antitumor response and ICI therapy has proven most effective on tumor types with a high TMB, whereby a high TMB value indicates better treatment outcomes (Yarchoan, et al., 2017).

Clinical biomarkers are widely used for making personalized and actionable decisions for cancer treatment. Many mutations result in production of numerous altered peptides—the more mutations present, the larger the number of altered peptides being produced. A subset of these peptides is expressed and processed by the major histocompatibility complex, resulting in neoantigens. These neoantigens can be recognized by the immune system to generate an antitumor response. The recognition of these neoantigens is thought to be largely stochastic and a higher number of DNA mutations means a higher number of candidate peptides being produced, resulting in a greater likelihood of these neoantigens being recognized by the immune system (Ritterhouse, 2019; Yarchoan et al., 2019).

This phenomenon is observed in patients with melanoma, non-small-cell lung carcinoma (NSCLC), small cell lung cancer, bladder cancer, colorectal cancer, microsatellite instability cancers, and pan-tumors who are on inhibitors of programmed cell death protein (PD-1) and the ligand for programmed cell death protein 1 (PD-L1) (PD-1/PD-L1). High TMB has also been associated with improved outcomes in patients on a combination of PD-1/PD-L1 and cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) inhibitors (Hellmann, Callahan, et al., 2018; Hellmann, Nathanson, et al., 2018; Le et al., 2015; Merino et al., 2020; Rizvi et al., 2015; Snyder et al., 2014; Van Allen et al., 2015).

Originally, TMB was measured with whole-exome sequencing (WES). In theory, WES is the best measure of TMB, as it covers every coding sequence within the tumor, ensuring that every mutation contributes to the TMB score. Recently, however, studies have shown that TMB measured using targeted panels which analyze mutations across specific genes (rather than the whole exome) correlates well with TMB measured by WES (Chalmers et al., 2017; Quy et al., 2019; Wu, et al., 2019). WES method has limited clinical utility as it is considered costly, has a high turnaround time (roughly six-to eight-week sequencing period), and is difficult for many clinical laboratories to use for routine patient care (Ritterhouse, 2019). Hence, targeted next generation sequencing (NGS)-based panels that focus analysis on a subset of genes implicated in cancer biology offer several advantages for TMB estimation (Garofalo et al., 2016).

There are FDA-approved products for calculating TMB include the FoundationOne CDx™ (F1CDx) assay and the FoundationOne Liquid CDx assay (Foundation Medicine Inc.) (FDA, 2017c, 2020b), as well as an additional FDA 510(k) authorized assay, MSK-IMPACT (Memorial Sloan Kettering Cancer Center) (FDA, 2017a, 2017b). These tests, referred to collectively as comprehensive genomic profiling (CGP), can identify all types of "molecular alterations (i.e., single nucleotide variants, small and large insertion‐ deletion alterations, copy number alterations, and structural variants) in cancer‐related genes, as well as genomic signatures such as microsatellite instability (MSI), loss of heterozygosity [LOH], and TMB” (Klempner et al., 2020). Studies show that TMB calculated from CGP has high concordance with TMB measured from WES. On June 16, 2020, the FDA approved pembrolizumab for the treatment of adult and pediatric patients with a TMB value of greater than 10 mutations per megabase as determined by the FoundationOne CDx™ assay (FDA, 2020a).

Microsatellites are short, repetitive segments of DNA that are highly prone to mutation. The presence of alternate sized repetitive DNA sequences that are not present in the corresponding germline DNA is defined as microsatellite instability (MSI) (Nojadeh et al., 2018). When a tumor has a high degree of microsatellite instability (MSI-H), it is more immunogenic and thus may respond to drugs that activate the immune system, such as ICIs (Zhao, et al., 2019).

The National Cancer Institute (NCI) recommends two mononucleotide repeats (BAT-25 and BAT-26) and three dinucleotide repeats (D5S346, D2S123, and D17S250) as the standard sites/markers in panels for MSI testing (Boland et al., 1998). Per the NCI guidelines, when a tumor shows instability in two or more of the five markers, it is defined as MSI-H. However, this definition may vary by source and markers used (NCI, n.d.). When only a single marker shows instability, tumors may be classified as low-frequency MSI (MSI-L) and tumors lacking any instability are often classified as microsatellite stable (MSS) (Boland et al., 1998).

Microsatellite instability has been identified in many cancer types, with the highest prevalence in uterine corpus endometrial carcinoma, colon adenocarcinoma, stomach adenocarcinoma, and rectal adenocarcinoma. In studies by Bonneville et al. (2017) and Zhao et al. (2019), MSI was undetectable (or at the borderline of detection) in 12 out of the 39 cancer types (Bonneville et al., 2017; Zhao et al., 2019).

Proprietary Testing

Caris MI Tumor Seek

Caris Life Sciences has developed MI Tumor Seek™, an NGS-based tumor profiling service that combines WES analysis of DNA for mutations, copy number alterations, insertions/deletions, and genomic signatures (MSI, TMB, and LOH) with Whole Transcriptome Sequencing (WTS) analysis for RNA fusions and variant transcripts. WES covers about 22,000 genes with 200-500x depth of coverage, while WTS covers about 22,000 genes with 60 million reads/sample. DNA and RNA are isolated from formalinfixed paraffin embedded (FFPE) tumor tissue “with a minimum of 20% malignant origin for DNA and 10% malignant origin for RNA. Needle biopsy is also acceptable (4-6 cores)” (Caris Life Sciences, 2020, 2024).

Foundation One CDx™ 

FoundationOne CDx™ is a single-site assay that is “a next generation sequencing based in vitro diagnostic device for detection of substitutions, insertion and deletion alterations (indels) and copy number alterations (CNAs) in 324 genes and select gene rearrangements, as well as genomic signatures including microsatellite instability (MSI) and tumor mutational burden (TMB) using DNA isolated from formalinfixed paraffin embedded (FFPE) tumor tissue specimens.” It is intended for use by health care professionals as a companion diagnostic to identify patients with solid malignant neoplasms who may benefit from treatment with targeted ICIs (FDA, 2017c).

Foundation One Liquid CDx

The FoundationOne Liquid CDx assay uses circulating cell-free DNA (cfDNA) that is isolated from plasma from patients with solid malignant neoplasms. The assay uses a “single DNA extraction method to obtain cfDNA from plasma from whole blood. Extracted cfDNA undergoes whole-genome shotgun library construction and hybridization-based capture of 324 cancer-related genes. All coding exons of 309 genes are targeted; select intronic or non-coding regions are targeted in BRCA1 and BRCA2 … Hybridcapture selected libraries are sequenced with deep coverage using the NovaSeq® 6000 platform. Sequence data are processed using a custom analysis pipeline designed to detect genomic alterations, including base substitutions and indels in 311 genes, and copy number variants and genomic rearrangements in BRCA1 and BRCA2. A subset of targeted regions in 75 genes is baited for increased sensitivity” (FDA, 2020b).

Guardant360 TissueNext

Guardant Health offers Guardant360® CDx, a liquid biopsy that provides comprehensive genomic results from a simple blood draw. This test “uses targeted high throughput hybridization-based capture technology for detection of single nucleotide variants (SNVs), insertions and deletions (indels) in 55 genes, copy number amplifications (CNAs) in two (2) genes, and fusions in four (4) genes. Guardant360® CDx utilizes circulating cell-free DNA (cfDNA) from plasma of peripheral whole blood” and is an FDA approved liquid biopsy that can be used for tumor mutation profiling across all solid cancers and as a companion diagnostic to identify NSCLC patients who may benefit from Tagrisso® (ealthinib), RYBREVANT™ (amivantamab-vmjw), or LUMAKRAS™ (sotorasib). To compliment Guardant360 CDx, Guardant Health also offers Guardant360 TissueNext. Guardant360 TissueNext assesses TMB either from an FFPE tumor specimen or from a blood sample initially processed with Guardant360 CDx. If an actionable biomarker is found in a Guardant360 CDx sample, it is then processed with the Guardant360 TissueNext assay (FDA, 2021; Guardant, 2024a, 2024b).

MSK-IMPACT

The MSK-IMPACT assay is a single-site assay performed at Memorial Sloan Kettering Cancer Center and is designed to compare two samples from a single patient—a normal specimen and a specimen from a solid malignant neoplasm—allowing the assay to detect tumor gene alterations in a broad multi gene panel (FDA, 2017a). MSK-IMPACT analyzes 505 genes chosen for their critical role in the development and behavior of tumors. To “take full advantage of MSK-IMPACT, MSK doctors and researchers developed a knowledge base called OncoKB. This system includes information about the clinical and biological effects of more than 4,000 genomic changes. That information is based on public databases, scientific literature, and clinical guidelines” (MSKCC, 2024). The assay is intended to provide information on both somatic mutations and microsatellite instability and is not conclusive or prescriptive for labeled use of any specific therapeutic product (FDA, 2017a).

QIAseq TMB

Qiagen offers QIAseq Tumor Mutational Burden Panels. These new panels collectively create a comprehensive profile of TMB and MSI status by “achieving high analytical sensitivity, with lower false and negative rates, while still maintaining >95% correlation with whole exome datasets.” The QIAseq TMB Panel covers 486 genes and can be boosted to add 27 MSI markers, with sample preparation coming from FFPE of solid tumors, patient’s plasma/serum, fresh or frozen tissue, or from cell lines in a research setting. This assay has been shown to correlate well with WES and established TMB panels. It uses enzymatic fragmentation for easy workflow, employs robust analysis modules, and incorporates unique molecular indices to correct for polymerase chain reaction (PCR) and sequencing errors (Qiagen, 2024).

TruSight Oncology 500

Illumina offers the TruSight Oncology 500 (formerly Illumina TSO500), a NGS assay that enables inhouse CGP of tumor samples, supports the identification of relevant DNA and RNA variants implicated in various solid tumor types, and accurately measure current immune-oncology biomarkers: MSI and TMB. This assay assesses both DNA (523 genes) and RNA (55 genes) for assessment of all variant types in addition to MSI and TMB. The TruSight Oncology 500 assay is intended for multiple solid tumor types and detects CNVs, gene fusions, insertions-deletions, single nucleotide variants, and transcript variants. It provides stealth oncology biomarker coverage, including biomarkers of TMB and MSI and is inclusive of human leukocyte antigen (HLA) regions, DNA polymerase epsilon catalytic subunit isoform b (POLE1) and DNA polymerase delta catalytic subunit (POLD1) (Illumina, 2024).

Analytical Validity

Wu et al. (2019) noted that targeted NGS panels often use correlation (R2 ) between panel- and WES- based TMB to validate TMB estimation, which can be distorted by cases where a tumor within a certain cancer type has a relatively ultra-high TMB. Thus, these authors proposed using accuracy (the proportion of cases correctly identified as TMB-high or TMB-low using panel based TMB) as a more robust indicator of panel performance, suggesting that accuracy and cancer type individualization are key factors to consider when designing panels for TMB estimation. For their study, the authors examined the TMB estimation from five available NGS panels for TMB determination (F1CDx, MSK-IMPACT, Illumina TSO500 [now TruSight Oncology 500], Oncomine TML, QIAseq TMB) to compare correlation to accuracy. F1CDx seemed to accurately assess TMB (R2 ≥0.75) in at least 24 out of 33 cancer types. However, when the cutoff point for high TMB was defined as the top 20% in each cancer type, the accuracy within these 24 cancer types ranged largely (56-99%), with only 7 cancer types having satisfactory accuracy. These results indicate that F1CDx-based TMB estimation is only reliable in certain cancer types (cervical squamous-cell carcinoma and endocervical adenocarcinoma, colon adenocarcinoma, head and neck squamous cell carcinoma, lung adenocarcinoma, skin cutaneous melanoma, stomach adenocarcinoma, and uterine corpus endometrial carcinoma). In the other 17 cancer types, correlation overestimated TMB status misclassifying a considerable number of TMB-low patients as TMB-high. This overestimation by correlation occurs because correlation is vulnerable to distortion by the common presence of cases with relatively ultra-high TMB within each cancer type. When removing cases with these relatively ultra-high TMB (top five percent within a particular cancer type), a dramatic decline in correlation between F1CDx- and WESbased TMB was observed in half of the 24 cancer types. In contrast, removing the top five percent of cases in each cancer type did not, in general, affect the accuracy within these 24 cancer types. Accuracy was also superior to correlation in the other four NGS panels examined. The authors conclude that the five available NGS panels can assess TMB accurately only in the 7 cancer types aforementioned and note that for relatively ultra-high TMB, correlation is an unreliable evaluation of panel based TMB estimation performance in most cancer types. Instead, they find that accuracy is a superior index in this situation. They conclude that individualized panels by cancer type might be a better strategy to guarantee the most robust TMB estimation and provide better power in detecting the predictive function of TMB in these cancer types (Wu et al., 2019).

J. Willis et al. (2019) aimed to evaluate the validity of cell-free (cf) DNA to detect microsatellite instability (MSI). The authors validated the Guardant360 MSI detection with 1145 cf-DNA samples and further investigated the landscape of cf-DNA based MSI across 28459 plasma samples. Circulating tumor (ct, used as a proxy for cf-DNA) MSI assessment was compared to current standard of care tissue testing (a combination of immunohistochemistry, PCR, NGS). 949 patients were considered unique and evaluable. The ctDNA method detected MSI-high values (82% accuracy) in 71 of 82 patients. 863 of 867 patients reported as MSI-stable (99.5% accuracy) for a total of 98.4% accuracy amongst all MSI-reported groups. The positive predictive value was found to be 95%. Finally, the authors examined the clinical outcomes of 16 patients with MSI-H gastric cancer that were treated with immunotherapy. The authors found that 10 of 16 patients achieved “complete or partial remission with sustained clinical benefit” after being treated with immunotherapy (J. Willis et al., 2019).

Georgiadis et al. (2019) aimed to validate a noninvasive approach for detection of MSI and tumor mutational burden (TMB). The authors used “a hybrid-capture–based 98-kb pan-cancer gene panel, including targeted microsatellite regions” as well as a “novel peak-finding algorithm…established to identify rare MSI frameshift alleles in cell-free DNA”. The authors evaluated 163 healthy patients with this assay, identifying a specificity of 99.4% (one false positive, which was considered a lower bound by the authors). The authors also evaluated 23 MSI-H cases (from patients with metastatic cancers) and 6 MSI-stable cases with the algorithm. Using cf-DNA, the algorithm detected MSI in 18 of the 23 MSI-H samples (78.3%) and correctly identified all 6 of the MSI-stable cases. Further, this approach was also applied to TMB (using a cutoff of 51 mutations/Mbp), of which the algorithm correctly detected 10 of the 15 samples identified as TMB-High (67%) (Georgiadis et al., 2019).

Woodhouse et al. (2020) evaluated the analytical performance of FoundationOne Liquid CDx assay to detect genomic alterations in cancer patients. The assay was evaluated across more than 30 different cancer types in over 300 genes and greater than 30,000 gene variants. "Results demonstrated a 95% limit of detection of 0.40% variant allele fraction for select substitutions and insertions/deletions, 0.37% variant allele fraction for select rearrangements, 21.7% tumor fraction (Marabelle, Le, et al.) for copy number amplifications, and 30.4% TF for copy number losses. The false positive variant rate was 0.013% (approximately one in 8,000). Reproducibility of variant calling was 99.59% (Woodhouse et al., 2020)." In comparison to in situ hybridization and immunohistochemistry, FoundationOne had an overall 96.3% positive percent agreement and >99.9% negative percent agreement. "These study results demonstrate that FoundationOne Liquid CDx accurately and reproducibly detects the major types of genomic alterations in addition to complex biomarkers such as microsatellite instability, blood tumor mutational burden, and tumor fraction" (Woodhouse et al., 2020).

Using data from the phase III MYSTIC trial, Si et al. (2021) validated the use of blood TMB (bTMB) using circulating cell-free tumor DNA (ctDNA) through comparisons with TMB measurements from tumor tissue (tTMB). Patients with metastatic NSCLC were treated either with first line durvalumab (anti– PD-L1 antibody) ± tremelimumab (anticytotoxic T-lymphocyte-associated antigen-4 antibody) or chemotherapy, then bTMB and tTMB were evaluated using the GuardantOMNI and the FoundationOne CDx assay, respectively. To identify the optimal bTMB cutoff, the authors used a Cox proportional hazards model and minimal p value cross-validation. In the majority of patients in the MYSTIC study, somatic mutations were detected in ctDNA extracted from plasma samples, which allowed for the subsequent calculation of bTMB. The authors found that the success rate for obtaining valid TMB scores was higher for bTMB (81%) than for tTMB (63%) and that with minimal p value cross-validation analysis, they suggest a bTMB ≥20 mutations per megabase (mut/Mb) as the optimal cutoff for clinical benefit with durvalumab + tremelimumab. Overall, their study demonstrated that use of bTMB from plasma samples was feasible and provided accurate and reproducible TMB detection using the GuardantOMNI ctDNA platform. They also state that using “the new bTMB algorithm and an optimal bTMB cutoff of ≥20 mut/Mb, high bTMB was predictive of clinical benefit with durvalumab + tremelimumab versus chemotherapy” (Si et al., 2021).

Clinical Utility and Validity

Overman et al. (2017) examined MSI-H/dMMR colorectal cancer patient response to nivolumab, a PD-1 ICI. In this phase 2 trial, the authors assessed adult patients with histologically confirmed recurrent or mCRC locally assessed as dMMR/MSI-H who had progressed either while still on or after, or who had been intolerant of, at least one prior line of treatment, including fluoropyrimidine and oxaliplatin or irinotecan. The primary endpoint of this study was investigator-assessed ORR. This study ran from March 12, 2014 to March 16, 2016 and included 74 patients, most of whom (54.1%), had received more than three prior therapies. Twelve months following initiation of nivolumab treatment, 31.1% (23) of the patients had an investigator-assessed ORR, which 68.9% of the patients had disease control for more than 12 weeks. The median duration of response was not reached, as all responders were alive and 8 of them had responses greater than 12 months. Twenty-three of the patients died during the study, though none of the deaths were considered treatment related. The authors concluded that nivolumab provided durable responses and disease control in dMMR/MSI-H patients that had previously received other forms of treatment for their colorectal cancer and they also note that nivolumab is a relevant treatment option for these patients (Overman et al., 2017).

Hellmann, Ciuleanu, et al. (2018) conducted an open-label, multipart, phase three trial to examine progression-free survival in non-small-cell lung cancer patients with a high TMB (TMB-H) that were being treated with nivolumab plus ipilimumab or with chemotherapy. TMB was determined by the FoundationOne CDx™ assay. Here, they found that progression-free survival in TMB-H patients was significantly longer with nivolumab plus ipilimumab (42.6%) compared with chemotherapy (13.2%). This discovery was found to be accurate irrespective of PD-L1 expression level and validated that a treatment of nivolumab and ipilimumab for non-small cell lung cancer (NSCLC) elicited a favorable outcome for select patients, as well as determining that tumor mutational burden was an important biomarker for patient selection when it comes to drug therapy (Hellmann, Ciuleanu, et al., 2018).

To examine the association between TMB and clinical response to ICIs in select cancer types, Samstein et al. (2019) analyzed the clinical and genomic data of 1,662 advanced cancer patients treated with ICI compared to 5,371 patients not treated with ICI. Tumors from these patients underwent targeted NGS with MSK-IMPACT. The authors found that among all patients, higher somatic TMB (highest 20% in each cancer type, consistent with many MSI-high colorectal tumors receiving ICI treatment) was associated with better overall survival. However, the TMB cutoffs associated with improved survival varied markedly between the different cancer types studied, suggesting that while TMB is associated with improved survival in patients receiving ICI across a variety of cancer types, there may not be a single, universal definition of high TMB (Samstein et al., 2019).

Ready et al. (2019) conducted an open-label phase II trial to evaluate the efficacy and safety of nivolumab plus low-dose ipilimumab as first-line treatment of advanced/metastatic NSCLC and assessed the treatment efficacy associated with PD-L1 expression and TMB. For this trial, “the primary end point was objective response rate (ORR) in patients with 1% or more and less than 1% tumor PD-L1 expression. Efficacy on the basis of TMB (FoundationOne CDx™ assay) was a secondary end point.” They found that regardless of PD-L1 expression, ORRs were higher in TMB-H patients versus patients with tumors below the threshold for TMB-H status and progression-free survival was longer in patients with TMB-H. They concluded that “nivolumab plus low-dose ipilimumab was effective and tolerable as a first-line treatment of advanced/metastatic NSCLC” and a high TMB status was associated with both improved response and prolonged progression-free survival irrespective of PD-L1 status. This identifies TMB-H as a potentially relevant cutoff in the assessment of TMB as a biomarker for first line nivolumab plus ipilimumab (Ready et al., 2019).

A comprehensive study by Singal et al. (2019) examined the electronic health records (EHR) of 4064 individuals with non-small cell lung cancer (NSCLC) from 275 different oncology practices to explore the associations between tumor genomics and patient characteristics with clinical outcome. The authors note that 21.4% of these individuals had a mutation in EGFR, ALK, or ROS1 and that patients with driver mutations who received targeted therapies had significantly improved overall survival times than individuals who did not receive targeted therapies (median of 18.6 versus 11.4 months, respectively). Moreover, a TMB of 20 or higher was associated with improved overall survival for patients on PD-L1- targeted therapy compared to patients with a TMB less than 20. TMBs measure the quality of a mutation in a tumor, suggesting whether a patient will benefit from immunology-based cancer therapies or not. The authors concluded that similar associations from previous research were replicated “between clinical and genomic characteristics, between driver mutations and response to targeted therapy, and between TMB and response to immunotherapy” (Singal et al., 2019).

Yarchoan et al. (2019) examined the relationship between the expression of PD-L1, a widely used biomarker for selecting patients for ICI, and TMB across the entire spectrum of ICI-responsive human cancers. They sought to use these predictive biomarkers to provide a broad definition to the immunologic subtypes of cancers and identify opportunities for the development of therapeutics. Their clinical cohort contained 9,887 unique clinical samples with paired CGP and PD-L1 expression obtained during standard clinical care. They found that 15.2% of samples had a positive expression of PD-L1 and that the highest occurrence of PD-L1 positivity occurred in thymic cancer and diffuse large B cell lymphoma, while there were no adenoid cystic or appendiceal samples with positive PD-L1. A total 3.6% of sample specimens were identified as having a high PD-L1 positivity (defined as greater than 50% of tumor cells staining positive). In addition, the median tumor mutational burden for all specimens was 3.48 mutations per megabase. One group of sampled tumors (16.4% of the cancers) had a TMB of greater than 10 mutations per megabase and 7.3% of tumors analyzed had a TMB of more than 20 mutations per megabase, indicating that the overall TMB profiles were comparable to prior cancer studies of a similar nature (including whole-exome studies) that estimated TMB biomarkers in solid tumors. This study found a relative independence of PD-L1 expression and TMB, and that the biomarkers were important indicators of patient response to ICI therapy. The study also defined which biomarkers may help identify future therapies through their ability to segment and profile different tumor types. In conclusion, they found that PD-L1 and TMB status may help determine whether the use of ICI therapy is indicated as well as pinpoint tumor types that would demonstrate a beneficial response to immune checkpoint inhibitor therapy (Yarchoan et al., 2019).

Alborelli et al. (2020) investigated the predictive power of TMB for patients treated with ICIs. Seventysix NSCLC patients treated with ICIs were included and TMB was evaluated with the Oncomine™ Tumor Mutational Load (TML) sequencing assay. Patients were separated into cohorts of “durable clinical benefit” (DCB) or “no durable benefit” (NDB). TMB was found to be higher in the DCB cohort (median TMB of 8.5 mutations / Mb compared to 6 mutations / Mb in NDB). Sixty-four percent of patients in the highest one third of TMB were responders, compared to 33% and 29% in the middle and lowest thirds, respectively. TMB-H patients were also found to have higher progression free survival and overall survival. Overall, the authors concluded that the TML panel was an effective tool to stratify patients for ICI treatment and suggested that “a combination of biomarkers might maximize the predictive precision for patient stratification”. Further, the authors remarked that their data “supports TMB evaluation through targeted NGS in NSCLC patient samples as a tool to predict response to ICI therapy” (Alborelli et al., 2020).

Marabelle, Fakih, et al. (2020) prospectively explored the association of a high tumor mutational burden with outcomes in ten tumor-type-specific cohorts from the phase 2 KEYNOTE-158 study. The study assessed use of the anti-PD-1 monoclonal antibody pembrolizumab in patients with select, previously treated, advanced solid tumors. Patients for this study were from 81 academic facilities across 21 countries in Africa, the Americas, Asia, and Europe, were over 18 years of age, and had an established tumor of anal, biliary, cervical, endometrial, mesothelioma, neuroendocrine, salivary, small-cell lung, thyroid, or vulvar origin. In addition, patients showed continued tumor progression or an inability to tolerate standard therapy and continued measurable disease (based on the Response Evaluation Criteria in Solid Tumors). The treatment course for this study was pembrolizumab 200 mg intravenously every three weeks for up to thirty-five cycles, with tissue TMB assessed in formalin-fixed, paraffin-embedded tumor samples using the FoundationOne CDx assay. The authors report that 13% of the participants showed a high number of tumor mutations (≥ 10 mutations per megabase) while 87% of the participants had a TMB status lower than this threshold. From this study, they found that a high TMB status characterized a small group of patients who might have a beneficial response to pembrolizumab therapy; those with previously treated recurrent or advanced solid tumors. As such, tissue TMB may be both a “novel and useful predictive biomarker” for therapy indications (Marabelle, Fakih, et al., 2020).

Data from the KEYNOTE-158 study was also analyzed by to examine the efficacy of pembrolizumab as an ICI treatment for MSI-H.dMMR cancer. The authors examined the efficacy of pembrolizumab in patients with previously treated, advanced noncolorectal MSI-H/DMMR cancer from the phase II KEYNOTE-158 study. In the 233 enrolled patients, 27 tumor types were represented (endometrial, gastric, cholangiocarcinoma, and pancreatic cancers were the most common). ORR was 34.3%, median PFS was 4.1 months, and median OS was 23.5 months. Adverse, treatment-related events occurred in 64.8% of patients (151) and 14.6% (34) patients had grade 3 to 5 treatment-related fatal adverse events. They report that pembrolizumab is clinically beneficial in patients with previously treated, unresectable or metastatic MSI-H/dMMR noncolorectal cancer and that toxicity of pembrolizumab monotherapy in these patients was consistent with previous data (Marabelle, Le, et al., 2020).

A study by Stahler et al. (2020) explored the ORR, PFS, and OS in the FIRE-3 clinical trial on metastatic colorectal cancer patients who were identified by the FoundationOne CDx panel to have single-nucleotide variants (SNVs), copy number alterations, TMB-H, or MSI-H tumors. Of the 752 patients in the FIRE-3 trial, 373 provided material for this specific analysis. They found that MSI-H (30%) and TMB-H (17.3%) tumors were enriched by FBXW7 mutations a frequent SNV found in the analysis. The authors found that “RAS, BRAF V600E and SMAD4 mutations were identified as poor prognostic biomarkers in patients of the FIRE-3 trial, whereas improved outcome was observed for BRAF non-V600E mutation. SMAD4 mutation might provide predictive relevance for cetuximab efficacy. MSI-H tumours showed numerically lower ORR, OS and PFS” (Stahler et al., 2020).

While ICIs are highly effective in patients with MSI/dMMR metastatic colorectal cancer (mCRC), Cohen et al. (2020) predicted that the Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 criteria was underestimating the response to ICIs due to the pseudoprogression phenomenon. Using data from the GERCOR NIPICOL phase II study, the authors aimed to evaluate the frequency of pseudoprogressions in patients with MSI/dMMR mCRC that were treated with nivolumab and ipilimumab. Nivolumab was given to MSI/dMMR mCRC patients previously treated with fluoropyrimidines, oxaliplatin, and irinotecan with/without targeted therapies. The primary endpoint of the study was disease control rate at 12 weeks according to RECIST 1.1 and iRECIST by central review. Fifty-seven patients were included in the study, with 48% having received ≥three prior lines of chemotherapy, 18% with BRAF mutations, and 56% with Lynch syndrome-related cancer. Of these 57 patients, 7 discontinued due to adverse events and one died due to a treatment-related adverse events (TRAE). The authors found that at 12 weeks, the disease control rate was 86% with RECIST 1.1 and 87.7% with iRECIST and 2 pseudoprogressions were observed (one at week 6, one at week 36). These two pseudoprogressions represented 18% of patients with disease progression per RECIST 1.1 criteria. The 12-month PFS rate was 72.9% with RECIST 1.1 and 76.5% with iRECIST and the 12-month OS rate was 84%. The ORR was 59.7% with both criteria. The authors concluded that pseudoprogression is rare in patients with MSI/dMMR mCRC who are treated with nivolumab and ipilimumab and that this combined ICI therapy confirms impressive disease control rate and survival outcomes in these MSI/dMMR patients (Cohen et al., 2020).

Andre et al. (2020) examined the efficacy of first line pembrolizumab therapy as compared to first-line chemotherapy therapy for dMMR/MSI-H advanced or metastatic colorectal cancer. In this phase three, open-label clinical trial, 307 patients with metastatic MSI/dMMR colorectal cancer who had not received prior cancer treatment were randomly assigned to pembrolizumab or chemotherapy as a primary treatment and PFS and OS were assessed. Patients that received chemotherapy were able to cross over to pembrolizumab therapy if their disease progressed. The authors found that pembrolizumab treatment in MSI-H/dMMR colorectal cancer patients was superior to chemotherapy with respect to PFS. At the time of cutoff, 56 members of the pembrolizumab group and 69 chemotherapy group patients had died. At the time of the report, OS data were still evolving (they report that 66% of required events had occurred). An overall response (complete or partial) was observed in 43.8% of pembrolizumab patients as compared to only 33.1% of chemotherapy patients. Among those patients with an overall response, 83% of the pembrolizumab group, as compared to just 35% of the chemotherapy group, had an ongoing response at 24 months. TRAEs occurred in just 22% of pembrolizumab patients as compared to 66% of chemotherapy patients (including one death). The authors conclude that pembrolizumab is superior to chemotherapy as a first-line therapy in MSI-H/dMMR metastatic colorectal cancer, with fewer TRAE (Andre et al., 2020).

Wei et al. (2022) evaluated Illumina’s TruSight Oncology 500 (TSO500) Assay for its ability to assess tumor mutational burden and the clinical utility of predicting response to pembrolizumab. Using samples collected from approximately 300 patients with “at least twelve different types of advanced solid tumors enrolled in eight clinical trials of pembrolizumab monotherapy,” and using FoundationOneCDx and whole exome sequencing (WES) as reference methods for TMB, the researchers found that the TSO500 was able to properly classify the status of high tumor mutational burden, a reproducible finding with subsequent use of the assay. The researchers also found that “similar to the validated and FDA-approved TMB cutpoint of 10 mut/Mb assessed by FoundationOneCDx, the TSO500 TMB cut-point of 10 mut/Mb is predictive of response to pembrolizumab monotherapy.” Ensuring the comparability of this assay to current technologies in the necessary settings may increase access to immunotherapy for patients with solid tumors and TMB-H (Wei et al., 2022).

Aggarwal et al. (2023) completed a clinical validation study of a TMB biomarker in the aim of assessing the association between TMB and clinical outcomes. The study included 674 patients with solid tumors from eight different cancer types. Patients were collected from 300 cancer sites between 2018 and 2022. All patients were sequenced with Tempus xT and treated with ICIs. The authors studied the association between the TMB binary category, high or low, and overall survival. Overall, “high TMB (TMB-H) cancers (206 patients [30.6%]) were significantly associated with longer OS than low TMB (TMB-L) cancers.” Additionally, “an overall survival benefit was seen regardless of the type of ICI used.” The authors concluded that “TMB-H cancers were significantly associated with improved clinical outcomes compared with TMB-L cancers” (Aggarwal et al., 2023).

Guidelines and Recommendations

National Comprehensive Cancer Network (NCCN)

The NCCN includes TMB and MSI testing in several of their guidelines for different types of cancers. Across these guidelines, TMB-high (TMB-H) is defined as tumors with a TMB value of ≥10 mutations/megabase.

The NCCN has clinical practice guidelines for non-small cell lung cancer (NSCLC). In 2020, the NCCN deleted TMB as an emerging immune biomarker based on clinical trial data and other issues, though was previously considered potentially useful for selecting patients to be treated with nivolumab with or without ipilimumab. However, recent data showed that overall survival was improved with nivolumab/ipilimumab regardless of TMB status. They conclude that PD-L1 expression level is a more useful immune biomarker than TMB for deciding how to prescribe immunotherapy in NSCLC patients in some instances. For example, “Testing for PD-L1 expression levels is recommended before prescribing pembrolizumab monotherapy,” but “is not required for prescribing subsequent therapy with atezolizumab or nivolumab”, though it may still provide useful information. Moreover, the NCCN Guidelines “do not recommend measurement of TMB levels before deciding whether to use nivolumab plus ipilmumab regimens or to use other ICIs, such as pembrolizumab” (NCCN, 2024i).

The NCCN clinical practice guidelines for cervical cancer advise that for Stage IVB or distant metastases, “comprehensive molecular profiling as determined by an FDA-approved assay or a validated test performed in a CLIA-certified laboratory” should be considered. They repeat this under their section for principles of pathology for squamous cell carcinoma, adenocarcinoma, and adenosquamous carcinoma, and add on that they recommend “PD-L1 testing for patients with recurrent, progressive, or metastatic disease.” The NCCN also recommend “mismatch repair (MMR)/microsatellite instability (MSI) testing for patients with recurrent, progressive, or metastatic cervical carcinoma” and/or “NRTK gene fusion testing for patients with cervical sarcoma.” For systemic cervical cancer therapy, the NCCN reports that pembrolizumab is a preferred regimen as a second-line or subsequent therapy for TMB-H, PD-L1 positive, or MSI-H/dMMR tumors that are recurrent or metastatic squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma (NCCN, 2024d).

The NCCN clinical practice guidelines for breast cancer note that pembrolizumab is indicated for the treatment of patients with either unresectable or metastatic, MSI-high or mismatch repair deficient (dMMR) solid tumors, or TMB-H breast tumors that have progressed following prior treatment and have no satisfactory alternative treatment options. TMB-H status is detected via NGS and MSI is detected via immunohistochemistry, NGS, or PCR. The NCCN finds using TMB and MSI to guide pembrolizumab treatment is associated with category 2A evidence (based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate). Dostarlimab-gxly is indicated for adult patients with MSI-H (category 2A) solid tumors or for TMB-H tumors that have progressed following prior treatment and with no satisfactory alternative treatment options (NCCN, 2024c).

In their clinical practice guidelines for uterine neoplasms, the NCCN recommends that one should “Consider tumor mutational burden (TMB) testing through an FDA-approved assay, or a validated test performed in a CLIA-certified laboratory” as one of their principles of molecular analysis. The NCCN also notes the use of pembrolizumab for TMB-H or MSI-H/deficient DNA mismatch repair (dMMR) tumors as “useful in certain circumstances” as a biomarker-directed systemic therapy for uterine sarcoma and endometrial carcinoma. There, the NCCN also “recommends TMB-H testing if not previously done. Pembrolizumab is indicated for patients with unresectable or metastatic tumors with RMB-H [>10 mutations/megabase (mut/Mb)], as determined by an FDA-approved assay, or a validated test performed in a CLIA-certified laboratory, whose disease has progressed following prior treatment and who have no satisfactory alternative treatment options.” In the molecular analysis of sarcoma, it is recommended that “Comprehensive genomic profiling in setting of metastatic disease as determined by an FDA-approved assay, or a validated test performed in a CLIA-certified laboratory, is informative for predicting rare pantumor targeted therapy opportunities and should include at least NTRK, MSI, and TMB” (NCCN, 2024p).

The NCCN guidelines for colon cancer state that “determination of tumor gene status for KRAS/NRAS and BRAF mutations, as well as HER2 amplifications and MSI/MMR status (if not previously done), are recommended for patients with mCRC.” They note that “Testing may be carried out for individual genes or as part of an NGS panel, although no specific methodology is recommended.” However, NGS panels do have an advantage in that they are “able to pick up rare and actionable genetic alterations, such as neurotrophic tyrosine receptor kinase (NTRK) and rearranged during transfection (RET) fusions and may be carried out using either a tissue or blood-based (eg, liquid) biopsy” (NCCN, 2024e). The NCCN advises that “As with other conditions in which stage IV disease is diagnosed, a tumor analysis (metastases or original primary) for KRAS/NRAS and BRAF mutations and HER2 amplifications, as well as MSI/MMR testing if not previously done, should be performed to define whether targeted therapies can be considered among the potential options”. In pedunculated or sessile colon polyps (adenoma) with invasive cancer, MSI/MMR testing can be part of the workup. Colon cancer appropriate for resection (non-metastatic) will “require a complete staging workup, including biopsy, pathologic tissue review, total colonoscopy, CBC, chemistry profile, CEA determination, and baseline CT scans of the chest, abdomen, and pelvis”, while “Testing for MMR/MSI should be done at diagnosis to help with detection of Lynch syndrome and to inform treatment decision-making.” When evaluating the recurrence of documented metachronous metastases by CT, MRI, and/or biopsy, dMMR and MSI-H can be considered endpoints for workups (NCCN, 2024e).

In the NCCN clinical practice guidelines for esophageal and esophagogastric junction (EGJ) cancers, the panel outlines their principles of pathologic review and biomarker testing. There, they assert that “Immunohistochemistry/in situ hybridization/targeted PCR should be considered first for the identification of biomarkers, followed by NGS testing. If limited tissue is available, or the patient is unable to undergo a traditional biopsy, sequential testing of single biomarkers/limited molecular diagnostic panels will exhaust the sample. In these scenarios, or at the discretion of the treating physician, comprehensive genomic profiling via a validated NGS assay performed in a CLIA-approved laboratory should be considered”. Their list of targeted biomarkers include HER2 overexpression/amplification, PD-L1 expression by immunohistochemistry, microsatellite instability, tumor mutational burden, NTRK gene fusion, RET gene fusion, and BRAF V600E mutation (NCCN, 2024f). Furthermore, during workup for esophageal and esophagogastric junction cancers, “Universal testing for microsatellite instability (MSI) by PCR/next-generation sequencing (NGS) or MMR by IHC is recommended in all newly diagnosed patients”. They also recommend “Programmed death ligand 1 (PD-L1) testing if advanced/ metastatic disease is documented/suspected”, “HER2 testing if metastatic adenocarcinoma is documented/ suspected”, and that “NGS should be considered”. Universal testing for MSI by PCR/NGS or MMR by IHC is recommended in all newly diagnosed patients for biopsy specimen types, endoscopic resection, and esophagogastrectomy, without prior chemoradiation (if MSI and MMR has not been previously performed) (NCCN, 2024f). In their assessment of overexpression or amplification of HER2 in esophageal and EGJ Cancers, the NCCN recommends that “The use of IHC/ISH should be considered first, followed by NGS testing as appropriate. Repeat biomarker testing may be considered at clinical or radiologic progression for patients with advanced/metastatic esophageal/EGJ adenocarcinoma (NCCN, 2024f).

The NCCN clinical practice guidelines for occult primary (cancer of unknown primary [CUP]) note that “MSI/mismatch repair (MMR) testing is indicated for patients with CUP; however, it should be noted that the population of patients with MSI-high/MMR-deficient (MSI-H/dMMR) occult primary tumors is generally low.” Moreover, they remark that determination of TMB by a validated and/or FDA approved assay is a category 2B recommendation (2B: based on lower-level evidence, there is NCCN consensus that the intervention is appropriate) and a biopsy from a suspected metastatic malignancy should undergo TMB testing during the workup of the tumor. The NCCN also asserts that during workup “Gene sequencing to predict tissue of origin is not recommended” and “Molecular profiling of tumor tissue using next-generation sequencing (NGS) (or other technique to identify gene fusions) can be considered after an initial determination of histology has been made”. Again, the NCCN cautions that “The population of patients with microsatellite instability-high/mismatch repair-deficient (MSI-H/dMMR) occult primary tumors is low. Use immunohistochemistry (IHC) for MMR or polymerase chain reaction (PCR) for MSI, which are different assays measuring the same biological effect” (NCCN, 2023e). For occult primary tumors, dostarlimab-gxly is recommended as “useful in certain circumstances” only in dMMR/MSI-H tumors, while pembrolizumab may be extended to tumors with either TMB-H or dMMR/MSI-H (NCCN, 2023e).

In the NCCN clinical practice guidelines for gastric cancer, universal testing, no matter the specimen type, for MSI by PCR/NGS or MMR by IHC is recommended in all newly diagnosed patients in the biopsy stage, endoscopic mucosal resection, gastrectomy without prior chemoradiation (if not previously performed). In the palliative management of unresectable locally advanced, locally recurrent, or metastatic disease, “Perform HER2, programmed death ligand 1 (PD-L1), and microsatellite testing (if not done previously) if metastatic cancer is documented or suspected” and “NGS may be considered via a validated assay”. Tumors should be interpreted as MSI-Stable, MSI-Low (1-29% of markers exhibit instability, one of the five National Cancer Institute (NCI) or mononucleotide markers exhibits instability), or MSI-H (≥30% of the markers exhibit instability, two or more of the five NCI or mononucleotide markers exhibit instability). They also state that “If limited tissue is available, or the patient is unable to undergo a traditional biopsy, sequential testing of single biomarkers/limited molecular diagnostic panels will exhaust the sample. In these scenarios, or at the discretion of the treating physician, comprehensive genomic profiling via a validated NGS assay performed in a CLIA-approved laboratory should be considered.” Their list of targeted biomarkers includes HER2 overexpression/amplification, PD-L1 expression by immunohistochemistry, microsatellite instability, tumor mutational burden, NTRK gene fusion, RET gene fusion, and BRAF V600E mutation (NCCN, 2024g).

The NCCN clinical practice guidelines for hepatocellular carcinoma state that there is “no established role for microsatellite instability (MSI), mismatch repair (MMR), tumor mutational burden (TMB), or PD-L1 testing in [hepatocellular carcinoma] at this time. Immune checkpoint inhibition has shown clinical benefit leading to regulatory approvals in patients with HCC without selection for MSI, MMR, TMB, or PD-L1 status. Regarding therapies, the panel recommends nivolumab + ipilimumab (category 2B) as a first-line systemic therapy that is useful in certain circumstances and a subsequent-line systemic therapy if there is disease progression for individuals with TMB-H tumors. The NCCN also dostarlimab-gxly (category 2B) as a subsequent-line systemic therapy if there is disease progression for individuals with MSI-H/dMMR tumors (NCCN, 2024h).

The NCCN clinical practice guidelines for biliary tract cancers (BTCs) recommend germline testing and/or referral to a genetic councilor for patients with dMMR/MSI-H tumors or a family history suggestive of BRCA1/2 mutation” when presenting during intraoperative staging biopsy following an incidental finding of suspicious mass during surgery. Germline testing and/or referral to genetic counseling for patients with dMMR/MSI-H tumors or a family history suggestive of BRCA1/2 mutations should also be extended to situations where hepatobiliary surgery expertise is unavailable and postoperative workup is deemed unresectable, incidental finding on pathologic review with T1b or greater and/or T1a with positive margins deemed unresectable, unresectable mass on imaging, jaundice, metastatic disease, or isolated intrahepatic mass with unresectable or metastatic disease biopsy. The NCCN recommends that “Comprehensive molecular profiling is recommended for patients with unresectable or metastatic BTC who are candidates for systemic therapy”. They also note that “Evidence remains insufficient for definitive recommendations regarding specific criteria to guide genetic risk assessment in hepatobiliary cancers or for universal germline testing in these tumors” and, as such, “In BTCs, genetic counseling referral and potential germline testing should be considered in patients with any of the following characteristics: young age at diagnosis; a strong personal or family history of cancer; no known risk factors for liver disease; or presence of mutations identified during tumor testing that are suspected to be possible germline alterations. For patients who do harbor a known germline mutation associated with a cancer predisposing syndrome (ie, Lynch syndrome or hereditary breast and ovarian cancer syndrome), there is currently insufficient evidence to support screening for biliary tract malignancies” (NCCN, 2024a). The NCCN asserts that “Testing for MSI or MMR deficiency is recommended in patients with unresectable or metastatic gallbladder cancer, intrahepatic CCA, or extrahepatic CCA” and that “Testing for TMB is recommended for patients with unresectable or metastatic gallbladder cancer, intrahepatic CCA, or extrahepatic CCA based upon clinical benefit observed across advanced solid tumors.” They also outline the three possible tests to evaluate mismatch repair (MMR) protein deficiency or microsatellite status. First, an immunohistochemical staining for the MLH1, MSH2, MSH6, and PMS2 gene products will establish protein retention or loss, followed by NGS to determine if there are inactivating mutations in the MMR genes (MLH1, MSH2, MSH6, and PMS2). Lastly, “microsatellite repeats of tumor DNA are examined by polymerase chain reaction (PCR)” and caution that “Tumor mutational burden (TMB) can be tested with a clinically validated NGS panel but has inherent platform variation” (NCCN, 2024a). In addition, the NCCN provides a helpful graphic to show their recommendations for molecular testing in unresectable or biliary tract cancers, captured below: