Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Tumor mutational burden quantification from targeted gene panels: major advancements and challenges

Tumor mutational burden quantification from targeted gene panels: major advancements and challenges Tumor mutational burden (TMB), the total number of somatic coding mutations in a tumor, is emerging as a promising biomarker for immunotherapy response in cancer patients. TMB can be quantitated by a number of NGS-based sequencing technologies. Whole Exome Sequencing (WES) allows comprehensive measurement of TMB and is considered the gold standard. However, to date WES remains confined to research settings, due to high cost of the large genomic space sequenced. In the clinical setting, instead, targeted enrichment panels (gene panels) of various genomic sizes are emerging as the routine technology for TMB assessment. This stimulated the development of various methods for panel-based TMB quantification, and prompted the multiplication of studies assessing whether TMB can be confidently estimated from the smaller genomic space sampled by gene panels. In this review, we inventory the collection of available gene panels tested for this purpose, illustrating their technical specifications and describing their accuracy and clinical value in TMB assessment. Moreover, we highlight how various experimental, platform-related or methodological variables, as well as bioinformatic pipelines, influence panel-based TMB quantification. The lack of harmonization in panel-based TMB quantification, of adequate methods to convert TMB estimates across different panels and of robust predictive cutoffs, currently represents one of the main limitations to adopt TMB as a biomarker in clinical practice. This overview on the heterogeneous landscape of panel-based TMB quantification aims at providing a context to discuss common standards and illustrates the strong need of further validation and consolidation studies for the clinical interpretation of panel-based TMB values. Keywords: Tumor mutational burden, TMB, Gene panels, Targeted enrichment sequencing, Immunotherapy Tumor mutational burden: an emerging strategies to adequately select those patients most likely biomarker for cancer immunotherapy to show a favorable response is recognized as an urgent Immunotherapy with immune checkpoint inhibitors medical need. A few potential biomarkers have been targeting cytotoxic T lymphocyte associated 4 (CTLA-4) identified up to now, such as PD-L1 gene expression, or programmed cell death 1 (PD-1) or its ligand (PD-L1) microsatellite instability (MSI), mismatch repair defi- can provide important clinical benefit to patients ciency (dMMR), POLE or JAK1/2 mutations, immune affected by multiple cancers, most notably lung cancer cell infiltration, IFNγ expression, tumor mutational [1, 2], melanoma [3], renal cancer [4] and urothelial burden (TMB) or neoantigen burden [6, 7]. carcinoma [5]. However, only a fraction of patients cur- TMB is a measure of the total amount of somatic cod- rently treated by immune checkpoint inhibitors derive ing mutations in a tumor and it is currently investigated benefit from it, while a minority of them suffers from as a potential biomarker in non-small cell lung carcinoma severe side effects. Given the significant cost and non- (NSCLC) [8–10]. Accumulating evidence, however, sug- negligible toxicity of these therapies, the identification of gests its potential usefulness also in melanoma [8, 11–14], urothelial cancer [5, 15, 16], mismatch-repair deficient colorectal tumors [17] and other cancer types [18]. Its * Correspondence: laura.fancello@ieo.it; luca.mazzarella@ieo.it pattern and distribution is highly variable across different Department of Experimental Oncology, IEO, European Institute of Oncology IRCCS, Via Adamello 16, 20139 Milan, Italy cancer types, with over 1000-fold difference between Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 2 of 13 cancer types with the lowest mutational burden and those load based only on somatic nonsynonymous coding muta- with the highest mutational burden, such as those associ- tions, called by Whole Exome Sequencing (WES). Briefly, ated with DNA environmental damage, i.e. by exposure to somatic nonsynonymous coding mutations are identified tobacco smoke or UVs [19, 20]. Increased TMB was also by WES and, if RNA sequencing is also available, only mu- observed in tumors with defects in DNA mismatch repair tations occurring in expressed genes are retained. Peptides and DNA replication or in tumors characterized by micro- containing selected mutations are then identified in silico satellite instability, as in colorectal cancer [21, 22]. Highly and the efficiency of their presentation to the immune mutated tumors are more likely to produce abundance of system may be evaluated by mass spectrometry or by algo- tumor-specific mutant epitopes, which may function as rithms that consider their predicted affinity to the MHC neoantigens recognized as non-self by the immune sys- class I complex and patient-specific HLA class I alleles tem. Therefore, increased activation of immune cells by [14, 25]. In comparison with overall neoantigen load, treatment with immune checkpoint inhibitors may lead to TMB is easier to measure and correlates with it. Although improved immune-mediated tumor-cell clearance and not all mutations can give rise to tumor immunogenic clinical response in these tumors (Fig. 1). A significant as- peptides, their number influences the amount of neoanti- sociation between neoantigen production and immune- gens potentially produced. High TMB correlates with mediated clinical response was indeed observed in several long-term clinical benefit from immune checkpoint inhib- studies [9, 11, 14, 23]. Measurement of this neoantigen itors in patients with melanoma [14], NSCLC [9, 26–28] production, though, is expensive and time-consuming. and urothelial carcinoma [5, 15, 16, 29]. In addition to Tumor neoantigens can be generated by mutations or by that, patients with mismatch repair (MMR) deficient tu- gene fusions, especially out-of-frame fusions. Although mors are more responsive to immunotherapy, probably some pipelines have recently been developed for the iden- due to their high tumor mutational burden [17]. There- tification of neoantigens derived from gene fusions [24], fore, although not always capable to explain the clinical most research up to now has estimated overall neoantigen benefit alone, TMB is a good approximation for neoantigen Fig. 1 Tumor mutational burden as immunotherapy biomarker. Interaction between tumor mutational burden, neoantigen production and immune checkpoints. Hyper-mutated tumors (bottom) are more likely than hypo-mutated tumors (top) to generate tumor-specific peptides (neoantigens) recognized by the immune system. However, immune surveillance can be restrained by simultaneous high expression of PD-L1, which delivers a suppressive signal to T cells. PD-L1/PD-1 interaction and other immune checkpoints can be inhibited by immune checkpoint inhibitors, restoring immune response Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 3 of 13 load assessment [14], is technically less challenging and less panel-based and WES-based TMB quantification, based expensive and may represent a better suited predictive on matched sequencing by gene panel and WES of the biomarker for immunotherapy response. same tumor sample and comparison of matched TMB TMB may also represent a relevant prognostic bio- values (Additional file 3: Table S3, Additional file 9:Figure marker. In BRCA-1/2 mutated ovarian cancers, TMB S3). Accuracy of panel-based TMB quantification is influ- correlates with improved overall survival [30, 31]. In enced by statistical sampling effects and small panels breast cancer patients, tumors with high TMB and favor- provide less precise TMB estimates [22, 34–36]. It was able immune-infiltrate (“hot tumors”) are associated demonstrated that TMB values from the FoundationOne with prolonged survival [32]. Consistently, basal cell gene panel, which targets 1.1 Mb of genomic space, are carcinoma, which is characterized by very high TMB, similar to those from WES, whereas accuracy drops im- presents with slow growth rates and rare metastases. Al- portantly when sequencing less than 0.5 Mb [22]. Another though not definitively demonstrated, we can speculate study simulated sequencing of theoretical gene panels of that this less aggressive phenotype may be due to hyper- different sizes and identified 1.5 to 3 Mb as the best suited mutation, which would trigger, via increased neoantigen targeted genomic size to confidently estimate TMB [35]. production, a more effective immune response of the Moreover, the deviation between WES- and panel-based host [33]. TMB appears more relevant for samples with low to moderate underlying TMB levels, compared to samples Quantification of tumor mutational burden from with high TMB [22, 35, 36]. Another retrospective study gene panels: “yes we can” on a commercial panel of 248 genes likewise cautions Initial studies showing a correlation between TMB and against small gene panels which would lead to TMB enhanced response to immunotherapy were based on overestimation [37]. WES datasets for TMB quantification [9, 14, 17]. WES Besides the accuracy of panel-based TMB quantifica- allows a direct measurement of TMB, yet it remains tion, it is critical to assess its capability to discriminate unsuitable as routine technology in clinical practice, be- between immunotherapy responders and non- cause expensive, labor-intensive and time-consuming. responders, as previously observed for WES-based TMB. Therefore, several studies explored the possibility to pro- Several exploratory analyses demonstrated that panel- vide equally accurate and clinically predictive TMB based TMB, as simulated in silico by downsampling a estimates from targeted enrichment sequencing, using WES dataset to only include genes targeted by the Foun- various gene panels (Table 1, Additional file 1: Table S1). dationOne gene panel, associates with immunotherapy The main challenge for accurate panel-based TMB response [8, 26] or with signatures of immune check- quantification is the ability to extrapolate the global point activation [38]. Comparable results were observed mutational burden from the narrow sequencing space in similar in silico analyses for other gene panels, such targeted by a gene panel. In silico analyses were per- as the Trusight170 [39, 40] or MSK-IMPACT [26] formed to test the concordance between panel-based (Additional file 4: Table S4). Notably, direct measurement and WES-based TMB, which is considered the reference of TMB from the Oncomine Tumor Load Assay shows for TMB quantification. Publicly available WES datasets that this panel-based TMB value allows to classify colorec- were downsampled to the subset of genes targeted in the tal cancer cases based on their MSI status [39, 41]. Since panel under consideration and TMB values from such in this cancer type MSI positively correlates with simulated gene panels were compared with TMB values immunotherapy response, this is a further, yet indirect from the original WES (Additional file 7:Figure S1),find- evidence, of the capability to predict immunotherapy re- ing high correlation between the two (Additional file 2: sponse, using a panel-based TMB estimate. Most import- Table S2, Additional file 8:Figure S2).Mostofthese in antly, a few clinical studies demonstrated that TMB silico analyses were performed using publicly available directly estimated using gene panels is higher in those WES datasets from TCGA, with the exception of the patients who benefit more from immune checkpoint Oncomine Tumor Mutation Load Assay or NovoPM and blockade treatment, thus providing “real-life” evi- CANCERPLEX gene panels, for which WES datasets from dence for its potential clinical predictive value (Fig. 2, COSMIC or from other sources were used. Regardless, Additional file 5: Table S5). A direct association with similar correlation values were reported for the differ- immunotherapy response was shown for the MSK- ent gene panels tested (Additional file 2: Table S2, IMPACT [42, 43] and the Guardant360 gene panels [44] Additional file 8: Figure S2). For some of these gene but most of the reported studies utilized the Foundatio- panels (FoundationOne, Trusight170, Oncomine Tumor nOne gene panel (Fig. 2, Additional file 5: Table S5). In Mutation Load Assay, Oncomine Comprehensive Assay particular, in the CheckMate 227 trial, NSCLC patients V3 and MSK-IMPACT gene panels), an empirical ap- with high TMB (> 10 mutations per Mb, measured by proach was also used to test the concordance between FoundationOne) presented increased progression-free Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 4 of 13 Table 1 Overview of the main published studies on TMB quantification from gene panels Reference Gene panel Cancer type Study design Study ID ICI TMB cutoff Method of TMB TMB predictive Clinical N patients (version) (mut/Mb) cutoff value outcome determination Rosenberg, FM1 (v3) urothelial carcinoma trial (single-arm, NCT02108652 PD-(L)1 NA NA NA ORR 315 2016 [5] (metastatic or locally phase 2) advanced) Balar, 2017 FM1± urothelial carcinoma trial (single-arm, NCT02108652 PD-(L)1 Q3 (> = 16) distribution NA OS 123 [16] (metastatic) phase 2) Powles, 2018 FM1± urothelial carcinoma trial (randomized, NCT02302807 PD-(L)1 Q2 (9.65) distribution NA OS 931 [15] (metastatic) phase 3) Kowanetz, FM1 (v3) NSCLC trial (randomized, NCT01903993 PD-(L)1 Q1, Q2 (9.9), distribution NA PFS, OS, ORR 454 2016 [27] phase 2) Q3 trial (single-arm, NCT02031458 phase 2) trial (single-arm, NCT01846416 phase 2) Gandara, 2018 FM1 bTMB NSCLC trial (randomized, NCT01903993 PD-(L)1 > = 14 positive and NA PFS, OS 259 [61] assay phase 2) negative percentage trial (randomized, NCT02008227 agreement with phase 3) the orthogonally validated FM1 Hellmann, FM1 CDx NSCLC trial (randomized, NCT02477826 combo > 10 based on NA PFS 1004 2018 [50] phase 3) NCT02659059 Rizvi, 2018 [42] MSK-IMPACT NSCLC trial (randomized, NCT01295827 PD-(L)1 Q2 (7.4) distribution AUC = 0.601 DCB, PFS 240 (v1, v2, v3) phase 1) (DCB) Ready, 2019 FM1 CDx NSCLC trial (non-randomized, NCT02659059 combo 10 ROC AUC (95% ORR 98 [28] phase 2) CI) = 0.73 (0.62–0.84); TPR (95% CI) = 0.78 (0.63–0.93); FPR (95% CI) = (0.62 (0.49–0.73) Wang, 2019 NCC-GP150 NSCLC observational (cohort) NA PD-(L)1 6 (tot mut) best cutoff from NA PFS, ORR 50 [49] in silico analysis on Rizvi 2015 WES Johnson, 2016 FM1 (v2, v3) melanoma observational NA PD-(L)1 < 3.3, 3.3–23.1, ROC NA PFS, OS, ORR 65 [12] (retrospective) >23.1 Chalmers, FM1 (v1, v2, various locally advanced observational NA NA > 20 NA NA NA 102, 292 2017 [22] v3, v4), FM1 or metastatic solid tumors (retrospective) Heme Goodman, FM1 (v1, v2, various locally advanced observational (cohort, NCT02478931 PD-(L)1, < 6, 6–19, > 19 Foundation NA PFS, OS, ORR 151 Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 5 of 13 Table 1 Overview of the main published studies on TMB quantification from gene panels (Continued) Reference Gene panel Cancer type Study design Study ID ICI TMB cutoff Method of TMB TMB predictive Clinical N patients (version) (mut/Mb) cutoff value outcome determination 2017 [18] v3) or metastatic solid tumors retrospective) CTLA-4, Medicine official high-dose reports IL2 or combo Khagi, 2017 Guardant360 various solid tumors observational (cohort, NCT02478931 PD-(L)1, mean (> 3 distribution NA PFS, OS, ORR 69 [44] retrospective) CTLA-4, VUS) combo or other Zehir, 2017 MSK-IMPACT various primary and observational (cohort, NCT01775072 NA > 13.8 distribution NA NA 10, 945 [73] (v1, v2) metastatic solid tumors prospective) (median TMB + 2 × IQR_TMB) Samstein 2019 MSK-IMPACT bladder observational (cohort, NCT01775072 PD-(L)1, 17.6 distribution (top NA OS, PFS, DCB 214 [43] (v3) prospective) CTLA-4 or 20%) breast 5.9 45 combo breast ER+ 6.8 24 breast ER- 4.4 21 unknown primary 14.2 90 colorectal 52.2 110 esophagogastric 8.8 126 glioma 5.9 117 head and neck 10.3 138 melanoma 30.7 321 NSCLC 13.8 350 renal cell carcinoma 5.9 151 ORR Objective Response Rates, DCB Durable Clinical Benefit, OS Overall Survival, PFS Progression-Free Survival, FM1 Foundation Medicine’s FoundationOne (v1: 185 genes, v2: 236 genes, v3: 315 genes, v4: 405 genes, Heme: 405 genes, CDx: 324 genes); ±: version not specified; MSK-IMPACT v1 341 genes, v2: 410 genes, v3 468 genes, NSCLC non-small cell lung cancer, ER Estrogen Receptor, VUS variants of unknown significance, PD- (L)1 anti-PD-1 or anti-PD-L1, CTLA-4 anti-CTLA-4, combo combined anti-PD-1/PD-L1 + anti-CTLA-4, Q1-Q4 quartiles, : TMB quantification from blood Each study is described reporting gene panel, cancer type, study design, study ID (on ClinicalTrials.gov), immune checkpoint inhibitor treatment (ICI), proposed TMB cutoff, method for TMB cutoff determination, outcome analyzed to evaluate TMB clinical utility. AUC, TPR (True Positive Rate) and FPR (False Positive Rate) are provided, when available, as a measure of TMB predictive value for immunotherapy responder classification Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 6 of 13 Fig. 2 TMB association with progression-free survival. Forest plot of hazard ratios (HR) comparing progression-free survival (PFS) between patients with high or low TMB, as indicated in the “Comparison” column. If not specified otherwise, TMB is reported as number of mutations per Mb. All patients were treated with immune checkpoint inhibitors (ICI). Bars represent the 95% confidence intervals. Size of the box is proportional to precision. Reference to the study and the analyzed cancer type are also reported together with the log-rank p-value. Q1-Q4: quartiles; VUS: variants of unknown significance. *: TMB quantified from blood; **: Cox proportional hazards model adjusted for age, gender, disease stage and prior therapy by ipilimumab survival after combined anti-CTLA-4 plus anti-PD-1/ and by the capability to extrapolate TMB values from PD-L1 therapy [45]. Interestingly, TMB was predictive the restricted genomic space sampled by gene panels to of anti-PD-L1 monotherapy response in NSCLC the whole genome (Fig. 3a). Experimental factors (e.g. (POPLAR trial, [27]) and metastatic urothelial carcin- tumor purity or sequencing depth) and the variant oma patients [5, 15, 16], independently from the PD-L1 calling pipeline (e.g. the variant calling algorithm and expression status. Analysis of archival tumor samples from the method to remove germline variants) can signifi- melanoma patients treated by anti-PD-1/PD-L1 mono- cantly affect the number of called somatic mutations therapy also showed superior response rates, progression- and have a similar impact on both panel-based and free survival and overall survival in high TMB cases [12]. WES-based TMB quantification. Indeed, the adoption of Moreover, a retrospective study on 151 patients across a well-documented standard pipeline was already diverse tumor types showed that cancer patients with claimed for WES analyses as an urgent need to allow higher TMB, benefit more from anti-PD-1/PD-L1, anti- data interoperability between different platforms [46]. CTLA-4 or high dose IL2 monotherapy [18]. The same The same applies to panel sequencing for TMB quantifi- was not observed for combined anti-PD-1/PD-L1 plus cation. In this context, an important factor investigated anti-CTLA-4 therapy but the available number of samples for its influence on the number of called somatic may be too small to draw conclusions [18]. variants is the method chosen to identify and remove germline variants. Indeed, since only somatic mutations Need for standardization of TMB quantification can potentially produce tumor neoantigens recognized and reporting as non-self by the immune system, it is important to Despite the increasing number of studies showing the remove germline variants in TMB quantification. It was potential clinical relevance of panel-based TMB as a pre- observed that the use of an in silico method for somatic dictive biomarker for immunotherapy response, its use variant calling instead of matched tumor-normal sam- in the clinical setting is currently limited by the absence ples, leads to increased false positive somatic variants, of standard methods of quantification and the lack of a which has an important influence on the accuracy of robust and universal cutoff to identify immunotherapy TMB quantification, especially for small gene panels responders. [34]. To avoid this, it was proposed to perform TMB Panel-based TMB quantification is influenced by vari- quantification using only high-confidence regions [47] ous experimental factors affecting library construction (e.g. regions of the genome, devoid of potential system- and sequencing, by the pipeline used to call mutations atic biases or structural variants, where mutations can Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 7 of 13 Fig. 3 Differences in the workflow for panel-based TMB quantification. a. Overview of the factors influencing panel-based TMB quantification. Several variables in library construction, sequencing and in the pipeline to call mutations influence panel-based TMB quantification. Furthermore, panel-based TMB quantification is influenced by differences in the bioinformatic method to extrapolate global TMB from mutations identified in the narrow genomic region targeted by the gene panel. b. Differences across various studies in panel-based TMB quantification: gene panel technical specifications, preanalytical factors and the bioinformatics workflow used to extrapolate from the genomic space targeted by gene panels global TMB are described. FM1: Foundation Medicine’s FoundationOne panel (v1: 185 genes, v2: 236 genes, v3: 315 genes, v4: 405 genes); NA: not available; ±: algorithm developed by Sun et al. for in silico removal of germline variants [74] be confidently called), as defined by Zook et al. [48]. It mutations. Therefore, it was proposed to remove known was also observed that increased somatic false positives cancer variants of targeted genes when performing TMB are generated by the in silico germline filtering method quantification, to avoid overestimation of TMB when for patients with non-caucasian ancestry compared to extrapolating it across the whole genome [12, 22]. Buch- caucasian patients, as the former are less represented in halter et al., showed that removal of cancer mutational public databases used for germline variant filtering [34]. hotspots slightly decreases the number of high TMB tu- The use of ExAC, the largest and more representative mors identified but does not change the general picture public germline WES database, in addition to dbSNP [35]. However, the importance of this filtering, routinely and 1000 Genomes, is recommended to reduce this performed only for Foundation Medicine panels, de- difference [34]. pends on panel size and composition: some gene panels In addition to these factors, which similarly influence may be larger and less enriched in cancer genes by WES- and panel-based analyses, panel-based TMB including, for example, pharmacogenomic variants. As quantification also requires to extrapolate the number of for synonymous mutations, it is claimed that, although somatic coding mutations observed in the targeted not biologically meaningful, their inclusion may reduce genomic space to the number that would be observed sampling noise and improve the approximation of TMB across the whole genome. Extrapolation methods may across the whole genome. Indeed, several works com- differ for various choices in variant filtering, such as pared TMB quantification with or without synonymous removal of known cancer mutations or synonymous mu- variants and observed that, when including synonymous tations (Fig. 3b). Standard gene panels are commonly variants, panel-based TMB shows increased correlation enriched in known cancer genes, which are more likely with WES-based TMB values [35, 49] and stronger asso- to be mutated in a tumor and expectedly enriched in ciation with clinical response [9]. Starting from the Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 8 of 13 above observations, we can infer that some recommen- most commonly established using distribution-based strati- dations to build a standardized and robust analysis pipe- fication, which can be heavily influenced by outliers, while line for TMB quantification are starting to emerge at it is now often identified based on statistically sound least for the following points: i. germline variants can be methods, such as ROC curves. In the future, the use of most efficiently identified and removed by matched nor- ROC curves as a common method of TMB cutoff deter- mal sequencing and, if this is not possible, the largest mination will greatly help to converge to a robust TMB cut- available germline variant databases should be used for in off and will facilitate comparison across different platforms. silico filtering, especially for non-caucasian ancestries; ii. The heterogeneity in experimental and analytical pro- TMB extrapolation to the whole genome is accurately per- tocols, in the extrapolation of panel-based TMB values formed by counting all somatic mutations, including syn- and in gene panel technical specifications is currently onymous mutations, to enlarge the sampling space and limiting the potential use of TMB in a clinical setting better approximate global TMB across the whole genome. (Fig. 3). For this reason, a common standard for TMB As already mentioned, panel-based TMB quantifica- quantification and a consensus on a clinically useful tion is affected by the genomic size targeted by the panel TMB cutoff are urgently needed. Some efforts in this and by its gene composition. Notably, gene panels tested sense are ongoing by the Quality Assurance Initiative up to now widely differ for number of targeted genes Pathology (QuIP) in Germany (https://quip.eu) and by (from 73 to 710) and size (from 0.39 to 2.8 Mb of tar- the European Society of Pathology. Moreover, in the US, geted genomic space) (Additional file 1: Table S1). These governmental organizations, health-sector industries, considerations raise the question of how to convert several NGS gene panel manufacturers and academic in- TMB estimates between different gene panels to allow stitutions set up a TMB Harmonization Working Group cross-platform comparability. Indeed, although the ma- (https://www.focr.org/tmb) and planned a 3-phase pro- jority of studies correlating TMB to immunotherapy re- ject for TMB harmonization. In the first phase, they per- sponse are currently based on FoundationOne, several formed in silico analyses of publicly available TCGA data other types of gene panels exist and the offer is steadily to identify sources of variability in TMB quantification increasing (Additional file 1: Table S1). Moreover, we between WES and gene panels. Recently concluded, this still need to standardize the conversion of the reference work established that panel-based TMB is comparable WES-based TMB values to panel-based TMB, since the between different gene panels for TMB values ranging 0 lower sequencing coverage and higher sequencing depth to 40 mutations per Mb, that it strongly correlates with of gene panels, as compared to WES, may lead to de- WES-based TMB and it is possibly influenced by the creased accuracy of TMB values and increased sensitivity type of cancer under investigation. It also found that the in variant calling. For cross-panels or panel-to-WES observed variance across gene panels stems from their TMB conversion, an in silico approach was proposed, different gene composition and technical specifications, where TMB distributions derived from two different as well as from the bioinformatic pipeline adopted [53]. technologies were interpolated and aligned and TMB The second step of the project empirically validates cutoffs were mapped across distributions [38]. However, TMB estimates from different gene panels by mapping a consensus on a standard method to convert TMB them to WES-based TMB values, used as a gold stand- values is still missing. ard, whereas the last step will define best practices for Not surprisingly, in this heterogeneous landscape a ro- TMB use as immunotherapy biomarker in clinical rou- bust TMB cutoff to discriminate between immunother- tine. Following preliminary results on the influence of apy responders and non-responders is still to be defined. the bioinformatic workflow and of gene panel size and Moreover, the adopted cutoffs sometimes differ across composition on TMB quantification, the working group different studies on the same gene panel (Table 1). Up to recommends the use of gene panels larger than 1 Mb now, the TMB cutoff of 10 mutations per Mb, measured and the standardization of the bioinformatic algorithms, by the FoundationOne gene panel and found to best in addition to standardization of sample processing. discriminate between responders and non-responders to Moreover, it suggests the inclusion of actionable genes, immunotherapy in NSCLC patients, is the only one genes associated with mutagenesis and negative predic- which has been validated in a separate further study tors of response in these gene panels and the alignment [28, 50, 51]; this cutoff was also observed, but not yet of panel-based TMB values to WES-based ones to allow validated, in melanoma [38] and in metastatic urothelial interoperability across different assays [54]. carcinoma [15](Table 1). Interestingly, these cancer types present a TMB distribution similar to that of NSCLC [52]. TMB quantification beyond tissue biopsies and Indeed, due to the diversity of TMB distribution across dif- current gene panels ferent cancer types, the adoption of cancer-specific TMB Most studies on TMB as a predictive biomarker for im- cutoffs was proposed [35, 43]. TMB cutoff was initially munotherapy response were performed on bioptical or Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 9 of 13 surgical specimens from solid tumors. Since obtaining main pitfalls of the panel, though, is its limit of detection, tissue biopsies may be challenging and invasive for defined as a minimum of 1% tumor content in at least 20 patients, it would be critical for the clinical routine to ng of cell-free DNA input, and its dependency on the assess TMB using cell-free DNA (cfDNA) from blood, overall tumor burden, which influences the likelihood of which includes circulating tumor DNA (ctDNA), as a detecting ctDNA. The exclusive use of single nucleotide surrogate specimen to biopsy. High throughput molecu- variants (SNVs) for TMB quantification represents an- lar profiling of ctDNA remains technically challenging other limitation, although future versions of the algorithm but increasing efforts are being made in this direction. A are planned to be released, which will also use indels. The few studies previously investigated the feasibility of WES commercial Guardant360 and GuardantOMNI gene on ctDNA and highlighted some inherent limitations, panels were also designed for blood-based TMB quantifi- such as the low amount of available ctDNA, which re- cation [62]. Their limit of detection was defined as a mini- duces sensitivity, or ctDNA being more associated with mum of 0.3% tumor content in at least 5 ng of cell-free metastases rather than with primary tumors [55–59]. In DNA input. They were validated in silico by subsetting one of the largest studies attempting optimization of TCGA WES datasets to only include genes targeted by WES-based TMB quantification from liquid biopsy, the panels. Panel accuracy in TMB quantification was then WES was performed in parallel on DNA from tissue evaluated by correlation of TMB values obtained from the biopsies and on cfDNA from liquid biopsies of 32 meta- simulated gene panels with those from WES. Their pre- static patients and comparable sequencing depth and dictive value was similarly evaluated in silico on 30 lung coverage were obtained [60]. Performance of variant de- cancer samples with matched information on immuno- tection was dependent on the fraction of tumor DNA therapy response. The performance showed by TMB from within the analyzed cfDNA, as previously described. In the simulated gene panel in responder identification was those samples positive for the presence of ctDNA, vari- comparable to that of WES-based TMB (Additional file 4: ant detection sensitivity of cfDNA-WES compared to Table S4). The Guardant360 panel was further tested in a tDNA-WES was 94%, regardless of the tumor type (2 small retrospective study on 69 patients with various cholangiosarcoma and 19 lung, 5 head and neck, 2 tumor types [44]. No comparison with tissue-based TMB prostate, 2 colorectal, 1 breast and 1 bladder cancer were has been reported yet, but a significant correlation be- analyzed). Most importantly, in ctDNA positive samples, tween high blood-derived TMB measured by Guard- TMB values from WES on liquid biopsies were robust ant360 and immunotherapy response was observed [44]. and consistent with those from WES on tissue biopsies, Finally, a further gene panel for bTMB quantification was which demonstrates for the first time the feasibility of recently developed in China. Consistency between panel- TMB quantification from liquid biopsies, using WES. based bTMB values and WES-based tTMB values, tested More recently, it was demonstrated that targeted in silico and empirically by matched blood and tissue sam- enrichment sequencing by gene panels is another valid ples, was comparable to that of the panels described approach for TMB quantification from liquid biopsies. above. Similar results were also found for its predictive In particular, Gandara et al. developed, tested and analyt- value, based on in silico analyses. Interestingly, the authors ically validated a novel gene panel for TMB quantification also raised the issue of the different frequency of onco- from blood [61]. The panel is based on hybridization- genic driver mutations, such as EGFR or KRAS, between capture enrichment and targets 394 genes, corresponding Asian and white population. For this reason, they compare to 1.1 Mb of genomic space (Fig. 3). Its clinical utility was TCGA WES-based TMB with panel-based TMB with or evaluated via a retrospective study on 259 NSCLC samples without inclusion of EGFR and/or KRAS mutations. Al- from patients treated with immunotherapy or chemother- though similar results are yielded, the raised issue is an apy in the OAK and POPLAR clinical trials. Blood- important point to be further investigated in panel-based derived TMB (bTMB) calculated using this novel gene TMB quantification [49]. TMB quantification from liquid panel correlated well with tissue-derived TMB (tTMB) biopsies suffers from ctDNA detection limits, which also measured by FoundationOne. Moreover, measured TMB depend on tumor size and number of cancer cells, but was found to be significantly associated with response to these results encourage to further explore and more ex- anti-PD-L1 immunotherapy in the POPLAR trial and this tensively validate this approach. was further confirmed on patient samples from the OAK Besides new technologies to estimate TMB from liquid trial. A prospective validation is also currently ongoing in biopsies, another significant step towards routine use of the BFAST trial (NCT03178552) on advanced and meta- TMB in clinical practice is TMB quantification from an static NSCLC patients. Interestingly, it was observed that even smaller set of genes than in targeted enrichment the capability of TMB, as measured by this panel, to gene panels. Although panel size is known to affect predict anti-PD-1/PD-L1 immunotherapy response is in- accuracy of TMB quantification, the use of a highly cus- dependent from PD-L1 expression levels [61]. One of the tomized set of genes may represent a valid and even less Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 10 of 13 expensive approach. In this view, Lyu et al., proposed a et al. observed a relationship between subclonal mutations computational framework to assess the best and smallest and mutational signatures associated with alkylating agents subset of genes necessary to estimate TMB as a bio- and, in NSCLC, between clonal mutations and mutational marker for lung adenocarcinoma [63]. They were able to signatures associated with smoking [65]. Mutational signa- identify a model of only 24 genes which predicted in tures associated with smoking were also found to be sig- silico immunotherapy response with 93% specificity and nificantly associated with high tumor mutational burden 85% sensitivity and they suggested that other small and with response to immunotherapy [9]. Therefore, al- custom sequencing gene panels may be designed in a though the extraction of mutational signatures from gene cancer-specific way to assess TMB with further reduced panels may be hampered by the small number of sampled costs. mutations, these observations suggest that they may prove helpful to infer neoantigen clonality and enhance TMB Future perspectives and recommendations predictive value. TMB is one of the most rapidly developing biomarkers Integration of TMB with other potential immunother- for immunotherapy response, with about 37 ongoing apy biomarkers represents another promising way to re- clinical trials currently registered in ClinicalTrials.gov fine prediction of immunotherapy responders. For that use TMB as stratification biomarker [64]. Several example, TMB, defects in DNA mismatch-repair path- gene panels were recently optimized to estimate TMB at way and the MSI status all are measures of genomic reduced sequencing costs, and emerging evidence sup- instability that can provide indirect assessment of tumor ports the feasibility of TMB quantification from liquid antigenicity, whereas PD-L1 expression, immune cell biopsies. However, harmonization in TMB quantification infiltration and inflammatory signatures represent bio- and reporting remains the main challenge for the near markers of the T cell-inflamed tumor microenvironment. future: standard procedures are required to allow inter- Therefore, their integration can refine prediction of im- operability between different gene panels, compare munotherapy outcome by combining information on results across studies and define a universal cutoff to tumor complexity and on the immune response. Indeed, confidently identify patients most likely to benefit from emerging evidence suggests that, at least in NSCLC, immunotherapy. TMB and PD-L1 expression are independent predictors Even an accurate TMB value is an imperfect predictor and TMB may complement or even outperform PD-L1 of immunotherapy response and further studies are expression [10, 26, 50, 66]. Moreover, it was observed needed to enhance its value as clinically useful immuno- that most tumors with high MSI also present elevated therapy biomarker. TMB is used as an approximation of TMB, whereas the opposite does not hold true. The neoantigen burden, upon the assumption that the higher combination of TMB with MSI and PD-L1 expression in the mutational burden, the higher the probability for gastrointestinal tumors significantly improved the identi- immunogenic peptides to be generated, which leads to fication of immunotherapy responders [67]. In another stronger immune response upon inhibition of immune study, it was observed that TMB is an independent pre- checkpoints. Interestingly, neoantigen clonality, in addition dictor and only weakly correlates with T cell-inflamed to the overall amount of neoantigens, influences immuno- gene expression profiles (GEP) or PD-L1 expression. therapy response in NSCLC patients [65]. In particular, Thus, TMB and T cell-inflamed GEP were jointly used tumors enriched in clonal neoantigens (e.g. present in all to identify immunotherapy responders: patients with tumor cells) are more sensitive to immune checkpoint in- both high TMB and high T cell-inflamed GEP were those hibitors than tumors enriched in subclonal neoantigens with the highest objective response rates on tumors from (e.g. present only in a subset of tumor cells), in advanced four KEYNOTE clinical trials across 22 cancer types. NSCLC and melanoma patients [65]. Indeed, clonality of Similarly, in melanoma patients, a response score based produced neoantigens seems to be associated with a more on the combination of TMB, infiltration of CD8+ T cells effective immune surveillance. On the other hand, enrich- and gene expression profiles for PD-L1, CD8 and a set ment in subclonal neoantigens may activate T cells against of 394 immune genes demonstrated higher sensitivity only a subset of tumor cells, leading to less effective tumor and similar specificity than each biomarker alone [68]. control. Based on these observations, it would be interest- To date, the FoundationOne and Guardant360 gene ing to investigate if information on mutation clonality (e.g. panels allow to measure both TMB and MSI but no variant allele frequency) improves the predictive power of other potential immunotherapy biomarker. Moreover, TMB. Evaluation of mutation clonality from gene panels is they do not provide the user any combinatorial model to not trivial though: the reduced genomic space targeted by integrate them. Although further validation in prospect- gene panels may not be representative of the overall clonal ive clinical studies is required for all these potential architecture and the mutations sampled herein may not be biomarkers, several observations suggest that simultan- those generating neoantigens. Interestingly, McGranahan eous profiling of both TMB and other immunotherapy Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 11 of 13 biomarkers currently under investigation may represent Additional file 7: Figure S1. Visual representation of the method used the next step forward in the design of new gene panels for in silico analyses on TMB quantification accuracy and on association or predictive value for immunotherapy response. In silico analyses are for clinical use. The Friends and QuIP initiatives for based on simulations of panel performance, wherein TMB is calculated TMB harmonization recommended to include as much using a subset of WES which only contains genes targeted by the panel. relevant genetic and molecular information as possible Accuracy of TMB quantification from the simulated gene panel is evaluated by comparison with WES-based TMB, used as gold reference, in these panels, to avoid the need to re-biopsy the with correlation analysis. The clinical predictive value of TMB estimated patient for further information. In line with this recom- from the simulated panel is evaluated based on its association with mendation, we propose to also include in gene panels clinical values measuring immunotherapy response. (PDF 55 kb) for TMB quantification other potential immunotherapy Additional file 8: Figure S2. In silico analysis of the correlation between panel-based and WES-based TMB. Correlation between panel- biomarkers but also negative predictors of immunother- based and WES-based TMB, considered the gold standard value, is used apy response [69, 70] and variants predisposing to to estimate the accuracy of panel-based TMB quantification. Panel-based adverse reaction to immunotherapy [71, 72]. These and TMB quantification was simulated in silico using a subset of WES which only contains genes targeted by the panel. The bubble plot shows on other recommendations which emerge from the studies the x axis the correlation coefficients and on the y axis the gene panel reviewed here, including the one from the TMB and the cancer type. Bubble size represents the number of data points Harmonization Working Group, are summed up in used in the analysis and the color corresponds to the reference study. (PDF 259 kb) Additional file 6: Table S6. Additional file 9: Figure S3. Empirical analysis of the correlation between panel-based and WES-based TMB. Correlation between panel- based and WES-based TMB, considered the gold standard value, is used to estimate the accuracy of panel-based TMB quantification. Correlation Additional files analysis is performed on TMB values calculated for samples with matched panel and whole exome sequencing. The bubble plot shows on the x axis the correlation coefficients and on the y axis the gene panel and the cancer Additional file 1: Table S1. Technical specifications of gene panels type. Bubble size represents the number of data points used in the analysis used or proposed for TMB quantification. For each gene panel, it is and the color corresponds to the reference study. (PDF 155 kb) reported the type of cancer and sample for which it was designed, the enrichment method, the targeted sequencing size (Genomic space) and the number of targeted genes (# genes). (XLSX 6 kb) Abbreviations Additional file 2: Table S2. In silico analysis of the correlation between ACC: Adrenocortical carcinoma; AUC: Area under the curve; BLCA: Bladder panel-based and WES-based TMB. Correlation between panel-based and urothelial carcinoma; BRCA: Breast invasive carcinoma; CESC: Cervical WES-based TMB, considered the gold standard value, is used to estimate squamous cell carcinoma and endocervical adenocarcinoma; the accuracy of panel-based TMB quantification. Panel-based TMB quanti- cfDNA: Circulating free DNA; CHOL: Cholangiosarcoma; COADREAD: Colon fication was simulated in silico using a subset of WES which only contains adenocarcinoma; CRC: Colorectal cancer; ctDNA: Circulating tumor DNA; genes targeted by the panel. (XLSX 11 kb) DLBC: Lymphoid neoplasm diffuse large B-cell lymphoma; ESCA: Esophageal carcinoma; FDA: Food and Drug Administration; GBM: Glioblastoma; Additional file 3: Table S3. Empirical analysis of the correlation GEP: Gene expression profile; HLA: Human Leukocyte Antigen; HNSC: Head between panel-based and WES-based TMB. Correlation between panel- and neck squamous cell carcinoma; KICH: Kidney chromophobe; KIRC: Kidney based and WES-based TMB, considered the gold standard value, is used renal clear cell carcinoma; KIRP: Kidney renal papillary cell carcinoma; to estimate the accuracy of panel-based TMB quantification. Correlation LAML: Acute myeloid leukemia; LGG: Brain lower grade glioma; LIHC: Liver analysis is performed on TMB values calculated for samples with matched hepatocellular carcinoma; LUAD: Lung adenocarcinoma; LUSC: Lung panel and whole exome sequencing. (XLSX 6 kb) squamous carcinoma; Mb: Megabase; mCRPC: Metastatic castration-resistant Additional file 4: Table S4. In silico analysis of TMB association or prostate cancer; MESO: Mesothelioma; MHC: Major histocompatibility predictive value for immunotherapy response. These analyses were performed complex; MMR: Mismatch repair; MSI: Microsatellite instability; NSCLC: Non- on panel-based TMB values simulated in silico using a subset of WES which small cell lung cancer; ORR: Objective response rates; OS: Overall survival; only contains genes targeted by the panel. The table reports measures of TMB OV: Ovarian serous cystadenocarcinoma; PAAD: Pancreatic adenocarcinoma; association with immunotherapy response (odds ratios, hazard ratios and PCPG: Pheochromocytoma and paraganglioma; PFS: Progression free survival; corresponding p-values), differences in TMB distribution between responders PRAD: Prostate adenocarcinoma; ROC: Receiver operating characteristic; and non-responders (Mann-Whitney U and Fisher’s p values) and measures of SARC: Sarcoma; SCLC: Small cell lung cancer; SKCM: Skin cutaneous TMB predictive value (AUC, specificity, sensitivity). (XLSX 9 kb) melanoma; SNV: Single nucleotide variant; STAD: Stomach adenocarcinoma; Additional file 5: Table S5. Empirical analysis of TMB association or TCGA: The cancer genome atlas; TCR: T cell receptor; TGCT: Testicular germ predictive value for immunotherapy response. These analyses were cell tumors; THYM: Thymoma; TMB: Tumor mutational burden; UCEC: Uterine performed on panel-based TMB values, directly calculated by panel se- corpus endometrial carcinoma; UCS: Uterine carcinosarcoma; UVM: Uveal quencing. The table reports measures of TMB association with immuno- melanoma; WES: Whole exome sequencing therapy response (odds ratios, hazard ratios and corresponding p-values), differences in TMB distribution between responders and non-responders Acknowledgements (Mann-Whitney U, unpaired Student’s t and Fisher’s test p values) and We thank all researchers, clinicians and organizations working in this field for measures of TMB predictive value (AUC, specificity, sensitivity). We also their contributions and we apologize to those whose work we did not specify how patients were stratified (“Comparison”), the method used to report or cite. determine TMB cutoff, the cohort considered for the analysis (if different co- horts were analyzed in the study), the type of immunotherapy, cancer type Authors’ contributions and number of patients. (XLSX 19 kb) LF designed, wrote and revised the manuscript. PGP, SG and LM revised the Additional file 6: Table S6. Proposed recommendations for consistent manuscript. All authors read and approved the final manuscript. TMB quantification and reporting. We report recommendations formulated by the TMB Harmonization Working Group (https://www.focr.org/tmb)as Funding well as indications emerging from the studies reviewed in this work. LF work was supported by the Italian Ministry of Health (Ricerca Corrente, (XLSX 24 kb) “Alleanza Contro il Cancro” - ACC network). Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 12 of 13 Availability of data and materials 16. Balar AV, Galsky MD, Rosenberg JE, Powles T, Petrylak DP, Bellmunt J, et al. Not applicable. Atezolizumab as first-line treatment in cisplatin-ineligible patients with locally advanced and metastatic urothelial carcinoma: a single-arm, multicentre, phase 2 trial. Lancet. 2017;389(10064):67–76. Ethics approval and consent to participate 17. Le DT, Uram JN, H W, R BB, Kemberling H, Eyring AD, et al. PD-1 blockade in Not applicable. tumors with mismatch-repair deficiency. NEJM. 2015:2509–20. 18. Goodman AM, Kato S, Bazhenova L, Patel SP, Frampton GM, Miller V, et al. Consent for publication Tumor mutational burden as an independent predictor of response to Not applicable. immunotherapy in diverse cancers. Mol Cancer Ther. 2017. https://doi.org/ 10.1158/1535-7163.MCT-17-0386. 19. Lawrence MS, Stojanov P, Polak P, Kryukov GV, Cibulskis K, Sivachenko A, et Competing interests al. Mutational heterogeneity in cancer and the search for new cancer- LF, SG, PGP, and LM declare that they have no competing interest. associated genes. Nature. 2013;499(7457):214–8. 20. Alexandrov LB, Nik-Zainal S, Wedge DC, Aparicio SAJR, Behjati S, Biankin AV, Author details et al. Signatures of mutational processes in human cancer. Nature. 2013; Department of Experimental Oncology, IEO, European Institute of Oncology 500(7463):415–21. IRCCS, Via Adamello 16, 20139 Milan, Italy. Department of Oncology and 21. Quiroga D, Lyerly HK, Morse MA. Deficient mismatch repair and the role of Hemato-Oncology, University of Milan, via Santa Sofia 9, 20142 Milan, Italy. immunotherapy in metastatic colorectal cancer. Curr Treat Options in Division of Early Drug Development, IEO, European Institute of Oncology Oncol. 2016. https://doi.org/10.1007/s11864-016-0414-4. IRCCS, Milan, Italy. 22. Chalmers ZR, Connelly CF, Fabrizio D, Gay L, Ali SM, Ennis R, et al. Analysis Received: 20 December 2018 Accepted: 19 June 2019 of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med. 2017;9(1):1–14. 23. Rooney MS, Shukla SA, Wu CJ, Getz G, Hacohen N. Molecular and genetic properties of tumors associated with local immune cytolytic activity. Cell. References 2015. https://doi.org/10.1016/j.cell.2014.12.033. 1. Antonia SJ, Villegas A, Daniel D, Vicente D, Murakami S, Ri H, et al. 24. Zhang J, Mardis ER, Maher CA. Genome analysis INTEGRATE-neo : a pipeline Durvalumab after chemoradiotherapy in stage III non-small cell lung cancer. for personalized gene fusion neoantigen discovery. Bioinformatics. 2017; N Engl J Med. 2017;377:1919–29. doi:https://doi.org/10.1093/bioinformatics/btw674. 2. Borghaei H, Paz-Ares L, Horn L. Nivolumab versus docetaxel in advanced 25. Yadav M, Jhunjhunwala S, Phung QT, Lupardus P, Tanguay J, Bumbaca S, et nonsquamous non-small-cell lung cancer. N Engl J Med. 2015;373:1627–39. al. Predicting immunogenic tumour mutations by combining mass 3. Hodi F, O’Day S, McDermott D, Weber R, Sosman J, Haanen J, et al. spectrometry and exome sequencing. Nature. 2014;515(7528):572–6. Improved survival with ipilimumab in patients with metastatic melanoma. N 26. Hellmann M, Nathanson T, Rizvi H. Genomic features of response to Engl J Med. 2010;363:711–23. combination immunotherapy in patients with advanced non- small-cell 4. Motzer R, Tannir N, McDermott D, Frontera O, Melichar B, Choueiri T, et al. lung cancer. Cancer Cell. 2018;33:843–52. Nivolumab plus Ipilimumab versus Sunitinib in advanced renal-cell 27. Kowanetz M. Tumor mutation load assessed by FoundationOne (FM1) is carcinoma. N Engl J Med. 2018;378(14):1277–90. associated with improved efficacy of atezolizumab (atezo) in patients with 5. Rosenberg JE, Hoff J, Powles T, Van Der HMS, Balar AV, Necchi A, et al. advanced NSCLC. Ann Oncol. 2016;27(6):15–42. Atezolizumab in patients with locally advanced and metastatic urothelial 28. Ready N, Hellmann MD, Awad MM, Otterson GA, Gutierrez M, Gainor JF, et carcinoma who have progressed following treatment with platinum-based al. First-line Nivolumab plus Ipilimumab in advanced non–small-cell lung chemotherapy: a single-arm, multicentre, phase 2 trial. Lancet. 2016. https:// cancer (CheckMate 568): outcomes by programmed death ligand 1 and doi.org/10.1016/S0140-6736(16)00561-4. tumor mutational burden as biomarkers. J Clin Oncol. 2019. https://doi.org/ 6. Li X, Song W, Shao C, Shi Y, Han W. Emerging predictors of the response to 10.1200/JCO.18.01042. the blockade of immune checkpoints in cancer therapy. Cell Mol Immunol. 29. Galsky M, Saci A, Szabo P, Azrilevich A, Horak C, Lambert A, et al. Impact of 2019. https://doi.org/10.1038/s41423-018-0086-z. tumor mutation burden on Nivolumab efficacy in second-line urothelial 7. Galluzzi L, Chan TA, Kroemer G, Wolchok JD, López-Soto A. The hallmarks of carcinoma patients: exploratory analysis of the phase II CheckMate 275. Ann successful anticancer immunotherapy. Sci Transl Med. 2018;10(459):1–15. Oncol. 2017. https://doi.org/10.1093/annonc/mdx371. 8. Campesato LF, Barroso-sousa R, Jimenez L, Camargo AA. Comprehensive 30. Strickland KC, Howitt BE, Shukla SA, Rodig S, Ritterhouse LL, Liu JF, et al. cancer-gene panels can be used to estimate mutational load and predict clinical Association and prognostic significance of BRCA1/2-mutation status with benefit to PD-1 blockade in clinical practice. Oncotarget. 2015;6(33):34221. neoantigen load, number of tumor-infiltrating lymphocytes and expression 9. Rizvi NA, Hellmann MD, Snyder A, Kvistborg P, Makarov V, Havel JJ, et al. of PD-1/PD-L1 in high grade serous ovarian cancer. Oncotarget. 2016;7(12): Mutational landscape determines sensitivity to PD-1 blockade in non – 13587–98. small cell lung cancer. Science. 2015;348(6230):124–9. 31. Birkbak NJ, Kochupurakkal B, Izarzugaza JMG, Eklund AC, Li Y, Liu J, et al. 10. Carbone DP, Reck M, Paz-Ares L, Creelan B, Horn L, Steins M, et al. First-line Tumor mutation burden forecasts outcome in ovarian cancer with BRCA1 Nivolumab in stage IV or recurrent non–small-cell lung Cancer. N Engl J or BRCA2 mutations. PLoS One. 2013;8(11). Med. 2017. https://doi.org/10.1056/NEJMoa1613493. 32. Thomas A, Routh ED, Pullikuth A, Jin G, Su J, Chou JW, et al. Tumor 11. Van Allen EM, Miao D, Schilling B, Shukla SA, Blank C, Zimmer L, et al. mutational burden is a determinant of immune-mediated survival in breast Genomic correlates of response to CTLA-4 blockade in metastatic cancer. Oncoimmunology. 2018;7(10):1–12. https://doi.org/10.1080/ melanoma. Science. 2015;350(6257):207–11. 2162402X.2018.1490854. 12. Johnson DB, Frampton GM, Rioth MJ, Yusko E, Xu Y, Guo X, et al. Targeted 33. Jayaraman SS, Rayhan DJ, Hazany S, Kolodney MS. Mutational landscape of Next Generation Sequencing Identi fi es Markers of Response to PD-1 basal cell carcinomas by whole-exome sequencing. J Invest Dermatol. 2014. Blockade. 2016;959–968. https://doi.org/10.1038/jid.2013.276. 13. Eroglu Z, Zaretsky JM, Hu-Lieskovan S, Kim DW, Algazi A, Johnson DB, et al. 34. Garofalo A, Sholl L, Reardon B, Taylor-Weiner A, Amin-Mansour A, Miao D, et High response rate to PD-1 blockade in desmoplastic melanomas. Nature. al. The impact of tumor profiling approaches and genomic data strategies 2018;553(7688):347–50. for cancer precision medicine. Genome Med. 2016. https://doi.org/10.1186/ 14. Snyder A, Makarov V, Merghoub T, Yuan J, Zaretsky JM, Desrichard A, et al. s13073-016-0333-9. Genetic Basis for Clinical Response to CTLA-4 Blockade in Melanoma. 2014; 2189–2199. 35. Buchhalter I, Rempel E, Endris V, Allgäuer M, Neumann O, Volckmar A-L, et al. 15. Powles T, Durán I, van der Heijden MS, Loriot Y, Vogelzang NJ, De Giorgi U, Size matters: dissecting key parameters for panel-based tumor mutational et al. Atezolizumab versus chemotherapy in patients with platinum-treated burden (TMB) analysis. Int J Cancer. 2019. https://doi.org/10.1002/ijc.31878. locally advanced or metastatic urothelial carcinoma (IMvigor211): a 36. Qiu P, Poehlein CH, Marton MJ, Laterza OF, Levitan D. Measuring tumor multicentre, open-label, phase 3 randomised controlled trial. Lancet. 2018. mutational burden (TMB) in plasma from mCRPC patients using two commercial https://doi.org/10.1016/S0140-6736(17)33297-X. NGS assays. Sci Rep. 2019. https://doi.org/10.1038/s41598-018-37128-y. Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 13 of 13 37. Nguyen A, Garner C, Reddy S, Sanborn J, Charles BS, Elisabeth ST, et al. 56. Murtaza M, Dawson SJ, Pogrebniak K, Rueda OM, Provenzano E, Grant J, et Three-fold overestimation of tumor mutation burden using 248 gene panel al. Multifocal clonal evolution characterized using circulating tumour DNA in versus whole exome. J Clin Oncol. 2018. https://doi.org/10.1200/JCO.2018. a case of metastatic breast cancer. Nat Commun. 2015. 36.15_suppl.12117. https://doi.org/10.1038/ncomms9760. 38. Panda A, Betigeri A, Subramanian K, Ross JS, Pavlick DC, Ali S, et al. 57. Butler TM, Johnson-Camacho K, Peto M, Wang NJ, Macey TA, Korkola JE, et Identifying a clinically applicable mutational burden threshold as a potential al. Exome sequencing of cell-free DNA from metastatic cancer patients biomarker of response to immune checkpoint therapy in solid tumors. JCO identifies clinically actionable mutations distinct from primary disease. PLoS Precis Oncol. 2017. https://doi.org/10.1200/PO.17.00146. One. 2015;10(8):1–14. 58. Klevebring D, Neiman M, Sundling S, Eriksson L, Ramqvist ED, Celebioglu F, 39. Endris V, Buchhalter I, Allgäuer M, Rempel E, Lier A, Volckmar A-L, et al. et al. Evaluation of exome sequencing to estimate tumor burden in plasma. Measurement of tumor mutational burden (TMB) in routine molecular PLoS One. 2014;9(8). diagnostics: in-silico and real-life analysis of three larger gene panels. Int J 59. Chan KC, Jiang P, Zheng YW, Liao GJ, Sun H, Wong J, et al. Cancer genome Cancer. 2019. https://doi.org/10.1002/ijc.32002. scanning in plasma: detection of tumor-associated copy number aberrations, 40. Zhang S, So AS, Kaplan S, KK M. Comprehensive evaluation of Illumina’s single-nucleotide variants, and tumoral heterogeneity by massively parallel TruSight® tumor 170 panel to estimate tumor mutational burden. Cancer sequencing. Clin Chem. 2013. https://doi.org/10.1373/clinchem.2012.196014. Res. 2017. https://doi.org/10.1158/1538-7445.AM2017-5358. 60. Koeppel F, Blanchard S, Marcaillou C, Martin E, Rouleau E, Solary E, et al. 41. Chaudhary R, Quagliata L, Martin JP, Alborelli I, Cyanam D, Mittal V, et al. A Whole exome sequencing for determination of tumor mutation load in scalable solution for tumor mutational burden from formalin- fixed , liquid biopsy from advanced cancer patients. PLoS One. 2017. https://doi. paraffin-embedded samples using the Oncomine tumor mutation load org/10.1371/journal.pone.0188174. assay. TLCR. 2018:1–15. 61. Gandara DR, Paul SM, Kowanetz M, Schleifman E, Zou W, Li Y, et al. Blood- 42. Rizvi H, Sanchez-Vega F, La K, Chatila W, Jonsson P, Halpenny D, et al. based tumor mutational burden as a predictor of clinical benefit in non- Molecular determinants of response to anti-programmed cell death (PD)-1 small-cell lung cancer patients treated with atezolizumab. Nat Med. 2018; and anti-programmed death-ligand 1 (PD-L1) blockade in patients with 24(September):1441. non-small-cell lung cancer profiled with targeted next-generation 62. Quinn K, Helman E, Nance T, Artieri C, Yen J, Zhao J, et al. Development sequencing. J Clin Oncol. 2018;36(7):633–41. and analytical validation of a plasma-based tumor mutational burden (TMB) 43. Samstein RM, Lee C, Shoushtari AN, Hellmann MD, Shen R, Janjigian YY, et al. score from next-generation sequencing panels. Ann Oncol. 2018; doi: 0. Tumor mutational load predicts survival after immunotherapy across multiple 1093/annonc/mdy269. cancer types. Nat Genet. 2019. https://doi.org/10.1038/s41588-018-0312-8. 63. Lyu G, Yeh Y, Yeh Y, Wang Y. Mutation load estimation model as a 44. Khagi Y, Goodman AM, Daniels GA, Patel SP, Sacco AG, Randall JM, et al. predictor of the response to cancer immunotherapy. npj Genomic Med. Hypermutated Circulating Tumor DNA : Correlation with Response to 2018. https://doi.org/10.1038/s41525-018-0051-x. Checkpoint Inhibitor – Based Immunotherapy. 2017;5729–5737. 64. Chan TA, Yarchoan M, Jaffee E, Swanton C, Quezada SA, Stenzinger A, et al. 45. Hellmann MD, Callahan MK, Awad MM, Calvo E, Ascierto PA, Atmaca A, et Development of tumor mutation burden as an immunotherapy biomarker: al. Tumor mutational burden and efficacy of Nivolumab monotherapy and utility for the oncology clinic. Ann Oncol. 2018. https://doi.org/10.1093/ in combination with Ipilimumab in small-cell lung Cancer. Cancer Cell. 2018. annonc/mdy495/5160130%0A. https://doi.org/10.1016/j.ccell.2018.04.001. 65. McGranahan N, Furness AJS, Rosenthal R, Ramskov S, Lyngaa R, Saini SK, et 46. Qiu P, Pang L, Arreaza G, Maguire M, Chang KCN, Marton MJ, et al. Data al. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to interoperability of whole exome sequencing ( WES ) based mutational immune checkpoint bloackade. Science (80- ). 2016;351(6280):1463–70. burden estimates from different laboratories. Int J Mol Sci. 2016. https://doi. 66. Kazmi SM. A retrospective analysis to evaluate prevalence and correlation org/10.3390/ijms17050651. between PD-L1 score and tumor mutational burden (TMB) levels in patients 47. Mola N, Schu M, Stiegelmeyer S, Jones W, Weigman V. Tumor mutational with solid tumor malignancies. Ann Oncol. 2017;28(Suppl 11):xi6–xi29. burden: guidelines for derivation and robustness of measurement. Cancer 67. Salem ME, Puccini A, Grothey A, Raghavan D, Goldberg RM, Xiu J, et al. Res. 2018. https://doi.org/10.1158/1538-7445.AM2018-2250. Landscape of Tumor Mutation Load , Mismatch Repair Deficiency, and PD-L1 48. Zook JM, Chapman B, Wang J, Mittelman D, Hofmann O, Hide W, et al. Expression in a Large Patient Cohort of Gastrointestinal Cancers. 2018;805–813. Integrating human sequence data sets provides a resource of benchmark 68. Morrison C, Pabla S, Conroy JM, Nesline MK, Glenn ST, Dressman D, et al. SNP and indel genotype calls. Nat Biotechnol. 2014;32(3):246–51. Predicting response to checkpoint inhibitors in melanoma beyond PD-L1 49. Wang Z, Dua J, Cai S, Han M, Dong H, Zhao J, et al. Assessment of blood and mutational burden. J Immunother Cancer. 2018;6(1):1–12. tumor mutational burden as a potential biomarker for immunotherapy in 69. Xiao W, Du N, Huang T, Guo J, Mo X, Yuan T, et al. TP53 mutation as potential patients with non–small cell lung cancer with use of a next-generation negative predictor for response of anti-CTLA-4 therapy in metastatic sequencing cancer gene panel. JAMA Oncol. 2019:1–7. melanoma. EBioMedicine. 2018. https://doi.org/10.1016/j.ebiom.2018.05.019. 50. Hellmann MD, Ciuleanu T-E, Pluzanski A, Lee JS, Otterson GA, Audigier- 70. Ock C-Y, Hwang J-E, Keam B, Kim S-B, Shim J-J, Jang H-J, et al. Genomic Valette C, et al. Nivolumab plus Ipilimumab in lung Cancer with a high landscape associated with potential response to anti-CTLA-4 treatment in tumor mutational burden. N Engl J Med. 2018. https://doi.org/10.1056/ cancers. Nat Commun. 2017. https://doi.org/10.1038/s41467-017-01018-0. NEJMoa1801946. 71. Kirchhoff T, Ferguson R, Simpson D, Kazlow E, Martinez C, Vogelsang M, et 51. Ramalingam SS. Tumor mutational burden (TMB) as a biomarker for clinical al. Germline determinants of immune related adverse events (irAEs) in benefit from dual immune checkpoint blockade with nivolumab (nivo) + melanoma immunotherapy response. Ann Oncol. 2017. https://doi.org/10. ipilimumab (ipi) in first-line (1L) non-small cell lung cancer (NSCLC): 1093/annonc/mdx376. identification of TMB cutoff from CheckMate 568. Cancer Res. 2018. https:// 72. Hasan Ali O, Berner F, Bomze D, Fässler M, Diem S, Cozzio A, et al. Human doi.org/10.1158/1538-7445.AM2018-CT078. leukocyte antigen variation is associated with adverse events of checkpoint 52. Chen Y, Zhang Y, Lv J, Li Y, Wang Y, He Q, et al. Genomic analysis of tumor inhibitors. Eur J Cancer. 2019;107:8–14. microenvironment immune types across 14 solid Cancer types : 73. Zehir A, Benayed R, Shah RH, Syed A, Middha S, Kim HR, et al. Mutational immunotherapeutic implications. Theranostics. 2017;7(14). landscape of metastatic cancer revealed from prospective clinical sequencing 53. Fabrizio D, Chen S-J, Xie M, Chen W, Quinn KJ, Zhao C, et al. In silico of 10,000 patients. Nat Med. 2017. https://doi.org/10.1038/nm.4333. assessment of variation in TMB quantification across diagnostic platforms: 74. Sun JX, He Y, Sanford E, Montesion M, Frampton GM, Vignot S, et al. A phase 1 of the friends of Cancer research harmonization project. J computational approach to distinguish somatic vs. germline origin of Immunother Cancer. 2018. https://doi.org/10.1186/s40425-018-0434-7. genomic alterations from deep sequencing of cancer specimens without a 54. Stenzinger A, Allen J, Maas J, Stewart M, Merino D, Dietel M. Tumor matched normal. PLoS Comput Biol. 2018. https://doi.org/10.1371/journal. mutational burden (TMB) standardization initiative: establishing a consistent pcbi.1005965. methodology for TMB measurement in clinical samples. Ann Oncol. 2018. https://doi.org/10.1093/annonc/mdy269.139. 55. Murtaza M, Dawson SJ, Tsui DWY, Gale D, Forshew T, Piskorz AM, et al. Non- Publisher’sNote invasive analysis of acquired resistance to cancer therapy by sequencing of Springer Nature remains neutral with regard to jurisdictional claims in plasma DNA. Nature. 2013. https://doi.org/10.1038/nature12065. published maps and institutional affiliations. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal for ImmunoTherapy of Cancer Springer Journals

Tumor mutational burden quantification from targeted gene panels: major advancements and challenges

Download PDF
13 pages

Loading next page...
 
/lp/springer-journals/tumor-mutational-burden-quantification-from-targeted-gene-panels-major-k6twZidDyX

References (90)

Publisher
Springer Journals
Copyright
Copyright © 2019 by The Author(s).
Subject
Medicine & Public Health; Oncology; Immunology
eISSN
2051-1426
DOI
10.1186/s40425-019-0647-4
Publisher site
See Article on Publisher Site

Abstract

Tumor mutational burden (TMB), the total number of somatic coding mutations in a tumor, is emerging as a promising biomarker for immunotherapy response in cancer patients. TMB can be quantitated by a number of NGS-based sequencing technologies. Whole Exome Sequencing (WES) allows comprehensive measurement of TMB and is considered the gold standard. However, to date WES remains confined to research settings, due to high cost of the large genomic space sequenced. In the clinical setting, instead, targeted enrichment panels (gene panels) of various genomic sizes are emerging as the routine technology for TMB assessment. This stimulated the development of various methods for panel-based TMB quantification, and prompted the multiplication of studies assessing whether TMB can be confidently estimated from the smaller genomic space sampled by gene panels. In this review, we inventory the collection of available gene panels tested for this purpose, illustrating their technical specifications and describing their accuracy and clinical value in TMB assessment. Moreover, we highlight how various experimental, platform-related or methodological variables, as well as bioinformatic pipelines, influence panel-based TMB quantification. The lack of harmonization in panel-based TMB quantification, of adequate methods to convert TMB estimates across different panels and of robust predictive cutoffs, currently represents one of the main limitations to adopt TMB as a biomarker in clinical practice. This overview on the heterogeneous landscape of panel-based TMB quantification aims at providing a context to discuss common standards and illustrates the strong need of further validation and consolidation studies for the clinical interpretation of panel-based TMB values. Keywords: Tumor mutational burden, TMB, Gene panels, Targeted enrichment sequencing, Immunotherapy Tumor mutational burden: an emerging strategies to adequately select those patients most likely biomarker for cancer immunotherapy to show a favorable response is recognized as an urgent Immunotherapy with immune checkpoint inhibitors medical need. A few potential biomarkers have been targeting cytotoxic T lymphocyte associated 4 (CTLA-4) identified up to now, such as PD-L1 gene expression, or programmed cell death 1 (PD-1) or its ligand (PD-L1) microsatellite instability (MSI), mismatch repair defi- can provide important clinical benefit to patients ciency (dMMR), POLE or JAK1/2 mutations, immune affected by multiple cancers, most notably lung cancer cell infiltration, IFNγ expression, tumor mutational [1, 2], melanoma [3], renal cancer [4] and urothelial burden (TMB) or neoantigen burden [6, 7]. carcinoma [5]. However, only a fraction of patients cur- TMB is a measure of the total amount of somatic cod- rently treated by immune checkpoint inhibitors derive ing mutations in a tumor and it is currently investigated benefit from it, while a minority of them suffers from as a potential biomarker in non-small cell lung carcinoma severe side effects. Given the significant cost and non- (NSCLC) [8–10]. Accumulating evidence, however, sug- negligible toxicity of these therapies, the identification of gests its potential usefulness also in melanoma [8, 11–14], urothelial cancer [5, 15, 16], mismatch-repair deficient colorectal tumors [17] and other cancer types [18]. Its * Correspondence: laura.fancello@ieo.it; luca.mazzarella@ieo.it pattern and distribution is highly variable across different Department of Experimental Oncology, IEO, European Institute of Oncology IRCCS, Via Adamello 16, 20139 Milan, Italy cancer types, with over 1000-fold difference between Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 2 of 13 cancer types with the lowest mutational burden and those load based only on somatic nonsynonymous coding muta- with the highest mutational burden, such as those associ- tions, called by Whole Exome Sequencing (WES). Briefly, ated with DNA environmental damage, i.e. by exposure to somatic nonsynonymous coding mutations are identified tobacco smoke or UVs [19, 20]. Increased TMB was also by WES and, if RNA sequencing is also available, only mu- observed in tumors with defects in DNA mismatch repair tations occurring in expressed genes are retained. Peptides and DNA replication or in tumors characterized by micro- containing selected mutations are then identified in silico satellite instability, as in colorectal cancer [21, 22]. Highly and the efficiency of their presentation to the immune mutated tumors are more likely to produce abundance of system may be evaluated by mass spectrometry or by algo- tumor-specific mutant epitopes, which may function as rithms that consider their predicted affinity to the MHC neoantigens recognized as non-self by the immune sys- class I complex and patient-specific HLA class I alleles tem. Therefore, increased activation of immune cells by [14, 25]. In comparison with overall neoantigen load, treatment with immune checkpoint inhibitors may lead to TMB is easier to measure and correlates with it. Although improved immune-mediated tumor-cell clearance and not all mutations can give rise to tumor immunogenic clinical response in these tumors (Fig. 1). A significant as- peptides, their number influences the amount of neoanti- sociation between neoantigen production and immune- gens potentially produced. High TMB correlates with mediated clinical response was indeed observed in several long-term clinical benefit from immune checkpoint inhib- studies [9, 11, 14, 23]. Measurement of this neoantigen itors in patients with melanoma [14], NSCLC [9, 26–28] production, though, is expensive and time-consuming. and urothelial carcinoma [5, 15, 16, 29]. In addition to Tumor neoantigens can be generated by mutations or by that, patients with mismatch repair (MMR) deficient tu- gene fusions, especially out-of-frame fusions. Although mors are more responsive to immunotherapy, probably some pipelines have recently been developed for the iden- due to their high tumor mutational burden [17]. There- tification of neoantigens derived from gene fusions [24], fore, although not always capable to explain the clinical most research up to now has estimated overall neoantigen benefit alone, TMB is a good approximation for neoantigen Fig. 1 Tumor mutational burden as immunotherapy biomarker. Interaction between tumor mutational burden, neoantigen production and immune checkpoints. Hyper-mutated tumors (bottom) are more likely than hypo-mutated tumors (top) to generate tumor-specific peptides (neoantigens) recognized by the immune system. However, immune surveillance can be restrained by simultaneous high expression of PD-L1, which delivers a suppressive signal to T cells. PD-L1/PD-1 interaction and other immune checkpoints can be inhibited by immune checkpoint inhibitors, restoring immune response Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 3 of 13 load assessment [14], is technically less challenging and less panel-based and WES-based TMB quantification, based expensive and may represent a better suited predictive on matched sequencing by gene panel and WES of the biomarker for immunotherapy response. same tumor sample and comparison of matched TMB TMB may also represent a relevant prognostic bio- values (Additional file 3: Table S3, Additional file 9:Figure marker. In BRCA-1/2 mutated ovarian cancers, TMB S3). Accuracy of panel-based TMB quantification is influ- correlates with improved overall survival [30, 31]. In enced by statistical sampling effects and small panels breast cancer patients, tumors with high TMB and favor- provide less precise TMB estimates [22, 34–36]. It was able immune-infiltrate (“hot tumors”) are associated demonstrated that TMB values from the FoundationOne with prolonged survival [32]. Consistently, basal cell gene panel, which targets 1.1 Mb of genomic space, are carcinoma, which is characterized by very high TMB, similar to those from WES, whereas accuracy drops im- presents with slow growth rates and rare metastases. Al- portantly when sequencing less than 0.5 Mb [22]. Another though not definitively demonstrated, we can speculate study simulated sequencing of theoretical gene panels of that this less aggressive phenotype may be due to hyper- different sizes and identified 1.5 to 3 Mb as the best suited mutation, which would trigger, via increased neoantigen targeted genomic size to confidently estimate TMB [35]. production, a more effective immune response of the Moreover, the deviation between WES- and panel-based host [33]. TMB appears more relevant for samples with low to moderate underlying TMB levels, compared to samples Quantification of tumor mutational burden from with high TMB [22, 35, 36]. Another retrospective study gene panels: “yes we can” on a commercial panel of 248 genes likewise cautions Initial studies showing a correlation between TMB and against small gene panels which would lead to TMB enhanced response to immunotherapy were based on overestimation [37]. WES datasets for TMB quantification [9, 14, 17]. WES Besides the accuracy of panel-based TMB quantifica- allows a direct measurement of TMB, yet it remains tion, it is critical to assess its capability to discriminate unsuitable as routine technology in clinical practice, be- between immunotherapy responders and non- cause expensive, labor-intensive and time-consuming. responders, as previously observed for WES-based TMB. Therefore, several studies explored the possibility to pro- Several exploratory analyses demonstrated that panel- vide equally accurate and clinically predictive TMB based TMB, as simulated in silico by downsampling a estimates from targeted enrichment sequencing, using WES dataset to only include genes targeted by the Foun- various gene panels (Table 1, Additional file 1: Table S1). dationOne gene panel, associates with immunotherapy The main challenge for accurate panel-based TMB response [8, 26] or with signatures of immune check- quantification is the ability to extrapolate the global point activation [38]. Comparable results were observed mutational burden from the narrow sequencing space in similar in silico analyses for other gene panels, such targeted by a gene panel. In silico analyses were per- as the Trusight170 [39, 40] or MSK-IMPACT [26] formed to test the concordance between panel-based (Additional file 4: Table S4). Notably, direct measurement and WES-based TMB, which is considered the reference of TMB from the Oncomine Tumor Load Assay shows for TMB quantification. Publicly available WES datasets that this panel-based TMB value allows to classify colorec- were downsampled to the subset of genes targeted in the tal cancer cases based on their MSI status [39, 41]. Since panel under consideration and TMB values from such in this cancer type MSI positively correlates with simulated gene panels were compared with TMB values immunotherapy response, this is a further, yet indirect from the original WES (Additional file 7:Figure S1),find- evidence, of the capability to predict immunotherapy re- ing high correlation between the two (Additional file 2: sponse, using a panel-based TMB estimate. Most import- Table S2, Additional file 8:Figure S2).Mostofthese in antly, a few clinical studies demonstrated that TMB silico analyses were performed using publicly available directly estimated using gene panels is higher in those WES datasets from TCGA, with the exception of the patients who benefit more from immune checkpoint Oncomine Tumor Mutation Load Assay or NovoPM and blockade treatment, thus providing “real-life” evi- CANCERPLEX gene panels, for which WES datasets from dence for its potential clinical predictive value (Fig. 2, COSMIC or from other sources were used. Regardless, Additional file 5: Table S5). A direct association with similar correlation values were reported for the differ- immunotherapy response was shown for the MSK- ent gene panels tested (Additional file 2: Table S2, IMPACT [42, 43] and the Guardant360 gene panels [44] Additional file 8: Figure S2). For some of these gene but most of the reported studies utilized the Foundatio- panels (FoundationOne, Trusight170, Oncomine Tumor nOne gene panel (Fig. 2, Additional file 5: Table S5). In Mutation Load Assay, Oncomine Comprehensive Assay particular, in the CheckMate 227 trial, NSCLC patients V3 and MSK-IMPACT gene panels), an empirical ap- with high TMB (> 10 mutations per Mb, measured by proach was also used to test the concordance between FoundationOne) presented increased progression-free Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 4 of 13 Table 1 Overview of the main published studies on TMB quantification from gene panels Reference Gene panel Cancer type Study design Study ID ICI TMB cutoff Method of TMB TMB predictive Clinical N patients (version) (mut/Mb) cutoff value outcome determination Rosenberg, FM1 (v3) urothelial carcinoma trial (single-arm, NCT02108652 PD-(L)1 NA NA NA ORR 315 2016 [5] (metastatic or locally phase 2) advanced) Balar, 2017 FM1± urothelial carcinoma trial (single-arm, NCT02108652 PD-(L)1 Q3 (> = 16) distribution NA OS 123 [16] (metastatic) phase 2) Powles, 2018 FM1± urothelial carcinoma trial (randomized, NCT02302807 PD-(L)1 Q2 (9.65) distribution NA OS 931 [15] (metastatic) phase 3) Kowanetz, FM1 (v3) NSCLC trial (randomized, NCT01903993 PD-(L)1 Q1, Q2 (9.9), distribution NA PFS, OS, ORR 454 2016 [27] phase 2) Q3 trial (single-arm, NCT02031458 phase 2) trial (single-arm, NCT01846416 phase 2) Gandara, 2018 FM1 bTMB NSCLC trial (randomized, NCT01903993 PD-(L)1 > = 14 positive and NA PFS, OS 259 [61] assay phase 2) negative percentage trial (randomized, NCT02008227 agreement with phase 3) the orthogonally validated FM1 Hellmann, FM1 CDx NSCLC trial (randomized, NCT02477826 combo > 10 based on NA PFS 1004 2018 [50] phase 3) NCT02659059 Rizvi, 2018 [42] MSK-IMPACT NSCLC trial (randomized, NCT01295827 PD-(L)1 Q2 (7.4) distribution AUC = 0.601 DCB, PFS 240 (v1, v2, v3) phase 1) (DCB) Ready, 2019 FM1 CDx NSCLC trial (non-randomized, NCT02659059 combo 10 ROC AUC (95% ORR 98 [28] phase 2) CI) = 0.73 (0.62–0.84); TPR (95% CI) = 0.78 (0.63–0.93); FPR (95% CI) = (0.62 (0.49–0.73) Wang, 2019 NCC-GP150 NSCLC observational (cohort) NA PD-(L)1 6 (tot mut) best cutoff from NA PFS, ORR 50 [49] in silico analysis on Rizvi 2015 WES Johnson, 2016 FM1 (v2, v3) melanoma observational NA PD-(L)1 < 3.3, 3.3–23.1, ROC NA PFS, OS, ORR 65 [12] (retrospective) >23.1 Chalmers, FM1 (v1, v2, various locally advanced observational NA NA > 20 NA NA NA 102, 292 2017 [22] v3, v4), FM1 or metastatic solid tumors (retrospective) Heme Goodman, FM1 (v1, v2, various locally advanced observational (cohort, NCT02478931 PD-(L)1, < 6, 6–19, > 19 Foundation NA PFS, OS, ORR 151 Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 5 of 13 Table 1 Overview of the main published studies on TMB quantification from gene panels (Continued) Reference Gene panel Cancer type Study design Study ID ICI TMB cutoff Method of TMB TMB predictive Clinical N patients (version) (mut/Mb) cutoff value outcome determination 2017 [18] v3) or metastatic solid tumors retrospective) CTLA-4, Medicine official high-dose reports IL2 or combo Khagi, 2017 Guardant360 various solid tumors observational (cohort, NCT02478931 PD-(L)1, mean (> 3 distribution NA PFS, OS, ORR 69 [44] retrospective) CTLA-4, VUS) combo or other Zehir, 2017 MSK-IMPACT various primary and observational (cohort, NCT01775072 NA > 13.8 distribution NA NA 10, 945 [73] (v1, v2) metastatic solid tumors prospective) (median TMB + 2 × IQR_TMB) Samstein 2019 MSK-IMPACT bladder observational (cohort, NCT01775072 PD-(L)1, 17.6 distribution (top NA OS, PFS, DCB 214 [43] (v3) prospective) CTLA-4 or 20%) breast 5.9 45 combo breast ER+ 6.8 24 breast ER- 4.4 21 unknown primary 14.2 90 colorectal 52.2 110 esophagogastric 8.8 126 glioma 5.9 117 head and neck 10.3 138 melanoma 30.7 321 NSCLC 13.8 350 renal cell carcinoma 5.9 151 ORR Objective Response Rates, DCB Durable Clinical Benefit, OS Overall Survival, PFS Progression-Free Survival, FM1 Foundation Medicine’s FoundationOne (v1: 185 genes, v2: 236 genes, v3: 315 genes, v4: 405 genes, Heme: 405 genes, CDx: 324 genes); ±: version not specified; MSK-IMPACT v1 341 genes, v2: 410 genes, v3 468 genes, NSCLC non-small cell lung cancer, ER Estrogen Receptor, VUS variants of unknown significance, PD- (L)1 anti-PD-1 or anti-PD-L1, CTLA-4 anti-CTLA-4, combo combined anti-PD-1/PD-L1 + anti-CTLA-4, Q1-Q4 quartiles, : TMB quantification from blood Each study is described reporting gene panel, cancer type, study design, study ID (on ClinicalTrials.gov), immune checkpoint inhibitor treatment (ICI), proposed TMB cutoff, method for TMB cutoff determination, outcome analyzed to evaluate TMB clinical utility. AUC, TPR (True Positive Rate) and FPR (False Positive Rate) are provided, when available, as a measure of TMB predictive value for immunotherapy responder classification Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 6 of 13 Fig. 2 TMB association with progression-free survival. Forest plot of hazard ratios (HR) comparing progression-free survival (PFS) between patients with high or low TMB, as indicated in the “Comparison” column. If not specified otherwise, TMB is reported as number of mutations per Mb. All patients were treated with immune checkpoint inhibitors (ICI). Bars represent the 95% confidence intervals. Size of the box is proportional to precision. Reference to the study and the analyzed cancer type are also reported together with the log-rank p-value. Q1-Q4: quartiles; VUS: variants of unknown significance. *: TMB quantified from blood; **: Cox proportional hazards model adjusted for age, gender, disease stage and prior therapy by ipilimumab survival after combined anti-CTLA-4 plus anti-PD-1/ and by the capability to extrapolate TMB values from PD-L1 therapy [45]. Interestingly, TMB was predictive the restricted genomic space sampled by gene panels to of anti-PD-L1 monotherapy response in NSCLC the whole genome (Fig. 3a). Experimental factors (e.g. (POPLAR trial, [27]) and metastatic urothelial carcin- tumor purity or sequencing depth) and the variant oma patients [5, 15, 16], independently from the PD-L1 calling pipeline (e.g. the variant calling algorithm and expression status. Analysis of archival tumor samples from the method to remove germline variants) can signifi- melanoma patients treated by anti-PD-1/PD-L1 mono- cantly affect the number of called somatic mutations therapy also showed superior response rates, progression- and have a similar impact on both panel-based and free survival and overall survival in high TMB cases [12]. WES-based TMB quantification. Indeed, the adoption of Moreover, a retrospective study on 151 patients across a well-documented standard pipeline was already diverse tumor types showed that cancer patients with claimed for WES analyses as an urgent need to allow higher TMB, benefit more from anti-PD-1/PD-L1, anti- data interoperability between different platforms [46]. CTLA-4 or high dose IL2 monotherapy [18]. The same The same applies to panel sequencing for TMB quantifi- was not observed for combined anti-PD-1/PD-L1 plus cation. In this context, an important factor investigated anti-CTLA-4 therapy but the available number of samples for its influence on the number of called somatic may be too small to draw conclusions [18]. variants is the method chosen to identify and remove germline variants. Indeed, since only somatic mutations Need for standardization of TMB quantification can potentially produce tumor neoantigens recognized and reporting as non-self by the immune system, it is important to Despite the increasing number of studies showing the remove germline variants in TMB quantification. It was potential clinical relevance of panel-based TMB as a pre- observed that the use of an in silico method for somatic dictive biomarker for immunotherapy response, its use variant calling instead of matched tumor-normal sam- in the clinical setting is currently limited by the absence ples, leads to increased false positive somatic variants, of standard methods of quantification and the lack of a which has an important influence on the accuracy of robust and universal cutoff to identify immunotherapy TMB quantification, especially for small gene panels responders. [34]. To avoid this, it was proposed to perform TMB Panel-based TMB quantification is influenced by vari- quantification using only high-confidence regions [47] ous experimental factors affecting library construction (e.g. regions of the genome, devoid of potential system- and sequencing, by the pipeline used to call mutations atic biases or structural variants, where mutations can Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 7 of 13 Fig. 3 Differences in the workflow for panel-based TMB quantification. a. Overview of the factors influencing panel-based TMB quantification. Several variables in library construction, sequencing and in the pipeline to call mutations influence panel-based TMB quantification. Furthermore, panel-based TMB quantification is influenced by differences in the bioinformatic method to extrapolate global TMB from mutations identified in the narrow genomic region targeted by the gene panel. b. Differences across various studies in panel-based TMB quantification: gene panel technical specifications, preanalytical factors and the bioinformatics workflow used to extrapolate from the genomic space targeted by gene panels global TMB are described. FM1: Foundation Medicine’s FoundationOne panel (v1: 185 genes, v2: 236 genes, v3: 315 genes, v4: 405 genes); NA: not available; ±: algorithm developed by Sun et al. for in silico removal of germline variants [74] be confidently called), as defined by Zook et al. [48]. It mutations. Therefore, it was proposed to remove known was also observed that increased somatic false positives cancer variants of targeted genes when performing TMB are generated by the in silico germline filtering method quantification, to avoid overestimation of TMB when for patients with non-caucasian ancestry compared to extrapolating it across the whole genome [12, 22]. Buch- caucasian patients, as the former are less represented in halter et al., showed that removal of cancer mutational public databases used for germline variant filtering [34]. hotspots slightly decreases the number of high TMB tu- The use of ExAC, the largest and more representative mors identified but does not change the general picture public germline WES database, in addition to dbSNP [35]. However, the importance of this filtering, routinely and 1000 Genomes, is recommended to reduce this performed only for Foundation Medicine panels, de- difference [34]. pends on panel size and composition: some gene panels In addition to these factors, which similarly influence may be larger and less enriched in cancer genes by WES- and panel-based analyses, panel-based TMB including, for example, pharmacogenomic variants. As quantification also requires to extrapolate the number of for synonymous mutations, it is claimed that, although somatic coding mutations observed in the targeted not biologically meaningful, their inclusion may reduce genomic space to the number that would be observed sampling noise and improve the approximation of TMB across the whole genome. Extrapolation methods may across the whole genome. Indeed, several works com- differ for various choices in variant filtering, such as pared TMB quantification with or without synonymous removal of known cancer mutations or synonymous mu- variants and observed that, when including synonymous tations (Fig. 3b). Standard gene panels are commonly variants, panel-based TMB shows increased correlation enriched in known cancer genes, which are more likely with WES-based TMB values [35, 49] and stronger asso- to be mutated in a tumor and expectedly enriched in ciation with clinical response [9]. Starting from the Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 8 of 13 above observations, we can infer that some recommen- most commonly established using distribution-based strati- dations to build a standardized and robust analysis pipe- fication, which can be heavily influenced by outliers, while line for TMB quantification are starting to emerge at it is now often identified based on statistically sound least for the following points: i. germline variants can be methods, such as ROC curves. In the future, the use of most efficiently identified and removed by matched nor- ROC curves as a common method of TMB cutoff deter- mal sequencing and, if this is not possible, the largest mination will greatly help to converge to a robust TMB cut- available germline variant databases should be used for in off and will facilitate comparison across different platforms. silico filtering, especially for non-caucasian ancestries; ii. The heterogeneity in experimental and analytical pro- TMB extrapolation to the whole genome is accurately per- tocols, in the extrapolation of panel-based TMB values formed by counting all somatic mutations, including syn- and in gene panel technical specifications is currently onymous mutations, to enlarge the sampling space and limiting the potential use of TMB in a clinical setting better approximate global TMB across the whole genome. (Fig. 3). For this reason, a common standard for TMB As already mentioned, panel-based TMB quantifica- quantification and a consensus on a clinically useful tion is affected by the genomic size targeted by the panel TMB cutoff are urgently needed. Some efforts in this and by its gene composition. Notably, gene panels tested sense are ongoing by the Quality Assurance Initiative up to now widely differ for number of targeted genes Pathology (QuIP) in Germany (https://quip.eu) and by (from 73 to 710) and size (from 0.39 to 2.8 Mb of tar- the European Society of Pathology. Moreover, in the US, geted genomic space) (Additional file 1: Table S1). These governmental organizations, health-sector industries, considerations raise the question of how to convert several NGS gene panel manufacturers and academic in- TMB estimates between different gene panels to allow stitutions set up a TMB Harmonization Working Group cross-platform comparability. Indeed, although the ma- (https://www.focr.org/tmb) and planned a 3-phase pro- jority of studies correlating TMB to immunotherapy re- ject for TMB harmonization. In the first phase, they per- sponse are currently based on FoundationOne, several formed in silico analyses of publicly available TCGA data other types of gene panels exist and the offer is steadily to identify sources of variability in TMB quantification increasing (Additional file 1: Table S1). Moreover, we between WES and gene panels. Recently concluded, this still need to standardize the conversion of the reference work established that panel-based TMB is comparable WES-based TMB values to panel-based TMB, since the between different gene panels for TMB values ranging 0 lower sequencing coverage and higher sequencing depth to 40 mutations per Mb, that it strongly correlates with of gene panels, as compared to WES, may lead to de- WES-based TMB and it is possibly influenced by the creased accuracy of TMB values and increased sensitivity type of cancer under investigation. It also found that the in variant calling. For cross-panels or panel-to-WES observed variance across gene panels stems from their TMB conversion, an in silico approach was proposed, different gene composition and technical specifications, where TMB distributions derived from two different as well as from the bioinformatic pipeline adopted [53]. technologies were interpolated and aligned and TMB The second step of the project empirically validates cutoffs were mapped across distributions [38]. However, TMB estimates from different gene panels by mapping a consensus on a standard method to convert TMB them to WES-based TMB values, used as a gold stand- values is still missing. ard, whereas the last step will define best practices for Not surprisingly, in this heterogeneous landscape a ro- TMB use as immunotherapy biomarker in clinical rou- bust TMB cutoff to discriminate between immunother- tine. Following preliminary results on the influence of apy responders and non-responders is still to be defined. the bioinformatic workflow and of gene panel size and Moreover, the adopted cutoffs sometimes differ across composition on TMB quantification, the working group different studies on the same gene panel (Table 1). Up to recommends the use of gene panels larger than 1 Mb now, the TMB cutoff of 10 mutations per Mb, measured and the standardization of the bioinformatic algorithms, by the FoundationOne gene panel and found to best in addition to standardization of sample processing. discriminate between responders and non-responders to Moreover, it suggests the inclusion of actionable genes, immunotherapy in NSCLC patients, is the only one genes associated with mutagenesis and negative predic- which has been validated in a separate further study tors of response in these gene panels and the alignment [28, 50, 51]; this cutoff was also observed, but not yet of panel-based TMB values to WES-based ones to allow validated, in melanoma [38] and in metastatic urothelial interoperability across different assays [54]. carcinoma [15](Table 1). Interestingly, these cancer types present a TMB distribution similar to that of NSCLC [52]. TMB quantification beyond tissue biopsies and Indeed, due to the diversity of TMB distribution across dif- current gene panels ferent cancer types, the adoption of cancer-specific TMB Most studies on TMB as a predictive biomarker for im- cutoffs was proposed [35, 43]. TMB cutoff was initially munotherapy response were performed on bioptical or Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 9 of 13 surgical specimens from solid tumors. Since obtaining main pitfalls of the panel, though, is its limit of detection, tissue biopsies may be challenging and invasive for defined as a minimum of 1% tumor content in at least 20 patients, it would be critical for the clinical routine to ng of cell-free DNA input, and its dependency on the assess TMB using cell-free DNA (cfDNA) from blood, overall tumor burden, which influences the likelihood of which includes circulating tumor DNA (ctDNA), as a detecting ctDNA. The exclusive use of single nucleotide surrogate specimen to biopsy. High throughput molecu- variants (SNVs) for TMB quantification represents an- lar profiling of ctDNA remains technically challenging other limitation, although future versions of the algorithm but increasing efforts are being made in this direction. A are planned to be released, which will also use indels. The few studies previously investigated the feasibility of WES commercial Guardant360 and GuardantOMNI gene on ctDNA and highlighted some inherent limitations, panels were also designed for blood-based TMB quantifi- such as the low amount of available ctDNA, which re- cation [62]. Their limit of detection was defined as a mini- duces sensitivity, or ctDNA being more associated with mum of 0.3% tumor content in at least 5 ng of cell-free metastases rather than with primary tumors [55–59]. In DNA input. They were validated in silico by subsetting one of the largest studies attempting optimization of TCGA WES datasets to only include genes targeted by WES-based TMB quantification from liquid biopsy, the panels. Panel accuracy in TMB quantification was then WES was performed in parallel on DNA from tissue evaluated by correlation of TMB values obtained from the biopsies and on cfDNA from liquid biopsies of 32 meta- simulated gene panels with those from WES. Their pre- static patients and comparable sequencing depth and dictive value was similarly evaluated in silico on 30 lung coverage were obtained [60]. Performance of variant de- cancer samples with matched information on immuno- tection was dependent on the fraction of tumor DNA therapy response. The performance showed by TMB from within the analyzed cfDNA, as previously described. In the simulated gene panel in responder identification was those samples positive for the presence of ctDNA, vari- comparable to that of WES-based TMB (Additional file 4: ant detection sensitivity of cfDNA-WES compared to Table S4). The Guardant360 panel was further tested in a tDNA-WES was 94%, regardless of the tumor type (2 small retrospective study on 69 patients with various cholangiosarcoma and 19 lung, 5 head and neck, 2 tumor types [44]. No comparison with tissue-based TMB prostate, 2 colorectal, 1 breast and 1 bladder cancer were has been reported yet, but a significant correlation be- analyzed). Most importantly, in ctDNA positive samples, tween high blood-derived TMB measured by Guard- TMB values from WES on liquid biopsies were robust ant360 and immunotherapy response was observed [44]. and consistent with those from WES on tissue biopsies, Finally, a further gene panel for bTMB quantification was which demonstrates for the first time the feasibility of recently developed in China. Consistency between panel- TMB quantification from liquid biopsies, using WES. based bTMB values and WES-based tTMB values, tested More recently, it was demonstrated that targeted in silico and empirically by matched blood and tissue sam- enrichment sequencing by gene panels is another valid ples, was comparable to that of the panels described approach for TMB quantification from liquid biopsies. above. Similar results were also found for its predictive In particular, Gandara et al. developed, tested and analyt- value, based on in silico analyses. Interestingly, the authors ically validated a novel gene panel for TMB quantification also raised the issue of the different frequency of onco- from blood [61]. The panel is based on hybridization- genic driver mutations, such as EGFR or KRAS, between capture enrichment and targets 394 genes, corresponding Asian and white population. For this reason, they compare to 1.1 Mb of genomic space (Fig. 3). Its clinical utility was TCGA WES-based TMB with panel-based TMB with or evaluated via a retrospective study on 259 NSCLC samples without inclusion of EGFR and/or KRAS mutations. Al- from patients treated with immunotherapy or chemother- though similar results are yielded, the raised issue is an apy in the OAK and POPLAR clinical trials. Blood- important point to be further investigated in panel-based derived TMB (bTMB) calculated using this novel gene TMB quantification [49]. TMB quantification from liquid panel correlated well with tissue-derived TMB (tTMB) biopsies suffers from ctDNA detection limits, which also measured by FoundationOne. Moreover, measured TMB depend on tumor size and number of cancer cells, but was found to be significantly associated with response to these results encourage to further explore and more ex- anti-PD-L1 immunotherapy in the POPLAR trial and this tensively validate this approach. was further confirmed on patient samples from the OAK Besides new technologies to estimate TMB from liquid trial. A prospective validation is also currently ongoing in biopsies, another significant step towards routine use of the BFAST trial (NCT03178552) on advanced and meta- TMB in clinical practice is TMB quantification from an static NSCLC patients. Interestingly, it was observed that even smaller set of genes than in targeted enrichment the capability of TMB, as measured by this panel, to gene panels. Although panel size is known to affect predict anti-PD-1/PD-L1 immunotherapy response is in- accuracy of TMB quantification, the use of a highly cus- dependent from PD-L1 expression levels [61]. One of the tomized set of genes may represent a valid and even less Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 10 of 13 expensive approach. In this view, Lyu et al., proposed a et al. observed a relationship between subclonal mutations computational framework to assess the best and smallest and mutational signatures associated with alkylating agents subset of genes necessary to estimate TMB as a bio- and, in NSCLC, between clonal mutations and mutational marker for lung adenocarcinoma [63]. They were able to signatures associated with smoking [65]. Mutational signa- identify a model of only 24 genes which predicted in tures associated with smoking were also found to be sig- silico immunotherapy response with 93% specificity and nificantly associated with high tumor mutational burden 85% sensitivity and they suggested that other small and with response to immunotherapy [9]. Therefore, al- custom sequencing gene panels may be designed in a though the extraction of mutational signatures from gene cancer-specific way to assess TMB with further reduced panels may be hampered by the small number of sampled costs. mutations, these observations suggest that they may prove helpful to infer neoantigen clonality and enhance TMB Future perspectives and recommendations predictive value. TMB is one of the most rapidly developing biomarkers Integration of TMB with other potential immunother- for immunotherapy response, with about 37 ongoing apy biomarkers represents another promising way to re- clinical trials currently registered in ClinicalTrials.gov fine prediction of immunotherapy responders. For that use TMB as stratification biomarker [64]. Several example, TMB, defects in DNA mismatch-repair path- gene panels were recently optimized to estimate TMB at way and the MSI status all are measures of genomic reduced sequencing costs, and emerging evidence sup- instability that can provide indirect assessment of tumor ports the feasibility of TMB quantification from liquid antigenicity, whereas PD-L1 expression, immune cell biopsies. However, harmonization in TMB quantification infiltration and inflammatory signatures represent bio- and reporting remains the main challenge for the near markers of the T cell-inflamed tumor microenvironment. future: standard procedures are required to allow inter- Therefore, their integration can refine prediction of im- operability between different gene panels, compare munotherapy outcome by combining information on results across studies and define a universal cutoff to tumor complexity and on the immune response. Indeed, confidently identify patients most likely to benefit from emerging evidence suggests that, at least in NSCLC, immunotherapy. TMB and PD-L1 expression are independent predictors Even an accurate TMB value is an imperfect predictor and TMB may complement or even outperform PD-L1 of immunotherapy response and further studies are expression [10, 26, 50, 66]. Moreover, it was observed needed to enhance its value as clinically useful immuno- that most tumors with high MSI also present elevated therapy biomarker. TMB is used as an approximation of TMB, whereas the opposite does not hold true. The neoantigen burden, upon the assumption that the higher combination of TMB with MSI and PD-L1 expression in the mutational burden, the higher the probability for gastrointestinal tumors significantly improved the identi- immunogenic peptides to be generated, which leads to fication of immunotherapy responders [67]. In another stronger immune response upon inhibition of immune study, it was observed that TMB is an independent pre- checkpoints. Interestingly, neoantigen clonality, in addition dictor and only weakly correlates with T cell-inflamed to the overall amount of neoantigens, influences immuno- gene expression profiles (GEP) or PD-L1 expression. therapy response in NSCLC patients [65]. In particular, Thus, TMB and T cell-inflamed GEP were jointly used tumors enriched in clonal neoantigens (e.g. present in all to identify immunotherapy responders: patients with tumor cells) are more sensitive to immune checkpoint in- both high TMB and high T cell-inflamed GEP were those hibitors than tumors enriched in subclonal neoantigens with the highest objective response rates on tumors from (e.g. present only in a subset of tumor cells), in advanced four KEYNOTE clinical trials across 22 cancer types. NSCLC and melanoma patients [65]. Indeed, clonality of Similarly, in melanoma patients, a response score based produced neoantigens seems to be associated with a more on the combination of TMB, infiltration of CD8+ T cells effective immune surveillance. On the other hand, enrich- and gene expression profiles for PD-L1, CD8 and a set ment in subclonal neoantigens may activate T cells against of 394 immune genes demonstrated higher sensitivity only a subset of tumor cells, leading to less effective tumor and similar specificity than each biomarker alone [68]. control. Based on these observations, it would be interest- To date, the FoundationOne and Guardant360 gene ing to investigate if information on mutation clonality (e.g. panels allow to measure both TMB and MSI but no variant allele frequency) improves the predictive power of other potential immunotherapy biomarker. Moreover, TMB. Evaluation of mutation clonality from gene panels is they do not provide the user any combinatorial model to not trivial though: the reduced genomic space targeted by integrate them. Although further validation in prospect- gene panels may not be representative of the overall clonal ive clinical studies is required for all these potential architecture and the mutations sampled herein may not be biomarkers, several observations suggest that simultan- those generating neoantigens. Interestingly, McGranahan eous profiling of both TMB and other immunotherapy Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 11 of 13 biomarkers currently under investigation may represent Additional file 7: Figure S1. Visual representation of the method used the next step forward in the design of new gene panels for in silico analyses on TMB quantification accuracy and on association or predictive value for immunotherapy response. In silico analyses are for clinical use. The Friends and QuIP initiatives for based on simulations of panel performance, wherein TMB is calculated TMB harmonization recommended to include as much using a subset of WES which only contains genes targeted by the panel. relevant genetic and molecular information as possible Accuracy of TMB quantification from the simulated gene panel is evaluated by comparison with WES-based TMB, used as gold reference, in these panels, to avoid the need to re-biopsy the with correlation analysis. The clinical predictive value of TMB estimated patient for further information. In line with this recom- from the simulated panel is evaluated based on its association with mendation, we propose to also include in gene panels clinical values measuring immunotherapy response. (PDF 55 kb) for TMB quantification other potential immunotherapy Additional file 8: Figure S2. In silico analysis of the correlation between panel-based and WES-based TMB. Correlation between panel- biomarkers but also negative predictors of immunother- based and WES-based TMB, considered the gold standard value, is used apy response [69, 70] and variants predisposing to to estimate the accuracy of panel-based TMB quantification. Panel-based adverse reaction to immunotherapy [71, 72]. These and TMB quantification was simulated in silico using a subset of WES which only contains genes targeted by the panel. The bubble plot shows on other recommendations which emerge from the studies the x axis the correlation coefficients and on the y axis the gene panel reviewed here, including the one from the TMB and the cancer type. Bubble size represents the number of data points Harmonization Working Group, are summed up in used in the analysis and the color corresponds to the reference study. (PDF 259 kb) Additional file 6: Table S6. Additional file 9: Figure S3. Empirical analysis of the correlation between panel-based and WES-based TMB. Correlation between panel- based and WES-based TMB, considered the gold standard value, is used to estimate the accuracy of panel-based TMB quantification. Correlation Additional files analysis is performed on TMB values calculated for samples with matched panel and whole exome sequencing. The bubble plot shows on the x axis the correlation coefficients and on the y axis the gene panel and the cancer Additional file 1: Table S1. Technical specifications of gene panels type. Bubble size represents the number of data points used in the analysis used or proposed for TMB quantification. For each gene panel, it is and the color corresponds to the reference study. (PDF 155 kb) reported the type of cancer and sample for which it was designed, the enrichment method, the targeted sequencing size (Genomic space) and the number of targeted genes (# genes). (XLSX 6 kb) Abbreviations Additional file 2: Table S2. In silico analysis of the correlation between ACC: Adrenocortical carcinoma; AUC: Area under the curve; BLCA: Bladder panel-based and WES-based TMB. Correlation between panel-based and urothelial carcinoma; BRCA: Breast invasive carcinoma; CESC: Cervical WES-based TMB, considered the gold standard value, is used to estimate squamous cell carcinoma and endocervical adenocarcinoma; the accuracy of panel-based TMB quantification. Panel-based TMB quanti- cfDNA: Circulating free DNA; CHOL: Cholangiosarcoma; COADREAD: Colon fication was simulated in silico using a subset of WES which only contains adenocarcinoma; CRC: Colorectal cancer; ctDNA: Circulating tumor DNA; genes targeted by the panel. (XLSX 11 kb) DLBC: Lymphoid neoplasm diffuse large B-cell lymphoma; ESCA: Esophageal carcinoma; FDA: Food and Drug Administration; GBM: Glioblastoma; Additional file 3: Table S3. Empirical analysis of the correlation GEP: Gene expression profile; HLA: Human Leukocyte Antigen; HNSC: Head between panel-based and WES-based TMB. Correlation between panel- and neck squamous cell carcinoma; KICH: Kidney chromophobe; KIRC: Kidney based and WES-based TMB, considered the gold standard value, is used renal clear cell carcinoma; KIRP: Kidney renal papillary cell carcinoma; to estimate the accuracy of panel-based TMB quantification. Correlation LAML: Acute myeloid leukemia; LGG: Brain lower grade glioma; LIHC: Liver analysis is performed on TMB values calculated for samples with matched hepatocellular carcinoma; LUAD: Lung adenocarcinoma; LUSC: Lung panel and whole exome sequencing. (XLSX 6 kb) squamous carcinoma; Mb: Megabase; mCRPC: Metastatic castration-resistant Additional file 4: Table S4. In silico analysis of TMB association or prostate cancer; MESO: Mesothelioma; MHC: Major histocompatibility predictive value for immunotherapy response. These analyses were performed complex; MMR: Mismatch repair; MSI: Microsatellite instability; NSCLC: Non- on panel-based TMB values simulated in silico using a subset of WES which small cell lung cancer; ORR: Objective response rates; OS: Overall survival; only contains genes targeted by the panel. The table reports measures of TMB OV: Ovarian serous cystadenocarcinoma; PAAD: Pancreatic adenocarcinoma; association with immunotherapy response (odds ratios, hazard ratios and PCPG: Pheochromocytoma and paraganglioma; PFS: Progression free survival; corresponding p-values), differences in TMB distribution between responders PRAD: Prostate adenocarcinoma; ROC: Receiver operating characteristic; and non-responders (Mann-Whitney U and Fisher’s p values) and measures of SARC: Sarcoma; SCLC: Small cell lung cancer; SKCM: Skin cutaneous TMB predictive value (AUC, specificity, sensitivity). (XLSX 9 kb) melanoma; SNV: Single nucleotide variant; STAD: Stomach adenocarcinoma; Additional file 5: Table S5. Empirical analysis of TMB association or TCGA: The cancer genome atlas; TCR: T cell receptor; TGCT: Testicular germ predictive value for immunotherapy response. These analyses were cell tumors; THYM: Thymoma; TMB: Tumor mutational burden; UCEC: Uterine performed on panel-based TMB values, directly calculated by panel se- corpus endometrial carcinoma; UCS: Uterine carcinosarcoma; UVM: Uveal quencing. The table reports measures of TMB association with immuno- melanoma; WES: Whole exome sequencing therapy response (odds ratios, hazard ratios and corresponding p-values), differences in TMB distribution between responders and non-responders Acknowledgements (Mann-Whitney U, unpaired Student’s t and Fisher’s test p values) and We thank all researchers, clinicians and organizations working in this field for measures of TMB predictive value (AUC, specificity, sensitivity). We also their contributions and we apologize to those whose work we did not specify how patients were stratified (“Comparison”), the method used to report or cite. determine TMB cutoff, the cohort considered for the analysis (if different co- horts were analyzed in the study), the type of immunotherapy, cancer type Authors’ contributions and number of patients. (XLSX 19 kb) LF designed, wrote and revised the manuscript. PGP, SG and LM revised the Additional file 6: Table S6. Proposed recommendations for consistent manuscript. All authors read and approved the final manuscript. TMB quantification and reporting. We report recommendations formulated by the TMB Harmonization Working Group (https://www.focr.org/tmb)as Funding well as indications emerging from the studies reviewed in this work. LF work was supported by the Italian Ministry of Health (Ricerca Corrente, (XLSX 24 kb) “Alleanza Contro il Cancro” - ACC network). Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 12 of 13 Availability of data and materials 16. Balar AV, Galsky MD, Rosenberg JE, Powles T, Petrylak DP, Bellmunt J, et al. Not applicable. Atezolizumab as first-line treatment in cisplatin-ineligible patients with locally advanced and metastatic urothelial carcinoma: a single-arm, multicentre, phase 2 trial. Lancet. 2017;389(10064):67–76. Ethics approval and consent to participate 17. Le DT, Uram JN, H W, R BB, Kemberling H, Eyring AD, et al. PD-1 blockade in Not applicable. tumors with mismatch-repair deficiency. NEJM. 2015:2509–20. 18. Goodman AM, Kato S, Bazhenova L, Patel SP, Frampton GM, Miller V, et al. Consent for publication Tumor mutational burden as an independent predictor of response to Not applicable. immunotherapy in diverse cancers. Mol Cancer Ther. 2017. https://doi.org/ 10.1158/1535-7163.MCT-17-0386. 19. Lawrence MS, Stojanov P, Polak P, Kryukov GV, Cibulskis K, Sivachenko A, et Competing interests al. Mutational heterogeneity in cancer and the search for new cancer- LF, SG, PGP, and LM declare that they have no competing interest. associated genes. Nature. 2013;499(7457):214–8. 20. Alexandrov LB, Nik-Zainal S, Wedge DC, Aparicio SAJR, Behjati S, Biankin AV, Author details et al. Signatures of mutational processes in human cancer. Nature. 2013; Department of Experimental Oncology, IEO, European Institute of Oncology 500(7463):415–21. IRCCS, Via Adamello 16, 20139 Milan, Italy. Department of Oncology and 21. Quiroga D, Lyerly HK, Morse MA. Deficient mismatch repair and the role of Hemato-Oncology, University of Milan, via Santa Sofia 9, 20142 Milan, Italy. immunotherapy in metastatic colorectal cancer. Curr Treat Options in Division of Early Drug Development, IEO, European Institute of Oncology Oncol. 2016. https://doi.org/10.1007/s11864-016-0414-4. IRCCS, Milan, Italy. 22. Chalmers ZR, Connelly CF, Fabrizio D, Gay L, Ali SM, Ennis R, et al. Analysis Received: 20 December 2018 Accepted: 19 June 2019 of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med. 2017;9(1):1–14. 23. Rooney MS, Shukla SA, Wu CJ, Getz G, Hacohen N. Molecular and genetic properties of tumors associated with local immune cytolytic activity. Cell. References 2015. https://doi.org/10.1016/j.cell.2014.12.033. 1. Antonia SJ, Villegas A, Daniel D, Vicente D, Murakami S, Ri H, et al. 24. Zhang J, Mardis ER, Maher CA. Genome analysis INTEGRATE-neo : a pipeline Durvalumab after chemoradiotherapy in stage III non-small cell lung cancer. for personalized gene fusion neoantigen discovery. Bioinformatics. 2017; N Engl J Med. 2017;377:1919–29. doi:https://doi.org/10.1093/bioinformatics/btw674. 2. Borghaei H, Paz-Ares L, Horn L. Nivolumab versus docetaxel in advanced 25. Yadav M, Jhunjhunwala S, Phung QT, Lupardus P, Tanguay J, Bumbaca S, et nonsquamous non-small-cell lung cancer. N Engl J Med. 2015;373:1627–39. al. Predicting immunogenic tumour mutations by combining mass 3. Hodi F, O’Day S, McDermott D, Weber R, Sosman J, Haanen J, et al. spectrometry and exome sequencing. Nature. 2014;515(7528):572–6. Improved survival with ipilimumab in patients with metastatic melanoma. N 26. Hellmann M, Nathanson T, Rizvi H. Genomic features of response to Engl J Med. 2010;363:711–23. combination immunotherapy in patients with advanced non- small-cell 4. Motzer R, Tannir N, McDermott D, Frontera O, Melichar B, Choueiri T, et al. lung cancer. Cancer Cell. 2018;33:843–52. Nivolumab plus Ipilimumab versus Sunitinib in advanced renal-cell 27. Kowanetz M. Tumor mutation load assessed by FoundationOne (FM1) is carcinoma. N Engl J Med. 2018;378(14):1277–90. associated with improved efficacy of atezolizumab (atezo) in patients with 5. Rosenberg JE, Hoff J, Powles T, Van Der HMS, Balar AV, Necchi A, et al. advanced NSCLC. Ann Oncol. 2016;27(6):15–42. Atezolizumab in patients with locally advanced and metastatic urothelial 28. Ready N, Hellmann MD, Awad MM, Otterson GA, Gutierrez M, Gainor JF, et carcinoma who have progressed following treatment with platinum-based al. First-line Nivolumab plus Ipilimumab in advanced non–small-cell lung chemotherapy: a single-arm, multicentre, phase 2 trial. Lancet. 2016. https:// cancer (CheckMate 568): outcomes by programmed death ligand 1 and doi.org/10.1016/S0140-6736(16)00561-4. tumor mutational burden as biomarkers. J Clin Oncol. 2019. https://doi.org/ 6. Li X, Song W, Shao C, Shi Y, Han W. Emerging predictors of the response to 10.1200/JCO.18.01042. the blockade of immune checkpoints in cancer therapy. Cell Mol Immunol. 29. Galsky M, Saci A, Szabo P, Azrilevich A, Horak C, Lambert A, et al. Impact of 2019. https://doi.org/10.1038/s41423-018-0086-z. tumor mutation burden on Nivolumab efficacy in second-line urothelial 7. Galluzzi L, Chan TA, Kroemer G, Wolchok JD, López-Soto A. The hallmarks of carcinoma patients: exploratory analysis of the phase II CheckMate 275. Ann successful anticancer immunotherapy. Sci Transl Med. 2018;10(459):1–15. Oncol. 2017. https://doi.org/10.1093/annonc/mdx371. 8. Campesato LF, Barroso-sousa R, Jimenez L, Camargo AA. Comprehensive 30. Strickland KC, Howitt BE, Shukla SA, Rodig S, Ritterhouse LL, Liu JF, et al. cancer-gene panels can be used to estimate mutational load and predict clinical Association and prognostic significance of BRCA1/2-mutation status with benefit to PD-1 blockade in clinical practice. Oncotarget. 2015;6(33):34221. neoantigen load, number of tumor-infiltrating lymphocytes and expression 9. Rizvi NA, Hellmann MD, Snyder A, Kvistborg P, Makarov V, Havel JJ, et al. of PD-1/PD-L1 in high grade serous ovarian cancer. Oncotarget. 2016;7(12): Mutational landscape determines sensitivity to PD-1 blockade in non – 13587–98. small cell lung cancer. Science. 2015;348(6230):124–9. 31. Birkbak NJ, Kochupurakkal B, Izarzugaza JMG, Eklund AC, Li Y, Liu J, et al. 10. Carbone DP, Reck M, Paz-Ares L, Creelan B, Horn L, Steins M, et al. First-line Tumor mutation burden forecasts outcome in ovarian cancer with BRCA1 Nivolumab in stage IV or recurrent non–small-cell lung Cancer. N Engl J or BRCA2 mutations. PLoS One. 2013;8(11). Med. 2017. https://doi.org/10.1056/NEJMoa1613493. 32. Thomas A, Routh ED, Pullikuth A, Jin G, Su J, Chou JW, et al. Tumor 11. Van Allen EM, Miao D, Schilling B, Shukla SA, Blank C, Zimmer L, et al. mutational burden is a determinant of immune-mediated survival in breast Genomic correlates of response to CTLA-4 blockade in metastatic cancer. Oncoimmunology. 2018;7(10):1–12. https://doi.org/10.1080/ melanoma. Science. 2015;350(6257):207–11. 2162402X.2018.1490854. 12. Johnson DB, Frampton GM, Rioth MJ, Yusko E, Xu Y, Guo X, et al. Targeted 33. Jayaraman SS, Rayhan DJ, Hazany S, Kolodney MS. Mutational landscape of Next Generation Sequencing Identi fi es Markers of Response to PD-1 basal cell carcinomas by whole-exome sequencing. J Invest Dermatol. 2014. Blockade. 2016;959–968. https://doi.org/10.1038/jid.2013.276. 13. Eroglu Z, Zaretsky JM, Hu-Lieskovan S, Kim DW, Algazi A, Johnson DB, et al. 34. Garofalo A, Sholl L, Reardon B, Taylor-Weiner A, Amin-Mansour A, Miao D, et High response rate to PD-1 blockade in desmoplastic melanomas. Nature. al. The impact of tumor profiling approaches and genomic data strategies 2018;553(7688):347–50. for cancer precision medicine. Genome Med. 2016. https://doi.org/10.1186/ 14. Snyder A, Makarov V, Merghoub T, Yuan J, Zaretsky JM, Desrichard A, et al. s13073-016-0333-9. Genetic Basis for Clinical Response to CTLA-4 Blockade in Melanoma. 2014; 2189–2199. 35. Buchhalter I, Rempel E, Endris V, Allgäuer M, Neumann O, Volckmar A-L, et al. 15. Powles T, Durán I, van der Heijden MS, Loriot Y, Vogelzang NJ, De Giorgi U, Size matters: dissecting key parameters for panel-based tumor mutational et al. Atezolizumab versus chemotherapy in patients with platinum-treated burden (TMB) analysis. Int J Cancer. 2019. https://doi.org/10.1002/ijc.31878. locally advanced or metastatic urothelial carcinoma (IMvigor211): a 36. Qiu P, Poehlein CH, Marton MJ, Laterza OF, Levitan D. Measuring tumor multicentre, open-label, phase 3 randomised controlled trial. Lancet. 2018. mutational burden (TMB) in plasma from mCRPC patients using two commercial https://doi.org/10.1016/S0140-6736(17)33297-X. NGS assays. Sci Rep. 2019. https://doi.org/10.1038/s41598-018-37128-y. Fancello et al. Journal for ImmunoTherapy of Cancer (2019) 7:183 Page 13 of 13 37. Nguyen A, Garner C, Reddy S, Sanborn J, Charles BS, Elisabeth ST, et al. 56. Murtaza M, Dawson SJ, Pogrebniak K, Rueda OM, Provenzano E, Grant J, et Three-fold overestimation of tumor mutation burden using 248 gene panel al. Multifocal clonal evolution characterized using circulating tumour DNA in versus whole exome. J Clin Oncol. 2018. https://doi.org/10.1200/JCO.2018. a case of metastatic breast cancer. Nat Commun. 2015. 36.15_suppl.12117. https://doi.org/10.1038/ncomms9760. 38. Panda A, Betigeri A, Subramanian K, Ross JS, Pavlick DC, Ali S, et al. 57. Butler TM, Johnson-Camacho K, Peto M, Wang NJ, Macey TA, Korkola JE, et Identifying a clinically applicable mutational burden threshold as a potential al. Exome sequencing of cell-free DNA from metastatic cancer patients biomarker of response to immune checkpoint therapy in solid tumors. JCO identifies clinically actionable mutations distinct from primary disease. PLoS Precis Oncol. 2017. https://doi.org/10.1200/PO.17.00146. One. 2015;10(8):1–14. 58. Klevebring D, Neiman M, Sundling S, Eriksson L, Ramqvist ED, Celebioglu F, 39. Endris V, Buchhalter I, Allgäuer M, Rempel E, Lier A, Volckmar A-L, et al. et al. Evaluation of exome sequencing to estimate tumor burden in plasma. Measurement of tumor mutational burden (TMB) in routine molecular PLoS One. 2014;9(8). diagnostics: in-silico and real-life analysis of three larger gene panels. Int J 59. Chan KC, Jiang P, Zheng YW, Liao GJ, Sun H, Wong J, et al. Cancer genome Cancer. 2019. https://doi.org/10.1002/ijc.32002. scanning in plasma: detection of tumor-associated copy number aberrations, 40. Zhang S, So AS, Kaplan S, KK M. Comprehensive evaluation of Illumina’s single-nucleotide variants, and tumoral heterogeneity by massively parallel TruSight® tumor 170 panel to estimate tumor mutational burden. Cancer sequencing. Clin Chem. 2013. https://doi.org/10.1373/clinchem.2012.196014. Res. 2017. https://doi.org/10.1158/1538-7445.AM2017-5358. 60. Koeppel F, Blanchard S, Marcaillou C, Martin E, Rouleau E, Solary E, et al. 41. Chaudhary R, Quagliata L, Martin JP, Alborelli I, Cyanam D, Mittal V, et al. A Whole exome sequencing for determination of tumor mutation load in scalable solution for tumor mutational burden from formalin- fixed , liquid biopsy from advanced cancer patients. PLoS One. 2017. https://doi. paraffin-embedded samples using the Oncomine tumor mutation load org/10.1371/journal.pone.0188174. assay. TLCR. 2018:1–15. 61. Gandara DR, Paul SM, Kowanetz M, Schleifman E, Zou W, Li Y, et al. Blood- 42. Rizvi H, Sanchez-Vega F, La K, Chatila W, Jonsson P, Halpenny D, et al. based tumor mutational burden as a predictor of clinical benefit in non- Molecular determinants of response to anti-programmed cell death (PD)-1 small-cell lung cancer patients treated with atezolizumab. Nat Med. 2018; and anti-programmed death-ligand 1 (PD-L1) blockade in patients with 24(September):1441. non-small-cell lung cancer profiled with targeted next-generation 62. Quinn K, Helman E, Nance T, Artieri C, Yen J, Zhao J, et al. Development sequencing. J Clin Oncol. 2018;36(7):633–41. and analytical validation of a plasma-based tumor mutational burden (TMB) 43. Samstein RM, Lee C, Shoushtari AN, Hellmann MD, Shen R, Janjigian YY, et al. score from next-generation sequencing panels. Ann Oncol. 2018; doi: 0. Tumor mutational load predicts survival after immunotherapy across multiple 1093/annonc/mdy269. cancer types. Nat Genet. 2019. https://doi.org/10.1038/s41588-018-0312-8. 63. Lyu G, Yeh Y, Yeh Y, Wang Y. Mutation load estimation model as a 44. Khagi Y, Goodman AM, Daniels GA, Patel SP, Sacco AG, Randall JM, et al. predictor of the response to cancer immunotherapy. npj Genomic Med. Hypermutated Circulating Tumor DNA : Correlation with Response to 2018. https://doi.org/10.1038/s41525-018-0051-x. Checkpoint Inhibitor – Based Immunotherapy. 2017;5729–5737. 64. Chan TA, Yarchoan M, Jaffee E, Swanton C, Quezada SA, Stenzinger A, et al. 45. Hellmann MD, Callahan MK, Awad MM, Calvo E, Ascierto PA, Atmaca A, et Development of tumor mutation burden as an immunotherapy biomarker: al. Tumor mutational burden and efficacy of Nivolumab monotherapy and utility for the oncology clinic. Ann Oncol. 2018. https://doi.org/10.1093/ in combination with Ipilimumab in small-cell lung Cancer. Cancer Cell. 2018. annonc/mdy495/5160130%0A. https://doi.org/10.1016/j.ccell.2018.04.001. 65. McGranahan N, Furness AJS, Rosenthal R, Ramskov S, Lyngaa R, Saini SK, et 46. Qiu P, Pang L, Arreaza G, Maguire M, Chang KCN, Marton MJ, et al. Data al. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to interoperability of whole exome sequencing ( WES ) based mutational immune checkpoint bloackade. Science (80- ). 2016;351(6280):1463–70. burden estimates from different laboratories. Int J Mol Sci. 2016. https://doi. 66. Kazmi SM. A retrospective analysis to evaluate prevalence and correlation org/10.3390/ijms17050651. between PD-L1 score and tumor mutational burden (TMB) levels in patients 47. Mola N, Schu M, Stiegelmeyer S, Jones W, Weigman V. Tumor mutational with solid tumor malignancies. Ann Oncol. 2017;28(Suppl 11):xi6–xi29. burden: guidelines for derivation and robustness of measurement. Cancer 67. Salem ME, Puccini A, Grothey A, Raghavan D, Goldberg RM, Xiu J, et al. Res. 2018. https://doi.org/10.1158/1538-7445.AM2018-2250. Landscape of Tumor Mutation Load , Mismatch Repair Deficiency, and PD-L1 48. Zook JM, Chapman B, Wang J, Mittelman D, Hofmann O, Hide W, et al. Expression in a Large Patient Cohort of Gastrointestinal Cancers. 2018;805–813. Integrating human sequence data sets provides a resource of benchmark 68. Morrison C, Pabla S, Conroy JM, Nesline MK, Glenn ST, Dressman D, et al. SNP and indel genotype calls. Nat Biotechnol. 2014;32(3):246–51. Predicting response to checkpoint inhibitors in melanoma beyond PD-L1 49. Wang Z, Dua J, Cai S, Han M, Dong H, Zhao J, et al. Assessment of blood and mutational burden. J Immunother Cancer. 2018;6(1):1–12. tumor mutational burden as a potential biomarker for immunotherapy in 69. Xiao W, Du N, Huang T, Guo J, Mo X, Yuan T, et al. TP53 mutation as potential patients with non–small cell lung cancer with use of a next-generation negative predictor for response of anti-CTLA-4 therapy in metastatic sequencing cancer gene panel. JAMA Oncol. 2019:1–7. melanoma. EBioMedicine. 2018. https://doi.org/10.1016/j.ebiom.2018.05.019. 50. Hellmann MD, Ciuleanu T-E, Pluzanski A, Lee JS, Otterson GA, Audigier- 70. Ock C-Y, Hwang J-E, Keam B, Kim S-B, Shim J-J, Jang H-J, et al. Genomic Valette C, et al. Nivolumab plus Ipilimumab in lung Cancer with a high landscape associated with potential response to anti-CTLA-4 treatment in tumor mutational burden. N Engl J Med. 2018. https://doi.org/10.1056/ cancers. Nat Commun. 2017. https://doi.org/10.1038/s41467-017-01018-0. NEJMoa1801946. 71. Kirchhoff T, Ferguson R, Simpson D, Kazlow E, Martinez C, Vogelsang M, et 51. Ramalingam SS. Tumor mutational burden (TMB) as a biomarker for clinical al. Germline determinants of immune related adverse events (irAEs) in benefit from dual immune checkpoint blockade with nivolumab (nivo) + melanoma immunotherapy response. Ann Oncol. 2017. https://doi.org/10. ipilimumab (ipi) in first-line (1L) non-small cell lung cancer (NSCLC): 1093/annonc/mdx376. identification of TMB cutoff from CheckMate 568. Cancer Res. 2018. https:// 72. Hasan Ali O, Berner F, Bomze D, Fässler M, Diem S, Cozzio A, et al. Human doi.org/10.1158/1538-7445.AM2018-CT078. leukocyte antigen variation is associated with adverse events of checkpoint 52. Chen Y, Zhang Y, Lv J, Li Y, Wang Y, He Q, et al. Genomic analysis of tumor inhibitors. Eur J Cancer. 2019;107:8–14. microenvironment immune types across 14 solid Cancer types : 73. Zehir A, Benayed R, Shah RH, Syed A, Middha S, Kim HR, et al. Mutational immunotherapeutic implications. Theranostics. 2017;7(14). landscape of metastatic cancer revealed from prospective clinical sequencing 53. Fabrizio D, Chen S-J, Xie M, Chen W, Quinn KJ, Zhao C, et al. In silico of 10,000 patients. Nat Med. 2017. https://doi.org/10.1038/nm.4333. assessment of variation in TMB quantification across diagnostic platforms: 74. Sun JX, He Y, Sanford E, Montesion M, Frampton GM, Vignot S, et al. A phase 1 of the friends of Cancer research harmonization project. J computational approach to distinguish somatic vs. germline origin of Immunother Cancer. 2018. https://doi.org/10.1186/s40425-018-0434-7. genomic alterations from deep sequencing of cancer specimens without a 54. Stenzinger A, Allen J, Maas J, Stewart M, Merino D, Dietel M. Tumor matched normal. PLoS Comput Biol. 2018. https://doi.org/10.1371/journal. mutational burden (TMB) standardization initiative: establishing a consistent pcbi.1005965. methodology for TMB measurement in clinical samples. Ann Oncol. 2018. https://doi.org/10.1093/annonc/mdy269.139. 55. Murtaza M, Dawson SJ, Tsui DWY, Gale D, Forshew T, Piskorz AM, et al. Non- Publisher’sNote invasive analysis of acquired resistance to cancer therapy by sequencing of Springer Nature remains neutral with regard to jurisdictional claims in plasma DNA. Nature. 2013. https://doi.org/10.1038/nature12065. published maps and institutional affiliations.

Journal

Journal for ImmunoTherapy of Cancer – Springer Journals

Published: Jul 15, 2019

There are no references for this article.