1. Current knowledge about the field under investigation
Clinical evidence demonstrates that treatment with immune checkpoint blocker (ICB) agents benefit to patients across multiple tumor type. However, development of predictive biomarkers is needed to identify patients who are most likely to respond to immunotherapy.
An emerging biomarker for response to immunotherapy is the total number of mutations present in a tumor specimen. This biomarker is named mutation load or tumor mutational burden (TMB). It is hypothesized that highly mutated tumors are more likely to harbor neoantigens targeted by activated immune cells. This metric, allowed by recent advances in next-generation sequencing (NGS) technologies, notably whole-exome sequencing (WES) and RNA-sequencing (RNA-seq), has been shown, in several tumor types, to correlate with patient response to ICB. Indeed, the TMB has been correlated with clinical benefits of anti-PD-1 and anti-CTLA-4 therapy in various tumor types, including malignant melanoma (Snyder et al., 2014; Van Allen et al., 2015) with a threshold of more than 100 nonsynonymous single-nucleotide variants (nsSNV) per exome, non-small cell lung cancers (NSCLC) (Rizvi et al., 2015) with a threshold defined as superior to 178 nsSNVs per exome, and several DNA repair-deficient tumors (Howitt et al., 2015; Le et al., 2015, 2017). A recent study prospectively confirmed that the PFS among patients with a high tumor mutational burden was significantly longer with nivolumab plus ipilimumab than with chemotherapy in NSCLC (Hellmann et al., 2018). Overall, a direct link between DNA repair deficiency, mutational landscape, predicted neoantigen load, and clinical activity of ICB is suggested.
TMB was defined as the number of somatic, coding, base substitution, and indel mutations per megabase of genome examined. All base substitutions and indels in the coding region of targeted genes. It has been shown that TMB calculated using cancer gene panel (CGP) assay agrees well with whole exome measures of mutation burden (Chalmers et al., 2017). This indicates that CGP, targeting the entire coding region of several hundred genes, covers sufficient genomic size to accurately assess WES mutational burden. It was found that filtering out germline alterations and rare variants was important to obtaining accurate measurements of TMB, and this will especially be important in patients from ethnic backgrounds not well represented in sequencing datasets. These findings indicate that CGP is an accurate, cost-effective, and clinically available tool for measuring TMB. The results of down sampling analysis show that the variation in measurement due to sampling when sequencing 1.1 Mb is acceptably low, resulting in highly accurate calling of TMB at a range of TMB levels. This sampling variation increases as the number of megabases sequenced decreases, especially at lower levels of TMB. While targeted CGP can be used to accurately assess TMB, it is not currently suited for identification of neoantigens, which might occur in any gene. Nevertheless, the theragnostic impact of TMB has been also determined with a targeted CGP by the FoundationOne CDx assay (Hellmann et al., 2018).
The failure to assess the TMB may occur at different steps in the analytical process. It can be at the sample acquisition with a low level of DNA (5-10%). The NGS processing depending on the method can lead to a process failure around 2-3%. And finally, the bioinformatic curation will lead to a 3-5% of failure. At the end, the TMB will be undetermined with an attrition rate at 15% of the samples (Fondation Medicine Personal communication).
The implementation of tumor mutational burden assessment in a nation-wide perspective is challenging and large efforts should be done. Whatever the method, WES or large CGP, the measurements of TMB will change the scale of gene sequencing in the molecular platforms.
It is important that the platforms, where the tumor mutational burden measurement will be implemented can be able to calibrate their CGP and compared it to a standard reference for the measurement of TMB. Moreover, with the limited quantity of DNA, the implemented method needs to produce reliable result for most of the samples. The platforms need also to deploy their wet benches processes and their bioinformatics pipeline to be able to produce the TMB in a turnaround time compatible with clinical patients management.
2. Description of the population of study participants and justification for the choice of participants
As mentioned above the TMB seems to be is an important marker for ICB efficacy. It is therefore important to implement the TMB measurement at a nationwide level to assure the equal access to the ICB.
The French National Cancer Institute (INCa) has labelled molecular platforms which are in charge of the different molecular tests should establish their processes to be able to face to this new test. This study is designated to test different CGP methods to assess tumor mutational burden in lung cancers in order to compare the different methods and to identify the drawbacks and failures of the different methods. Results of this study should allow to estimate the needs for the implementation of TMB at the nationwide level, recommend cancer gene panels, propose validated bioinformatics pipelines and finally recommend preanalytical processes for reducing the failure rate of the tests.
For this purpose this study will determine the TMB on patients with a lung cancer. 200 patients with a non-small cell lung cancer (NSCLC) and naive of treatment will be recruited.
Three cohorts will be recruited:
* 100 patients will be included without resection and we will determine the TMB on the biopsy specimen
* 100 patients will be included after surgical resection of their tumor and we will determine the TMB on the resected specimen
* Within this cohort, TMB will be assessed both on the biopsy specimen and on the resected specimen for 20 patients 200 patients will be enrolled in order to perform 220 TMB analysis.
Participating sites are large platforms which were selected on the basis of their capacity to implement new tests and which have all facilities for conducting the described research. The patients included in this study will correspond to those that will receive immune checkpoint inhibitors during the course of their disease. This study in a non-interventional study, as it will not modify the clinical care of the patients. The tissue sampling will be taken during clinical procedure needed for patient care. The study will be proposed to patients after the verification that tissues have been collected in a way that is compatible with the present study. Only tumor material and no germline DNA will be collected.
Selected patients will be close to those that receive ICB and the target population will be enriched with patients who had been surgically resected, in order to be able to have enough tissues to compare in these samples different methods to assess TMB and also to compare pre surgical biopsies and surgical specimen.
3. Description of the element or elements under investigation
Elements under investigation are the TMB of NSCLC patients. The attrition rate (number of cases without result) in the TMB estimation will be assessed. The gold standard for the determination of TMB is the WES as it covers all coding regions. Several others Cancer Gene Panel methods (CGP) have been described based on cancer gene panels sequencing. The CGP included in this study will be the panel from Illumina (TS500), Thermofisher (Oncomine Tumor Mutation Load Assay), home-made panel with Agilent technology (Dedicated XT HS) and Foundation medicine F1 CDx. It is recognized that size of the cancer gene panels should be close to 1Mb to be able to measure reliably the TMB in different type of cancer.
Different methods of measurement of TMB will be compared in this series on NSCLC in order to validate different CGP available on the market or in the different INCa platforms. Bioinformatics modelling will try to correlate the TMB results from WES and CGP. For this purpose, based on results of mutational load calculated from results of WES, results of mutation load will be extrapolated based on a random selection of 1000 genes repeated 1000 times. The mean of the correlation coefficient (R2) between the tumor mutational burden calculated by the whole exome sequencing and by the different random gene panels is 0.88 IQR \[0.84-0.91\] showing that the selection of genes to estimate the TMB should be evaluated carefully before to be used in clinical practice (unpublished results). The material that will be collected during the study will allow to determine the performance of the CGP implemented in the molecular biology platforms.
