Periodic Reporting for period 1 - IMMUNOMARK (Omics integration for precision cancer immunotherapy)
Période du rapport: 2018-03-01 au 2020-02-29
Checkpoint inhibitor therapy constitutes a promising cancer treatment strategy that targets the immune checkpoints to re-activate silenced T cell cytotoxicity. Immune checkpoint blockade (ICB) demonstrated durable responses and acceptable toxicity. However, up to ~85% of patients present with innate or acquired resistance to ICB, limiting its clinical utility. Current response biomarker candidates, including DNA mutation and neoantigen load, immune profiles, as well as programmed death-ligand 1 (PD-L1) expression, are only weak predictors of ICB response. Thus, identification of novel, more predictive biomarkers constitutes one of the most important areas of immunotherapy research. In IMMUNOMARK, we proposed a multilayer integrative omics analysis to identify biomarkers of clinical response to immunotherapy amenable to be used in liquid biopsies, and therefore contribute to reach the Horizon 2020 Research & Innovation goal: “to improve our ability to monitor health and to prevent, detect, treat and manage disease”.
The specific scientific goal was to identify definite biomarkers predictive of the response to anti-PD-1, anti-PD-L1, and anti-CTLA4 therapy in patients with metastatic melanoma and advanced Non–Small-Cell Lung Cancer (NSCLC).
Our research strategy, implemented in two Work Packages, consisted of the multidimensional exploration of putative players in the innate and acquired resistance scenarios. On one hand, we hypothesized that the phenotypic changes associated to the resistance to ICB are not only related to genetic variants but also to epigenetic differences among individuals. Other dimension integrated in the study was the stroma, which has resulted vital for deciphering which cells of the inflammatory infiltrate are the most important for determining a good response to ICB. Finally, we considered not only solid biopsies but also liquid biopsies in order to search for non-invasive biomarkers of response that could give a real-time information of the treatment efficacy.
The most important conclusions derived from the results of this study are two:
1) The most significant mechanisms of response to ICB in both metastatic melanoma and NSCLC patients are driven by a differential activity of specific cells of the immune system.
- In NSCLC, the EPIMMUNE CpG signature, specifically the unmethylated status of a CpG site of FOXP1, a gene related to quiescent naïve CD4+ T-cells, is associated with favorable outcome of ICB treated patients.
- In metastatic melanoma, a transcriptomic signature of 140 genes defines the innate response to ICB, and 58 out of them are associated with overall survival, constituting new putative predictors of outcome. Importantly, the differential expression signature has unmasked a major contribution of the B lymphocytes in the response to ICB in melanoma patients, which has been validated in single cell RNA-seq datasets. Moreover, we have identified the plasmablast-like B lymphocytes as the specific subtype enriched in the patients that respond well to ICB; overall, this opens a completely new avenue for disentangling the immune-suppressor role of B cells in cancer.
2) Direct Fluorescence-activated cell-based quantification of specific lymphocyte and myeloid cell populations in peripheral blood from patients from metastatic melanoma and NSCLC provides non-invasive biomarkers of outcome to ICB treatment. In addition, their combination with specific clinicopathological variables in logistic regression prediction model increases the ability to predict response with high specificity and sensitivity.
The specific scientific goal was to identify definite biomarkers predictive of the response to anti-PD-1, anti-PD-L1, and anti-CTLA4 therapy in patients with metastatic melanoma and advanced Non–Small-Cell Lung Cancer (NSCLC).
Our research strategy, implemented in two Work Packages, consisted of the multidimensional exploration of putative players in the innate and acquired resistance scenarios. On one hand, we hypothesized that the phenotypic changes associated to the resistance to ICB are not only related to genetic variants but also to epigenetic differences among individuals. Other dimension integrated in the study was the stroma, which has resulted vital for deciphering which cells of the inflammatory infiltrate are the most important for determining a good response to ICB. Finally, we considered not only solid biopsies but also liquid biopsies in order to search for non-invasive biomarkers of response that could give a real-time information of the treatment efficacy.
The most important conclusions derived from the results of this study are two:
1) The most significant mechanisms of response to ICB in both metastatic melanoma and NSCLC patients are driven by a differential activity of specific cells of the immune system.
- In NSCLC, the EPIMMUNE CpG signature, specifically the unmethylated status of a CpG site of FOXP1, a gene related to quiescent naïve CD4+ T-cells, is associated with favorable outcome of ICB treated patients.
- In metastatic melanoma, a transcriptomic signature of 140 genes defines the innate response to ICB, and 58 out of them are associated with overall survival, constituting new putative predictors of outcome. Importantly, the differential expression signature has unmasked a major contribution of the B lymphocytes in the response to ICB in melanoma patients, which has been validated in single cell RNA-seq datasets. Moreover, we have identified the plasmablast-like B lymphocytes as the specific subtype enriched in the patients that respond well to ICB; overall, this opens a completely new avenue for disentangling the immune-suppressor role of B cells in cancer.
2) Direct Fluorescence-activated cell-based quantification of specific lymphocyte and myeloid cell populations in peripheral blood from patients from metastatic melanoma and NSCLC provides non-invasive biomarkers of outcome to ICB treatment. In addition, their combination with specific clinicopathological variables in logistic regression prediction model increases the ability to predict response with high specificity and sensitivity.
Overall, the objectives of our project have been fulfilled, with 4 major discoveries of impact in the field of precision cancer immunotherapy:
- Firstly, the identification of a DNA methylation signature of response to Nivolumab in NSCLC patients (The Lancet Resp Medicine, 2018; Best Practice, 2019).
The methylation signature, EPIMMUNE, was validated in other cohorts and compared with the most relevant response biomarkers to date. The fact that both tumor- and immune system- related genes formed part of this signature, and the better prediction value over PD-L1 or CD8, as well as the epigenetic nature of the biomarkers, corroborate our three standing hypotheses.
- Secondly, the generation of a transcriptomic signature of response to Nivolumab in metastatic melanoma patients (J Trans Med 2020). The major contribution of immune-related, particularly B lymphocytes, processes in the determination of good outcome is a ground-breaking finding in the field that goes along with our hypothesis of the best predictive biomarkers being the ones at the interaction of the tumor and the extended immune system beyond T lymphocyte markers.
- Thirdly, the exploitation of the new bioinformatic pipelines that have been developed in parallel with the flourishment of the field in the last two years, have permitted to characterize not only the TMB and other putative important biomarkers of immunotherapy from All RNA-seq data, but to validate the results in silico in single-cell datasets, to identify the specific B lymphocyte subtype associated to the response to Nivolumab, and to characterize the abundancy and clonality of the T lymphocyte and B lymphocyte receptors, and HLA of our patients in relation with the tumor neoantigens. In addition, we have implemented recent bioinformatic tools using de novo assembly for the identification of circulating RNA (circRNA) to characterize the metastatic melanoma circRNAs profile and the specific pattern for responders.
- The fourth major discovery was the novel, previously unreported, association of peripheral blood circulating specific populations of lymphocytes and myeloid cells with response to immunotherapy in metastatic melanoma and NSCLC patients. In addition, we have combined the cytometry data, and the clinicopathological variables of our circulating cohort in a multivariate based model that is able to predict response to immunotherapy with a maximum AUC of 0.94 (sensitivity 76.9%; specificity 100%).
- Firstly, the identification of a DNA methylation signature of response to Nivolumab in NSCLC patients (The Lancet Resp Medicine, 2018; Best Practice, 2019).
The methylation signature, EPIMMUNE, was validated in other cohorts and compared with the most relevant response biomarkers to date. The fact that both tumor- and immune system- related genes formed part of this signature, and the better prediction value over PD-L1 or CD8, as well as the epigenetic nature of the biomarkers, corroborate our three standing hypotheses.
- Secondly, the generation of a transcriptomic signature of response to Nivolumab in metastatic melanoma patients (J Trans Med 2020). The major contribution of immune-related, particularly B lymphocytes, processes in the determination of good outcome is a ground-breaking finding in the field that goes along with our hypothesis of the best predictive biomarkers being the ones at the interaction of the tumor and the extended immune system beyond T lymphocyte markers.
- Thirdly, the exploitation of the new bioinformatic pipelines that have been developed in parallel with the flourishment of the field in the last two years, have permitted to characterize not only the TMB and other putative important biomarkers of immunotherapy from All RNA-seq data, but to validate the results in silico in single-cell datasets, to identify the specific B lymphocyte subtype associated to the response to Nivolumab, and to characterize the abundancy and clonality of the T lymphocyte and B lymphocyte receptors, and HLA of our patients in relation with the tumor neoantigens. In addition, we have implemented recent bioinformatic tools using de novo assembly for the identification of circulating RNA (circRNA) to characterize the metastatic melanoma circRNAs profile and the specific pattern for responders.
- The fourth major discovery was the novel, previously unreported, association of peripheral blood circulating specific populations of lymphocytes and myeloid cells with response to immunotherapy in metastatic melanoma and NSCLC patients. In addition, we have combined the cytometry data, and the clinicopathological variables of our circulating cohort in a multivariate based model that is able to predict response to immunotherapy with a maximum AUC of 0.94 (sensitivity 76.9%; specificity 100%).
With the implementation of the current project, we have generated several transcendent results in the field, highlighting the EPIMMUNE signature, the B lymphocyte signature, and the circulating specific population of lymphocytes as novel biomarkers of response to immunotherapy.
Our results have been amply disseminated to the scientific community and other audiences in society, and have been discussed with the Unit of Innovation and Technological Transfer of IBIMA to apply for a European patent that includes both a methodological adaptation for all RNA-seq with Ribosomal RNA depletion from FFPE samples, and a transcriptomic signature of response to ICB.
We believe this fellowship have proved useful for the advance of the field of precision cancer immunotherapy and for the consolidation of the scientific career of the fellow.
It is envisaged that the patent can be exploited in the form of specific kits that test the most common variants that are implied in the response to ICB; given the proliferation of the FDA and EMA approvals of ICB cancer treatments, the socio-economic impact of the results generated in this proposal is expected to be important. In addition, the integration of the project into a Public Oncology Unit and of other hospitals of the region provides with the ideal scenario to implement the identified biomarkers and models into the clinical practice.
Our results have been amply disseminated to the scientific community and other audiences in society, and have been discussed with the Unit of Innovation and Technological Transfer of IBIMA to apply for a European patent that includes both a methodological adaptation for all RNA-seq with Ribosomal RNA depletion from FFPE samples, and a transcriptomic signature of response to ICB.
We believe this fellowship have proved useful for the advance of the field of precision cancer immunotherapy and for the consolidation of the scientific career of the fellow.
It is envisaged that the patent can be exploited in the form of specific kits that test the most common variants that are implied in the response to ICB; given the proliferation of the FDA and EMA approvals of ICB cancer treatments, the socio-economic impact of the results generated in this proposal is expected to be important. In addition, the integration of the project into a Public Oncology Unit and of other hospitals of the region provides with the ideal scenario to implement the identified biomarkers and models into the clinical practice.