Comparing the Efficacy of Two Generations of EGFR-TKIs: An Integrated Drug-Disease Mechanistic Model Approach in EGFR-Mutated Lung Adenocarcinoma.

Mutationsin epidermal growth factor receptor (EGFR) are found in approximately 48% of Asian and 19% of Western patients with lung adenocarcinoma (LUAD), leading to aggressive tumor growth. While tyrosine kinase inhibitors (TKIs) like gefitinib and osimertinib target this mutation, treatments often face challenges such as metastasis and resistance. To address this, we developed physiologically based pharmacokinetic (PBPK) models for both drugs, simulating their distribution within the primary tumor and metastases following oral administration. These models, combined with a mechanistic knowledge-based disease model of EGFR-mutated LUAD, allow us to predict the tumor's behavior under treatment considering the diversity within the tumor cells due to different mutations. The combined model reproduces the drugs' distribution within the body, as well as the effects of both gefitinib and osimertinib on EGFR-activation-induced signaling pathways. In addition, the disease model encapsulates the heterogeneity within the tumor through the representation of various subclones. Each subclone is characterized by unique mutation profiles, allowing the model to accurately reproduce clinical outcomes, including patients' progression, aligning with RECIST criteria guidelines (version 1.1). Datasets used for calibration came from NEJ002 and FLAURA clinical trials. The quality of the fit was ensured with rigorous visual predictive checks and statistical tests (comparison metrics computed from bootstrapped, weighted log-rank tests: 98.4% (NEJ002) and 99.9% (FLAURA) similarity). In addition, the model was able to predict outcomes from an independent retrospective study comparing gefitinib and osimertinib which had not been used within the model development phase. This output validation underscores mechanistic models' potential in guiding future clinical trials by comparing treatment efficacies and identifying patients who would benefit most from specific TKIs. Our work is a step towards the design of a powerful tool enhancing personalized treatment in LUAD. It could support treatment strategy evaluations and potentially reduce trial sizes, promising more efficient and targeted therapeutic approaches. Following its consecutive prospective validations with the FLAURA2 and MARIPOSA trials (validation metrics computed from bootstrapped, weighted log-rank tests: 94.0% and 98.1%, respectively), the model could be used to generate a synthetic control arm.

Empirical methods for the validation of time-to-event mathematical models taking into account uncertainty and variability: application to EGFR + lung adenocarcinoma.

BACKGROUND: Over the past several decades, metrics have been defined to assess the quality of various types of models and to compare their performance depending on their capacity to explain the variance found in real-life data. However, available validation methods are mostly designed for statistical regressions rather than for mechanistic models. To our knowledge, in the latter case, there are no consensus standards, for instance for the validation of predictions against real-world data given the variability and uncertainty of the data. In this work, we focus on the prediction of time-to-event curves using as an application example a mechanistic model of non-small cell lung cancer. We designed four empirical methods to assess both model performance and reliability of predictions: two methods based on bootstrapped versions of parametric statistical tests: log-rank and combined weighted log-ranks (MaxCombo); and two methods based on bootstrapped prediction intervals, referred to here as raw coverage and the juncture metric. We also introduced the notion of observation time uncertainty to take into consideration the real life delay between the moment when an event happens, and the moment when it is observed and reported. RESULTS: We highlight the advantages and disadvantages of these methods according to their application context. We have shown that the context of use of the model has an impact on the model validation process. Thanks to the use of several validation metrics we have highlighted the limit of the model to predict the evolution of the disease in the whole population of mutations at the same time, and that it was more efficient with specific predictions in the target mutation populations. The choice and use of a single metric could have led to an erroneous validation of the model and its context of use. CONCLUSIONS: With this work, we stress the importance of making judicious choices for a metric, and how using a combination of metrics could be more relevant, with the objective of validating a given model and its predictions within a specific context of use. We also show how the reliability of the results depends both on the metric and on the statistical comparisons, and that the conditions of application and the type of available information need to be taken into account to choose the best validation strategy.

Knowledge-based mechanistic modeling accurately predicts disease progression with gefitinib in EGFR-mutant lung adenocarcinoma.

Lung adenocarcinoma (LUAD) is associated with a low survival rate at advanced stages. Although the development of targeted therapies has improved outcomes in LUAD patients with identified and specific genetic alterations, such as activating mutations on the epidermal growth factor receptor gene (EGFR), the emergence of tumor resistance eventually occurs in all patients and this is driving the development of new therapies. In this paper, we present the In Silico EGFR-mutant LUAD (ISELA) model that links LUAD patients' individual characteristics, including tumor genetic heterogeneity, to tumor size evolution and tumor progression over time under first generation EGFR tyrosine kinase inhibitor gefitinib. This translational mechanistic model gathers extensive knowledge on LUAD and was calibrated on multiple scales, including in vitro, human tumor xenograft mouse and human, reproducing more than 90% of the experimental data identified. Moreover, with 98.5% coverage and 99.4% negative logrank tests, the model accurately reproduced the time to progression from the Lux-Lung 7 clinical trial, which was unused in calibration, thus supporting the model high predictive value. This knowledge-based mechanistic model could be a valuable tool in the development of new therapies targeting EGFR-mutant LUAD as a foundation for the generation of synthetic control arms.

Identification of human progenitors of exhausted CD8+ T cells associated with elevated IFN-γ response in early phase of viral infection.

T cell exhaustion is a hallmark of hepatitis C virus (HCV) infection and limits protective immunity in chronic viral infections and cancer. Limited knowledge exists of the initial viral and immune dynamics that characterise exhaustion in humans. We studied longitudinal blood samples from a unique cohort of individuals with primary infection using single-cell multi-omics to identify the functions and phenotypes of HCV-specific CD8+ T cells. Early elevated IFN-γ response against the transmitted virus is associated with the rate of immune escape, larger clonal expansion, and early onset of exhaustion. Irrespective of disease outcome, we find heterogeneous subsets of progenitors of exhaustion, based on the level of PD-1 expression and loss of AP-1 transcription factors. Intra-clonal analysis shows distinct trajectories with multiple fates and evolutionary plasticity of precursor cells. These findings challenge the current paradigm on the contribution of CD8+ T cells to HCV disease outcome and provide data for future studies on T cell differentiation in human infections.

Innate cell markers that predict anti-HIV neutralizing antibody titers in vaccinated macaques.

Given the time and resources invested in clinical trials, innovative prediction methods are needed to decrease late-stage failure in vaccine development. We identify combinations of early innate responses that predict neutralizing antibody (nAb) responses induced in HIV-Env SOSIP immunized cynomolgus macaques using various routes of vaccine injection and adjuvants. We analyze blood myeloid cells before and 24 h after each immunization by mass cytometry using a three-step clustering, and we discriminate unique vaccine signatures based on HLA-DR, CD39, CD86, CD11b, CD45, CD64, CD14, CD32, CD11c, CD123, CD4, CD16, and CADM1 surface expression. Various combinations of these markers characterize cell families positively associated with nAb production, whereas CADM1-expressing cells are negatively associated (p < 0.05). Our results demonstrate that monitoring immune signatures during early vaccine development could assist in identifying biomarkers that predict vaccine immunogenicity.

Integration of Heterogeneous Biological Data in Multiscale Mechanistic Model Calibration: Application to Lung Adenocarcinoma.

Mechanistic models are built using knowledge as the primary information source, with well-established biological and physical laws determining the causal relationships within the model. Once the causal structure of the model is determined, parameters must be defined in order to accurately reproduce relevant data. Determining parameters and their values is particularly challenging in the case of models of pathophysiology, for which data for calibration is sparse. Multiple data sources might be required, and data may not be in a uniform or desirable format. We describe a calibration strategy to address the challenges of scarcity and heterogeneity of calibration data. Our strategy focuses on parameters whose initial values cannot be easily derived from the literature, and our goal is to determine the values of these parameters via calibration with constraints set by relevant data. When combined with a covariance matrix adaptation evolution strategy (CMA-ES), this step-by-step approach can be applied to a wide range of biological models. We describe a stepwise, integrative and iterative approach to multiscale mechanistic model calibration, and provide an example of calibrating a pathophysiological lung adenocarcinoma model. Using the approach described here we illustrate the successful calibration of a complex knowledge-based mechanistic model using only the limited heterogeneous datasets publicly available in the literature.

Initial TK-deficient HSV-1 infection in the lip alters contralateral lip challenge immune dynamics.

Primary infection with herpes simplex type 1 (HSV-1) occurring around the mouth and nose switches rapidly to lifelong latent infection in sensitive trigeminal ganglia (TG) neurons. Sporadic reactivation of these latent reservoirs later in life is the cause of acute infections of the corneal epithelium, which can cause potentially blinding herpes simplex keratitis (HSK). There is no effective vaccine to protect against HSK, and antiviral drugs provide only partial protection against recurrences. We previously engendered an acute disease-free, non-reactivating latent state in mice when challenged with virulent HSV-1 in orofacial mucosa, by priming with non-neurovirulent HSV-1 (TKdel) before the challenge. Herein, we define the local immune infiltration and inflammatory chemokine production changes after virulent HSV-1 challenge, which were elicited by TKdel prime. Heightened immunosurveillance before virulent challenge, and early enhanced lymphocyte-enriched infiltration of the challenged lip were induced, which corresponded to attenuation of inflammation in the TG and enhanced viral control. Furthermore, classical latent-phase T cell persistence around latent HSV-1 reservoirs were severely reduced. These findings identify the immune processes that are likely to be responsible for establishing non-reactivating latent HSV-1 reservoirs. Stopping reactivation is essential for development of efficient vaccine strategies against HSV-1.

Human CD8 T-stem cell memory subsets phenotypic and functional characterization are defined by expression of CD122 or CXCR3.

Long-lived T-memory stem cells (TSCM ) are key to both naturally occurring and vaccine-conferred protection against infection. These cells are characterized by the CD45RA+ CCR7+ CD95+ phenotype. Significant heterogeneity within the TSCM population is recognized, but distinguishing surface markers and functional characterization of potential subsets are lacking. Human CD8 TSCM subsets were identified in healthy subjects who had been previously exposed to CMV or Influenza (Flu) virus in flow cytometry by expression of CD122 or CXCR3, and then characterized in proliferation, multipotency, self-renewal, and intracellular cytokine production (TNF-α, IL-2, IFN-γ), together with transcriptomic profiles. The TSCM CD122hi -expressing subset (versus CD122lo ) demonstrated greater proliferation, greater multipotency, and enhanced polyfunctionality with higher frequencies of triple positive (TNF-α, IL-2, IFN-γ) cytokine-producing cells upon exposure to recall antigen. The TSCM CXCR3lo subpopulation also had increased proliferation and polyfunctional cytokine production. Transcriptomic analysis further showed that the TSCM CD122hi population had increased expression of activation and homing molecules, such as Ccr6, Cxcr6, Il12rb, and Il18rap, and downregulated cell proliferation inhibitors, S100A8 and S100A9. These data reveal that the TSCM CD122hi phenotype is associated with increased proliferation, enhanced multipotency and polyfunctionality with an activated memory-cell like transcriptional profile, and hence, may be favored for induction by immunization and for adoptive immunotherapy.

Optimize Prime/Boost Vaccine Strategies: Trained Immunity as a New Player in the Game.

Most vaccines require multiple doses to induce long-lasting protective immunity in a high frequency of vaccines, and to ensure strong both individual and herd immunity. Repetitive immunogenic stimulations not only increase the intensity and durability of adaptive immunity, but also influence its quality. Several vaccine parameters are known to influence adaptive immune responses, including notably the number of immunizations, the delay between them, and the delivery sequence of different recombinant vaccine vectors. Furthermore, the initial effector innate immune response is key to activate and modulate B and T cell responses. Optimization of homologous and heterologous prime/boost vaccination strategies requires a thorough understanding of how vaccination history affects memory B and T cell characteristics. This requires deeper knowledge of how innate cells respond to multiple vaccine encounters. Here, we review how innate cells, more particularly those of the myeloid lineage, sense and respond differently to a 1st and a 2nd vaccine dose, both in an extrinsic and intrinsic manner. On one hand, the presence of primary specific antibodies and memory T cells, whose critical properties change with time after priming, provides a distinct environment for innate cells at the time of re-vaccination. On the other hand, innate cells themselves can exert enhanced intrinsic antimicrobial functions, long after initial stimulation, which is referred to as trained immunity. We discuss the potential of trained innate cells to be game-changers in prime/boost vaccine strategies. Their increased functionality in antigen uptake, antigen presentation, migration, and as cytokine producers, could indeed improve the restimulation of primary memory B and T cells and their differentiation into potent secondary memory cells in response to the boost. A better understanding of trained immunity mechanisms will be highly valuable for harnessing the full potential of trained innate cells, to optimize immunization strategies.

The Route of Vaccine Administration Determines Whether Blood Neutrophils Undergo Long-Term Phenotypic Modifications.

Innate immunity modulates adaptive immunity and defines the magnitude, quality, and longevity of antigen-specific T- and B- cell immune memory. Various vaccine and administration factors influence the immune response to vaccination, including the route of vaccine delivery. We studied the dynamics of innate cell responses in blood using a preclinical model of non-human primates immunized with a live attenuated vaccinia virus, a recombinant Modified vaccinia virus Ankara (MVA) expressing a gag-pol-nef fusion of HIV-1, and mass cytometry. We previously showed that it induces a strong, early, and transient innate response, but also late phenotypic modifications of blood myeloid cells after two months when injected subcutaneously. Here, we show that the early innate effector cell responses and plasma inflammatory cytokine profiles differ between subcutaneous and intradermal vaccine injection. Additionally, we show that the intradermal administration fails to induce more highly activated/mature neutrophils long after immunization, in contrast to subcutaneous administration. Different batches of antibodies, staining protocols and generations of mass cytometers were used to generate the two datasets. Mass cytometry data were analyzed in parallel using the same analytical pipeline based on three successive clustering steps, including SPADE, and categorical heatmaps were compared using the Manhattan distance to measure the similarity between cell cluster phenotypes. Overall, we show that the vaccine per se is not sufficient for the late phenotypic modifications of innate myeloid cells, which are evocative of innate immune training. Its route of administration is also crucial, likely by influencing the early innate response, and systemic inflammation, and vaccine biodistribution.

Non-human Primate Determinants of Natural Killer Cells in Tissues at Steady-State and During Simian Immunodeficiency Virus Infection.

Natural killer (NK) cells play essential roles in immunity to viruses and tumors. Their function is genetically determined but also modulated by environmental factors. The distribution and functional regulation of these cells vary depending on the tissue. NK cell behavior in lymphoid tissues is so far understudied. Non-human primate (NHP) models are essential for the development of therapies and vaccines against human diseases, and access to NHP tissues allows insights into spatial regulations of NK cells. Here, we investigated tissue-specific parameters of NK cells from NHP species, i.e., cynomolgus macaque (Macaca fascicularis), African green monkey (Chlorocebus sabaeus), rhesus macaque (Macaca mulatta), and baboon (Papio anubis). By comprehensive multi-dimensional analysis of NK cells from secondary lymphoid organs, intestinal mucosa, liver, and blood, we identified tissue- and species-specific patterns of NK cell frequencies, phenotypes, and potential activity. Also, we defined the tissue-specific characteristics of NK cells during infection by the simian immunodeficiency virus. Altogether, our results provide a comprehensive anatomic analysis of NK cells in different tissues of primates at steady-state and during a viral infection.

Innate and secondary humoral responses are improved by increasing the time between MVA vaccine immunizations.

Comprehending the mechanisms behind the impact of vaccine regimens on immunity is critical for improving vaccines. Indeed, the time-interval between immunizations may influence B and T cells, as well as innate responses. We compared two vaccine schedules using cynomolgus macaques immunized with an attenuated vaccinia virus. Two subcutaneous injections 2 weeks apart led to an impaired secondary antibody response and similar innate myeloid responses to both immunizations. In contrast, a delayed boost (2 months) improved the quality of the antibody response and involved more activated/mature innate cells, induced late after the prime and responding to the recall. The magnitude and quality of the secondary antibody response correlated with the abundance of these neutrophils, monocytes, and dendritic cells that were modified phenotypically and enriched prior to revaccination at 2 months, but not 2 weeks. These late phenotypic modifications were associated with an enhanced ex vivo cytokine production (including IL-12/23 and IL-1ß) by PBMCs short after the second immunization, linking phenotype and functions. This integrated analysis reveals a deep impact of the timing between immunizations, and highlights the importance of early but also late innate responses involving phenotypical changes, in shaping humoral immunity.

NK cell immune responses differ after prime and boost vaccination.

A better understanding of innate responses induced by vaccination is critical for designing optimal vaccines. Here, we studied the diversity and dynamics of the NK cell compartment after prime-boost immunization with the modified vaccinia virus Ankara using cynomolgus macaques as a model. Mass cytometry was used to deeply characterize blood NK cells. The NK cell subphenotype composition was modified by the prime. Certain phenotypic changes induced by the prime were maintained over time and, as a result, the NK cell composition prior to boost differed from that before prime. The key phenotypic signature that distinguished NK cells responding to the boost from those responding to the prime included stronger expression of several cytotoxic, homing, and adhesion molecules, suggesting that NK cells at recall were functionally distinct. Our data reveal potential priming or imprinting of NK cells after the first vaccine injection. This study provides novel insights into prime-boost vaccination protocols that could be used to optimize future vaccines.

Prime and Boost Vaccination Elicit a Distinct Innate Myeloid Cell Immune Response.

Understanding the innate immune response to vaccination is critical in vaccine design. Here, we studied blood innate myeloid cells after first and second immunization of cynomolgus macaques with the modified vaccinia virus Ankara. The inflammation at the injection site was moderate and resolved faster after the boost. The blood concentration of inflammation markers increased after both injections but was lower after the boost. The numbers of neutrophils, monocytes, and dendritic cells were transiently affected by vaccination, but without any major difference between prime and boost. However, phenotyping deeper those cells with mass cytometry unveiled their high phenotypic diversity with subsets responding differently after each injection, some enriched only after the primary injection and others only after the boost. Actually, the composition in subphenotype already differed just before the boost as compared to just before the prime. Multivariate analysis identified the key features that contributed to these differences. Cell subpopulations best characterizing the post-boost response were more activated, with a stronger expression of markers involved in phagocytosis, antigen presentation, costimulation, chemotaxis, and inflammation. This study revisits innate immunity by demonstrating that, like adaptive immunity, innate myeloid responses differ after one or two immunizations.

In depth comparative phenotyping of blood innate myeloid leukocytes from healthy humans and macaques using mass cytometry.

Comparative immune-profiling of innate responses in humans and non-human primates is important to understand the pathogenesis of infectious and chronic inflammatory diseases as well as for the preclinical development of vaccines and immune therapies. However, direct comparisons of the two species are rare and were never performed using mass cytometry. Here, whole-blood-derived leukocytes from healthy humans and cynomolgus macaques were analyzed with mass cytometry. Two similar panels of around 30 monoclonal antibodies targeting human markers associated with innate myeloid cells to stain fixed human and macaque leukocytes were constructed. To compare the circulating innate cells from the two primate species, an analysis pipeline combining a clustering analysis by the Spanning-tree Progression Analysis of Density-normalized Events (SPADE) algorithm with a two-step hierarchical clustering of cells nodes and markers was used. Identical SPADE settings were applied to both datasets, except for the 20 clustering markers which slightly differed. A correlation analysis designed to compare the phenotypes of human and macaque cell nodes and based on 16 markers, including 15 shared clustering markers and CD19 for humans or CD20 for macaques, revealed similarities and differences between staining patterns. This study unique by the number of individuals (26 humans and 5 macaques) and the use of mass cytometry certainly contributes to better assess the advantages and limits of the use of non-human primates in preclinical research. © 2017 International Society for Advancement of Cytometry.

Unity in diversity: shared mechanism of entry among paramyxoviruses.

The Paramyxoviridae family includes many viruses that are pathogenic in humans, including parainfluenza viruses, measles virus, respiratory syncytial virus, and the emerging zoonotic Henipaviruses. No effective treatments are currently available for these viruses, and there is a need for efficient antiviral therapies. Paramyxoviruses enter the target cell by binding to a cell surface receptor and then fusing the viral envelope with the target cell membrane, allowing the release of the viral genome into the cytoplasm. Blockage of these crucial steps prevents infection and disease. Binding and fusion are driven by two virus-encoded glycoproteins, the receptor-binding protein and the fusion protein, that together form the viral "fusion machinery." The development of efficient antiviral drugs requires a deeper understanding of the mechanism of action of the Paramyxoviridae fusion machinery, which is still controversial. Here, we review recent structural and functional data on these proteins and the current understanding of the mechanism of the paramyxovirus cell entry process.