A Histo-molecular System to Enhance Lung Transplant Rejection Diagnostics – NANO-LUNG

The NANO-LUNG project is built on a structured methodology involving five complementary work packages (WPs), combining clinical data collection, transcriptomic and spatial profiling, statistical modeling, and scientific coordination. The overall approach leverages clinically validated technologies and robust methods.

WP1 – Establishment of a multicenter cohort of 750 lung transplant recipients
Patients are recruited from three expert French transplant centers based on strict eligibility criteria. Clinical, histological, immunological, and biological data are collected using a shared thesaurus and harmonized protocols. A biobank of 750 FFPE transbronchial biopsies is established.

WP2 – Transcriptomic analysis on FFPE biopsies (NanoString® platform)
RNA is extracted and hybridized on the nCounter® platform using the BHOT panel (770 genes). Raw data are normalized (nSolver, RUVSeq), and differential expression is assessed (negative binomial model, FDR< 0.05). The data will support:

identification of rejection-specific signatures,

development of supervised classifiers (logistic regression, molecular score),

unsupervised analysis to reveal novel phenotypes (archetypal analysis).

WP3 – Digital spatial profiling (GeoMx® DSP)
Fifty representative biopsies are selected. Regions of interest are annotated by expert pathologists. Spatial transcriptomic analysis enables quantification of over 18,000 genes per region. Cell-type deconvolution using reference scRNA-seq datasets identifies immune infiltrates and spatial immune responses.

WP4 – Project coordination, data governance and dissemination
A scientific consortium oversees key milestones, regulatory compliance, deliverables, and dissemination activities targeting the scientific community, patient organizations, and the general public.

The entire project is built on standardized procedures and clinically validated platforms, ensuring high reproducibility, clinical applicability, and potential integration into routine transplant care workflows.

Initial results were obtained from a pilot cohort of 121 FFPE transbronchial biopsies analyzed using the BHOT panel (NanoString®), including 42 cases of antibody-mediated rejection (AMR). Differential gene expression analysis revealed a distinct molecular signature of rejection, characterized by activation of adaptive and innate immune pathways (chemokines, complement, activated macrophages). Interestingly, endothelial transcripts were downregulated, highlighting a lung-specific rejection profile.

Unsupervised archetypal analysis identified four distinct molecular phenotypes, two of which closely matched AMR-like and ACR-like histological patterns. These findings support the use of gene expression data to refine biopsy classification beyond the limits of conventional histological interpretation.

In parallel, the consortium successfully:

developed a unified, deidentified clinical database on 750 lung transplant recipients,

established a standardized protocol for biopsy reassessment and eligibility,

validated the feasibility of transcriptomic analysis on archived FFPE samples,

initiated spatial profiling on 50 representative biopsies.

These results confirm the project’s capacity to deliver robust and reproducible assessments of lung allograft rejection using integrated molecular data. A supervised modeling platform is currently under development and will generate automated molecular rejection scores for each subtype, paving the way for a more precise and personalized classification system.

The NANO-LUNG project lays the foundation for a major transformation in lung transplant rejection diagnostics. By identifying specific molecular signatures and developing automated classification tools, it enables objective, standardized, and reproducible biopsy assessment. This approach breaks away from the current paradigm of histology-based diagnosis, which is prone to variability and imprecision.

In the short term, the goal is to integrate transcriptomic tools into existing clinical workflows. Since the BHOT panel used in this project is already clinically validated, this transition can occur rapidly. A molecular rejection score will support personalized therapeutic decisions, including immunosuppression adjustment and targeted follow-up, based on the graft’s molecular profile.

In the medium term, these tools will improve the design of clinical trials in lung transplantation by enabling more homogeneous patient selection, more precise treatment response definitions, and the use of molecular endpoints. This will facilitate multicenter and international collaborations under harmonized diagnostic frameworks.

In the longer term, the integrated system developed through NANO-LUNG could be extended to other solid organ transplants (kidney, heart), supporting the convergence of precision diagnostic approaches across transplantation medicine. The project also contributes to the development of a shared European database (ICDOT), which will serve as a reference for future diagnostic standards.

Lastly, the collaborative dynamic created between clinicians, molecular biologists, statisticians, and data scientists provides a robust foundation for future initiatives in precision medicine for respiratory and immunological diseases. NANO-LUNG thus paves the way for a more integrated, predictive, and personalized approach to lung transplant care.

N/A

Despite significant advancements in understanding alloimmune responses, rejection continues to be the primary determinant influencing the longevity of lung allografts. The complexity of histopathologic analysis poses a major challenge in accurately identifying alloimmune-mediated complications, causing histology to be an imperfect gold standard in diagnosing rejection. The integration of formalin-fixed paraffin-embedded-based molecular diagnostics alongside conventional histology may improve the current empirical diagnostic system.
To address the unmet need for precision diagnostics, the current proposal aims to enhance lung transplant rejection diagnostics and increase the level of phenotyping of pulmonary rejection using a tissue-based molecular approach.
In order to achieve this goal, we will build a highly characterized population-based cohort of lung transplant patients incorporating 750 transbronchial biopsies from three leading French transplant centers. Each biopsy will be contextualized by means of clinical, biological and immunological data. We will sequence and spatially resolve the molecular immune landscape of pulmonary rejection in formalin-fixed paraffin-embedded transbronchial biopsies using a two-tier approach, comprising both bulk transcriptomics and spatial molecular profiling. These complimentary, cutting-edge technologies will precisely identify the molecular signature corresponding to the subtypes of rejection, allowing us to develop gene-expression based classifiers which will provide a more accurate, standardized and reproducible diagnose of cellular and humoral rejection, than conventional histopathology. In addition, integrating gene expression data in unsupervised probabilistic algorithms will enable us to discover novel molecular, clinical entities and correct errors in diagnostic nosology.
By merging the clinical experience of Foch hospital and scientific expertise of INSERM U970, the present proposal will establish the groundwork for the development of integrative diagnostic systems which will facilitate the adaptation of gene expression profiling in routine clinical care.