Faculty Bibliography

High-dimensional multiplexed imaging can reveal the spatial organization of tumour tissues at the molecular level. However, owing to the scale and information complexity of the imaging data, it is challenging to discover and thoroughly characterize the heterogeneity of tumour microenvironments. Here we show that self-supervised representation learning on data from imaging mass cytometry can be leveraged to distinguish morphological differences in tumour microenvironments and to precisely characterize distinct microenvironment signatures. We used self-supervised masked image modelling to train a vision transformer that directly takes high-dimensional multiplexed mass-cytometry images. In contrast with traditional spatial analyses relying on cellular segmentation, the vision transformer is segmentation-free, uses pixel-level information, and retains information on the local morphology and biomarker distribution. By applying the vision transformer to a lung-tumour dataset, we identified and validated a monocytic signature that is associated with poor prognosis.

Cancer of unknown primary (CUP) constitutes between 2% and 5% of human malignancies and is among the most common causes of cancer death in the United States. Brain metastases are often the first clinical presentation of CUP; despite extensive pathological and imaging studies, 20%-45% of CUP are never assigned a primary site. DNA methylation array profiling is a reliable method for tumor classification but tumor-type-specific classifier development requires many reference samples. This is difficult to accomplish for CUP as many cases are never assigned a specific diagnosis. Recent studies identified subsets of methylation quantitative trait loci (mQTLs) unique to specific organs, which could help increase classifier accuracy while requiring fewer samples. We performed a retrospective genome-wide methylation analysis of 759 carcinoma samples from formalin-fixed paraffin-embedded tissue samples using Illumina EPIC array. Utilizing mQTL specific for breast, lung, ovarian/gynecologic, colon, kidney, or testis (BLOCKT) (185k total probes), we developed a deep learning-based methylation classifier that achieved 93.12% average accuracy and 93.04% average F1-score across a 10-fold validation for BLOCKT organs. Our findings indicate that our organ-based DNA methylation classifier can assist pathologists in identifying the site of origin, providing oncologists insight on a diagnosis to administer appropriate therapy, improving patient outcomes.

Over the last decade, Hippo signaling has emerged as a major tumor-suppressing pathway. Its dysregulation is associated with abnormal expression of YAP1 and TEAD-family genes. Recent works have highlighted the role of YAP1/TEAD activity in several cancers and its potential therapeutic implications. Therefore, identifying patients with a dysregulated Hippo pathway is key to enhancing treatment impact. Although recent studies have derived RNA-seq-based signatures, there remains a need for a reproducible and cost-effective method to measure the pathway activation. In recent years, deep learning applied to histology slides have emerged as an effective way to predict molecular information from a data modality available in clinical routine. Here, we trained models to predict YAP1/TEAD activity from H&E-stained histology slides in multiple cancers. The robustness of our approach was assessed in seven independent validation cohorts. Finally, we showed that histological markers of disease aggressiveness were associated with dysfunctional Hippo signaling.

Small cell lung carcinoma (SCLC) is a highly aggressive malignancy that is typically associated with tobacco exposure and inactivation of RB1 and TP53 genes. Here, we performed detailed clinicopathologic, genomic, and transcriptomic profiling of an atypical subset of SCLC that lacked RB1 and TP53 co-inactivation and arose in never/light smokers. We found that most cases were associated with chromothripsis-massive, localized chromosome shattering-recurrently involving chromosome 11 or 12 and resulting in extrachromosomal amplification of CCND1 or co-amplification of CCND2/CDK4/MDM2, respectively. Uniquely, these clinically aggressive tumors exhibited genomic and pathologic links to pulmonary carcinoids, suggesting a previously uncharacterized mode of SCLC pathogenesis via transformation from lower-grade neuroendocrine tumors or their progenitors. Conversely, SCLC in never-smokers harboring inactivated RB1 and TP53 exhibited hallmarks of adenocarcinoma-to-SCLC derivation, supporting two distinct pathways of plasticity-mediated pathogenesis of SCLC in never-smokers. Significance: Here, we provide the first detailed description of a unique SCLC subset lacking RB1/TP53 alterations and identify extensive chromothripsis and pathogenetic links to pulmonary carcinoids as its hallmark features. This work defines atypical SCLC as a novel entity among lung cancers, highlighting its exceptional histogenesis, clinicopathologic characteristics, and therapeutic vulnerabilities. See related commentary by Nadeem and Drapkin, p. 8.

Sjögren's disease (SjD) is an autoimmune disorder characterized by progressive salivary and lacrimal gland dysfunction, inflammation, and destruction, as well as extraglandular manifestations. SjD is associated with autoreactive B and T cells, but its pathophysiology remains incompletely understood. Abnormalities in regulatory T (T_reg) cells occur in several autoimmune diseases, but their role in SjD is ambiguous. We had previously shown that the function and development of T_reg cells depend on store-operated Ca²⁺ entry (SOCE), which is mediated by ORAI1 Ca²⁺ channels and stromal interaction protein 1 (STIM1) and STIM2. Here, we show that mice with a Foxp3⁺ T_reg cell-specific deletion of Stim1 and Stim2 develop a phenotype that fulfills all classification criteria of human SjD. Mutant mice have salivary and lacrimal gland inflammation characterized by strong lymphocyte infiltration and transcriptional signatures dominated by T helper 1 (T_H1) and interferon (IFN) signaling. CD4⁺ T cells from mutant mice are sufficient to induce SjD-like disease in an IFN-γ-dependent manner. Inhibition of IFN signaling with the JAK1/2 inhibitor baricitinib alleviated CD4⁺ T cell-induced SjD in mice. These findings are consistent with the transcriptional profiles of CD4⁺ T cells from patients with SjD, which indicate enhanced T_H1 but reduced memory T_reg cell function. Together, our study provides evidence for a critical role of dysfunctional T_reg cells and IFN-γ-producing T_H1 cells in the pathogenesis of SjD.

CONTEXT.—/UNASSIGNED:Adenocarcinomas are the most common histologic subtype of lung cancer, and exist within a widely divergent clinical, radiologic, molecular, and histologic spectrum. There is a strong association between histologic patterns and prognosis that served as the basis for a recently described grading system. As the study of molecular pathology rapidly evolves, all targetable mutations so far have been found in adenocarcinomas, thus requiring accurate diagnosis and classification for triage of molecular alterations and adequate therapy. OBJECTIVE.—/UNASSIGNED:To discuss the rationale for adenocarcinoma classifications within the 2021 5th edition of the World Health Organization, with a focus on nonmucinous tumors, including tumor grading and biopsy/cytology diagnosis. DATA SOURCES.—/UNASSIGNED:PubMed search. CONCLUSIONS.—/UNASSIGNED:A grading system for adenocarcinoma has improved prognostic impact of the classification of pulmonary adenocarcinoma. An accurate diagnosis of adenocarcinoma in small biopsy material is important for tissue triage for molecular studies and ultimately for patient management and treatment.

Pathogen encounter can result in epigenetic remodeling that shapes disease caused by heterologous pathogens. Here, we examined innate immune memory in the context of commonly circulating respiratory viruses. Single-cell analyses of airway-resident immune cells in a disease-relevant murine model of SARS-CoV-2 recovery revealed epigenetic reprogramming in alveolar macrophages following infection. Post-COVID-19 human monocytes exhibited similar epigenetic signatures. In airway-resident macrophages, past SARS-CoV-2 infection increased activity of type I interferon (IFN-I)-related transcription factors and epigenetic poising of antiviral genes. Viral pattern recognition and canonical IFN-I signaling were required for the establishment of this innate immune memory and augmented secondary antiviral responses. Antiviral innate immune memory mounted by airway-resident macrophages post-SARS-CoV-2 was necessary and sufficient to ameliorate secondary disease caused by influenza A virus and curtailed hyperinflammatory dysregulation and mortality. Our findings provide insights into antiviral innate immune memory in the airway that may facilitate the development of broadly effective therapeutic strategies.

BACKGROUND:Risk of early-stage lung adenocarcinoma (LUAD) recurrence after surgical resection is significant, and post-recurrence median survival is approximately two years. Currently there are no commercially available biomarkers that predict recurrence. Here, we investigated whether microbial and host genomic signatures in the lung can predict recurrence. METHODS:In 91 early-stage (Stage IA/IB) LUAD-patients with extensive follow-up, we used 16s rRNA gene sequencing and host RNA-sequencing to map the microbial and host transcriptomic landscape in tumor and adjacent unaffected lung samples. RESULTS:23 out of 91 subjects had tumor recurrence over 5-year period. In tumor samples, LUAD recurrence was associated with enrichment with Dialister, Prevotella, while in unaffected lung, recurrence was associated with enrichment with Sphyngomonas and Alloiococcus. The strengths of the associations between microbial and host genomic signatures with LUAD recurrence were greater in adjacent unaffected lung samples than in the primary tumor. Among microbial-host features in the unaffected lung samples associated with recurrence, enrichment with Stenotrophomonas geniculata and Chryseobacterium were positively correlated with upregulation of IL-2, IL-3, IL-17, EGFR, HIF-1 signaling pathways among the host transcriptome. In tumor samples, enrichment with Veillonellaceae Dialister, Ruminococcacea, Haemophilus Influenza, and Neisseria were positively correlated with upregulation of IL-1, IL-6, IL17, IFN, and Tryptophan metabolism pathways. CONCLUSIONS:Overall, modeling suggested that a combined microbial/transcriptome approach using unaffected lung samples had the best biomarker performance (AUC=0.83). IMPACT/CONCLUSIONS:This study suggests that LUAD recurrence is associated with distinct pathophysiological mechanisms of microbial-host interactions in the unaffected lung rather than those present in the resected tumor.

PURPOSE/OBJECTIVE:Patients with stage II and III cutaneous primary melanoma vary considerably in their risk of melanoma-related death. We explore the ability of methylation profiling to distinguish primary melanoma methylation classes and their associations with clinicopathologic characteristics and survival. MATERIALS AND METHODS/METHODS:InterMEL is a retrospective case-control study that assembled primary cutaneous melanomas from American Joint Committee on Cancer (AJCC) 8th edition stage II and III patients diagnosed between 1998 and 2015 in the United States and Australia. Cases are patients who died of melanoma within 5 years from original diagnosis. Controls survived longer than 5 years without evidence of melanoma recurrence or relapse. Methylation classes, distinguished by consensus clustering of 850K methylation data, were evaluated for their clinicopathologic characteristics, 5-year survival status, and differentially methylated gene sets. RESULTS:= .007). CIMP and IM had a 2-fold higher likelihood of 5-year death from melanoma than LM (CIMP odds ratio [OR], 2.16 [95% CI, 1.18 to 3.96]; IM OR, 2.00 [95% CI, 1.12 to 3.58]) in a multivariable model adjusted for age, sex, log Breslow thickness, ulceration, mitotic index, and N stage. Despite more extensive CpG island hypermethylation in CIMP, CIMP and IM shared similar patterns of differential methylation and gene set enrichment compared with LM. CONCLUSION/CONCLUSIONS:Melanoma MethylClasses may provide clinical value in predicting 5-year death from melanoma among patients with primary melanoma independent of other clinicopathologic factors.

PURPOSE/UNASSIGNED:Immune checkpoint inhibition (ICI) shows benefits in adjuvant (AT) and neoadjuvant melanoma treatments. However, ICI frequently induces severe immune-related adverse events (irAE). Unlike metastatic disease, in which irAEs are a clinical trade-off for treatment that improves survival, the toxicity burden from ICI in the AT setting is a substantial clinical problem urging for irAE-predictive biomarkers. EXPERIMENTAL DESIGN/UNASSIGNED:We assessed postsurgical, pre-ICI treatment peripheral CD4+ and CD8+ T cells from clinical trial patients (CheckMate 915) treated with AT nivolumab (n = 130) or ipilimumab/nivolumab (COMBO, n = 82). Performing RNA sequencing differential gene expression analysis, we tested baseline differences associated with severe (grades 3-5) irAEs and constructed an irAE-predictive model using least absolute shrinkage and selection operator-regularized logistic regression. RESULTS/UNASSIGNED:The analysis of predicted protein-protein interactions among differentially expressed genes in peripheral CD4+ cells revealed significant enrichment of the spleen tyrosine kinase (SYK) pathway, associated with severe irAEs in COMBO-treated patients. This gene expression signature predicted severe-irAE COMBO patients (χ2P value = 0.001) with 73% accuracy and was independent of disease recurrence (P = 0.79). The irAE-predictive model incorporating this gene expression signature demonstrated 82% accuracy (χ2P value = 8.91E-06). CONCLUSIONS/UNASSIGNED:We identified baseline gene expression differences in key immune pathways of peripheral blood T cells from COMBO-treated patients with grades 3 to 5 irAEs and defined a SYK-related gene signature correctly identifying ∼60% of COMBO-treated patients with grades 3 to 5 irAEs. This finding aligns with our previous work linking anti-CTLA4 irAEs with a germline variant associated with high SYK expression. This gene signature may serve as a baseline biomarker of severe grade 3 to 5 irAE risk, which is especially important in AT treatment.