Objective: The goal of this study is to use adjunctive classes to improve a predictive model whose performance is limited by the common problems of small numbers of primary cases, high feature dimensionality, and poor class separability. Specifically, our clinical task is to use mammographic features to predict whether ductal carcinoma in situ (DCIS) identified at needle core biopsy will be later upstaged or shown to contain invasive breast cancer.
Methods: To improve the prediction of pure DCIS (negative) versus upstaged DCIS (positive) cases, this study considers the adjunctive roles of two related classes: atypical ductal hyperplasia (ADH), a non-cancer type of breast abnormity, and invasive ductal carcinoma (IDC), with 113 computer vision based mammographic features extracted from each case. To improve the baseline Model A's classification of pure vs. upstaged DCIS, we designed three different strategies (Models B, C, D) with different ways of embedding features or inputs.
Results: Based on ROC analysis, the baseline Model A performed with AUC of 0.614 (95% CI, 0.496-0.733). All three new models performed better than the baseline, with domain adaptation (Model D) performing the best with an AUC of 0.697 (95% CI, 0.595-0.797).
Conclusion: We improved the prediction performance of DCIS upstaging by embedding two related pathology classes in different training phases.
Significance: The three new strategies of embedding related class data all outperformed the baseline model, thus demonstrating not only feature similarities among these different classes, but also the potential for improving classification by using other related classes.
The biological importance and varied metabolic capabilities of specific microbial strains have long been established in the scientific community. Strains have, in the past, been largely defined and characterized based on microbial isolates. However, the emergence of new technologies and techniques has enabled assessments of their ecology and phenotypes within microbial communities and the human microbiome. While it is now more obvious how pathogenic strain variants are detrimental to human health, the consequences of subtle genetic variation in the microbiome have only recently been exposed. Here, we review the operational definitions of strains (e.g., genetic and structural variants) as they can now be identified from microbial communities using different high-throughput, often culture-independent techniques. We summarize the distribution and diversity of strains across the human body and their emerging links to health maintenance, disease risk and progression, and biochemical responses to perturbations, such as diet or drugs. We list methods for identifying, quantifying, and tracking strains, utilizing high-throughput sequencing along with other molecular and “culturomics” technologies. Finally, we discuss implications of population studies in bridging experimental gaps and leading to a better understanding of the health effects of strains in the human microbiome.
Antitumoral immunity requires organized, spatially nuanced interactions between components of the immune tumor microenvironment (iTME). Understanding this coordinated behavior in effective versus ineffective tumor control will advance immunotherapies. We re-engineered co-detection by indexing (CODEX) for paraffin-embedded tissue microarrays, enabling simultaneous profiling of 140 tissue regions from 35 advanced-stage colorectal cancer (CRC) patients with 56 protein markers. We identified nine conserved, distinct cellular neighborhoods (CNs)—a collection of components characteristic of the CRC iTME. Enrichment of PD-1+CD4+ T cells only within a granulocyte CN positively correlated with survival in a high-risk patient subset. Coupling of tumor and immune CNs, fragmentation of T cell and macrophage CNs, and disruption of inter-CN communication was associated with inferior outcomes. This study provides a framework for interrogating how complex biological processes, such as antitumoral immunity, occur through concerted actions of cells and spatial domains.
Metastasis constitutes the primary cause of cancer-related deaths, with the lung being a commonly affected organ. We found that activation of lung-resident group 2 innate lymphoid cells (ILC2s) orchestrated suppression of natural killer (NK) cell-mediated innate antitumor immunity, leading to increased lung metastases and mortality. Using multiple models of lung metastasis, we show that interleukin (IL)-33-dependent ILC2 activation in the lung is involved centrally in promoting tumor burden. ILC2-driven innate type 2 inflammation is accompanied by profound local suppression of interferon-γ production and cytotoxic function of lung NK cells. ILC2-dependent suppression of NK cells is elaborated via an innate regulatory mechanism, which is reliant on IL-5-induced lung eosinophilia, ultimately limiting the metabolic fitness of NK cells. Therapeutic targeting of IL-33 or IL-5 reversed NK cell suppression and alleviated cancer burden. Thus, we reveal an important function of IL-33 and ILC2s in promoting tumor metastasis via their capacity to suppress innate type 1 immunity.
Purpose: Although high T-cell density is a well-established favorable prognostic factor in colorectal cancer, the prognostic significance of tumor-associated plasma cells, neutrophils, and eosinophils is less well-defined.
Experimental Design: We computationally processed digital images of hematoxylin and eosin (H&E)–stained sections to identify lymphocytes, plasma cells, neutrophils, and eosinophils in tumor intraepithelial and stromal areas of 934 colorectal cancers in two prospective cohort studies. Multivariable Cox proportional hazards regression was used to compute mortality HR according to cell density quartiles. The spatial patterns of immune cell infiltration were studied using the GTumor:Immune cell function, which estimates the likelihood of any tumor cell in a sample having at least one neighboring immune cell of the specified type within a certain radius. Validation studies were performed on an independent cohort of 570 colorectal cancers.
Results: Immune cell densities measured by the automated classifier demonstrated high correlation with densities both from manual counts and those obtained from an independently trained automated classifier (Spearman's ρ 0.71–0.96). High densities of stromal lymphocytes and eosinophils were associated with better cancer-specific survival [Ptrend < 0.001; multivariable HR (4th vs 1st quartile of eosinophils), 0.49; 95% confidence interval, 0.34–0.71]. High GTumor:Lymphocyte area under the curve (AUC0,20μm; Ptrend = 0.002) and high GTumor:Eosinophil AUC0,20μm (Ptrend < 0.001) also showed associations with better cancer-specific survival. High stromal eosinophil density was also associated with better cancer-specific survival in the validation cohort (Ptrend < 0.001).
Conclusions: These findings highlight the potential for machine learning assessment of H&E-stained sections to provide robust, quantitative tumor-immune biomarkers for precision medicine.