Biology subjects

Spilman, P.

Publications and source records attributed to Spilman, P..

3 recordsLinked to original sources

bioRxiv2021

A deep learning-based iterative digital pathology annotation tool

Well-annotated exemplars are an important prerequisite for supervised deep learning schemes. Unfortunately, generating these annotations is a cumbersome and laborious process, due to the large amount of time and effort needed. Here we present a deep-learning-based iterative digital pathology annotation tool that is both easy to use by pathologists and easy to integrate into machine vision systems. Our pathology image annotation tool greatly reduces annotation time from hours to a few minutes, while maintaining high fidelity with human-expert manual annotations. Here we demonstrate that our active learning tool can be used for a variety of pathology annotation tasks including masking tumor, stroma, and lymphocyte-rich regions, among others. This annotation automation system was validated on 90 unseen digital pathology images with tumor content from the CAMELYON16 database and it was found that pathologists gold standard masks were re-produced successfully using our tool. That is, an average of 2.7 positive selections (mouse clicks) and 8.0 negative selections (mouse clicks) were sufficient to generate tumor masks similar to pathologists gold standard in CAMELYON16 test WSIs. Furthermore, the developed image annotation tool has been used to build gold standard masks for hundreds of TCGA digital pathology images. This set was used to train a convolutional neural network for identification of tumor epithelium. The developed pan-cancer deep neural network was then tested on TCGA and internal data with comparable performance. The validated pathology image annotation tool described herein has the potential to be of great value in facilitating accurate, rapid pathological analysis of tumor biopsies.

pathology↗

bioRxiv2020

Complete Protection of Nasal and Lung Airways Against SARS-CoV-2 Challenge by Antibody Plus Th1 Dominant N- and S-Specific T-Cell Responses to Subcutaneous Prime and Thermally-Stable Oral Boost Bivalent hAd5 Vaccination in an NHP Study

We have developed a dual-antigen COVID-19 vaccine incorporating genes for a modified SARS-CoV-2 spike (S-Fusion) protein and the viral nucleocapsid (N) protein with an Enhanced T-cell Stimulation Domain (N-ETSD) with the potential to increase MHC class I/II responses. The adenovirus serotype 5 platform used, hAd5 [E1-, E2b-, E3-], previously demonstrated to be effective in the presence of Ad immunity, can be delivered in an oral formulation that overcomes cold-chain limitations. The hAd5 S-Fusion + N-ETSD vaccine was evaluated in rhesus macaques showing that a subcutaneous prime followed by oral boosts elicited both humoral and Th1 dominant T-cell responses to both S and N that protected the upper and lower respiratory tracts from high titer (1 x 106 TCID50) SARS-CoV-2 challenge. Notably, viral replication was inhibited within 24 hours of challenge in both lung and nasal passages, becoming undetectable within 7 days post-challenge. ONE SENTENCE SUMMARYhAd5 spike + nucleocapsid SC prime/oral boost vaccine elicits humoral and T-cell responses and protects rhesus macaques from SARS-CoV-2 challenge.

immunology↗

bioRxiv2020

A Next Generation Bivalent Human Ad5 COVID-19 Vaccine Delivering Both Spike and Nucleocapsid Antigens Elicits Th1 Dominant CD4+, CD8+ T-cell and Neutralizing Antibody Responses

In response to the health crisis presented by the COVID-19 pandemic, rapid development of safe and effective vaccines that elicit durable immune responses is imperative. Recent reports have raised the concern that antibodies in COVID-19 convalescent patients may not be long lasting and thus even these individuals may require vaccination. Vaccine candidates currently in clinical testing have focused on the SARS-CoV-2 wild type spike (S) protein (S-WT) as the major antigen of choice and while pre-clinical and early clinical testing have shown that S elicits an antibody response, we believe the optimal vaccine candidate should be capable of inducing robust, durable T-cell responses as well as humoral responses. We report here on a next generation bivalent human adenovirus serotype 5 (hAd5) vaccine capable of inducing immunity in patients with pre-existing adenovirus immunity, comprising both an S sequence optimized for cell surface expression (S-Fusion) and a conserved nucleocapsid (N) antigen designed to be transported to the endosomal subcellular compartment, with the potential to generate durable immune protection. Our studies suggest that this bivalent vaccine is optimized for immunogenicity as evidenced by the following findings: (i) The optimized S-Fusion displayed improved S receptor binding domain (RBD) cell surface expression compared to S-WT where little surface expression was detected; (ii) the expressed RBD from S-Fusion retained conformational integrity and recognition by ACE2-Fc; (iii) the viral N protein modified with an enhanced T-cell stimulation domain (ETSD) localized to endosomal/lysosomal subcellular compartments for MHC I/II presentation; and (iv) these optimizations to S and N (S-Fusion and N-ETSD) generated enhanced de novo antigen-specific B cell and CD4+ and CD8+ T-cell responses in antigen-naive pre-clinical models. Both the T-cell and antibody immune responses to S and N demonstrated a T-helper 1 (Th1) bias. The antibody responses were neutralizing as demonstrated by two independent SARS-CoV-2 neutralization assays. Based on these findings, we are advancing this next generation bivalent hAd5 S-Fusion + N-ETSD vaccine as our lead clinical candidate to test for its ability to provide robust, durable cell-mediated and humoral immunity against SARS-CoV-2 infection. Further studies are ongoing to explore utilizing this vaccine construct in oral, intranasal, and sublingual formulations to induce mucosal immunity in addition to cell-mediated and humoral immunity. The ultimate goal of an ideal COVID-19 vaccine is to generate long-term T and B cell memory.

immunology↗