A simple point-of-care COVID-19 test

Researchers at the University of Illinois, Urbana-Champaign, have developed a prototype of a rapid molecular test for COVID-19, as well as a portable, user-friendly instrument for seeing the results on a smartphone in 30 minutes. The test, which utilizes a 3D-printed, microfluidic cartridge, can be performed at the point of care without the need to send samples to a lab. Unlike the reverse transcription polymerase chain reaction (RT-PCR), which requires specialized equipment and controls, the team used a process called loop-mediated isothermal amplification, or LAMP, which doesn’t require RNA extraction and purification. They compared their assay with PCR using synthetic nasal fluid spiked with virus and with clinical samples and got results in line with PCR results. When tested on 20 patient samples, the assay demonstrated 100% accuracy, sensitivity and specificity. The instrument was constructed from commercially available materials and a housing that was easily produced with a consumer-grade 3D printer. “The POC instrument is designed for low cost, accessibility, and the potential for scale-up,” the authors wrote. “Because the entire assay can be conducted within the cartridge, the principle of operation is very simple and can be performed with minimal training.” Their work appeared online Aug. 31, 2020, in Proceedings of the National Academy of Sciences.

A deep-learning method to predict AMD risk

Age-related macular degeneration (AMD) is the leading cause of blindness in developed countries, and aging demographics put it on track to hit 288 million people globally by 2040. Deep learning has shown potential in diagnosing and screening AMD using color fundus photographs, but has not been good at predicting individuals’ risk of late AMD, the stage where severe visual loss occurs. Researchers at the National Institutes of Health developed, trained and validated a framework for predicting a person’s risk of late AMD by combining deep learning and survival analysis. The method produces autonomous predictions with greater accuracy than retinal specialists using clinical standards. They built the deep-learning approach using over 80,000 images (3,298 participants) from the Age-Related Eye Disease Studies ARED2 and AREDS2, the largest longitudinal studies in AMD. “When validated against an independent test data set of 601 participants, our model achieved high prognostic accuracy (5-year C-statistic 86.4 (95% confidence interval 86.2-86.6) that substantially exceeded that of retinal specialists using two existing clinical standards (81.3 (81.1-81.5) and 82.0 (81.8-82.3), respectively,” the authors wrote. “Not only did its accuracy meet and surpass existing clinical standards, but additional strengths in clinical settings include risk ascertainment above 50% and without genotype data,” they said. The approach is “likely to be highly generalizable, given the breadth of training data from 82 US retinal specialty clinics” and demonstrates the potential of deep learning to improve clinical decisionmaking for AMD patients. Their study was published online Aug. 27, 2020, in npj: digital medicine.

AP-1 and antidepressant action

Scientists at Rockefeller University have identified a gene expression pattern set off by activator protein 1 (AP-1) as a predictor of a response to selective serotonin reuptake inhibitors (SSRIs). The antidepressant mechanism of action of SSRIs has long been a mystery, as rising serotonin levels, the supposed molecular cause, occur weeks before any improvements in mood in patients. “The AP-1 transcriptional program modulates the expression of key neuronal remodeling genes … linking neuronal plasticity to the antidepressant response,” the authors wrote. “We find that AP-1 function is required for the antidepressant effect in vivo,” pointing to strategies to induce or potentiate antidepressant responses. Their study appeared in the Aug. 13, 2020, print issue of Molecular Psychiatry after earlier publication online.

Orasure collection device included in Miradx EUA

Orasure Technologies Inc., of Bethlehem, Pa., said its Oracollect∙RNA device was included with other devices in the U.S. FDA emergency use authorization (EUA) granted to Los Angeles-based Miradx Inc. This is the fifth EUA to include a collection device from the company’s DNA Genotek subsidiary. Miradx will use Oracollect∙RNA to collect oropharyngeal samples in its COVID-19 testing program for essential workers and first responders.