A specific biomarker that can separate active renal vasculitis from other causes of renal dysfunction is lacking, with a kidney biopsy often being required. Soluble CD163 (sCD163), shed by monocytes and macrophages, has been reported as a potential biomarker in diseases associated with excessive macrophage activation. Thus, we hypothesized that urinary sCD163 shed by crescent macrophages correlates with active glomerular inflammation. We detected sCD163 in rat urine early in the disease course of experimental vasculitis. Moreover, microdissected glomeruli from patients with small vessel vasculitis (SVV) had markedly higher levels of CD163 mRNA than did those from patients with lupus nephritis, diabetic nephropathy, or nephrotic syndrome. Both glomeruli and interstitium of patients with SVV strongly expressed CD163 protein. In 479 individuals, including patients with SVV, disease controls, and healthy controls, serum levels of sCD163 did not differ between the groups. However, in an inception cohort, including 177 patients with SVV, patients with active renal vasculitis had markedly higher urinary sCD163 levels than did patients in remission, disease controls, or healthy controls. Analyses in both internal and external validation cohorts confirmed these results. Setting a derived optimum cutoff for urinary sCD163 of 0.3 ng/mmol creatinine for detection of active renal vasculitis resulted in a sensitivity of 83%, specificity of 96%, and a positive likelihood ratio of 20.8. These data indicate that urinary sCD163 level associates very tightly with active renal vasculitis, and assessing this level may be a noninvasive method for diagnosing renal flare in the setting of a known diagnosis of SVV.

Class II Major Histocompatibility Complex (MHC) molecules have an open-ended binding groove which can accommodate peptides of varying lengths. Several studies have demonstrated that peptide flanking residues (PFRs) which lie outside the core binding groove can influence peptide binding and T cell recognition. By using data from the AntiJen database we were able to characterise systematically the influence of PFRs on peptide affinity for MHC class II molecules. By analysing 1279 peptide elongation events covering 19 distinct HLA alleles it was observed that, in general, peptide elongation resulted in increased MHC class II molecule affinity. It was also possible to determine an optimal peptide length for MHC class II affinity of approximately 18-20 amino acids; elongation of peptides beyond this length resulted in a null or negative effect on affinity. The observed relationship between peptide length and MHC class II affinity has significant implications for the design of vaccines and the study of the epitopic basis of immunological disease.

RET gene fusions are established oncogenic drivers in 1% of NSCLC. Accurate detection of advanced patients with RET fusions is essential to ensure optimal therapy choice. We investigated the performance of fluorescence in situ hybridization (FISH) as a diagnostic test for detecting functional RET fusions. Between January 2016 and November 2019, a total of 4873 patients with NSCLC were routinely screened for RET fusions using either FISH (n = 2858) or targeted RNA next-generation sequencing (NGS) (n = 2015). If sufficient material was available, positive cases were analyzed by both methods (n = 39) and multiple FISH assays (n = 17). In an independent cohort of 520 patients with NSCLC, whole-genome sequencing data were investigated for disruptive structural variations and functional fusions in the RET and compared with ALK and ROS1 loci. FISH analysis revealed RET rearrangement in 48 of 2858 cases; of 30 rearranged cases double tested with NGS, only nine had a functional RET fusion. RNA NGS yielded RET fusions in 14 of 2015 cases; all nine cases double tested by FISH had RET locus rearrangement. Of these 18 verified RET fusion cases, 16 had a split signal and two a complex rearrangement by FISH. By whole-genome sequencing, the prevalence of functional fusions compared with all disruptive events was lower in the RET (4 of 9, 44%) than the ALK (27 of 34, 79%) and ROS1 (9 of 12, 75%) loci. FISH is a sensitive but unspecific technique for RET screening, always requiring a confirmation using an orthogonal technique, owing to frequently occurring RET rearrangements not resulting in functional fusions in NSCLC.

Advanced diagnostics are enabling cancer treatments to become increasingly tailored to the individual through developments in immunotherapies and targeted therapies. However, long turnaround times and high costs of molecular testing hinder the widespread implementation of targeted cancer treatments. Meanwhile, gold-standard histopathological assessment carried out by a trained pathologist is widely regarded as routine and mandatory in most cancers. Recently, methods have been developed to mine hidden information from histopathological slides using deep learning applied to scanned and digitized slides; deep learning comprises a collection of computational methods which learn patterns in data in order to make predictions. Such methods have been reported to be successful in a variety of cancers for predicting the presence of biomarkers such as driver mutations, tumour mutational burden, and microsatellite instability. This information could prove valuable to pathologists and oncologists in clinical decision making for cancer treatment and triage for in-depth sequencing. In addition to identifying molecular features, deep learning has been applied to predict prognosis and treatment response in certain cancers. Despite reported successes, many challenges remain before the clinical implementation of such diagnostic strategies in the clinical setting is possible. This review aims to outline recent developments in the field of deep learning for predicting molecular genetics from histopathological slides, as well as to highlight limitations and pitfalls of working with histopathology slides in deep learning.