Clinical Validation of Digitally Acquired Clinical Data and Machine Learning Models for Remote Measurement of Psoriasis and Psoriatic Arthritis: A Proof-of-Concept Study.

OBJECTIVE: Psoriatic disease remains underdiagnosed and undertreated. We developed and validated a suite of novel, sensor-based smartphone assessments (Psorcast app) that can be self-administered to measure cutaneous and musculoskeletal signs and symptoms of psoriatic disease. METHODS: Participants with psoriasis (PsO) or psoriatic arthritis (PsA) and healthy controls were recruited between June 5, 2019, and November 10, 2021, at 2 academic medical centers. Concordance and accuracy of digital measures and image-based machine learning models were compared to their analogous clinical measures from trained rheumatologists and dermatologists. RESULTS: Of 104 study participants, 51 (49%) were female and 53 (51%) were male, with a mean age of 42.3 years (SD 12.6). Seventy-nine (76%) participants had PsA, 16 (15.4%) had PsO, and 9 (8.7%) were healthy controls. Digital patient assessment of percent body surface area (BSA) affected with PsO demonstrated very strong concordance (Lin concordance correlation coefficient [CCC] 0.94 [95% CI 0.91-0.96]) with physician-assessed BSA. The in-clinic and remote target plaque physician global assessments showed fair-to-moderate concordance (CCCerythema 0.72 [0.59-0.85]; CCCinduration 0.72 [0.62-0.82]; CCCscaling 0.60 [0.48-0.72]). Machine learning models of hand photos taken by patients accurately identified clinically diagnosed nail PsO with an accuracy of 0.76. The Digital Jar Open assessment categorized physician-assessed upper extremity involvement, considering joint tenderness or enthesitis (AUROC 0.68 [0.47-0.85]). CONCLUSION: The Psorcast digital assessments achieved significant clinical validity, although they require further validation in larger cohorts before use in evidence-based medicine or clinical trial settings. The smartphone software and analysis pipelines from the Psorcast suite are open source and freely available.

Deep Learning-Based Psoriasis Assessment: Harnessing Clinical Trial Imaging for Accurate Psoriasis Area Severity Index Prediction.

Introduction: Image-based machine learning holds great promise for facilitating clinical care; however, the datasets often used for model training differ from the interventional clinical trial-based findings frequently used to inform treatment guidelines. Here, we draw on longitudinal imaging of psoriasis patients undergoing treatment in the Ultima 2 clinical trial (NCT02684357), including 2,700 body images with psoriasis area severity index (PASI) annotations by uniformly trained dermatologists. Methods: An image-processing workflow integrating clinical photos of multiple body regions into one model pipeline was developed, which we refer to as the "One-Step PASI" framework due to its simultaneous body detection, lesion detection, and lesion severity classification. Group-stratified cross-validation was performed with 145 deep convolutional neural network models combined in an ensemble learning architecture. Results: The highest-performing model demonstrated a mean absolute error of 3.3, Lin's concordance correlation coefficient of 0.86, and Pearson correlation coefficient of 0.90 across a wide range of PASI scores comprising disease classifications of clear skin, mild, and moderate-to-severe disease. Within-person, time-series analysis of model performance demonstrated that PASI predictions closely tracked the trajectory of physician scores from severe to clear skin without systematically over- or underestimating PASI scores or percent changes from baseline. Conclusion: This study demonstrates the potential of image processing and deep learning to translate otherwise inaccessible clinical trial data into accurate, extensible machine learning models to assess therapeutic efficacy.

Smartphone-Based VO2max Measurement With Heart Snapshot in Clinical and Real-world Settings With a Diverse Population: Validation Study.

BACKGROUND: Maximal oxygen consumption (VO2max) is one of the most predictive biometrics for cardiovascular health and overall mortality. However, VO2max is rarely measured in large-scale research studies or routine clinical care because of the high cost, participant burden, and requirement for specialized equipment and staff. OBJECTIVE: To overcome the limitations of clinical VO2max measurement, we aim to develop a digital VO2max estimation protocol that can be self-administered remotely using only the sensors within a smartphone. We also aim to validate this measure within a broadly representative population across a spectrum of smartphone devices. METHODS: Two smartphone-based VO2max estimation protocols were developed: a 12-minute run test (12-MRT) based on distance measured by GPS and a 3-minute step test (3-MST) based on heart rate recovery measured by a camera. In a 101-person cohort, balanced across age deciles and sex, participants completed a gold standard treadmill-based VO2max measurement, two silver standard clinical protocols, and the smartphone-based 12-MRT and 3-MST protocols in the clinic and at home. In a separate 120-participant cohort, the video-based heart rate measurement underlying the 3-MST was measured for accuracy in individuals across the spectrum skin tones while using 8 different smartphones ranging in cost from US $99 to US $999. RESULTS: When compared with gold standard VO2max testing, Lin concordance was pc=0.66 for 12-MRT and pc=0.61 for 3-MST. However, in remote settings, the 12-MRT was significantly less concordant with the gold standard (pc=0.25) compared with the 3-MST (pc=0.61), although both had high test-retest reliability (12-MRT intraclass correlation coefficient=0.88; 3-MST intraclass correlation coefficient=0.86). On the basis of the finding that 3-MST concordance was generalizable to remote settings whereas 12-MRT was not, the video-based heart rate measure within the 3-MST was selected for further investigation. Heart rate measurements in any of the combinations of the six Fitzpatrick skin tones and 8 smartphones resulted in a concordance of pc≥0.81. Performance did not correlate with device cost, with all phones selling under US $200 performing better than pc>0.92. CONCLUSIONS: These findings demonstrate the importance of validating mobile health measures in the real world across a diverse cohort and spectrum of hardware. The 3-MST protocol, termed as heart snapshot, measured VO2max with similar accuracy to supervised in-clinic tests such as the Tecumseh (pc=0.94) protocol, while also generalizing to remote and unsupervised measurements. Heart snapshot measurements demonstrated fidelity across demographic variation in age and sex, across diverse skin pigmentation, and between various iOS and Android phone configurations. This software is freely available for all validation data and analysis code.

Aiming for Cure and Preventive Initiatives in Psoriatic Disease: Building Synergy at NPF, GRAPPA, and PPACMAN.

PURPOSE OF REVIEW: To provide a general overview of the organizations dedicated to advance clinical and translational research in the field of psoriatic disease and to describe the current and future opportunities for team science approaches to overcome unmet needs in the field. Descriptions of initiatives from the NPF, PPACMAN, and GRAPPA are summarized. RECENT FINDINGS: Program projects have recently identified areas of knowledge gaps in diagnosis, treatment, and prevention of psoriasis and psoriatic arthritis (PsA). NPF's Psoriasis Prevention Initiative aims to identify interventions that can prevent the onset and relapse of psoriatic disease or related comorbidities. The Psorcast Study is a joint venture between PPACMAN and Sage Bionetworks based on patient-generated smartphone measurements of psoriatic disease. Similarly, GRAPPA is involved in a number of projects related to axial PsA, enthesitis prevalence, and biomarker discoveries. As important initiatives bring new targets for diagnosis and therapeutics in psoriatic disease, supra-endeavors such as the NIH-Accelerating Medicines Partnership (AMP) and the European Innovative Medicines Initiative (IMI) are promising public-private partnerships that can significantly catapult the field forward.

COVID-19 in Patients With Inflammatory Arthritis: A Prospective Study on the Effects of Comorbidities and Disease-Modifying Antirheumatic Drugs on Clinical Outcomes.

OBJECTIVE: To characterize the hospitalization and death rates among patients with inflammatory arthritis (IA) affected by coronavirus disease 2019 (COVID-19) and to analyze the associations of comorbidities and immunomodulatory medications with infection outcomes. METHODS: Data on clinical and demographic features, maintenance treatment, disease course, and outcomes in individuals with IA (rheumatoid arthritis and spondyloarthritis) with symptomatic COVID-19 infection were prospectively assessed via web-based questionnaire followed by individual phone calls and electronic medical record review. Baseline characteristics and medication use were summarized for hospitalized and ambulatory patients, and outcomes with the different medication classes were compared using multivariable logistic regression. RESULTS: A total of 103 patients with IA were included in the study (80 with confirmed COVID-19 and 23 with high suspicion of COVID-19). Hospitalization was required in 26% of the participants, and 4% died. Patients who were hospitalized were significantly more likely to be older (P < 0.001) and have comorbid hypertension (P = 0.001) and chronic obstructive pulmonary disease (P = 0.02). IA patients taking oral glucocorticoids had an increased likelihood of being admitted for COVID-19 (P < 0.001), while those receiving maintenance anticytokine biologic therapies did not. CONCLUSION: Among patients with underlying IA, COVID-19 outcomes were worse in those receiving glucocorticoids but not in patients receiving maintenance anticytokine therapy. Further work is needed to understand whether immunomodulatory therapies affect COVID-19 incidence.

MoleMapper: an application for crowdsourcing mole images to advance melanoma early-detection research.

Advancements in smartphone technologies and the use of specialized health care applications offer an exciting new era to promote melanoma awareness to the public and improve education and prevention strategies. These applications also afford an opportunity to power meaningful research aimed at improving image diagnostics and early melanoma detection. Here, we summarize our experience associated with developing and managing the implementation of MoleMapper™, a research-based application that not only provides an efficient way for users to digitally track images of moles and facilitate skin self-examinations but also provides a platform to crowdsource research participants and the curation of mole images in efforts to advance melanoma research. Obtaining electronic consent, safeguarding participant data, and employing a framework to ensure collection of meaningful data represent a few of the inherent difficulties associated with orchestrating such a wide-scale research enterprise. In this review, we discuss strategies to overcome these and other challenges leading to the implementation of MoleMapper™.

The Mole Mapper Study, mobile phone skin imaging and melanoma risk data collected using ResearchKit.

Sensor-embedded phones are an emerging facilitator for participant-driven research studies. Skin cancer research is particularly amenable to this approach, as phone cameras enable self-examination and documentation of mole abnormalities that may signal a progression towards melanoma. Aggregation and open sharing of this participant-collected data can be foundational for research and the development of early cancer detection tools. Here we describe data from Mole Mapper, an iPhone-based observational study built using the Apple ResearchKit framework. The Mole Mapper app was designed to collect participant-provided images and measurements of moles, together with demographic and behavioral information relating to melanoma risk. The study cohort includes 2,069 participants who contributed 1,920 demographic surveys, 3,274 mole measurements, and 2,422 curated mole images. Survey data recapitulates associations between melanoma and known demographic risks, with red hair as the most significant factor in this cohort. Participant-provided mole measurements indicate an average mole size of 3.95 mm. These data have been made available to engage researchers in a collaborative, multidisciplinary effort to better understand and prevent melanoma.

The noncoding RNAs SNORD50A and SNORD50B bind K-Ras and are recurrently deleted in human cancer.

Small nucleolar RNAs (snoRNAs) are conserved noncoding RNAs best studied as ribonucleoprotein (RNP) guides in RNA modification. To explore their role in cancer, we compared 5,473 tumor-normal genome pairs to identify snoRNAs with frequent copy number loss. The SNORD50A-SNORD50B snoRNA locus was deleted in 10-40% of 12 common cancers, where its loss was associated with reduced survival. A human protein microarray screen identified direct SNORD50A and SNORD50B RNA binding to K-Ras. Loss of SNORD50A and SNORD50B increased the amount of GTP-bound, active K-Ras and hyperactivated Ras-ERK1/ERK2 signaling. Loss of these snoRNAs also increased binding by farnesyltransferase to K-Ras and increased K-Ras prenylation, suggesting that KRAS mutation might synergize with SNORD50A and SNORD50B loss in cancer. In agreement with this hypothesis, CRISPR-mediated deletion of SNORD50A and SNORD50B in KRAS-mutant tumor cells enhanced tumorigenesis, and SNORD50A and SNORD50B deletion and oncogenic KRAS mutation co-occurred significantly in multiple human tumor types. SNORD50A and SNORD50B snoRNAs thus directly bind and inhibit K-Ras and are recurrently deleted in human cancer.

A LncRNA-MAF:MAFB transcription factor network regulates epidermal differentiation.

Progenitor differentiation requires remodeling of genomic expression; however, in many tissues, such as epidermis, the spectrum of remodeled genes and the transcription factors (TFs) that control them are not fully defined. We performed kinetic transcriptome analysis during regeneration of differentiated epidermis and identified gene sets enriched in progenitors (594 genes), in early (159 genes), and in late differentiation (387 genes). Module mapping of 1,046 TFs identified MAF and MAFB as necessary and sufficient for progenitor differentiation. MAF:MAFB regulated 393 genes altered in this setting. Integrative analysis identified ANCR and TINCR lncRNAs as essential upstream MAF:MAFB regulators. ChIP-seq analysis demonstrated MAF:MAFB binding to known epidermal differentiation TF genes whose expression they controlled, including GRHL3, ZNF750, KLF4, and PRDM1. Each of these TFs rescued expression of specific MAF:MAFB target gene subsets in the setting of MAF:MAFB loss, indicating they act downstream of MAF:MAFB. A lncRNA-TF network is thus essential for epidermal differentiation.

Dicer-microRNA-Myc circuit promotes transcription of hundreds of long noncoding RNAs.

Long noncoding RNAs (lncRNAs) are important regulators of cell fate, yet little is known about mechanisms controlling lncRNA expression. Here we show that transcription is quantitatively different for lncRNAs and mRNAs--as revealed by deficiency of Dicer (Dcr), a key RNase that generates microRNAs (miRNAs). Dcr loss in mouse embryonic stem cells led unexpectedly to decreased levels of hundreds of lncRNAs. The canonical Dgcr8-Dcr-miRNA pathway is required for robust lncRNA transcriptional initiation and elongation. Computational and genetic epistasis analyses demonstrated that Dcr activation of the oncogenic transcription factor cMyc is partly responsible for lncRNA expression. A quantitative metric of mRNA-lncRNA decoupling revealed that Dcr and cMyc differentially regulate lncRNAs versus mRNAs in diverse cell types and in vivo. Thus, numerous lncRNAs may be modulated as a class in development and disease, notably where Dcr and cMyc act.

Enhancer-targeted genome editing selectively blocks innate resistance to oncokinase inhibition.

Thousands of putative enhancers are characterized in the human genome, yet few have been shown to have a functional role in cancer progression. Inhibiting oncokinases, such as EGFR, ALK, ERBB2, and BRAF, is a mainstay of current cancer therapy but is hindered by innate drug resistance mediated by up-regulation of the HGF receptor, MET. The mechanisms mediating such genomic responses to targeted therapy are unknown. Here, we identify lineage-specific enhancers at the MET locus for multiple common tumor types, including a melanoma lineage-specific enhancer 63 kb downstream from the MET TSS. This enhancer displays inducible chromatin looping with the MET promoter to up-regulate MET expression upon BRAF inhibition. Epigenomic analysis demonstrated that the melanocyte-specific transcription factor, MITF, mediates this enhancer function. Targeted genomic deletion (<7 bp) of the MITF motif within the MET enhancer suppressed inducible chromatin looping and innate drug resistance, while maintaining MITF-dependent, inhibitor-induced melanoma cell differentiation. Epigenomic analysis can thus guide functional disruption of regulatory DNA to decouple pro- and anti-oncogenic functions of a dominant transcription factor and block innate resistance to oncokinase therapy.

Control of somatic tissue differentiation by the long non-coding RNA TINCR.

Several of the thousands of human long non-coding RNAs (lncRNAs) have been functionally characterized; however, potential roles for lncRNAs in somatic tissue differentiation remain poorly understood. Here we show that a 3.7-kilobase lncRNA, terminal differentiation-induced ncRNA (TINCR), controls human epidermal differentiation by a post-transcriptional mechanism. TINCR is required for high messenger RNA abundance of key differentiation genes, many of which are mutated in human skin diseases, including FLG, LOR, ALOXE3, ALOX12B, ABCA12, CASP14 and ELOVL3. TINCR-deficient epidermis lacked terminal differentiation ultrastructure, including keratohyalin granules and intact lamellar bodies. Genome-scale RNA interactome analysis revealed that TINCR interacts with a range of differentiation mRNAs. TINCR-mRNA interaction occurs through a 25-nucleotide 'TINCR box' motif that is strongly enriched in interacting mRNAs and required for TINCR binding. A high-throughput screen to analyse TINCR binding capacity to approximately 9,400 human recombinant proteins revealed direct binding of TINCR RNA to the staufen1 (STAU1) protein. STAU1-deficient tissue recapitulated the impaired differentiation seen with TINCR depletion. Loss of UPF1 and UPF2, both of which are required for STAU1-mediated RNA decay, however, did not have differentiation effects. Instead, the TINCR-STAU1 complex seems to mediate stabilization of differentiation mRNAs, such as KRT80. These data identify TINCR as a key lncRNA required for somatic tissue differentiation, which occurs through lncRNA binding to differentiation mRNAs to ensure their expression.

Identification of proteins binding coding and non-coding human RNAs using protein microarrays.

BACKGROUND: The regulation and function of mammalian RNAs has been increasingly appreciated to operate via RNA-protein interactions. With the recent discovery of thousands of novel human RNA molecules by high-throughput RNA sequencing, efficient methods to uncover RNA-protein interactions are urgently required. Existing methods to study proteins associated with a given RNA are laborious and require substantial amounts of cell-derived starting material. To overcome these limitations, we have developed a rapid and large-scale approach to characterize binding of in vitro transcribed labeled RNA to ~9,400 human recombinant proteins spotted on protein microarrays. RESULTS: We have optimized methodology to probe human protein microarrays with full-length RNA molecules and have identified 137 RNA-protein interactions specific for 10 coding and non-coding RNAs. Those proteins showed strong enrichment for common human RNA binding domains such as RRM, RBD, as well as K homology and CCCH type zinc finger motifs. Previously unknown RNA-protein interactions were discovered using this technique, and these interactions were biochemically verified between TP53 mRNA and Staufen1 protein as well as between HRAS mRNA and CNBP protein. Functional characterization of the interaction between Staufen 1 protein and TP53 mRNA revealed a novel role for Staufen 1 in preserving TP53 RNA stability. CONCLUSIONS: Our approach demonstrates a scalable methodology, allowing rapid and efficient identification of novel human RNA-protein interactions using RNA hybridization to human protein microarrays. Biochemical validation of newly identified interactions between TP53-Stau1 and HRAS-CNBP using reciprocal pull-down experiments, both in vitro and in vivo, demonstrates the utility of this approach to study uncharacterized RNA-protein interactions.

Transcriptome sequencing in Sezary syndrome identifies Sezary cell and mycosis fungoides-associated lncRNAs and novel transcripts.

Sézary syndrome (SS) is an aggressive cutaneous T-cell lymphoma (CTCL) of unknown etiology in which malignant cells circulate in the peripheral blood. To identify viral elements, gene fusions, and gene expression patterns associated with this lymphoma, flow cytometry was used to obtain matched pure populations of malignant Sézary cells (SCs) versus nonmalignant CD4(+) T cells from 3 patients for whole transcriptome, paired-end sequencing with an average depth of 112 million reads per sample. Pathway analysis of differentially expressed genes identified mis-regulation of PI3K/Akt, TGFß, and NF-κB pathways as well as T-cell receptor signaling. Bioinformatic analysis did not detect either nonhuman transcripts to support a viral etiology of SS or recurrently expressed gene fusions, but it did identify 21 SC-associated annotated long noncoding RNAs (lncRNAs). Transcriptome assembly by multiple algorithms identified 13 differentially expressed unannotated transcripts termed Sézary cell-associated transcripts (SeCATs) that include 12 predicted lncRNAs and a novel transcript with coding potential. High-throughput sequencing targeting the 3' end of polyadenylated transcripts in archived tumors from 24 additional patients with tumor-stage CTCL confirmed the differential expression of SC-associated lncRNAs and SeCATs in CTCL. Our findings characterize the SS transcriptome and support recent reports that implicate lncRNA dysregulation in human malignancies.

Genomic profiling of a human organotypic model of AEC syndrome reveals ZNF750 as an essential downstream target of mutant TP63.

The basis for impaired differentiation in TP63 mutant ankyloblepharon-ectodermal dysplasia-clefting (AEC) syndrome is unknown. Human epidermis harboring AEC TP63 mutants recapitulated this impairment, along with downregulation of differentiation activators, including HOPX, GRHL3, KLF4, PRDM1, and ZNF750. Gene-set enrichment analysis indicated that disrupted expression of epidermal differentiation programs under the control of ZNF750 and KLF4 accounted for the majority of disrupted epidermal differentiation resulting from AEC mutant TP63. Chromatin immunoprecipitation (ChIP) analysis and ChIP-sequencing of TP63 binding in differentiated keratinocytes revealed ZNF750 as a direct target of wild-type and AEC mutant TP63. Restoring ZNF750 to AEC model tissue rescued activator expression and differentiation, indicating that AEC TP63-mediated ZNF750 inhibition contributes to differentiation defects in AEC. Incorporating disease-causing mutants into regenerated human tissue can thus dissect pathomechanisms and identify targets that reverse disease features.

BRAFV600E remodels the melanocyte transcriptome and induces BANCR to regulate melanoma cell migration.

Aberrations of protein-coding genes are a focus of cancer genomics; however, the impact of oncogenes on expression of the ~50% of transcripts without protein-coding potential, including long noncoding RNAs (lncRNAs), has been largely uncharacterized. Activating mutations in the BRAF oncogene are present in >70% of melanomas, 90% of which produce active mutant BRAF(V600E) protein. To define the impacts of oncogenic BRAF on the melanocyte transcriptome, massively parallel cDNA sequencing (RNA-seq) was performed on genetically matched normal human melanocytes with and without BRAF(V600E) expression. To enhance potential disease relevance by verifying expression of altered genes in BRAF-driven cancer tissue, parallel RNA-seq was also undertaken of two BRAF(V600E)-mutant human melanomas. BRAF(V600E) regulated expression of 1027 protein-coding transcripts and 39 annotated lncRNAs, as well as 70 unannotated, potentially novel, intergenic transcripts. These transcripts display both tissue-specific and multi-tissue expression profiles and harbor distinctive regulatory chromatin marks and transcription factor binding sites indicative of active transcription. Coding potential analysis of the 70 unannotated transcripts suggested that most may represent newly identified lncRNAs. BRAF-regulated lncRNA 1 (BANCR) was identified as a recurrently overexpressed, previously unannotated 693-bp transcript on chromosome 9 with a potential functional role in melanoma cell migration. BANCR knockdown reduced melanoma cell migration, and this could be rescued by the chemokine CXCL11. Combining RNA-seq of oncogene-expressing normal cells with RNA-seq of their corresponding human cancers may represent a useful approach to discover new oncogene-regulated RNA transcripts of potential clinical relevance in cancer.

ZNF750 is a p63 target gene that induces KLF4 to drive terminal epidermal differentiation.

Disrupted epidermal differentiation characterizes numerous diseases that impact >25% of the population. In a search for dominant mediators of differentiation, we defined a requirement for ZNF750 in terminal epidermal differentiation. ZNF750 controlled genes mutated in numerous human skin diseases, including FLG, LOR, LCE3B, ALOXE3, and SPINK5. ZNF750 induced progenitor differentiation via an evolutionarily conserved C2H2 zinc finger motif. The epidermal master regulator, p63, bound the ZNF750 promoter and was necessary for its induction. ZNF750 restored differentiation to p63-deficient tissue, suggesting that it acts downstream of p63. A search for functionally important ZNF750 targets via analysis of ZNF750-regulated genes identified KLF4, a transcription factor that activates late epidermal differentiation. ZNF750 binds to KLF4 at multiple sites flanking the transcriptional start site and controls its expression. ZNF750 thus directly links a tissue-specifying factor, p63, to an effector of terminal differentiation, KLF4, and represents a potential future target for disorders of this process.

Suppression of progenitor differentiation requires the long noncoding RNA ANCR.

Long noncoding RNAs (lncRNAs) regulate diverse processes, yet a potential role for lncRNAs in maintaining the undifferentiated state in somatic tissue progenitor cells remains uncharacterized. We used transcriptome sequencing and tiling arrays to compare lncRNA expression in epidermal progenitor populations versus differentiating cells. We identified ANCR (anti-differentiation ncRNA) as an 855-base-pair lncRNA down-regulated during differentiation. Depleting ANCR in progenitor-containing populations, without any other stimuli, led to rapid differentiation gene induction. In epidermis, ANCR loss abolished the normal exclusion of differentiation from the progenitor-containing compartment. The ANCR lncRNA is thus required to enforce the undifferentiated cell state within epidermis.