From the Microscopic Sweat Gland to Large Datasets: The Power of Quantitative Traits in Dermatology

Clinical Implications▪Using traits rather than diseases as outcomes in human genetic studies facilitates discoveries relevant to disease pathology.▪Sweat gland density GWAS suggests human relevance of a mouse model of eccrine and hair follicle density.▪Potassium channel genes are further implicated in human sweat gland development and dysfunction. ▪Using traits rather than diseases as outcomes in human genetic studies facilitates discoveries relevant to disease pathology.▪Sweat gland density GWAS suggests human relevance of a mouse model of eccrine and hair follicle density.▪Potassium channel genes are further implicated in human sweat gland development and dysfunction. Sweat gland density (SGD) is a cutaneous trait that varies naturally within and across populations and also contributes to disease pathologies. In their recent article in Journal of Investigative Dermatology, Chen et al., 2024Chen W. Li J. Xu K. Luo J. Wang M. Liu Y. et al.Genetic variants associated with sweat gland phenotypes in 6,210 Han Chinese individuals.J Invest Dermatol. 2024; Google Scholar report a GWAS of SGD that illustrates how genetic studies of quantitative skin traits provide an efficient method for determining the molecular basis of dermatologic diseases. This work also provides an opportunity to highlight the many modalities available to collect quantitative data about skin features that have relevance to cutaneous diseases. Human genetic studies have emerged as a powerful tool for reducing the physical and financial burdens of disease. Given that skin disease is the fourth leading cause of disability globally and among the top 20 causes of disability-adjusted life years, which also includes years of life lost, there remain many opportunities in dermatology to use human genetic studies to improve health (Jumonville et al., 2024Jumonville G. Hong D. Khan A. DeWan A. Leal S.M. Weng C. et al.Digital biobanks are underutilized in dermatology and create opportunities to reduce the burden of skin disease.Br J Dermatol. 2024; 190: 566-568Google Scholar). Historically, human genetic studies have been conducted in disease cohorts assembled with cases (individuals with a disease) and controls (individuals without disease). However, most diseases are rare in the population. Even diseases that are considered common typically affect <1% of the population. This limits the size of cohorts and constrains statistical power for gene discovery. In contrast to diseases, traits (nonpathogenic features that vary naturally in a population) are easier to study with genetic approaches because everyone in the population can contribute outcome data. Similar methods are used for genetic studies of traits (eg, variant-level association tests with GWAS or gene-level association tests with burden testing), with the main difference being that the outcome is quantitative (a numerical value) rather than binary (case or control disease status). Furthermore, understanding trait variance in a population can better define pathological states at molecular, cellular, and histological levels. Thus, genetic studies of traits can be leveraged to improve our understanding of disease pathogenesis. In neurology, collecting brain images of healthy individuals facilitated genetic studies of brain structure and helped to identify new features associated with neurodevelopmental disorders (Warrier et al., 2023Warrier V. Stauffer E.M. Huang Q.Q. Wigdor E.M. Slob E.A.W. Seidlitz J. et al.Genetic insights into human cortical organization and development through genome-wide analyses of 2,347 neuroimaging phenotypes.Nat Genet. 2023; 55: 1483-1493Google Scholar). In dermatology, where the visual nature of clinical assessments and measurements are so central, there remain many opportunities to use human genetic studies of cutaneous traits to understand molecular regulators of skin morphology and physiology. Sweat glands play an important role in regulating body temperature and maintaining skin hydration. Thermoregulation failure can result in heat stroke and organ failure. Inadequate skin hydration causes dry skin and increases susceptibility to wrinkles, whereas excessive sweating can lead to skin irritation and moisture-related dermatitis. Within populations, natural variation in sweat gland development and function contributes to variance in heat tolerance and skin aging among individuals—2 traits that are increasingly important as we adapt to climate change. Prior human genetic studies of sweat glands have largely focused on disease outcomes such as anhidrosis and hyperhidrosis. Anhidrosis is a rare disease that often comes to medical attention in syndromic forms. Hyperhidrosis, despite being considered a common disease, only affects approximately 1 of 100 people. As a result, genetic studies have been sparse, and very little is known about the biological drivers of sweat gland development and function in humans. In their recent article in Journal Investigative Dermatology, Chen et al., 2024Chen W. Li J. Xu K. Luo J. Wang M. Liu Y. et al.Genetic variants associated with sweat gland phenotypes in 6,210 Han Chinese individuals.J Invest Dermatol. 2024; Google Scholar overcomes this limitation by focusing their genetic study of sweat glands on a quantitative trait, SGD. Using an efficient and scalable measure of SGD, a simple starch–iodine test in conjunction with imaging analysis, they collected GWAS outcome data on a large sample of 6,210 Han Chinese individuals. The increase in statistical power afforded by their study design and choice of outcome allowed them to detect 3 new loci associated with sweat gland phenotypes in humans. Two of the loci, at chromosomes 1p36.31 and 1q41, implicate potassium channel genes (KCNAB2 and KCNK2, respectively), validating previous associations between potassium channels and idiopathic anhidrosis (Patterson et al., 2021Patterson Rosa L. Walker N. Mallicote M. MacKay R.J. Brooks S.A. Genomic association of chronic idiopathic Anhidrosis to a potassium channel subunit in a large animal model.J Invest Dermatol. 2021; 141: 2639-2645.e3Google Scholar). Interestingly, the third locus, at 18q21.3, is syntenic to an interval in the mouse genome that was linked to eccrine and hair follicle density in a mouse quantitative trait locus mapping study (Kamberov et al., 2015Kamberov Y.G. Karlsson E.K. Kamberova G.L. Lieberman D.E. Sabeti P.C. Morgan B.A. et al.A genetic basis of variation in eccrine sweat gland and hair follicle density.Proc Natl Acad Sci USA. 2015; 112: 9932-9937Google Scholar), providing coherency with existing knowledge of SGD and an opportunity to further probe the generalizability of mouse studies to human traits. Finally, they also used their GWAS outcome data to demonstrate a correlation between SGD and important patient-centered measures, including susceptibility to dry skin (OR = 1.03, 95% confidence interval [CI] = 1.01–1.05, P = .003), cracked and wrinkled skin (OR = 1.02, 95% CI = 1.00–1.03, P = .004), and number of facial pigment spots (β = 0.03, 95% CI = 0.01–0.05, P = .003). This study provides a salient example of how genetic studies of quantitative traits in dermatology can be leveraged to improve our knowledge about fundamental cutaneous physiology that is relevant to skin health. In particular, there are opportunities to leverage imaging modalities that capture nuanced phenotypic variations. By amassing quantitative trait data in large cohorts, researchers can achieve robust statistical power, facilitating the identification of genetic variants associated with dermatologic traits. Dermatology distinguishes itself from other patient-facing medical specialties by the nearly complete reliance on visualization to guide diagnosis and treatment of diseases. The visual findings that dermatologists make during physical examination are challenging to document in a quantitative and precise manner. At the same time, there exists a wide array of imaging modalities that can provide valuable insights into skin morphology and function (Table 1). These range from low-tech and affordable options to highly sophisticated and expensive equipment that has been previously reviewed (Tkaczyk, 2017Tkaczyk E. Innovations and developments in dermatologic non-invasive optical imaging and potential clinical applications.Acta Derm Venereol. 2017; 218: 5-13Google Scholar). Examples range from simple photography for characterizing cutaneous lesions, dermoscopy for imaging pigmented lesions, and high-frequency ultrasound for preoperative tumor assessment. Recently, standardized photography has been integrated as a key documentation of the primary endpoint of major multicenter clinical trials in Stevens–Johnson syndrome (NATIENS ClinicalTrials.gov NCT02987257) and mpox (PALM007 NCT05559099). This heralds that rigorous imaging datasets will likely become the norm in dermatology research. Adoption of other technologies will undoubtedly follow as miniaturization enables the implementation of more advanced optical technologies on smartphones (Hunt et al., 2021Hunt B. Ruiz A. Pogue B. Smartphone-based imaging systems for medical applications: a critical review.J Biomed Opt. 2021; 26040902Google Scholar).Table 1Imaging Modalities Available to Capture Cutaneous Traits that are Relevant to Skin DiseasesMethodTraitsExpenseScalabilityEfficiencyDigital photographySkin pigmentation, freckling, pigmented lesions, erythema, scaling, photoaging, hair follicle density, and hair fiber characteristicsLowHighHigh (quick and simple)Skin tapingMicrobiome, proteome, lipids, RNA, and pHLowHighHigh (noninvasive)WearablesScratching, UVR exposure, sweat production, and skin temperatureVariableHighHigh (dependent on device)Total body digital photographyPigmented lesions, moles, freckling, scaling, hyperkeratosis, nail color, shape, and textureMediumMediumMediumConfocal microscopyCellular structure, epidermal architecture, pigmented lesions, and photoagingHighLowLow (requires expertise)Spectral imaging and elastic scattering spectroscopyMelanin distribution, vascular patterns, inflammatory changes, and UV damageHighLowLow (requires expertise)Optical coherence tomographyHair follicle density, microvasculature patterns, and epidermal thicknessHighMediumMediumEmerging optical technologies1Emerging optical technologies are reviewed in Tkaczyk (2017).MultipleVariableVariableVariable1 Emerging optical technologies are reviewed in Tkaczyk, 2017Tkaczyk E. Innovations and developments in dermatologic non-invasive optical imaging and potential clinical applications.Acta Derm Venereol. 2017; 218: 5-13Google Scholar. Open table in a new tab The emergence of wearable sensors presents another promising avenue for quantifying traits in dermatology that would otherwise be amenable only to subjective assessments. For example, Yang et al., 2023Yang A.F. Chun K.S. Yu L. Walter J.R. Kim D. Lee J.Y. et al.Validation of a hand-mounted wearable sensor for scratching movements in adults with atopic dermatitis.J Am Acad Dermatol. 2023; 88: 726-729Google Scholar conducted a proof-of-concept study of atopic dermatitis symptomatology in 11 patients over 73 nights with a wearable sensor that detected acoustomechanic signals generated by scratching. Subjects wore a wrist-mounted device overnight to quantify scratching, an important QOL metric in atopic dermatitis. Researchers used an artificial intelligence algorithm to quantify nocturnal scratching and sleep metrics. The researchers found that increasing severity of itching as measured by the wearable correlated with worse sleep. This innovative approach holds potential for advancing both diagnostic capability and therapeutic interventions in dermatologic practice by quantifying a subjective trait, nocturnal itching, that previously could not be measured in the sleeping patient. Future modalities for wearables include hybrid wearable sensors that interact with the epidermis and combine chemical, physical, and electrophysiologic sensors on the same platform. Dermatologic applications of these wearables could allow nuanced understanding of physiologic states and provide more data that can inform on dermatologic diagnosis and treatment (Heikenfeld et al., 2018Heikenfeld J. Jajack A. Rogers J. Gutruf P. Tian L. Pan T. et al.Wearable sensors: modalities, challenges, and prospects.Lab Chip. 2018; 18: 217-248Google Scholar). Tape strips are a minimally invasive technique in dermatology research that allow for epidermal sampling without the need for a skin biopsy. Applications for tape stripes range from epidermal conditions such as atopic dermatitis, melanoma, and lichen planus to skin cancer and microbiome analysis (Hughes et al., 2021Hughes A.J. Tawfik S.S. Baruah K.P. O'Toole E.A. O'shaughnessy R.F.L. Tape strips in dermatology research.Br J Dermatol. 2021; 185: 26-35Google Scholar). Tape strips can be used to sample cutaneous microbes and permit identification of microorganisms within the stratum corneum. They also can be used to sample protein, RNA, lipid, and chemistry (eg, pH). Tape strips are low cost, do not cause bleeding or scarring, are fast and noninvasive, and can be performed in patients of all ages. Thus, the use of tape strips provides an opportunity to collect quantitative skin data at a large scale (Hughes et al., 2021Hughes A.J. Tawfik S.S. Baruah K.P. O'Toole E.A. O'shaughnessy R.F.L. Tape strips in dermatology research.Br J Dermatol. 2021; 185: 26-35Google Scholar). The ability to generate large datasets is limited by the scalability of imaging modalities. Although numerous imaging modalities are available in dermatology, not all strategies are routinely performed in the clinical setting. Current standard of care in dermatology employs digital photography, dermoscopy, and dermatopathology in the correct clinical context. Although a low-tech option, the humble digital photograph, readily accessible with patient and provider smart phones, offers opportunities for further scaling of quantification of dermatologic findings. Modalities such as high-frequency ultrasound, reflectance confocal microscopy, and optical coherence tomography represent special applications at centers where these technologies and trained clinicians are available. Despite this visual richness, the field has yet to fully exploit the potential of imaging in understanding dermatologic traits. The study by (Chen et al., 2024Chen W. Li J. Xu K. Luo J. Wang M. Liu Y. et al.Genetic variants associated with sweat gland phenotypes in 6,210 Han Chinese individuals.J Invest Dermatol. 2024; Google Scholar) exemplifies the power of integrating imaging technologies with genetic approaches to unravel the complex physiologic underpinnings of dermatologic traits. It also is an example of an imaging modality that is not currently standard of care. When imaging is standard of care, it becomes economically feasible to scale up data collection, as was done to conduct genetic studies of human cortical organization and development (Warrier et al., 2023Warrier V. Stauffer E.M. Huang Q.Q. Wigdor E.M. Slob E.A.W. Seidlitz J. et al.Genetic insights into human cortical organization and development through genome-wide analyses of 2,347 neuroimaging phenotypes.Nat Genet. 2023; 55: 1483-1493Google Scholar). The integration of large-scale quantitative trait data into digital biobanks holds immense potential for driving future innovation in dermatology (Jumonville et al., 2024Jumonville G. Hong D. Khan A. DeWan A. Leal S.M. Weng C. et al.Digital biobanks are underutilized in dermatology and create opportunities to reduce the burden of skin disease.Br J Dermatol. 2024; 190: 566-568Google Scholar). Current imaging modalities in dermatology range from the inexpensive and readily accessible digital photography to sophisticated imaging techniques. Wearables and tape strips facilitate skin measurements and sampling at scales appropriate for gene discovery. With far-ranging applications including blood flow monitoring, microbiome surveillance, skin cancer diagnosis, composition measurements in atopic dermatitis, skin rejuvenation measurement, and noninvasive sentinel lymph node assessment in melanoma, these modalities represent rich sources of data for enhancing the field of dermatology. Such resources would accelerate discoveries about the genetic basis of cutaneous traits, inform diagnostics, and potentially improve the clinical management of skin disease. By elucidating the genetic determinants underlying sweat gland phenotypes, (Chen et al., 2024Chen W. Li J. Xu K. Luo J. Wang M. Liu Y. et al.Genetic variants associated with sweat gland phenotypes in 6,210 Han Chinese individuals.J Invest Dermatol. 2024; Google Scholar) pave the way for targeted interventions for conditions such as hyperhidrosis and anhidrosis and offers opportunities to improve QOL for nonpathological conditions.