A genome-wide association study of genetic and clinical data from 1.6 million people has identified 22 unique genes linked to keloids, raised scars growing beyond the original wound boundary that can cause pain and disfigurement.
Results of the study, led by researchers from Vanderbilt University Medical Center and reported in the journal Nature Communications, have yielded a deeper understanding of the genetic basis for keloids, and revealed “biological threads” for scar formation that are shared across populations of widely different ancestries.
“By connecting these newly implicated genes to each other, we uncovered potential molecular networks that could be strategically targeted for prevention and treatment,” said Jacklyn Hellwege, PhD, research assistant professor of Medicine in the Division of Genetic Medicine and Clinical Pharmacology.
“These results have implications beyond keloids,” added Hellwege, the report’s co-corresponding author with Digna Velez Edwards, PhD, MS, professor of Obstetrics and Gynecology and director of the Center for Women’s Health Research at VUMC.
“We now know that many keloid-associated genes are also important in other forms of fibrosis, tissue damage and dysregulated growth,” she said. “Using that information collectively could lead to broad insights across human disease.”
Keloids result from a derangement of the wound-healing process that involves excessive deposits of collagen, the body’s main structural protein, and overactive cell proliferation. These raised scars tend to recur after surgical removal, and the effectiveness of alternative treatments is limited.
Keloids are more common in darker-skinned people, particularly those of African or Asian descent. In the United States, about one in 30 Black individuals will develop keloids; their risk is approximately 20-times higher than it is for people of European (white) ancestry.
While men and women have a similar incidence of keloids, cases of worsening scarring during pregnancy or after puberty in females have been documented. People with keloids also have an increased risk of cancer, including skin cancer.
“For a long time, it was very difficult to assemble a large enough cohort to study the genetics of keloids,” said co-author Todd Edwards, PhD, MS, associate professor of Medicine in the Division of Epidemiology. “We tried for years without much success.
“However, by using data from biobanks we were able to gather a large collection of participants and more than triple the number of associated genes from prior studies,” Edwards said.
Previous genome-wide association studies (GWAS), which link genetic variations to disease, identified six distinct loci, or positions on the chromosomes, which are significantly correlated with keloid risk.
The current study, the largest to date, drew from seven large genomic data sources, including VUMC’s DNA biorepository, BioVU; the national Electronic Medical Records and Genomics (eMERGE) network; the National Institutes of Health (NIH) All of Us research program; and the U.S. Department of Veterans Affairs Million Veteran Program.
The researchers identified 26 loci associated with keloid risk, 20 of which were novel associations. Along with the identification of 22 specific risk genes, these associations have illuminated potentially targetable biological processes in keloid pathology.
Using a statistical method to estimate heritability, the likelihood that an observable trait (keloids) was due to genetic variation, they found that the heritability of keloids was highest in African ancestry, reflecting the observed pattern of keloid prevalence in different populations.
With another technique called S-PrediXcan, they investigated the potential functional effects that keloid-associated variants may have on gene expression in specific tissues. Of 27 gene-tissue pairs identified this way, nine were in skin cells and in fibroblasts, which play key roles in connective tissue formation and wound healing.
The most significant variant in the multi-ancestry and European ancestry analyses was located near LINCO1705. Like other keloid-associated genes, LINCO1705 plays a role in tumor suppression. Increased expression of the gene in fibroblasts also was associated with a decreased risk of keloids.
The most significant variant in the African-ancestry specific analysis was located near the ITGA11 gene, which encodes a collagen receptor and is involved in regulating the integrin-binding, profibrotic TGF-beta signaling pathway.
ITGA11 also is expressed by cancer-associated fibroblasts and has been implicated in the growth of aggressive cancers. It has previously been associated with various kinds of tissue fibrosis, uterine fibroids and Dupuytren’s disease, the abnormal accumulation of fibrotic tissue resulting in contracture of the hand.
The researchers plan to use these gene-level results to identify possible pharmaceutical targets for treating keloids, beyond the current use of corticosteroids.
Prior to this study, they leveraged genetic associations to identify opportunities for “repurposing” drugs already approved for other diseases, and to evaluate potential safety risks.
“We’re working on this for keloids,” Edwards said.
The first author of the keloid paper, Catherine Greene, is a PhD candidate in Human Genetics.
Other VUMC co-authors were Gabrielle Hampton, MS, also a PhD candidate in Human Genetics, James Jaworski, MPH, a statistical genetic analyst in the Edwards and Velez Edwards labs, and Megan Shuey, PhD, research assistant professor of Medicine in the Division of Genetic Medicine and Clinical Pharmacology.
Edwards, Hellwege, and Velez Edwards, who supervised the keloid study, co-led another multicenter study that earlier this year reported the identification of multiple genes associated with uterine fibroids, the nation’s leading indication for hysterectomy.
These studies exemplify the power of GWAS and other methods to unravel the mystery of common, complex diseases, and open the door to new, more effective treatments, the researchers said.
The keloid study was supported in part by NIH grants TL1TR002244, T32GM080178, K12AR084232 and R21AR067938.
