A few drops of blood collected from a newborn’s heel may contain clues about whether that child is at unusually high risk of developing cancer during early childhood, according to a large population-based study led by researchers at Dana-Farber/Boston Children’s Cancer and Blood Disorders Center and Mass General Brigham. The findings suggest that genomic testing could eventually be added to routine newborn screening, allowing clinicians to identify some inherited cancer-predisposition syndromes before symptoms appear. Such information could create an opportunity for intensive surveillance, earlier diagnosis and, in some cases, less aggressive treatment.
The study, published in Nature Communications, examined archived dried blood spots from 1,948 children born in Michigan who later developed a solid tumor or brain tumor by the age of eight. Researchers used DNA extracted from the newborn samples to analyze a panel of 11 genes associated with pediatric cancer-predisposition syndromes. They identified pathogenic or likely pathogenic variants—genetic changes known or strongly suspected to disrupt normal gene function—in 132 children, representing nearly 7 percent of the study group.
Routine newborn screening currently focuses primarily on rare biochemical and metabolic disorders for which early treatment can prevent severe disability or death. A small blood sample collected approximately 24 hours after birth is tested for abnormal levels of specific substances or signs of impaired biological pathways. Genomic newborn screening would use the same type of dried blood spot, but would examine the infant’s DNA sequence instead. In the context of cancer risk, the goal would not be to detect a tumor in a newborn, but to identify inherited or newly occurring variants that increase the probability of cancer later in childhood.
Most of the relevant variants are germline changes, meaning they are present in virtually every cell of the body from birth. They may be inherited from a parent or arise spontaneously during the formation of the egg, sperm or early embryo. A germline mutation can impair genes that normally control cell division, repair damaged DNA or prevent abnormal cells from surviving. When one of these protective systems is weakened, a child may be more likely to develop a tumor after acquiring additional genetic changes. The presence of a pathogenic variant does not always guarantee that cancer will occur, but it can substantially alter the timing and probability of disease.
The clearest signals in the study involved cancers with well-established genetic links. All six children who developed medullary thyroid carcinoma carried a germline mutation in RET, a gene associated with multiple endocrine neoplasia type 2 and other inherited cancer syndromes. Among children who developed retinoblastoma, the most common eye cancer of childhood, 40 percent had a germline mutation in RB1. Mutations in RB1 disrupt a critical cell-cycle checkpoint, allowing retinal precursor cells to divide inappropriately and form tumors at a very young age.
The study also detected cancer-predisposing variants among children who later developed choroid plexus carcinoma, adrenocortical carcinoma, pineoblastoma and medulloblastoma. Depending on the tumor type, between 11 and 30 percent of cases carried a mutation in one of the 11 genes examined. In 130 of the 132 children with a detected variant, the altered gene was already known to be associated with the type of tumor that subsequently developed. This alignment between the genetic finding and the later cancer strengthens the possibility that targeted genomic screening could be clinically useful, rather than merely producing ambiguous risk information.
Children carrying a cancer-predisposing variant also tended to receive their cancer diagnosis earlier than children without one. The median age at diagnosis was 14 months among children with a detected mutation, compared with 32 months among those without an identified variant. The contrast was especially pronounced for retinoblastoma. Children with a germline RB1 mutation were diagnosed at a median age of nine months, while those without a detected RB1 variant were diagnosed at a median age of 23 months.
That difference could directly influence medical care. If an infant were known to carry a high-risk RB1 mutation, clinicians could begin regular ophthalmologic examinations soon after birth, sometimes using examinations under anesthesia or specialized imaging when appropriate. Early surveillance might reveal a small tumor—or a lesion at a very early stage—before it threatens vision or spreads beyond the eye. Earlier treatment can sometimes allow doctors to use localized therapies and avoid or reduce the need for more intensive interventions, including chemotherapy, radiation or surgical removal of the eye.
The researchers estimate that approximately one in 27,000 newborns would develop an early-onset cancer that might have been predicted through genomic newborn screening. In the United States, that frequency could correspond to roughly 1,000 children each year who might benefit from being identified at birth. However, the study does not establish that universal screening is ready for immediate implementation. The analysis was observational and focused on children who later developed cancer, rather than screening an unselected group of newborns and following every child prospectively. Further research will be needed to determine how often similar variants occur in the general newborn population, how accurately they predict cancer, and whether surveillance improves survival, quality of life or treatment-related outcomes.
There are also significant practical and ethical questions. A positive result would require rapid confirmation through clinical genetic testing, careful interpretation of the child’s specific variant and counseling for the family. Some genetic changes may be difficult to classify, while others may raise questions about risks that vary widely between individuals. Health systems would need enough pediatric oncologists, genetic counselors, ophthalmologists and other specialists to provide syndrome-specific monitoring. Families would also need clear information about what a result means, including the difference between increased risk and certainty of disease. Researchers at Mass General Brigham and Dana-Farber are now working toward a workflow that could connect newborn genomic findings with established surveillance protocols.
The study’s authors argue that cancer-risk genes should be considered selectively for expanded newborn screening when three conditions are met: the genetic risk is sufficiently well established, the associated cancer can arise early in life, and surveillance or early treatment can meaningfully improve outcomes. The concept is not to sequence every possible cancer-related gene without a clear clinical plan. Instead, it would focus on a carefully validated set of variants for which a positive result can lead to immediate, evidence-based action. As genomic technologies become faster and less expensive, the same heel-stick sample used to screen for metabolic disorders could become a gateway to identifying a small but important group of children whose cancers might otherwise remain invisible until symptoms emerge.
Subject of Research: Human tissue samples
Article Title: Population-based genomic detection of childhood cancer predisposition using newborn dried blood spots
News Publication Date: 12-Aug-2026
Web References: https://www.nature.com/articles/s41467-026-76296-8
References: Diller L et al., “Population-based genomic detection of childhood cancer predisposition using newborn dried blood spots,” Nature Communications, DOI: 10.1038/s41467-026-76296-8
Keywords: Genetic testing, genomic newborn screening, pediatric cancer, cancer predisposition, newborn dried blood spots, germline mutations, retinoblastoma, RB1, RET, early cancer detection
