(Photo by Jimmy Conover on Unsplash)
Forever Chemicals in Newborn Blood Are Being Linked to Childhood Leukemia
In A Nutshell
- Researchers found that newborns with the highest levels of PFAS “forever chemicals” in their blood had increased odds of later developing childhood leukemia, though most results fell short of statistical certainty.
- The strongest associations came from lesser-known replacement chemicals, compounds brought in as substitutes for older PFAS, with some linked to more than fivefold increases in leukemia risk in smaller analyses.
- PFAS have been detected in the blood of 98% of Americans and linger in the body for years, even in children born after major manufacturers began phasing out the most common compounds.
- Experts say the findings do not prove cause and effect but add to a growing body of evidence calling for closer scrutiny of newer PFAS chemicals that have not been widely studied.
A few drops of blood drawn from a baby’s heel just hours after birth can give researchers a rare look at chemical exposures present at birth, and a new study published in the Journal of Exposure Science & Environmental Epidemiology used exactly that to examine a possible connection to one of the most common childhood cancers.
Researchers examined dried blood spots collected from newborns in Los Angeles County and found that babies with the highest levels of certain PFAS, sometimes called “forever chemicals” because many persist for years in the environment and the body, had increased odds of developing childhood leukemia. Most individual results did not meet the threshold scientists use to call findings definitive, but the pattern of rising risk with rising exposure was consistent enough that researchers say the connection warrants serious attention.
PFAS are a large family of industrial chemicals found in nonstick cookware, waterproof clothing, fast-food packaging, and firefighting foam. Previous research cited in the study found PFAS in the blood of 98% of Americans. Even though manufacturers began phasing out the most widely studied compounds in the early 2000s, these chemicals persist in the environment and in human bodies for years.
Blood Spots Give Researchers a Window Into Birth Exposures
In California and most other states, a few drops of blood are collected from nearly every newborn, typically within 36 hours of birth, and stored on paper cards. Those cards become a time capsule of what was circulating in a baby’s blood at the moment of birth.
Researchers tapped into a statewide California registry linking childhood cancer cases to those stored blood spot cards. They identified 125 children diagnosed with acute lymphoblastic leukemia between ages 0 and 14 during 2000 through 2015, all born in LA County, and compared them to 219 cancer-free controls from the same county. LA County was chosen for its large, diverse population and because it contained the greatest number of California public water districts where PFAS had been detected. Lab staff at Yale University, blinded to which samples came from leukemia patients, extracted and analyzed the blood spots.
PFAS Levels at Birth Were Higher in Children Who Later Developed Leukemia
Of 23 PFAS chemicals targeted in the analysis, 17 turned up in the blood spot samples, though only two appeared in more than half: PFOA and PFOS, the two most widely studied compounds in this chemical family.
Children in the highest quarter of PFOS exposure had a 56% greater chance of developing leukemia compared to those in the lowest quarter. For PFOA, the increase was 64%. Neither result was statistically definitive on its own. Childhood leukemia affects roughly 3 to 5 children per 100,000 each year, so even a 56% or 64% increase in relative risk translates to a small change in absolute risk for any individual child. When the team looked at PFOA and PFOS together, the highest risk appeared among children with the highest levels of both compounds.
Replacement PFAS Showed Some of the Strongest Associations
A newer detection method that scans for any fluorinated chemical in a sample, including ones never formally studied, works more like a chemical dragnet than a targeted search. Researchers applied it here alongside the traditional approach.
It turned up 26 additional PFAS in newborn blood, nine common enough to analyze statistically. Two stood out. Both belong to a class of chemicals used as replacements for PFOA in some modern manufacturing processes, compounds that have received far less scientific scrutiny than the originals. A doubling in exposure to one was associated with a fivefold increase in leukemia risk; children in the highest exposure quarter for the other had more than five times the odds of developing leukemia compared to those in the lowest quarter, a finding that was statistically definitive. Those estimates came from smaller subsets of samples with wide margins of uncertainty and should be read as a signal for follow-up, not a precise measure of risk.
Who Was Exposed and How
About 73% of leukemia cases and 69% of controls were Hispanic, reflecting LA County’s demographics. When researchers broke results down by ethnicity, those findings were treated as exploratory because subgroups became small. Associations were generally stronger among non-Hispanic participants, and several newly identified PFAS showed statistically definitive links among non-Hispanic children that weren’t visible in the full sample. Exposure patterns varied by group, likely reflecting differences in diet, consumer products, and drinking water.
Other studies have pointed in a similar direction. A separate California analysis published in 2025 found similarly elevated, though not statistically definitive, risks with PFOS in newborn blood spots, and a Finnish study found elevated risks linked to PFOA during certain time periods. Earlier work by members of this research group had already connected PFAS in household dust to childhood leukemia risk.
No single study of this size can establish cause and effect. But when babies’ first drops of blood contain dozens of industrial chemicals, including compounds brought in as replacements for ones already flagged as harmful, a practical question follows: are newer PFAS being studied quickly enough?
Disclaimer: This article is based on a published scientific study and is intended for informational purposes only. It does not constitute medical advice. The findings reflect associations observed in a research setting and do not prove that PFAS cause childhood leukemia. Readers with questions about chemical exposures or their family’s health should consult a qualified medical professional.
Paper Notes
Limitations
The study’s statistical power was limited by its small sample size of 125 cases and 219 controls. Many of the 23 PFAS included in the targeted analysis were not detectable in most participants. The mutually adjusted non-targeted analysis was limited to only 86 participants who had detectable levels of all seven compounds included in that model, and results from that analysis carry wide confidence intervals. The research relied on a single PFAS measurement from dried blood spots, which is less reliable than repeated blood draws over time. Dried blood spots also contain smaller blood volumes than traditional samples, which limits detection sensitivity. Breakdowns by ethnicity and other demographic factors were explicitly described by the authors as exploratory due to small numbers within subgroups. Associations that did not reach statistical significance cannot rule out chance as an explanation.
Funding and Disclosures
The study was supported by grant R01 ES032196 from the National Institutes of Health. Co-author Xiaomei Ma disclosed consulting for Bristol Myers Squibb outside the scope of this work. No other competing interests were declared.
Publication Details
The study, “Targeted and non-targeted analyses of per-and polyfluoroalkyl substances in newborn dried blood spots and risk of childhood acute lymphoblastic leukemia,” was published in the Journal of Exposure Science & Environmental Epidemiology, a Nature publication. DOI: 10.1038/s41370-026-00891-6. Authors: Veronica M. Vieira (University of California, Irvine), Sheng Liu and Jeremy Koelmel (Yale University), Libby M. Morimoto and Catherine Metayer (University of California, Berkeley), Natalie R. Binczewski (University of California, Irvine), Joseph L. Wiemels (University of Southern California Keck School of Medicine), and Xiaomei Ma (Yale School of Public Health). Corresponding author: Veronica M. Vieira ([email protected]). Received January 8, 2026; accepted April 1, 2026; published online April 14, 2026.
