乌鸦传媒

Zinc and copper metabolism cycles in the layers of baby teeth may be able to predict which children will develop autism spectrum disorder, a longitudinal analysis suggests.

This is the first study to generate a 90% accurate fetal and early childhood biomarker of autism by tracking metabolic pathways over time and could lead to new diagnostic tools, reported Paul Curtin, PhD, of Icahn School of Medicine at Mount Sinai in New York City, and colleagues in .

Using novel tooth-matrix biomarkers that directly measured uptake of elements, the researchers found that children who later developed autism had disrupted zinc-copper rhythmicity in utero or in their earliest months of life.

"We looked at the naturally shed teeth of children and explored them much as you would explore the growth rings of a tree, using them as a sort of retrospective biomarker to see what children were exposed to in the womb and in early life. When we derived measures of metabolic cycles and used machine-learning algorithms to predict which children would develop autism, we found out we were 90% accurate in our predictions," he told 乌鸦传媒.

Prenatal and newborn children form a new tooth layer daily, which captures an imprint of chemicals circulating in the body and produces a chronological exposure record. Zinc and copper pathways are central regulators of multiple metals; disruption of the pathways may have downstream effects that may affect the metabolism of other essential elements and toxic metals.

For this study, the researchers analyzed teeth from 193 participants in four case-control samples -- a of twins from Sweden, two similar groups from the United States, and a birth cohort from the United Kingdom -- using a laser ablation technique to sample each tooth at an average of 152 locations. "The data from the teeth we analyzed covered primarily the second and third trimesters through a few months after birth," Curtin said.

In all four study sets and in the pooled analysis, zinc-copper rhythmicity was disrupted in autism cases. The autism cases had three quantifiable characteristics altered: the average duration of zinc-copper cycles, the regularity with which the cycles recurred, and the number of complex features within a cycle.

Using two different classification models, the researchers achieved 90% accuracy in classifying cases and controls, with sensitivity to autism diagnosis ranging from 85% to 100% and specificity ranging from 90% to 100%.

"These cycles haven't been well documented in the past," said Curtin. "Here we are showing they are critical to neurodevelopment and when they are disrupted, we find that disruption is linked to autism and in fact, can be used to predict autism."

The study also represents a new direction in autism biomarker research, he added. While many studies have assessed exposure levels, this analysis examined cycles to see how metabolism might be disrupted.

"With this research, we are shifting the focus to looking at metabolic cycles to understand how children are processing nutrients, as opposed to just looking at their exposure to toxicants."

Comment