Infantile autism and urinary excretion of peptides and protein-associated peptide complexes.

Attwood et al. (1988) collected morning urine from three groups of young adult males: autistic, intellectually disabled, and neurotypical. They measured peptide levels and protein-bound peptide complexes. The goal was to see if autistic people excrete unique peptide patterns that could serve as a biological test for autism.

Peptide levels overlapped across all three groups. No pattern reliably separated autistic participants from the others. The authors concluded that urinary peptide profiling is not a useful biomarker for autism in young adults.

Eto et al. (1992) repeated the idea by testing urinary pterins and folate in autistic children. They also found no group differences, backing up the null result. Root et al. (2017) later looked at urinary IAG, a different metabolite, and again saw no autism signal. Together these three studies form a consistent ‘no-go’ line of evidence against urinary metabolite screening.

McCarron et al. (2002) appears to disagree: girls with Rett syndrome did show high urinary peptides. The contradiction is solved by population choice. Rett syndrome is a distinct genetic condition, not autism, so the positive finding does not overturn the 1988 null result in autistic males.

Lussu et al. (2017) used newer 1H-NMR metabolomics and claimed a unique urinary ‘fingerprint’ for Italian autistic children. Their method was broader than the 1988 peptide assay, yet even this advanced scan has not moved into routine clinical use, underscoring how elusive reliable urinary biomarkers remain.

If parents ask about special urine tests to confirm autism, you can confidently say the science shows no benefit. Save assessment time for gold-standard tools like ADOS and caregiver interviews. Skip costly peptide labs and focus resources on skill-based interventions that actually change behavior.