Lower IGFBP-7 blood levels linked to cognitive impairment in adolescent schizophrenia

In a finding that could reshape how scientists think about the biological roots of teenage psychosis, researchers in China have reported that adolescents suffering their first episode of schizophrenia carry significantly lower blood levels of a protein called IGFBP-7 — and that the lower the level, the worse their cognitive performance. The study, published in Annals of General Psychiatry, offers one of the clearest links yet between a measurable blood protein and the devastating thinking problems that define adolescent-onset schizophrenia, a rare and particularly severe form of the illness.

Adolescent-onset schizophrenia, defined by symptoms emerging before the age of 18, affects roughly 0.3 percent of the population. Compared with the adult form of the disorder, it is associated with more profound neurodevelopmental abnormalities, worse long-term prognosis and more severe cognitive deficits. Clinicians have long known that cognitive dysfunction — impairments in memory, attention, language and problem-solving — can appear years before hallucinations and delusions, yet the molecular drivers of these deficits remain poorly understood. The new study points to a surprising candidate: insulin-like growth factor binding protein-7, or IGFBP-7, a protein best known for its roles in blood vessel formation, cellular aging and inflammation.

IGFBP-7 is an extracellular matrix protein with a complicated biography. It belongs to the insulin-like growth factor binding protein family, which traditionally regulates the availability and signaling of IGF-1 and IGF-2, growth factors essential for neuronal survival and synaptic plasticity. Yet IGFBP-7 is an unusual member: it binds IGF-1 and IGF-2 with roughly 100-fold lower affinity than its five classical relatives and actually binds insulin more strongly than the IGFs. Beyond growth factor regulation, it has emerged as a marker of vascular aging and neuroinflammation. Elevated levels are seen in obesity, aging-related diseases and major depressive disorder, where higher IGFBP-7 tracks with worse cognition. Animal studies have found that decreased IGFBP-7 in cerebrospinal fluid accompanies cognitive impairment in rats, and in humans, circulating levels correlate with postoperative cognitive dysfunction. Until now, however, no one had examined IGFBP-7 in adolescent-onset schizophrenia.

The research team, led by Haidong Yang and Jie Hou as joint first authors, with Qing Tian and Xiaobin Zhang as senior investigators, enrolled 91 patients aged 13 to 18 who were experiencing their first episode of schizophrenia and had never taken antipsychotic medication. This drug-naïve design is critical: previous studies in adult patients have shown that IGFBP-7 levels rise after antipsychotic treatment, so measuring patients before any medication eliminates the possibility that drugs, rather than the illness itself, explain the differences. Forty healthy controls, matched for age, sex and education, served as the comparison group. Patients with other psychiatric or neurological conditions, serious physical illness, substance abuse or recent use of metabolism-altering medications were excluded.

Each participant provided fasting blood samples drawn in the early morning, which were processed and stored at minus 80 degrees Celsius before analysis using Luminex liquid suspension array technology — a bead-based immunoassay capable of measuring multiple proteins simultaneously with high sensitivity. All samples were run in duplicate, and quality control coefficients of variation stayed below 10 percent. Cognitive function was assessed with the Repeatable Battery for the Assessment of Neuropsychological Status, a 30-to-45-minute battery covering five domains: immediate memory, visuospatial construction, language, attention and delayed memory. Psychiatric symptoms were rated with the Positive and Negative Syndrome Scale using its five-factor model, which separates positive symptoms, negative symptoms, cognitive symptoms, excitement and hostility, and anxiety and depression.

The results were striking. Patients averaged 10.15 nanograms per milliliter of serum IGFBP-7 compared with 11.75 nanograms per milliliter in controls, a difference that remained significant even after statistically adjusting for the age gap between the groups. Hepatocyte growth factor, a neuroprotective protein involved in neurodevelopment that the team also measured, showed no difference between the groups and no relationship with symptoms or cognition. Within the patient group, lower IGFBP-7 was significantly associated with worse scores on the cognitive factor of the PANSS and with lower total scores on the cognitive battery. Statistical modeling confirmed that IGFBP-7 remained independently associated with overall cognitive performance after controlling for age, sex, body mass index, years of education, illness duration and age of onset.

The magnitude of the association was clinically meaningful. Using modified Poisson regression — a technique that estimates relative risk in cross-sectional data — the researchers found that low IGFBP-7 levels, defined by splitting the sample at the median, independently predicted having adolescent-onset schizophrenia, with a relative risk of 1.298. Translated into population terms, low IGFBP-7 accounted for an estimated 23 percent of schizophrenia cases among exposed individuals and 12.7 percent of cases in the overall study population. In other words, a substantial fraction of illness burden in this sample was statistically associated with reduced levels of a single circulating protein.

Sex emerged as an interesting wrinkle in the data. While sex was a significant confounder in the relationship between IGFBP-7 and cognitive scores — males and females performed differently on neuropsychological testing, as prior literature on schizophrenia would predict — it did not moderate the IGFBP-7-cognition link. The interaction term between sex and IGFBP-7 was not statistically significant, meaning the association between the protein and cognitive impairment held equally for boys and girls. This suggests the underlying biology is a sex-independent process, even though the expression of cognitive deficits differs between the sexes.

Why would reduced IGFBP-7 matter in a developing adolescent brain? The authors propose several converging mechanisms. Adolescence is a critical window for synaptic pruning, myelination and cerebrovascular maturation, processes that depend on precise coordination between blood vessels and neurons. IGFBP-7, also known as angiomodulin, is a cerebrum-specific angiocrine factor that helps maintain vascular integrity and cerebral perfusion. It also modulates IGF signaling, thereby influencing neuronal survival and synaptic plasticity. More recently, work published in Nature Immunology showed that IGFBP-7 expression is regulated by GABAergic neurotransmission and sustains immune homeostasis, positioning the protein as a bridge between neurodevelopment and neuroimmunity. A shortfall of IGFBP-7 during adolescence could therefore disrupt vascular support, trophic signaling and inflammatory balance simultaneously — a plausible recipe for the cognitive decline seen in the disorder.

The finding also fits the broader neurodevelopmental hypothesis of schizophrenia, which holds that the disorder arises from subtle disruptions of brain maturation long before psychosis becomes manifest. Notably, the direction of the IGFBP-7 signal in schizophrenia — reduced levels — contrasts with the elevated levels seen in Alzheimer’s disease and depression, hinting that the protein participates in distinct pathological pathways in neurodevelopmental versus neurodegenerative conditions. In adult schizophrenia patients, IGFBP-7 has been found to increase after antipsychotic treatment, raising the possibility that the protein could serve not only as a biomarker but as a readout of treatment response.

The authors are careful about limits. This was a cross-sectional, single-center study, so it cannot establish whether low IGFBP-7 causes schizophrenia or is a consequence of it, and the relatively small control group and homogeneous regional sample may limit generalizability. Peripheral blood levels may not perfectly mirror what happens in the brain, and metabolic factors such as glucose homeostasis could confound the measurements. Still, the consistency of the findings across symptom ratings, neuropsychological testing and risk modeling, in patients who had never been exposed to antipsychotics, makes IGFBP-7 a compelling candidate biomarker.

If future longitudinal studies confirm the result, a simple blood test measuring IGFBP-7 could eventually help identify adolescents at risk for cognitive decline in early psychosis — a window where intervention matters most. For a disorder in which cognitive impairment often does the most lasting damage, a molecular handle on that impairment would be a genuine advance.