Disentangling Depression in Women with Diabetes: Evidence for Measure-Dependent Associations with Interleukin-4 and Common Inflammatory Biomarkers

Nicole Beaulieu Perez^1*Maria Paula Gordillo Sierra¹Jackie Finik¹Jason Fletcher¹David B Hanna^2,3Anjali Sharma^2,3Kathryn Anastos^2,3Leah Helane Rubin^4,5Gail D'Eramo Melkus¹Bradley Aouizerat⁶

• ¹New York University Rory Meyers College of Nursing, New York, United States
• ²Albert Einstein College of Medicine, New York, United States
• ³Montefiore Einstein Medical Center, New York, United States
• ⁴Departments of Neurology, Psychiatry and Behavioral Sciences, and Molecular and Cellular Pathobiology, The Johns Hopkins University School of Medicine, Baltimore, United States
• ⁵Department of Epidemiology, Johns Hopkins University Bloomberg School of Public Health, Baltimore, United States
• ⁶Department of Oral and Maxillofacial Surgery, New York University College of Dentistry, New York, United States

Abstract Background: Women living with type 2 diabetes (T2D) face elevated risks for depression and its consequences, including early mortality, yet depression is underrecognized in this population. Depression is a heterogeneous phenotype lacking objective diagnostic biomarkers, with symptoms spanning multiple inconsistently assessed dimensions across various measures. Converging evidence implicates inflammation in depression and links depression with T2D. Here, we explored associations between inflammation and various measures and dimensions of depressive symptoms. Methods: This cross-sectional pilot study enrolled 38 women with T2D from the MACS/WIHS Combined Cohort Study (MWCCS) Bronx site from 2022-2023. Serum inflammatory biomarkers (hsCRP, INF-𝛾, IL-1β, IL-1RA, IL-2, IL-4, IL-6, IL-8, IL-10, TNF-⍺) were measured and analyzed via Luminex. We performed Spearman correlation analysis using log-transformed biomarker levels and the multidimensional Center for Epidemiological Studies Depression scale (CES-D), alongside unidimensional Patient-Reported Outcomes Measurement Information System (PROMIS) measures for depression, fatigue, sleep, and anxiety, in this exploratory analysis. Results: Participants were on average 61.4 (SD 4.8) years old, 71% Black, and 32% Hispanic. While all women had T2D, 82% were also living with HIV. Mean BMI of 34.7 kg/m2 (SD 7.4) was high, but HbA1c of 6.5% (SD 1.3%) indicated fairly adequate glycemic control. Among participants with HIV, 94% were taking antiretroviral therapy. Mean high sensitivity C-reactive protein (hsCRP) of 4.1 mg/L (SD 3.7) signaled moderate inflammation in this population. IL-4 demonstrated significant negative associations with PROMIS-Depression (rs= 0.35; 95% CI -0.61, -0.03; p=0.034) and PROMIS-Anxiety scores (rs=-0.37; 95% CI -0.62, -0.05; p=0.025), but associations with CES-D were not significant. hsCRP and IL-6 were positively correlated with CES-D and negatively correlated with PROMIS-Depression, although these associations did not reach statistical significance. PROMIS-Sleep was moderately associated with IL-8 (rs =0.39; 95% CI 0.06, 0.64; p=0.021). Conclusions: While preliminary, our findings suggest that associations between inflammatory biomarkers and depression may not be consistent across all depressive measures or symptom dimensions. Although larger samples with repeated measures are needed, findings from this exploratory study suggest that including inflammatory measures beyond hsCRP and IL-6, together with tools that capture distinct depressive symptom dimensions, may help to inform future precision mental health research.