Evans, Joni K.

Background TP53 mutational status ( TP53 ms) is prognostically important in myeloid neoplasms. Immunohistochemistry (IHC) of p53 has been used as a surrogate for TP53 mutations, with faster results than next-generation sequencing (NGS) and cytogenetic analysis. Given the challenges posed by heterogeneous p53 IHC expression, this study aims to evaluate the interobserver interpretation of p53 stain as a surrogate marker for TP53m . Methods Fifty-three bone marrow (BM) specimens previously diagnosed as myeloid neoplasms or nonneoplastic BM specimens with available concurrent TP53ms were evaluated. p53 IHC staining was performed on all 53 core biopsy or clot sections. TP53 mutation was found in 24 cases ( TP53 m). Twenty-nine specimens were TP53 wild-type and served as the control group. Six hematopathologists (HPs) blinded to the diagnosis and TP53 m independently scored the p53 staining pattern of the biopsies/clot sections for overall staining percentage (OSP) and staining intensity (dark staining percentage as a percentage of total cells (DCP)). Results The intraclass correlation was excellent for OSP (0.927) and good for DCP (0.848). For OSP, a cutoff of 30% resulted in sensitivity of 83.3%, specificity of 89.7%, PPV of 87.0%, and NPV of 86.7%. For DCP, a cutoff of 10% achieved sensitivity of 87.5%, specificity of 96.6%, PPV of 95.4%, and NPV of 90.3%. Additionally, DCP significantly correlated with the VAF of TP53 m (R = 0.469, p = 0.028). Conclusion High inter-rater reliability for OSP and DCP supports integrating p53 IHC into bone marrow evaluations for rapid TP53 m assessment. With superior specificity and predictive values, DCP proves to be a more reliable diagnostic marker for TP53 mutations. While this approach shows promise for myeloid neoplasms, further validation is needed to overcome challenges and expand its clinical utility for tailored treatment strategies.

AbstractBackgroundThe effect of an intensive lifestyle intervention (ILI) on Alzheimer’s disease (AD) blood‐based biomarker levels and cognitive outcomes among individuals with type 2 diabetes (T2D) is unknown. We examined AD biomarkers and cognitive outcomes in the Look AHEAD Study.MethodParticipants aged 45‐76 years were randomized at baseline to a 10‐year ILI or a diabetes support and education (DSE) condition. Cognitive assessments at years 8‐18 included measures of attention, executive function, global function, and memory. Mild cognitive impairment (MCI) and probable dementia were adjudicated by an expert panel. Stored baseline and end‐of‐intervention plasma samples were analyzed with the Quanterix Simoa HD‐X Analyzer. Changes in Aβ42/Aβ40, ptau181, neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP) were evaluated in relation to randomization status, demographic characteristics, and cognitive outcomes. Cognitive scores and biomarker levels were converted to z‐scores. Cognitive trajectories were evaluated using mixed models to account for repeated measures; logistic regression was used for a composite outcome of MCI/dementia. Models were adjusted for baseline characteristics (age, sex, education, race/ethnicity, body mass index (BMI), CVD history, hypertension, diabetes duration, eGFR), APOE‐ε4, randomization arm, time from randomization, and time between blood draw and cognitive testing.ResultThe mean age of 779 participants was 61.4 years; 56% = female, 22.8% = APOE ε4+, mean BMI = 34.8, baseline diabetes duration >5 years = 57%, 16.7% Black, 12.5% Hispanic, 66.2% White, and 4.7% other racial and ethnic groups. There were no associations between the intervention and biomarker levels at baseline or biomarker change over time. Increases in NfL levels significantly predicted lower memory (‐0.04 ±0.01; p = 0.002), global cognition scores (‐0.05 ±0.02; p = 0.011), executive function (‐0.04 ±0.01; p = 0.01), as well as increased risk of MCI/probable dementia (odds ratio [OR] 1.10, 95% CI 1.02‐1.19). Increases in GFAP predicted lower memory (‐0.10 ±0.03; p = 0.0001), global cognition (‐0.09 ±0.03; p = 0.009), executive function scores (‐0.12 ±0.03; p<0.0001), and increased risk of MCI/probable dementia (OR 1.24, 95% CI 1.09‐1.41).ConclusionThere was no association between the intervention and changes in biomarker levels. Increasing levels of NfL and GFAP over the ∼10 year period were associated with reduced cognitive performance and increased risk of MCI/probable dementia.