Background: Fragile X syndrome (FXS) is the most common inherited form of intellectual disability affecting 1 in 4,000 males and 1 in 6-8,000 females. FXS is caused by a trinucleotide expansion in the 5’UTR of the Fragile X Mental Retardation (FMR1) gene which in full mutation carriers (>200 repeats) leads to hypermethylation and transcriptional silencing of the gene and lack of expression of Fragile X Protein (FXP, formerly known as Fragile X Mental Retardation Protein, FMRP). Phenotypic presentation of FXS is highly variable, and molecular markers explaining or predicting this variability are lacking. Recent studies suggest that trace amounts of FXP can be detected even in fully methylated individuals and may have clinical relevance; however, the lack of available reproducible, sensitive assays to detect FXP in peripheral tissue makes evaluation of peripheral FXP as a source of clinical variability challenging. Methods: We optimized a Luminex-based assay to detect FXP in dried blot spots for increased reproducibility and sensitivity by improving reagent concentrations and buffer conditions. The optimized assay was used to quantify FXP in 187 individuals (101 males, 86 females; 0-78.4 years) including 35 typically developing controls (24 males, 11 females), 103 individuals carrying full mutations (70 males, 33 females), and 49 individuals with premutations (7 males, 42 females). A subset of these individuals showed repeat number or methylation mosaicism. We investigated the clinical relevance of peripheral FXP levels by examining its relationship with general intellectual functioning in a subset of individuals with available IQ scores. Results: We show that the optimized assay is highly reproducible and detects a wide range of FXP levels. Mosaic individuals had, on average, higher FXP levels than fully methylated individuals, and trace amounts of FXP were consistently detectable in a subset of individuals with full mutation FXS. IQ scores were positively correlated with peripheral FXP levels in male and female individuals with full mutation FXS. Conclusions: We demonstrate that our optimized Luminex-based assay to detect FXP is reproducible, highly sensitive, and related to the core intellectual disability phenotype. Further, our data suggest that trace amounts of FXP detectable in dried blood spots of individuals with FXS could be clinically relevant and may be used to stratify individuals with FXS for optimized treatment. Future studies are needed with larger sample sizes, evaluating FXP across development and expanded analysis of the relevance of FXP levels for behavioral and electrophysiological phenotypes in FXS.