Analytical and preanalytical biases in serum proteomic pattern analysis for breast cancer diagnosis.
Karsan A, Eigl BJ, Flibotte S, Gelmon K, Switzer P, Hassell P, Harrison D, Law J, Hayes M, Stillwell M, Xiao Z, Conrads TP, Veenstra T
Purpose
Conclusion
Studies (1)
Purpose
The purpose of this study was to assess the influence of preanalytical (specimen collection source) and analytical processing (day of analysis) variables on the spectral profile of serum by comparing specimens obtained from three clinics and processed and analyzed by SELDI-TOF MS on different days. Blood specimens were collected from patients diagnosed with breast cancer or a benign growth.
Summary of Findings
Two SELDI-TOF analytical machine learning algorithms were unsuccessful in differentiating serum specimens based on cancer status (carcinoma versus benign), but successfully segregated specimens based on the clinic that collected the specimen, the day the chip was prepared, and the day the chip was read.
Biospecimens
FluidSerumFluidBloodPreservation Types
FrozenDiagnoses
NeoplasticCarcinomaNeoplasticBenignPlatforms
ProteinSELDI-TOF MSPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Preaquisition Diagnosis/ patient condition BenignBreast cancerBiospecimen Acquisition Locale of biospecimen collection 3 separate medical clinicsBiospecimen Acquisition Method of fluid acquisition Clinic dependent phlebotomy methodology (unspecified)