Markedly raised levels of C-reactive protein are associated with culture-proven sepsis or necrotising enterocolitis in extremely preterm neonates.

AIM: A serious inflammatory process is suspected when C-reactive protein (CRP) is very high, and we established the causes and outcomes when CRP was >100 mg/L in neonates. METHODS: We retrospectively reviewed all 277 episodes where CRP exceeded 100 mg/L between January 2007 and December 2011 at a tertiary neonatal unit. RESULTS: Of the 6025 neonates admitted during the study period, 258 had CRP >100 mg/L at least once. The overall mortality rate was 44/258 (17%); 36 died within 7 days of CRP >100 mg/L, and 34 were extremely preterm infants. CRP exceeded 100 mg/L in 106 infants within the first 3 days of life - 74 term, 25 preterm and seven extremely preterm - with no infection identified in 81%. In contrast, infections were found in 87% of the 171 episodes from day four of life - 129 extremely preterm, 23 preterm and 19 term - predominantly coagulase-negative staphylococcus sepsis and necrotising enterocolitis. CONCLUSION: Markedly elevated CRP in the first 3 days of life was most likely to affect term neonates (74/106) with no infectious cause (81%). However, CRP >100 mg/L from the fourth day of life was most likely to affect extremely preterm neonates (129/171) and have an infectious cause (87%).

Seromic analysis of antibody responses in non-small cell lung cancer patients and healthy donors using conformational protein arrays.

Analysis of antibody responses to self-antigens has driven the development of the field of tumor immunology, with the identification of many protein targets found in cancer but with limited expression in normal tissues. Protein microarray technologies offer an unprecedented platform to assay the serological response of cancer patients to tumor antigens in a comprehensive fashion, against many proteins simultaneously. We developed an array containing 329 full-length proteins, originally identified as antigenic in various cancer patients by serological expression cloning (SEREX), that were immobilized as folded, functional products accessible for antibody binding. To validate the use of these microarrays, we selected 31 sera from non-small cell lung cancer patients previously known to react to the following antigens by ELISA: LAGE-1/CTAG2, MAGEA4, TP53, SSX and SOX2. These sera were compared with 22 sera from healthy donors for reactivity against a series of antigens present on microarrays. The sensitivity and specificity of the arrays compared favorably with standard ELISA techniques (94% concordance). We present here a stringent strategy for data analysis and normalization that is applicable to protein arrays in general, and describe findings suggesting that this approach is suitable for defining potential antigenic targets for cancer vaccine development, serum antibody signatures with clinical value, characterization of predictive serum markers for experimental therapeutics, and eventually for the serological definition of the cancer proteome (seromics).