Validation of a fully automated HER2 staining kit in breast cancer.

BACKGROUND: Testing for HER2 amplification and/or overexpression is currently routine practice to guide Herceptin therapy in invasive breast cancer. At present, HER2 status is most commonly assessed by immunohistochemistry (IHC). Standardization of HER2 IHC assays is of utmost clinical and economical importance. At present, HER2 IHC is most commonly performed with the HercepTest which contains a polyclonal antibody and applies a manual staining procedure. Analytical variability in HER2 IHC testing could be diminished by a fully automatic staining system with a monoclonal antibody. MATERIALS AND METHODS: 219 invasive breast cancers were fully automatically stained with the monoclonal antibody-based Oracle HER2 Bond IHC kit and manually with the HercepTest. All cases were tested for amplification with chromogenic in situ hybridization (CISH). RESULTS: HercepTest yielded an overall sharper membrane staining, with less cytoplasmic and stromal background than Oracle in 17% of cases. Overall concordance between both IHC techniques was 89% (195/219) with a kappa value of 0.776 (95% CI 0.698-0.854), indicating a substantial agreement. Most (22/24) discrepancies between HercepTest and Oracle showed a weaker staining for Oracle. Thirteen of the 24 discrepant cases were high-level HER2 amplified by CISH, and in 12 of these HercepTest IHC better reflected gene amplification status. All the 13 HER2 amplified discrepant cases were at least 2+ by HercepTest, while 10/13 of these were at least 2+ for Oracle. Considering CISH as gold standard, sensitivity of HercepTest and Oracle was 91% and 83%, and specificity was 94% and 98%, respectively. Positive and negative predictive values for HercepTest and Oracle were 90% and 95% for HercepTest and 96% and 91% for Oracle, respectively. CONCLUSION: Fully-automated HER2 staining with the monoclonal antibody in the Oracle kit shows a high level of agreement with manual staining by the polyclonal antibody in the HercepTest. Although Oracle shows in general some more cytoplasmic staining and may be slightly less sensitive in picking up HER2 amplified cases, it shows a higher specificity and may be considered as an alternative method to evaluate the HER2 expression in breast cancer with potentially less analytical variability.

Extra-domain-A fibronectin: a new marker of fibrosis in cutaneous graft-versus-host disease.

One of the major complications that limit the success of allogeneic stem cell transplantation is graft-versus-host disease (GVHD). The major target organ in GVHD is the skin. Cutaneous GVHD can eventually lead to fibrosis of the skin. Fibronectin mediates a variety of cellular interactions with the extracellular matrix. The molecular and functional diversity of fibronectin (FN) arises from alternative splicing of pre-mRNA. In normal circumstances endothelial cells and fibroblasts synthesize FN without the ED-A domain. In tissue repair and pathologic circumstances such as fibrosis, the ED-A domain is expressed. We hypothesize that expression of ED-A FN is upregulated in patients with cutaneous GVHD. In frozen skin biopsies the expression of ED-A FN was measured at the protein level by immunohistochemistry and at the mRNA level by quantitative real-time PCR (qPCR). In normal control skin, immunohistochemistry showed slight deposits of ED-A FN just under the basal layer. The expression of ED-A FN significantly increased in acute cutaneous GVHD (p<0.05) and ED-A FN was strongly deposited in chronic cutaneous GVHD (p<0.001). Quantitative PCR also showed increased expression of ED-A FN mRNA in acute and chronic cutaneous GVHD compared with normal control skin (p=0.07 and 0.039, respectively). The expression of ED-A FN is increased in the skin of patients with cutaneous GVHD measured both with immunohistochemistry and qPCR. ED-A FN is a new marker of fibrosis in the skin of patients with cutaneous GVHD.

Activation of hemostasis after coronary artery bypass grafting with or without cardiopulmonary bypass.

Activation of coagulation, fibrinolysis, and the vascular endothelium occurs after heart surgery with cardiopulmonary bypass (CPB), but the effects of eliminating CPB in patients undergoing coronary artery bypass grafting (CABG) are unknown. Therefore, we compared the hemostatic profiles of off-pump and on-pump CABG patients. Two groups of consecutive patients participating in a larger trial (the Octopus Trial) were randomly allocated to undergo CABG with (n = 20) or without (n = 20) CPB. Platelet numbers and plasma concentrations of P-selectin, prothrombin fragment 1.2 (F1.2), soluble fibrin, d-dimers, and von Willebrand factor (as a marker of endothelial cell activation) were measured and corrected for hemodilution. Compared with the on-pump CABG group, F1.2 and d-dimer levels were significantly lower (P = 0.004 and P = 0.03, respectively) in patients having CABG surgery performed off-pump. In the CPB group, F1.2 (median [interquartile range], 450% of baseline [233%-847%]) and d-dimer (538% [318%-1192%]) peaked in the immediate postoperative period and remained increased until Day 4, whereas in the off-pump group, F1.2 and d-dimer levels increased more gradually and peaked on Day 4 (342% [248%-515%] and 555% [387%-882%], respectively). In both groups, von Willebrand factor concentrations were increased until Day 4 (CPB, 308% [228%-405%]; off-pump, 288% [167%-334%]). Despite heparinization, CABG surgery with CPB was associated with excessive thrombin generation and fibrinolytic activity immediately after surgery. The off-pump group demonstrated a delayed postoperative response that became equal in magnitude to the CPB in the later (20-96 h) postoperative period.

Endothelial cell activation after subarachnoid hemorrhage.

OBJECTIVE: Evidence from animal experiments suggests that endothelial cell activation plays a pathogenetic role in the development of cerebral ischemia after subarachnoid hemorrhage (SAH). We measured plasma concentrations of two markers of endothelial cell activation, i.e., ED1-fibronectin (ED1-fn) and von Willebrand factor (vWf), among patients with aneurysmal SAH. We analyzed the relationships of concentrations to initial clinical conditions, treatment modalities, and the occurrence of delayed cerebral ischemia. METHODS: We collected 123 blood samples from 27 patients with aneurysmal SAH. Aneurysms were treated surgically in 19 cases, were treated endovascularly in 7 cases, and remained untreated in 1 case. Twelve patients developed symptomatic delayed cerebral ischemia. RESULTS: Initial concentrations of ED1-fn (4.3 +/- 3.7 microg/ml) and vWf (17.8 +/- 8.2 microg/ml) were higher than the reference values (ED1-fn, 1.7 +/- 0.9 microg/ml, P < 0.001; vWf, 11.5 +/- 5.2 microg/ml, P = 0.003). Concentrations were higher among patients in poor clinical condition at admission, compared with patients in good clinical condition (mean difference, ED1-fn, 5.7 microg/ml, P = 0.04; vWf, 10.4 microg/ml, P = 0.02). Levels of both markers increased significantly after surgery (mean increase, ED1-fn, 7.5 microg/ml, P = 0.01; vWf, 13.2 microg/ml, P = 0.05) and after ischemic episodes (mean increase, ED1-fn, 8.3 microg/ml, P = 0.02; vWf, 5.0 microg/ml, P = 0.04). CONCLUSION: Plasma concentrations of markers of endothelial cell activation were increased early after SAH and were significantly associated with the clinical condition at admission. We also observed a significant increase in concentrations after surgery and after ischemic episodes. Whether endothelial cell activation is a causal or indirectly related factor in the pathogenesis of delayed cerebral ischemia after SAH is still uncertain.