Endogenous ligands for the epidermal growth factor receptor in human breast tumors do not interfere in an epidermal growth factor receptor radioligand binding assay.

Ligands for the epidermal growth factor receptor (EGFR) (e.g. EGF and transforming growth factor alpha, TGF alpha) may be included in membrane preparations from human breast tumors, either complexed or not to EGFR. Assessment of EGFR in human placental membranes (HPM) after preincubation with human EGF (hEGF) or human TGF alpha (hTGF alpha) indicated that the presence of these ligands in a membrane preparations did not affect the apparent number of binding sites, but only resulted in an increased apparent dissociation constant (Kd). To obviate the effect of endogenously bound EGFR ligands in a radioligand binding assay, an acid treatment procedure has been used recently. In the present study we show that acid treatment of mammary tumor membranes does not result in increased EGFR levels but does result in an EGFR enriched membrane preparation by elimination of contaminating cytosol proteins. This however also occurred by washing the membranes with assay buffer at neutral pH. From these experiments we conclude that endogenous ligands, if present in membrane fractions from human breast tumors and occupying EGFR, do not interfere in a multipoint radioligand binding assay.

Epidermal growth factor receptor and prognosis in human breast cancer: a prospective study.

The prognostic value of Epidermal Growth Factor Receptor (EGFR) in human breast cancer is a matter of debate. We conducted a prospective study that included 459 unselected patients with primary breast cancer (median follow-up 24 months) to assess the prognostic value of EGFR. EGFR was assessed using a standardized radioligand binding assay. Univariate analysis showed that EGFR is a factor indicative of a poor prognosis with respect to Disease Free Survival (DFS, P = 0.03) and Overall Survival (OS, P = 0.002), if an EGFR level of 50 fmol/mg of membrane protein is introduced as a cut-off for EGFR-positivity. Multivariate analysis showed that EGFR was not an independent factor. This prospective study shows that EGFR, although not an independent factor, is indicative of poor prognosis in human breast cancer.

Epidermal growth factor receptor levels increase but epidermal growth factor receptor ligand levels decrease in mouse mammary tumors during progression from hormone dependence to hormone independence.

Twenty-six serially transplanted Grunder (GR) strain mouse mammary tumors were analyzed for epidermal growth factor receptor (EGFR) and EGFR-ligand levels, in addition to steroid hormone receptors (estrogen receptor, ER, progesterone receptor, PgR). In concordance with earlier studies, hormone dependent (HD) and hormone responsive (HR) tumors were found to be positive for both ER and PgR, whereas hormone independent (HI) tumors contained only 30% of the ER concentration that was found in the HD tumors. PgR was undetectable in HI tumors. HI tumors contained 2.5 to 3-fold higher EGFR levels than HD/HR tumors, an observation which shows remarkable concordance with studies on EGFR in human breast cancer. On the other hand, the level of EGFR-ligand(s) was positively associated with ER levels and was three-fold higher in HD/HR tumors than in HI tumors. The low EGFR in HD/HR tumors relative to HI tumors may be the result of downregulation by EGFR ligands produced under ER control. During progression to hormone independence this downregulation of EGFR is then abolished in absence of ER. The increase in EGFR may therefore be a secondary effect rather than a key event in the progression to hormone independence in this mouse mammary tumor model.

High-performance liquid chromatography of 125I-labeled mouse epidermal growth factor radioiodinated by six different methods.

Six different procedures for radioiodination of mouse epidermal growth factor (EGF) all resulted in a heterogeneous 125I-labeled EGF preparation, as analyzed by reversed-phase HPLC. EGF preparations that had been iodinated with Chloramine T, lodogen, or lodo-beads were found mainly to consist of oxidized 125I-labeled EGF moieties. In contrast, the heterogeneous 125I-labeled EGF preparations obtained by using iodine monochloride, Protag-125, or lactoperoxidase-glucose oxidase-coupled beads (Enzymobeads) contained insignificant amounts of oxidized EGF entities. Ligand equivalence analysis (LEA) of distinct HPLC column fractions, obtained after preparative separation of Chloramine T-125I-labeled EGF, showed that the receptor-binding affinity of the tracer in all subfractions was less than the affinity of unlabeled EGF. This implies that HPLC purification of these 125I-labeled EGF preparations does not yield 125I-labeled EGF preparations with ligand equivalence. However, all but one HPLC column fraction of Enzymobeads-125I-labeled EGF showed ligand equivalence. Despite the small amount of the nonequivalent component in the Enzymobeads-labeled tracer, the nonchromatographed 125I-labeled EGF preparation showed ligand equivalence. No significant differences were observed in the maximal binding capacity of the different 125I-labeled EGF preparations.

Epidermal growth factor receptors in human breast cancer: a plea for standardisation of assay methodology.

In a prospective study 200 primary human breast cancer specimens were analysed for epidermal growth factor receptor (EGFR) content by means of a multiple point ligand binding assay, proposed by the EORTC Receptor Study Group to be the standard EGFR assay. In 54% of the tumours the presence of saturable high affinity binding sites for epidermal growth factor could be demonstrated. The median EGFR level was 34 fmol/mg of membrane protein, the median Kd 0.50 nmd. Univariate analysis of the EGFR data stratified according to patient age, menopausal status, tumour size, axillary lymph node status, histological tumour type, tumour differentiation grade or the tumours' steroid hormone receptor status showed EGFR to be positively associated with younger age (P = 0.03), tumour dedifferentiation (P = 0.04) and steroid hormone receptor negativity (P less than 0.001). No association between EGFR and menopausal status, tumour size, axillary lymphnode status or histological tumour type could be demonstrated.

Scintillation proximity assay to study the interaction of epidermal growth factor with its receptor.

Scintillation Proximity Assay (SPA), which does not require the physical separation of receptor bound and free ligand, was applied to study the interaction of Epidermal Growth Factor (EGF) with its receptor (EGFR) in membrane preparations from human placenta. Fluomicrospheres to which the monoclonal anti-EGFR antibody R1 was coupled, were used. Kinetic binding data of the association of 125I-labeled EGF binding to the receptor at 20 degrees C could be fitted according to a double exponential model, which is consistent with the presence of fast and slow associating EGF binding sites. Dissociation kinetics revealed that perturbation of equilibrium conditions rapidly occurs upon washing. Multiple point Scatchard analysis of equilibrium 125I-labeled EGF binding data revealed curvilinearity, indicating the presence of both high and low affinity EGF binding sites. We conclude that SPA is an interesting new tool in the exploration of the interaction of ligands with their receptors, which allows detailed ligand-receptor studies under precise in situ conditions.

Six methods for direct radioiodination of mouse epidermal growth factor compared: effect of nonequivalence in binding behavior between labeled and unlabeled ligand.

Mouse epidermal growth factor (EGF) was radioiodinated by six different direct iodination methods. The 125I-labeled EGF preparations were distinguished by analyzing the binding of the radioligand to the EGF receptor (EGFR)-containing human placental membranes. The receptor-binding affinity of EGF labeled with Chloramine T was less than the affinity of unlabeled EGF, which precluded an accurate determination of the specific radioactivity of the 125I-labeled EGF preparation by "self-displacement analysis." Scatchard analysis of competitive binding data (increasing concentrations of unlabeled EGF) obtained with commercially prepared 125I-labeled EGF (Chloramine T method), according to the specific radioactivity stated by the manufacturer, resulted in a substantial underestimation of the apparent number of receptors. Iodination of EGF with Iodogen or Iodo-beads, reagents claimed to be more gentle because of their solid state, also yielded 125I-labeled EGF preparations that were not equivalent to the native EGF in receptor binding. In contrast, equivalence in the ligand-receptor interaction between labeled and unlabeled EGF could be achieved by iodinating EGF with iodine monochloride (ICl), Protag-125, or lactoperoxidase-glucose oxidase-coupled beads (Enzymobeads). Scatchard plots of saturation and competitive binding data obtained with these 125I-labeled EGF preparations produced identical results for apparent receptor number and apparent dissociation constants. Such radioiodinated EGF preparations yield relevant binding data in competition studies of labeled and unlabeled EGF.

Epidermal growth factor receptor-negative tumors are predominantly confined to the subgroup of estradiol receptor-positive human primary breast cancers.

A total of 725 human primary breast tumor biopsy samples were analyzed for epidermal growth factor receptor (EGFR) content, using a multiple-point EGFR assay standardized in accordance with the recommendations of the European Organization for Research and Treatment of Cancer Receptor Study Group. After the establishment of a lower cell membrane protein threshold of 0.2 mg of membrane protein per ml of assay buffer, the results of 27% (194 samples) of the EGFR determinations were excluded from the study because of insufficient assay membrane protein content. Of the remaining 531 breast tumor biopsy samples, 57% (302 samples) were shown to be EGFR positive by Scatchard analysis, with a median value of 40 fmol/mg of membrane protein. Of the breast tumor biopsy samples, 72% (380 samples) were estrogen receptor (ER) positive, and 65% (344 samples) were progesterone receptor (PgR) positive. EGFR positivity was found in 46% (173 of 380) of ER-positive and in 85% (129 of 151) of ER-negative breast tumor biopsy samples (P less than 0.0001), as well as in 49% (168 of 344) of PgR-positive and in 72% (134 of 186) of PgR-negative breast tumor biopsy samples (P less than 0.0001). Mean EGFR levels in ER-positive breast tumor biopsy samples were lower than they were in ER-negative ones, 40 +/- 31 (SD) against 72 +/- 55 fmol/mg of membrane protein (P less than 0.0001). Similarly, mean EGFR levels in PgR-positive breast tumor biopsy samples were lower than they were in PgR-negative ones, 41 +/- 29 against 70 +/- 56 fmol/mg of membrane protein (P less than 0.0001). Both EGFR positivity and EGFR levels decreased with increasing steroid hormone receptor levels. A multivariate analysis showed only ER to be independently associated with EGFR.