WO1996024065A1 - Methods for the detection of pro-cathepsins and uses thereof - Google Patents

Abstract

This invention is directed to a method of prognosis for a breast cancer patient involving quantitatively detecting pro-cathepsin B, D, or L in a sample of a bodily fluid which may include plasma, serum, urine, saliva, lymphatic fluids, whole blood, or nipple aspirates. The detection of pro-cathepsin may be performed utilizing a cathepsin binding agent as a capture agent. The binding agent may be monoclonal or polyclonal antibodies specific for pro-cathepsin or an enzyme inhibitor such as pepstain. The prognosis may involve the likelihood of recurrence of cancer, of relapse, or of survival.

Description

METHODS FOR THE DETECTION OF PRO-CATHEPSINS AND USES THEREOF This application claims priority of U.S. Serial No. 08/382,876 filed February 3, 1995, the contents of which are hereby incorporated by reference into the present application.

Background of the Invention

Throughout this application, various publications are referenced by author and date. Full citations for these publications may be found listed alphabetically at the end of the specification immediately preceding the claims. The disclosures of these publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art as known to those skilled therein as of the date of the invention described and claimed herein.

There is clearly a need for new and improved methods for the diagnosis and prognosis of cancer. Nowhere is this situation more apparent than in the case of breast cancer. Approximately one of every eight women will contract breast cancer during their lifetime and in the U.S. alone there were about 186,000 new cases in 1993 (McGuire, 1992) . In about two thirds of these cases the disease is confined to the breast and with no sign of lymph node involvement. About 70% of these "node-negative" women remain disease-free without adjuvant therapy. Unfortunately, the other 30% relapse within a five year period (Neville, 1991) . Therefore, a major challenge in the management of breast cancer is to find ways of distinguishing this subset of node-negative women who are at a high risk of recurrence. This would make it possible to concentrate therapeutic efforts on those who need it most and spare the majority of women from the severe side effects associated with chemotherapy. The two most useful parameters in determining follow up therapy in breast cancer are tumor size and node invasiveness . Tumors of less than 1 cm and the absence of metastasis in the axillary lymph node generally indicate very good prognosis. However, 30% of these patients relapse and there are currently no reliable methods for identifying this group. Certain groups of patients with a probability of recurrence receive benefit from adjuvant systemic therapy but there is much debate as to whether all node-negative patients should receive some form of treatment (Ingle, 1990) . New prognostic factors that can identify the group at high risk of recurrence would help to define and develop more effective therapies and make more efficient use of limited health care resources.

Besides tumor size and node status, numerous parameters are currently being used in an attempt to gain prognostic information. These include estrogen and progesterone receptor status, histopathological classification, histologic and nuclear grade, and DNA-ploidy and S-phase fraction. In addition, several new factors of potential value are the epidermal growth factor receptor, oncogenes such as eriB-2/neu and myc, tumor suppressor genes like p53 and NM23, angiogenic factors like topoisomerase 2, bFGF, and proteinases such as cathepsin D and urokinase-type plasminogen activator (uPA) . Each of these markers has shown merit but are a long way from being established as reliable prognostic indicators.

Proteinases are attractive as prognostic and diagnostic tools since it is well established that proteolytic activity is necessary for the process of invasion and metastasis. Three critical steps in the transition from an in si tu carcinoma to one that is metastatic include the attachment of tumor cells to the basement membrane, localized proteolysis and the eventual migration of the tumor cells through the digested stroma where they gain access to the vascular system (Liotta, 1990) . Circulating tumor cells are then capable of attaching to the endothelial surface and initiating metastatic colonies. Central to this process is the involvement of several classes of proteolytic enzymes. An imbalance in localized proteolytic activity may involve increases in the proteolytic enzymes, reduced levels of their endogenous inhibitors or both. Classes of proteinases that have been implicated in the metastatic process include matrix etalloproteinases, cathepsins B and L (cysteine proteinases) , cathepsin D (aspartic proteinase) and plasminogen activator (serine proteinase) . Numerous in vi tro as well as clinical studies have suggested that changes in the expression, processing or activation of proteinases could play a critical role in the conversion of a tumor cell to the metastatic phenotype. The measurement of these changes, therefore, has the potential to provide valuable information on the progression of cancer. Since many proteinases are synthesized as latent forms and must themselves be activated by proteolysis, this raises the possibility for a complex cascade of proteolytic events which could be involved in many aspects of the regulation of cell growth. It should be noted that prostate specific antigen (PSA) , which is an extremely useful marker for prostate cancer, is a proteolytic enzyme.

Lysosomal proteinases are beginning to attract a great deal of attention for having a role in the process of invasion and metastasis. This is particularly well documented for the case of cathepsin D (CD) (Rochefort, 1990; Rochefort, 1992) . CD is a normal lysosomal proteinase which is synthesized as 52 kDa pro-enzyme and processed to the active, two chain form of 34 and 14 kDa subunits (Rochefort, 1990) . A relationship between CD and cancer was first proposed when it was discovered that the 52 kDa proenzyme was secreted by the hormone-dependent human breast cancer cells, MCF-7 (Westley, 1980) . In MCF-7 cells, secretion is specifically increased by estrogens and decreased by anti- estrogens whereas, in normal cells, little or no 52 kDa precursor is secreted. Expression is increased at the level of both mRNA and protein (Capony, 1989) . The mechanism for how CD is diverted from its normal processing and lysosomal targeting to the secretory pathway is not well understood.

In vi tro studies support a role for CD in cell transformation (Cavailles, 1988; Vignon, 1986; Briozzo, 1988; Garcia, 1990) . The level of CD increases in response to mitogens such as estrogen, EGF, FGF and IGF-1. When applied to estrogen-deprived MCF-7 cells, CD proved to be mitogenic suggesting that it may function as an autocrine growth regulator. CD is capable of digesting extracellular matrix proteins in in vi tro models. And finally, transfection of the CD gene into rat cells increases their tumorogenicity when injected into nude mice.

Currently at least a dozen clinical studies suggest cathepsin D as a possible prognostic marker in breast cancer (Ingle, 1990) . Immunohistochemistry (IHC) has shown that CD is elevated in both benign and malignant proliferative mastopathies compared to normal mammary tissue (Garcia,

1986) . There was a correlation with the most proliferative

(high risk) lesions. In a recent IHC study of 262 node- negative breast cancer patients, CD was a powerful and independent prognostic factor in predicting overall survival

(Isola, 1993) . In breast tumor cytosols routinely prepared for estrogen/progesterone receptor assays, CD was shown by immunoassay to be an independent factor for identifying breast cancers with poor prognosis and was especially predictive for patients that were node negative (Spyratos, 1989) . In a retrospective study on 162 node-negative patients, Kute, et al . (Kute, 1992) measured CD levels by immunoassay and enzymatic activity and compared them to other factors such as steroid receptors, tumor size, histological grade, ploidy and cell cycle kinetics. It was concluded that CD was the most useful prognostic indicator. A study in which 858 breast tumor cytosols were examined by immunoassay, high total CD may be predictive of shorter disease-free survival but only in node positive patients (Seshadri, 1994) . High CD was independent of other popular prognostic markers but was associated with Her-2/ι.eu amplification. The results from the first 5-year prospective study (Pujol, 1993) suggested that CD may have prognostic value in breast cancer.

However, not all studies support CD as a useful prognostic marker. In particular, immunohistochemical studies have failed to confirm the prognostic relevance of CD. Merkel, et al . (Merkel, 1991) and Armas, et al . (Armas, 1994) did not find any relationship between levels of CD staining and disease-free survival. Henry, et al . (Henry, 1990) showed that high levels of CD actually seemed to correlate with good prognosis. A study of CD expression by IHC and immunoblot analysis failed to improve the prognostic evaluation of node negative breast cancer patients (Ravdin, 1994) . In an in vi tro study of a series of breast cell lines that secrete different levels of CD, no correlation was found between CD secretion and invasive behavior in the Boyden chamber assay (Johnson, 1993) . In a recent immunohistochemical study of 481 breast carcinomas, a high percentage of CD positive tumor cells correlated with better overall survival whereas immunostaining of the surrounding stroma and macrophages indicated poor prognosis (Haerslev, 1994) . The investigators were able to create a cathepsin D index which was found to be a significant prognostic parameter. Studies like these indicate that the role of CD in breast cancer is extremely complex and may involve the interplay of many different cell types. Clearly, the prognostic significance of CD remains controversial and additional work is required using standardized methods before it is established as a useful marker in breast cancer. Nearly all studies on CD have been confined to tumor samples or cytosols. One report described the measurement of total

CD in plasma and concluded that it was not a useful marker

(Brouillet, 1991) . One other study used immunoblots to detect pro and mature CD is sera and noted little difference between breast cancer patients and controls (Schultz, 1994) .

However, no standards were used to confirm the identity of the bands on their blots. In fact, the level of pro and mature CD in their experiments was substantially higher than we detect and it is possible that they were actually measuring the heavy and light chain of human IgG which migrate at similar molecular weights.

A similar story is beginning to emerge for cathepsin B (CB) and cathepsin L (CL) (Sloane, 1990; Kane, 1990) . These lysosomal enzymes are cysteine proteinases and, like CD, appear to be abnormally processed and secreted by tumor cells in their pro-forms. Evidence is accumulating that CB and CL could be useful prognostic cancer markers (Chauhan, 1991; Lah, 1992) .

Much of the controversy surrounding the utility of cathepsins as prognostic markers may be the result of the way in which they are measured. Although it is the pro-form that may be abnormally secreted by tumor cells, currently available immunoassays (two kits are commercially available for CD) and IHC reagents are directed against the mature form and thus measure total enzyme levels. Attempts to measure increased secretion of the precursor by tumor cells is often obscured by high background levels of the endogenous lysosomal form. The normal lysosomal enzyme is made by all cells, especially stromal macrophages which are a variable contaminant of breast tumor samples. In fact, activated macrophages normally secrete a certain level of active lysosomal enzymes. Therefore, assays and IHC reagents that specifically detect the pro-form would provide more reliable data and be more predictive of the status of a tumor. There is one report in which pro-CD was measured by immunoradioassay in breast cancer cytosols (Brouillet, 1993) . Although pro-CD correlated with total CD, it showed no prognostic value for survival. The authors point out, however, that the proportion of pro-CD recovered in vivo was only 1 to 6% of that produced in cell lines.

A key to developing good prognostic markers is to understand the biological changes that occur when a normal cell progresses to a tumor cell and then becomes invasive. Changes in the expression, processing and activity of proteolytic enzymes have been demonstrated for several human cancers where they are thought to play a critical role in the degradation of extracellular matrix proteins. In fact, PSA, which is a proteinase, has proven to be a very useful prognostic marker for prostate cancer (Lilja, 1992) . The fact that transformed cells display changes in the trafficking of cathepsins leading to the overexpression and secretion of the precursor forms represents a significant alteration in cellular processing and one that should be measurable. Reagents with the proper specificity for the pro-forms should be capable of measuring these changes. Since pro-cathepsin D is abnormally secreted by tumor cells this allows for its participation in the degradation of the extracellular matrix and the possibility for its detection in biological fluid such as blood and urine.

Summarv of the Invention

This invention is directed to a method for making a prognosis for a breast cancer patient involving quantitatively detecting an amount of pro-cathepsin in a sample of a bodily fluid. In one embodiment the method comprises the following steps (a) reacting the sample of the bodily fluid from the patient with an immobilized anti- cathepsin antibody or a suitable cathepsin binding agent as a capture reagent so as to form complexes with any cathepsin present in the sample; (b) reacting the sample of step (a) with an unlabeled antibody capable of specifically binding to an epitope on a pro-cathepsin; (c) reacting the sample of step (b) with a detectable antibody specific for the unlabeled antibody so as to thereby determine the amount of pro-cathepsin in the bodily fluid; and (d) comparing the amount of pro-cathepsin in the bodily fluid with that of a normal subject so as to thereby make a prognosis for the breast cancer patient. In the method described herein the pro-cathepsin may be a pro-cathepsin B, D or L. The bodily fluid may be plasma, serum, urine, saliva, lymphatic fluids, whole blood, nipple aspirates or other bodily fluids. In the methods described herein the immobilized anti-cathepsin antibody may be a monoclonal or a polyclonal antibody. Alternatively, a suitable cathepsin binding agent may be used such as an enzyme inhibitor e.g. pepstatin. In the methods described herein the unlabeled antibody may be a polyclonal antibody. The detectable antibody specific for the unlabeled antibody may be a horseradish peroxidase conjugate. In the methods described herein the prognosis may be a prognosis as to the likelihood of recurrence of cancer such as breast cancer. Alternatively, the prognosis may be the prognosis of relapse or of survival. Brief Description of the Ficrures

Fig. 1. Immunoblot analysis of cathepsin D antibodies. SK- HEP-1 cell conditioned media was subjected to SDS-PAGE then the separated proteins transferred to nitrocellulose and blotted with the anti-pro-CD polyclonal or the anti-mature- CD monoclonal antibody. The latter antibody detects both pro (52 kDa) and the heavy chain of mature CD (34 kDa) while the pro-specific antibody only detects the pro-form.

Fig. 2. Pro-CD levels in human plasma samples. Human plasma samples from breast cancer patients and normal individuals were diluted starting at 1:20 in sample diluent and measured in the ELISA. Results are presented in fmol/ml of plasma. The mean values (horizontal lines) and standard deviations for the normal and breast cancer samples are 4230 (s.d.=1610) and 5610 (s.d.=1800) fmol/ml respectively.

Fig. 3. Immunoblot analysis of pro-CD in the plasma of breast cancer patients. 0.25 μl of each plasma sample was subjected to SDS-PAGE then the separated proteins transferred to nitrocellulose and blotted with an anti-CD monoclonal antibody. Each sample is listed by sample number and whether it measured high or low in the pro-CD ELISA. The standard was SK-HEP-1 cell conditioned media. Each blot contains a lane in which a plasma sample that was low in pro-CD was spiked with an amount of SK-HEP-1 standard to result in a total pro-CD level approximately equivalent to the activity measured in a high plasma sample. A) Identical blots developed with an anti-CD or an irrelevant (UPC 10) monoclonal antibody. B) Blot developed with the anti-CD monoclonal antibody.

Fig. 4. Pepstatin affinity adsorption of pro-CD from the plasma of breast cancer patients. 50 μl of each sample was incubated with of pepstatin agarose as described in

Materials and Methods. Adsorbed proteins were stripped from the beads with pH 9 buffer and SDS sample buffer, separated by SDS-PAGE then blotted with an anti-CD monoclonal antibody. Lane 1, SK-HEP-1 cell conditioned media; lane 2, conditioned media added to a sample of normal plasma; lanes 3-5, plasma samples from breast cancer patients which gave a high value in the pro-CD ELISA, lanes 6-8, plasma samples from normal control individuals. The standard is conditioned media added directly to the gel without prior adsorption to pepstatin agarose. The arrows indicate pro-CD and the heavy chain of the mature form of CD.

Detailed Description of the Invention

This invention is directed to a method for making a prognosis for a breast cancer patient involving quantitatively detecting an amount of pro-cathepsin in a sample of a bodily fluid. In one embodiment the method comprises the following steps (a) reacting the sample of the bodily fluid from the patient with an immobilized anti- cathepsin antibody or a suitable cathepsin binding agent as a capture reagent so as to form complexes with any cathepsin present in the sample; (b) reacting the sample of step (a) with an unlabeled antibody capable of specifically binding to an epitope on a pro-cathepsin; (c) reacting the sample of step (b) with a detectable antibody specific for the unlabeled antibody so as to thereby determine the amount of pro-cathepsin in the bodily fluid; and (d) comparing the amount of pro-cathepsin in the bodily fluid with that of a normal subject so as to thereby make a prognosis for the breast cancer patient. In the method described herein the pro-cathepsin may be a pro-cathepsin B, D or L. The bodily fluid may be plasma, serum, urine, saliva, lymphatic fluids, whole blood, nipple aspirates or other bodily fluids. In the methods described herein the immobilized anti-cathepsin antibody may be a monoclonal or a polyclonal antibody. Alternatively, a suitable cathepsin binding agent may be used such as an enzyme inhibitor e.g. pepstatin, calpain inhibitor I or II, acetyl pepstatin, amastatin, bestatin, antipain, deacylleupeptin, chymostatin, conduritol B epoxide, eglin c fragment, elastatin, epiamastatin, epibestin, foroxymithine, leupeptin, nazumamide, pepsinostreptin, pepstatin A or phosphoramidon or suitable functionalized derivative thereof (see Sigma Chemical Company Catalog.)

In the methods described herein the unlabeled antibody may be a polyclonal antibody. The detectable antibody specific for the unlabeled antibody may be a horseradish peroxidase conjugate. In the methods describe herein the prognosis may be a prognosis as to the likelihood of recurrence of cancer such as breast cancer. Alternatively, the prognosis may be the prognosis of relapse or of survival.

Experimental Details

We have developed antibodies to pro and mature CD and have used these to develop a sandwich ELISA for the quantitation of pro-CD. We have shown that the assay is specific for pro-CD and is capable of detecting this antigen in blood plasma samples. A series of plasma samples from breast cancer patients and normal controls were evaluated with the assay. A subset of samples showed elevated levels of pro- CD. Unfortunately, no follow-up information was available for these patients so it was not possible to make any correlation with prognosis. Immunoblot analysis revealed a band at the appropriate molecular weight for pro-CD whose intensity corresponded with the ELISA data. These results are extremely encouraging and suggest that pro-CD levels could be used to make clinical decisions regarding the treatment of breast cancer patients.

Changes in the expression and processing of cathepsin D (CD) have been shown to be associated with cancer invasion and metastasis. However, the value of CD as a prognostic marker remains controversial. Most studies have used immunological methods to measure the mature form of CD, although it is the precursor (pro-CD) that appears to be abnormally secreted by breast cancer cells. A sandwich-type ELISA has been developed that is specific for pro-CD. The assay employs a monoclonal antibody to mature CD as the capture reagent and a rabbit polyclonal to the pro fragment as the detector.

The assay is specific for pro-CD and capable of quantitating this antigen in biological samples. Pro-CD levels were measured in plasma samples from 76 breast cancer patients and compared with 36 plasmas from normal control individuals. The breast cancer plasmas showed elevated levels of pro-CD; 12% were more than two standard deviations above the mean for the normal samples. Immunoblots of the plasma samples using a CD monoclonal antibody revealed a band at the appropriate size for pro-CD that corresponded in intensity with the ELISA results. Affinity adsorption of breast cancer plasmas with pepstatin agarose followed by immunoblot analysis revealed a single protein band that corresponded with pro-CD. Only trace amounts were detected in the normal control plasmas. These results demonstrate that cathepsin D is present in the plasma of breast cancer patients primarily in its precursor form and that it is a useful prognostic indicator.

INTRODUCTION

The proteolytic degradation of extracellular matrix components is a crucial part of the multistep process of cancer cell invasion and metastasis. Numerous studies indicate that several classes of proteolytic enzymes are involved and suggest that changes in their expression, processing and activation contribute to the pathogenesis of tumor growth and invasion. An increasing body of evidence implicates lysosomal proteinases, in particular cathepsin D (CD) , as having a critical role in this process. CD is an aspartic proteinase that functions in the lysosomes in the normal degradation of proteins. It is synthesized as a 52 kDa pro-enzyme (pro-CD) which is processed to the active, two-chain form of 34 and 14 kDa subunits (Yonezawa, 1988) . A relationship between CD and breast cancer was first proposed when it was discovered that the breast cancer cell line, MCF-7, secreted the 52 kDa pro-enzyme in response to estrogen (Westley, 1980) . CD is made in all cells and is normally processed and targeted to the lysosomal compartment. However, in tumor cells, a significant amount of the enzyme is diverted from the normal processing pathway and is secreted in its pro-form (Rochefort, 1990) . The mechanism for this is not well understood.

In vi tro studies support a role for CD in cancer progression. CD levels increase in response to estrogens and other mitogens (Cavailles, 1988) . CD is capable of digesting extracellular matrix proteins (Briozzo, 1988) and transfection of the CD gene into rat cells increases their tumorogenicity in nude mice (Garcia, 1990) . A number of clinical studies have shown that CD is a useful prognostic marker for breast cancer (Rochefort, 1992) . Studies using immunoassays showed that CD was a powerful prognostic factor for predicting recurrence and survival and was independent of other traditional prognostic parameters (Spyratos, 1989; Kute, 1992; Seshadri, 1994) . A 5-year prospective study showed that CD has prognostic value in breast cancer (Pujol,

1993) . Unfortunately, not all reports have confirmed the prognostic utility of CD, especially those using immunohistochemical techniques (Merkel, 1991; Armas, 1994; Ravdin, 1994) . However, a recent study suggested that immunohistochemical determination of CD was a useful prognostic tool in early-stage breast cancer (Belvilacqua,

1994) . Clearly, the use of CD as a clinically relevant marker for breast cancer remains controversial.

Most clinical studies have been conducted using immunoassays and immunohistochemical methods with reagents that measure the mature form or the total of mature plus precursor. This is in spite of the fact that it is the precursor form that is abnormally secreted by cancer cells. Attempts to measure increased secretion of the precursor by tumor cells could be obscured by high background levels of the endogenous lysosomal form. A means of specifically measuring only the level of pro-CD has the potential to provide more reliable prognostic information. Also, nearly all studies on CD have been confined to tumor samples and cytosols. One report described the immunoradiometric measurement of total CD in plasma of breast cancer patients and concluded that it was not a useful marker (Brouillet, 1993) . Since pro-CD is secreted by tumor cells, we were interested in developing methods to detect it in biological fluids, such as plasma, and to determine if this can provide clinically relevant information.

We have developed a sandwich-type enzyme-linked immunosorbent assay (ELISA) and demonstrated that it can specifically quantitate pro-CD in biological samples. The analysis of a series of plasma samples from breast cancer patients showed that the level of pro-CD was elevated compared to normal control individuals. Immunoblots of plasma samples revealed a band at the appropriate molecular weight for pro-CD that corresponded in intensity with the ELISA results. The prospect of using circulating levels of pro-CD as a prognostic indicator for the management of breast cancer is discussed.

MATERIALS AND METHODS

Materials: Plasma samples from patients with Stage I-IV breast cancer were obtained from Dianon Systems (Stratford, CT) and those from normal individuals were from the American Red Cross. Samples were stored at -70°C and thawed prior to use. The monoclonal antibody (clone # 6410) made against mature cathepsin D purified from human liver and purified mature cathepsin D standard were obtained from Scripps Laboratories. A second anti-mature CD monoclonal antibody (#IM03) and the Total Cathepsin D ELISA were from Oncogene Science. The UPC 10 monoclonal antibody was obtained from Sigma (#M 9144) . The proteinase inhibitors (Sigma) were made as concentrated stock solutions and used at the following final concentrations: 0.2 mM PMSF, 0.5 μg/ml Leupeptin, 1.0 mM EDTA and 1.0 μg/ml Pepstatin. Pepstatin agarose was also obtained from Sigma. Cell Culture: Serum-free conditioned media from SK-HEP-1 human liver tumor cells was prepared as follows. Roller bottle cultures were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 5% fetal calf serum(FCS) and nonessential amino acids. When cells reached approximately 75-90% confluence, they were washed twice with PBSdOmM sodium phosphate, pH 7.4, 150mM NaCl) , once with serum-free DMEM then maintained in serum-free DMEM. After 72-96 hours, the media was collected and the cocktail of proteinase inhibitors added. Cells and cell debris were removed by centrifugation at 200xg for 10 min at 4°C. Clarified, conditioned media was concentrated approximately 10 fold using an Amicon CH2 spiral ultrafiltration apparatus with an SI, 10,000 MW cutoff cartridge. Media was stored at -70°C until needed.

Polyclonal Antibody Development: A sequence from the pro- region of pro-CD was synthesized as a multiple antigenic peptide(MAP) (Tarn, 1988) using an Applied Biosystems 430 Peptide Synthesizer. New Zealand white rabbits were immunized intra-nodally by injection into the popliteal lymph node (Dresser, 1986) . Animals were primed with 500 μg of MAP peptide in complete Freund's adjuvant and boosted 3 times with 250 μg in incomplete Freund's adjuvant. Test bleed titers were monitored by ELISA using microtiter plates coated with MAP peptide (5 μg/ml) and detection with goat anti-rabbit HRP conjugate (125 ng/ml) . The antisera was purified using Protein A chromatography (Pierce) and antibody concentration determined by absorbance at 280nm. This antibody is commercially available (OSI, #IM04) .

Pro-cathepsin D ELISA: 96 well microtiter plates were coated by applying lOOμl/well of anti-cathepsin D monoclonal antibody (clone #6410) at 5 μg/ml in lOOmM sodium carbonate, pH 9.6; and incubated overnight at room temperature (RT) . Plates were washed with PBS and blocked with 2% BSA, 10% lactose in PBS prior to use. 100 μl of standard or sample diluted in sample diluent (PBS, 4% BSA and 0.7% NP40) was added to each well and incubated overnight at RT. Plates were washed 6 times with wash buffer (10 mM phosphate, pH 7.5, 150 mM NaCl, 0.05% Tween-20) then 100 μl of the anti- pro-CD rabbit polyclonal detector antibody(4 μg/ml) was added and incubated for 1 hr at RT. Plates were washed 6 times as before followed by the addition of 100 μl of goat anti-rabbit horseradish peroxidase conjugate (Kirkegaard and Perry) at 0.25 μg/ml. After 30 min. at RT, the plate were again washed 6 times and 100 μl of O-phenylenediamine substrate (DAKO, lmg/ml in 100 mM citrate buffer, 0.03% hydrogen peroxide) was added. Development was allowed to proceed for 1 hr at RT in the dark then stopped by the addition of 100 μl of 4N H₂S0₄. Absorbance was read at 490nm using a Biotek EL 309 Autoreader.

Immunoblots: Proteins were separated by SDS-PAGE on 10% gels (Laemmli 1970) . Electrotransfer to nitrocellulose

(Schleicher and Schuell) was performed overnight at 4°C according to standard procedures (Timmons 1990) . Blots were incubated with blocking solution [5% nonfat dry milk in TBS- T(50 mM Tris, pH 7.6, 150 mM NaCl, 0.5% Tween 20)] , washed with TBS-T then incubated for 1 hr in the presence of the anti-mature CD monoclonal antibody(OSI, #IM03) at 5 μg/ml in blocking solution. Blots were washed 3 times followed by incubation for 1 hr with goat anti-mouse IgG or goat anti- rabbit IgG horse radish peroxidase conjugate (Kirkegaard and Perry) at 30 ng/ml in blocking solution. After 4 washes with TBS-T, the blots were developed using chemiluminescent detection (ECL, Amersham) .

Pepstatin affinity adsorption: Plasma samples were clarified by brief centrifugation at 15,000xg in a microcentrifuge. 50 μl of plasma or SK-HEP-1 conditioned media were diluted to 100 μl with dH₂0 and added to 200 μl of binding buffer (100 mM sodium citrate, pH 3.4, 400 mM NaCl) . Samples were added to microcentrifuge tubes containing 50 μl of packed pepstatin agarose beads previously equilibrated with binding buffer. The samples were allowed to incubate with gentle rocking for 3 hr at 4°C. The beads were pelleted by centrifugation, washed twice in binding buffer containing 6M urea then twice in 100 mM sodium citrate. Proteins were eluted from the beads by the addition of 500 mM Tris, pH 9.0 and reducing SDS-PAGE sample buffer follow by incubation at 100°C. The beads were pelleted by centrifugation and the supernatants subjected to electrophoresis and immunoblotted as described.

RESULTS

Pro-CD Specific ELISA: Polyclonal antisera was generated by immunizing rabbits with a synthetic peptide from the pro- sequence of pro-CD (Faust, 1985) . The conditioned media from SK-HEP-1 cells was used to confirm that the antisera recognized the pro-form of CD. Immunoblot analysis indicated that SK-HEP-1 cells secrete cathepsin D predominantly in the precursor form. This is shown in Fig. 1 in which a sample of conditioned media is probed with a monoclonal antibody directed against an epitope on mature CD. The antibody recognized a major protein band at 52 kDa and a minor band at about 34 kDa which correspond to pro-CD and the heavy chain of mature CD respectively. An immunoblot of conditioned media probed with the anti-pro- peptide polyclonal antibody showed that only the 52 kDa pro- CD was recognized (Fig. 1) .

A pro-CD specific sandwich ELISA was developed in which a monoclonal antibody to mature CD was used as the capture reagent and the anti-pro-CD peptide polyclonal as the detector. The presence of bound detector antibody was determined using a goat anti-rabbit antibody conjugated with HRP. SK-HEP-1 conditioned media, containing pro-CD (see Fig. 1) , was used to prepare standards which were calibrated using the Total Cathepsin D ELISA. The assay produces a linear standard curve in the range of 100 - 1000 femtomoles/ml . A mature cathepsin D standard does not yield a signal in this assay.

In order to demonstrate assay specificity, a series of controls were run in which each of the critical assay components were individually replaced with an irrelevant reagent. The results are summarized in Table 1. The first test contained all the proper reagents and a level of antigen to produce a suitable absorbance signal. In each of the other cases, either the capture antibody, antigen or detector antibody was changed to a negative control. All produced absorbance signals that were near background. In particular, a standard containing 4200 fmoles/ml of mature CD did not give an appreciable signal in the assay. These tests indicate that the assay is specific for the pro-CD antigen.

Since an anti-mature CD antibody is used as the capture reagent, there was a concern that an excess of mature CD in a sample could inhibit the detection of pro-CD. This was examined by measuring a pro-CD standard in the presence of the mature form. A 7.5 fold molar excess of mature CD had no influence on the quantitation of pro-CD in the ELISA.

A competition study was used to provide additional confirmation of assay specificity. When an excess of capture antibody is added in solution together with the antigen preparation, it should prevent the antigen from binding to the antibody adsorbed to the solid phase and thus quench the assay signal. A series of conditioned media standards were pre-incubated with the monoclonal capture antibody at final concentrations of 10 or 25 μg/ml and then measured in the assay. This resulted in an inhibition of the assay signal to background levels. As a control, the same experiment was performed in the presence of similar amounts of an irrelevant, isotype-matched, competing antibody, UPC 10. The result was little or no loss of signal. This demonstrates that the signal generated in the assay is specific for the antigen that is recognized by the capture monoclonal antibody and is not due to interfering or cross-reacting substances in the sample.

The pro-CD ELISA is capable of measuring antigen in lysates and conditioned media from human cell lines. However, the assay must be validated for the measurement of pro-CD in human plasma. For this purpose, about 300 fmol of a pro-CD standard was added to normal plasma samples then measured in the ELISA to determine percent recovery. The results are summarized in Table 2. Plasma caused severe inhibition of the assay signal and, even after a 1:5 dilution in sample diluent, only 18% of the added activity was recovered. Plasma samples were further diluted with sample diluent while maintaining a constant level of pro-CD standard. The ELISA results indicate that a dilution factor of 1:20 to 1:40 is required in order to achieve greater than 90% recovery of added activity.

A possible explanation for the inhibition observed in the plasma samples is proteolytic degradation of antigen by plasma proteinases. To test this, a cocktail of proteinase inhibitors (see Materials) was added to a plasma sample containing the pro-CD standard and measured in the ELISA. The results in Table 2 indicate that the proteinase inhibitors did not improve the recovery of pro-CD activity. Addition of the inhibitor cocktail to the standard alone had no effect on the measurement of pro-CD. These results indicate that something in plasma other than proteolytic degradation is causing interference in the assay. Therefore, all plasma samples must be diluted by at least a factor of 1:20 in order to obtain a reliable quantitation of pro-CD.

Evaluation of Plasma Samples: The pro-CD ELISA was used to evaluate 76 plasma samples from breast cancer patients and 36 normal controls. All samples were diluted starting at 1:20. Samples were measured at multiple dilutions and the data displayed good parallelism to the standard curve. The results, expressed in terms of femtomoles/ml of plasma, are presented in Figure 2. The means with standard deviations for the cancer and control groups were 5610 (s.d.=1800) and 4230 (s.d.=1610) fmol/ml, respectively. Most of the values are clustered around 5000 fmol/ml but there are several patients with pro-CD levels 2-3 times higher. Using a cut- off of two standard deviations above the mean for the normal samples, nine breast cancer samples or 12% of the total gave values above this level. There is no follow-up information available on these patients but it is interesting to note that 6 of the 9 specimens were from patients with stage I or II disease.

Competition studies were again performed to confirm that the ELISA signal in the plasma samples represented pro-CD antigen. An excess of capture antibody was added to several of the high pro-CD breast cancer plasmas and the signals were shown to be inhibitable. As in the experiment described above, this verifies that the assay signal is due to the antigen that is recognized by the capture antibody.

Immunoblot Analysis of Plasma Samples: A very powerful way of demonstrating specificity is to identify the presence of the antigen in the sample by immunoblot. Several plasma samples that measured either high or low in the ELISA were separated by SDS-PAGE, and blotted with the monoclonal antibody to mature CD used in Fig. 1. It is important to point out that this monoclonal antibody is different from the capture reagent in the ELISA. As a negative control, a similar blot was probed with the irrelevant, isotype-matched monoclonal antibody, UPC 10. The immunoblots are presented in Figure 3. The plasma samples are indicated by their sample number and whether they measured high or low in the ELISA. SK-HEP-1 cell conditioned media was included as the standard and the band we believe to represent pro-CD is indicated by the arrow. Panel A shows a blot of a high, low and a low sample spiked with an amount of pro-CD standard to result in a total pro-CD level approximately equivalent to the ELISA activity in the high sample. The identical blot developed with the UPC 10 antibody reveals bands that are thought to represent non-specific binding. In particular, the band just above the pro-CD marker could be human IgG heavy chain for which the goat anti-mouse detector antibody has a slight affinity. Of note is the band at about the molecular weight of the pro-CD standard that appears to vary in intensity consistent with the ELISA results. Panel B shows a blot of more plasma samples developed with the anti- CD monoclonal and again the band at about 52 kDa seems to have an intensity that is consistent with the ELISA analysis. This band, however, clearly migrates slightly above the standard from the SK-HEP-1 cell conditioned media. This may represent the reported observation that pro-CD from cell culture medium undergoes a 1-2 kDa reduction in size (see Discussion) . It is important to point out that in the standard lane of Panel B and in the spiked samples (#33 and #75) , the amount of pro-CD ELISA activity was approximately equal to that contained in the high samples. The resultant band has an intensity comparable to that which we have identified as pro-CD in the plasma samples.

The blots in Fig. 3 also reveal a strong band at about 150 kDa that does not appear in the UPC 10 control blot. It appears to be a plasma protein that is recognized by the anti-CD monoclonal but it is much more intense than expected and does not vary in intensity with the ELISA quantitation. This is puzzling since we are unaware of any reports of a form of cathepsin D that migrates in this region on SDS- PAGE. It is possible that this species is responsible for some of interference problems we have experienced with the measurement of plasma samples. Pepstatin affinity adsorption: Because of the difficulties associated with direct immunoblot analysis of plasma, we also attempted to enrich for pro-CD by affinity adsorption to pepstatin agarose. Pepstatin is a specific inhibitor of aspartic proteinases and has been used to affinity purify mature and pro-CD (Conner, 1989; Larsen, 1993) . Both forms bind strongly to pepstatin agarose at pH 3.5 and elute at pH 8.3. 50 μl of pepstatin agarose was incubated with plasma samples that had been adjusted to acidic pH. The beads were washed then the proteins eluted by the addition of pH 9 buffer and SDS-PAGE sample buffer. The entire sample was subjected to SDS-PAGE and immunoblotted using the mature CD monoclonal antibody. Whereas the previous immunoblots were limited to 0.25 μl of plasma, this procedure allowed for the analysis of 50 μl of each sample. This greatly improved the likelihood of detecting pro-CD while reducing nonspecific background bands.

The results are shown in the immunoblot in Figure 4. All samples, with the exception of the SK-HEP-1 cell conditioned media standard, were adsorbed to pepstatin agarose and treated as described. Lane 1 is a sample of conditioned media and lane 2 represents conditioned media added to a sample of normal plasma. The single protein band at the correct molecular weight for pro-CD indicates that this procedure effectively enriched for pro-CD from either conditioned media or plasma. Lanes 3-5 represent plasma samples from breast cancer patients which measured high in the pro-CD ELISA and lanes 6-8 are normal control plasmas. A prominent pro-CD band is evident in the breast cancer plasmas while only trace amounts were detected in the controls. The pro-CD protein band was greatly enriched compared to the blots in Fig. 3 and the background bands, particularly the one at 150 kDa, were eliminated. These results confirm that there is a significant difference in circulating pro-CD levels in breast cancer patients compared to non-cancer controls. Also, the plasma samples do not appear to contain any of the 34 kDa heavy chain of mature CD indicating that cathepsin D in plasma is predominantly in the pro-form.

DISCUSSION

We have generated a polyclonal antibody that recognizes pro- CD and have demonstrated its specificity by immunoblot. This antibody together with an anti-mature CD monoclonal antibody were used to develop an ELISA for the quantitation of pro-CD. The ELISA gives a linear response over the range of 100 to 1000 femtomoles/ml and was shown to be specific for pro-CD by several criteria. In addition, the assay signal was shown to be inhibitable by an excess of capture monoclonal antibody.

The pro-CD ELISA was validated for the measurement of pro-CD in blood plasma. Plasma caused an interference problem in the assay but this could be minimized with appropriate dilutions. Spike and recovery tests showed that >90% of added pro-CD antigen could be detected by the ELISA when the plasma was diluted by a factor of 20 or more. The assay is sufficiently sensitive to detect pro-CD in human plasma samples and specificity was confirmed by the fact that the signal was inhibited by excess capture antibody.

Pro-CD levels were determined in a series of plasma samples from breast cancer patients and compared to samples from non-cancer controls. The majority of the values were within a narrow range but a portion of the breast cancer plasmas exhibited elevated levels (12% were greater than two standard deviations above the mean of the normal group) . Unfortunately, there is no follow-up information available for these patients so it is not possible to draw any correlations regarding prognosis, although pro-CD levels appeared to be independent of stage. The value of a prognostic marker is in the ability to distinguish a subset of individuals from the total group of cancer patients that differ by a clinically relevant parameter such as probability of relapse, overall survival or response to therapy. Our data shows that a small group of patients have higher then normal levels of pro-CD in their blood. These results are preliminary but suggest that further studies are warranted to determine if pro-CD levels correlate with prognosis and provide useful information for the management of breast cancer.

We performed immunoblot analysis to confirm the presence of pro-CD in plasma. Using a monoclonal antibody to CD, a band was detected that migrated at the appropriate molecular weight for the precursor form. The intensity of this band appeared to correspond with the results obtained from the ELISA. However, this band migrated slightly above that of the pro-CD standard from SK-HEP-1 cell conditioned media. It has been reported that pro-CD undergoes a pH-dependent autoactivation which results in a reduction in its molecular weight by 1 - 2 kDa (Briozzo, 1988; Conner, 1989; Larsen, 1993; Richo, 1994) . This has been demonstrated for cell culture derived pro-CD but experiments suggest that it may not be a normal intermediate for processing in vivo . It is possible that the conditioned media standard contains predominantly the lower molecular weight form of pro-CD while human plasma has the full size 52 kDa form.

The presence of pro-CD in plasma was further verified by taking advantage of its affinity for the aspartic proteinase inhibitor, pepstatin. Affinity adsorption of the breast cancer plasmas with pepstatin agarose followed by immunoblot analysis revealed one band at the correct molecular weight for pro-CD. This experiment confirms that the protein band represents pro-CD based on two criteria, affinity for pepstatin then recognition by a monoclonal antibody. Pro-CD was shown to be prominent in the plasma from breast cancer patients but very low in the normal controls. In fact, the difference in the level of pro-CD between the breast cancer plasmas and normal controls appears to be more dramatic by this method than is indicated by the ELISA results. The ELISA may be underestimating the difference due to the interference problems discussed earlier. It is also important to note that the pepstatin adsorption experiment did not detect any mature CD, indicating that circulating cathepsin D is predominantly in the precursor form.

Although numerous clinical studies have suggested that cathepsin D is a useful marker for breast cancer prognosis (Rochefort, 1992) , this still remains controversial (Cardiff, 1994) . Nearly all studies on CD have focused on the determination of mature or total CD in tumor samples and cytosols . Part of the controversy may be due to the fact that it is the pro-form of CD that is abnormally secreted by tumor cells. The ability to measure increased secretion of the precursor by tumor cells may be hindered or masked by the presence of high levels of the endogenous lysosomal form. Assays and immunohistochemistry reagents that specifically measure the pro-form have the potential to provide more reliable data and be more predictive of tumor status. However, the only report on the measurement of pro- CD in breast tumor cytosols by immunoassay found no prognostic value but the authors point out that the recovery of antigen in their experiments may have been low (Brouillet, 1993) .

Studies on circulating cathepsin D levels are limited. One report described the measurement of total CD in plasma and concluded that it was not a useful marker (Brouillet, 1991) . Another used immunoblots to detect pro and mature CD in sera and noted little difference between breast cancer patients and controls (Schultz, 1994) . However, no standards were used to confirm the identity of the bands on their blots. In fact, the level of pro and mature CD in their experiments was substantially higher than we detect and it is possible that they were actually measuring the heavy and light chains of human IgG which migrate at similar molecular weights .

A key to developing good prognostic markers is to understand the biological changes that occur when a normal cell progresses to a tumor cell and then becomes invasive. The fact that tumor cells display changes in the trafficking of cathepsins leading to the overexpression and secretion of the precursor forms represents a significant alteration in cellular processing and one that should be measurable. Secreted pro-CD, once activated, could participate in the degradation of the extracellular matrix. Secretion may also allow for its detection in bodily fluids such as blood and urine. Reagents and assays with the proper specificity for pro-CD could provide valuable information regarding tumor status and patient prognosis. We have developed an assay for pro-CD and have found elevated levels of this antigen in the plasma of breast cancer patients. These results are encouraging but much additional work is required in order to establish a clinical relevance for the measurement of pro-CD in the blood of cancer patients.

ABBREVIATIONS

CD, cathepsin D; DMEM, Dulbecco's modified Eagle's medium; ELISA, enzyme-linked immunosorbent assay; FCS, fetal calf serum; HRP, horseradish peroxidase; MAP, multiple antigenic peptide; PBS, phosphate-buffered saline; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; TBS-T, tris-buffered saline-tween. REFERENCES

Armas OA, Gerald WL, Lesse CD, Arroyo CD, Norton L and Rosen PP: Immunohistochemical detection of cathepsin D in T(2)N(0)M(0) breast carcinoma. Amer. J. of Surgical Pathology, 18 : 158-166, 1994.

Belvilacqua P, Boracchi P and Gasparini G: Prognostic indicators for early-stage breast carcinoma. Part II: value of cathepsin D expression, detected by immunocytometry. A multiparametric study. Int. J. Oncol., 5; 559-564, 1994.

Briozzo P, Morisset M, Capony F, Rougeot C, and Rochefort H: In Vi tro degradation of extracellular matrix with Mr 52,000 cathepsin D secreted by breast cancer cells. Cancer Res., 48 : 3688-3692, 1988.

Brouillet JP, Hanslick B, Maudelonde T, Pivat MT, Grenier J, Blanc F and Rochefort H: Increased plasma cathepsin D concentration in hepatic carcinoma and cirrhosis but not in breast cancer. Clinical Biochemistry, 24 : 491-496, 1991.

Brouillet JP, Spyratos F, Hacene K, Fauque J, Freiss G, Dupont F, Maudelonde T and Rochefort H: Immunoradiometric assay of pro-cathepsin D in breast cancer cytosol : relative prognostic value versus total cathepsin D. Eur. J. Cancer, 29A : 1248-1251, 1993.

Capony F, Rougeot C, Montcourrier P, Cavailles V, Salazar G and Rochefort H: Cancer Res., 49 : 3904-3909, 1989.

Cardiff RD: Cathepsin D and breast cancer: useful? Human Pathology, 25 : 847-848, 1994.

Cavailles V, Augereau P, Garcia M, and Rochefort H: Estrogens and growth factors induce the mRNA of the 52K pro- cathepsin D secreted by breast cancer cells. Nucleic Aids Res., 16 : 1903-1919, 1988.

Chauhan SS, Goldstein LJ, and Gottesman MM; Cancer Res., 51 : 1487-1481, 1991.

Conner GE: The Role of the Cathepsin D Propeptide in Sorting to the Lysosome. JBC, 267 : 21738-21745, 1992.

Conner GE: Isolation of procathepsin D from mature cathepsin D by Pepstatin affinity chromatography. Biochem. J., 263 : 601-604, 1989.

Dresser DW: Immunization of Experimental Animals. In : Handbook of Experimental Immunology. Weir DM. (ed.) .

Oxford, Blackwell Scientific Publications, Chapter 8, 1986.

Faust PL, Kornfeld S and Chirgwin JM: Cloning and sequence analysis of cDNA for human cathepsin D. PNAS, 82 : 4910- 4914, 1985.

Garcia M, Derocq D, Pujol P, and Rochefort H: Overexpression of transfected cathepsin D in transformed cells increases their malignant phenotype and metastatic potency. Oncogene, 5: 1809-1814, 1990.

Garcia M, Salazar-Retana G, Pages A, Richer G, Domergue J, Pages AM, Cavalie G, Martin JM, Lamarque JL, Pau B, Pujol H, and Rochefort H: Cancer Res., 46 : 3734-3738, 1986.

Haerslev T, Jacobsen GK and Zedeler K: Oncology Reports, 1 : 787-793, 1994.

Henry JA, McCarthy AL, Angus B, Westley BR, May FE, Nicholson S, Cairns J, Harris A and Home LH: Cancer, 65 : 265-271, 1990. Ingle JN: New England Journal of Medicine, 322 : 329-331, 1990.

Isola J, Weitz S, Visakorpi T, Holli K, Shea R, Khabbaz N, and Kallioniemi 0-P: J. Clinical Oncology, 1 : 36-43, 1993.

Johnson MD, Torri JA, Lippman ME and Dickson RB: Cancer Res., 53 : 873-877, 1993.

Kute TE, Shao ZM, Sugg KN, Long RT, Russell GB, and Case LD: Cathepsin D as a prognostic indicator for node-negative breast cancer patients using both immunoassays and enzymatic assays. Cancer Res, 52 : 5198-5203, 1992.

Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227 : 680- 685, 1970.

Lah TT, Kokalj-Kunovar M, Strukelj B, Pungercar J, Barlic- Maganja D, Drobnic-Kosorok M, Kastelic L, Babnik J, Golouh R and Turk V: Int. J. Cancer, 50 : 36-44, 1992.

Larsen LB, Boisen A and Petersen TE: Procathepsin D cannot autoactivate to cathepsin D at acid pH. FEBS Letters, 319 : 54-58, 1993.

Lilja H, Cockett ATK and Abrahamsson PA: Cancer, 70 : 230- 234, 1992.

Liotta LA and Stetler-Stevenson WG: In : Seminars in Cancer Biology. Gottesman, MM. (eds.) Vol. 1(2) : 99-106, 1990.

McGuire WL and Clark GM: New England Journal of Medicine, 326 : 1756-1761, 1992.

Merkel DE, Goldschmidt RA, Gatbunton C, Winchester DJ, and Rademaker AW: Intracellular cathepsin-D and prognosis in node-negative breast cancer. Breast Cancer Research and Treatment, 19 : 200, 1991.

Neville AM: Diagnostic Oncology, 1 : 53-63, 1991.

Pujol P, Maudelonde T, Daures JP, Rouanet P, Brouillet JP, Pujol H, and Rochefort H A: Prospective study of the prognostic value of cathepsin D levels in breast cancer cytosol. Cancer, 71: 2006-2012, 1993.

Ravdin PM, Tandon AK, Allred DC, Clark GM, Fuqua SAW, Hilsenbeck SH, Chamness GC, Osborne CK: Cathepsin D by western blotting and immunohistochemistry: failure to confirm correlations with prognosis in node-negative breast cancer. J. Clinical Oncology, 12 : 467-474, 1994.

Richo GR and Connor GE: Structural requirements of procathepsin D activation and maturation. J. Biol . Chem. , 269 : 14806-14812, 1994.

Rochefort H: Biological and clinical significance of cathepsin D in breast cancer. In : Seminars in Cancer Biology. Gottesman, MM. (ed.) Vol. 1(2) : 153-160, 1990.

Rochefort H: Cathepsin D in breast cancer: a tissue marker associated with metastasis. Eur. J. Cancer, 28A : 1780-83, 1992.

Schultz DC, Bazel S, Wright LM, Tucker S, Lange MK, Tachovsky T, Longo S, Niedbala S and Alhadeff JA: Western blotting and enzymatic activity analysis of cathepsin D in breast tissue and sera of patients with breast cancer and benign breast disease and of normal controls. Cancer Research, 54 : 48-54, 1994.

Seshadri R, Horsfall DJ, Firgaira F, McCaul K, Setlur V, Chalmers, AH, Yeo R, Ingram D, Dawkins H, Hahnel R, and the South Australian Breast Cancer Study Group: The relative prognostic significance of total cathepsin D and her-2/neu oncogene amplification in breast cancer. Int. J. Cancer, 56: 61-65, 1994.

Sloan BF: In : Seminars in Cancer Biology. Gottesman, MM. (eds) Vol. 1(2) : 137-152, 1990.

Spyratos F, Maudelonde T, Brouillet JP, Brunet M, Defrenne A, Andrieu C, Hacene K, Desplaces A, and Rochefort H: Cathepsin D; an important marker predicting metastasis in primary breast cancer. Lancet, 334 : 1115-1118, 1989.

Tarn JP: Synthetic peptide vaccine design: Synthesis and properties of a high density multiple antigenic peptide system. PNAS 85 : 5409-5413, 1988.

Timmons TM and Dunbar BS: Protein blotting and immunodetection. In : Methods in Enzymology, Guide to Protein Purification. Deutscher MP. (ed) , San Diego, Academic Press, 1990.

Vignon F, Capony F, Chambon M, Freiss G, Garcia M, and Rochefort H: Endocrinology, 118 : 1537-1545, 1986.

Westley B, and Rochefort HA: Secreted glycoprotein induced by estrogen in human breast cancer cell lines. Cell, 20 : 352-362, 1980.

Yonezawa S, Takahashi T, Wang XJ, Wong RN, Hartsuck JA and Tang J: Structures at the proteolytic processing region of cathepsin D. J. Biol . Chem. , 263 : 16504-11, 1988. Table I. Pro-CD ELISA Specificity Tests

Capture Antige Detector Abs. 490 nm

Ab n Ab

+ + + 0.981

c-erbB-2 + + 0.082

+ matCD + 0.070

+ media + 0.031

+ 5% FCS + 0.116

+ 5% + 0.120

BSA

+ pro-CB 0.099

+ + c-erbB-2 0.090

+ + na 0.029

The ELISA was performed as described with the usual antibodies and a pro-CD standard as antigen (+) or with irrelevant antibodies or antigens as indicated. Antigen samples were diluted in sample diluent. FCS and BSA were diluted in DMEM. Media represents DMEM alone. The purified mature cathepsin D (matCD) standard was at 4200 fmol/ml. The c-erbB-2 antibody used to replace the capture was a monoclonal and that used to replace the detector was a rabbit polyclonal. na indicates no addition. Table II. Effect of Plasma on the Pro-CD ELISA

Sample Dilution Abs. 490 nm % Recovery

Pro-CD std - 2.6 (100)

Plasma 1 :5 0.15

Plasma + pro-CD std 1 :5 0.47 18

Plasma + pro-CD std 1 :10 1.19 46

Plasma + pro-CD std 1 :20 2.36 91

Plasma + pro-CD std 1 :40 2.50 96

Plasma + pro-CD std 1 :80 2.58 99

Pro-CD std + PI 1 :6 2.73 (100)

Plasma + pro-CD std + 1 :6 0.42 15

PI

A fixed amount of a pro-CD standard was added to a series of serially diluted plasma samples which were measured in the ELISA. Percent recoveries are based on the value for the standard alone in the absence plasma. Plasma alone gave negligible activity. In a separate experiment, the proteinase inhibitor cocktail(PI) as described in Materials was added to a plasma sample containing a standard amount of pro-CD. The percent recovery was determined by ELISA and compared to the same preparation without plasma.

Claims

What is claimed is:

1. A method for making a prognosis for a breast cancer patient involving quantitatively detecting an amount of pro-cathepsin in a sample of a bodily fluid which comprises:

(a) reacting the sample of the bodily fluid from the patient with an immobilized anti-cathepsin antibody or a suitable cathepsin binding agent as a capture reagent so as to form complexes with any cathepsin present in the sample;

(b) reacting the sample of step (a) with an unlabeled antibody capable of specifically binding to an epitope on a pro-cathepsin;

(c) reacting the sample of step (b) with a detectable antibody specific for the unlabelled antibody so as to thereby determine the amount of pro-cathepsin in the bodily fluid; and

(d) comparing the amount of pro-cathepsin in the bodily fluid with that of a normal subject so as to thereby make a prognosis for the breast cancer patient .

2. The method of claim 1, wherein the pro-cathepsin is pro-cathepsin B, D or L.

3. The method of claim 1, wherein the bodily fluid is plasma.

4. The method of claim 1, wherein the bodily fluid is serum.

5. The method of claim 1, wherein the bodily fluid is urine.

6. The method of claim 1, wherein the bodily fluid is saliva.

7. The method of claim 1, wherein the immobilized anti- cathepsin antibody is a monoclonal antibody.

8. The method of claim 1, wherein the suitable cathepsin binding agent is an enzyme inhibitor.

9. The method of claim 1, wherein the enzyme inhibitor is pepstatin.

10. The method of claim 1, wherein the unlabeled antibody is a polyclonal antibody.

11. The method of claim 1, wherein the detectable antibody specific for the unlabeled antibody is a horseradish peroxidase conjugate.

12. The method of claim 1, wherein the prognosis is a prognosis as to the likelihood of recurrence of cancer.