KR20050020806A - Diagnosis of hepatocellular carcinoma - Google Patents

Abstract

The present invention provides a method for screening liver cancer patients by measuring the level of glypican-3 (GPC3) in a patient's body fluid sample. The present invention further provides a method of measuring liver cancer by measuring GPC3 in liver tissue samples. The present invention also provides an antibody that specifically binds to GPC3.

Description

Diagnosis of hepatocellular carcinoma {DIAGNOSIS OF HEPATOCELLULAR CARCINOMA}

The present invention relates to the diagnosis of liver cancer. In particular, the present invention relates to antibodies and methods for measuring levels of glypican-3 as liver cancer markers.

In the present invention, various references are cited in order to more fully describe the present technology related to the present invention. These references are incorporated by reference in their entirety.

Hepatocellular carcinoma (HCC) is the world's most common solid-organ tumor, with more than 1 million deaths each year (Parkin et al., CA. Cancer J. Clin., 49: 33-). 64, 1999; Okuda et al., Neoplasms of the Liver in Diseases of the Liver, 7th edition 1236, 1993).

The incidence of liver cancer is increasing in most western countries (Deuffic, et al., Lancet 351: 214-215, 1998). Although not enough epidemiological studies are available, mortality from liver cancer in adult men has nearly doubled in the last 15 years, according to data from the Canadian Department of Health. Demographics of patients with viral hepatitis indicate that the incidence of liver cancer will double again within the next decade (Zou, et al., Can. J. Gastroenterol., 14: 575-580, 2000).

Liver cancer usually has no symptoms in the early stages and tends to invade into the blood vessels or bile even though the primary tumor is small (Fong, et al., Cancer of the Liver and Biliary Tree. In Cancer.Principles & Practice of Oncology, 6th Edition, 1162-1199, 2001). As a result, liver cancer is usually found at some stage. Only 10-20% of early liver cancers are found to be removed at diagnosis (Fong, et al., Cancer of the Liver and Biliary Tree in Cancer: Principles & Practice of Oncology, 1162-1199, 2001). While most patients still appear incurable at diagnosis, the emergence of liver cancer screening programs is increasing the proportion of tumors that can be treated.

Liver cancer is associated with chronic liver disease, particularly chronic viral hepatitis and alcoholic liver disease (Rustgi, Gastroenterol. Clin. North Am., 16: 545-551, 1987). Liver cancer is more common with hepatitis B and hepatitis C. For hepatitis B, patients with chronic infection are 50-100 times more likely to develop liver cancer than those without infection (Beasley, et al., Epidemiology of hepatocellular carcinoma in Viral hepatitis and liver disease, 209, 1994). Unlike other causes of liver cancer, chronic hepatitis B is not a prerequisite to developing liver cancer. Although there is no estimated risk of chronic hepatitis C virus, the incidence of liver cancer in hepatitis C carriers is as high as 5% per year compared to 0.5% in hepatitis B carriers per year (Di Bisceglie, Semin. Liver Dis., 15). : 64-69, 1995). Persistent hepatitis C virus infection is the leading cause of hepatocarcinoma in North America, with 70% of the causes of liver cancer in Japan and an increase in hepatitis C virus infection (Hasan, et al., Hepatology, 12: 589-591, 1990 EI-Serag, et al., N. Eng /. J. Med., 340: 745-750, 1999).

Several chemicals are involved in the development of liver cancer. The most important of them is ethanol and alcohol abuse is associated with liver cancer (Schiff, Hepatology, 26: 39S-42S, 1997; Nalpas, et al., Alcohol, 12: 17-120, 1995). Ethanol is considered a co-carcinogen that causes cirrhosis or other factors such as hepatitis B virus and hepatitis C virus. Aflatoxin, produced by some fungi, is also associated with liver cancer (Yu, J. Gastroenterol. Hepatol., 10: 674-682, 1995). These fungi tend to grow in grains, peanuts and other foods and are a major cause of food rot.

Diagnosis of liver cancer is relatively easy for patients with space-occupying lesions on ultrasonography or computed tomography, or those with 500-ng / ml serum alpha-fetoprotein (AFP). Fong, et al., In Cancer of the Liver and Biliary Tree in Cancer.Principles & Practice of Oncology, 6th Edition, 1162-1199, 2001). However, liver cancer at the time of onset of symptoms usually cannot be treated because AFP does not increase measurably.

Diagnosing small lesions with burns, even by ultrasonography, CT or MRI, is relatively inaccurate (Fong, et al., In Cancer of the Liver and Biliary Tree in Cancer: Principles & Practice of Oncology, 6th Edition, 1162-). 1199, 2001; Murakami, et al., Detectability of hypervascular hepatocellular carcinoma by arterial phase images of MR and spiral CT.Acta Radiol., 36: 372-376, 1995). In particular, cirrhosis nodular and dysplastic nodules are very similar to liver cancer on radiographs.

In addition, liver biopsies of small lesions are not sensitive or specific (Levy, et al., Ann. Surg., 234: 206-209, 2001). A well-differentiated cancer is often indistinguishable from benign tumors even with needle biopsy due to the limited amount of material obtained in this process (Fong, et al., In Cancer of the Liver and Biliary Tree). in Cancer.Principles & Practice of Oncology 6th Edition, 1162-1199, 2001). Therefore, despite advances in imaging technology, there is still a need for suitable molecular markers to distinguish liver cancer from benign liver lesions.

Tumor size is an important risk factor for the spread and metastasis of liver cancer (Yuki, et al., Cancer, 66: 2174-2179, 1990). On the other hand, patients with small tumors have more treatment options, but symptomatic tumors are usually large and untreated. Sensitive and specific liver cancer markers are useful for screening patients at risk for liver cancer, such as chronic carriers of the hepatitis B virus. Although this screening is widely applied, little data is available on reducing the disease (Collier, et al., Hepatology, 27: 273-278, 1998). One reason screening is not effective is because serological tests currently used are primarily sequential AFP assays with low sensitivity and specificity (Collier, et al., Viral Hepatitis Reviews, 4: 31-41, 1998). Therefore, improved test systems are needed to improve liver cancer screening.

The only molecular marker used for the screening and diagnosis of liver cancer is alphafetoprotein (AFP). This protein is mass synthesized by yolk sac and liver in embryonic development (Chan et al., In Tumor Markers. In Tietz textbook of clinical chemistry, 3rd, 722-749, 1999; Taketa, K, Hepatology, 12: 1420-1432, 1990). AFP concentration gradually decreases to <10 ng / ml within 12-18 months of birth.

AFP reappears in maternal serum during pregnancy. Increased circulating AFP is associated with liver cancer, digestive cancer, lung cancer, pancreatic cancer, biliary tract cancer and testicular carcinoma (Chan, et al., In Tumor Markers. In Tietz textbook of clinical chemistry, 3rd, 722-749, 1999).

If the AFP level is 20 ng / ml or higher, 60-80% of liver cancers are diagnosed but sensitivity is significantly lower (40%) for small tumors (Trevisani, et al., J. Hepatol., 34: 570-575, 2001).

Another problem with using AFP as a liver cancer marker is that it is not specific; Significant increases in AFP (20-200 ng / ml) are seen in many patients with chronic liver disease (Collier, et al., Viral Hepatitis Reviews, 4: 31-41, 1998). 15-58% of chronic hepatitis patients and 11-47% of cirrhosis patients reported increased serum AFP (Taketa, Hepatology, 12: 1420-1432, 1990). Therefore, the same serum AFP levels in liver cancer patients and cirrhosis patients are common. This leads to confusion in interpreting AFP assay results.

As a result, the usefulness of measuring AFP as a method of identifying patients at risk for liver cancer has been questioned (Sherman, J. Hepatol., 34: 603-605, 2001). Therefore, AFP measurements should be performed only when confirming an initial diagnosis based on imaging techniques (Sherman, J. Hepatol., 34: 603-605, 2001).

Since AFP levels are not reliable, most screening assays include ultrasonography, which is very sensitive to liver mass. Ultrasonography, however, is not specific and does not clearly distinguish between liver cancer, cirrhosis nodules and dysplastic nodules when lesions are smaller than 2 cm.

Over the past few years, several new candidate markers for liver cancer have been studied but have not found anything that surpasses AFP (Seow, et al., Proteomics, 1: 1249-1263, 2001). For example, des-gamma-carboxy prothrombin has been proposed as a candidate marker. However, his molecule does not detect tumors as small as AFP (Nomura, et al., Am. J. Gastroenterol., 94: 650-654, 1999).

In 1997, Hsu et al. Conducted differential mRNA display analysis of normal liver and liver cancer to identify and report transcripts that are upregulated in liver cancer (Hsu, et al., Cancer Res., 57: 5179). -5184, 1997).

This transcript, named MXR7, has been identified as Glypican-3 (GPC3).

GPC3 is a heparan sulfate proteoglycan bound to the cell surface by a lipid tail (Duenas Gonzales, et al., J. Cell. Biol., 141: 1407-1414, 1998). Hsu et al. Found that GPC3 mRNA was expressed in 143 (74.8%) of early and recurrent liver cancer 191, and 5 out of 154 (3.2%) in normal liver.

A second study found GPC3 mRNA overexpression in 75% of liver cancers, but not in local nodular hyperplasia and cirrhosis (Zhu et al., Gut, 48, 558-564, 2001).

This study only analyzed mRNA levels and it is known that mRNA levels are not always related to protein expression and secretion. In addition, only a limited number of liver proteins are normally secreted into the blood and circulated.

These results indicate that GPC3 mRNA analysis in liver tissue may be useful for detection of liver cancer, while this assay is invasive because it requires the separation of tumor tissue. Furthermore, mRNA analysis is time consuming and difficult to perform by mechanical procedures.

Therefore, serum markers are needed to perform screening (ie can be positive for asymptomatic patients with a high proportion of small liver cancers). In patients with small tumor lesions detected by ultrasound, the test can also clearly distinguish between patients without malignant lesions and those with liver cancer.

In conclusion, better molecular markers are needed for the surveillance of high-risk groups, such as hepatitis C patients who still have liver cancer, especially chronic cirrhosis. It is very difficult to find such markers because the characteristics that indicate liver cancer are so diverse (Thorgeirsson et al., Nature Genet., 31: 339-346, 2002).

Preferred embodiments of the invention are described in the drawings. From here:

1 shows micrographs stained with anti-GPC3 monoclonal antibody 1G12, 293 cells transformed with GPC3 (Panels C and D) and control cells transformed with vectors only (Panels A and B). Panels A and C are phase contrast micrographs and panels B and D are fluorescence micrographs.

Figure 2 shows Westerns of protein from 293 cells transformed with vector only (EF) or HA labeled GPC3 (GPC3). Membrane was incubated with anti-GPC3 antibody 1G12 or anti-HA antibody (12CA5)

Closed arrow: GPC3 nuclear protein; Open arrow: glycanated form of GPC3;

Arrowhead: Cleavage product comprising the N-terminus of GPC3 that does not contain an epitope detected by 1G12. The number on the right corresponds to the molecular weight marker.

3 shows micrographs of portions of liver cancers immunohistochemically stained using monoclonal antibody 1G12.

Panel A: 16X overview of highly GPC3-positive liver cancer (T) surrounded by non-tumour hepatocytes (N);

Panel B: 400 × view of the same tissue showing intracellular granules and membrane staining.

4 shows the relationship between GPC3 concentration and the optical density of an ELISA assay for GPC3.

FIG. 5 shows GPC3 levels in serum of control, hepatitis patients (H), hepatitis and cirrhosis patients (H + LC), and liver cancer patients (HCC) as determined by the mean of four ELISAs. The Significance of difference between mean GPC3 values in mean liver cancer patients and cirrhosis patients is shown at the top of the figure.

The inventors have discovered a new serum marker for liver cancer, glypican-3 (GPC3), which provides the basis for a new, quick, simple and non-invasive assay that shows excellent specificity in the diagnosis of liver cancer.

The present invention provides a method for screening a liver cancer patient, as in an embodiment of the present invention.

Obtaining body fluid from the patient; And

Determining the level of glypican-3 (GPC3) in the sample;

The presence of measurable GPC3 levels in the sample here means that the patient is liver cancer.

The present invention provides a fragment from which a purely isolated antibody or fragment derived therefrom, which specifically binds to GPC3 or a fragment derived therefrom, as in another embodiment of the invention.

The present invention provides a method for diagnosing liver cancer of a patient in another embodiment of the present invention comprising:

Obtaining a liver tissue sample from the patient;

And determining whether GPC3 is present in the tissue sample;

Here, detection of GPC3 indicates liver cancer.

The present invention provides a method of detecting the presence of GPC3 in a sample comprising the following in another embodiment of the present invention.

Contacting the sample with an antibody or fragment therefrom that specifically binds to GPC3 or a fragment derived therefrom to form an antibody-GPC3 complex or an antibody-GPC3 fragment complex; And

Detecting the antibody-GPC3 or antibody-GPC3 fragment complex.

In another embodiment of the invention, the present invention provides a method for determining the level of GPC3 in a sample comprising:

Contacting the sample with an antibody or fragment therefrom that specifically binds to GPC3 or a fragment derived therefrom to form an antibody GPC3 or antibody-GPC3 fragment complex; And

Detecting the antibody-GPC3 or antibody-GPC3 fragment complex

The present invention provides a kit for detecting or measuring the level of GPC3 in a sample, which in another embodiment of the present invention comprises:

Fully purely isolated antibody or fragment therefrom that specifically binds GPC3 or a fragment therefrom.

The present invention provides a new method for screening mammalian liver cancer patients. We can detect cell surface proteoglycan GPC3 in the blood, serum, or plasma of liver cancer patients, i.e., detectable in normal human serum or plasma using a standard test of significance as described herein. It turns out that it does not appear at a level that is significantly higher than the background. It was also shown that little GPC3 was detected in serum or plasma of patients with hepatitis, hepatitis and cirrhosis.

The present invention therefore provides a convenient, non-invasive method for screening liver cancer patients by measuring the level of GPC3 in the patient's body fluids. Suitable body fluids for the test include serum, plasma and whole blood. Screening of plasma or serum is preferred.

GPC3 levels in body fluids can be measured by any suitable method of measuring the protein. Immunological methods of measuring GPC3 levels are very simple and suitable types of methods will be readily understood by those skilled in the art and will be described in more detail herein.

For example, antibodies or fragments of antibodies that specifically bind to human GPC3 provide a variety of assay methods based on antibodies that determine the amount of GPC3. Liver cancer is diagnosed by detecting and / or determining the level of the complex by contacting the body fluid of the patient with an antibody or fragment that specifically binds GPC3 under the conditions of forming the antibody-GPC3 complex. The amount of antibody-GPC3 complex is related to the amount of GPC3 in the sample.

The term “antibody” as used in the text refers to polyclonal antibodies, monoclonal antibodies, chimeric antibodies, humanized antibodies, and single chains unless otherwise specified. Antibody (single chain antibody).

"Antibody fragment" refers to a portion of an antibody that exhibits specific binding, such as a parent antibody, and includes Fab, F (ab ') ₂ and Fv fragments.

As used herein, expression of an antibody or antibody fragment “specifically binds” a target molecule means that the antibody or antibody fragment recognizes and binds to the target protein, but substantially recognizes other molecules in the sample containing the target protein. It does not mean not to combine.

Chimeric antibodies are antibodies that include portions of antibodies from other species. For example, chimeric antibodies have variable regions from human constant regions and other species. Chimeric antibodies are described in US invention Nos. 5,354, 847 and 5,500, 362 and scientific literature (Couto et al., Hybridoma, 12: 485-489, 1993).

Humanized antibodies are antibodies from which complementarity determining regions (CDRs) that express antigen binding and specificity are not derived from humans, but the rest of the antibody molecule is derived from humans. Humanized antibodies and their preparation are also described in US invention Nos. 5,225, 539; And 5,585, 089; Well known as in 5,693, 761 and 5,693, 762.

Single chain antibodies are polypeptide sequences that can specifically bind to peptides or epitopes. Wherein the single chain antibody is derived from the L-chain or H-chain of the monoclonal or polyclonal antibody. Single chain antibodies include polypeptides derived from humanized, chimeric or fully human antibodies. Wherein the single chain antibody is derived from the L-chain or H-chain of the polypeptide.

Antibodies

Polyclonal antibody

To prepare polyclonal antibodies, purely isolated GPC3 human or non-human can be obtained. For example, GPC3 is derived from Bonneh-Barkay et al. J. Biol. Chem. , v. 272, pp. It may be prepared in the traditional manner as described in 12415-12421 (1997). Purely isolated proteins can be combined with carrier proteins such as keyhole limpet hemocyanin (KLH) and combined with emulsifiers (Freund's adjuvant), as known to those skilled in the art, to be administered to rabbits or other suitable laboratory animals. Alternatively, all or part of the GPC3 protein may be recombinantly synthesized by expression of the appropriate DNA sequence inserted in a suitable cloning carrier in a suitable host such as a bacterium. DNA encoding GPC3 can be cloned as previously described (Filmus et al., (1988), Molec. & Cell Biol., V. 8, pp. 4243-4249; Linsley et al., (1993), Ann, Rev. Immunol., V. 11, pp. 191-212; Peach et al., J. Exp. Med., V. 180, pp. 2049-2058, 1994). GPC3 or a portion thereof may be expressed as a fusion protein in a similar manner. Two widely used expression systems for producing recombinant fusion proteins in E. coli are glutathione-S-transferase or maltose binding proteins.

Fusion with pUR series vectors and trpE fusion with pATH vectors

The expressed GPC3 protein may be purely isolated using a glutathione column or the like, combined with a carrier protein, mixed with an emulsifier (to stimulate animal antigen response), and then administered to rabbits or other suitable experimental animals. Weeks are then followed up weekly to collect blood from rabbits or other laboratory animals and to separate serum. Serum is purified prior to use before use or by a variety of methods including affinity chromatography with protein A-Sepharse, antigen sepharose or anti mouse Ig sepharose.

Polyclonal antibodies can be prepared using fragments of GPC3 as antigens. Three or more consecutive amino acids derived from the GPC3 amino acid sequence can be used. GPC3 fragments or portions thereof comprising a 70 amino acid carboxy terminal amino acid sequence are preferred antigens. Fragments can be recombinantly produced as described herein and, optionally, made of fusion proteins. Such peptide antigens may be combined with keyhole limpet hemocyanin to enhance antigenicity in vivo.

Monoclonal antibody

Monoclonal anti-GPC3 antibodies can be produced by conventional methods after administration of a purely isolated fragment of GPC3 or GPC3 to the antibody mice obtained as described above for polyclonal antibody preparation, optionally with a fusion protein.

Briefly, purely isolated proteins or peptides are administered to mice with emulsifiers, for example, in cycles nine times over three weeks. Mouse pancreas was extracted and resuspended in phosphate buffered saline (PBS).

Interest cells serve as a source of lymphocytes, some of which produce antigens with appropriate specificity. They are fused with constantly growing myeloma partner cells and the fusion hybridomas are plated into multiple tissue culture wells in the presence of a selective agent such as HAT. Wells are then screened by ELISA to identify wells containing cells that make antibodies specific for GPC3. These cells are plated and then screened to identify antibody-producing cells again after the growth cycle. Some cloning procedures are performed up to 90% of wells containing monoclones that are positive for antibody production. This creates a stable line of clones that produce antibodies. Monoclonal antibodies can be isolated by affinity chromatography or protein-exchange chromatography using Protein A Sepharose, variants of these techniques, or a combination of both.

Further embodiments of the present invention prepare antibody fragments comprising F (ab) ₂ , F (ab) ₂ and Fv fragments or single chain components from GPC3 monoclonal antibodies. Such fragments are described in Lin et al., PNAS, 75 (6); Through proteolytic cleavage of anti-GPC3 antibodies as described in 2649-2653 (1978), the genes encoding such fragments can be genetically manipulated or secreted such anti-GPC3 antibody fragments. Can be produced by fermentation of a transformed production system. Anti-GPC3 antibodies for antigen affinity are preferred at least 1 × 10 ⁸ .

Various types of assay formats can be used to assay GPC3 in the sample. These include sandwich ELISA, radioimmunoassay, fluoroimmunoassay, dot-blot, dip-stick and Western blot.

In a preferred embodiment, antibodies that specifically bind to GPC3 or fragments therefrom and antibodies having a detectable label are used and the resulting GPC3-antibody conjugate is determined by measuring the detectable label and as a result the level of GPC3 in the sample Will be determined. Also used are primary antibodies specific for GPC3 or fragments derived therefrom, secondary antibodies with labels that can be measured directly and specific for the primary antibody or that can be a component of the signal generating system. Such labeled antibodies and systems are well known in the art.

Detection and measurement of the signal generated by the label or signal generation system allows the determination of the GPC3-antibody complex, ie, GPC3, present in the sample in comparison with suitable criteria.

Secondary antibodies have a suitable label that can be directly detected or a label that is an element of a suitable signal generating system.

Many examples of labels are well known in the immunoassay art.

Labeling secondary antibodies with detectable labels or elements of a signal generating system is performed by techniques well known in the art. Examples of labels used to recognize antibodies include radioisotopes, enzymes, fluorescent and chemiluminescent materials. For example, radioactive elements can be used as labels that are directly detected; Examples of typical radiolabels are I ¹²⁴¹ , I ¹²⁵¹ , I ¹²⁸¹ , and I ¹³¹ which emit gamma rays. Fluorescent labels can also be used as labels that are directly detected; Suitable fluorophores, for example, include coumarines, rare earth metal ions, chelates or chelate complexes, fluorescein, rhodamine and rhodamine derivatives.

Suitable labels are also metal complexes, stable free radicals, vesicles. Liposomes, colloidal particles, latex particles, spin labels, biotin / avidin and derivatives thereof.

Alkaline phosphatase, amylase, luciferase, catalase, beta-galactosidase, glucose-6-phosphatase dehydrogenase, hexokinase, horseradish peroxidase, lactamase, Enzyme-linked signal-generating systems may also be used, including urease and malate dehydrogenase. The activity of the enzyme can be detected by measuring absorbance, fluorescence or color intensity after reaction of the enzyme with a suitable substrate. If an enzyme is used as a label, the linkage between the enzyme and the antibody can be accomplished using conventional methods such as glutaraldehyde, periodic acid and maleimide.

Materials that can serve as solid supports suitable for immobilizing antibodies include plastics from Falcon, Oxnard, Calif. Or ELISA micro titration plates (ImmulonII, Dynax, Chantilly, VA) and streptavidin, for example. ) Coated ELISA titration plates (Reacti-Bind, Pierce, Rockford, IL) and micro titration plates such as products from Dynatech (Alexandria, VA).

The wells of the strip or microtiter plate are made of plastic and are preferably made of polyvinyl chloride or polystyrene. Other solid materials useful for antibody immobilization include polystyrene tubes, sticks or paddle polystyrene beads or polyacrylamide mattresses of any convenient size.

Antibodies are described, for example, in US Pat. It may be secured to the solid support by traditional methods well known in the art as described in 5,352, 583.

In a preferred embodiment of the present invention, a sample of a bodily fluid is first contacted with a primary antibody capable of specifically binding to GPC3 to form a complex, wherein the primary antibody is in a fixed state in solid form. You need enough time to get it to join. The solid support is then washed and contacted with a secondary antibody attached with a label or signaling system that can specifically bind and detect the primary antibody. The label bound to the solid support or the signal generated is measured to determine the complex present in the sample and thereby determine the level of GPC3 in the sample.

In further examples, the sample is contacted simultaneously with a primary antibody and a labeled secondary antibody immobilized on a solid support.

Furthermore, in embodiments, the secondary antibody may be devoid of labels or signal generating system elements, in which case the secondary antibody bound to the solid support is measured by a tertiary antibody having a label or signaling element that can be detected. The tertiary antibody binds specifically to the secondary antibody.

Furthermore, in an embodiment, the sample is contacted with a labeled secondary antibody that binds to GPC3 simultaneously or step-specifically, and with a primary antibody attached to one of the capture pairs and binds to the primary antibody. .

The resulting mixture is in contact with a fixed solid that is another member of the capture pair. After the labeled complex has given sufficient time to bind to the solid phase by the interaction of the capture pair members, the amount of label bound to the solid is measured to wash the solid and measure the level of GPC3 in the sample. Suitable capture pairs include biotin / streptavidin. Other suitable partners are well known to those skilled in the art.

The specificity of the antibody can be reversed as if the primary antibody specifically binds to the complex and the labeled secondary antibody specifically binds to GPC3. The screening methods described herein are considered to be very specific for liver cancer because GPC3 is not detected in samples of normal and hepatitis patients and negligible levels are sometimes measured in patients with hepatitis and cirrhosis.

In experiments with patients diagnosed with liver cancer, serum GPC3 levels were measured between 151 ng / ml and 2924 ng / ml in 53 to 55% of patients. When serum AFP is used as a liver cancer marker of the same patient and 20 ng / ml is used as the threshold for the condition, 59% of liver cancer patients are detected. If the threshold level of AFP, the marker of disease, rises to 100 ng / ml, only 32% of liver cancer patients are detected. However, when the threshold is 20ng / ml AFP, in many chronic liver diseases other than liver cancer, the AFP level is higher than 20ng / ml, so it will include patients who are not actually liver cancer.

The methods of the present invention can be used alone or in combination with AFP screening by traditional methods to improve the ability to identify patients with liver cancer at early and treatable stages.

Further in embodiments of the invention, GPC3 and AFP screening can be used together to provide a test that combines the specificity of screening high levels of AFP with improved specificity of GPC3 screening. For example, if the AFP and GPC3 serum levels are measured and the threshold of AFP is 100 ng / ml, as shown in the data in Table 4, then 100 ng / ml AFP alone is 32% when used as a liver cancer marker or GPC3 alone. When used as a marker, 53% of patients with liver cancer show an increase in at least one of AFP and GPC3. If both AFP and GPC3 are considered and 20ng / ml AFP is used as the threshold, 82% of patients will have at least one of AFP and GPC3 elevated.

 The serological GPC3 assay of the present invention is particularly useful for diagnosing patients who are clinically showing signs of liver tumors.

Assays are also useful for screening populations of people at risk for liver cancer, such as patients with chronic hepatitis B or hepatitis C or carriers.

Further in embodiments the present invention provides an immunohistochemical method for diagnosing liver cancer by examining a liver biopsy sample from a patient exhibiting a liver tumor. Liver tissue samples obtained by biopsy often exhibit distorted morphology, making traditional histological diagnosis difficult. Tissue sections obtained by biopsy are stained with anti-GPC3 antibodies as described in the examples in the text and when GPC3 is detected in the sections, the biopsy tumor is liver cancer.

Anti-GPC3 antibodies can be detected with a secondary antibody labeled or capable of detecting a detectable label such as a fluorescent label. The invention also provides antibodies that specifically bind to GPC3. Antibodies that bind to the same epitope with such antibodies are also included in the present invention. The antibody may be of the immunoglobulin group, including the antibodies may be of any immunoglobulin class including IgG, IgM, IgA, IgE and IgD, and any subclass therefrom.

Further in embodiments the present invention provides a kit comprising a purely isolated antibody or fragment derived therefrom that specifically binds to GPC3 or a fragment derived therefrom.

In further examples, the kit is for conducting a standard two antibody sandwich assay comprising a GPC3-specific primary antibody that captures GPC3 present in a liquid sample and a secondary antibody that detects the presence of the captured GPC3. Primary antibodies are typically immobilized on solids such as ELISA plates, nitrocellulose membranes, beads or other supports well known in the art. The antibody to be detected is detected by a secondary antibody, which is a labeled antibody that is specific for the isotype of the antibody that is directly labeled or detected with a colorimeter or radioisotope label.

The examples are described for the purpose of illustrating the invention and are not intended to limit the scope of the invention.

The methods of chemistry, molecular biology, protein and peptide biochemistry and immunology used for reference are not explicitly described in the present invention but are reported in the scientific literature and are well known to those skilled in the art.

Way

Patient tissue: A sample of liver tissue was obtained from a pathology classroom at Toronto General Hospital. All liver specimens are large chunks from surgically cut tumors and surrounding parenchyma. All tissues were fixed in 10% formalin and buried in paraffin for general biopsy.

Patient Serum: 34 patients with liver cancer (Table 3 patient characteristics), 20 patients with hepatitis and cirrhosis (12 hepatitis C, 8 hepatitis B), 18 patients with hepatitis (12 hepatitis C) , 1 patient with hepatitis C and HIV, 1 patient with hepatitis C and thalesemia, 1 patient with both hepatitis C and hepatitis B, 3 with hepatitis B Blood samples from 53 healthy recipients (Sunnybrook and Women's College Health Sciences Centre, Toronto, Canada). Liver cancer was diagnosed histologically when a liver biopsy was available, or with clinical information according to the European Association's Guidelines for Study of Liver Disease. Serum was soon isolated by centrifugation and frozen at minus 20 degrees. When serum was collected, serum from patients diagnosed with non-malignant liver disease (hepatitis and hepatitis with cirrhosis) was included only for six months after collection.

Cell lines: Hepatocarcinoma cell lines HepG2, Hep3B and PLC-PRF-5 were incubated in minimal essential medium (MEM) (Gibco BRL) with 10% fetal bovine serum (FBS), where MEM non-essential amino acid solution and sodium 1 mM sodium pyruvate was supplemented. 293 cell lines were incubated in DMEM and 10% FBS. All cell lines were obtained from the American Type Culture Collection and maintained at 37 degrees Celsius in humid air at 5% CO ₂ . Cell lines were grown at high density overnight in the presence of serum to collect cell culture. The collected medium was concentrated using Centricon YM-10,000 filters (Millipore).

Transformation of GPC3: The 293 cell line was transformed with GPC3 cDNA labeled with hemaglutinin A (HA) and inserted into the pEF vector (or empty vector as a control) and the transformed cells were 800 lig / ml G418 ( Gibco BRL). Cells expressing high levels of GPC3 were sorted with FACS after staining G418-selected cells with anti-HA 12CA5 monoclonal antibody and FITC-bound secondary antibody.

Production of Anti-GPC3 Mouse Monoclonal Antibodies: Balb / C mouse females were immunized with a single intraperitoneal injection of 50 μg of His labeled GPC3 containing the last 70 amino acids of the nucleoprotein. Immunogens were emulsified with Titermax Gold (Cedarlane) adjuvant. After 21 days mice were boosted without adjuvant with 50 μg of the same immunogen. Two days later, blood samples were collected from the tail vein to obtain the serum. Anti-GPC3 titers were measured by ELISA using sera previously immunized from the same mice in serum samples as negative controls. Splenocytes from two mice with high titers were fused with SP2 / 0Ag14 mouse myeloma cells in the presence of polyethylene glycol 4000 (Sigma) using standard methods (ratio 6: 1). The fused cells were placed in 96-well plates in 20% FBS, aminopterin and thymidine (HAT supplement, Gibco BRL) with alpha-MEM, 10 mM Hepes 50 uM 2-mercaptoethanol, Nutridoma-CS (Roche) and hypoxanthine. Clones were screened by ELISA using 96 well plates filled with 0.05 μg of GPC3 immunogen per well to screen for hybridoma production. Plates were blocked with 1% bovine serum albumin (BSA) in PBS (phosphate-buffered saline) for 2 hours and conditioned media was added from each clone. Anti-GPC3 antibody bound to the antigen is based on O-phenylenediamine (Sigma) and hydrogen peroxide as substrates of goat anti mouse-IgG conjugated with horse radish peroxidase And goat's anti mouse-IgM (StressGen) was measured. The positive clones were then expanded and the ability to stain 293 cells transformed with GPC3 was tested by immunofluorescence. The positive hybridomas, designated 1G12 and 8H5, were cloned again and their isotypes (IgG1, K) were determined using Isostrip (Roche).

Antibodies were purified from cell culture supernatants using a Protein A column (Affi-Gel Protein A MAPSII kit, Bio-Rad).

Immunofluorescence: 293 cells transformed with GPC3 or 293 cells transformed with the vector alone were seeded on poly-L-lysine treated slides and fixed with 4% paraformaldehyde in PBS. After 1 hour incubation with anti-GPC3 monoclonal antibody or normal mouse IgG (Santa CruzBiotechnology) (1511 g / ml), slides were washed three times with PBS and FITC bound anti-mouse Primary antibodies detected with IgG F (ab ') ₂ fragments (StressGen) were bound.

Western blot analysis: Cells were lysed for 30 minutes on ice in RIPA buffer containing protease inhibitors (2 mM PMSF, 10 pg / ml leupeptin, 101 lg / ml aprotinin). Protein samples were electrophoresed on 6% SDS polyacrylamide gels and then transferred to PVDF membranes (Pall Corporation). The membranes were blocked for 1 hour at room temperature in blocking buffer (10 mM Tris-HCI pH 8.0, 150 mM NaCI 0.1% Tween 20,5% non-fat dry milk) and then with 1 pg / ml 1G12 or 12CA5 monoclonal antibody. It was incubated overnight at 4 degrees Celsius.

 Protein bands were detected using chemiluminescent ECL reagent (DuPont NEN) after 1 h of chilling with horseradish peroxidase anti-mouse IgG secondary antibody.

Immunohistochemistry: Paraffin-covered tissue sections were stripped of paraffin with xylene and rehydrated in gradated series ethanol solution. The antigen recovery technique then heats the slides in a Kenmore microwave oven at 10 mM citrate buffer pH 6 (2 minutes at power level 10, 1.5 minutes at power level 6 and 2.5 minutes at power level 9 and repeats this twice). This was done. The slides were washed with distilled water and rehydrated in ethanol at the point of concentration and then immersed in 10% hydrogen peroxide in methanol for 10 minutes to block intracellular peroxidase. Slides were rehydrated using 20 μg / ml anti-GPC3 monoclonal antibody (or normal mouse IgG and PBS as negative controls) as directed by Histostain-SP kit (Zymed Laboratories) and TSATM Biotin System (NEN Life Science Products). Immunostaining was performed. Diluted to 1/400, staining consecutive sections with polyclonal antibody (Dako, A008) against rabbit AFP. Intracellular peroxidase was blocked with aqueous hydrogen peroxide and antibody binding was detected by Ultra HRP system (ID Labs).

Sandwich ELISA: 96 well ELISA plates were covered with 0.5 μg anti-GPC3 monoantibody 1G12 in 50 μl PBS per well. Plates were blocked with 1% BSA in PBS and allowed to cool overnight at 4 degrees Celsius by adding 50 μl of diluted serum sample (1: 3 PBS). The unbound material was washed with 0.1% BSA in PBS and bound GPC3 was detected using anti-GPC3 shep polyclonal antibody followed by O-phenylenediamine (Sigma) and hydrogen peroxide. The donkey's anti-sheep IgG (Sigma) combined with horseradish peroxidase was maintained as a substrate. To quantify GPC3 present in serum, a calibration curve of purely isolated GPC3 diluted in equal dilution and added to the pool of normal serum was drawn. To determine GPC3 in normal serum, a GPC3 calibration curve was drawn in PBS. Each sample was measured three times in four sets.

Statistical analysis: The significance of the differences between the groups was determined by the Mann-Whitney test.

Serological Concentrations of AFP: AFP levels in serum samples were measured with a commercially available ELISA kit (Axsym, Abbott).

Example 1 Production and Analysis of Polyclonal Antibodies to GPC3

 Part of the cloned GPC3 Cdna (Filmus et al., Homologous), which encodes 70 consecutive amino acids of the C-terminus of human GPC3, is a glutathione-S-transferase in a traditional E. coli expression system. Expressed as a fusion protein.

The fusion protein isolated from the glutathione column was used to immunize the sheep with the standard protocol. Polyclonal antibodies were purified by affinity separation of 70 amino acid immunogens by conventional methods as described in "Antibodies, a Laboratory Manual" (1988), Ed Harlow et al., Cold Spring Harbor. . Polyclonal antibodies were used for western blot and immunoprecipitation experiments.

Example 2: Hybridoma Production to Produce Monoclonal Antibodies to GPC3

The same immunogen, a GST fusion protein comprising 70 amino acids at the C-terminus of human GPC3, was used to increase monoclonal antibodies. The 70 amino acid regions are the most conserved parts between species and are therefore used to generate antibodies specific for human GPC3. Hybridomas were made using methods known to those skilled in the art, as described, for example, in Kohler and Milstein, Nature, 256: 495-497, (1975).

Briefly, Balb / C female mice were immunized with a single intraperitoneal injection of 70 amino acid human GPC3 fragments of His labeled carboxy terminus emulsified with a TitermaxGoldT (Cedarlane) adjuvant, including 50 μg. After 21 days mice were boosted with the same immunogen 50 μg without adjuvant. Two days later, blood samples were collected from the tail vein to obtain mouse serum. Anti-GPC3 titers were tested by ELISA using previously immunized serum from the same mice in serum samples as negative controls.

Interest cells from the two mice with the highest titers were standardized (Parkin, DM, Pisani, P. & Ferlay, J., Global cancer statistics.CA. Cancer J. Clin., 49,33-64, 1 , 1999) was used to fuse SP2 / 0 Ag14 mouse osteosarcoma cells (ratio 6: 1) with polyethylene glycol 4000. The fused cells were 96-well containing alpha-MEM, 10 mM Hepes, 50 LM 2- mercaptoethanol, 20% FBS with growth factor source Nutridoma-CSTM (Roche) and hypoxanthine plus thymidine for screening hybrids. Plated on plate.

 The obtained clones were selected by ELISA using 96 well plates loaded with 0.05 μg of GPC3 C-terminal fragments per well. Plates were blocked for 2 hours with 1% BSA contained in PBS and then placed in a cell culture supernatant. The binding of the primary antibody was detected using goat anti-mouse IgG and goat anti-mouse IgM (StressGen) conjugated with horseradish peroxidase with O-phenylenediamine and hydrogen peroxide as substrates. . Positive clones were reselected by immunofluorescence using cells transformed with GPC3. Selected hybridomas (1G12 and 8H5) were cloned twice to determine isotype using isotrip (Roche) (IgG1, K) and protein A column chomatography using Affi-Gel Protein A MAPSII kit (Bio-Rad). By pure separation from the cell culture supernatant.

Antibodies produced from hybridomas were screened for 70 amino acid GPC3 fragments using ELISA. Positive clones were tested for specific staining of cells transformed with GPC3. Two hybridomas (1G12 and 8H5) produced antibodies that potently and specifically immunostain the cells transformed with GPC3. 1 is a diagram showing staining of 293 cells with 1G12 antibody followed by a fluorescent secondary antibody. Panels A and C are phase contrast micrographs and panels B and D are fluorescence micrographs. Panels A and B show 293 cells transformed with the vector alone, while Panels C and D show cells transformed with the vector comprising GPC-3. In this example the cells are stained in the presence of 10 ug / ml monoclonal antibody 1G12. Panel D shows specific staining by mAb1G12. Staining is not seen in cells transformed with the vector alone (Panel B).

Example 3: Monoclonal Antibody Assay

The specificity of two monoclonal antibodies produced by 1G12 and 8H5 hybridomas was confirmed by Western blot protein supernatants of cells transformed with HA labeled GPC-3. As shown in FIG. 2, antibody 1GI2 detected a band corresponding to a GPC3 nuclear protein and a drag band corresponding to a glycanated form of GPC3. The same western blot experiment was conducted with anti-HA antibody (12CA5) as a control.

Example 4 GPC3 in the Tissue Section

Paraffin sections of liver cancer tumors and surrounding non-malignant cells were immunohistochemically stained as above with anti-GPC3 monoclonal antibody 1G12. Anti-GPC3 antibodies were detected as biotin-labeled anti-mouse immunoglobulin secondary antibodies and bound secondary antibodies were detected using streptavidin bound to HP.

As shown in FIG. 3, anti-GPC3 monoclonal antibody 1G12 binds strongly to tumor cells but does not bind at all to normal hepatocytes. In addition to the expected presence of GPC3 in the cell membrane, the cytoplasm was also stained to a large extent. Cytoplasmic staining mainly stained the granules and the area close to the cell surface. Occasional nuclear staining was also observed. Non-parenchymal cells were generally negative except for macrophages. These results indicate that anti-GPC3 monoclonal antibodies can be used to diagnose liver cancer by detecting GPC3 in liver tissue sections.

Example 5: In Vitro Measurement of GPC3 Expression in Malignant and Benign Sites

Table 1 summarizes the results of immunohistochemical studies of monoclonal antibodies against GPC3 that react with 15 different liver cancers. 20 (80%) of these liver cancers reacted with monoclonal antibodies against GPC3. There was a slight difference in the number of stained cells that reacted with anti-GPC3 mAb in each tumor. Positive cells usually gathered at sites similar to clones. In all cases non-malignant hepatocytes around the tumor were negative.

Expression of GPC3 in Liver Cancer . Acc. No. of HCC ’s mAb 1G12 mAb 8H5 % of positive cells 2188-96 I 3 3 70 21914- 3 3 80 4865-99 3 3 90 7530-93 A 3 3 100 7590-98 B 3 2 90 28828-99 A 3 - 100 12056-97 2 2 40-60 19626-98 A One One 1-5 18891-96 One One 5 6301-99 B One One 20 2978-97 B One One One 5940-96 One One 5-10 21424-98 A 0 0 - 10355-97 D 0 0 - 10865-01 C 0 0 -

Table 2 summarizes the response of anti-GPC3 monoclonal antibodies to malignant and non-malignant liver tissues. Six hepatic adenoma and nine hepatic dysplasia were negative for GPC3.

Expression of GPC3 in Non-malignant Liver Tissue Type of liver tissue Number of organization types tested GPC3 expression (%) Liver cancer 23 19 (83%) High degree of dysplasia 3 0 (0%) Low degree of dysplasia 6 0 (0%) Hepatocellular Adenoma 6 0 (0%) Normal liver 34 0 (0%)

The data indicate that GPC3 expresses a detectable level of glypican-3 in most liver cancers, but not in normal liver and benign sites.

Example 6: Immunological Assays for GPC3 Detection

Sandwich ELISA was used to measure GPC3 in the serum of patients diagnosed with liver cancer. For ELISA, 96-well plates were coated with anti-GPC3 monoclonal antibody 1G12 and the serum to be tested was aliquoted. Unbound material was removed by washing and bound GPC3 was detected using an anti-GPC3 shep polyclonal antibody prepared in Example 1 directly or indirectly bound to a measurement reagent useful for detecting the presence of immune complexes. The optical density of the sample was measured and compared with the results of the standard curve. 4 shows a standard curve for assays. Assay results are straight in the test range.

A protocol was developed to collect serum from patients treated at Toronto General Hospital and who signed a consent pledge. In preliminary studies, serum was collected from nine patients with liver cancer. Normal serum was also collected from 38 recipients. GPC3 levels were determined by subjecting the samples to 1: 3 (serum: PBS) followed by sandwich ELISA assay. Each quadruple was analyzed. The concentration of GPC3 was obtained by comparing the OD values obtained from each sample with a standard curve made by adding varying amounts of GPC3 in equal dilutions of a pool of normal serum.

Table 3 summarizes the findings. Five of nine patients with liver cancer (55%) had significantly higher levels of GPC3. None of these patients showed significantly increased AFP levels. GPC3 was not measured in any of the normal samples.

patient # GPC3 level (ng / ml) Standard error P One 100.6 17.7 0.0027 (*) 2 292.4 16.7 <0.0001 (*) 4 84.9 19.3 0.0081 (*) 5 0 9.6 NS 6 75.4 9.0 0.0014 (*) 7 14.9 3.5 NS 8 44.4 54.3 NS 9 102.6 12.9 0.0001 (*) 10 0 12.9 NS Pool of normal serum 0 10.1

* Significantly different from normal serum

Additional experiments were performed using sandwich ELISA to measure serum GPC3 levels in 53 healthy, 18 hepatitis, 20 hepatitis and cirrhosis patients and 34 hepatocellular carcinoma patients. GPC3 was not found in all healthy people and 18 of 34 liver cancer patients (53% sensitivity) showed significantly higher serum GPC3 levels in the range of 151 to 2924 ng / ml (Figures 5 and 4) (Figure 5 and Table 4). In addition, GPC3 was not measured in all patients with hepatitis and was measured at 117 ng / ml level in only one in 20 (5%) with hepatitis and cirrhosis (95% specificity). Statistical analysis using the Mann-Whitney test showed that GPC3 mean levels in liver cancer patients were significantly higher than those of hepatitis, cirrhosis, hepatitis, and healthy recipients, with little difference between the three groups.

Example 7: Comparing GPC3 and AFP Levels

Serum AFP levels were measured using an ELISA (Axsym, Abbott) available in the same set of 34 liver cancer patients described in Example 6. Table 4 shows a comparison between AFP and GPC3 obtained.

Serum Concentrations of GPC3 and AFP in Patients with Liver Cancer patient# GPC3 (ng / ml) AFP (ng / ml) One 2924 133 HBV 2 2295 12 HCV 3 1758 <5 HBV 4 1065 34 HCV 5 1026 14 HCV 6 1006 30 HBV 7 849 14 HCV 8 783 23 HBV & HCV 9 755 <5 HBV 10 445 14428 Unknown 11 433 10 HBV 12 351 7809 HCV 13 252 8629 HCV 14 235 1628 HCV 15 214 78 Alcoholic liver disease 16 211 <5 Alcoholic liver disease 17 184 5 HBV 18 151 40 Alcoholic liver disease 19 47 52000 Alcoholic liver disease 20 26 1516 HCV 21 0 8 HBV 22 0 30 HBV 23 0 <5 HBV 24 0 15 HBV 25 0 <5 HBV 26 0 488 Unknown 27 0 5 Alcoholic liver disease 28 0 2603 Unknown 29 0 194 HBV 30 0 13 HBV, HIV 31 0 67 HCV 32 0 27 HBV 33 0 74 HCV 34 0 1900 HCV

The invention is not limited to the nature of the embodiments described herein but includes all changes and modifications within the scope of the claims.

Claims (37)

How to screen for patients with liver cancer, including: Iii) obtaining a bodily fluid sample from the patient; And Ii) measuring the level of glycpican-3 (GPC3); The presence of detectable levels of GPC3 in the sample here means that the patient is liver cancer. Furthermore, the method of claim 1 comprising: Iii) measuring alpha-fetoprotein (AFP) levels in the sample, Wherein at least one of the detectable levels of GPC3 in the sample, an AFP level higher than the AFP level of a normal control patient in the sample, indicates that the patient is hepatocellular carcinoma. The method of claim 2, wherein the alphafetoprotein (AFP) level of the sample is greater than 20 ng / ml. 4. The method of claim 3, wherein the alphafetoprotein (AFP) level of the sample is greater than 100 ng / ml. 5. The method of claim 1, wherein the GPC3 level of the sample is measured by immunological methods. The method of claim 5, wherein an antibody or antibody fragment that specifically binds to human GPC3 or a fragment derived therefrom is used. The method of claim 6, wherein the antibody or fragment specifically binds to epitopes within 70 consecutive carboxy-terminal amino acids of human GPC3. 8. The method of claim 6 or 7, wherein the antibody or fragment is a polyclonal antibody. 8. The method of claim 6 or 7, wherein the antibody or fragment is a monoclonal antibody. 8. The method of claim 6 or 7, wherein the antibody or fragment is a fragment selected from the group consisting of Fab, F (ab ') ₂ and Fv. The method of claim 6, wherein the antibody or antibody fragment comprises a label that can be detected. The secondary of any one of claims 6 to 10, wherein the antibody or antibody fragment specifically binds to an anti-GPC3 antibody or antibody fragment and comprises one component of a signal-generating system. And the method is measured using an antibody. The solid material of claim 6, wherein the anti-GPC3 antibody or fragment is attached to a solid material and the sample is in contact with the solid material to bind the attached anti-GPC3 antibody or fragment to GPC3 in the sample. Silver washed and bound GPC3 is measured using a secondary antibody that specifically binds to an anti-GPC3 antibody or antibody fragment and comprises components of a signal generating system. The method of claim 12 or 13, wherein the secondary antibody is bound to horse radish peroxidase. The method according to any one of claims 2 to 14, wherein the level of alphafetoprotein (AFP) is measured by an ELISA assay. The method according to any one of claims 1 to 15, wherein the body fluid is serum or plasma. 17. The method of any one of claims 1 to 16, wherein the patient is a human patient. A sufficiently purely isolated antibody or fragment therefrom that specifically binds GPC3 or a fragment therefrom. The antibody of claim 18, wherein the antibody or fragment is a polyclonal antibody. The antibody of claim 18, wherein the antibody or fragment is a monoclonal antibody. The antibody or fragment of claim 18, wherein the antibody or fragment is selected from the group consisting of single chain antibody molecules, Fab fragments, F (ab) ₂ fragments, Fv fragments, and chimeric molecules. The antibody or fragment of any one of claims 18-21, wherein the antibody or fragment is selected from the group consisting of IgG, IgM, IgA, IgE, IgD and any subclass derived therefrom. The antibody or fragment according to any one of claims 18 to 21, wherein the antibody fragment specifically binds to GPC3 or a fragment derived therefrom. A hybridoma producing the antibody of any one of claims 18 to 23. How to diagnose liver cancer (HCC) in patients, including: Iii) obtaining a liver tissue sample from the patient; And Ii) detecting the presence of GPC3 in the tissue sample; Here, GPC detection in the sample is a marker of liver cancer. The method of claim 25, wherein the presence of GPC3 in the tissue sample is measured by contacting the tissue section with an antibody or antibody fragment that specifically binds human GPC3 and detecting and detecting an anti-GPC3 antibody or fragment bound to the tissue section. How to. The method of claim 26, wherein the bound anti-GPC3 antibody or fragment is detected using a secondary antibody that specifically binds to the anti-GPC3 antibody or fragment and comprises elements of a signaling system. The method of claim 27, wherein the secondary antibody is biotin bound. How to detect the presence of GPC3 in a sample containing: Iii) contacting the sample with an antibody or fragment derived therefrom that specifically binds GPC3 or a fragment derived therefrom to form an antibody-GPC3 complex or an antibody-GPC3 fragment complex; And Ii) measuring the antibody-GPC3 fragment complex. How to measure the level of GPC3 in a sample containing: Iii) contacting the sample with an antibody or antibody derived therefrom or a fragment derived therefrom that specifically binds to GPC3 or a fragment derived therefrom to form an antibody-GPC3 or antibody-GPC3 fragment complex; And Ii) measuring the antibody-GPC3 or antibody-GPC3 fragment complex. Kit to detect or measure the level of GPC3 in a sample containing: A sufficiently purely isolated antibody or fragment therefrom that specifically binds GPC3 or a fragment therefrom. The kit of claim 31, wherein the antibody or fragments derived therefrom specifically binds to human GPC3 or fragments derived therefrom. 22. The kit of claim 31 or 21, wherein the antibody or fragment specifically binds to epitopes within 70 consecutive C-terminal amino acids of human GPC3. The kit of claim 33, wherein the antibody or fragment is a polyclonal antibody. The kit of claim 33, wherein the antibody or fragment is a monoclonal antibody. 34. The kit of claim 33, wherein the antibody or fragment is selected from the group consisting of Fab, F (ab ') ₂ and Fv. The kit of claim 31, further comprising means for detecting and measuring the amount of GPC3-antibody complex or GPC3-antibody fragment complex.