Gene detection assay for improving the likelhood of an effective response to an ErbB antagonist cancer therapy

Abstract

The invention provides a method for more effective treatment of patients susceptible to or diagnosed with tumors overexpressing ErbB, as determined by a gene amplification assay, with an ErbB antagonist. Such method comprises administering a cancer-treating dose of the ErbB antagonist, preferably in addition to chemotherapeutic agents, to a subject in whose tumor cells ErbB has been found to be amplified e.g., by fluorescent in situ hybridization. ErbB antagonists described include an anti-HER2 antibody. Pharmaceutical packaging for providing the components for such treatment is also provided.

Description

[0001] This is a divisional application of non-provisional application Ser. No.09 / 863,101 filed on May 18, 2001, which claims priority under 35 U.S.C. § 119(e) to provisional application Ser. No. 60 / 205,754, filed May 19, 2000, the entire disclosures of which are incorporated herein by reference.FIELD OF THE INVENTION [0002] The present invention concerns the treatment of cancers characterized by the overexpression of of a tumor antigen, such as an ErbB receptor, particularly HER2. More specifically, the invention concerns more effective treatment of human patients susceptible to or diagnosed with cancer, in which the tumor cells overexpress ErbB as determined by a gene amplification assay, with an ErbB antagonist, e.g., an anti-ErbB antibody. The invention further provides pharmaceutical packages for such treatment. BACKGROUND OF THE INVENTION [0003] Advancements in the understanding of genetics and developments in technology and epidemiology have allowed for the correlation of gen...

AI Technical Summary

Benefits of technology

[0013] A particular advantage of the invention is that it permits selection of patients for treatment who, based on immunohistochemical criteria, would be excluded. Thus, in a specific embodiment, the subject has been found to have an antigen level corresponding to a 0 or 1+ score for HER2 by immunohistochemistry on a formaldehyde-fixed tissue sample.

[0014] The invention further provides a pharmaceutical package comprising an ErbB antagonist for treating a cancer, and instructions to administer the ErbB antagonist to a subject if an erbB gene in tumor cells in a tissue sample from the subject is amplified. Preferably the ErbB antagonist is an anti-ErbB antibody, such as an anti-HER2 antibody. In a further aspect, the instructions also teach administering a cancer treating dose of a chemotherapeutic, e.g., a taxol. Such pharmaceutical packages, including the instructions for use, can be provided for any antibody-based therapeutic specific for a tumor-specific antigen. DETAILED DESCRIPTION

[0015] The present invention advantageously permits treatment of patients who have a greater likelihood of responding to the treatment by administering therapeutic agents, i.e., anti-tumor antigen therapeutic antibodies or ErbB receptor antagonists, to patients who are found to have an amplified gene encoding such a tumor antigen or ErbB receptor protein. The invention is based, in part, on the unexpected discovery that her2 gene amplification, e.g., as detected by fluorescence in situ hybridization (FISH), although it correlates with HER2 expression as detected by immunohistochemistry (IHC), provides a more accurate basis for selecting patients for treatment because FISH status unexpectedly correlates better with response to treatment. This outcome was surprising in part because FISH status has about the same rate of correlation with a clinical trial assay (CTA) IHC assay as another IHC assay (Hercep Test). Based on this observation, FISH would be expected to have a similar correlation with treatment response. This outcome also surprises because direct measurement of protein (by immunoassay) would be expected to provide a more accurate assessment of a cancer therapy targeted to the protein than an indirect measure of expression, like gene amplification.

[0016] Evaluation of patient groups and subgroups demonstrates the power of gene amplification analysis for selecting patients more likely to respond to treatment. IHC provides a score for HER2 expression on tumor cells: 0 (no expression) through 3+ (very high levels of expression). Clinical selection criteria exclude patients with 0 and 1+ scores and select patients with 2+ and 3+ scores. The data show that 14% of combined 2+ / 3+patients respond to Herceptin®, while 20% of FISH+ (amplified her2 gene) patients respond to Herceptin®. The 3+ subgroup has a 17% response rate, which is very close to the FISH+ subjects' response rate. However, the 2+ subgroup has less than half the response rate of FISH+ subjects. Thus, gene amplification clearly differentiates large sub-populations within the 2+ subgroup, permitting more effective treatment for those who are FISH+, and quickly identifying patients for whom alternative treatment modalities are appropriate and should commence immediately.

[0017] Gene amplification analysis also identifies patients who are unnecessarily excluded because of anomalies in the IHC analysis, particularly when the tests are performed on formalyn fixed, paraffin embedded samples (such sample processing can disrupt or destroy antibody epitopes on the HER2 protein, but has much less impact on gene amplification assays). As shown in the examples, a subset of 0 and 1+ subjects are FISH+. These patients are likely to respond to anti-HER2 antibody therapy, e.g., with Herceptin®, although by IHC criteria they would be excluded from receiving this treatment.

[0018] Thus, the present invention advantageously permits inclusion of patients who are more likely to benefit from treatment but who, by standard IHC criteria, would be excluded from treatment. At the same time, the invention permits exclusion of patients who should promptly seek an alternative mode of treatment because the anti-tumor antigen therapy (i.e., ErbB antagonist or tumor antigen-specific therapeutic antibody) is not likely to succeed.

Problems solved by technology

However, most of the methodologies available for evaluation of tissue for the presence of genes associated with or predisposing an individual to a malignancy have well-known drawbacks.

For example, methods that require disaggregation of the tissue, such as Southern, Northern, or Western blot analysis, are rendered less accurate by dilution of the malignant cells by the normal or otherwise non-malignant cells that are present in the same tissue.

Furthermore, the resulting loss of tissue architecture precludes the ability to correlate malignant cells with the presence of genetic abnormalities in a context that allows morphological specificity.

This issue is particularly problematic in tissue types known to be heterogeneous, such as in human breast carcinoma, where a significant percentage of the cells present in any area may be non-malignant.

Thus, IHC can lead to false negative results, excluding from treatment patients who might benefit from the treatment.