The problem of inadequate broad-range prognostic factors for breast
cancer is solved by providing a multi-gene probe for a single-step
determination of disease outcome. The 5-genes chosen (HER2, Topo
II.alpha., NM23-H1, CK19 and MMP9) are known to show altered expression
in different breast tumors. The altered expression of any one of
them has a similar prognostic significance, with respect to disease-free
survival or overall survival. The multi-gene probe is labeled and
used to screen a tumor specimen. The signal generated after the
assay provides information that has prognostic significance. Since
five prognostic markers are simultaneously assessed, a wider variety
of breast cancers are covered and disease outcome prediction is
improved in a wider population.
1. A method of assessing breast cancer prognosis in a human subject,
comprising the steps of: (a) obtaining a biopsy of a tumor from
the breast of the human subject; and (b) analyzing the sample with
a multi-gene probe targeting the HER2, Topo II.alpha. NM23-H1, CK19
and MMP9 DNA.
2. The method of claim 1 wherein the multi-gene probe is a 5-gene
3. The method of claim 2 wherein the multi-gene probe comprises
selected polymerase chain reaction generated fragments.
4. The method of claim 3 wherein each of the fragments is cloned
into a vector.
5. The method of claim 4 wherein each of the fragments is released
from the vector by enzymatic digestion and labeled with a different
6. The method of claim 5 wherein the labeled fragments are mixed
in predetermined concentrations.
7. The method of claim 6 wherein the labeled fragments and buffer
components are assembled as a easy-to-use assay kit.
8. A method of assessing breast cancer prognosis in a human subject,
comprising the steps of: (a) obtaining a sample of a thin section
of tissue containing tumor from the breast of the human subject;
and (b) analyzing the sample by fluorescence in situ hybridization
with a multi-gene probe comprising gene specific fragments each
labeled with a fluorescent dye mixed in predetermined concentrations
to simultaneously target the HER2, Topo II.alpha., NM23-H1, CK19
and MMP9 genes.
9. A method of assessing breast cancer prognosis in a human subject,
comprising the steps of: (a) generating DNA fragments unique to
HER, Topo II.alpha., NM23-H1, CK19 and MMP9; (b) isolating, purifying
and cloning the generated fragments into a cloning vector; (c) releasing
the fragments by enzymatic digestion and labeling each fragment
with a different fluorescent dye; (d) mixing the labeled fragments
in predetermined concentrations to form a multi-gene probe; (e)
obtaining a sample of a tumor from the breast of the human subject;
and (f) analyzing the sample with the multi-gene probe.
10. The method of claim 9 wherein the multi-gene probe is a 5-gene
11. The method of claim 10 wherein the multi-gene probe comprises
polymerase chain reaction fragments.
12. An assay kit for assessing breast cancer prognosis in a human
subject, comprising a package containing a multi-gene probe targeting
the HER2, Topo II.alpha., NM23-H1, CK19 and MMP9 genes.
13. The assay kit of claim 12 wherein the multi-gene probe is a
14. The assay kit of claim 13 wherein the multi-gene probe comprises
polymerase chain reaction fragments.
15. The assay kit of claim 14 wherein each of the fragments is
labeled with a fluorescent dye.
16. The assay kit of claim 15 wherein the labeled fragments are
mixed in predetermined concentration.
17. The assay kit of claim 12 further comprising pre- and post-hybridization
18. A multi-gene probe for assessing breast cancer prognosis in
a human subject, comprising: (a) a labeled fragment complementary
to at least a region of the HER2 gene sequence; (b) a labeled fragment
complementary to at least a region of the Topo II.alpha. gene sequence;
(c) a labeled fragment complementary to at least a region of the
NM23-H1 gene sequence; (d) a labeled fragment complementary to at
least a region of the CK19 gene sequence; and (e) a labeled fragment
complementary to at least a region of the MMP9 gene sequence; each
of the fragments being labeled with a different fluorescent label
and each of the fragments being useable to detect its complementary
gene copy number by hybridization.
19. The multi-gene probe of claim 18 wherein the multi-gene probe
is a 5-gene probe.
20. The multi-gene probe of claim 19 wherein the multi-gene probe
comprises polymerase chain reaction fragments.
21. The multi-gene probe of claim 20 wherein each of the fragments
is labeled with a fluorescent dye.
BACKGROUND OF THE INVENTION
 1. Field of the Invention
 The present invention relates to multiple diagnostic probes
for the assessment of multiple markers for breast cancer prognosis.
The invention also relates to assay kits containing multi-gene probes.
In addition, the invention relates to a method for the assessment
of multiple markers for breast cancer prognosis.
 2. Description of the Related Art
 Cancer is a disease that results when the controls that
regulate normal cell growth break down. The growth and development
of normal cells are subject to a multitude of different types of
control. A fully malignant cancer cell appears to have lost most,
if not all, of these controls. However, conditions that seem to
represent intermediate stages, when only some of the controls have
been disrupted, can be detected. Thus, the progression from a normal
cell to a malignant cell is a multistep process, each step corresponding
to the breakdown of a normal cellular control mechanism.
 Normal growth controls appear to become ineffective because
of mutations in the cellular genes coding for components of the
regulatory mechanism. Cancer can therefore result from the accumulation
of a series of specific mutations in the malignant cell.
 Oncogenes are genes whose expression causes cells to become
cancerous. The normal version of the gene (termed a proto-oncogene)
becomes mutated so that it is overactive. Because of their overactivity,
oncogenes are genetically dominant over proto-oncogenes, that is
only one copy of an oncogene is sufficient to cause a change in
the cell's behavior.
 The oncogene differs from the normal proto-oncogene in important
ways. The coding function of the gene may be unaltered but may be
transcribed at a higher rate or under different circumstances from
normal. This results in overproduction of a normal gene product.
Under other circumstances, there may be under-production of a normal
 Tumor suppressor genes act in a fundamentally different
way from oncogenes. Whereas proto-oncogenes are converted to oncogenes
by mutations that increase the genes' activity, tumor suppressor
genes become oncogenic as the result of mutations that eliminate
their normal activity. The normal, unmutated version of a tumor
suppressor gene acts to inhibit a normal cell from entering mitosis
and cell division. Removal of this negative control allows a cell
 Amplification, overexpression and/or underexpression of
some proto-oncogenes and tumor suppressor genes are useful clinically
for breast cancer prognosis. However, there is currently no single
strong independent marker that is useful for predicting disease
outcome in a majority of patients.
 The number of clinical laboratory assays currently used
in oncology is very small. For breast cancer, only three, namely,
estrogen, progesterone and HER2 status are currently assessed. Unfortunately,
these three only provide useful prognostic and predictive information
in a small number of patients. For example, although HER2 (a proto-oncogene)
has emerged as a strong independent prognostic and predictive marker
for breast cancer, it is only useful clinically in about 25 to 30%
of cases. In the last five years, the College of American Pathologists,
the American Society of Clinical Oncology expert panels and the
Joint Committee on Cancer have carefully considered many markers
proposed for managing breast cancer but have found none with proven
clinical utility sufficient to justify their adoption for routine
practice and no single marker has been found which gives a consistent
result in all manifestations.
 There are ongoing studies aimed at identifying new and broad-spectrum
markers that will be useful in many cases. But it is unlikely that
a single broad-range marker will be found since multiple biochemical
pathways are associated with the onset, progression and/or severity
of breast cancer. Moreover, the relevant pathways involved may be
different in different individuals due to other compounding factors
such as aging, race, nutrition, habit and environment.
 An object of the present invention is to provide a more
sensitive, relatively faster and cost effective approach for assessing
the status of prognostic markers in breast cancer. This and other
objects of the present invention will become more apparent from
a consideration of the following description and claims.
BRIEF SUMMARY OF THE INVENTION
 The present invention involves the use of a multiple-gene
diagnostic probe targeting the HER2, Topo II.alpha., NM23-H1, CK19
and MMP9 genes. The invention permits a simultaneous assessment
of at least five specific but independent changes in DNA number
that enable prognosis. All these changes do not necessarily occur
together always or in sequence in each patient. However, because
one detectable change (at least) will occur in greater than 95%
of cancers, the invention provides a one-step assessment of the
molecular status of breast tumors. The present innovation is more
sensitive and improves the likelihood of detecting molecular changes
of prognostic significance in a larger patient population at a relatively
 The basic elements of the multi-gene probe of the invention
are the following:
 i. labeled fragments complementary to multiple but unique
regions of the HER2 gene sequence;
 ii. labeled fragments complementary to multiple but unique
regions of the Topo II.alpha. gene sequence;
 iii. labeled fragments complementary to multiple but unique
regions of the NM23-H1 gene sequence;
 iv. a labeled fragment complementary to the CK19 gene sequence;
 v. labeled fragments complementary to at least a region
of the MMP9 gene sequence; each of the fragments being labeled with
a different label and each of the fragments being useable to detect
its complementary gene sequence by hybridization. The multi-gene
probe may be packaged in the form of an assay kit.
 The basic elements of the method of making the multi-gene
probe of the invention are the following:
 i. generation of gene-specific fragments corresponding to
the gene of interest;
 ii. labeling each of the fragments to form individual diagnostic
 iii. mixing the labeled fragments in predetermined concentrations
to form a multi-gene diagnostic probe.
 The multi-gene probe is used to screen a thin section of
a tumor specimen. The preferred analysis technique is fluorescence
in situ hybridization (FISH) assay. The resultant signal after imaging
has prognostic significance. Since this multi-gene probe targets
one or more independent alterations, it provides a one step and
highly efficient system for breast cancer prognosis.
 The present invention improves predication of disease outcome
in a larger population of breast cancer patients by enabling detection
of multiple changes at the molecular level that correlates with
prognosis, and significantly reduces the time and cost normally
required for assessment of prognostic markers in breast tumors.
The present invention permits the assessment of at least five independent
prognostic markers via a single step.
BRIEF DESCRIPTION OF THE DRAWINGS
 FIG. 1 shows the steps involved in the generation of gene
specific probes for fluorescence in situ hybridization analysis
of breast tumors. [i] Gene specific PCR primers are used to generate
DNA fragments (of different sizes) unique to HER2 (1100 pb), TopoII.alpha.
(2342 bp), NM23-H1 (1785 bp), CK19 (365 bp) and MMP9 (601 bp). [ii]
PCR fragments are purified by HPLC and cloned into specific restriction
sites of pUC19 cloning vectors. The gene specific fragments are
released after cloning using specific restriction enzymes and purified.
 FIG. 2 shows steps involved in cloning of PCR generated
fragments and labeling with fluorescent dyes.
 FIG. 3 shows the steps for preparation of a multi-gene probe
for fluorescence in situ hybridization analysis of breast tissue.
DETAILED DESCRIPTION OF THE INVENTION
 The present invention solves the problem of inadequate broad-range
prognostic factors for breast cancer by providing a multi-gene probe
for a single-step determination of disease outcome. The 5-genes
chosen are known to show altered expression in different breast
tumors. The altered expression of any one of them has a similar
prognostic significance, with respect to disease-free survival or
overall survival. The multi-gene probe is labeled and used to screen
a tumor specimen. The signal generated after the assay serves as
a prognostic marker. Since five prognostic markers are simultaneously
assessed, this strategy covers a wider variety of breast cancers
and so improves disease outcome prediction in a wider population.
 The five prognostic markers include HER2. HER2 is an acronym
for human epidermal growth factor receptor, also known as c-erbB-2/neu.
Growth factors are protein products of genes called proto-oncogenes,
which are fundamentally important for normal cells. The proto-oncogenes
interact with other genes and their products; these genes, called
tumor suppressor genes, also have important roles in normal cell
division. HER2 gene amplification and protein overexpression play
a pivotal role in oncogenic transformation, tumorigenesis and metastasis.
The HER2 gene (ERBB2) maps on chromosome 17q 21.1. The mRNA size
is 4.5 kb. The protein expressed by HER2 is a 185-kDa tyrosine kinase
receptor for heregulin and other members of the heregulin family.
 Topoisomerase II.alpha. (Topo II.alpha.) plays a key role
in DNA replication and is a target for multiple chemotherapeutic
agents. In breast cancer, Topo II.alpha. expression has been linked
to cell proliferation and HER2/neu protein overexpression. Topo
II.alpha. (170 kD) maps at chromosome 17q21-q22, and encodes a protein
that controls topological states of DNA.
 Nm23 is an acronym for nonmetastatic protein 23 or nucleoside
diphosphate (NPD) kinase-A (NDPKA). The underexpression of the NM23-H1
gene is related to cell proliferative activity. The NM23-H1 gene
maps to 17q22 and consists of 5 exons and 4 introns spanning 8.5
kb. The mRNA size is 0.8 kb. The NM23-H1 gene encodes a 17 KD protein.
 Cytokeratin 19 (CK19) is one of the families of genes for
keratins 13, 14, 15, 16, 17, and 19 contained in less than 150 kb
of genomic DNA in the region 17q21-q22. The mRNA size is 1.3 kb.
The gene expresses a 40-kda acidic keratin component of intermediate
filaments. CK19 protein is found on the surface of epithelial cells.
 Matrix metalloproteinase-9 (MMP9) is also known as 92-kD
gelatinase or Gelatinase B. It is a collagenase type IV-B (CLG4B).
In highly metastatic tumor cells, there may be conspicuous expression
of MMP9. The MMP9 gene (CLG4B) maps to 20 q11.2-q13.1. MMP9 has
13 exons and similar intron locations. The 13 exons of MMP9 are
3 more than have been found in other members of this gene family.
The extra exons encode the amino acids of the fibronectin-like domain,
which has been found only in MMP-2 and MMP-9. The mRNA size is 2.8
 The present invention preferably uses polymerase chain reaction
(PCR) to prepare gene-specific fragments. PCR allows an extremely
large number of copies to be synthesized of any given DNA sequence
provided that two oligonucleotide primers are available that hybridize
to the flanking sequences on the complementary DNA strands. The
reaction requires the target DNA, the two primers, all four deoxyribonucleoside
triphosphates, Mg.sup.2+ and a thermostable DNA polymerase or enzyme.
A PCR cycle is repeated for a set number of times depending on the
degree of amplification required.
 A PCR cycle consists of three steps:
 (1) Denaturation. The reaction mixture is heated to 95.degree.
C. for a short time period to denature the target DNA into single
strands that can act as templates for DNA synthesis.
 (2) Primer annealing. The mixture is rapidly cooled to a
defined temperature typically around 55.degree. C., which allows
the two primers to bind to the sequences on each of the two strands
flanking the target DNA region. This annealing temperature is calculated
to ensure that the primers bind only to the desired DNA sequences.
One primer binds to each strand. The two parental strands do not
reanneal with each other because the primers are in large excess
over parental DNA.
 (3) Elongation. The temperature of the mixture is raised
to 72.degree. C. (usually) and kept at this temperature for a pre-set
period of time to allow DNA polymerase to elongate each primer by
copying the single-stranded templates. Thus at the end of this incubation,
both single-stranded template strands have been made partially double
stranded. The new strand of each double-stranded DNA extends for
a variable distance downstream.
 The three steps of the PCR cycle are repeated. Thus in the
second cycle, the four strands denature, bind primers and are extended.
No other reactants need to be added. The three steps are repeated
once more for a third cycle and so on for a set number of additional
cycles. By the third cycle, some of the PCR products represent DNA
sequence only between the two primer sites and the sequence does
not extend beyond these sites. As more and more reaction cycles
are carried out, this type of double-stranded DNA molecule becomes
the majority species present.
 As shown in FIG. 1, sequence and gene specific primers are
used to generate DNA fragments (of different sizes) unique to HER2,
Topo II.alpha., NM23-H1, CK19 and MMP9 by PCR. The PCR generated
fragments are isolated from the PCR reaction and purified.
 The primers are designed to be complementary to the target
DNA such that they can be extended by the DNA polymerase towards
each other. Each one of a pair of PCR primers needs to be about
18-30 nt long and to have similar G+C content so they anneal to
their complementary sequences at similar temperatures. Since the
DNA gene sequences of the present invention are known, primer design
is straightforward and may be accomplished by techniques well known
in the art.
 In preparing the PCR fragments, the thermostable DNA polymerase
is TaqPlus Long PCR system (Stratagen Inc, La Jolla, Calif.).
 The accession number for the target DNA for HER2 is nm004448
and the DNA sequence is available at http://genome.ucsc.edu. The
primer sets for generation of DNA fragments for HER2 are illustrated
in FIG. 1 as HER-P1.1/1.2, HER-P2.1/2.2 and HER-P3.1/3.2 HER (P1.1,
2.1 and 3.1 are the forward primers for each region and P1.2, 2.2,
3.2 are the corresponding reverse primers, respectively). HER-P
1.1 is 5'-GCAGTGAGCACCATGGAGCT-3', HER-P 1.2 is 5'-TGCAAGCCTCAACTTCCTGG-3',
HER-P2.1 is 5'-CTCTTGGGACCTAGTCTCTG-3', HER-P2.2 is 5'-ACACTGTTAACCATGGTCCC,
HER-P3.1 is 5'-GGATTACAAGCGCCCGCTAATT-3' and HER-P3.2 is 5'-GAGGTTTCGCTCTGTCACCC-3'.
 The accession number for the target DNA for Topo II.alpha.
is nm001067 and the DNA sequence is available at http://genome.ucsc.edu.
The primer sets for generation of DNA fragments for Topo II.alpha.
are illustrated in FIG. 1 as Topo-P1.1/1.2, Topo-P2.1/2.2 and Topo-P3.1/3.2.
The primer for Topo-P1.1 is 5'-GAGTGATCTGCCCTCGTCAG-3', Topo-P1.2
is 5'-CCCACCTGTGGTTTACTTGT-3', Topo-P2.1 is 5'-GAATAGAATGTTTCCAGTAAGC-3',
Topo-P2.2 is 5'-CCTGGTTTCAAACCTTTAAA-3', Topo-P3.1 is 5'-ATTGAGGATACTTACGTTTG-3'
and Topo-3.2 is 5'-GAGACCAAGACTGGAGATTT-3'.
 The accession number for the target DNA for NM23-H1 is x73066
and the DNA sequence is available at http://genome.ucsc.edu. The
primer sets for generation of DNA fragments for NM23-H1 are illustrated
in FIG. 1 as NM23-P1.1/1.2, NM23-P2.1/2.2 and NM23-P3.1/3.2. The
primer for NM23-P1.1 is 5'-GGCTGCAGCCGGAGTTCAAA-3', NM23-P1.2 is
5'-CCCAGAATTCCCAACCCATT-3', NM23-P2.1 is 5'-CCGCTTGAGACGGATGACGCTGTA-3',
NM23-P2.2 is 5'-TCCCTTGCTTCCTGCCTCCA-3', NM23-P3.1 is 5'-ATAAAATTAGCCAAGTCTGG-3'
and NM23-P3.2 is 5'-TAATCTACCAGTTCCTCAGG-3'.
 The accession number of the target DNA for CK19 is u85961.1
and the DNA sequence is available at http://www.ncbi.nlm.nih.gov.
The primer set for generation of a DNA fragment for CK19 is illustrated
in FIG. 1 as CK19-P1.1/1.2. The primer for CK19-P1.1 is 5'-TCGAGGACCTGCGGGACAAGAT-3'
and CK19-P1.2 is 5'-ATCAGCTCGCACATCCGCCA-3'.
 The accession number of the target DNA for MMP9 is nm004994
and the DNA sequence is available at http://genome.ucsc.edu. The
primer sets for generation of DNA fragments for MMP9 are illustrated
in FIG. 1 as MMP9-P1.1/1.2, MMP9-P2.1/2.2, and MMP9-P3.1/3.2. The
primer for MMP9-P1.1 is 5'-AGACACCTCTGCCCTCACCA-3', MMP9-P1.2 is
5'-CCCATATCGCAGAGACTTCA-3', MMP9-P2.1 is 5'-AGCGGCCCTCGAAGATGAAG-3',
MMP9-P2.2 is 5'-GACCTGTTTCTTCAGAGCAC-3', MMP9-P3.1 is 5'-TGACTTCCCTTTCTTACCAG-3'
and MMP9-3.2 is 5'-CAAAGGTGAGAAGAGAGGGC-3'.
 After the PCR reaction is complete, the gene specific DNA
fragments are isolated from the reaction mixture and purified using
DNA spin columns (Qiagen Inc., Valencia, Calif.). The fragments
are then cloned into specific restriction sites (HER2=SmaI; Topo2=XmaI;
NM23H1=BamHI; CK19=PstI and MMP9=HindIII) of a cloning vector (e.g.,
pUC19). The recombinant vector is used to transfect bacteria (e.g.,
Escherichia coli Top 10) and after culturing the bacteria in a suitable
medium for approximately 24 hours the vector is isolated using plasmid
isolation kits (Qiagen, Inc., Valencia, Calif.) and the cloned inserts
generated by restriction digestion and purified (FIG. 1).
 Each of the separated and purified gene-specific fragments
is labeled with fluorescent dyes. Preferably, each of the PCR fragments
is labeled with a different fluorescent dye as follows:
1 Fluorescent Dye Fragment Type (Color) Preferred Fluorescent Dye
HER2 SpectrumOrange .TM. SpectrumOrange .TM. (Vysis Inc.) (Orange)
Topo II.alpha. SpectrumGreen .TM. (Green) SpectrumGreen .TM. (Vysis
Inc.) NM23-H1 Cy3 .TM. (Red) Cy3 .TM. (Amersham Biosciences) CK19
Cy5 .TM. (Blue) Cy5 .TM. (Amersham Biosciences) MMP9 Cy7 .TM. (Yellow)
Cy7 .TM. (Amersham Biosciences)
 A multi-gene diagnostic probe is prepared as illustrated
in FIG. 3 by mixing precise concentrations of each labeled fragment
2 Fragment General Concentration Range Preferred Concentration
HER2 10-25 ng 22 ng Topo II.alpha. 10-25 ng 18 ng NM23-H1 10-25
ng 20 ng CK19 10-25 ng 25 ng MMP9 10-25 ng 25 ng
 The multi-gene probe is used to screen a thin section of
a tumor specimen. The tumor specimen to be screened is first fixed
with, for example, formaldehyde, embedded in paraffin wax and then
cut into thin sections 4-5 .mu.m thick. The screening is preferably
accomplished by in situ hybridization. More specifically, it is
possible to incubate radioactive or fluorescent probes with sections
of tissues, wash away excess probe and then detect where the probe
has hybridized. The most preferred screening technique is by fluorescence
in situ hybridization (FISH) assay.
 The FISH assay broadly comprises de-paraffinization, denaturation
of the specimen DNA, preparation of the probe mixture, hybridization
of the specimen DNA and the probe mixture, and post-hybridization
washes. Standard protocols may be used to determine the optimum
denaturation time and temperature, typically 72.+-.1.degree. C.
for 5 minutes; hybridization time and temperature, typically 37.degree.
C. for 14-18 hours; post-hybridization wash time and temperature,
typically 72.+-.1.degree. C. for 2 minutes. The post-hybridization
wash buffer composition is typically 2.times.SSC/0.3% NP-40.
 The fluorescent fragments absorb light at an excitation
(EX) wavelength and then emit it at an emission (EM) wavelength.
After unbound probe is removed, the slide is air dried in the dark,
a counter stain (DAPI) is applied and the specimen is illuminated
at the excitation wavelength to enumerate the signal. The resultant
signal after imaging will indicate the status of the DNA targeted
by each probe. The DNA copy number has prognostic significance.
 Signal enumeration is carried out by imaging the hybridized
slides under a fluorescence microscope with filters appropriate
for each probe. Use a 25.times. or 40.times. objective for an initial
scan of the entire tissue section hybridized. Select an area of
good nuclei distribution and switch to a 63.times. or 100.times.
objective for enumeration of 60 nuclei to determine DNA copy number.
Presently, the HER2 DNA copy number is best characterized in relation
to the prognostic significance. For example, HER 2 gene copy number
is determined with a HER2 specific probe and a chromosome 17-enumeration
probe (Vysis Inc.). The ratio between the HER2 and chromosome-7
copy number has a prognostic significance. Ratios above 2.0 are
typically associated with poor prognosis, characterized by shorter
disease-free and overall survival in node positive patients.
 The multi-gene probe may be packaged in the form of an assay
kit. The kit would typically include the fluorescent-labeled multi-gene
probe, control slides with human cell line specimens having different
levels of HER2 gene amplification, a counter stain such as DAPI,
NP-40, 20.times.SSC solution, 4% formalin in PBS, NaOH, protease/buffer,
and microcentrifuge tubes.
 While the foregoing invention has been described in some
detail for purposes of clarity and understanding, it will be appreciated
by one skilled in the art from a reading of this disclosure that
various changes in form and detail can be made without departing
from the true scope of the invention and appended claims.