CA2394921A1 - Supracolonic aerodigestive neoplasm detection - Google Patents

Abstract

The invention provides methods and materials for detecting supracolonic aerodigestive premalignant and malignant neoplasms. Specifically, the invention provides methods and materials for determining whether a stool sample from a mammal contains a neoplasm-specific marker from a neoplasm located in the supracolonic aerodigestive tissue of a mammal. These neoplasm - specific markers are selected from the group consisting of nucleic acids having a point mutation (K-ras, APC, p53), nucleic acids that reflect microsatellite instability (BAT-26), and high integrity nucleic acids (DNA longer than 200 bp).

Description

SUPRACOLONIC AERODIGESTIVE NEOPLASM DETECTION
Statement as to Federally Sponsored Research Funding for the work described herein may have been provided by the federal government, which may have certain rights in the invention.
Claim to Priority s , This application claims priority to U.S. Provisional Application Serial numbers 60/169,457, filed on December 7, 1999 and 60/196,074, filed on April 10, 2000.
FIELD OF THE INVENTION
The invention relates to methods and materials involved in the detection of supracolonic aerodigestive premalignant and malignant neoplasms.
o BACKGROUND
About half of all cancer deaths in the United States result from aerodigestive cancer. For example, of the estimated 564,800 annual cancer deaths, 160,100 (25%) result from lung cancer; 56,500 (10%) result from colorectal cancer;
28,900 (6%) result from pancreas cancer; 13,700 (3%) result from stomach cancer; and is 11,900 (3%) result from esophagus cancer. In addition, over 7 percent of the annual cancer deaths result from other aerodigestive cancers such as naso-oro-pharyngeal, bile duct, gall bladder, and small bowel cancers (Landis ef al., CA Cancer J.
Clin., 48:6-29 (1998)).
Attempts have been made to identify and use nucleic acid markers that are 2o indicative of cancer. For example, mutations in the p53 cell cycle regulator gene have been associated with numerous cancers, especially colorectal cancer, and it has been suggested that specific mutations might be a basis for molecular screening assays for the early stages of certain types of cancer. See, e.g., Sidransky, et al., Science, 256: 102-105 (1992).

The invention involves detecting premalignant and malignant supracolonic aerodigestive neoplasms. According to the invention, a supracolonic aerodigestive neoplasm is a neoplasm in an aerodigestive tissue proximal to (above) the colon.

An aerodigestive tissue is a tissue characterized by a lumenal space that is connected to the lumenal spaces of the respiratory and digestive tracts.
Supracolonic aerodigestive tissue includes tissue such as a mammal's small intestine, gall bladder, bile duct, pancreas, liver, stomach, esophagus, lung, and s naso-oro-pharyngeal airways. Supracolonic aerodigestive tissue does not include tissue such as blood, serum, bone, connective tissue or other tissue that is not directly connected to a lumen of the aerodigestive tract.
The invention involves determining whether a stool sample from a mammal contains a neoplasm-specific marker from a supracolonic aerodigestive neoplasm.
~o The detection of a neoplasm-specific marker in a mammal's stool allows a physician to screen for a supracolonic aerodigestive neoplasm much earlier than currently available cancer detection techniques. In addition, the analysis of a stool sample is much less invasive than other types of diagnostic techniques such as endoscopy.
The invention is based on the discovery that neoplasm-specific markers from ~s a neoplasm located in an aerodigestive tissue proximal to the colon (e.g., in the small intestine, gall bladder, bile duct, pancreas, liver, stomach, esophagus, lung, and naso-oro-pharyngeal airways) can be detected in that mammal's stool. Thus, stool can be analyzed to identify mammals having cancer other than colorectal cancer. Once a particular patient is determined to have stool containing a 2o neoplasm-specific marker, additional cancer screening techniques can be used to identify the exact location and nature of the neoplasm. For example, a stool sample can be analyzed to determine that the patient has a neoplasm, while magnetic resonance imaging (MRI), endoscopic analysis (e.g., colonscopy, gastroscopy, and bronchoscopy), and tissue biopsy techniques can be used to identify the exact 2s location and nature of the neoplasm in the supracolonic aerodigestive tract. Thus, the invention provides convenient methods that can be used to screen and identify patients having a supracolonic aerodigestive neoplasm.
In general, one aspect of the invention features a method for detecting a lung neoplasm in a mammal, preferably a human. The method includes determining 3o whether a stool sample from the mammal contains a neoplasm-specific marker associated with lung cancer. The lung neoplasm can include a premalignant neoplasm or malignant neoplasm. The neoplasm-specific marker can include a neoplasm-specific nucleic acid marker. The neoplasm-specific nucleic acid marker can include a nucleic acid having a point mutation. The point mutation can be located in a K-ras, APC, or p53 gene. The neoplasm-specific nucleic acid marker s can include nucleic acid that reflects microsatellite instability. The microsatellite instability can be located in the BAT-26 segment of the MSH2 mismatch repair gene.
The neoplasm-specific nucleic acid marker can include long DNA (e.g., DNA
greater than about 270, 300, 400, 500, 600, 700, 800, 900, 1000, 1500, 2000, or 2500 base pairs in length). The neoplasm-specific marker can include a neoplasm-specific to polypeptide marker or a neoplasm-specific cell marker. A method of the invention can include determining whether the stool sample contains two or more neoplasm-specific markers. The two or more neoplasm-specific markers can be nucleic acid markers, polypeptide markers, and/or cell markers. For example, the two or more neoplasm-specific markers can be neoplasm-specific nucleic acid markers, and the is two or more neoplasm-specific nucleic acid markers can include nucleic acid having a point mutation, nucleic acid that reflects microsatellite instability, and/or long DNA.
In another aspect, the invention features a method for detecting a naso-oro-pharyngeal neoplasm in a mammal. The method includes determining whether a stool sample from the mammal contains a neoplasm-specific marker, where the 2o marker is from the naso-oro-pharyngeal neoplasm.
Another aspect of the invention features a method for detecting an esophageal neoplasm in a mammal. The method includes determining whether a stool sample from the mammal contains a neoplasm-specific marker, where the marker is from the esophageal neoplasm.
2s Another aspect of the invention features a method for detecting a stomach neoplasm in a mammal. The method includes determining whether a stool sample from the mammal contains a neoplasm-specific marker, where the marker is from the stomach neoplasm.
Another aspect of the invention features a method for detecting a liver 3o neoplasm in a mammal. The method includes determining whether a stool sample from the mammal contains a neoplasm-specific mavker, where the marker is from the liver neoplasm.
Another aspect of the invention features a method for detecting a bile duct neoplasm in a mammal. The method includes determining whether a stool sample from the mammal contains a neoplasm-specific marker, where the marker is from s the bile duct neoplasm.
Another aspect of the invention features a method for detecting a gall bladder neoplasm in a mammal. The method includes determining whether a stool sample from the mammal contains a neoplasm-specific marker, where the marker is from the gall bladder neoplasm.
io Another aspect of the invention features a method for detecting a small intestine neoplasm (e.g., duodenum, jejunum, and/or ileum neoplasm) in a mammal.
The method includes determining whether a stool sample from the mammal contains a neoplasm-specific nucleic acid marker, where the marker is from the small intestine neoplasm.
is Another aspect of the invention features a method for detecting a pancreatic neoplasm in a mammal. The method includes determining whether a stool sample from the mammal contains long DNA, where the tong DNA is from the pancreatic neoplasm. Another aspect of the invention features a method for detecting a pancreas neoplasm in a mammal, including determining whether a stool sample 2o from the mammal contains two or more neoplasm-specific markers, where the markers are from the pancreas neoplasm.
In another embodiment of the invention, the nucleic acid being analyzed is selected from a coding region of a gene, or portion thereof, a noncoding nucleic acid region, or portion thereof, a regulatory element of a gene or a portion thereof, and an 2s unidentified fragment of genomic DNA.
Methods of the invention are useful as diagnostic screening methods. Often it is desirable to perform follow-up testing on a patient in order to confirm a suspected disease location in the aerodigestive tract. Such follow-up procedures are determined based upon the disease state being interrogated. For example, a 3o cotonoscopy, gastroscopy, or bronchoscopy may be suggested in a case in which a _5_ stool sample is positively screened according to methods of the invention.
Such follow-up procedures are contemplated herein as part of the invention.
Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to s which this invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present ~o specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting.
Other objects and advantages of the invention are apparent upon consideration of the following detailed description, drawings and claims.
Description of the Drawings is Figure 1 is a gel photograph showing results of amplification of K-ras (exon 1 ) DNA isolated from stool using forward and reverse primers spaced about 200 by apart. The band intensity relates to the amount of 200 by product or greater in the sample. Lanes 1-4 are results from patients with cancer or adenoma, lane 5 is a positive control, lanes 6-10 are from patients who did not have cancer or adenoma, 20 lanes 11-12 are negative controls, and lanes 13-18 are standards at the approximate molecular weight indicated in the figure.
Figures 2-4 are gel photographs showing results of amplification of apc (exon 15) DNA isolated from stool using forward and reverse primers spaced about 200 by apart. The band intensity relates to the amount of 200 by product or greater in the 2s sample. Lanes 1-4 are results from patients with cancer or adenoma, lane 5 is a positive control, lanes 6-10 are from patients who did not have cancer or adenoma, lanes 11-12 are negative controls, and lanes 13-18 are standards at the approximate molecular weight indicated in the figure.
Figure 5 is a gel photograph showing results of amplification of p53 (exon 5) 3o DNA isolated from stool using forward and reverse primers spaced about 200 by apart. The band intensity relates to the amount of 200 by product or greater in the sample. Lanes 1-4 are results from patients with cancer or adenoma, lane 5 is a positive control, lanes 6-10 are from patients who did not have cancer or adenoma, lanes 11-12 are negative controls, and lanes 13-18 are standards at the approximate s molecular weight indicated in the figure.
Figure 6 is a gel photograph showing results of amplification of p53 (exon 7) DNA isolated from stool using forward and reverse primers spaced about 200 by apart. The band intensity relates to the amount of 200 by product or greater in the sample. Lanes 1-4 are results from patients with cancer or adenoma, lane 5 is a io positive control, lanes 6-10 are from patients who did not have cancer or adenoma, lanes 11-12 are negative controls, and lanes 13-18 are standards at the approximate molecular weight indicated in the figure.
Figure 7 is a gel photograph showing results of amplification of p53 (exon 8) DNA isolated from stool using forward and reverse primers spaced about 200 by is apart. The band intensity relates to the amount of 200 by product or greater in the sample. Lanes 1-4 are results from patients with cancer or adenoma, lane 5 is a positive control, lanes 6-10 are from patients who did not have cancer or adenoma, lanes 11-12 are negative controls, and lanes 13-18 are standards at the approximate molecular weight indicated in the figure.
2o Figure 9-10 are get photographs of results of amplification of DNA from stool samples using forward and reverse primers spaced approximately 1.8 Kb apart.
The band intensity shows the amount of 1.8 Kb or greater product. Lanes 1, 8, and 9 are negative controls, lanes 2, 3, and 5 are results from patients with cancer or adenoma, lanes 4, 6, and 7 are results from patients who did not have cancer or 2s adenoma, and lanes 10-14 are molecular weight standards.
Figures 11 A and B are gel photographs of results of amplification of DNA in stool from a total of 30 patients and controls. The band intensity relates to the amount of amplifiable DNA in the sample. Lanes N are negative controls, lanes 1, 3, 11, and 18 are results from patients which are indicative of the presence of cancer or 3o adenoma, lanes 2, 4, 5-10, 12-17, and 19-30 are results from patients which are indicative of the absence c~f cancer or adenoma. The remaining lanes are markers or standards.
The amplification reactions described above may be conducted according to any suitable or convenient protocol and the fragment size of the resulting s amplification products (if any) may be determined by any suitable or convenient means.
DETAILED DESCRIPTION
The invention provides methods and materials related to the detection of neoplasm-specific markers from the aerodigestive tract in a stool sample.
to Specifically, the invention provides methods and materials for identifying mammals having a supracolonic aerodigestive neoplasm by detecting a neoplasm-specific marker in a stool sample obtained from the mammal. For example, the investion provides methods for detecting a neoplasm in the small intestine, gall bladder, bile duct, pancreas, liver, stomach, esophagus, lung, or naso-oro-pharyngeal neoplasm ~s of a mammal. A small intestine neoplasm can be a duodenum, jejunum, or ileum neoplasm. It will be appreciated that the methods and materials of the invention can be used to detect a neoplasm-specific marker in a mammal having a combination of different supracolonic aerodigestive neoplasms. For example, the methods and materials of the invention can be used to detect a neoplasm-specific marker in a 2o human having a lung and stomach neoplasm. The term "neoplasm" as used herein refers to any new and abnormal growth of tissue. Thus, a neoplasm can be a premalignant neoplasm or a malignant neoplasm. The term "neoplasm-specific marker" refers to any biological material that can be used to indicate the presence of a neoplasm. Examples of biological materials include, without limitation, nucleic 2s acids, polypeptides, carbohydrates, fatty acids, cellular components (e.g., cell membranes and mitochondria), and whole cells.
While not being limited to any particular mode of action, the invention appears to be based on the fact that premalignant and malignant neoplasms arising in a mammal's small intestine, gull bladder, bile duct, pancreas, liver, stomach, 3o esophagus, lung, ornaso-oro- naso-oro-pharyngeal airways can shed cells into the _$_ aerodigestive lumen. These exfoliated cells as well as their constituents can survive transit through the gastrointestinal tract and ultimately pass as fecal waste.
For example, as described herein, a neoplasm-specific marker can be detected in a stool sample collected from a human having lung cancer. In this case, cancer cells s and their constituents leave the lung, enter the digestive tract, and exit the body as fecal waste.
Nucleic acrd markers Neoplasm-specific markers can be nucleic acid. Examples of neoplasm-~o specific nucleic acid markers include, without limitation, nucleic acid having a point mutation, nucleic acid that reflects microsatellite instability, and long DNA.
Nucleic acid having a point mutation can encode a polypeptide or regulate the expression of a polypeptide (e.g., promoters, enhancers, and silencers). Examples of nucleic acid that can contain a point mutation indicative of a neoplasm include, without limitation, is the genes for K-ras, APC (adenomatous polyposis coli), and p53.
Nucleic acid that reflects microsatellite instability can be used to indicate the presence of a neoplasm. Briefly, nucleic acid that reflects microsatellite instability can be identified as described elsewhere (Samowitz et al., Am. J. Path., 154:1637-1641 (1999) and Hoang et al., Cancer Res., 57:300-303 (1997)). An example of 2o nucleic acid that can reflect microsatellite instability indicative of a neoplasm includes, without limitation, the gene for BAT-26.
While each type of supracolonic aerodigestive neoplasm (e.g., lung, stomach, etc.) is associated with some DNA alterations unique to the site, many of the mutations commonly present involve the same genes as those found in colorectal 2s neoplasia - especially with respect to mutations on K-ras, APC, and p53 as well as to microsatellite instability (Table I).

_g_ Table I. Proportion of different gene mutations found in different supracolonic aerodigestivetissue neoplasms.
K-ras APC P53 MSI
Lung >30% 30-80% >50% 30-66%
s Esophagus Low >30% >50% >50%
Stomach Low 50-80% >50% >30%
Bile Duct 17-60% 20-67%
Long DNA is a marker for non-apoptotic cells. Typically, cells shed from io normal mucosa are apoptotic, while those shed from colorectal and supracolonic aerodigestive neoplasms are non-apoptotic. As described herein, long DNA can be used as a neoplasm-specific marker for patients having a supracolonic aerodigestive neoplasm such as a small intestine, gall bladder, bile duct, pancreas, liver, stomach, esophagus, lung, or naso-oro-pharyngeal neoplasm. One hallmark of apoptosis is is the autodigestion or cleavage of DNA into "short" fragments of about 180 base-pairs.
The detection of "long" DNA (i.e., DNA greater than about 200 base-pairs) in a stool sample can indicate the presence of non-apoptotic cells of neoplastic lineage derived from a supracolonic aerodigestive neoplasm. The term "long DNA" as used herein refers to DNA greater than about 200 base-pairs (e.g., greater than about 20 250, 300, 350, 400, 500, 600, 700, 800, 900, 1000, 1250, 1500, 1750, 2000, or 2500 base-pairs).
Any method can be used to detect a neoplasm-specific nucleic acid marker in a stool sample. For example, once a stool sample is collected and the mammal's nucleic acid isolated, PCR can be used to detect the presence or absence of 2s particular nucleic acid markers such as a nucleic acid having a particular point mutation, a nucleic acid that reflects microsatellite instability, and long DNA. It is noted that a single stool sample can be analyzed for one neoplasm-specific marker or for multiple neoplasm-specific markers. For example, a stool sample can be analyzed using assays that detect a panel of different neoplasm-specific markers. In 3o addition, multiple stool samples can be collected for a single mammal and analyzed as described herein. U.S. Patents 5,670,325; 5,741,650; 5,928,870; 5,952,178;
and 6,020,137 describe various methods that can be used to prepare and analyze stool samples.
Polypeptide markers Neoplasm-specific markers can be polypeptides. Examples of neoplasm-s specific polypeptide markers include, without limitation, oncogenic polypeptides and mutated polypeptides. Examples of polypeptides that can be indicative of a neoplasm include, without limitation, K-ras, APC, and p53. Any method can be used to detect a neoplasm-specific polypeptide marker. For example, antibodies specific for the polypeptide marker can be used in an immunoassay (e.g., ELISA) to detect ~o the presence or absence of the polypeptide in a stool sample that is indicative of the presence of an aerodigestive neoplasm.
Cell and cell component markers Neoplasm-specific markers can be cells or cell components (i.e., cell markers). Examples of neoplasm-specific cell or cell component markers include, is without limitation, tumor cells and tumor cell components (e.g., cell membranes).
U.S. Patent 5,891,651 describes methods and materials that can be used to detect neoplasm-specific cell or cell component markers in stool samples.
The invention will be further described in the following examples, which do not limit the scope of the invention described in the claims.
?o EXAMPLES
EXAMPLE 1. The three component test.
The three component test can detect three different types of neoplasm-specific nucleic acid markers from supracolonic aerodigestive neoplasm: (1 ) nucleic acid having a point mutation, (2) nucleic acid that reflects microsatellite instability, zs and (3) long DNA. Briefly, stool samples were thawed at room temperature and homogenized in an excess volume (>1:10 w:v) of EXACT buffer A (EXACT
Laboratories, Maynard, MA) utilizing an EXACTOR stool shaker (EXACT
Laboratories Maynard, MA). Following homogenization, a four gram stool equivalent of each sample was centrifuged to remove all particulate matter, and the 3o supernatants incubated at 37°C following addition of Proteinase K
(0.5 Ng/pL) and SDS (0.5%). The supernatants were subsequently extracted with Tris saturated phenol (Gibco/BRL, Grand island, NY), phenol/chloroform/isoamyl alcohol (25:24:1 ), and chloroform. Total nucleic acid was then precipitated (1/10 volume 3M NaAc and an equal volume isopropanol), removed from solution by centrifugation, and s resuspended in TE (0.01 M Tris pH 7.4, 0.001 M EDTA) buffer containing RNase A
(2.5 Ng/mL). For each group of samples prepared, process positive control samples as well as component negative controls were included.
Sequence specific DNA fragments were purified from the total nucleic acid preparations by performing oligonucleotide-based hybrid capture. For each sample, to seven hybrid capture reactions were performed in duplicate. Each capture reaction was carried out by adding 300 pL of sample preparation to an equal volume of Guanidine Isothiocyanate solution (Gibco/BRL, Grand Island, NY)) containing biotinylated sequence specific oligonucleotides (20 pmoles) (Midland Certified Reagent Co., Midland, TX). The sequence of each oligonucleotide was specific for is the DNA fragment to be analyzed. For example, an oligonucleotide having specificity for a region of the K-ras gene was used to capture a fragment that could contain the K-ras mutations. Following a two-hour incubation at 25°C, strepavidin coated magnetic beads were added to the solution, and the tubes were incubated for an additional hour at room temperature. The bead/hybrid capture complexes were 2o then washed four times with 1X B+W buffer (1 M NaCI, .01 M Tris-HCI pH 7.2, 0.001 M EDTA 0.1 % Tween 20), and the sequence specific captured DNA was eluted into 35 NL L-TE (1 mM Tris pH 7.4, 0.1 M EDTA) by heat denaturation.
PCR amplifications (50 NL) were performed on MJ Research Tetrad Cyclers (Watertown, MA) using 10 NL of captured DNA, 1X GeneAmp PCR buffer (PE
2s Biosystems, Foster City, CA), 0.2mM dNTPs (Promega, Madison, WI), 0.5 NM
sequence specific primers (Midland Certified Reagent Co., Midland, TX) and 5 units Amplitaq DNA polymerase (PE Applied Biosystems, Norwalk, CT). All of the sequence specific amplification reactions were carried out under identical thermocycler conditions. Following an initial denaturation of 94°C for 5 min, PCR
3o amplification was performed for 40 cycles con~istincr of 1 min at 94°C, 1 min at 60°C, and 1 min at 72°C, with a final extension of 5 min at 72°C. For PCR product analysis, 8 NL of each amplification reaction was loaded and electrophoresed on a 4% ethidium bromide- stained NuSieve 3:1 agarose gels (FMC, Rockland, ME) and visualized with a Stratagene EagIeEye II (Stratagene, La Jolla, CA) still image system.
s The presence or absence of point mutations or BAT-26 associated mutations was determined by using a modified solid phase minisequencing method (Syvanen et al., Genomics, 8:684-692 (1990)). Point mutation targets included codons K12p1, K12p2, and K13p2 of the K-ras gene; codons 1309 delta 5, 1367p1, 1378p1, and 1450p1 of the APC gene; and codons 175p2, 245p1, 245p2, 248p1, 248p2, 273p1, ~0 273p2, and 282p1 of the p53 gene. For all gene targets, both wild-type and mutant specific reactions were performed. Within the wild-type reactions, radionucleotide bases complementary to the wild-type base were added. For each point mutation specific reaction, radionucleotide bases complementary to the expected mutant bases were added in addition to unlabeled dideoxy nucleotides complementary to is the wild-type base. BAT-26 mutations associated with a 4-15 by deletion were identified by size discrimination of reaction products.
The presence of long DNA was determined by analyzing the relative intensity of each sample specific PCR product. For each stool sample analyzed, 7 unique PCR amplification products were generated in duplicate (or 14 amplifications per 2o subject) and independently scored by two technicians. PCR product intensities were scored as high, medium, or low by visual examination of the gel image (Grades A, B, and C, respectively).
Examples 2-4 Experiments were conducted to determine whether the presence of long DNA
2s in stool were predictive of supracolonic aerodigestive neoplasm in patients from whom stools samples were obtained. In the first experiment (Example 2), the amount of amplifiable DNA was measured in each of several stool samples using PCR amplification to detect DNA fragments in the sample of at least 200 base pairs in length. The second experiment (Example 3) determined the amount of long 3o fragments (greater than 200 base pair) in the same samples, and then determined ratios of long product to short product. The third experiment (Example 4) determined a profile of amplification products with nucleic acid fragment lengths of 200 bp, 400 bp, 800 bp, 1.3 Kb, 1.8 Kb and 2.4 Kb.
The size of human DNA fragments obtained above can be determined by numerous means. For example, human DNA can be separated using gel s electrophoresis. A 3% agarose gel is prepared using techniques known in the art.
See Ausubel et. al., Short Protocols in Molecular Biology, John Wiley & Sones, 1195, pgs. 2-23-2-24, incorporated by reference herein. The size of human DNA
fragments is then determined by comparison to known standards. Fragments greater than about 200 by provide a positive screen.
to EXAMPLE 2 Stool samples were collected from 9 patients who presented with symptoms or a medical history that indicated that a colonoscopy should be performed.
Each stool sample was frozen. Immediately after providing a stool sample, each patient was given a colonoscopy in order to determine the patient's disease status.
Based ~s upon the colonoscopy results, and subsequent histological analysis of biopsy samples taken during colonoscopy, individuals were placed into one of two groups:
normal or abnormal. The abnormal group consisted of patients with colorectal cancer or with an adenoma of at least 1 cm in diameter. Based upon these results, 4 of the 9 patients were placed into the abnormal group.
2o The samples were screened by determining the amount of amplifiable DNA
having at least 200 base pairs.
Human DNA was isolated and amplified using PCR. Each sample was amplified using forward and reverse primers through 7 loci (Kras, exon 1, APC
exon 15 (3 separate loci), p53, exon 5, p53, exon 7, and p53, exon 8) in duplicate (for a 2s total of 14 amplifications for each locus). Seven separate PCRs (40 cycles each) were run in duplicate using primers directed to detect fragments in the sample having 200 base pairs or more. Amplified DNA was placed on a 4% Nusieve (FMC
Biochemical) gel (3% Nusieve, 1 % agarose), and stained with ethidium bromide (0.5 ~.g/ml). The resulting amplified DNA was graded based upon the relative intensity of 3o the stained gels. The results are shown in Figures 1-7. Each Figure represents the results for all 9 patients (including standards) for the seven different loci that were amplified. As shown in the Figures, each sample from a patient with colorectal cancer or adenoma was detected as a band having significantly greater intensity than the bands associated with samples from patients who did not have colorectal s cancer or precancer. All four colorectal cancer/adenoma patients identified using colonoscopy were correctly identified by determining the amount of amplifiable DNA
200 base pairs or greater in length. As shown in Figures 1-7, the results were the same regardless of which locus was amplified. Accordingly, the amount of 200 by or greater DNA in a sample was predictive of patient disease status.
~o EXAMPLE 3 An experiment was conducted that was essentially identical to the one described above in Example 2, but forward and reverse primers were placed such that fragments of about 1.8 Kb and above were amplified.
Forward and reverse primers were spaced so as to hybridize approximately ~s 1.8 Kb apart on three different loci (Kras, exon 1, APC, exon 15, and p53 exon 5).
Thirty-three rounds of amplification were performed, and the resulting DNA was placed on a 3% agarose gel. The results are shown in Figures 8-10. As shown in the Figures (which show results from three separate experiments to amplify and detect "long" product), samples from individuals having colorectal cancer or 2o precancer produced large amounts of long (in this case 1.8 Kb and above) DNA;
whereas samples from patients who did not have cancer or precancer produced no DNA in the range of about 1.8 Kb and higher. Thus, the presence of long DNA
was indicative of the disease status of the patient.

2s An experiment was conducted to determine the molecular weight profile of DNA from samples collected and prepared as part of a blind study on 30 patients who presented at the Mayo Clinic with suspected gastrointestinal disorders.
Stool samples were obtained, and DNA was isolated as described above.
According to methods of the invention, amplification reactions were 3o conducted using forward and reverse primers through the 5 loci for each sample.

Forward and reverse primers were spaced to amplify fragments of 200 bp, 400 bp, 800 bp, 1.3 Kb, 1.8 Kb, and 2.4 ~Cb. Each of 30 PCR reactions was run for 36 cycles. Amplicon was run on a 3% Seakeam gel, and stained with ethidium bromide. The results are shown in Figures 11A and 11 B. Each figure represents s the results for 15 of the 30 patients.
As shown in those figures, patients with cancer or adenoma have an increased yield of long DNA. That is especially true at the 1.8 Kb level and above.
Thus, patients with cancer or adenoma produce larger DNA fragments than are produced in the stool of patients who do not have cancer. Thus, the presence of io high molecular weight DNA, especially that at 1.8 Kb and above, were indicative of the presence of cancer.

In this example, methods of the invention were correlated with clinical outcome in numerous patients who had a colorectal adenoma or colorectal cancer ~s as diagnosed using colonoscopy, and 79 patients who were diagnosed as not having colorectal cancer or adenoma. A stool sample was obtained from each of these patients and prepared as described above. Fragments of the 5 different loci referred to above were amplified using primers spaced 200, 400, 800, 1300, 1800, and 2400 base pairs apart using the protocol described above in Example 4.
Each 2o amplification was scored such that successful amplification of a fragment received a score of 1, and no amplification received a score of 0. Since five loci were interrogated using 6 primer pairs each, the maximum score was 30 (successful amplification of all 6 fragments at all five loci). The cutoff for a positive screen was set at 21. The results are shown below.

Table 1 Normals Table 2 Patient A a Score Aden omas No.

P-178 64 19 Patient A Score No. a Table 3 Carcinomas Patient A Score No. a As shown above, methods of the invention are effective in screening for the presence of colorectal cancer and adenoma.
EXAMPLE 6. Neoplasm detection in humans.
Stool samples were analyzed using the long DNA component of the three component test described in Example 1. Briefly, a single freezer-archived stool sample was assayed in blinded fashion from each of 25 patients with proven supracolonic aerodigestive cancer, 19 patients with colorectal cancer, and 20 colonoscopically-normal controls without history of neoplasia. Human DNA was isolated from stool by sequence-specific hybrid capture, and selected primers were used to amplify long DNA of 1800-2400 by on each of 5 gene loci (apoptotic DNA
consists of short fragment lengths of 180-200 by and would not be included in this assay). PCR product intensities were determined by UV transilluminator photo-imaging of ethidium bromide stained gels.
In a logistic regression model, long DNA proved to be a discriminating marker for all aerodigestive cancers with an area of 0.83 under the ROC curve. At a specificity of 96% (95% C1:78-99%): sensitivity for all aerodigestive cancers was 77% (95% C1:62-89%); sensitivity for supracolonic aerodigestive cancers was 76%
(95% C1:55-91%) -- lung 7/8 (88%), esophageal 2/3 (67%), gastroduodenal 1/4 (25%), pancreatic 6/7 (86%), and biliary 3/3 (100%); and sensitivity for colorectal cancers was 79% (95% C1:54-94%). For supracolonic aerodigestive cancers, 5/8 (63%) stage I-II and 12/15 (80%) stage III-IV lesions were detected; staging unknown in 2. For colorectal cancers, 7/10 (70%) stage I-II and 8/9 (89%) stage III-IV lesions were detected.
These observations indicate that supracolonic aerodigestive neoplasms at any aerodigestive tissue site can be detected using DNA markers to analyze stool.
The high yield by long DNA also indicates that non-apoptotic exfoliation is a hallmark of most aerodigestive cancers. Larger clinical studies targeting neoplasm-specific nucleic acid markers in stool can be used in this noninvasive screening approach to detect supracolonic aerodigestive neoplasms.
EXAMPLE 7. A blind study.
Stool samples are collected from 100 cases with known primary aerodigestive cancers located proximal to the colon (20 lung, 20 esophageal, 20 stomach, 20 pancreas, and 20 bile duct) and from 50 controls (10 colorectal cancers with positive three component test results and 40 healthy colonoscopy-negative patients from a parallel study). Stool samples are assayed in blinded fashion using the three component test described in Example 1. In those cases from which adequate tissue is available from the primary tumor, tissue DNA is also assayed in blinded fashion using the three component test described in Example 1.
Human subjects are instructed to collect a single whole stool using a plastic bucket-type container that mounts on the toilet seat and that is sealed with an airtight lid. Stools are sent to the laboratory so that less than 12 hours elapse between collection and receipt. Upon receipt, stools are bisected and promptly frozen at -80°C. In those instances where either frozen or formalin fixed tissue from the primary tumor is available, tumor DNA is extracted using standard techniques.
Tumor DNA samples are labeled differently than stool samples from corresponding patients so that a blind is maintained.
The three component test described in Example 1 is used as follows. Briefly, DNA is recovered from a fecal aliquot less than six grams by solvent extraction (which yields human DNA in essentially 100% of instances). Using quantitative PCR
methods: 23 high-frequency point mutations are targeted on K-ras, APC, and p53 genes; BAT-26, a marker for microsatellite instability is assayed; and selected primers are used to detect "long DNA" which includes fragment lengths greater than 270 base pairs and which can reflect non-apoptotic DNA. For each component, results are expressed as percent altered:wild-type. Thus, quantitative distributions are plotted for each assay component for each tumor site group and compared to the control distribution. Results are also reported as a discrete positive or negative value if a cut-off level is chosen; results have generally been considered positive for colorectal neoplasia detection if any DNA alteration was detected at a ratio greater than 1 % on replicate testing.
Using the extra (discard) blood drawn for routine laboratory testing, 5-10 mL
is retrieved. Serum and buffy-coat is prepared from each sample and stored for future analysis. DNA recovered from these sera is assayed using the three component test, and the buffy coats are evaluated for the presence of tumor cells using an RT-PCR technique.

Frequency distributions of the three component test results (for each compound and for combined components) are tabulated for each tumor site group and for all groups in aggregate. A sample size of 20 in each tumor site group in this study will yield 95% confidence intervals of 6-44%, 12-54%, and 27-73% if the observed detection rates are 20, 30, and 50%, respectively.
A chance-corrected measure of agreement (kappa statistic, and 95%
confidence intervals) is computed to estimate the concordance in results between corresponding stool and tissues for those cases in whom both stool and tissue were obtained. This estimate is calculated overall and for each tumor site group.
The patient population is consenting patients with an established primary cancer in the lung or tracheobronchial tree (n=20), esophagus (n=20), stomach (n=20), pancreas (n=20), and bile duct (n=20). Cases are selected sequentially from eligible pools for each site group; gender and age will reflect the referral population.
In this study, patients are not stratified within tumor site at the time of selection based on histologic type or other tumor variables. Controls include 10 colorectal cancer patients (positive controls) and 40 asymptomatic patients found to have a normal colonoscopy (negative controls) in a parallel screening study. The inclusion criteria are (1 ) signed consent, (2) age greater than 18 years, and (3) histologically confirmed primary tumor at appropriate anatomic site. The exclusion criteria include (1) known second primary aerodigestive cancer or premalignant adenoma greater than 1 cm outside of the site of the index cancer, (2) cathartic bowel preparation, barium x-rays, CT scan with oral contrast, or colonoscopy within 7 days, and (3) ostomy or less than '/2 colorectum remaining.
The three component test described in Example 1 will detect a proportion of cancers from each proximal aerodigestive site. Correlation of results between corresponding stool and tissue will help determine to what extent mutant DNA
expressed by tumors survives enteric transit and can be recovered in stool.

In this example a portion of the results from Example 6 relating to the supracolonic aerodigestive neoplasm patients are represented as follows.
Methods of the invention were used to detect supracolonic aerodigestive neoplasms in patients.
A stool sample was obtained from each of the 28 patients. The sample was prepared as described above. Fragments of the 5 different loci referred to above were amplified using primers spaced 200, 400, 800, 1300, 1800, and 2400 base pairs apart using the protocol described above in Example 4. Each amplification was scored such that successful amplification of a fragment received a score of 1, and no amplification received a score of 0. Since five loci were interrogated using 6 primer pairs each, the maximum obtainable score was 30 (successful amplification of all 6 fragments at all five loci). A score of 21 was used as a cutoff between diseased and non-diseased patients. The results are shown below.

Table 4 Supracolonic Aerodigestive Cancers Patient SupracolonicAge Score No. Aerodigestive Cancer P-145 Pancreas 68 30 P-164 Lun CA 68 30 P-166 Bile Duct 52 30 P-189 Bile Duct 43 30 P-190 Lun CA 50 30 P-019 Atypical 71 29 Findings in Stomach P-152 Lun CA 77 28 P-167 Pancreas 72 28 P-011 Lun CA 73 27 P-153 Pancreas 65 27 P-165 Lun CA 85 27 P-170 Duodenum 65 27 P-182 Barrett's 58 27 Eso ha us P-146 Bile Duct 63 26 P-081 Barrett's 74 26 Eso ha us P-151 Pancreas 49 25 P-155 Lun CA 60 25 P-156 Lun CA 57 25 P-150 Pancreas 78 23 P-149 Eso ha us 59 19 P-154 Eso ha us 80 19 P-169 Pancreas 71 19 P-168 Lun CA 63 18 P-180 Pancreas 67 13 P-144 Eso ha us 59 9 P-147 Stomach 57 7 P-148 Stomach 69 6 P-171 Esophagus 76 0 As shown above, methods of the invention successfully screened 18 out of 27 patients who actually had a supracolonic aerodigestive neoplasm. Only one patient was misdagnosed as having cancer when he did not. Thus, the methods of the invention are useful for non-invasive diagnosis of supracolonic aerodigestive neoplasm in a patient.
The threshold of 21 for a positive screen can be changed to accommodate desired sensitivities and specificities. For example, if 18 were determined to be the cutoff, the false negative results shown in Table 4 would be avoided. The skilled artisan knows how to set thresholds depending on the patient (e.g., a lower threshold for patients with symptoms than patients presenting with no symptoms), the disease being diagnosed, and the desired level of sensitivity and specificity.
Regardless of the threshold, the principle of the invention remains that long DNA can be used to detect supracolonic aerodigestive neoplasms.
OTHER EMBODIMENTS
It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.

Claims (29)

WHAT IS CLAIMED IS:

1. A method for detecting a supracolonic aerodigestive neoplasm in a mammal, said method comprising determining whether a stool sample from said mammal contains a neoplasm specific marker, wherein said marker is from said supracolonic aerodigestive neoplasm.

2. The method of claim 1, wherein said supracolonic aerodigestive neoplasm comprises a premalignant neoplasm.

3. The method of claim 1, wherein said aerodigestive neoplasm comprises a malignant neoplasm.

4. The method of claim 1, wherein said supracolonic aerodigestive neoplasm is selected from the group consisting of lung, naso-oro-pharyngeal, esophageal, stomach, liver, bile duct, gall bladder, small intestine, and pancreas neoplasms.

5. The method of claim 1, wherein said mammal is a human.

6. The method of claim 1, wherein said neoplasm-specific marker comprises a neoplasm-specific nucleic acid marker.

7. The method of claim 6, wherein said neoplasm-specific nucleic acid marker comprises nucleic acid having a point mutation.

8. The method of claim 7, wherein said point mutation is located in a gene selected from the group consisting of K-ras, APC, and p53.

9. The method of claim 6, wherein said neoplasm-specific nucleic acid marker comprises nucleic acid that reflects microsatellite instability.

10. The method of claim 9, wherein said microsatellite instability is located in the BAT-26 gene.

11. The method of claim 6, wherein said neoplasm-specific nucleic acid marker comprises long DNA.

12. The method of claim 11, wherein said long DNA comprises DNA greater than about 300 base pairs in length.

13. The method of claim 11, wherein said long DNA comprises DNA greater than about 400 base pairs in length.

14. The method of claim 11, wherein said long DNA comprises DNA greater than about 500 base pairs in length.

15. The method of claim 11, wherein said long DNA comprises DNA greater than about 1000 base pairs in length.

16. The method of claim 1, wherein said neoplasm-specific marker comprises a neoplasm-specific polypeptide marker.

17. The method of claim 1, wherein said neoplasm-specific marker comprises a neoplasm-specific cell marker.

18. The method of claim 1, wherein said method comprises determining whether said stool sample contains two or more neoplasm-specific markers.

19. The method of claim 18, wherein said two or more neoplasm-specific markers are selected from the group consisting of nucleic acid markers, polypeptide markers, and cell markers.

20. The method of claim 18, wherein said two or more neoplasm-specific markers are neoplasm-specific nucleic acid markers.

21. The method of claim 20, wherein said two or more neoplasm-specific nucleic acid markers are selected from the group consisting of nucleic acid having a point mutation, nucleic acid that reflects microsatellite instability, and long DNA.

22. The method of claim 20, wherein said supracolonic aerodigestive neoplasm is a small intestine neoplasm.

23. The method of claim 22, wherein said small intestine neoplasm is a duodenum neoplasm.

24. The method of claim 22, wherein said small intestine neoplasm is a jejunum neoplasm.

25. The method of claim 22, wherein said small intestine neoplasm is an ileum neoplasm.

26. The method of claim 1, wherein said supracolonic aerodigestive neoplasm is a pancreas neoplasm.

27. The method of claim 26, wherein said neoplasm-specific marker comprises long DNA.

28. The method of claim 26, wherein said method comprises determining whether said stool sample contains two or more neoplasm-specific markers.

29. A method for detecting a supracolonic aerodigestive neoplasm in a mammal, said method comprising determining whether a stool sample from said mammal contains long DNA, said long DNA being from said supracolonic aerodigestive neoplasm.