Colorectal cancer screening by detection of altered human DNA in stool: feasibility of a multitarget assay panel.

BACKGROUND & AIMS: Assay of altered DNA exfoliated into stool represents an intriguing approach to screen for colorectal neoplasia, but multiple markers must be targeted because of genetic heterogeneity. We explored the feasibility of a stool assay panel of selected DNA alterations in discriminating subjects with colorectal neoplasia from those without. METHODS: Freezer-archived stools were analyzed in blinded fashion from 22 patients with colorectal cancer, 11 with adenomas > or =1 cm, and 28 with endoscopically normal colons. After isolation of human DNA from stool by sequence-specific hybrid capture, assay targets included point mutations at any of 15 sites on K-ras, p53, and APC genes; Bat-26, a microsatellite instability marker; and highly amplifiable DNA. RESULTS: Analyzable human DNA was recovered from all stools. Sensitivity was 91% (95% confidence interval, 71%-99%) for cancer and 82% (48%-98%) for adenomas > or =1 cm with a specificity of 93% (76%-99%). Excluding K-ras from the panel, sensitivities for cancer were unchanged but decreased slightly for adenomas to 73% (39%-94%), while specificity increased to 100% (88%-100%). CONCLUSIONS: Assay of altered DNA holds promise as a stool screening approach for colorectal neoplasia. Larger clinical investigations are indicated.

Intratumoral genetic heterogeneity and progression of endometrioid type endometrial adenocarcinomas.

OBJECTIVES: Development of genetic heterogeneity is one mechanism whereby tumors may acquire increasing aggressiveness during neoplastic progression. In this study we relate development of intratumoral genetic heterogeneity to invasion and metastatic spread of sporadic endometrioid (type I) endometrial adenocarcinomas. METHODS: Microsatellite unstable adenocarcinomas underwent detailed microsatellite allelotype mapping with reconstruction of neoplastic lineages using maximum parsimony analysis. RESULTS: Within individual patients, tumor allelotypes sometimes varied between regions of histologically identical tumor, indicating that genotypic variation may reflect differences inapparent by histology. Comparison of noninvasive (surface/luminal) with invasive (myometrial invasion or metastasis) carcinoma showed highly related genotypes in 3/8 cases in which the invasive component can be recognized as evolved from the superficial tumor lineage by progressive clonal selection. In 3/8 cases superficial and invasive genotypes independently evolved different sets of altered microsatellites, indicating either divergence at an early stage in tumor evolution or independent selection events. A total of 2/8 cases had random patterns of marker distribution between sampled areas that were not informative in delineating systematic relationships between surface and invasive tumor. CONCLUSIONS: We conclude from these results that endometrial tumor progression may occur through physical extension of existing clones or through creation of new subclones with altered growth properties. The latter occurs in about half of cases, where myometrial invasion may select for particular clones that are poorly represented on the luminal surface.

Sporadic microsatellite instability is specific to neoplastic and preneoplastic endometrial tissues.

Microsatellite instability is a frequent (13%-24%) finding in sporadic endometrial adenocarcinoma and its precursor lesions, but most studies are limited to patients who already have malignant or premalignant endometrial disease. We performed retrospective testing for microsatellite instability in women in whom cancers showing microsatellite instability developed later and prospective testing in randomly selected normal and anovular endometrial biopsy specimens. Microsatellite instability in cancer-bearing biopsy specimens accurately reflected that seen in matched malignant tissues obtained at hysterectomy. In 1 patient, microsatellite instability developed in a scanty sample of fragmented endometrial tissues 7 years before the onset of endometrial cancer. Prospective testing for microsatellite instability in the endometria of women unselected for subsequent appearance of endometrial cancer showed a very low rate of microsatellite instability. Only 1 endometrial specimen showing microsatellite instability was found among 75 anovulatory endometrial specimens, and none were found in 377 normal endometrial specimens and 46 polyps examined. Microsatellite instability may precede the onset of histologically diagnosed carcinoma but is rare in randomly sampled histologically normal endometrial tissues.

Patterns of allelic loss (LOH) in vulvar squamous carcinomas and adjacent noninvasive epithelia.

The pathogenesis of carcinoma of the vulva is diverse and includes both human papilloma virus (HPV)-positive and HPV-negative pathways. The objective of this study was to correlate the morphology with patterns of loss of heterozygosity (LOH) within four vulvar carcinomas and in adjacent vulvar epithelia. Tumors were categorized as HPV positive or negative by polymerase chain reaction (PCR) analysis. Forty-one different sites of normal squamous mucosa, hyperplasia, vulvar intraepithelial neoplasia (VIN), and carcinoma were microdissected in duplicate, and each extracted DNA was analyzed in duplicate for LOH at 10 chromosomal loci by PCR and polyacrylamide gel electrophoresis. Patterns of LOH were compared within different sites of tumors and between the tumor and the noninvasive epithelia. Of three tumors with multiple invasive foci analyzed, divergent patterns of LOH were identified in two, correlating in one with differences in tumor grade. In one HPV-16-positive case, multiple sites of VIN displayed heterogeneity for LOH consistent with divergent clonal or subclonal populations, some of which were not shared by the tumor. In one HPV-negative case, LOH was found in foci of hyperplasia and differentiated VIN (atypical hyperplasia), the latter sharing LOH with the invasive carcinoma at some but not all chromosomal loci. This study suggests that a genetic relationship exists between VIN and carcinoma, irrespective of HPV involvement. It also suggests that in HPV-negative tumors, allelic loss may predate the onset of invasive carcinoma and, in some cases, cellular atypia (VIN). However, the divergent patterns of LOH observed imply that many genetic alterations in the adjacent vulvar epithelium are not directly related to the invasive carcinoma.

Monoclonal origin of vulvar intraepithelial neoplasia and some vulvar hyperplasias.

Squamous neoplasms of the female genital tract, including vulvar intraepithelial neoplasia, presumably are derived from a single cell. This study addressed this hypothesis and determined the clonal status of other squamous epithelial alterations associated with vulvar carcinoma, including hyperplasia and lichen sclerosis. X chromosome inactivation patterns of 22 epithelial lesions and matched normal epithelium were determined using a polymerase chain reaction (PCR)-based assay targeting the X-linked human androgen receptor gene (HUMARA). Clonality was inferred by comparing matched lesional and control tissues as follows: 1) monoclonal, if intensity of either PCR product was skewed relative to normal reference epithelium (control), 2) polyclonal, if both lesional and control were unskewed, and 3) unknown, if both lesion and control tissues were skewed toward the same allele. Two cases were excluded because of noninformative homozygous HUMARA alleles. Of 8 vulvar intraepithelial neoplasias analyzed, 7 were scored monoclonal and 1 polyclonal. Of 12 hyperplasias, 6 were monoclonal, including one with lichen sclerosis, 2 were polyclonal, and in 4, the clonal status could not be determined. The PCR-based clonal assay supports a monoclonal derivation for vulvar intraepithelial neoplasia and, in some cases, vulvar hyperplasia, and lichen sclerosis. The finding of monoclonal hyperplasia and lichen sclerosis suggests that clonal expansion may evolve before the development of morphological atypia in these epithelia.

Clonal analysis of sacrococcygeal "teratomas".

Congenital masses of the sacrococcygeal region commonly contain multiple tissues and have variously been subclassified as neoplasms or congenital hamartomas based on clinicopathological and embryological observations. We have used a polymerase chain reaction-based assay for nonrandom X chromosome inactivation to infer the clonality of three cogenital sacrococcygeal tumors previously diagnosed as teratomas. One solid immature teratoma was monoclonal, and a predominantly cystic histologically mature mass was polyclonal. A third immature teratoma was noninformative because of baseline asymmetry of polyclonal tissue X inactivation. We confirm that immature teratomas at this site appear to be monoclonal neoplasms and suggest that at least some histologically mature "teratomas" are more appropriately classified as hamartomas.

Allelotype mapping of unstable microsatellites establishes direct lineage continuity between endometrial precancers and cancer.

Progressive microsatellite changes in replication error positive (RER+) endometrium were used to reconstruct evolutionary stages of nonfamilial adenocarcinoma. RER+ putative endometrial precancers (atypical endometrial hyperplasias) progress to RER+ carcinomas, which retain some of the altered microsatellites acquired in earlier precursor stages. The RER+ phenotype may provide a specific marker for early-stage endometrial neoplasms that cannot be resolved by routine histopathology and may be a useful tool to stratify stages in the evolution of RER+ tumors.

Uteri of women with endometrial carcinoma contain a histopathological spectrum of monoclonal putative precancers, some with microsatellite instability.

We have tested the hypothesis that endometrial precancers persist in uteri of patients with endometrial carcinoma and are monoclonal. Twenty-two hysterectomies with both well-differentiated endometrial adenocarcinoma and adjacent (normal or abnormal) noncancerous endometrium underwent successful clonal analysis using a PCR assay for nonrandom X chromosome inactivation. Monoclonal lesions included endometrial carcinoma, endometrial polyps, and atypical endometrial hyperplasias, whereas normal and anovulatory endometrium were polyclonal. Comparison of the specific X chromosome copy preferentially inactivated by the matched monoclonal cancers and associated monoclonal lesions allowed us to exclude polyps, but not endometrial hyperplasias, as potential precancers. The repetitive genetic marker (HUMARA) for X inactivation was altered in some cancers, permitting identification of microsatellite instability (RER+). Two patients with RER+ cancers also had adjacent RER+ hyperplasias. The seven monoclonal and two RER+ hyperplasias had focal or diffuse cytological atypia, a feature previously associated with risk for endometrial cancer. We conclude that: (a) putative endometrial precancers and cancers share a monoclonal growth pattern; (b) cancers with microsatellite instability may acquire this feature as precancers; and (c) monoclonal endometrial precancers have the morphology of hyperplasias, which vary in the extent of cytological atypia and degree of architectural complexity.

X chromosome inactivation in the normal female genital tract: implications for identification of neoplasia.

Monoclonal proliferative lesions may be identified by X chromosome inactivation skewing relative to normal polyclonal tissues. We have quantitatively analyzed X-inactivation patterns throughout polyclonal uterine tissues to develop interpretive criteria for recognition of monoclonal neoplasms. Six fresh tissue samples (two samples each of cervix, endometrium, and myometrium) were collected from hysterectomy specimens, and the percentage of androgen receptor (HUMARA) marker allele present on inactive X chromosomes was calculated from a PCR assay. Exact balancing yields 50% of the marker on the inactive X, whereas complete skewing shows either 0 or 100%. X inactivation was similar throughout the tissues of each uterus but was significantly different among the 11 women studied. Comparison of differences in X inactivation between pairs of polyclonal tissue samples within each uterus (Xi spread) permitted delineation of cumulative experimental and biological variation of this parameter. Polyclonal-polyclonal Xi spread averaged 10.7 and was independent of the tissue type, sampling site, or the individual studied. Severe baseline skewing of reference polyclonal tissues or contamination of monoclonal tissue by polyclonal cells may reduce the polyclonal-monoclonal Xi spread. The extent of X-inactivation skewing necessary to infer a monoclonal process should exceed the 20 or 27 point spread seen, respectively, between 85 and 95% of polyclonal samples.

PCR bias in amplification of androgen receptor alleles, a trinucleotide repeat marker used in clonality studies.

Trinucleotide CAG repeats in the X-linked human androgen receptor gene (HUMARA) have proved a useful means of determining X chromosome haplotypes, and when combined with methylation analysis of nearby cytosine residues permits identification of non-random X inactivation in tumors of women. Co-amplification of two alleles in a heterozygote generates PCR products which differ in the number of CAG units, and thus their melting and secondary structure characteristics. We have shown that under optimal conditions amplification efficiency of two HUMARA alleles is near-equivalent, generating PCR products in a ratio proportional to that of the genomic template. In contrast, reduction of template quantity, damage of template by ultraviolet irradiation or addition of monovalent salts (sodium chloride, sodium acetate or ammonium acetate) produces highly variable imbalances of allelic PCR products, with a strong tendency to preferentially amplify lower molecular weight alleles. Variability and biasing was diminished by substitution of 7-deaza-2'-dGTP for dGTP during amplification, an intervention which reduces stability of intramolecular and intermolecular GC base pairing. We conclude that DNA which is scanty, damaged or salt contaminated may display amplification bias of GC-rich PCR targets, potentially confounding accurate interpretation or reproducibility of assays which require co-amplification of alleles.