Value of combining HPV-DNA testing with follow-up Papanicolaou smear in patients with prior atypical squamous cells of undetermined significance

OBJECTIVE: To address human papillomavirus (HPV) testing on negative Pap tests preceded by atypical squamous cells of undetermined significance (ASC-US) without reflex HPV testing. STUDY DESIGN: Positive HPV test results with concurrent negative Pap tests over 1 year were identified. Pathology records for all patients diagnosed with ASCUS without reflex HPV testing in the previous year were reviewed; all cytologic and surgical specimens over the subsequent 2 years were evaluated for squamous abnormalities. RESULTS: Fifty patients had positive HPV DNA (HPV-DNA) test result combined with a negative Pap test. Twenty-three had a previous Pap test interpretation of ASC-US (without HPV testing) within the preceding year. On follow-up, 8 of 23 developed a squamous intraepithelial lesion (SIL) within 1 year. Four additional cases developed SIL in the second year after positive HPV testing. All dysplasias in the first year of follow-up were low grade; 1 of 4 developing in the second year was high grade. CONCLUSION: Negative Pap smear following an ASC-US interpretation without a concurrent HPV test is associated with significant false negative rate. We suggest consideration of combining HPV-DNA testing to all initial follow-up negative Pap tests of patients with previous ASC-US, if reflex HPV testing has not been performed.     

An integrated approach for the analysis of biological pathways using mixed models

Gene class, ontology, or pathway testing analysis has become increasingly popular in microarray data analysis. Such approaches allow the integration of gene annotation databases, such as Gene Ontology and KEGG Pathway, to formally test for subtle but coordinated changes at a system level. Higher power in gene class testing is gained by combining weak signals from a number of individual genes in each pathway. We propose an alternative approach for gene-class testing based on mixed models, a class of statistical models that: a) provides the ability to model and borrow strength across genes that are both up and down in a pathway, b) operates within a well-established statistical framework amenable to direct control of false positive or false discovery rates, c) exhibits improved power over widely used methods under normal location-based alternative hypotheses, and d) handles complex experimental designs for which permutation resampling is difficult. We compare the properties of this mixed models approach with nonparametric method GSEA and parametric method PAGE using a simulation study, and illustrate its application with a diabetes data set and a dose-response data set.     

Fip1 regulates the activity of Poly(A) polymerase through multiple interactions

Fip1 is an essential component of the Saccharomyces cerevisiae polyadenylation machinery and the only protein known to interact directly with poly(A) polymerase (Pap1). Its association with Pap1 inhibits the extension of an oligo(A) primer by limiting access of the RNA substrate to the C-terminal RNA binding domain (C-RBD) of Pap1. We present here the identification of separate functional domains of Fip1. Amino acids 80 to 105 are required for binding to Pap1 and for the inhibition of Pap1 activity. This region is also essential for viability, suggesting that Fip1-mediated repression of Pap1 has a crucial physiological function. Amino acids 206 to 220 of Fip1 are needed for the interaction with the Yth1 subunit of the complex and for specific polyadenylation of the cleaved mRNA precursor. A third domain within amino acids 105 to 206 helps to limit RNA binding at the C-RBD of Pap1. Our data demonstrate that the C terminus of Fip1 is required to relieve the Fip1-mediated repression of Pap1 in specific polyadenylation. In the absence of this domain, Pap1 remains in an inhibited state. These findings show that Fip1 has a crucial regulatory function in the polyadenylation reaction by controlling the activity of poly(A) tail synthesis through multiple interactions within the polyadenylation complex.     

Clinical management of patients with invasive cervical cancer following a negative Pap smear

Among 535 patients with invasive cervical carcinoma seen between January 1975 and June 1986, 26 were found to have developed the disease within six months (65 percent), 35 within 12 months (88 percent), 37 within 13 months (93 percent), and three developed the disease within 17 months after a negative Pap smear. Eighty-eight percent of these 40 patients were under age 40 at diagnosis. Rapidly progressive cancers are highly resistant to radiation therapy. Seven stage IB patients treated only with radiation died within nine to 29 months after initial therapy. By contrast, 15 patients treated by radical hysterectomy and four by radical hysterectomy and post-surgical radiation were alive with no evidence of disease from six to 109 months after surgery (median, 30 months). Six of nine patients with stage II to IV disease treated with radiation have died; the remaining three are alive. One patient is well 14 months after therapy, but two others have developed metastases seven and 12 months after treatment. Surprisingly, 37 of 40 patients had symptoms of pain, bleeding, and discharge at the initial diagnosis, but their physicians had a false sense of security because of a recent negative Pap smear. Early biopsy diagnosis and radical hysterectomy with bilateral pelvic lymphadenectomy is the most effective management for this cancer.     

A feasibility study to identify proteins in the residual Pap test fluid of women with normal cytology by mass spectrometry-based proteomics

Background

The proteomic analysis of body fluids is a growing technology for the identification of protein biomarkers of disease. Given that Papanicolaou tests (Pap tests) are routinely performed on over 30 million women annually in the U.S. to screen for cervical cancer, we examined the residual Pap test fluid as a source of protein for analysis by mass spectrometry (MS). In the liquid-based Pap test, cervical cells are collected from the ectocervix and placed into an alcohol-based fixative prior to staining and pathologic examination. We hypothesized that proteins shed by cells of the female genital tract can be detected in the Pap test fixative by MS-based proteomic techniques. We examined the feasibility of using residual fluid from discarded Pap tests with cytologically “normal” results to optimize sample preparation for MS analysis. The protein composition of the cell-free Pap test fluid was determined by silver staining of sodium dodecyl sulfate -polyacrylamide gels, and the abundance of serum proteins was examined by Western immunoblot using an antibody against human serum albumin. Both pooled and individual samples were trypsin digested and analyzed by two-dimensional MS/MS. Proteins were identified by searching against the Human Uniprot database, and characterized for localization, function and relative abundance.

Results

The average volume of the residual Pap test fluid was 1.5 ml and the average protein concentration was 0.14 mg/ml. By Western immunoblot we showed that the amount of albumin in each sample was significantly reduced compared to normal serum. By MS/MS, we identified 714 unique proteins in pooled Pap test samples and an average of 431 proteins in individual samples. About 40% of the proteins identified were extracellular or localized to the plasma membrane. Almost 20% of the proteins identified were involved in immunity and defense, characteristic of the healthy cervical-vaginal proteome. By merging the protein sets from the individual and pooled Pap test samples, we created a “Normal Pap test Core Proteome” consisting of 153 proteins.

Conclusions

Residual Pap test fluid contains a sufficient amount of protein for analysis by MS and represents a valuable biospecimen source for the identification of protein biomarkers for gynecological diseases.     

Bayes empirical bayes inference of amino acid sites under positive selection

Codon-based substitution models have been widely used to identify amino acid sites under positive selection in comparative analysis of protein-coding DNA sequences. The nonsynonymous-synonymous substitution rate ratio (d(N)/d(S), denoted omega) is used as a measure of selective pressure at the protein level, with omega > 1 indicating positive selection. Statistical distributions are used to model the variation in omega among sites, allowing a subset of sites to have omega > 1 while the rest of the sequence may be under purifying selection with omega < 1. An empirical Bayes (EB) approach is then used to calculate posterior probabilities that a site comes from the site class with omega > 1. Current implementations, however, use the naive EB (NEB) approach and fail to account for sampling errors in maximum likelihood estimates of model parameters, such as the proportions and omega ratios for the site classes. In small data sets lacking information, this approach may lead to unreliable posterior probability calculations. In this paper, we develop a Bayes empirical Bayes (BEB) approach to the problem, which assigns a prior to the model parameters and integrates over their uncertainties. We compare the new and old methods on real and simulated data sets. The results suggest that in small data sets the new BEB method does not generate false positives as did the old NEB approach, while in large data sets it retains the good power of the NEB approach for inferring positively selected sites.     

Effects of filtering by Present call on analysis of microarray experiments

Background

Affymetrix GeneChips^® are widely used for expression profiling of tens of thousands of genes. The large number of comparisons can lead to false positives. Various methods have been used to reduce false positives, but they have rarely been compared or quantitatively evaluated. Here we describe and evaluate a simple method that uses the detection (Present/Absent) call generated by the Affymetrix microarray suite version 5 software (MAS5) to remove data that is not reliably detected before further analysis, and compare this with filtering by expression level. We explore the effects of various thresholds for removing data in experiments of different size (from 3 to 10 arrays per treatment), as well as their relative power to detect significant differences in expression.

Results

Our approach sets a threshold for the fraction of arrays called Present in at least one treatment group. This method removes a large percentage of probe sets called Absent before carrying out the comparisons, while retaining most of the probe sets called Present. It preferentially retains the more significant probe sets (p ≤ 0.001) and those probe sets that are turned on or off, and improves the false discovery rate. Permutations to estimate false positives indicate that probe sets removed by the filter contribute a disproportionate number of false positives. Filtering by fraction Present is effective when applied to data generated either by the MAS5 algorithm or by other probe-level algorithms, for example RMA (robust multichip average). Experiment size greatly affects the ability to reproducibly detect significant differences, and also impacts the effect of filtering; smaller experiments (3–5 samples per treatment group) benefit from more restrictive filtering (≥50% Present).

Conclusion

Use of a threshold fraction of Present detection calls (derived by MAS5) provided a simple method that effectively eliminated from analysis probe sets that are unlikely to be reliable while preserving the most significant probe sets and those turned on or off; it thereby increased the ratio of true positives to false positives.     

Utility of imprint cytology for early presumptive diagnosis in clinically suspicious cervical cancer

OBJECTIVE: To determine the utility of imprint cytology (IC) in providing an early presumptive diagnosis of clinically suspected cervical carcinoma. STUDY DESIGN: A total of 219 clinically suspicious cervical cancer cases underwent Pap test, punch biopsy and IC at the same sitting. Correlations were performed between these diagnostic modalities to determine the sensitivity and specificity of IC in diagnosis of cervical cancer. RESULTS: The overall accuracy of IC in detecting cervical cancers was 96.2%. About 78% of squamous cell carcinomas (SCC), 60% of adenocarcinomas and 100% of small cell carcinoma could be accurately typed on imprints. Twelve malignant lesions were diagnosed on IC among 26 unsatisfactory biopsies. Although there was no false positive result, 3.5% false negative diagnoses were given on IC. The sensitivity and specificity of imprint smear cytology to detect malignancy was 96.2% and 100%. Agreement between imprint cytology and Pap smear diagnosis of malignancy was 95.3%. kappa Statistics revealed excellent agreement between imprints and biopsies and between imprints and Pap smears in diagnosis of malignant lesions. CONCLUSION: IC can be used as an adjunctive technique for an early and reliable preliminary presumptive diagnosis of cancer of the uterine cervix.     

Comparative study of the ThinPrep Pap test and conventional cytology results in a Canadian cohort

OBJECTIVE: To compare the frequency of Pap test results in a prospective series of direct to vial ThinPrep tests to a cohort of conventionally prepared tests. To follow-up all test results for a minimum of 2 years and assess performance based on this outcome. METHODS: All women presenting for either routine screening or colposcopic examination in 2001 were enrolled in the ThinPrep cohort. A similar, population of conventionally prepared tests was extracted from the year 2000 laboratory data. Information on all concurrent and follow-up cervical specimens over the ensuing 2 years was retrieved. RESULTS: The ThinPrep cohort comprised 2288 Pap tests and the conventional, 2211. The frequency of normal [within normal limits (WNL) and benign cellular changes (BCC)] results in the ThinPrep cohort was 6% lower and the frequency of abnormal [> or =atypical squamous cells of undetermined significance (ASCUS)] results was 6.8% higher. Respective ThinPrep and conventional cohort results were 1156 (51%) and 1291 (58%) WNL, 625 (27%) and 561 (25%) BCC, 101 (4%) and 65 (3%) ASCUS, 21 (1%) and 2 (0.1%) atypical glandular cells of undetermined significance, 301 (13%) and 224 (10%) low-grade squamous intraepithelial lesion (LSIL), and 74 (3%) and 40 (2%) high-grade SIL (HSIL) (P < 0.0001). Follow-up was available for nearly 80% of each cohort. LSIL or higher was confirmed in 57.5% (n = 266) of the abnormal ThinPrep and 60.9% (n = 190) of the abnormal conventional tests. The ThinPrep yield of confirmed tests however was almost 50% higher than the conventional test. CONCLUSION: In this population, ThinPrep was superior to the conventional Pap test.     

Mixture models for detecting differentially expressed genes in microarrays

An important and common problem in microarray experiments is the detection of genes that are differentially expressed in a given number of classes. As this problem concerns the selection of significant genes from a large pool of candidate genes, it needs to be carried out within the framework of multiple hypothesis testing. In this paper, we focus on the use of mixture models to handle the multiplicity issue. With this approach, a measure of the local FDR (false discovery rate) is provided for each gene. An attractive feature of the mixture model approach is that it provides a framework for the estimation of the prior probability that a gene is not differentially expressed, and this probability can subsequently be used in forming a decision rule. The rule can also be formed to take the false negative rate into account. We apply this approach to a well-known publicly available data set on breast cancer, and discuss our findings with reference to other approaches.     

An empirical Bayesian approach for identifying differential coexpression in high-throughput experiments

A common goal of microarray and related high-throughput genomic experiments is to identify genes that vary across biological condition. Most often this is accomplished by identifying genes with changes in mean expression level, so called differentially expressed (DE) genes, and a number of effective methods for identifying DE genes have been developed. Although useful, these approaches do not accommodate other types of differential regulation. An important example concerns differential coexpression (DC). Investigations of this class of genes are hampered by the large cardinality of the space to be interrogated as well as by influential outliers. As a result, existing DC approaches are often underpowered, exceedingly prone to false discoveries, and/or computationally intractable for even a moderately large number of pairs. To address this, an empirical Bayesian approach for identifying DC gene pairs is developed. The approach provides a false discovery rate controlled list of significant DC gene pairs without sacrificing power. It is applicable within a single study as well as across multiple studies. Computations are greatly facilitated by a modification to the expectation-maximization algorithm and a procedural heuristic. Simulations suggest that the proposed approach outperforms existing methods in far less computational time; and case study results suggest that the approach will likely prove to be a useful complement to current DE methods in high-throughput genomic studies.     

Novel developments in rapid mycotoxin detection

Rapid antibody-based mycotoxin screening techniques are designed to be used outside a laboratory environment, at the place of sampling. Results are expected immediately, so that commodities can be further processed without delay. Because they are used for mycotoxin analysis, very low levels (ppb and ppt range) should be detected. A further requirement is that the obtained results are accurate with a false negative rate of <5% at the level of interest. At first, plastic microtiter plates were used as solid phase materials for immobilizing antibodies (enzyme-linked immunosorbent assays). However, to increase speed and user-friendliness, plastics were replaced by microporous membranes. As an example a flow-through enzyme immunoassay for the detection of fumonisins in cornflakes with a cut-off value of 275 μg/kg is described. No false negative results were observed and the false positive rate was 18%. However, enzyme labels, used to enable visual evaluation of results, did not seem to be completely satisfactory in terms of stability and repeatability of the generated signal. Therefore microparticle labels such as colloidal gold particles are used more and more,e.g. in a lateral flow dipstick immunoassay. When applied to the detection of aflatoxin B₁ in pig feed a cut-off value of 5 μg/kg could be reached with no false negative results and a false positive rate of only 10%. Sample pretreatment for screening techniques should be rapid and simple. Preferably a simple solvent extraction is used, followed by a filtration and dilution step. However, for strongly coloured or complex food matrices, this did not seem to work. The combination of clean-up and detection in one single test device is a new approach. When using this clean-up tandem assay column for the detection of ochratoxin A in roasted coffee, a cut-off value of 6 μg/kg was reached. No false positive results were obtained, however, the false negative rate was 8%. Presented at the EU-USA Bilateral Workshop on Toxigenic Fungi & Mycotoxins, New Orleans, USA, July 5–7, 2005 Financial support: Belgian Federal Science Policy Office (SPSDII-CPAA 15 project), IWT-Flanders (research project 020448/Toxi-Test) and Bijzonder Onderzoeksfonds Ghent University (01 1D02803)     

Accuracy of liquid-based Pap tests: comparison of concurrent liquid-based tests and cervical biopsies on 782 women with previously abnormal Pap smears

OBJECTIVE: To evaluate the diagnostic performance of a liquid-based Pap test, the ThinPrep Pap test (TP) (Cytyc Corp., Boxborough, Massachusetts, U.S.A.), by comparing concurrent TP and cervical biopsy results on 782 patients who were referred for colposcopy because of previously abnormal conventional Pap smears (CPs). STUDY DESIGN: The ability of TP diagnoses of atypical cells of undetermined significance (ASC-US) and squamous intraepithelial lesions (SILs) to predict biopsy diagnoses of cervical intraepithelial neoplasia (CIN) was analyzed using chi2 and McNemar tests. RESULTS: The rate of agreement between diagnoses of SIL by TP and CIN by biopsy was 74.7%. ASC-US accounted for 16.0% of TP diagnoses. ASC-US had biopsy diagnoses of CIN 1 in 60% and CIN 2/3 in 12.8% of cases. For TP diagnosis of low grade SIL, biopsy diagnoses of CIN 2/3 were found in 13.5% of cases. For TP diagnoses of ASC-US and higher, the proportions of TP and cervical biopsies in comparable diagnostic categories were statistically significant (p < 0.001), with TP having sensitivity of 89.4% and positive predictive value of 89.7% for the detection of CIN. The false positive rate for TP was 8.1%, but rescreening confirmed the presence of abnormal cells in 51 of 63 (81.0%) cases of ASC-US or higher having negative biopsies. TP had a false negative rate of 8.3% and negative predictive value of 61.3%. Rescreening showed that most (77.6%) of the false negative TP specimens failed to have abnormal cells on the slides. CONCLUSION: For patients having previously detected cervical abnormalities by CP, concurrent TP demonstrated the following: (1) that it has high diagnostic accuracy for SIL, (2) that ASC-US was diagnostically equivalent to LSIL, and (3) that false negative TP for SIL can be attributed primarily to sampling rather than cytotechnologists' screening errors.     

Use of a liquid-based, thin-layer Pap test in a community hospital. Impact on cytology performance and productivity

OBJECTIVE: To evaluate the direct-to-vial efficacy of the SurePath Pap test (TriPath Imaging, Burlington, North Carolina, U.S.A.) in a community hospital laboratory and to assess its impact on productivity, as measured by Pap test turnaround times (TATs). STUDY DESIGN: A total of 8,771 SurePath Pap tests were compared to 5,055 conventional Pap smears collected and processed over the same 12-month period. SurePath histologic correlation rates were compared to historical correlation rates for conventional Pap smears. Pap test TATs for 3 months prior to implementing SurePath were compared to TATs for the last 3 months of the study, which included an approximate 70:30 ratio of SurePath to conventional Pap tests. Laboratory staffing was unchanged, and mean monthly accessions were relatively constant. RESULTS: SurePath showed statistically significant improvements in the detection of low grade (LSIL) (196%) and high grade (HSIL) (243%) squamous intraepithelial lesions (SILs) relative to conventional Pap smears. The atypical cells of undetermined significance (ASCUS) rate remained nearly constant with SurePath, while the ASCUS/SIL ratio decreased by 68%. Unsatisfactory rates declined 81%. Histologic correlation rates with SurePath increased over those of conventional Pap smears. Pap test mean TATs improved 73% with SurePath. CONCLUSION: Sure-Path detected significantly more cases of LSIL and HSIL than conventional smears without compromising specificity. The ASCUS/SIL ratio and unsatisfactory rate declined dramatically. Pap test TATs also improved markedly with SurePath.     

False discoveries and models for gene discovery

In the search for genes underlying complex traits, there is a tendency to impose increasingly stringent criteria to avoid false discoveries. These stringent criteria make it hard to find true effects, and we argue that it might be better to optimize our procedures for eliminating and controlling false discoveries. Focusing on achieving an acceptable ratio of true- and false-positives, we show that false discoveries could be eliminated much more efficiently using a stepwise approach. To avoid a relatively high false discovery rate, corrections for 'multiple testing' might also be needed in candidate gene studies. If the appropriate methods are used, detecting the proportion of true effects appears to be a more important determinant of the genotyping burden than the desired false discovery rate. This raises the question of whether current models for gene discovery are shaped excessively by a fear of false discoveries.     

Microarray meta-analysis database (M2DB): a uniformly pre-processed, quality controlled, and manually curated human clinical microarray database

Background

With the rapid expansion of DNA sequencing databases, it is now feasible to identify relevant information from prior sequencing projects and completed genomes and apply it to de novo sequencing of new organisms. As an example, this paper demonstrates how such extra information can be used to improve de novo assemblies by augmenting the overlapping step. Finding all pairs of overlapping reads is a key task in many genome assemblers, and to this end, highly efficient algorithms have been developed to find alignments in large collections of sequences. It is well known that due to repeated sequences, many aligned pairs of reads nevertheless do not overlap. But no overlapping algorithm to date takes a rigorous approach to separating aligned but non-overlapping read pairs from true overlaps.

Results

We present an approach that extends the Minimus assembler by a data driven step to classify overlaps as true or false prior to contig construction. We trained several different classification models within the Weka framework using various statistics derived from overlaps of reads available from prior sequencing projects. These statistics included percent mismatch and k-mer frequencies within the overlaps as well as a comparative genomics score derived from mapping reads to multiple reference genomes. We show that in real whole-genome sequencing data from the E. coli and S. aureus genomes, by providing a curated set of overlaps to the contigging phase of the assembler, we nearly doubled the median contig length (N50) without sacrificing coverage of the genome or increasing the number of mis-assemblies.

Conclusions

Machine learning methods that use comparative and non-comparative features to classify overlaps as true or false can be used to improve the quality of a sequence assembly.     

Geometric morphometric study of the regional variation of modern human craniofacial form.

The regional variability of the modern human craniofacial form is of importance to debates about human origins. The study of craniofacial form has generally been carried out either by interlandmark distance measurement and analysis or by observation and character scoring. In this study of four modern human groups (Eskimo/Inuit, African, Australian, and Romano-British), nine craniofacial landmark coordinates were recorded by extraction from laser scans. The coordinates were studied by geometric morphometrics, and a regression analysis was used to investigate the dominant variability in shape within and between groups. Statistical tests of shape difference between groups were carried out. By these methods, the statistical patterns of shape variability and their geometric interpretations were studied on a common basis. The results were found to be in agreement with the classic studies of Howells ([1989:189] Pap Peabody Mus 79), and show the potential of this approach for future research.     

Hybrid Mixture Model for Subpopulation Identification

Personalized medicine aims to identify those patients who have good or poor prognosis for overall disease outcomes or therapeutic efficacy for a specific treatment. A well-established approach is to identify a set of biomarkers using statistical methods with a classification algorithm to identify patient subgroups for treatment selection. However, there are potential false positives and false negatives in classification resulting in incorrect patient treatment assignment. In this paper, we propose a hybrid mixture model taking uncertainty in class labels into consideration, where the class labels are modeled by a Bernoulli random variable. An EM algorithm was developed to estimate the model parameters, and a parametric bootstrap method was used to test the significance of the predictive variables that were associated with subgroup memberships. Simulation experiments showed that the proposed method averagely had higher accuracy in identifying the subpopulations than the Naïve Bayes classifier and logistic regression. A breast cancer dataset was analyzed to illustrate the proposed hybrid mixture model.