Statistical Analysis of the Comet Assay Using a Mixture of Gamma Distributions

Tail moments in the single cell gel electrophoresis (comet) assay usually do not follow a normal distribution, making the statistical analysis complicated. Researchers have used a wide variety of statistical techniques in an attempt to overcome this problem. In many cases, the tail moments follow a bimodal distribution that can be modeled with a mixture of gamma distributions. This bimodality may be due to cells being in two different stages of the cell cycle at the time of treatment. Maximum likelihood, modified to accommodate censored data, can be used to estimate the five parameters of the gamma mixture distribution for each slide. A weighted analysis of variance on the parameter estimates for the gamma mixtures can be performed to determine differences in DNA damage between treatments. These methods were applied to an experiment on the effect of thymidine kinase in DNA damage and repair. Analysis based on the mixture of gamma distributions was found to be more statistically valid, more powerful, and more informative than analysis based on log-transformed tail moments.     

In vitro detection of indirect-acting genotoxins in the comet assay using Hep G2 cells

The induction of DNA damage by four known promutagens (cyclophosphamide (CP), benzo(a)pyrene (BP), dimethylbenz(a)anthracene and 2-acetylaminofluorene (2AAF) was investigated on Hep G2 using the alkaline single cell electroporesis (SCGE) test, most often referred as the "comet assay". After a 3-day incubation, lysed cells embedded in agarose were electrophoresed under alkaline conditions, dyed with a SYBRgold fluorogen and analysed by the Komet software. Among the comet parameters provided by the image analysis program, statistical analysis did not identify any in particular that could best represent the DNA damages. All promutagens, when compared with the control, caused a statistically significant increase in DNA migration as determined by different parameters such as Olive tail moment, tail extent moment, tail/head or tail length. The data demonstrated the ability and the sensitivity of the comet assay when performed on Hep G2 in the detection of DNA damage induced by promutagens, and its suitability in mutagenicity testing in in vitro short-term assays.     

Identification by microscopically controlled comet assay of peritoneal macrophages in a mixture of peritoneal exudate for DNA strand break analysis

A simple modification of the alkaline comet assay allows the study of DNA damage in a specific cell type in a mixture of primary cells. Peritoneal macrophages from mice are selected from other peritoneal exudate cells without complex preparation and separation steps by their size and shape of the nuclei and their comets. The DNA damage can be well characterised by the manually monitored parameter 'tail length'. Complex measurement of the 'tail moment', often used for characterising DNA damage is not required, a fact which further simplifies the protocol. The distribution of tail length within one sample is symmetric and can be described by a Gaussian distribution and the mean tail length. As a first application, UV-A sensitivity of resident and stimulated macrophages was studied. The resident macrophages were more sensitive to UV-A than the stimulated ones. DNA damage repair follows the same simple monoexponential time course for both cell types. The simplicity of results, i.e., applicability of tail lengths and Gaussian statistics as well as monoexponential kinetics, suggest that microscopically controlled comet assay is well suited to study elementary processes of DNA damage induction and repair.     

Is there an influence of enzyme polymorphisms on ochratoxin A genotoxicity?

Primary cultured human urothelial cells derived from ureter specimens of urological patients were used to evaluate induction of DNA-damage by OTA in the alkaline single-cell gel electrophoresis (comet) assay. With the cultured cells from each donor a separate comet assay was performed and tail length of the damaged DNA was measured. A broad spectrum of effects was detected between the individual cell cultures with effects reaching from tail lengths on control level up to strongly enhanced tail lengths.All donors of urothelial tissue were additionally genotyped for several xenobiotic metabolising enzymes (cytochrome P450 1A2, glutathione S-transferases T1, M1, and P1, N-acetyltransferase 2) in lymphocyte DNA. The genotype was then correlated with the genotoxic effects obtained in the comet assay.No correlation was found with CYP1A2, GSTT1, and GSTM1 genotypes whereas for GSTP1 stronger genotoxic effects were found in cells from donors with hetero-and homozygously mutated (w/m, m/m) genotypes compared to homozygous wildtypes. The strongest hint for a correlation was found for NAT2, as cells from donors with homozygous mutated alleles (m/m), known as slow acetylators, displayed a higher susceptibility to OTA in the comet assay than cells from donors with the heterozygously mutated or wildtype alleles (rapid acetylators).     

Use of the single cell gel electrophoresis/comet assay for detecting DNA damage in aquatic (marine and freshwater) animals

The comet assay is a rapid, sensitive and inexpensive method for measuring DNA strand breaks. The comet assay has advantages over other DNA damage methods, such as sister chromatid exchange, alkali elution and micronucleus assay, because of its high sensitivity and that DNA strand breaks are determined in individual cells. This review describes a number of studies that used the comet assay to determine DNA strand breaks in aquatic animals exposed to genotoxicants both in vitro and in vivo, including assessment of DNA damage in aquatic animals collected from contaminated sites. One difficulty of using the comet assay in environmental work is that of comparing results from studies that used different methods, such as empirical scoring or comet tail lengths. There seems to be a consensus in more recent studies to use both the intensity of the tail and the length of the tail, i.e. DNA tail moment, percentage of DNA in the tail. The comet assay has been used to assess DNA repair and apoptosis in aquatic animals and modifications of the comet assay have allowed the detection of specific DNA lesions. There have been some recent studies to link DNA strand breaks in aquatic animals to effects on the immune system, reproduction, growth, and population dynamics. Further work is required before the comet assay can be used as a standard bio-indicator in aquatic environments, including standardization of methods (such as ASTM method E2186-02a) and measurements.     

The use of cyprinodont fish, Aphanius fasciatus, as a sentinel organism to detect complex genotoxic mixtures in the coastal lagoon ecosystem

In the present work we aimed to standardise the alkaline comet assay with erythrocytes of the cyprinodont, Mediterranean Killifish, Aphanius fasciatus. The aims of the study were to explore the suitability of this fish to assess biomarkers of genotoxic effects and as a sentinel organism to detect complex genotoxic mixtures in coastal lagoon ecosystems. Following proper optimisation, the application and effectiveness of the comet assay in erythrocytes of A. fasciatus were tested by measuring the tail DNA (%) induced by (a) in vivo exposure of individual fish to X-rays (dose, 3Gy) and (b) following in vitro challenge of erythrocytes with restriction endonucleases Fok-I and Eco-RI, which selectively induce double-strand breaks with cohesive and blunt termini, respectively. Furthermore, in order to evaluate whether circulating fish blood contained actively proliferating cells that could influence the extent of DNA damage in control (untreated) fish, we measured the number of "comets" positive for 5-bromo-2'-deoxyuridine (BrdU) by the use of anti-BrdU antibody and immuno-histochemical methods. Both treatments (i.e. with X-rays and restriction endonucleases) induced statistically significant increases in tail DNA (%) values compared with the relevant untreated controls, indicating the effectiveness of the comet assay in the erythrocytes of A. fasciatus to detect different types of DNA lesions. Results from anti-BrdU antibody labelling of erythrocytes indicated a very low percentage (5%) of "comets" positive for BrdU. Following optimisation and validation of the assay under laboratory conditions, fish were collected in the Orbetello lagoon (Tuscany, Italy), considered to be a significantly polluted site. The results showed statistically significant increases for tail DNA (%) compared with corresponding values observed in erythrocytes of fish caught in the unpolluted reference site "Saline di Tarquinia". The effects of physico-chemical parameters of the water (i.e., salinity, pH and oxygen content) did not significantly influence the induction of DNA damage. These results indicate that the comet assay provides a reliable parameter and that A. fasciatus is a promising "sentinel organism" to detect the genotoxic impact of complex mixtures in coastal lagoon ecosystems.     

Comet assay: rapid processing of multiple samples

The present study describes modifications to the basic comet protocol that increase productivity and efficiency without sacrificing assay reliability. A simple technique is described for rapidly preparing up to 96 comet assay samples simultaneously. The sample preparation technique allows thin layers of agarose-embedded cells to be prepared in multiple wells attached to a flexible film of Gelbond, which improves the ease of manipulating and processing samples. To evaluate the effect of these modifications on assay sensitivity, dose-response curves are presented for DNA damage induced by exposure of TK6 cells to low concentrations of hydrogen peroxide (0-10 microM) and for exposure of human lymphocytes to X-irradiation (0-100 cGy). The limit of detection of DNA damage induced by hydrogen peroxide in TK6 cells was observed to be 1 uM for all parameters (tail ratio, tail moment, tail length and comet length) while the limit of detection of DNA damage in human lymphocytes was 10 cGy for tail and comet length parameters, but 50 cGy for tail ratio and tail moment parameters. These results are similar to those previously reported using the conventional alkaline comet assay. The application of SYBR Gold for detection of DNA damage was compared to that of propidium iodide. Measurements of matching samples for tail length and comet length were similar using both stains. However, comets stained with SYBR Gold persisted longer and were much brighter than those obtained with propidium iodide. SYBR Gold was found to be ideal for measuring tail length and comet length but, under present assay conditions, impractical for measuring tail ratio or tail moment due to saturation of staining in the head region of the comets.     

Assessment of genotoxic effects of chloropyriphos and acephate by the comet assay in mice leucocytes

Two organophosphorus (OP) pesticides (chloropyriphos and acephate) and cyclophosphamide (CP) (positive control) were tested for their ability to induce in vivo genotoxic effect in leucocytes of Swiss albino mice using the single cell gel electrophoresis assay or comet assay. The mice were administered orally with doses ranging from 0.28 to 8.96 mg/kg body weight (b. wt.) of chloropyriphos and 12.25 to 392.00 mg/kg b.wt. of acephate. The assay was performed on whole blood at 24, 48, 72 and 96 h. A significant increase in mean comet tail length indicating DNA damage was observed at 24h post-treatment (P<0.05) with both pesticides in comparison to control. The damage was dose related. The mean comet tail length revealed a clear dose dependent increase. From 48 h post-treatment, a gradual decrease in mean tail length was noted. By 96 h of post-treatment the mean comet tail length reached control levels indicating repair of the damaged DNA. From the study it can be concluded that the comet assay is a sensitive assay for the detection of genotoxicity caused by pesticides.     

Tail profile: a more accurate system for analyzing DNA damage using the Comet assay

The Comet assay (single cell gel electrophoresis assay) measures DNA strand breaks in individual cells. In the assay cells are embedded in agarose, lysed, and electrophoresed under low voltage, allowing migration of damaged DNA. The DNA is stained and subsequently viewed with an epifluorescent microscope. If DNA damage has occurred the electrophoresed DNA fragments appear as a diffuse tail behind the nucleus known as a "comet". Many computer-aided analysis systems are currently in use to quantify the amount of DNA damage that is represented by a comet image. Here, we present a novel method of analysis known as "tail profile". This method of analysis provides several advantages over currently employed methods, which rely primarily on the "tail moment" method of analysis. We compared the amount of DNA damage reported from both the tail profile and tail moment methods of analysis and observed a 26% (P<0.0001) increase in damage detected by tail profile across the 10-25 microm range of tail length, where the majority of the relevant comet data is concentrated. We further report that this increase in sensitivity is not only limited to assessing DNA damage, but also to gathering data from DNA repair assays. Furthermore, we demonstrate increased functionality and extended data analysis capabilities with the use of a compressed collection of images called a "comet chip" and through a visual representation of data called a "profile plot". Use of the custom macros enabled us to detect an unexpected characteristic of the electrophoretic profile, giving us novel insight into the nature of comet analysis. In addition to the increased analytical sensitivity proffered by this system, the tail profile macros are upgradeable and platform independent.     

Enhanced prediction of potential rodent carcinogenicity by utilizing comet assay and apoptotic assay in combination

The comet assay has been recently validated as a sensitive and specific test system for the quantification of DNA damage. The objectives of this study are to investigate the utility of comet assay for detecting mutagens with 11 substances that demonstrated positive results in at least one test among four standard short-term genotoxicity tests, and to evaluate its ability to predict rodent carcinogenicity. Out of 11 test substances, positive comet results were obtained for colchicine, hydroxyurea and actinomycin D. No effect on DNA migration, determined as the tail moment, was found with theophylline or 2,4-dinitrophenol. Bisphenol A, vinblastine, paclitaxel and p-anisidine appeared cytotoxic clastogens because these induced tail moment at concentrations showing 60% or less cell survival. In addition, among three test substances showing the bimodal distribution of DNA damage, which is a characteristic of apoptosis, true apoptosis result was obtained for camptothecin and dexamethasone with the Annexin V affinity assay. With this limited data-set, an investigation into the predictive value of these short-term genotoxicity tests for determining the carcinogenicity showed that comet assay has relatively high sensitivity and superior specificity to other four short-term genotoxicity assay. Therefore, our data suggest that comet assay, especially in combination with apoptotic assay, would be a good predictive test to minimize false-positives in evaluation of the potential rodent carcinogenicity.     

Validation and implementation of an internal standard in comet assay analysis

The comet assay is widely used to detect DNA damage in single cells. However, only moderate attention has been paid to the experimental variability of this assay, especially during electrophoresis. To take into account this variation and to be able to compare measurements from different electrophoretic runs, as would be necessary when large numbers of samples need to be analysed, it is important to integrate an internal standard into the assay. This study presents a first step in the validation and implementation of an internal standard in the alkaline comet assay. Untreated and ethyl methanesulfonate treated cells (K562 human erythroleukemia cell line) were used as negative and positive internal standards, respectively, in each electrophoresis run. Three steps were followed: (1) assessment of the different levels of variability which may influence the damage levels of the internal standards, (2) evaluation of the variability across separate electrophoresis runs on the quantification of DNA damage in the internal standards by three experimenters involved in different studies and (3) proposal of an adequate calculation system to integrate the internal standards into test sample data. The application of the two proposed models to samples from a human biomonitoring study is presented. The model which calibrates the measurements against the negative internal standard is the most useful since this negative standard was the most stable across experiments and among the three experimenters. The percentage of DNA in the tail is the most appropriate parameter to analyse induced DNA damage, because its interelectrophoresis and interexperimenter variation is less pronounced than that of tail length.     

Primary DNA damage in human blood lymphocytes exposed in vitro to 2450 MHz radiofrequency radiation

Human peripheral blood samples collected from three healthy human volunteers were exposed in vitro to pulsed-wave 2450 MHz radiofrequency (RF) radiation for 2 h. The RF radiation was generated with a net forward power of 21 W and transmitted from a standard gain rectangular antenna horn in a vertically downward direction. The average power density at the position of the cells in the flask was 5 mW/cm(2). The mean specific absorption rate, calculated by finite difference time domain analysis, was 2.135 (+/-0.005 SE) W/kg. Aliquots of whole blood that were sham-exposed or exposed in vitro to 50 cGy of ionizing radiation from a (137)Cs gamma-ray source were used as controls. The lymphocytes were examined to determine the extent of primary DNA damage (single-strand breaks and alkali-labile lesions) using the alkaline comet assay with three different slide-processing schedules. The assay was performed on the cells immediately after the exposures and at 4 h after incubation of the exposed blood at 37 +/- 1 degrees C to allow time for rejoining of any strand breaks present immediately after exposure, i.e. to assess the capacity of the lymphocytes to repair this type of DNA damage. At either time, the data indicated no significant differences between RF-radiation- and sham-exposed lymphocytes with respect to the comet tail length, fluorescence intensity of the migrated DNA in the tail, and tail moment. The conclusions were similar for each of the three different comet assay slide-processing schedules examined. In contrast, the response of lymphocytes exposed to ionizing radiation was significantly different from RF-radiation- and sham-exposed cells. Thus, under the experimental conditions tested, there is no evidence for induction of DNA single-strand breaks and alkali-labile lesions in human blood lymphocytes exposed in vitro to pulsed-wave 2450 MHz radiofrequency radiation, either immediately or at 4 h after exposure.     

Estimators and tests in the analysis of multiple nonindependent 2 x 2 tables with partially missing observations

We present methods for the analysis of a K-variate binary measure for two independent groups where some observations may be incomplete, as in the case of K repeated measures in a comparative trial. For the K 2 X 2 tables, let theta = (theta 1,..., theta K) be a vector of association parameters where theta k is a measure of association that is a continuous function of the probabilities pi ik in each group (i = 1, 2; k = 1,..., K), such as the log odds ratio or log relative risk. The asymptotic distribution of the estimates theta = (theta 1,..., theta K) is derived. Under the assumption that theta k = theta for all k, we describe the maximally efficient linear estimator theta of the common parameter theta. Tests of contrasts on the theta are presented which provide a test of homogeneity Ha: theta k = theta l for all k not equal to l. We then present maximally efficient tests of aggregate association Hb: theta = theta 0, where theta 0 is a given value. It is shown that the test of aggregate association Hb is asymptotically independent of the preliminary test of homogeneity Ha. These methods generalize the efficient estimators of Gart (1962, Biometrics 18, 601-610), and the Cochran (1954, Biometrics 10, 417-451), Mantel-Haenszel (1959, Journal of the National Cancer Institute 22, 719-748), and Radhakrishna (1965, Biometrics 21, 86-98) tests to nonindependent tables. The methods are illustrated with an analysis of repeated morphologic evaluations of liver biopsies obtained in the National Cooperative Gallstone Study.     

Statistical evaluation and comparison of comet assay results

A method of critical evaluation of comet assay test results is proposed. This method is based on representing frequency histograms of tail moments and %DNA in the tail with the use of the unsymmetrical Weibull distribution. The distribution is fully characterized by two parameters: shape (alpha) and scale (beta). Coefficients of variation and confidence intervals of the parameters have been determined.     

Statistical inference for simultaneous clustering of gene expression data

Current methods for analysis of gene expression data are mostly based on clustering and classification of either genes or samples. We offer support for the idea that more complex patterns can be identified in the data if genes and samples are considered simultaneously. We formalize the approach and propose a statistical framework for two-way clustering. A simultaneous clustering parameter is defined as a function theta=Phi(P) of the true data generating distribution P, and an estimate is obtained by applying this function to the empirical distribution P(n). We illustrate that a wide range of clustering procedures, including generalized hierarchical methods, can be defined as parameters which are compositions of individual mappings for clustering patients and genes. This framework allows one to assess classical properties of clustering methods, such as consistency, and to formally study statistical inference regarding the clustering parameter. We present results of simulations designed to assess the asymptotic validity of different bootstrap methods for estimating the distribution of Phi(P(n)). The method is illustrated on a publicly available data set.     

Just-identified versus overidentified two-level hierarchical linear models with missing data

The development of model-based methods for incomplete data has been a seminal contribution to statistical practice. Under the assumption of ignorable missingness, one estimates the joint distribution of the complete data for thetainTheta from the incomplete or observed data y(obs). Many interesting models involve one-to-one transformations of theta. For example, with y(i) approximately N(mu, Sigma) for i= 1, ... , n and theta= (mu, Sigma), an ordinary least squares (OLS) regression model is a one-to-one transformation of theta. Inferences based on such a transformation are equivalent to inferences based on OLS using data multiply imputed from f(y(mis) | y(obs), theta) for missing y(mis). Thus, identification of theta from y(obs) is equivalent to identification of the regression model. In this article, we consider a model for two-level data with continuous outcomes where the observations within each cluster are dependent. The parameters of the hierarchical linear model (HLM) of interest, however, lie in a subspace of Theta in general. This identification of the joint distribution overidentifies the HLM. We show how to characterize the joint distribution so that its parameters are a one-to-one transformation of the parameters of the HLM. This leads to efficient estimation of the HLM from incomplete data using either the transformation method or the method of multiple imputation. The approach allows outcomes and covariates to be missing at either of the two levels, and the HLM of interest can involve the regression of any subset of variables on a disjoint subset of variables conceived as covariates.     

Validation of an automatic comet assay analysis system integrating the curve fitting of combined comet intensity profiles.

In recent years, the single-cell gel electrophoresis (comet) assay has become a reference technique for the assessment of DNA fragmentation both in vitro and in vivo at the cellular level. In order to improve the throughput of genotoxicity screening, development of fully automated systems is clearly a must. This would allow us to increase processing time and to avoid subjectivity brought about by frequent manual settings required for the 'classical' analysis systems. To validate a fully automatic system developed in our laboratory, different experiments were conducted in vitro on murine P388D1 cells with increasing doses of ethyl methanesulfonate (up to 5 mM), thus covering a large range of DNA damage (up to 80% of DNA in the tail). The present study (1) validates our 'in house' fully automatic system versus a widely used semi-automatic commercial system for the image-analysis step, and versus the human eye for the image acquisition step, (2) shows that computing tail DNA a posteriori on the basis of a curve fitting concept that combines intensity profiles [G. Dehon, P. Bogaerts, P. Duez, L. Catoire, J. Dubois, Curve fitting of combined comet intensity profiles: a new global concept to quantify DNA damage by the comet assay, Chemom. Intell. Lab. Syst. 73 (2004) 235-243] gives results not significantly different from the 'classical' approach but is much more accurate and easy to undertake and (3) demonstrates that, with these increased performances, the number of comets to be scored can be reduced to a minimum of 20 comets per slide without sacrificing statistical reliability.     

DNA damage and micronucleus induction in human leukocytes after acute in vitro exposure to a 1.9 GHz continuous-wave radiofrequency field

Human blood cultures were exposed to a 1.9 GHz continuous-wave (CW) radiofrequency (RF) field for 2 h using a series of six circularly polarized, cylindrical waveguides. Mean specific absorption rates (SARs) of 0.0, 0.1, 0.26, 0.92, 2.4 and 10 W/kg were achieved, and the temperature within the cultures during a 2-h exposure was maintained at 37.0 +/- 0.5 degrees C. Concurrent negative (incubator) and positive (1.5 Gy (137)Cs gamma radiation) control cultures were run for each experiment. DNA damage was quantified immediately after RF-field exposure using the alkaline comet assay, and four parameters (tail ratio, tail moment, comet length and tail length) were used to assess DNA damage for each comet. No evidence of increased primary DNA damage was detected by any parameter for RF-field-exposed cultures at any SAR tested. The formation of micronuclei in the RF-field-exposed blood cell cultures was assessed using the cytokinesis-block micronucleus assay. There was no significant difference in the binucleated cell frequency, incidence of micronucleated binucleated cells, or total incidence of micronuclei between any of the RF-field-exposed cultures and the sham-exposed controls at any SAR tested. These results do not support the hypothesis that acute, nonthermalizing 1.9 GHz CW RF-field exposure causes DNA damage in cultured human leukocytes.     

Laser scanning cytometry for comet assay analysis

BACKGROUND: The comet assay (single-cell gel electrophoresis) is a sensitive method for evaluating nuclear DNA damage. Previously used evaluation methods for the comet assay are time consuming and have an inherent risk of biased selection of comets due to manual selection and categorization of comet images. Laser scanning cytometry (LSC), the principle of which is equivalent to flow cytometry, enables quantification of fluorescence emitted from the cells on a microscope slide. In the present study, we explored whether LSC could be used to determine the degree of DNA damage demonstrated by the comet assay. METHODS: DNA damage was induced by ultraviolet A irradiation of keratinocytes and visualized by the comet assay. The evaluation included (a) LSC determination of DNA-specific fluorescence in 1,000 comet heads (undamaged DNA), (b) image acquisition of comets by rescanning of the microscope slide, and (c) digital image analysis and computation of tail moment and DNA content in the comet tails. RESULTS: Cells with damaged DNA were observed in a sub-G(1) area because the comet head loses DNA to the tail. We found a strong inverse correlation between tail moment and DNA content per nucleus. CONCLUSIONS: LSC enables an automated method for cell recognition and evaluation of the comets, thus providing quantitative information about nuclear DNA damage without subjective selection of analyzed comets.     

Recommendations for safety testing with the in vivo comet assay

While the in vivo comet assay increases its role in regulatory safety testing, deliberations about the interpretation of comet data continue. Concerns can arise regarding comet assay publications with limited data from non-blind testing of positive control compounds and using protocols (e.g. dose concentrations, sample times, and tissues) known to give an expected effect. There may be a tendency towards bias when the validation or interpretation of comet assay data is based on results generated by widely accepted but non-validated assays. The greatest advantages of the comet assay are its sensitivity and its ability to detect genotoxicity in tissues and at sample times that could not previously be evaluated. Guidelines for its use and interpretation in safety testing should take these factors into account. Guidelines should be derived from objective review of data generated by blind testing of unknown compounds dosed at non-toxic concentrations and evaluated in a true safety-testing environment, where the experimental design and conclusions must be defensible. However, positive in vivo comet findings with such compounds are rarely submitted to regulatory agencies and this data is typically unavailable for publication due to its proprietary nature. To enhance the development of guidelines for safety testing with the comet assay, and with the permission of several sponsors, this paper presents and discusses relevant data from multiple GLP comet studies conducted blind, with unknown pharmaceuticals and consumer products. Based on these data and the lessons we have learned through the course of conducting these studies, I suggest significant adjustments to the current conventions, and I provide recommendations for interpreting in vivo comet assay results in situations where risk must be evaluated in the absence of carcinogenicity or clinical data.