Analysis of tumor incidence

TABLE 9.5. Interpretation of the Analysis of Tumor Incidence and Survival Analysis (Life Table)... 

The methods we advocate for routine use for the analysis of tumor incidence tend, therefore, not to be based on the use of formal parametric statistical models. For example, when studying the relationship of treatment to incidence of a pathological condition and wishing to adjust for other factors (in particular, age at death) that might otherwise bias the comparison, methods involving stratification are... 

In 1985, the United States Federal Register recommended that the analysis of tumor incidence data be carried out with a Cochran-Armitage (Armitage, 1955 Cochran, 1954) trend test. The test statistic of the Cochran-Armitage test is defined as this term ... 

Figure 3 Liver tumors (o) and DNA adducts ( ) induced by AFB1. Male Fischer rats were exposed to five doses of AFB1 between 1 and 50 ppb in the drinking water for 24 months for analysis of tumor incidence. DNA adducts were measured after eight weeks of continuous administration of AFB1 in the drinking water (17-19).

TABLE 29.6. Multisite Time-Dependent Analysis of Tumor Incidence in the NTP(1993) Study of Mice Exposed to 1,3-Butadiene... 

Tumors were not clearly or consistently observed in nonhepatic tissues of animals exposed to FireMaster FF-1. Induction of thyroid follicular cell adenoma was inconclusive in mice in both National Toxicology Program bioassays. Equivocal increases in incidences of mononuclear cell leukemia were observed in adult-only exposed rats in the NTP chronic study, and combined perinatal and adult exposure showed no significant increase. Combined analysis of the incidences of this leukemia in the adult-only, perinatal only. 

First, the highest appropriate unit cancer risk (UCR) is determined from the tumor data from the most sensitive species, strain, sex, and study (FDA, 2002). Tumor incidences not considered treatment related are omitted from analysis statistical significance is an important tool in evaluating the relationship of treatment to tumor induction. If only one dose was tested and only one type of tumor was induced, then the UCR is defined as the slope of the straight line of tumor incidence versus dose. 

Irrespective of the specific protocols used, all carcinogenicity studies end with a statistical comparison of tumor proportions between treated and control groups. This analysis is necessary because the control incidence of most tumor types is rarely zero. In the unlikely case that a type of tumor is found in treated animals but not in concurrent or appropriate historical controls, it is reasonable to conclude that the tumor is drug-related without statistical analysis. 

Although methods are available for including historical control data in the formal statistical analysis (Tarone, 1982 Dempster et al., 1983 Haseman, 1990), this is usually not done and for good reason. The heterogeneity of historical data requires that they be used qualitatively and selectively to aid in the final interpretation of the data, after completion of the formal statistical analysis. Table 9.3 presents a summary of background tumor incidences for the most commonly employed rodent strains. 

Patients with retinoblastoma suffer from a high incidence of tumors arising from clonal outgrowth of some retinal precursor cells due to mutation of the tumor suppressor gene RBI. Analysis of cells from these tumors indicates that both copies of the RBI gene are mutated or lost, whereas the surrounding retinal cells have at least one functional RBI allele. 

As the kidney tumors were not fatal, the appropriateness of Lifetable test for the analysis of these tumors is questionable. The overall unadjusted incidences were significantly different from the historical control incidence by the Fisher Exact test. The kidney tumors were not observed in female rats or in male or female mice. 

Trypsinization in excess of 1 min has been reported to decrease lung colony formation and should therefore be avoided (13). In other studies it has been shown that comparing the incidence of metastasis from the injection of one predetermined dose of tumor cells does not allow an analysis of their relative metastatic capacities. Reproducible and meaningful results require studies that introduce increasing numbers of viable tumor cells admixed with a constant number ofnon-tumorigenic (X-irradiated) carrier cells (15) (see Note 4). 

Validation The use of TMAs enables analysis of large data sets, however this does not by any means suggest that the data set is not skewed. This skewing may be the result of the institution s location (population distributions with regards to race, ethnicity, access to health care), type of practice (community hospital versus referral center). These collectively might influence the tumor size, grade and subtype composition of the cases in the dataset. Such abnormalities of the dataset need to be compensated the involvement of a biostatistician from the start (i.e from case selection) helps to prevent the creation of biased TMAs. It is useful to perform common biomarker analysis on sections from the created TMA to confirm the normal distribution of known parameters. Comparison of this data with prior clinical data (e.g. ER analysis) obtained from whole section analysis is particularly useful to validate utility of the TMA. Alternatively the incidence of expression of a number of biomarkers in the TMA should be compared to that in published literature (using whole sections). 

Statistical analysis is performed on all parameters in the study. Its most fundamental objective is to determine whether administration of the test agent results in an increase in tumor incidence rates as compared to those in unexposed controls. Various statistical methods can be used. Tests for increased tumor occurrence rates between dosages may be based on pair-wise comparisons, such as the Chi-square test for 2x2 tables, the Fisher s exact test, or the Cochan-Armitage test. These tests are most appropriate when survival rates do not differ appreciably in the various dose groups. 

Human health biomonitoring may also use animal surrogates in the environment to assess potential health hazards to humans - the proverbial canary in the coal mine . For example, chemical and biomarker analysis of bodily fluids or tissue biopsies from family pets, especially dogs, are sometimes used to assess potential chemical exposure and effects in children. This is because dogs often accompany children in the outdoor environment, and both have a tendency to (accidentally or intentionally) consume environmental media such as soil and surface water There has also been an increasing trend to use native animals as sentinel species, that is, fish, wildlife, or invertebrates that are indicators of possible human health risks from environmental hazards. For example, increased incidences of tumors or endocrine disruption in fish may indicate the presence of compounds in the water that may cause cancer or reproductive dysfunction in humans. Concern has also been raised over the increased incidence of deformities in frogs, because these may indicate an increased level of chemicals in the environment, which can cause birth defects in humans. 

Intermediate-duration topical application of benz[a]anthracene to the backs of mice for 30 weeks resulted in a slightly elevated (2.6%) (but not statistically significant) skin tumor incidence. No definitive conclusions can be drawn from this study since only one dose was employed and no statistical analysis was performed. 

---