Estimation of disease progression parameters from case-control data: application to mammographic patterns and breast cancer natural history

A method to determine the mammographic regions that show early changes due to the development of breast cancer

It is well understood nowadays that changes in the mammographic parenchymal pattern are an indicator of a risk of breast cancer and we have developed a statistical method that estimates the mammogram regions where the parenchymal changes, due to breast cancer, occur. This region of interest is computed from a score map by utilising the anatomical breast coordinate system developed in our previous work. The method also makes an automatic scale selection to avoid overfitting while the region estimates are computed by a nested cross-validation scheme. In this way, it is possible to recover those mammogram regions that show a significant difference in classification scores between the cancer and the control group. Our experiments suggested that the most significant mammogram region is the region behind the nipple and that can be justified by previous findings from other research groups. This result was conducted on the basis of the cross-validation experiments on independent training, validation and testing sets from the case-control study of 490 women, of which 245 women were diagnosed with breast cancer within a period of 2-4 years after the baseline mammograms. We additionally generalised the estimated region to another, mini-MIAS study and showed that the transferred region estimate gives at least a similar classification result when compared to the case where the whole breast region is used. In all, by following our method, one most likely improves both preclinical and follow-up breast cancer screening, but a larger study population will be required to test this hypothesis.

Percentage density, Wolfe's and Tabár's mammographic patterns: agreement and association with risk factors for breast cancer

Introduction

The purpose of this report was to classify mammograms according to four methods and to examine their agreement and their relationship to selected risk factors for breast cancer.

Method

Mammograms and epidemiological data were collected from 987 women, aged 55 to 71 years, attending the Norwegian Breast Cancer Screening Program. Two readers each classified the mammograms according to a quantitative method (Cumulus or Madena software) and one reader according to two qualitative methods (Wolfe and Tabár patterns). Mammograms classified in the reader-specific upper quartile of percentage density, Wolfe's P2 and DY patterns, or Tabár's IV and V patterns, were categorized as high-risk density patterns and the remaining mammograms as low-risk density patterns. We calculated intra-reader and inter-reader agreement and estimated prevalence odds ratios of having high-risk mammographic density patterns according to selected risk factors for breast cancer.

Results

The Pearson correlation coefficient was 0.86 for the two quantitative density measurements. There was moderate agreement between the Wolfe and Tabár classifications (Kappa = 0.51; 95% confidence interval 0.46 to 0.56). Age at screening, number of children and body mass index (BMI) showed a statistically significant inverse relationship with high-risk density patterns for all four methods (all P < 0.05). After adjustment for percentage density, the Wolfe classification was not associated with any of the risk factors for breast cancer, whereas the association with number of children and BMI remained statistically significant for the Tabár classification. Adjustment for Wolfe or Tabár patterns did not alter the associations between these risk factors and percentage mammographic density.

Conclusion

The four assessments methods seem to capture the same overall associations with risk factors for breast cancer. Our results indicate that the quantitative methods convey additional information over the qualitative methods.

Breast density and breast cancer risk factors in a high-risk population

This study investigates relationships between familial and hormonal risk factors and breast density in women at high risk of developing breast cancer. The subjects are a subset of 102 women from the international breast cancer intervention study (IBIS), for whom a series of repeated measurements of breast density were available. Details of familial and hormonal risk factors for breast cancer were collected at entry and multivariate ordered logistic regression used to identify risk factors for increased breast density. Lower body mass index and nulliparity were associated with high breast density, whereas smoking was associated with lower breast density. It is not yet known whether a reduction in breast density will lead to a corresponding reduction in breast cancer risk, so we propose that changes in breast density be investigated as a potential early indicator of efficacy in chemoprevention trials for breast cancer.

Deterministic models for breast cancer progression: application to the association between mammographic parenchymal pattern and histologic grade of breast cancers

Dense mammographic patterns are a risk factor for breast cancer. In a recent case-control study, dense patterns were observed to be associated with a higher relative risk estimate for the more aggressive grade 3 tumors than for grade 1 or 2 tumors. The present study was designed to ascertain, by reference to odds ratio estimates of relative risk by tumor size and histologic grade, whether this was more likely to be due to increased innate predisposition of dense parenchymal patterns to grade 3 tumors, or to dedifferentiation of cases missed at screening due to dense patterns. If the observation were due entirely to dedifferentiation, then for a given size the relative risk of grade 3 tumors would be the same as for grades 1-2. Size-specific relative risks of grade 3 tumors were higher than those of grade 1-2 tumors. It is concluded that the majority of the enhanced risk of grade 3 tumors is due to enhanced innate predisposition to grade 3 tumors in dense breasts.