Tamoxifen or Letrozole in Treating Women With Ductal Carcinoma in Situ

Change in size of Ductal Carcinoma in situ (DCIS) for participants on hormonal therapy, as determined by mammography are determined by (1) largest diameter of tumor, as visualized on mammography (2) extent of disease on mammography (3) quantification of mammographically-detected change from baseline to 6-month and used to generate the change in tumor volume of mammographic extent of disease from baseline. Since values were not normally distributed, the median change was calculated, and Wilcoxon sign rank tests were used to evaluate the significance of these changes

Change in size of Ductal Carcinoma in situ (DCIS) on hormonal therapy, as determined by MRI are determined by (1) largest diameter of tumor, as visualized on MRI (2) extent of disease on MRI (3) quantification of MR-detected change from baseline to 6-month and used to generate the change in tumor volume of MRI extent of disease from baseline. Since values were not normally distributed, the median change was calculated, and Wilcoxon sign rank tests were used to evaluate the significance of these changes.

MRI volume response at each time point was classified as follows: 90% image-complete response (ICR90) is defined as a >90% reduction in tumor volume, 80% image-complete response (ICR80) is defined as an 81-90% reduction in tumor volume , partial response (PR) is defined as a 20-80% reduction in tumor volume, and sustained disease or progressive disease (SD/PD) defined as a <20% reduction or increase in volume.

MRI volume response at each time point was classified as follows: 90% image-complete response (ICR90) is defined as a >90% reduction in tumor volume, 80% image-complete response (ICR80) is defined as an 81-90% reduction in tumor volume , partial response (PR) is defined as a 20-80% reduction in tumor volume, and sustained disease or progressive disease (SD/PD) defined as a <20% reduction or increase in volume.

Tumor volume changes between baseline and surgery were calculated at month 6 and compared across baseline ER Hormone (H-) Score quartile. The ER H- scores are a percentage that tells you how many cells out of 100 stain positive for hormone receptors. Each participant is assigned an ER H- score at baseline with the full score range between 0 (none have receptors) and 100 (all have receptors). The participants were grouped into quartiles (four equal groups) based on their baseline ER H- score. ER H- score and the reduction in tumor volume from baseline to month 6 was measured for each quartile group.

Tumor volume changes between baseline and surgery were calculated at month 6 and compared across baseline PgR Hormone (H-) Score quartile. The PgR H-scores are a percentage that tells you how many cells out of 100 stain positive for hormone receptors. Each participant is assigned a PgR H- score at baseline with the full PgR H score ranges between 0 (none have receptors) and 100 (all have receptors). The participants were grouped into quartiles (four equal groups) based on their baseline PgR H- score and the reduction in tumor volume from baseline to month 6 was measured for each quartile group.

A wilcoxon sign rank tests were used to evaluate the significance of these changes

Tumor volume changes between baseline and surgery were calculated at month 6 by Baseline Ki-67 Average Score which is divided into 2 groups: (1) <=10% or (2) >10% to 100%. In est results, the Ki-67 findings expressed as a percentage with less than 10% considered low Ki-67 expression and > than 10% or higher considered high. A "high" score means that the breast tumor is more likely to be aggressive and spread quickly. A wilcoxon sign rank tests were used to evaluate the significance of these changes

Correlations between pathologic tumor size and maximum diameters of baseline and 6-month MRI extent of disease were evaluated using Spearman correlation coefficient measure of association. The Spearman's rank-order correlation (rs) measures the strength and direction of association between two variables. The Spearman correlation coefficient, rs, can take values from +1 to -1 where a value of +1 indicates a perfect association, an rs of 0 indicates no association and an rs of -1 indicates a perfect negative association. The closer rs is to 0, the weaker the association.

Correlations between pathologic tumor size and maximum diameters of baseline and pre-surgical mammographic extent of disease were evaluated using Spearman correlation coefficient measure of association. The Spearman's rank-order correlation (rs) measures the strength and direction of association between two variables. The Spearman correlation coefficient, rs, can take values from +1 to -1 where a value of +1 indicates a perfect association, an rs of 0 indicates no association and an rs of -1 indicates a perfect negative association. The closer rs is to 0, the weaker the association.