Is the biology of breast cancer changing? A study of hormone receptor status and grade of breast cancers 1984-1986 and 1996-97.

Abstract

Abstract Abstract #3088 Introduction: There has been a significant improvement in breast cancer survival in the UK in recent decades. Changes in the molecular epidemiology of breast cancers may have contributed to this, but the existing evidence may be confounded by heterogeneity of laboratory protocols. This study aimed to re-analyse the molecular profile of breast cancers from two different time periods using archived tissue. Methods: Archived tumour samples from all breast cancer patients at two Glasgow hospitals between 1984-86 and 1996-97 were sought, and linked to clinicopathologic, screening, demographic and survival data. Patients in the 1984-6 cohort would not have been offered mammographic screening but those in 1996-97 would have. Samples were placed in tissue microarrays and underwent immunohistochemistry for ER, PR and Her-2 status with strict standardisation. H&E sections were constructed to assess tumour grade. Statistical analysis included Kaplan-Meier survival analysis and Cox's regression. Results: 900 tumour samples underwent staining. In 1984-86, 8% of tumours were grade 1 and 42.9% grade 3 but in 1996-97 14.9% were grade 1 and 36.8% grade 3 (p=0.009). This effect appeared to be exerted by the presence of screen detected tumours in 1996-97 (p for difference in grade distribution between symptomatic patients between 1984-86 and 1996-97 = 2). In 1984-86 64.2% of tumours were ER positive and in 1996-97 71.5% were ER positive (p=0.042). This did not appear to be a function of the screening programme as there was a significant rise in ER positivity in symptomatic patients between the two cohorts (p=0.024). 44.9% of tumours in 1984-86 and 49.9% of tumours in 1996-97 were PR positive (p=0.181). 21.5% of tumours in 1984-86 and 20.6% of tumours in 1996-97 were Her-2 positive (p=0.772). 5-year survival in 1984-1986 patients was significantly lower than in 1996-1997 patients (p<0.001). When the effect of cohort on survival was adjusted for these changes in ER status and grade, cohort remained a significant independent factor. Conclusions: This study suggests a small but significant rise has occurred in the incidence of ER positive tumours in women in Glasgow. There has also been a shift in grade distribution of tumours, which is likely to be an effect of the NHS screening programme. The changes do not fully explain improvements in breast cancer survival but should be borne in mind when applying the results of clinical trials performed in the past to the women of today. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 3088.