High-grade ER-negative tumour breast cancers are characteristic of both very young onset cases and patients with hereditary breast cancer

The Prospective Study of Treatment Outcomes in Sporadic versus Hereditary Breast Cancer (POSH) will have recruited 2,000 women over a 5-year interval from over 100 participating UK centres who have newly diagnosed breast cancer before age 41 years.

The first 1,200 cases from the study in whom diagnostic pathology reports were submitted were analysed. We looked at the distribution of the reported tumour phenotype (major prognostic histopathological features) in women aged ≤35 years (43% of the total cohort) compared with women diagnosed age 35–40 years in order to further explore biological explanations for the known worse clinical prognosis for women aged under 36 years compared with older women. The χ² statistic was used to compare groups; genetic risk for each recruit was derived using software that incorporates a general genetic model rather than a gene specific model. The highest genetic risk groups are likely to harbour most of the BRCA1 and BRCA2 gene carriers.

The majority of women at all ages were treated with anthracycline-based adjuvant chemotherapy and there was no difference in the choice of immediate surgical management between either age groups or between genetic risk groups.

Significantly more women in the ≤35 years age group had grade 3 (P < 0.01) and ER-negative (P < 0.02) tumours compared with women diagnosed in the older age group (>35 but ≤40 years). There was no significant difference in tumour size or lymph node status based on age categories. Compared with women with no family history, women falling into the 10% of the cohort estimated from family history to be most likely to carry BRCA1 or BRCA2 gene mutations, high genetic risk women had significantly more grade 3 tumours (P < 0.001) and a nonsignificant trend towards more ER-negative tumours.

These data are from a preliminary pending systematic pathology review but bear out the observations by others that very young age of onset and host genotype affect the tumour phenotype and are therefore likely to have an impact on prognosis. Longer follow-up of this cohort is planned and outcome data based on age and based on genetic risk category and genetic mutation status will be available in a further 12 months time.