Introduction. Choice of adjuvant systemic therapy in early breast cancer patients followed breast conserving surgery depends on many prognostic factors especially from steroid (estrogen receptor — ER; progesterone receptor — PR) and HER-2 receptor status.

Purpose. To evaluate the treatment we determined disease free survival (DFS) and the risk of local recurrence and examined the influence of classical prognostic factors with special consideration of the biological subtype of breast cancer on DFS before the era of trastuzumab treatment. Patients and methods. Consecutively 615 female patients with early invasive breast cancer received breast conservative treatment between 2003 and 2006 in the Oncological Center in Warsaw. Data were prospectively collected. Adjuvant systemic therapy of second and third generation in 40% of patients had been given and in 28% of patients hormonotherapy was applied. After surgery and chemotherapy, irradiation with mild hypofractionation during 3 or 4 weeks was used. The following prognostic factors were included in the study: age, menopausal status, breast laterality, pT, pN, histology, grade, EIC, margins, and four biological subtypes: Luminal (ER positive and/or PR positive) HER-2 negative, Luminal (ER positive and/or PR positive) HER-2 positive, Triple-Negative, Non-Luminal (ER positive and/or PR positive) HER-2 positive. Survival curves were obtained using the Kaplan Maier method. To analyse time to recurrence, the competing risk method was performed. To study the influence of prognostic factors on DFS the proportional hazards model of Cox was used. The median follow-up time was 8 years.

Results. The 8-year DFS and cumulative loco-regional recurrence (CLRR) rate were 89% and 4.6% respectively. The significant factors influencing DFS were: young age of patients, number of involved nodes above three and grade 3 histological malignancy. Biological subtypes of breast cancer were not significant predictors for DFS in the univariate or multivariate analysis — logrank test: p = 0.19. It was shown, however, the probability of occurrence of the negative trend DFS for biological subtypes in the following order: Luminal HER2 (negative), Luminal HER2 (positive), Triple-negative and Nonluminal HER2 (positive) — logrank test for trend: p = 0.03. The analysis examined the distribution of prognostic factors and confirmed that in biological subtype Triple-negative and Nonluminal HER2 (positive) significantly more often than in the types of Luminal HER2 (negative) and Luminal HER2 (positive), were younger patients, with larger tumour, with more than 3 involved nodes and grade 3 histological malignancy.

Conclusions. 1. High probability of 8-year DFS and low CLRR rate of breast cancer indicated a positive assessment of conserving therapy at the Cancer Center in Warsaw between 2003–2006. 2. Despite aggressive treatment the strongest prognostic factors still remain: the young age of patients, the number of involved lymph node in the axillary fossa greater than three and low differentiation of cancer G3. 3. The worst prognosis is for patients diagnosed with breast cancer in subtype Nonluminal HER2 (positive), and this justifies the introduction of molecular targeted therapies aimed at HER2.