[P4-17-05] Brain Metastasis Free Survival (BMFS) Differs between Breast Cancer Subtypes.

Bartsch R, Berghoff A, Bago-Horvath Z, DeVries C, Dubsky P, Pluschnig U, Rudas M, Rottenfusser A, Fitzal F, Dieckmann K, Mader RM, Gnant M, Zielinski CC, Steger GG. Medical University of Vienna, Vienna, Austria

Brain metastases (BM) are frequently diagnosed in patients (pts) with Her2-positive metastatic breast cancer (BC); a rising incidence was also reported in triple-negative disease. We hypothesized that pts with triple-negative or Her2-positive tumours had shorter BMFS as compared to other BC subtypes.
Therefore, we aimed to compare BMFS in pts with Her2-positive, estrogen receptor (ER) positive and triple-negative BC treated at the Medical University of Vienna from 1999-2009. In Her2-positive tumours, we further investigated the influence of ER co-expression on BMFS, as Her2-positive / ER-positive tumours were reported to express less aggressive biological properties.
BMFS was defined as primary study endpoint and measured as the interval from diagnosis of metastatic BC until diagnosis of BM.
A total of 168 pts were identified from a breast cancer database. 34 pts were excluded from this analysis as brain was the first site of disease progression; hence complete datasets from 134 pts were available (69 Her2-positive; 33 triple-negative; 32 ER-positive).
Her2 status was analyzed by immunohistochemistry (IHC) and reanalyzed by FISH if a score of 2+ was gained. ER was analyzed by IHC; ER negative tumours were defined by a cut-off value of <10% positively stained tumour cells. BMFS was estimated with the Kaplan-Meier product limit method and compared with the log-rank test; factors significantly associated with BMFS in the univariate analysis were included into a Cox proportional hazard model.
Median BMFS in triple-negative pts was 14 months (m) (95% CI 12.17-15.83), as compared to 25 m (95% CI 13.37-36.62) in Her2-positive (p=0.001) and 35 m (95% CI 19.79-50.22) in ER-positive pts (p<0.001), respectively.
In Her2-positive pts, prior trastuzumab treatment for metastatic disease prolonged median BMFS (29 vs. 11 m; p<0.001); BMFS was further improved by trastuzumab in multiple lines (p=0.045) and co-positivity for ER and Her2 (30 vs. 15 m; p<0.001).
ER-expression (HR 2.03; 95%CI 1.22-3.36; p<0.05) and prior trastuzumab (HR 2.72; 95%CI 1.20-6.17; p=0.017) remained independent predictors of longer BMFS in the Cox regression model.
In ER-positive, triple-negative as well as Her2-positive pts, no correlation was found between BMFS and factors such as grading, histological subtype, stage IV disease at primary diagnosis, disease-free interval <24 months from primary treatment to diagnosis of metastatic disease, presence of visceral metastases, presence of lung metastases, and prior capecitabine exposure.
BMFS in triple-negative disease is significantly shorter as compared to Her2-positive or ER-positive tumours, mirroring the aggressiveness of this breast cancer subtype. Probably due to improved systemic disease control, trastuzumab significantly prolonged BMFS in Her2-positive pts. Longer BMFS in ER/Her2 co-positive disease reflects a less aggressive subtype of Her2-positive breast cancer which is less likely to benefit from strategies of BM screening or prevention.

Friday, December 9, 2011 7:00 AM

Poster Session 4: Treatment – Types and Sites of Breast Cancer: Brain Metastases (7:00 AM-9:00 AM)

Terms of Service.


Close Window