br recurrence in eBC occurs within the first two or
recurrence in eBC occurs within the first two or three years and then decreases consistently in the interval of three to five years and decreases even more slowly beyond five years. In addition, past survival analysis with longer follow-up suggested that HER2-targeted therapy showed its benefit primarily as a CL 316243 in early recurrence.14
Women who were not reported to be deceased at the time of the data analysis were censored at the date they were last known to be alive.
Association within categorical variables were assessed by Fisher exact test for binomial categorical variables and by Chi-square test for all other instances. Survival distribution was estimated by the Kaplan–Meier me thod. Significant differences in probability of surviving between the strata were evaluated by log-rank test. The hazard ratios and their associated 95% confidence interval (95% CI) of any distant invasive recurrence at any time and after 3 years were estimated using a Cox multivariate proportional hazard regression model. The estimated risk of recurrence was adjusted for HR status, BMI, AJCC stage, tumor size, nodal status and type of adjuvant chemotherapy.
Patients were then grouped by HR status and BMI and four subgroups were assembled: patients with HR+ tumors and lower BMI, patients with HR+ tumors and higher BMI, patients with HR-negative (HR-) tumors and lower BMI and patients with HR- and higher BMI (the latter one representing the reference group).
BMI categories were defined according to the World H ealth Organization (WHO)16: underweight below 18.5 kg/m2; normal between 18.5 and 24.9 kg/m2; overweight between 25 and 29.9 kg/m2; and obese above 30.0 kg/m2. Underweight women (4% of total) were included in the normal weight group and overweight and obese patients were combined as one group. BMI was assessed at baseline before treatment administration.
A significance level of 0.05 was chosen to assess the statistical significance. All reported p-values are two sided.
Statistical analysis was performed with the MedCalc package (MedCalc® v126.96.36.199 Software, Ostend, Belgi um).
Two hundred and thirty-eight women out of 279 were eligible for the analysis. Forty-one patients were excluded because they had not completed adjuvant treatment with chemotherapy followed by one year of Trastuzumab. Median age was 54 years (range 20-88). The majority of patients (58.1%) had a post-menopausal status.
The adjuvant treatment was very homogeneous across subgroups. One hundred and thirty-five (56.4%) patients received a sequential chemotherapy with anthracyclines and taxanes. Seventy-eight (32.6%) received only anthracyclines and 26 (10.8%) taxanes alone. All patients completed the adjuvant chemotherapy plus one year of Trastuzumab.
Differences by subgroups
The majority of women diagnosed with higher stage received a sequential treatment with anthracycline plus taxane (86% of them in stage III, p<0.0001, 65% in stage II, respectively) as recommended by international guidelines.13
Obese patients (according to the standard World Health Organization classification)16 in our cohort were 14% while 69 of them (29%) were overweight.
The distribution of BMI categories at time of diagnosis was different (p=0.03) among HR+/HER2+ and HR-
/HER2+ patients (36% of overweight/obese patients in the HR+ group and 54% in the HR- group), as already reported by past observations.17,18
Overweight and obese women were significantly more likely to be postmenopausal (p=0.006) and to have larger tumors (T3 to T4, p=0.036). No significant differences were noted for nodal status, grade, lymph vascular invasion, Ki 67 score or adjuvant systemic chemotherapy. Patients’ characteristics by BMI are summarized in Table 1 .
Long-term recurrence (DDFS)
At multivariate analysis, after adjusting for covariates (tumor size, nodal status, AJCC stage, hormonal receptors status, type of adjuvant chemotherapy, BMI), the only factor influencing DDFS was AJCC stage (hazard ratio: 3.20; 95% CI: 1.31-7.81, p=0.010) (Table 3).
Early recurrence (3yDDFS)
Neither HR status nor BMI alone resulted independent prognostic factors at multivariate analysis. However, the hazards for patients with HR-negative tumors who were also overweight/obese (3yDDFS for HR-negative tumors with BMI≥25: 86.9%, 95% CI: 75.0-97.7%) were amplified when compared to patients with HR+ tumors who were under/normal weight (3yDDFS for HR+ tumors with BMI<25: 98%, 95% CI: 94.8-100.0%) and other subgroups (log-