Influence of age and prognosis of breast cancer in Nigeria.

OBJECTIVE

To determine the relationship between the age at diagnosis and established prognostic factors of breast cancers in Calabar, Nigeria. Attempts made to assess the prognostic value of age at presentation.

DESIGN

Retrospective study of invasive breast cancer seen in Calabar over a seventeen-year period. Pearson's correlation, univariate and multivariate Cox's regression were used.

SETTING

University of Calabar Teaching Hospital, Calabar, Nigeria, a referral and teaching hospital.

SUBJECTS

Three hundred cases of invasive breast cancer diagnosed between 1983 and 1999 in Calabar, Nigeria. The necessary follow-up data was available for 129 patients.

RESULTS

The mean age at diagnosis of breast cancer in Nigeria was 42.7 years (SD 12.2, range 18-85 years). Patients less than 40 years accounted for 39.8% of the total number of patients with infiltrating breast carcinoma. In the whole material (n=300), there was a positive association between age and tumour size (r=0.44, p=<0.0001), stage (r=0.47, p=<0.0001), the degree of necrosis (r=0.21, p=0.0002), histological grade (r=0.11, p=0.0476), MAI (mitotic activity index, r=0.12, p=0.0338), and MNA (mean nuclear area, r=0.17, p=0.0033). The correlation between age and SMI (standardized mitotic index), AI (apoptotic index), SMI/AI ratio, and FTD (fraction of fields showing tubular differentiation) were not statistically significant. The optimal decisive prognostic cut point for age was 33 years (p=0.0064). Age was also a significant prognosticator when used as a continuous variable (p=0.0240). Survival was better in the younger patients. However, in the Cox's multivariate analysis involving SMI, tumour size and age (both as a continuous variable and using the determined cut point of 33 years), the age at diagnosis lacked an independent prognostic value.

CONCLUSION

The more advanced nature of breast cancers and the possible more aggressive tumours (reflected by the higher MNA values) in the older patients may explain the poorer survival seen in patients diagnosed at 40 years or above. It is also probable that the lifestyle differences between the two studied age groups may influence the early detection and prompt commencement of therapy. Screening and treatment approaches between the two age groups may differ in view of the differences.