Chemotherapy use for hormone receptor-positive, lymph node-negative breast cancer

Abstract

Purpose: To describe the frequency of chemotherapy use for hormone receptor (HR)-positive, lymph node (LN)-negative breast cancer from 1997 to 2004 at eight National Comprehensive Cancer Network institutions, to explore whether chemotherapy use varied over time and between institutions, and to identify factors associated with the decision to forego chemotherapy.

Patients and methods: Among women younger than age 70 years with HR-positive, LN-negative breast cancer measuring more than 1 cm, we analyzed the frequency of chemotherapy use on a yearly basis. A multivariable logistic regression model assessed the relationship between receipt of chemotherapy and year of diagnosis, institution, tumor features, and patient characteristics. Interaction terms were added to the model, and stratified analyses were conducted to further explore the determinants of chemotherapy use.

Results: Fifty-five percent of 3,190 women received chemotherapy. Chemotherapy use was less common for patients with 1.1- to 2-cm tumors than for patients tumors greater 2 cm (47% v 87%, respectively; P < .01) and for women age 60 to 69 years versus women younger than age 50 years (24% v 76%, respectively; P < .01). On multivariable analysis, predictors independently associated with receiving chemotherapy included larger tumor size, higher grade, human epidermal growth factor receptor 2 overexpression, younger age, and institution (P < .01 for all). Institutions exhibited dramatically different rates of chemotherapy use (from 46% to 65%) and patterns of change in chemotherapy use over time (from a 79% relative increase to a 22% relative decrease).

Conclusion: Although institutions seemed to agree that not all women with HR-positive, LN-negative breast cancer need chemotherapy, there did not seem to be consensus regarding which women should get chemotherapy. Only prospective randomized controlled trials will conclusively establish which subtypes of HR-positive, LN-negative breast cancer benefit from chemotherapy.

Fig 1.

Percentage of patients receiving chemotherapy at each institution stratified by age at diagnosis. P values refer to analyses of unadjusted relationships between institution and receipt of chemotherapy for each age stratum using the χ² test.

Fig 2.

Percentage of patients receiving chemotherapy by age at diagnosis, tumor size, and institution for each year of diagnosis. P values refer to analyses of unadjusted relationships between receipt of chemotherapy and age, tumor size, or institution using the Mantel-Haenszel χ² test for trend.

Fig 3.

Adjusted odds of receiving chemotherapy for each institution over time. The referent group for all odds ratios is institution D during the 1997 to 1998 period.