Sometimes a Great Notion—An Assessment of Neoadjuvant Systemic Therapy for Breast Cancer
Journal of the National Cancer Institute
Sometimes a Great Notion-An Assessment of Neoadjuvant Systemic Therapy for Breast Cancer
Nancy E. Davidson 0 1
Monica Morrow 0 1
0 DOI: 10.1093/jnci/dji049 Journal of the National Cancer Institute , Vol. 97, No. 3, © Oxford University Press 2005, all rights reserved
1 Affiliations of authors: Johns Hopkins Kimmel Cancer Center , Baltimore , MD (NED); Fox Chase Cancer Center , Philadelphia, PA (MM). Center, 1650 Orleans St., Rm. 409, Baltimore, MD 21231 , USA
-
Sometimes I live in the country,
Sometimes I live in the town,
Sometimes I get a great notion…
From the song “Good Night, Irene”
By Huddie Ledbetter and John Lomax as adapted by Ken Kesey
The use of neoadjuvant (aka, preoperative or primary)
systemic therapy for the management of operable breast cancer has
been widely heralded as the future of breast cancer treatment. The
underlying rationale consists of several important observations
including the success of this strategy for locally advanced breast
cancer, the utility of the approach to permit organ preservation
in other tumor types, and a compelling preclinical literature
suggesting that preoperative administration of chemotherapy is
associated with improved survival in a rodent breast cancer model.
Enthusiasm has also been engendered by the belief that
preoperative therapy would improve clinical outcomes, enhance the
likelihood of breast conservation, and facilitate prediction of treatment
response, thereby leading to individualization of therapy (
1,2
).
A meta-analysis of neoadjuvant versus adjuvant systemic
therapy for early-stage breast cancer in this issue of the Journal
provides an opportunity for reflection about this approach (
3
). This
meta-analysis appears to be carefully performed and
methodologically sound. Its assumptions and limitations are meticulously
delineated. In brief, a search of the medical literature yielded
12 potentially eligible trials of neoadjuvant versus adjuvant
chemotherapy or endocrine therapy; three were excluded because
the results have not yet been published in the peer-reviewed
literature. Participants in the nine trials total a surprisingly meager
3946 patients; more than half are derived from the NSABP B18
(
4
) and EORTC 10902 (
5
) trials. The meta-analysis did not show
any statistically or clinically significant difference between the
adjuvant and neoadjuvant arms for death, disease progression,
or distant recurrence. Paradoxically, neoadjuvant therapy was
associated with a statistically significantly increased risk of
locoregional recurrence; this increased risk was largely attributed to
trials where radiotherapy without surgery was used for patients
who had a complete clinical response after neoadjuvant therapy.
The rate of breast conservation was increased by neoadjuvant
therapy in only six of the studies, and the rates of clinical and
pathologic complete response were highly variable. The authors
conclude that these results are not likely to change even with the
inclusion of data from two studies of 1315 patients as yet
unreported, a contention that is likely to be correct.
What can we conclude? First, this meta-analysis confirms that
surgery remains an essential part of early breast cancer
management, even when systemic therapy appears to have eradicated all
grossly evident disease. The absence of any tumor (invasive or
intraductal) on pathologic examination after neoadjuvant therapy
was demonstrated in less than 10% of women in the larger
studies in this overview (
4–6
). Although we might anticipate that
radiation alone would give favorable local control rates in these
women, the identification of patients with a pathologic complete
response before surgery has proved to be a difficult task. Use
of mammography, physical examination, or magnetic resonance
imaging cannot reliably assess degree of response in this setting
(
7,8
). Newer approaches using functional MRI show promise but
they remain to be validated in large studies (9). Until complete
pathologic responses to neoadjuvant therapy can be reliably
identified and until clinical trials demonstrate that local control rates
equivalent to those seen with surgery and radiation are obtained
with radiotherapy alone, surgery must be considered a routine
part of management.
Second, not only was the local failure rate increased by 22%
in neoadjuvant recipients, but also the rate of conversion to
breast conservation was not uniformly improved in these trials.
This is in part because a large number of women were already
candidates for breast conservation—66% and 24% of patients
randomly assigned to neoadjuvant therapy on the NSABP B18
EDITORIALS
and EORTC trials, respectively (
4,5
). Also, higher rates of local
recurrence are seen in those women who require chemotherapy
to undergo breast preservation. For example, local recurrence
rates were 15% in patients requiring chemotherapy to undergo
breast conservation compared with 7% for those who we (...truncated)