E12. Predictors of resistance to hormonal therapy in breast cancer
Mitch Dowsett
*
, Lesley-Anne Martin, Stephen Johnston, Ian E. Smith
Academic Department of Biochemistry and Breast Unit, Royal Marsden Hospital, London SW3 6JJ, UK
Endocrine therapy with selective oestrogen receptor
modulators (SERMs), such as tamoxifen, or agents
leading to oestrogen deprivation, gonadotrophin-
releasing hormone (GnRH) agonists in premenopausal
and aromatase inhibitors in postmenopausal women, are
widely used in the treatment of all stages of breast cancer.
Resistance to such treatments can be intrinsic and occur at
first exposure or acquired after initial response. The
identification of predictors of resistance to these agents is
important for the appropriate selection of patients for
treatment. Additionally, where these predictors are me-
chanistically involved in the resistance, their identification
is important for the development and application of new
agents targeted at the pathways involved.
It has become increasingly clear that only those pa-
tients that present as positive for oestrogen receptor
(ER)-a and/or progesterone receptor (PgR) benefit from
hormonal therapy of any sort. The small fraction of
patients that are ER-PgR+ (<5% of total in almost all
reports) is important to define as they do show benefit,
despite having no measurable ER. Some data indicate
that ER-b expression may be associated with a poorer
likelihood of response to endocrine therapy, but the
data are not sufficiently compelling to assess this marker
routinely [1]. Most patients continue to show ER+ dis-
ease at relapse during endocrine therapy, but approxi-
mately 15% of patients treated with tamoxifen show no
ER in relapsed lesions [2].
The growth factor receptors, HER2 and epidermal
growth factor receptor (EGFR), are less frequently
positive in ER+ breast cancers, but there is increas-
ingly strong evidence that tumours in which one or
both are co-expressed with ER show a reduced like-
lihood of benefit from tamoxifen. This evidence comes
from a series of adjuvant studies of tamoxifen (none
of which is individually persuasive in this regard) and
2 neoadjuvant studies of tamoxifen versus an ar-
omatase inhibitor [3,4]. Of particular interest, ER+-
HER2/EGFR+ tumours showed a better response to
the aromatase inhibitors in these latter studies.
This may be due to cross-talk between the growth
factor receptor and ER signalling pathways that in
model systems leads to sensitisation of the ER to
oestrogen stimulation and to the agonist effects of
tamoxifen [5].
PgR expression is thought to be highly dependent on
the classical mechanism of ER-oestrogen response ele-
ment signalling. There is recent evidence that there may be
differential benefit from aromatase inhibitors and ta-
moxifen according to the PgR status of ER tumours. A
retrospective subgroup analysis in the arimidex, tamoxi-
fen, alone or in combination (ATAC) adjuvant trial of
anastrozole versus tamoxifen versus the combination of
anastrozole and tamoxifen found that patients that were
ER+PgR+ or ER+PgR- had a similar recurrence-free
survival (RFS) on anastrozole, but that patients that were
ER+PgR- faired much worse on tamoxifen than the
ER+PgR+ patients. This resulted in a substantially
greater benefit for anastrozole over tamoxifen in the
ER+PgR- than in the ER+PgR+ patients, although
greater benefit over tamoxifen remained in the latter
group. There is significant co-segregation of growth fac-
tor receptor-positivity with PgR-negativity. Therefore, it
may be growth factor receptor-positivity that is respon-
sible for this differential benefit and PgR expression may
be a marker of this mechanism. This possibility will be
examined by growth factor receptor measurements in a
retrospective collection of excised tumours from the
ATAC trial in study TA/01 (also known as TransATAC).
Almost all studies of response and resistance in breast
cancer depend on clinical measurement of tumour
growth or regression and the linkage of this to putative
biochemical or molecular markers of response/resis-
tance. By such clinical measures, approximately 50% of
*
Corresponding author. Tel.: +44 207 808 2885; fax: +44 207 376
3918.
E-mail address: mitch.dowsett@icr.ac.uk (M. Dowsett).
1359-6349/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ejcsup.2004.08.011
EJC Supplements Vol 2 No. 9 (2004) 43–44
www.ejconline.com
EJC
Supplements