|
DISCUSSION
Evaluation of the axilla has been a major focus of interest
in the treatment and staging of patients with invasive breast
cancer. Axillary lymph node status continues to be the single
most important prognostic variable in patients with carcinoma
of the breast.1-4 The issue of staging the axilla is of particular
importance in patients with small invasive carcinomas (<
1 cm), given that detection of nodal disease alters the postoperative
management by providing adjuvant chemotherapy in a group of patients
that would not receive it otherwise. There is continued debate
about the role of axillary dissection in this patient population.
Previous series reported in the literature have shown a substantial
variation in the incidence of ALNM from 3% to 37% in those with
small invasive carcinomas. Attempts have been made to identify
factors that may predict an increase risk of nodal involvement
in this group of patients.
Silverstein and coworkers5 in a retrospective study reported
an incidence of ALNM in 3% of patients with T1a and 17% with
T1b tumors. They determined that increasing size of the tumor
was predictive of ALNM, and given the low incidence of ALNM with
T1a tumors in this study, the authors recommended elimination
of axillary dissection in those with tumors less than 0.5 cm.
Others have advocated a more selective approach in examining
the axilla. Mustafa and colleagues6 in a review of more than
2,100 patients with < 1-cm invasive tumors identified
from a state-wide tumor registry reported a 21% rate of ALNM
in this group. They determined by univariate and multivariate
analyses that the incidence of nodal involvement depended not
only on the size of the tumor but also on the grade of the tumor
and the patient's age. They concluded that any decision about
the elimination of axillary dissection in patients with small
invasive tumors should take into account not only the tumor size,
but in addition other factors such as the grade of the tumor
and the patient's age.
In the present study, we attempted to identify patient and
tumor characteristics that would predict ALNM in those with T1a
and T1b invasive breast cancers. We demonstrate an 18% overall
incidence of ALNM in patients with tumors < 1 cm. By
univariate analysis, four predictors of lymph node involvement
were identified, which include: younger age of the patient, increasing
size and grade of the tumor, and the presence of lymphatic or
vascular invasion. Multivariate analysis continued to identify
patient age, tumor size, and grade of the tumor as significant
predictive factors of ALNM. Lymphatic and vascular invasion could
not be assessed by multivariate analysis because of unknown lymphatic
or vascular status in a significant number of patients. Of the
14 patients identified in this study with ALNM and positive lymphatic
or vascular invasion, none had tumors < 0.5 cm.
Tumor size has consistently been shown to be associated with
frequency of ALNM.1-7 Again, our study demonstrated tumor size
as the most significant independent predictive factor for ALNM
by multivariate analysis. Patients were three-and-a-half times
more likely to have ALNM with every 0.10-cm increase in tumor
size.
Increasing age was inversely correlated with the presence
of nodal disease in our study. ALNM was 40% less likely to develop
in women greater than 50 years of age compared with those younger
than 50 years of age. Others have reported these findings. Mustafa and colleagues9
showed that the incidence of nodal metastases doubled in women
less than 40 years of age when compared with those greater than
40 years of age. A review of the Surveillance, Epidemiology,
and End Results data of more that 12,000 patients with T1a and
T1b tumors demonstrated that increasing patient age was associated
with a progressively decreasing frequency of ALNM. This
ranged from 22.6% in women younger than 40 years of age to 10%
in women older than 70 years of age.10
Although the reason for the age-related
difference is not entirely clear, some have attributed a biologically
more aggressive disease in the younger patient.11
In addition, we show that tumor grade is an important predictor
of ALNM in patients with tumors < 1 cm. Our data indicate
that patients with histologic grade 3 tumors are two-and-a-half
times more likely to have nodal involvement compared with patients
with grade 1. Of the 22 patients identified in this study with
ALNM and grade 3 tumors, 17% were T1a and 77% were T1b. Shoup
and coworkers12
reported similar results of poor tumor grade and associated increase
in nodal disease. In their study, 44% of patients with grade
3 tumors had ALNM.
Although some studies have reported an extremely low risk
of ALNM in patients with small invasive tumors, particularly
T1a (< 5-mm) lesions, and advocate for the complete
elimination of axillary lymph node dissections, we were unable
to identify in our study any subgroup of patients that had an
acceptable low risk of lymph node metastases. When four of the
favorable clinical and pathologic parameters were analyzed in
our study, that is, T1a (< 5-mm) tumors that were well-differentiated
(grade 1) and lacked lymphatic or vascular invasion in patients
older than 50 years of age, a total of 108 patients were identified
with 13% having ALNM. When age increased to more than 60 years
and all other parameters remained the same, an incidence of ALNM
was seen in 8.7% of patients.
Given our results and those of others, we cannot advocate
the complete elimination of axillary lymph node dissections in
those patients with T1a or T1b invasive carcinomas, but given
the recent advances in intraoperative lymphatic mapping and sentinel
lymph node biopsy, the use of these techniques may be a reasonable
alternative to traditional axillary dissections. Interestingly,
these methods may actually increase the rate of ALNM found in
patients with small invasive tumors. With sentinel node biopsy
methods, the identifiable node is carefully examined by immunohistochemical
staining methods and serial sectioning, allowing a more thorough
pathologic examination of the axilla. Reports
indicate that by using these methods compared with conventional
hematoxylin and eosin staining an increase in identifying metastatic
disease to lymph nodes is increased by at least 10%.13
Giuliano and coworkers14
retrospectively used immunohistochemical staining in a study
of patients with T1 carcinomas. They identified 15 patients with
ALNM by these methods, of which 7 were negative by hematoxylin
and eosin, and reported an increase from 10% to 15% in the nodal
metastases in patients with T1a tumors using these methods.
We recognize that several patient and tumor characteristics
such as younger age of the patient and increasing size and grade
of the tumor are predictive of ALNM in patients diagnosed with
small invasive breast cancers. We were unable to identify in
this study any subgroup of patients that had an acceptable low
risk of ALNM. Our recommendation for patients diagnosed with
T1a and T1b tumors is to have their axilla investigated for metastatic
disease either by traditional axillary lymph node dissections
or by intraoperative lymphatic mapping and sentinal lymph node
biopsy techniques.
Introduction
| Methods
| Results
| Discussion | References
Invited
Commentary | Reply
JACS |
