DOI:10.2214/AJR.07.2659
AJR 2008; 191:346-351
© American Roentgen Ray Society
Breast Carcinoma with Basal Phenotype: Mammographic Findings
Angela A. Luck1,
Andrew J. Evans1,
Jonathan J. James1,
Emad A. Rakha2,
E. Claire Paish2,
Andrew R. Green2 and
Ian O. Ellis2
1 Nottingham International Breast Education Centre, Nottingham City Hospital,
Hucknall Rd., Nottingham NG5 1PB, United Kingdom.
2 Department of Histopathology, University of Nottingham, Nottingham City
Hospital, Nottingham, United Kingdom.
Received June 1, 2007;
accepted after revision February 26, 2008.
Address correspondence to A. A. Luck
(angela.luck{at}uhb.nhs.uk).
Abstract
OBJECTIVE. Basal phenotype has been found to be an independent poor
prognostic factor for breast cancer. The aim of this study was to assess the
mammographic appearance of screening-detected breast carcinoma with the basal
phenotype.
MATERIALS AND METHODS. A series of 1,944 consecutively enrolled
patients with operable invasive breast cancer underwent immunohistochemical
analysis with cytokeratin 5/6 and cytokeratin 14 markers to identify tumors
exhibiting basal phenotype characteristics. Among those patients, 356 women
with breast cancer were common to a prospectively collected database of
screening-detected cases of breast cancer. The predominant mammographic
appearance and any associated features were reported by experienced image
readers blinded to phenotype status. A chi-square test was used to assess
difference between the mammographic appearances of a group of tumors with the
basal phenotype and those of a group with the nonbasal phenotype.
RESULTS. Forty-one (12%) of the screening-detected tumors had basal
phenotypic expression, and these were compared with 309 (88%) nonbasal tumors.
Basal-phenotype tumors were significantly more likely to manifest as an
ill-defined mass (basal phenotype, 25 [61%] of 41 tumors; nonbasal phenotype,
75 [24%] of 309 tumors; p < 0.001) or with comedo calcification
(basal phenotype, nine [22%] of 41 tumors; nonbasal phenotype, 30 [10%] of 309
tumors; p = 0.019). Nonbasal-phenotype tumors were more likely to
manifest as a spiculated mass (nonbasal phenotype, 150 [49%] of 309 tumors;
basal phenotype, eight [20%] of 41 tumors; p < 0.001). The low
rate of spiculation in basal tumors was independent of histologic grade.
CONCLUSION. Screening-detected breast tumors with a basal phenotype
have a mammographic appearance different from that of nonbasal tumors. This
finding may explain the good prognostic value of mammographic spiculation
reported in previous studies.
Keywords: basal phenotype • breast cancer • mammographic appearance
Introduction
Breast cancer that exhibits basal-phenotype characteristics is the subject
of much investigation and debate. It is estimated to account for 2–18%
of all cases of breast cancer
[1]. Tumors with the basal
phenotype differ from nonbasal tumors in both histologic appearance and
prognostic importance [2,
3]. They tend to be larger and
of higher histologic grade, exhibit comedonecrosis, and are often estrogen
receptor negative, progesterone receptor negative, and c-ERBB2 (HER2) negative
(triple negative) [4].
Independently of other variables, such as tumor size and grade and lymph node
status, the basal phenotype has been found
[2] to be a poor prognostic
factor in breast cancer. Such tumors have a shorter disease-free interval,
have an aggressive pattern of metastatic disease favoring brain and lung
metastasis over bone involvement, and are associated with poor overall
survival rates
[5–7].
The differing mammographic appearances of breast cancer are
well-recognized. Knowledge of such appearances is fundamental to screening
detection of the various types of breast cancer. Masses, architectural
distortions, focal asymmetries, and microcalcification raise suspicion of the
presence of pathologic changes. Such features differ in positive predictive
value for the presence of breast cancer
[8], and the degree of
suspicion determines both the decision to recall a woman for additional
evaluation and the subsequent investigation of the perceived abnormality.
Mammographic appearance frequently has good correlation with subsequent
histologic findings. Medullary, papillary, and mucinous subtypes are often
well-defined tumors. Lobular carcinoma often is diffuse, infrequently
calcified, and in the early stages is difficult to appreciate on mammograms.
Microcalcification is the hallmark of ductal carcinoma in situ. Spiculation
correlates significantly with low histologic grade, and ill-defined masses and
microcalcification are features of high-grade tumors
[9]. Although previous studies
[10,
11] have shown conflicting
evidence on the prognostic value of comedo calcification on mammograms, they
also have shown that spiculation is a favorable feature in terms of breast
cancer survival
[12–14].
The aim of this study was to assess the mammographic appearance of breast
cancer exhibiting a basal phenotype to discern the presence of radiographic
differences from nonbasal tumors and to determine whether there are features
that facilitate earlier detection of these tumors, which frequently have a
poor prognosis.
Materials and Methods
At our institution, a series of 1,944 cases of invasive breast cancer was
consecutively recorded. The women presented both with symptoms and through the
screening program between the years 1986 and 1998. The series in this study
was limited to women younger than 70 years with operable breast cancer. Tumors
in this series had been subjected to immunohistochemical analysis for basal
and nonbasal phenotypic expression
[2]. Basal phenotype was
defined according to expression of basal cytokeratins, 10% or more of tumor
cells expressing cytokeratin 5/6 or cytokeratin 14. Tumors with less than 10%
reactivity were classified as nonbasal.
Mammographic screening was begun at our center in 1987. Although the
screening program has evolved, women in this study with disease diagnosed at
screening were invited to undergo mammography every 3 years between the ages
of 50 and 64 years. Two views, mediolateral oblique and craniocaudal, were
obtained at the first screening session, and a single mediolateral oblique
view was obtained at subsequent sessions. All women in this study were
recalled to the screening assessment clinic and underwent triple assessment
before the diagnosis of breast cancer was made. Craniocaudal or extra views
were obtained at assessment for full evaluation of the mammographic findings.
Sonography was used to evaluate all mass lesions and possible distortion, but
most patients with calcification did not undergo sonography during the period
studied. Wherever possible, biopsy was performed under sonographic guidance.
This study was focused on the mammographic appearance of the tumors, and
therefore the sonographic appearance was outside its scope.
The cases of women who received the diagnosis of breast cancer were entered
prospectively into a database in which the mammographic appearance of the
tumor was recorded. The dominant mammographic feature was recorded as mass
(well-defined, ill-defined, or spiculated) (Figs.
1 and
2), architectural distortion,
focal asymmetry, or microcalcification (comedo or noncomedo) (Figs.
3,
4A, and
4B). When more than one finding
was present, the nondominant characteristics also were recorded. The dominant
feature was the most easily perceived abnormality, not necessarily the most
malignant-appearing one. For example, if a large well-defined mass with a
small area of faint comedo calcification was identified, the well-defined mass
was the dominant feature. Five experienced breast radiologists who had
interpreted more than 5,000 sets of screening images per year for 5 years and
who were blinded to phenotype status recorded the mammographic features. The
data sets were combined to identify cases in which the mammographic appearance
at screening detection was recorded and the breast tumor was classified as
having the basal or the nonbasal phenotype. This study group comprised 356
tumors.
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2 —56-year-old woman with breast cancer. Mammogram shows
ill-defined mass representing invasive carcinoma (dominant abnormality).
Associated ductal microcalcification (nondominant abnormality) was high-grade
ductal carcinoma in situ.
|
|
View larger version (82K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3 —52-year-old woman with breast cancer. Mammogram shows casting
comedo calcification in areas of increased density. Histologic examination
showed high-grade ductal carcinoma in situ with grade 3 invasive
carcinoma.
|
|
The dominant mammographic features for the phenotypic groups were compared
by use of a chisquare test to assess for significance of differences. All
suspicious mammographic features were recorded for every tumor and analyzed in
a similar manner. Histologic grade 3 and grade 2 tumors were separately
assessed on the basis of phenotype to determine whether differences between
the mammographic features remained when tumors of the same grade were
compared.
Results
Forty-three (12%) of the 356 screening-detected cases of invasive breast
cancer were basal-phenotype tumors and 313 (88%) had the nonbasal phenotype. A
total of 27 basal tumors were identified during prevalence screening, and the
other 16 presented during subsequent (incidence) screening. In the nonbasal
group, 167 tumors were detected at prevalence screening and 146 at incidence
screening.
One patient in the basal group was recalled because of a separate benign
lesion, and breast carcinoma was detected during physical examination before
further radiologic assessment. The breast tumor was occult on mammograms. The
mammograms of one patient were missing. Therefore, 41 basal-phenotype tumors
were analyzed according to mammographic appearance. Three patients in the
nonbasal group were recalled because of symptoms or abnormalities detected by
the radiographer at the breast screening visit. These patients had normal
mammographic findings. Mammograms were missing for a patient who presented
during incidence screening. The nonbasal-phenotype group therefore comprised
309 patients.
Table 1 shows the results
for the dominant mammographic feature according to phenotype. For both
phenotypes, the most common presentation was a mass with either an ill-defined
or a spiculated margin. Basal-phenotype tumors (25 [61%] of 41), however, were
significantly more likely than nonbasal tumors (73 [24%] of 309) to appear as
ill-defined masses on mammograms (p < 0.001). Nonbasal-phenotype
tumors (150 [49%] of 309) were significantly more likely than basal tumors
(six [15%] of 41) to exhibit marginal spiculation (p < 0.001).
When every suspicious feature on a screening mammogram was assessed, the
observed difference in tumor margin (spiculated vs ill-defined) maintained
significance between the two phenotypic groups
(Table 2): eight [20%] of 41
basal phenotype versus 150 [49%] of 309 nonbasal phenotype (p <
0.001) for marginal spiculation and 25 (61%) of 41 basal phenotype versus 75
[24%] of 309 nonbasal phenotype (p < 0.001) for an ill-defined
margin. In addition, comedo calcification was seen more frequently with basal
tumors: nine (22%) of 41 basal-phenotype tumors versus 30 (10%) of 309
nonbasal-phenotype tumors (p = 0.019).
View this table:
[in this window]
[in a new window]
|
TABLE 2: Suspicious Features Detected During Screening Mammography in Cases of
Basal and Nonbasal Phenotype Breast Cancer
|
|
The mammographic features of all grade 3 tumors in the study group were
compared according to phenotypic status
(Table 3). There were 98 grade
3 tumors, 20 having a basal and 78 a nonbasal phenotype. Basal grade 3 tumors
(14 [70%] of 20) were more likely than nonbasal grade 3 tumors (33 [42%] of
78) to manifest as an ill-defined mass (p = 0.027). Basal grade 3
tumors (one [5%] of 20) also were less likely than nonbasal grade 3 tumors (21
[27%] of 78) to present as a spiculated mass (p = 0.036). No
difference in frequencies of comedo calcification was found in a comparison of
the grade 3 basal and grade 3 nonbasal tumors. The results for grade 2 tumors
are shown in Table 4. The
difference in tumor margins of the basal and nonbasal phenotypes again
maintained significance.
Discussion
There is compelling evidence that basal characteristics of a primary tumor
carry important prognostic information for the patient. Many authors
[4,
7,
15] advocate routine testing
for such characteristics. Determination of basal phenotype, however, is not
used in the routine clinical management of breast cancer, at least in part
because of discrepancies in the literature regarding the definition of basal
phenotype and the method used for identification of these tumors. Although
most basal-phenotype tumors have common features, there are no specific
hallmark morphologic features for reliably identifying them. An increasing
number of research studies are being conducted to reach consensus on a
reliable method of identification of basal-phenotype tumors and eventual
incorporation of the findings into patient treatment.
Immunohistochemical staining for basal cytokeratins often is used to
identify tumors with the basal phenotype
[2,
16–21].
In this study, we used cytokeratin 5/6 and cytokeratin 14 to determine basal
phenotypic expression, 10% or more positivity for either of these markers
indicating a basal phenotype. This simple definition has the advantages of
ease of application, reproducibility, and availability of these cytokeratin
markers in histopathology laboratories. Should assessment of basal phenotypic
expression become routine, straightforward definition may be beneficial in
terms of both cost and quality assurance. We have used this definition as the
basis of earlier studies [2,
4,
6,
18,
21]. Like several other
groups, we have found that basal tumors are most likely to be ductal carcinoma
of no special type, grade 3, and large (> 2 cm). Basal tumors often exhibit
comedonecrosis, a stromal lymphocytic response, presence of metaplastic
elements, and glomeruloid microvascular proliferation and are frequently
hormone receptor and ERBB2 (HER2) negative
[5,
17–23].
Basal phenotype is an independent factor indicating poor prognosis in breast
cancer [2]. Expression of this
phenotype is predictive of a particularly aggressive course for patients with
grade 3 tumors without lymph node involvement
[6] and is a powerful factor
indicating poor prognosis for small (< 15 mm) screening-detected carcinomas
[21]. Moreover, in breast
cancer patients with the triple-negative phenotype (estrogen
receptor–negative, progesterone receptor–negative, and ERBB2
(HER2) negative tumors), the basal phenotype is the sole prognostic factor in
patients without lymph node involvement
[4].
Metastatic disease in women with basal-phenotype primary breast cancer is
more likely to involve the brain and lungs than is metastatic disease in women
with a nonbasal primary tumor
[17,
18]. Patients with basal
tumors also are more likely than patients with nonbasal tumors to present with
metastatic disease in the brain. Breast cancer–specific survival and
survival with metastatic disease both are shorter for women with basal
phenotypic expression in the primary tumor
[18]. To our knowledge, the
mammographic appearance of this important class of breast cancer has not been
evaluated.
There has been continued interest in the use of mammographic features as
prognostic indicators in cases of small screening-detected tumors. We
concentrated on tumors detected through screening. Previous work in our unit
and in other units on the prognostic significance of mammographic spiculation
and of basal phenotype has been restricted to screening cases. Although many
of the patients with symptomatic disease detected during the study period also
underwent imaging, we considered it important to confine this study to
screening-detected lesions. Although breast cancer screening programs have
been criticized for early detection of particular cancers that have little
effect on subsequent outcome among patients
[24,
25], early detection and
timely effective management of aggressive subtypes of breast cancer are key to
influencing morbidity and mortality.
Recognizing how certain mammographic features correlate with prognosis is
important not only in detecting cancer but also in gaining insight into how
the tumor grows and potentially in determining which tumors can be managed
with aggressive adjuvant treatment. A number of independent study groups
[12–14]
have found mammographic spiculation, or a stellate appearance, associated with
good outcome. Many women with such tumors have long survival times; if the
tumor is smaller than 15 mm in diameter, the survival period typically is
excellent. The consensus is less clear regarding the prognostic role of
microcalcification. Some groups
[10,
12] have found the presence of
casting type (comedo) calcification predictive of poorer outcome among women
with small tumors. Others [11,
14] have been unable to find
an effect of calcification on survival.
We identified mammographic spiculation as a feature of nonbasal phenotype
tumors. This finding supports the presence of mammographic spiculation and the
absence of basal characteristics in a tumor as predictive of a relatively
better survival outcome for the patient. Basal tumors are more likely to
manifest as ill-defined masses. When grade is taken into account, these
findings persist, showing that the differences in mammographic appearance are
not simply due to tumor grade
[9]. Basal tumors also are more
likely to exhibit comedo calcification. This finding correlates with the
histologic findings of comedonecrosis in many basal-type tumors. The rates of
mammographic comedo calcification of basal and nonbasal tumors are similar in
comparisons of grade 3 tumors. The high frequency of comedo calcification in
basal tumors may therefore be a consequence of high histologic grade.
Basal tumors are particularly frequent among carriers of BRCA1
[26]. Spiculation is rare in
BRCA1-associated tumors
[27]. However,
BRCA1-associated tumors rarely exhibit comedo calcification on
mammograms, so the mammographic features of BRCA1-associated tumors
cannot be fully explained on the basis of high frequency of tumors with a
basal phenotype.
The difference in mammographic appearance between basal and nonbasal tumors
identified at screening is of particular importance in consideration of
interval tumors. Reviews of those tumors reveal minimal mammographic signs
that can be overlooked [28];
one of these signs often is an ill-defined mass. We have found that
basal-phenotype tumors are more likely to be ill-defined and thus are
potentially more difficult to assess on screening mammograms.
In this study, basal and nonbasal classification of tumors was made in a
series of cases of breast cancer in which the survival outcome for the
individual patients is now known, enabling identification of the basal
phenotype as an independent factor indicating poor prognosis. We used the same
series to evaluate the mammographic appearances of the tumors. The tumors were
detected between 1987 and 1998 and were evaluated to the best available
standard at that time. We expect that analysis of these data will be of value
for future work on this important subgroup. Since the presentation of these
tumors, screening practice has evolved (two views and double reading with
arbitration for all screening images), and the ability to detect subtle
mammographic signs has increased. If assessment of basal cytokeratin
expression becomes routine, the findings of this study should be validated
prospectively for tumors detected with current screening practice. Evaluation
of tumors manifesting symptomatically also may further clarify the
mammographic appearance of basal-phenotype tumors.
In an assessment of the characteristics of basal and nonbasal tumors
manifesting at screening, we found that screening-detected breast tumors with
basal phenotypic expression differ in mammographic appearance from nonbasal
tumors. Basal tumors are less likely to have spiculation, more likely to
manifest as an ill-defined mass, and more likely to be found in association
with comedo calcification. These findings may explain the prognostic
importance of mammographic spiculation, and the increased rates of comedo
calcification may be an effect of tumor grade.
References
- Nagle RB, Bocker W, Davis JR, et al. Characterisation of breast
carcinomas by two monoclonal antibodies distinguishing myoepithelial from
luminal epithelial cells. J Histochem Cytochem1986; 34:869
-881[Abstract]
- Rakha EA, Putti TC, Abd El-Rehim DM, et al. Morphological and
immunophenotypic analysis of breast carcinomas with basal and myoepithelial
differentiation. J Pathol 2006;208
: 495-506[CrossRef][Medline]
- Neilsen TO, Hsu FD, Jensen K, et al. Immunohistochemical and
clinical characterisation of the basal like sub-type of invasive breast
cancer. Clin Cancer Res 2004;10
: 5367-5374[Abstract/Free Full Text]
- Rakha EA, El-Sayed ME, Green AR, et al. Prognostic markers in
triple-negative breast cancer. Cancer2007; 109:25
-32[CrossRef][Medline]
- Tsuda H, Takarabe T, Hasegawa F, Fukutomi T, Hirohashi S. Large,
central acellular zones indicating myoepithelial tumour differentiation in
high-grade invasive ductal carcinomas as markers of predisposition to lung and
brain metastases. Am J Surg Pathol 2000;24
: 197-202[CrossRef][Medline]
- Rakha EA, Abd El-Rehim DM, Paish C, et al. Basal phenotype
identifies a poor prognostic subgroup of breast cancer of clinical importance.
Eur J Cancer 2006;42
: 3149-3156[CrossRef][Medline]
- Rodriguez-Pinilla SM, Sarrio D, Honrado E, et al. Prognostic
significance of basal-like phenotype and fascin expression in node-negative
invasive breast carcinomas. Clinical Cancer Res2006; 12:1533
-1539[Abstract/Free Full Text]
- Burrell HC, Pinder SE, Wilson ARM, et al. The positive predictive
value of mammographic signs: a review of 425 nonpalpable breast lesions.
Clin Radiol 1996;51
: 277-281[CrossRef][Medline]
- De Nunzio MC, Evans AJ, Pinder SE, et al. Correlations between the
mammographic features of screen detected invasive breast cancer and
pathological prognostic factors. Breast1997; 6:146
-149[CrossRef]
- Thurfjell E, Thurfjell MG, Lindgren A. Mammographic finding as a
predictor of survival in 1–9 mm invasive breast cancers: worse prognosis
for those cases presenting as calcifications alone. Breast Cancer
Res Treat 2001; 67:177
-180[CrossRef][Medline]
- James JJ, Evans AJ, Pinder SE, McMillan RD, Wilson ARM, Ellis IO.
Is the presence of mammographic comedo calcification really a prognostic
factor for small screen detected invasive breast cancer? Clin
Rad 2003; 58:54
-62[CrossRef][Medline]
- Tabar L, Tony Chen HH, Amy Yen MF, et al. Mammographic tumor
features can predict long-term outcomes reliably in women with 1-14-mm
invasive breast carcinoma. Cancer 2004;101
: 1745-1759[CrossRef][Medline]
- Alexander MC, Yankaskas BC, Biesemier KW. Association of stellate
mammographic pattern with survival in small invasive breast tumors.
AJR 2006; 187:29
-37[Abstract/Free Full Text]
- Evans AJ, Pinder SE, James JJ, Ellis IO, Cornford E. Is
mammographic spiculation an independent, good prognostic factor in screening
detected invasive breast cancer? AJR2006; 187:1377
-1380[Abstract/Free Full Text]
- Dairkee SH, Mayall BH, Smith H, Havkett A. Monoclonal marker that
predicts early recurrence in breast cancer. Lancet1987; 1:514[Medline]
- Jones C, Ford E, Gillett C, et al. Molecular cytogenetic
identification of subgroups of grade III invasive ductal breast carcinomas
with different clinical outcomes. Clin Cancer Res2004; 10:5988
-5997[Abstract/Free Full Text]
- Fulford LG, Reis-Filho JS, Ryder K, et al. Basal-like grade III
invasive ductal carcinoma of the breast: patterns of metastases and long term
survival. Breast Cancer Res 2007;9
: R4[CrossRef][Medline]
- Luck AA, Evans AJ, Green AR, et al. The influence of basal
phenotype on the metastatic pattern of breast cancer. Clin Oncol (R
Coll Radiol) 2008; 20:40
-45[Medline]
- Banerjee S, Reis-Filho JS, Ashley S, et al. Basal-like breast
carcinomas: clinical outcome and response to chemotherapy. J Clin
Pathol 2006; 59:729
-735[Abstract/Free Full Text]
- Fulford LG, Easton DF, Reis-Filho JS, et al. Specific morphological
features predictive for the basal phenotype in grade 3 invasive ductal
carcinoma of breast. Histopathology 2006;49
: 22-34[CrossRef][Medline]
- Evans AJ, Rakha EA, Pinder SE, et al. Basal phenotype: a powerful
prognostic factor in small screen-detected invasive breast cancer with
long-term follow-up. J Med Screen 2007;14
: 210-214[Abstract/Free Full Text]
- Jones C, Nonni AV, Fulford L, et al. CGH analysis of ductal
carcinoma of the breast with basaloid/myoepithelial cell differentiation.
Br J Cancer 2001;85
: 422-427[CrossRef][Medline]
- Foulkes WD, Brunet JS, Stefansson IM, et al. The prognostic
implication of the basal-like (cyclin E high/p27
low/p53+/glomeruloid-microvascular-proliferation+) phenotype of BRCA1-related
breast cancer. Cancer Res 2004;64
: 830-835[Abstract/Free Full Text]
- Gotzsche PC, Olsen O. Is screening for breast cancer with
mammography justifiable? Lancet 2000;355
: 129-134[CrossRef][Medline]
- Olsen O, Gotzsche PC. Cochrane review on screening for breast
cancer with mammography. Lancet 2001;358
: 1340-1342[CrossRef][Medline]
- Lakhani SR, Reis-Filho JS, Fulford L, et al. Prediction of BRCA1
status in patients with breast cancer using estrogen receptor and basal
phenotype. Clin Cancer Res 2005;11
: 5175-5180[Abstract/Free Full Text]
- Hamilton LJ, Evans AJ, Wilson AR, et al. Breast imaging findings in
women with BRCA1- and BRCA2-associated breast carcinoma. Clin
Radiol 2004; 59:895
-902[CrossRef][Medline]
- Burrell HC, Sibbering DM, Wilson AR, et al. The mammographic
features of interval cancers and prognosis compared with screen detected and
symptomatic breast cancers. Radiology1996; 199:811
-817[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
