close

Вход

Забыли?

вход по аккаунту

?

646

код для вставкиСкачать
629
Male Breast Carcinoma
A Review of 229 Patients Who Presented to the Princess Margaret Hospital
during 40 Years: 1955–1996
Paul E. Goss, M.D., Ph.D.1
Caroline Reid, M.Sc.1
Melania Pintilie, M.Sc.2
Ruth Lim, B.Sc.1
Naomi Miller, M.D.3
1
Department of Medical Oncology and Hematology, The Princess Margaret Hospital, Toronto, Ontario, Canada.
2
Department of Biostatistics, The Princess Margaret Hospital, Toronto, Ontario, Canada.
3
Department of Pathology, The Princess Margaret
Hospital, Toronto, Ontario, Canada.
BACKGROUND. A single-institution review of clinical presentation, treatment, and
outcome of male breast carcinoma was conducted.
METHODS. Data obtained by chart review of 229 cases were analyzed with respect
to clinical presentation, treatment choice, significant prognostic factors, and survival. The patients were analyzed both as a single cohort and as four cohorts
grouped according to decade of diagnosis.
RESULTS. Presentation occurred at a median age of 63 years, most often with a
self-detected lump. Pathology consisted of subtypes similar to those of female
breast carcinoma. The majority of tumors were larger than 2 cm in greatest
dimension. Lymph node status, hormone receptors, and histologic and nuclear
grade were underreported. Primary, adjuvant, and advanced disease treatment
practices were reviewed over time. The 5-year disease free survival (DFS), overall
survival (OS), and local control were 47%, 53%, and 91%, respectively. No difference in outcome by decade of diagnosis was observed. Negative lymph nodes and
adjuvant hormone treatment predicted for better DFS and OS. Younger age and
Stage 0 also predicted for better OS.
CONCLUSIONS. Compared with data from female breast carcinoma patients, 5-year
OS for this series was low; however, when these patients were separated by lymph
node status, survival was similar for those with axillary lymph node metastases.
Despite a change in standard primary surgical treatment and an increased use of
chemotherapy and hormone therapy over the study period, no difference in
outcome was observed among these males. In the absence of prospective, randomized clinical trials, collection of comprehensive data on the presentation and
management of male breast carcinoma may help to optimize clinical care. Cancer
1999;85:629 –39. © 1999 American Cancer Society.
KEYWORDS: male breast carcinoma, clinical presentation, treatment, survival.
R
Presented in poster form at the May 1997 meeting
of the American Society of Clinical Oncology, Denver, Colorado.
Address for reprints: Paul E. Goss, M.D., Ph.D.,
Princess Margaret Hospital, 610 University Avenue, 5-303, Toronto, Ontario, Canada M5G 2M9.
Received April 1, 1998; revision received July 17,
1998; accepted July 17, 1998.
© 1999 American Cancer Society
elative to the disease in women, male breast carcinoma is rare,
accounting for less than 1% of all cases of breast carcinoma and
with an incidence of 1 in 100,000 men.1 Each year approximately 1000
cases are diagnosed in the U.S., resulting in roughly 300 deaths
annually.2 Previous reports have drawn similarities between breast
carcinoma in men and women with regard to presentation and sites
of distant metastases.3,4 In addition, there is clinical evidence that, as
in females, tumor size and axillary lymph node involvement are
important prognostic factors in male breast carcinoma.5 Histopathologic subtypes are also similar in men who have infiltrating ductal
carcinomas with or without an intraductal component observed most
frequently.6 Lobular carcinoma, previously not thought to occur in
men due to lack of lobular differentiation, has now been reported in
a few cases.7
630
CANCER February 1, 1999 / Volume 85 / Number 3
A number of factors have been considered to be
related to increased risk for breast carcinoma in men. A
family history of breast carcinoma,1,8 –10 clinical conditions causing hypoandrogenism (including Klinefelter’s
syndrome, testicular trauma, or infertility),11,12 liver disease causing hyperestrogenism,13 and gynecomastia
have all been implicated as possible risk factors. Unfortunately, these factors are not consistently reported;
therefore, their relation to the disease is unclear.
Development of optimal treatment strategies specific to male breast carcinoma is limited by a lack of
traditional large scale clinical trial data. As a result,
current clinical management is generally extrapolated
from principles established for the treatment of female breast carcinoma. In the absence of sufficient
cases necessary to conduct informative prospective
clinical trials, reviews of groups of male breast carcinoma patients, accumulated over time, provide a
means to study the clinical presentation of the disease
and to assess the efficacy of various treatments.
We report here on a large series of male breast
carcinoma cases with long term follow-up presenting
to our institution over a period of 42 years. Patient
history, including recorded risk factors, clinical presentation, treatment, and outcome, are described for
the group as a whole. In addition, we have examined
our patient population for changes in clinical presentation over time and for any relation between disease
outcome and patient survival with changes in management over the 4 decades.
METHODS
Men presenting to the Princess Margaret Hospital
(PMH) with breast carcinoma were ascertained from
patient registration records from 1955–1996. Some of
these cases have been described previously.14,15 Data
regarding patient history, presenting signs and symptoms, risk factors, primary tumor pathology, initial treatment (surgery, chemotherapy, and radiation), details of
recurrence, treatment for recurrence, and survival were
obtained by chart review. Data extraction was designed
and analyzed by the department of biostatistics at PMH.
The data was analyzed both as a single cohort and as 4
cohorts grouped according to decade of diagnosis (1955–
1965, 1966 –1975, 1976 –1985, and 1986 –1996). The
Kruskal–Wallis test was used to test for differences in
age, weight, and height between decades. Most clinical
characteristics of the population were described using
tables for categoric variables. Age was considered a continuous variable and was described by the mean, median, and range. Associations between clinical factors
and patient history or treatment choice were reviewed in
cross-tabulation form and tested for significance with
the chi-square test. The Wilcoxon rank sum test was
used to compare the duration of symptoms between
patient groups with or without a family history of cancer.
Influence of clinical factors on treatment choice was
tested in a logistic model16 using a stepwise selection
technique. Cases missing clinical factor information
were excluded from this analysis. Relapse or death due
to any cause were considered factors in disease free
survival (DFS). Death due to any cause was considered a
factor in overall survival (OS). Local relapse was the only
event considered in determining the local control rate
(LCR). Survival curves were drawn based on Kaplan–
Meier estimates,17 and differences were assessed with
the log rank test.18 Fifteen patients with breast tumor
pathologies other than adenocarcinoma were omitted
from outcome analysis. These included pure ductal carcinoma in situ (DCIS), pure Paget’s disease, sarcoma,
medullary and gelatinous carcinoma, and papillary adenocarcinoma. The 14 patients with unknown pathologies were included in the outcome analysis results presented here, because no major differences were
observed when they were excluded. In addition to 4
patients for whom disease status at last follow-up was
unknown, 26 patients with metastatic disease at diagnosis and 15 patients for whom M stage at diagnosis was
unknown were excluded from disease free survival and
local control analysis. Prognostic factors were tested with
the Cox proportional hazards model19 using the stepwise
selection technique. Cases with any missing prognostic
factor information were excluded from this multivariate
analysis. The group excluded due to missing data were
not significantly different in terms of OS, DFS, or LCR by
the log rank test. Hormone receptor status and family
cancer history were not included in the multivariate
analysis because this information was not reported in a
large number of cases. The variables describing whether
surgery was performed and type of surgery were also
excluded, because those who did not have surgery or
had lumpectomies were often missing data on either
lymph node status or tumor size. Inclusion of these
factors in the analysis did not result in any major differences in the results.
RESULTS
Two hundred twenty-nine male breast carcinoma patients were treated over the 42-year period. The incidence in our patients of factors previously reported to
be associated with an increased risk for male breast
carcinoma are reported in Table 1. Clinical gynecomastia was recorded in 15 cases (6.6%). Family cancer
information, either positive or negative, was reported
in only 51.5% of cases overall. However, the frequency
of family history reporting increased from only 19.4%
in the first decade of the study period to 62.1% by the
most recent decade. There were no patients with clin-
A Review of 229 Male Breast Carcinoma Cases/Goss et al.
TABLE 1
Occurrence of Putative Risk Factors in This Series
631
TABLE 2
Patient Characteristics at Presentation
Risk factor
No. of cases (%)a
History of physical trauma
History of alcohol abuse
History of smoking (ⱖ10 pack-years)
Prior breast radiation therapy
No offspring
Unilateral clinical gynecomastia
Bilateral clinical gynecomastia
Other malignancy (prior or subsequent)
Family history of breast carcinoma
21 (9.2%)
22 (9.6%)
62 (27.1%)
7 (3.1%)
21 (9.2%)
10 (4.4%)
5 (2.2%)
56 (24.5%)b
23 (19.5%)c
a
Percentage of 229 cases, unless otherwise indicated.
b
Ten colon, 4 head and neck, 14 skin, and 28 unspecified. Twenty-five cases occurred before breast
cancer diagnosis.
c
Out of 118 cases. A first-, second-, or third-degree relative with breast carcinoma was considered a
positive family history.
ical Klinefelter’s syndrome or a history of unusual
exposure to radiation (e.g., occupational).
Patient characteristics at presentation are summarized for the group as a whole in Table 2. The
median age at diagnosis for the entire group was 63
years. When patients were grouped according to decade of diagnosis, age and height did not vary significantly. An increase in average weight was observed
over the study period (Kruskal–Wallis test, P ⫽ 0.0041).
Initial detection of breast carcinoma occurred in most
cases due to a self-detected lump (60.7%) or the appearance of breast symptoms, such as nipple discharge (29.3%). The manner of detection did not vary
significantly over the study period. Only 27 cases
(11.8%) were investigated mammographically. The
majority of these mammograms (24 cases) were performed on cases diagnosed within the last two decades of the study. The most common sign or symptom at first examination was a palpable breast mass.
All but 3 patients had some sign or symptom (Table 2),
with 62.9% presenting with 2 or more symptoms. Delay from first appearance of symptoms to first physician contact was ⬎6 months for 36.0% of the 203
patients for whom this information was available (median, 4 months). The duration of symptoms was found
to be significantly shorter (Wilcoxon’s rank sum test,
P ⫽ 0.0025) for patients with a family history of cancer
(median, 1 month) compared with patients who did
not have a family history of cancer (median, 4
months). No significant difference in age at diagnosis
was observed when patients were separated according
to family cancer history.
Tumors occurred most frequently (44.1% of 152
cases with tumor location information) in the central
subareolar region. The second most common site was
Median yrs of age at diagnosis (range)
Median height (range), n ⫽ 155
Median weight (range), n ⫽ 191
Ethnic origin, n ⫽ 168
White
Black
Asian
Other
Initial detection of breast carcinoma
Self-detected lump
Symptoms (i.e., pain, nipple discharge)
Physician-detected lump
Other or unknown
Presenting signs and symptoms
Breast mass
Nipple ulceration
Nipple bleeding
Nipple discharge
Nipple retraction
Local pain
Inflammatory skin
Other
No symptoms
Breast carcinoma laterality
Right
Left
Bilateralb
a
b
No.a
Range or %
63
171.5 cm
78.8 kg
(23–97)
(150.0–200.7)
(48.8–120.0)
156
3
4
5
(92.8%)
(1.8%)
(2.4%)
(3.0%)
139
67
12
11
(60.7%)
(29.3%)
(5.2%)
(4.8%)
196
18
20
13
60
22
11
70
3
(85.6%)
(7.9%)
(8.7%)
(5.7%)
(26.2%)
(9.6%)
(4.8%)
(30.6%)
(1.3%)
101
126
2
(44.1%)
(55.0%)
(0.9%)
Unless otherwise indicated, n ⫽ 229.
Of these 2 cases, 1 was synchronous and 1 was metachronous.
the upper outer quadrant (25.7%). Other tumor characteristics are shown in Table 3. Infiltrating ductal
carcinoma with or without other associated histologies was the most common tumor type, occurring in
147 cases (64.2%). Lobular carcinoma was identified in
6 cases (2.6%). From 1955–1965, 36.1% of cases had no
known tumor size and 13.9% of cases had tumors
larger than 5 cm. By 1986 –1996, the number of cases
with no reported tumor size had decreased to 6.1%,
and only 4.5% of tumors were larger than 5 cm. Axillary lymph nodes were not assessed in 68 cases. The
ratio of cases found to have negative versus positive
lymph nodes remained fairly constant over the study
period. Importantly, the proportion of cases with no
pathologic lymph node assessment decreased with
each decade, from 44.4% for 1955–1965 to 15.1% for
1986 –1996. Hormone receptor status was not reported
at all until the last two decades of the study period
(28.1% of cases for 1976 –1985 and 69.7% of cases for
1986 –1996). Over the entire period histologic grade
and nuclear grade were reported in only 31 and 28
cases, respectively. Twenty-seven patients (12.7%) had
Stage IV disease at presentation with bone as the most
632
CANCER February 1, 1999 / Volume 85 / Number 3
TABLE 3
Tumor Characteristics
Characteristic
Histologic tumor type (n ⫽ 215)
Infiltrating ductal
Infiltrating ductal with DCIS
Infiltrating lobular
Mixed infiltrating ductal and lobular
Infiltrating lobular with DCIS, LCIS, and papillary
DCIS
Inflammatory
Paget’s disease of the nipplea
Papillary adenocarcinoma
Papillary with DCIS
Otherb
Unspecified adenocarcinoma
Tumor Size (n ⫽ 185)
⬍1 cm
1–2 cm
2–5 cm
⬎5 cm
Multifocality (n ⫽ 148)
Marginsc (n ⫽ 160)
Negative
Positive (margins reexcised)
Positive (no further surgery)
Lymph node status (n ⫽ 161)
Positive
Negative
Positive receptor status
Estrogen (n ⫽ 66)
Progesterone (n ⫽ 65)
No.
%
105
38
4
1
1
4
1
6
3
1
9
42
48.8%
17.7%
1.9%
0.5%
0.5%
1.9%
0.5%
2.8%
1.4%
0.5%
4.2%
19.5%
6
63
98
18
10
3.2%
34.0%
53.0%
9.7%
6.8%
113
29
18
70.6%
18.1%
11.3%
92
69
57.1%
42.9%
60
49
90.9%
75.4%
a
Paget’s associated with unspecified adenocarcinoma in 2 of 6 cases; associated with infiltrating ductal
in 3 of 6 cases; pure Paget’s in 1 of 6 cases.
b
Other pathologies include sarcoma, epithelioid angiosarcoma, fibrosarcoma or stromal sarcoma,
infiltrating polygonal carcinoma, medullary carcinoma, trabecular duct carcinoma, gelatinous carcinoma, anaplasia.
c
In 57 cases pathology reports did not comment on margin status, and in 12 cases surgery was not
performed.
common site of metastases (occurring in 16 cases).
The proportion of cases with metastatic disease at
diagnosis remained constant for each decade of diagnosis (data not shown).
Primary treatment consisted of surgical resection
in 94.8% of cases. The majority (73.3%) underwent a
simple or modified radical mastectomy. Almost half of
all patients (47.2%) diagnosed during the period 1955–
1965 had a radical mastectomy. By the third decade
(1976 –1985), however only 5.6% had a radical mastectomy, and during the period 1986 –1996 no patients
were treated with radical mastectomy. Following the
first decade, simple or modified radical mastectomy
became the most popular surgical technique and was
used with increasing frequency over the last 2 decades
of the study period. Lumpectomy with or without axillary dissection was uncommon throughout the study
period, with only 8 cases in the entire group (3.5%)
treated with lumpectomy with axillary dissection and
12 (5.2%) treated with lumpectomy alone. Twelve patients (5.2%) did not have surgery, and in 1 case the
type of surgical intervention was not described. The 12
patients who did not undergo surgery as part of their
primary treatment either had Stage IV disease at diagnosis or were older than 71 years. Initial treatment
regimens included radiation for 131 patients (57.2%),
chemotherapy for 21 (9.2%), or additive hormone
therapy (i.e., tamoxifen) for 57 patients (24.9%). Nine
patients, all of whom were diagnosed in the first three
decades of the study period, underwent orchiectomy.
In 7 cases the orchiectomy was performed following
their breast carcinoma diagnosis. Two patients, both
with previous cancer diagnosis (prostate and testicular), were orchiectomized prior to their breast carcinoma diagnosis.
Primary and adjuvant treatment combinations are
presented per decade and for the group as a whole in
Table 4. For the first 3 decades, primary treatment
most frequently consisted of surgery plus radiation.
Surgery alone was the next most common primary
treatment choice. From 1986 to 1996, however, the
first and second most popular treatment regimens
were surgery plus radiation plus hormone therapy and
surgery plus hormone therapy. Of the 57 patients receiving additive hormone therapy, 75.4% of these were
diagnosed between 1986 and 1996. Approximately half
of patients receiving additive hormone therapy
(50.9%) continued treatment for less than 2 years.
Most of the 21 patients treated with chemotherapy at
this stage were diagnosed during the last two decades
of the study period (90.5% from 1976 to 1996). Thirteen patients received regimens in which cyclophosphamide, methotrexate, and 5-fluorouracil were combined. Only two patients (both diagnosed between
1986 and 1996) received anthracyclines. The type of
chemotherapy was unknown in four cases.
Age, lymph node status, tumor size, presence or
absence of metastatic disease at diagnosis, and decade
of diagnosis were shown to influence significantly the
choice of primary treatment modality by logistic regression analysis. Patients were more likely to be
treated with radiotherapy if they were diagnosed before 1976 (P ⫽ 0.0001), if their tumor was large (P ⫽
0.0003) or if their disease status was less advanced
than Stage IV (P ⫽ 0.0002). Because most chemotherapy use occurred in the last 2 decades of the study
period, analysis of the influence of clinical factors on
chemotherapy use was based on the subset of patients
diagnosed between 1976 and 1996. Patients were more
likely to be treated with chemotherapy if they had
positive lymph nodes (P ⫽ 0.002) or if they were
A Review of 229 Male Breast Carcinoma Cases/Goss et al.
633
TABLE 4
Primary and Adjuvant Treatment Combinations
Decade of diagnosis
Treatment combinationa
1955–1965
(n ⴝ 36)
1966–1975
(n ⴝ 56)
1976–1985
(n ⴝ 71)
1986–1996
(n ⴝ 66)
All decades
(n ⴝ 229)
Surgery alone
Surgery and radiation
Surgery and hormone therapyb
Surgery and chemotherapy
Surgery, radiation, and hormone therapy
Surgery, chemotherapy, and hormone therapy
Radiation and hormone therapy
11.1%
69.4%
5.5%
2.8%
5.5%
0%
2.8%
17.9%
71.4%
3.6%
0%
0%
0%
1.8%
32.4%
39.4%
8.4%
11.3%
0%
2.8%
4.2%
18.2%
7.6%
27.3%
6.1%
31.8%
3.0%
0%
21.4%
44.8%
12.2%
5.7%
10.0%
1.7%
2.2%
a
Does not include the following treatment combinations, which were used in ⱕ2 cases: surgery, radiation and chemotherapy; radiation, chemotherapy, and hormone therapy; chemotherapy and hormone therapy;
radiation, chemotherapy, or hormone therapy alone. Two patients received no treatment.
b
Includes additive hormone treatment with tamoxifen, diethylstilbestrol, or medroxyprogesterone and ablative hormone therapy (orchiectomy).
younger (P ⫽ 0.01). Because most hormone therapy
use occurred between 1986 and 1996, analysis of the
influence of clinical factors on hormone therapy use
was based on the subset of patients diagnosed in that
decade. Patients were more likely to receive hormone
therapy if they had large tumors (P ⫽ 0.02) or if they
had a positive family history (P ⫽ 0.01).
At last follow-up, 89 patients had relapsed and 30
patients had no appreciable disease free period. The site
of first recurrence was local in 13 cases. Six of these cases
were lymph node positive at diagnosis and 1 was lymph
node negative. Lymph node status was unknown in 6 of
these cases. In 7 cases locally recurrent disease was
treated by surgical excision either alone (5 of 13 cases) or
in combination with radiation (2 of 13 cases). Others
were treated by radiation alone (1 of 13 cases), hormone
therapy alone (1 of 13 cases), or a combination of radiation, chemotherapy, and hormone therapy (1 of 13
cases). Three cases were not treated for their local recurrence. The site of first recurrence was regional or distant
in 76 cases. Treatment for these included hormone therapy in 59 cases, radiation therapy in 51 cases, and chemotherapy in 23 cases. Most cases of regional recurrence
that occurred during follow-up involved the axilla (22 of
30 cases, 73.3%). The first site of distant metastases was
most often bone (44 of 78 cases, 56.4%) or lung (18 of 78
cases, 23.1%).
Median follow-up was 5.3 years, ranging from 3
weeks to 30 years. Survival curves for the entire group
are shown in Figure 1. The estimated 5-year DFS, OS,
LCR, and cause specific survival (CSS) for all patients
are 47%, 53%, 91%, and 63%, respectively. Median
survival from first relapse was 4.2 years. No significant
difference was observed in outcome for cases grouped
according to decade of diagnosis. As shown in Table 5,
univariate analysis (log rank test) indicated that DFS
was significantly longer for patients who had primary
surgery, smaller tumor sizes, negative lymph node
status, or positive estrogen receptor (ER) status. Adjuvant chemotherapy was associated with shorter DFS.
Younger age, smaller tumor size, negative lymph node
status, less advanced disease (⬍ Stage IV) at diagnosis,
primary surgical resection, or a positive family cancer
history had a positive influence on OS. Again, adjuvant
chemotherapy was associated with poorer OS. Factors
that were associated with better local control were ER
positive tumors, primary surgical resection (vs. no primary surgery), and mastectomy (vs. lumpectomy). Local control was worse in patients treated with adjuvant
chemotherapy. Only 6 local relapses occurred within
the group of patients known to be free of metastatic
disease at diagnosis. Any interpretation of local control data must take this small number of relevant
events into account. No significant difference in DFS,
OS, or LCR was observed in comparing patients with
ⱕ1 or ⬎1 risk factor or patients with a delay from first
appearance of symptoms to presentation to a physician of either ⱕ6 months or ⬎6 months.
Multiple regression analysis of prognostic factors
found that negative lymph node status (P ⫽ 0.0001,
Fig. 2) and adjuvant hormone treatment (P ⫽ 0.0368)
have a significant positive influence on DFS. Younger
age (P ⫽ 0.0003), negative lymph node status (P ⫽
0.0070, Fig. 2), absence of Stage IV disease at diagnosis
(P ⫽ 0.0001), and adjuvant hormone treatment (P ⫽
0.0401) had a positive influence on OS. Number of risk
factors, tumor size, radiotherapy, and chemotherapy
had no significant effect on either OS or DFS according to multiple regression analysis. A multivariate
analysis of factors affecting local control was not
found to be informative because only 6 local relapses
634
CANCER February 1, 1999 / Volume 85 / Number 3
FIGURE 1. A Kaplan–Meier plot of disease free survival, overall survival, and cause specific survival is shown for the group overall (5-year disease free survival:
47%; 5-year overall survival: 53%; 5-year cause specific survival: 63%).
occurred within the relevant group of patients (i.e.,
known to be free of metastatic disease at diagnosis).
Data analysis for associations between variables
in the group as a whole found only a trend for larger
tumors in older patients. Sixteen percent of patients
older than 63 years had tumors larger than 5 cm,
whereas only 3% of patients age 63 years or younger
had such large tumors (chi-square test, P ⫽ 0.014).
There was no association between age at diagnosis
and family cancer history or number of risk factors,
between lymph node status and family cancer history
or ER status, or between tumor size and ER status.
DISCUSSION
The proportion of men with clinical gynecomastia reported in our series is within the range reported in the
literature on male breast carcinoma.20 However, the
incidence of clinical gynecomastia in healthy adult
cohorts has been reported to be as high as 36%.21 In an
unselected series of 100 adult male autopsies, microscopic gynecomastia was reported in 55 cases. Clinical
gynecomastia was absent in all cases.22 Microscopic
examination of histologic specimens from 79 male
breast carcinoma cases23 found only 21 with gynecomastia. These numbers suggest that histologic gynecomastia may in fact occur less frequently in men with
breast carcinoma. Male breast carcinoma and gynecomastia may be unrelated except that both may be
caused by abnormal hormonal environments.24 A longitudinal, prospective histologic or mammographic
study of men with gynecomastia may be the only way
to resolve this question.
The median age at diagnosis in our series of men
was similar to that in previous studies,25–27 further supporting the theory that breast carcinoma generally presents almost a decade later in men than in women.
Unlike in women, for whom familial breast carcinoma is
often associated with an earlier age of onset,28 the median age at diagnosis was similar for patients with and
without a family history of cancer. However, underreporting of family cancer history in our series must be
considered in any interpretation of these data.
The relatively infrequent use of mammography in
our patient population, even within the last 2 decades
A Review of 229 Male Breast Carcinoma Cases/Goss et al.
635
TABLE 5
Prognostic Factors for Disease Free Survival, Overall Survival, and Local Control Rate: Results of the Log Rank Test
and Multiple Regressiona Analysis
DFS
OS
Factor
Log rank P value
Multiple regression P value
(risk ratio) n ⴝ 108
Log rank P value
Age (yrs)b
ⱖ63, ⬎63
Tumor size (cm)
⬍2, 2–5, ⬎5
Lymph node status
Negative, positive
ER status
Positive, negative
Stage IV disease at diagnosis
Absent, present
Primary surgery
Any surgery, no surgery
Type of surgery
Mastectomy, lumpectomy
Adjuvant chemotherapy
No chemo, chemo
Adjuvant hormone therapy
No hormone therapy
hormone/orchiectomy
Family cancer history
Present, absent
NS
NS
0.0088
0.0349
NS
0.0003
0.0001
0.0011
(2.37)
0.0023
0.0413
Multiple regression P value
(risk ratio) n ⴝ 120
0.0003
(1.05)
NS
0.0070
(2.03)
NS
0.0001
LCR
Log rank P value
NS
NS
NS
0.0061
0.0001
(5.98)
0.0057
0.0001
0.0027
NS
NS
0.0012
0.0110
NS
0.0493
NS
0.0252
NS
0.0368
(0.36)
NS
0.0401
(0.38)
NS
0.0483
NS
NS
DFS: disease free survival; OS: overall survival; LCR: local control rate.
a
Adjuvant radiotherapy and the presence of ⱖ1 risk factors had no significant influence on outcome by either univariate or multivariate analysis. Blank sections indicate exclusion of the factor from multivariate
analysis, as described in the “Methods” section.
b
Age was considered continuous for the Cox proportional hazards model.
of the study period, likely reflects both the large proportion of palpable tumors at presentation and the
absence of routine breast carcinoma screening in
men. The relative ease of diagnosis by physical exam
and the rarity of the disease does not support widespread mammographic screening. However, mammography is useful for differentiating between gynecomastia and malignancy in the male patient.
Previous descriptions of male breast carcinoma
series have reported a longer median delay between
the onset of symptoms and diagnosis, ranging from 6
to 18 months.4,29 Our data indicate a shorter median
(4 months), although the delay did range up to 2 years.
The significant difference in symptom duration detected between patients with (median, 1 month) or
without (median, 4 months) a family cancer history
may reflect a greater disease awareness in families
affected by cancer or more rapid tumor growth in
familial cases. Previously, duration of symptoms has
been correlated with decreased survival rates.30 –31
Analysis of our data, however, did not find that the
length of delay significantly influenced any of the survival variables. In view of this, it is difficult to explain
the positive association (P ⫽ 0.0483) between family
cancer history and DFS indicated by our univariate
analysis. It is likely that this result is unreliable due to
inconsistent reporting of family cancer history in our
series.
Consistent with previous reports,25,32 the central
subareolar region is the most common tumor site in
our series. In women the most frequent location is the
upper outer quadrant of the breast. In keeping with
data from the Surveillance, Epidemiology, and End
Results Program (SEER),33 there was no significant
excess in laterality (P ⫽ 0.097) observed in our series.
The SEER data did report an excess in left-sided breast
carcinoma in women. These gender differences in tumor laterality and location are likely related to the
relatively rudimentary nature of the male breast.
Our low incidence of bilateral male breast carcinomas, as compared with that among females, is consistent with other series. Reports of low disease bilaterality in men have suggested that this is due, in part,
to the later age of diagnosis of male patients resulting
in death due to other causes prior to contralateral
disease onset.32 Based on this point, primary con-
636
CANCER February 1, 1999 / Volume 85 / Number 3
FIGURE 2. Kaplan–Meier plots of male breast carcinoma patients according to lymph node status are shown. (A) Disease free survival at 5 years: lymph node
negative, 65%: lymph node positive, 35%. (B) Overall survival at 5 years: lymph node negative, 68%; lymph node positive, 47%.
A Review of 229 Male Breast Carcinoma Cases/Goss et al.
tralateral mammographic breast surveillance is unlikely to be important in clinical management of the
male breast carcinoma patient.
Our finding of lobular carcinoma in 2.6% of cases,
though higher than reported in previous series
(SEER),34 may still be an underestimate of the frequency of this histopathologic subtype. Historically,
histologic evaluation of male breast carcinomas has
not been more specific than a general diagnosis of
adenocarcinoma.35 The earlier assumption that lobular carcinoma could not occur in men due to the
absence of lobules in normal male breast tissue has
been dispelled. It is now known that true acinii and
lobules can develop in the male breast during exposure to increased estrogen.36 It is possible that some
cases of lobular carcinoma are also included in the
19.5% of cases with unspecified adenocarcinoma in
our series.
Despite a greater proportion of cases with large
tumors (⬎5 cm) during the first 2 decades (1955–1975)
of the study period, the proportion of cases with either
metastatic disease or positive lymph nodes at diagnosis was constant for each decade of diagnosis. Furthermore, tumor size was not found to influence OS by
multivariate analysis, and survival was similar for each
decade of diagnosis. Thus, though it appears that previously undetected, smaller male breast tumors are
now being identified, no significant improvement has
been made in terms of diagnosing prior to initiation of
the metastatic process.
The transition in primary surgical procedure in
our series from the radical mastectomy to the simple
or modified radical mastectomy mirrors the elimination of the radical mastectomy from standard treatment of female breast carcinoma in the 1970s.37
Lumpectomy was not performed often, even within
the most recent decade of our series. As is indicated by
the negative association between local control and the
use of lumpectomy in our series, the paucity of breast
tissue in men may often render adequate tumor excision by lumpectomy impossible. It is not likely that
lumpectomy with or without axillary dissection will be
performed as often on men as on women.
Treatment data from our group of male breast
carcinoma patients suggests that, despite an increase
in use during the last 2 decades, hormone therapy and
chemotherapy are used less often in the adjuvant
treatment of male breast carcinoma than female
breast carcinoma. Furthermore, the duration of adjuvant hormone therapy in our series was often less than
2 years. Studies of women with ER positive disease
have conclusively shown that adjuvant treatment with
tamoxifen improves overall survival,38 and that 5 years
of adjuvant tamoxifen is more beneficial than 2
637
years.39 It is possible then that more frequent use of
adjuvant tamoxifen prescribed for 5 years may improve the survival of ER positive male breast carcinoma patients. In addition, recent studies by the International Breast Cancer Study Group,40 the National
Surgical Adjuvant Breast and Bowel Project,41 and the
Southwestern Oncology Group42 of postmenopausal
women with breast carcinoma have shown that the
addition of chemotherapy to adjuvant tamoxifen may
be of benefit for both lymph node negative and lymph
node positive patients with either ER positive or ER
negative disease. These findings suggest that more
aggressive treatment, which includes adjuvant chemotherapy, may be beneficial to men. Chemotherapy use
may, however, be limited by the older average age of
male breast carcinoma patients. It is important to note
that the negative association between survival variables and adjuvant chemotherapy use found in our
univariate analysis was not confirmed by multivariate
analysis. This likely reflects administration of adjuvant
chemotherapy to patients with more advanced disease, as is indicated by our finding that chemotherapy
was more likely to be given to patients with positive
lymph nodes. Indeed, when disease stage and other
factors were included in the analysis, this negative
effect was no longer found.
Reported overall 5-year survival rates for male
breast carcinoma vary widely from 40% to 65%.27,43,44
The 53% OS at 5 years found in our series falls in the
middle of this range. It is important to note that the
Kaplan–Meier method for analyzing survival in our
series gives a biased, often overestimated CSS.45
Therefore, though the CSS has been estimated for our
series (shown in Fig. 1), the DFS and OS are more
informative.
Our finding that men diagnosed with breast carcinoma at a younger age have a survival advantage
over older patients has been observed previously in a
large, population-based study of 1429 patients from
Nordic countries.46 The larger sample size of the Nordic study enabled survival estimates to be more accurately controlled for the fact that men are diagnosed
with breast carcinoma at an older average age than
women therefore are more likely to die of causes other
than breast carcinoma. In female breast carcinoma
patients, age-associated mortality risk does not appear
to be linear. Women diagnosed when younger than 40
years or older than 50 years have been shown to have
a worse prognosis than those age 40 – 49 years.47– 49
Increased survival in this age group has been attributed to the increased likelihood of natural menopause
occurring during this time period, thereby depriving
newly diagnosed breast tumors of hormonal stimulation.49 The linear decrease in survival observed with
638
CANCER February 1, 1999 / Volume 85 / Number 3
increasing age among male breast carcinoma patients
may reflect the lack of significant age-dependent endocrine changes.
Lymph node involvement has often been cited as
a significant prognostic indicator in men.5,26,27,32,50
Reported 5-year survival rates for lymph node negative and lymph node positive patients vary. The 68%
overall survival estimate from our lymph node negative group of patients is lower than previous reports,
ranging from 77% to 100%.14,27,32 This difference is
partially attributable to the exclusion of patients with
pathologies associated with a better prognosis (i.e.,
pure DCIS) from our survival analysis. However, the
47% overall survival of lymph node positive patients is
in keeping with previous reports, ranging from 37%
to 61%.14,27,32
Comparison of the overall prognoses for male and
female breast carcinoma patients is controversial.
Studies reporting a worse prognosis for males have
suggested that the anatomy of the male breast may
provide less of a barrier to metastases or that more
aggressive tumor biology may be the basis of survival
variation.26,51,52 However, others have found that,
once separated according to stage or lymph node involvement, the prognosis is the same.5,25,27 Our 53%
OS at 5 years for the entire group is lower than the 64%
OS found for female breast carcinoma patients reported by the National Surgical Adjuvant Breast and
Bowel Project in the 1970s.53 When separated according to negative or positive lymph node status, the
5-year OS for the same group of women is 78% and
47%, respectively. Though OS for the lymph node negative male patients in this study is lower, the OS for the
lymph node positive male patients in our study is the
same. However, the lack of controlled, large male
breast carcinoma cohorts make a definitive prognostic
comparison between genders difficult.
Some modifications in clinical management were
observed over the 4 decades of the study period.
Smaller tumors were detected and axillary lymph
nodes were assessed more often by the end of the
study period. The standard primary surgical technique
evolved from the radical mastectomy to the simple or
modified mastectomy, and chemotherapy and hormone therapy were used slightly more often in the last
2 decades. However, these changes in management
had no detected impact on the outcomes of the patients in our series. As suggested previously, increased
use of adjuvant tamoxifen and adjuvant chemotherapy similar to that indicated in emerging guidelines
for female breast carcinoma may improve outcome for
this group with long term poor prognosis.
Clearly, these statements regarding treatment
may only be considered as suggestions at this point.
The data presented in this retrospective study is not
appropriate for accurately assessing the efficacy of any
given treatment; however, prospective clinical trials of
male breast carcinoma have yet to be performed.
Though the rarity of the disease limits the feasibility of
clinical trials, comprehensive, prospective data collection alone would augment our understanding of risk
factors and prognostic factors as well as improve clinical management specific for the male breast carcinoma patient. Though some improvement was seen
over the study period, the insufficient reporting of risk
factors, such as family history and pathology data
(including histologic grade, nuclear grade, and hormone receptor and axillary lymph node status), in our
series further emphasizes the need for rigorous,
thoughtful data collection. Thus further research is
required to determine the molecular biologic properties of male breast carcinoma, particularly in relation
to BRCA154 and BRCA2,55,56 to characterize the role of
adjuvant chemotherapy; and to determine the optimal
treatment of metastatic disease.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Sasco AJ, Lowenfels AB, Pasker-De Jong P. Epidemiology of
male breast cancer: a meta-analysis of published case– control studies and discussion of selected aetiological factors
[review article]. Int J Cancer 1993;53:538 – 49.
Boring CC, Squires TS, Tong T, Montegomery S. Cancer
statistics, 1994. CA Cancer J Clin 1994;44:7–26.
Izquierdo MA, Alonso C, DeAndres L, Ojeda B. Male breast
cancer: report of a series of 50 cases. Acta Oncol 1994;33:
767–71.
Crichlow RW. Carcinoma of the male breast. Surg Gynecol
Obstet 1972;134:1011–9.
Guinee VF, Olsson H, Moller T, Shallenberger RC, van den
Blink JW, Peter Z, et al. The prognosis of breast cancer in
males. Cancer 1993;71:154 – 61.
Hecht RJ, Winchester DJ. Male breast cancer. Am J Clin
Pathol 1994;102(S1):S25–S30.
Michaels BM, Nunn CR, Roses DF. Lobular carcinoma of the
male breast. Surgery 1994;115:402–5.
Everson RB, Li FP, Fraumeni JF, Wilson RE, Fishman J, Stout
D, et al. Familial male breast cancer. Lancet 1976;I:9.
Kozak FK, Hall JG, Baird PA. Familial breast cancer in males:
a case report and review of the literature. Cancer 1986;58:
2736 –9.
Anderson DE, Badzioch MD. Breast cancer risks in relatives
of male breast cancer patients. J Natl Cancer Inst 1992;84:
1111–7.
Evans DB, Crichlow RW. Carcinoma of the male breast and
Klinefelter’s syndrome. CA Cancer J Clin 1987;37:246 –51.
Kelsey JL, Gammon MD. Epidemiology of breast cancer.
Epidemiol Rev 1990;12:228 – 40.
Thomas DB. Male breast cancer. Epidemiol Rev 1993;15:
220 –31.
Erlichman C, Murphy KC, Elhakim T. Male breast cancer: a
13-year review of 89 patients. J Clin Oncol 1984;2:903–9.
McLachlan SA, Erlichman C, Liu F-F, Miller N, Pintilie M.
Male breast cancer: an 11-year review of 66 patients. Breast
Cancer Res Treat 1996;40:225–30.
A Review of 229 Male Breast Carcinoma Cases/Goss et al.
16. Hosmer DW, Lameshow S. Applied logistical regression.
New York: John Wiley & Sons, 1989.
17. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457– 81.
18. Mantel N. Evaluation of survival data and two new rank
order statistics arising in its consideration. Cancer Chemother Rep 1966;50:163–70.
19. Cox DR. Regression models and lifetables. J R Stat Soc Series
B 1972;34:187–202.
20. Korenman SG. The endocrinology of the abnormal male
breast. Ann N Y Acad Sci 1986;464:400 – 8.
21. Braustein GD. Gynecomastia. N Engl J Med 1993;328:490 –5.
22. Anderson JA, Gram JB. Male breast at autopsy. Acta Pathol
Microbiol Immunol Scand (A) 1982;90:191–7.
23. Scheike O, Visfeldt J. Male breast cancer 4: gynecomastia in
patients with male breast cancer. Acta Pathol Microbiol
Scand (A) 1973;81:359 – 65.
24. Donegan WL. Cancer of the breast in men. CA Cancer J Clin
1991;41:339 –54.
25. Stierer M, Rosen H, Weitensfelder W, Hausmaninger H,
Teleky B, Jakesz R, et al. Male breast cancer: Austrian experience. World J Surg 1995;19:687–93.
26. Salvadori B, Saccozzi R, Manzari A, Andreola S, Conti RA,
Cusumano F, et al. Prognosis of breast cancer in males: an
analysis of 170 cases. Eur J Cancer 1994;30A:930 –5.
27. Cutuli B, Lacroze M, Dilhuydy JM, Velten M, De Lafontan B,
Marchal C, et al. Male breast cancer: results of the treatments and prognostic factors in 397 cases. Eur J Cancer
1995;31A:1960 – 4.
28. Lynch HT. Breast cancer genetics: family history, heterogeneity, molecular genetic diagnosis, and genetic counseling.
Curr Probl Cancer 1996;Nov/Dec:331– 65.
29. Scheike O. Male breast cancer 5: clinical manifestations in
257 cases in Denmark. Br J Cancer 1973;28:552– 62.
30. Appleqvist P, Salmo M. Prognosis in carcinoma of the male
breast. Acta Chir Scand 1982;148:499 –502.
31. Roswit B, Edlis H. Carcinoma of the male breast: a thirtyyear experience and literature review. Int J Radiat Oncol Biol
Phys 1978;4:711– 6.
32. Borgen PI, Wong GY, Vlamis V, Potter C, Hoffmann B, Kinne
DW, et al. Current management of male breast cancer: a
review of 104 cases. Ann Surg 1992;215:451–9.
33. Weiss HA, Devesa SS, Brinton LA. Laterality of breast cancer
in the United States. Cancer Causes Control 1996;7:539 – 43.
34. Stalsberg H, Thomas DB, Rosenblatt KA, Jimenez LM, McTiernan A, Stemhagen A, et al. Histologic types and hormone receptors in breast cancer in men: a population-based
study in 282 United States men. Cancer Causes Control
1993;4:143–51.
35. Michaels BM, Nunn CR, Roses DF. Lobular carcinoma of the
male breast. Surgery 1994;115:402–5.
36. Schwartz IS. Small-cell breast carcinoma in men [letter].
Hum Pathol 1982;13:185.
37. Wilson RE, Donegan WL, Mettlin C, Smart CR, Murphy GP.
The 1982 national survey of carcinoma of the breast in the
United States by the American College of Surgeons. Surg
Gynecol Obstet 1982;159:309 –18.
38. Early Breast Cancer Trialists’ Cooperative Group (EBCTCG).
Treatment of early breast cancer: worldwide evidence in
1985–1990. A systematic overview of all available randomized trials in early breast cancer of adjuvant endocrine and
cytotoxic therapy. Treatment of early breast cancer. Volume
1. Oxford, UK: Oxford University Press, 1990.
639
39. Swedish Breast Cancer Cooperative Group. Randomized
trial of two versus five years of adjuvant tamoxifen for postmenopausal early stage breast cancer. J Natl Cancer Inst
1996;88:1543–9.
40. International Breast Study Group. Effectiveness of adjuvant
chemotherapy in combination with tamoxifen for nodepositive postmenopausal breast cancer patients. J Clin Oncol 1997;15:1385–94.
41. Fisher B, Dignam J, Wolmark N, DeCillis A, Emir B, Wickerham DL, et al. Tamoxifen and chemotherapy for lymph
node-negative, estrogen receptor positive breast cancer.
J Natl Cancer Inst 1997;89:1673– 82.
42. Albain K, Green S, Osborne K, Cobau C, Levine E, Ingle J, et
al. Tamoxifen (T) vs. cyclophosphamide, adriamycin, and
5-FU (CAF) plus either concurrent or sequential T in postmenopausal, receptor (⫹), node (⫹) breast cancer: a Southwest Oncology Group Phase III Intergroup Trial (SWOG
8814) [Abstract 450]. Proc Am Soc Clin Oncol 1997;16:128a.
43. Wagner JL, Thomas CR, Wui-jin K, Rudolph RH. Carcinoma
of the male breast: update 1994. Med Pediatr Oncol 1995;24:
123–32.
44. Ribiero G. Male breast carcinoma: a review of 301 cases
from the Christie Hospital and Holt Radium Institute,
Manchester. Br J Cancer 1985;51:115–9.
45. Caplan RJ, Pajak TF, Cox JD. Analysis of probability and risk
of cause-specific failure. Int J Radiat Oncol Biol Phys 1994;
29:1183– 6.
46. Adami HO, Hakulinen T, Ewertz M, Tretli S, Holmberg L,
Karjalainen S. The survival pattern in male breast cancer: an
analysis of 1429 patients from the Nordic countries. Cancer
1989;64:1177– 82.
47. Adami HO, Malker B, Holmberg L, Persson I, Stone B. The
relation between survival and age at diagnosis in breast
cancer. N Engl J Med 1986;315:559 – 63.
48. Palmer MK, Lythgoe JP, Smith A. Prognostic factors in breast
cancer. Br J Surg 1982;69:697– 8.
49. Mohle-Boetani J, Grosser S, Malec M, Whittemore AS. Survival advantage among patients with breast cancer diagnosed at 45 to 49 years of age. N Engl J Med 1986;315:587.
50. Crichlow RW. Breast cancer in men. Semin Oncol 1974;1:
145–52.
51. Joshi MG, Lee AK, Loda M, Camus MG, Pedersen C, Heatley
GJ, et al. Male breast carcinoma: an evaluation of prognostic
factors contributing to a poorer outcome. Cancer 1996;77:
490 – 8.
52. Ciatto S, Iossa A, Bonardi R. Male breast carcinoma: review
of a multicenter series of 150 cases. Tumori 1990;76:555– 8.
53. Fisher B, Slack N, Katrych D, Wolmark N. Ten year follow-up
results of patients with carcinoma of the breast in a cooperative clinical trial evaluating surgical adjuvant chemotherapy. Surg Gynecol Obstet 1975;140:528 –34.
54. Struewing JP, Brody LC, Erdos MR, Kase RG, Giamberresit R,
Smith SA, et al. Detection of eight BRCA1 mutations in 10
breast/ovarian cancer families, including 1 family with male
breast cancer. Am J Hum Genet 1995;57:1–7.
55. Tavtigian SV, Simard J, Rommens J, Shattuck-Eidens D,
Neuhausen S, Merajver S, et al. The complete BRCA2 gene
and mutations in chromosome 13q-linked kindreds. Nat
Genet 1996;12:333–7.
56. Couch FJ, Farid L, DeShano ML, Tavtigian SV, Calzone K,
Campeau L, et al. BRCA2 germ line mutations in male breast
cancer cases and breast cancer families. Nat Genet 1996;13:
123–5.
Документ
Категория
Без категории
Просмотров
2
Размер файла
143 Кб
Теги
646
1/--страниц
Пожаловаться на содержимое документа