|
RESULTS
The patients were divided into two groups by age. The 101
patients less than 36 years ranged in age from 21 to 35 with
a mean of 32. The 631 comparison patients ranged in age from
36 to 85 with a mean of 58. Significant differences between the
two groups were noted in risk factors, clinical presentation,
pathologic findings, extent of disease, surgical and adjuvant
treatment, and outcomes.
The racial distribution of patients younger than 36 years
was significantly different from that of patients 36 years and
older (p=0.002) (Table 1). Seventy-seven percent of patients
younger than 36 years were Caucasian compared with 60% of patients
36 years and older. Patients younger than 36 years had significantly
lower body mass index (23.0 versus 26.7 kg/m2, p<0.0001) as
a consequence of both being taller (162 cm versus 159 cm, p=0.042)
and having lower body weight (62 kg versus 70 kg, p=0.001). Patients
younger than 36 years had fewer pregnancies (median 1 versus
2) and significantly fewer live births (median 1 versus 2, p<0.001)
than patients 36 years and older. A family history of breast
cancer was significantly more frequent among patients younger
than 36 years (43% versus 31%, p=0.021). Patients less than 36
years were significantly more likely to smoke (40% versus 27%,
p=0.014).
Table 1. Comparison of Patients <36 Years and >36
Years for Women Diagnosed with Stage 0 to Stage III Breast Cancer
|
|
Characteristic |
< 36 y |
> 36 y |
p Value |
|
|
n |
101 |
631 |
— |
|
Age (y) (mean) |
32 |
58 |
— |
|
(median:range) |
33:21-35 |
56:36-85 |
— |
|
Race (n) |
101 |
603 |
— |
|
Caucasian |
78 (77%) |
361 (60%) |
0.002 |
|
Hispanic |
11 (11%) |
132 (22%) |
— |
|
African American |
7 (7%) |
99 (16%) |
— |
|
Other |
5 (5%) |
11 (2%) |
— |
|
Height (cm) (mean) |
162 |
159 |
0.042 |
|
Weight (kg) (mean) |
62 |
70 |
0.001 |
|
Body mass index (kg/m2) (mean) |
23 |
26.7 |
<0.001 |
|
Age at menarche (y) (median) |
13 |
13 |
0.380 |
|
Pregnancies (median) |
1 |
2 |
0.103 |
|
Births (median) |
1 |
2 |
<0.001 |
|
Age first birth (y) (median) |
27 |
25 |
0.276 |
|
Family history breast cancer |
41/95 (43%) |
176/565 (31%) |
0.021 |
|
Smoking history |
38/96 (40%) |
135/497 (27%) |
0.014 |
|
Presentation (n) |
99 |
496 |
— |
|
Palpable mass |
86 (87%) |
273 (55%) |
<0.001 |
|
Mammographic calcium |
4 (4%) |
91 (18%) |
— |
|
Mammographic mass |
2 (2%) |
111 (22%) |
— |
|
Other |
7 (7%) |
21 (4%) |
— |
|
Mammography |
|
|
|
|
Positive/suspicious |
65/78 (83%) |
485/553 (88%) |
0.280 |
|
Diagnostic method (n) |
100 |
577 |
— |
|
Excisional biopsy |
45 (45%) |
312 (54%) |
0.154 |
|
Fine needle cytology |
50 (50%) |
229 (40%) |
— |
|
Core needle biopsy |
5 (5%) |
36 (6%) |
— |
|
Patients younger than 36 years rarely had mammography before
diagnosis and most (87%) presented with a palpable mass, which
was proved to be cancer by fine needle aspiration cytology. When
mammography was performed, usually after the diagnosis was made,
it was frequently (83%) positive. Patients 36 years and older
presented with mammographic findings significantly more frequently
(40%) and fewer presented with a palpable mass (55%). Spot localization
breast biopsy for mammographic findings was significantly more
common in the patients 36 years and older (40% versus 6%, p<0.001).
The consequence of presenting more often with a palpable mass
as opposed to mammographic findings is that patients younger
than 36 years were diagnosed with breast cancer when it was significantly
more advanced (p<0.001) (Table 2). Patients less than 36 years
had significantly larger invasive tumors (median 2 cm versus
1.5 cm, p<0.001), were more likely to have nodal involvement
(50% versus 38%, p=0.022), and had more nodes involved (median
1.0 versus 0, p=.010). The majority of patients younger than
36 years were diagnosed with stage II or III disease, but the
majority of patients 36 years and older were diagnosed with stage
0 or I (p < 0.001).
Table 2. Pathologic Findings of Patients <36 Years and
> 36 Years for Women with Stage 0 to Stage III Breast
Cancer
|
|
|
< 36 y |
> 36 y |
p Value |
|
|
Histopathology (n) |
101 |
539 |
— |
|
Infiltrating ductal |
84 (83%) |
417 (77%) |
0.337 |
|
Infiltrating lobular |
5 (5%) |
43 (8%) |
— |
|
Ductal carcinoma in situ |
12 (12%) |
79 (15%) |
— |
|
Tumor differentiation (n) |
73 |
450 |
— |
|
Well |
3 (4%) |
41 (9%) |
<0.001 |
|
Moderate |
12 (16%) |
213 (47%) |
— |
|
Poor |
58 (80%) |
196 (44%) |
— |
|
Tumor size (cm) (median)* |
2.0 |
1.5 |
<0.001 |
|
n |
89 |
504 |
— |
|
0–2 cm |
48 (54%) |
359 (71%) |
— |
|
2.1–5 cm |
32 (36%) |
110 (22%) |
0.005 |
|
> 5.1 cm |
9 (10%) |
35 (7%) |
— |
|
Estrogen-positive |
44/92 (48%) |
349/507 (69%) |
<0.001 |
|
Progesterone-positive |
35/92 (38%) |
277/493 (56%) |
0.001 |
|
Aneuploid |
28/40 (70%) |
110/225 (49%) |
0.013 |
|
High S phase |
20/34 (59%) |
59/203 (29%) |
0.001 |
|
Node-positive† |
44/89 (51%) |
145/413 (35%) |
0.023 |
|
Involved nodes (median)† |
1.0 |
0 |
0.010 |
|
Margins |
|
|
|
|
Close/involved |
21/101 (21%) |
68/423 (16%) |
0.025 |
|
Stage (n) |
101 |
538 |
|
|
0 |
12 (12%) |
79 (15%) |
|
|
I |
28 (28%) |
225 (42%) |
<0.001 |
|
II |
47 (47%) |
214 (40%) |
|
|
III |
14 (14%) |
20 (4%) |
|
|
Surgery (n) |
101 |
606 |
|
|
Breast conservation |
41 (41%) |
473 (78%) |
<0.001 |
|
Mastectomy |
60 (59%) |
133 (22%) |
|
|
Neoadjuvant chemotherapy |
11/101 (11%) |
26/631 (4%) |
0.004 |
|
Postoperative chemotherapy |
75/94 (80%) |
338/631 (54%) |
<0.001 |
|
Tamoxifen |
21/58 (36%) |
282/485 (58%) |
0.001 |
|
Radiation therapy |
45/84 (54%) |
482/631 (76%) |
<0.001 |
|
* Size of invasive component.
† Ductal carcinoma in situ.
In addition to being larger with more nodal involvement, young
patients' tumors tended to be more aggressive. Their cancers
were more poorly differentiated (80% versus 44%, p < 0.001),
more likely to be estrogen receptor-negative (52% versus 31%,
p < 0.001), more likely to be aneuploid (70% versus 49%, p
= 0.013), and more likely to have high S phase (59% versus 29%,
p = 0.001).
The majority (59%) of patients younger than 36 years were
treated with mastectomy but the majority (78%) of patients 36
years and older were treated with breast conservation (p <
0.001). Breast conservation was accompanied by radiation therapy
for all patients younger than 36 years and for 90% of the patients
36 years and older. Adjuvant chemotherapy was more frequently
used in patients younger than 36 years (80% versus 54%, p <
0.001) and tamoxifen was more frequently used in patients 36
years and older (36% versus 58%, p = 0.001).
Patients younger than 36 years fared worse than patients 36
years and older (Table 3). Patients younger than 36 years had
a significantly higher rate of both local recurrence and distant
metastases. The cumulative 5-year local recurrence-free survival
was 87% for patients younger than 36 years compared with 91%
of patients 36 years and older (p = 0.011) (Table 3; Fig. 1).
The higher rate of local recurrence in patients younger than
36 years was from an excess number of local recurrences in patients
treated with breast conservation. The cumulative 5-year local
recurrence-free survival for the 39 patients younger than 36
years treated with breast conservation was 87% compared with
94% in the 346 patients 36 years and older (p=0.036). This difference
was not attributable to a higher rate of positive margins, involved
nodes, or larger tumor size among patients younger than 36 years.
Table 3. Local and Distant Disease-Free Survival
|
|
Age |
n |
Recurred |
Cumulative 5-year recurrence-free survival |
p Value* |
|
|
Local Recurrence |
|
|
|
|
|
< 36 y |
95 |
11 |
87% |
0.011 |
|
> 36 y |
471 |
29 |
91% |
|
|
Distant Recurrence |
|
< 36 y |
97 |
13 |
84% |
0.044 |
|
> 36 y |
470 |
46 |
85% |
|
|
*p value is from the Breslow test comparing the
Kaplan-Meier survival curves.
Figure 1. Five-year cumulative survival free of local
recurrence for women under age 36 and for women ages 36 and older.
The cumulative 5-year distant disease-free survival was 84%
for patients younger than 36 years compared with 85% for patients
36 years and older (p=0.044) (Table 3; Fig. 2). This is statistically
significant because a difference in distant disease-free survival
was observed early on, but there was little difference in cumulative
survival at 5 years.
Figure 2. Five-year cumulative survival free of distant
metastases for women under age 36 and for women ages 36 and older.
Cox regression was used to evaluate the significance of age
as a prognostic factor after controlling for all other variables
potentially related to survival. Because the prognostic influence
of young age at diagnosis appeared to diminish during followup,
the estimated influence of age was allowed to change with time.
Nodal involvement (p<0.001; relative risk [RR]: 1.19; 95%
confidence limit [CL]: 1.11 to 1.28), chemotherapy (p=0.016;
RR: 0.38; 95% CL: 0.13 to 1.07), and tamoxifen (p=0.040; RR:
1.60; 95% CL: 1.00 to 2.55) were significantly related to local
disease-free survival. After consideration for nodal involvement,
chemotherapy, and tamoxifen, no other variable was significantly
related to local disease-free outcomes. Tumor size (p<0.001;
RR: 1.03; 95% CL: 1.01 to 1.04) and nodal involvement (p<0.001;
RR: 1.17; 95% CL: 1.09 to 1.24) were significantly related to
distant disease-free survival. After consideration for tumor
size and nodal involvement no other variable was significantly
related to distant disease-free outcomes. The results were similar
when the influence of age was considered as constant over time.
Introduction
| Methods
| Results | Discussion
| References
JACS |
