Archives of Gynecology and Obstetrics

, Volume 282, Issue 4, pp 427–432

Comparison of early onset breast cancer patients to older premenopausal breast cancer patients

Authors

    • Universitätsfrauenklinik Ulm
  • Jochem Koenig
    • Institut für Medizinische Biometrie, Epidemiologie und Informatik (IMBEI)
  • Kathrin Kuhr
    • Institut für Medizinische Biometrie, Epidemiologie und Informatik (IMBEI)
  • Kathrin Strunz
    • Universitätsfrauenklinik Ulm
  • Verena Geyer
    • Universitätsfrauenklinik Ulm
  • Christian Kurzeder
    • Universitätsfrauenklinik Ulm
  • Ziad Atassi
    • Universitätsfrauenklinik Ulm
  • Maria Blettner
    • Institut für Medizinische Biometrie, Epidemiologie und Informatik (IMBEI)
  • Rolf Kreienberg
    • Universitätsfrauenklinik Ulm
  • Achim Woeckel
    • Universitätsfrauenklinik Ulm
Gynecologic Oncology

DOI: 10.1007/s00404-009-1339-y

Cite this article as:
Varga, D., Koenig, J., Kuhr, K. et al. Arch Gynecol Obstet (2010) 282: 427. doi:10.1007/s00404-009-1339-y

Abstract

Background

The objective of this study was to show differences between breast cancer patients ≤35 and >35 years with regard to tumor characteristics and to present the patient-relevant outcomes overall survival (OAS) and recurrence-free survival (RFS).

Methods

We analyzed data from 119 women aged 35 years or younger with breast cancer and compared multiple parameters against breast cancer patients between 36 and 55 (n = 1,097), all pre-menopausal. Data were adjusted for tumor characteristics and therapy.

Results

There was no statistically significant difference in tumor size, axillary lymph node involvement, and histological subtypes. On the contrary, grading lymphovascular invasion and receptor negativity showed statistically significant differences. Unadjusted hazard ratio are 2.11 (1.32–3.39) (OAS) and 1.92 (1.35–2.73) (RFS). Multi-adjusted hazard ratio are 2.97 (1.70–5.18) (OAS) and 2.11 (1.42–3.13) (RFS).

Conclusions

In conclusion, young breast cancer patients still have a poor prognosis. Even after adjustment of the data, OAS and RFS showed a worse prognosis. Normal prognostic factors like tumor size, axillary lymph node involvement, and grading can therefore be not the explanation for the more aggressive disease progress within early onset breast cancer patients.

Keywords

Early onset breast cancerOutcomeRisk assessment

Introduction

Only 5–10% of all breast cancer cases occur in women below the age of 35 years. The prognosis of early onset breast cancer is worse, especially in the case of BRCA 1- and 2-mutation carriers [1, 2]. Several studies have shown that there are significantly higher rates of locoregional recurrence and lower rates of overall survival in young breast cancer patients [35]. In addition, breast cancer in younger women may be more aggressive and less likely to respond to treatment [6]. Most studies analyzing histopathological factors have reported that early onset breast cancer is defined by higher histological grading, greater tumor size, and higher lymph node involvement [7, 8]. The situation of estrogen and progesterone receptor expression is controversial due to the analyzed samples of the different publications [9]. For other histopathological features, early onset breast tumors present similar to sporadic cancers of older women.

There are various reasons why early onset breast cancer is associated with a poorer prognosis. Diagnosing breast cancer in younger women is more difficult because their breast tissue generally has a higher density than breast tissue in older women, and hence the tumors are likely to be of greater size at first diagnosis [10]. Due to this fact, a combination of ultrasound, MRI and Mammography is recommended [11]. In addition, some younger women or their physicians ignore warning symptoms, and delaying consultation could lead to late stage diagnoses [12]. Breast cancer in younger women may show often a basal cell like subtype [13] that is associated with a more aggressive behavior. The aim of this study was to examine the experience of our institution in treating young breast cancer patients, focussing on the clinical presentation and pathological findings.

Materials and methods

Study population and data collection

Between 1992 and 2006, 6,855 breast cancer patients were treated in the Department of Obstetrics and Gynecology, University of Ulm Medical Center or associated facilities. These patients were identified from the pathology report database.

Women aged 35 or younger with primary breast cancer were analyzed and compared to that of breast cancer patients between 36 and 55, all of them premenopausal. Menopause was defined as absence of period for 1 year.

Patients with primary metastatic disease and noninvasive breast tumors (i.e. ductal and lobular carcinomas in situ) were excluded from the study, as were patients who did not receive surgery after primary systemic therapy and patients with second primary tumors. In women with multiple primaries, the tumor with the worst prognostic features was included. The total number of patients after these exclusions was 119 aged ≤35 years, and 1,097 aged >35 years.

A retrospective chart review was conducted to abstract date of diagnosis, TNM stage, histological type, grade, presence of lymphatic or vascular invasion, estrogen receptor and progesterone receptor status (see Table 1). HER2neu status was excluded as there were around 30% unknown status for the patients treated from 1992 to 1998. The recorded data included type and date of surgery, type, and date of adjuvant systemic therapy and detailed information on the adjuvant radiotherapy administered, including total dose and target volume (see Table 2, data on total dose and target volumes not shown). Therapy was aligned on the German S3 guidelines [14]. Data on first recurrences and secondary primary tumors were also collected. For deceased patients, time and cause of death were recorded. Queries were sent to physicians involved in follow-up care, to local death registries, and to patients to determine the recurrence and survival status of patients with incomplete follow-up records.
Table 1

Total numbers and percentage of the analyzed collective in terms of tumor size, axillary node involvement, histology, grading, lymphovascular infiltration, hormone receptor expression, recurrence risk

Characteristic

Cases ≤35 years at diagnosis (N = 119)

Cases >35 years at diagnosis (N = 1,097)

Number

Percentage

Number

Percentage

T stage

 TX

2

1.7

8

0.7

 T1

78

66

660

60

 T2

32

27

370

34

 T3

2

1.7

37

3.4

 T4

5

4.2

22

2

 P valueb

0.15

   

Metastatically involved axillary lymph nodes

 Unknown

    

 None

65

55

630

57

 1–3

36

30

300

27

 4–9

11

9.2

95

8.7

 ≥10

7

5.2

59

5.4

 P valuea

0.91

   

Histology

 Lobular invasive

7

5.9

117

11

 Ductal invasive

91

76

790

72

 Special type

21

18

90

17

 P valuea

0.26

   

Grade

 X

4

3.4

14

1.3

 1

4

3.4

89

8.1

 2

46

39

579

53

 3

65

55

415

38

 P valuea

<0.001

   

Lymphovascular infiltration

 Yes

37

31

224

20

 No

78

66

860

78

 Unknown

4

3.4

13

1.2

 P valuea

0.005

   

Estrogen receptor status

 Uncertain

2

1.7

10

0.91

 Negative

45

38

284

26

 Positive 1–5

35

29

388

35

 Positive >5

37

31

415

38

 P valuea

0.02

   

Progesterone receptor status

 Uncertain

2

1.7

11

1

 Negative

47

39

263

24

 Positive 1–5

38

32

305

28

 Positive >5

32

27

518

47

 P Valuea

<0.001

   

Endocrine responsiveness

 Unknown

2

1.7

11

1

 Negative

36

30

213

19

 Uncertain

32

27

260

24

 Positive

49

41

613

56

 P valuea

0.004

   

More than 1 tumor

 No

115

97

1,058

96

 Yes

4

3.4

39

3.6

 P valueb

1.00

   

Participation in clinical trial

 No

99

83

908

83

 Yes

20

17

189

17

 P valuea

1.00

   

Risk for recurrence

 Unknown

6

5

87

7.9

 Low

51

4.6

 Medium

90

76

760

69

 High

23

19

199

18

 P valueb

0.02

   

The numbers were tested for statistical significance between early onset and older breast cancer patients. P values are shown

aχ2 test for homogeneity in contingency table analyses

bFisher’s exact test

Table 2

Total numbers and percentage of treatment modalities

 

Cases ≤35 years at diagnosis (N = 119)

Cases >35 years at diagnosis (N = 1,097)

Number

Percentage

Number

Percentage

Breast surgery

 Mastectomy

31

26

277

25

 BCS

88

74

820

75

Axilla surgery

 <10 lymph nodes removed

37

31

282

26

 ≥10 lymph nodes removed

82

69

815

74

Radiation therapy

 No

23

19

221

20

 Yes

96

81

876

80

Endocrine therapy

 No

61

51

441

40

 Yes

58

49

656

60

Chemotherapy

 No

18

15

401

37

 Yes, normal

50

42

341

31

 Yes, aggressive

51

43

355

32

BCS breast-conserving surgery, ME mastectomy

Study variables

The objective of the study was to describe differences between subgroups of breast cancer patients aged ≤35 years and pre-menopausal patients >35 years with regard to tumor characteristics and to present the patient-relevant outcomes overall survival (OAS) and recurrence-free survival (RFS). Survival time for the clinical outcomes OAS and RFS was calculated from date of histological diagnosis to these endpoints, censoring at the date of last contact.

Data analysis

The influence of age at diagnosis was assessed by comparing breast cancer patients diagnosed under and over age 35. Differences in characteristics of the populations were tested using Pearson’s or Fisher’s exact test.

Survival effects were evaluated using the Cox proportional hazards model. In addition to an unadjusted model, a multivariable model with the following baseline characteristics T stage, N stage, grading, risk for recurrence, lymphovascular invasion, endocrine responsiveness, year of diagnosis, adjuvant treatment, Hazard ratios were presented with a 95% confidence interval, and the models were fitted for overall and recurrence-free survival. P values under 0.05 are reported as statistically significant, adjustment for multiple testing was not applied.

Results

Study population

The sample of breast cancer patients aged ≤35 years consisted of 119 women with a mean age at diagnosis of 32.2 2.9 years. They were compared to 1,097 women aged >35 years, the mean age was 44.2 4.5 years. The mean follow up was 59.6 months for cases diagnosed at age ≤35 years, and 62.1 month for cases diagnosed at age >35. In the group of younger patients, 21 events were recorded and 95 events occurred in the group of women >35.

Table 1 shows distribution according to the TNM formula, endocrine receptor status, grading, lymphovascular infiltration, participation in clinical trials. The two groups were compared in order to get information on statistical significant differences. As can be seen in Table 1, there were no statistical differences in tumor size, metastatically involved axillary lymph nodes and in histological type, whereas grading (P < 0.001), lymphovascular infiltration (P = 0.005) and endocrine responsiveness (P = 0.004) showed significant differences. The majority of patients was diagnosed with T1 stage (66% in the group under 35 years and 60% in the group above 35 years) and had no metastatically involved axillary lymph nodes (55 and 57%, respectively). Histological type showed also a normal frequency pattern in the cases with early onset breast cancer. Grading and lymphovascular infiltration showed statistically significant differences between the two groups of patients. The majority of patients in the early onset group had grade 3 grading at time of diagnosis, and a third of them showed lymphovascular infiltration, indicating a more aggressive tumor behavior as compared to older patients with breast cancer. Endocrine responsive tumors were less common among women diagnosed ≤35 years (41% compared to 56% in those >35). The risk for recurrence was higher in the group of women aged ≤35 years at diagnosis (P = 0.02). Per definition, these patients had at least a medium risk. Both groups participated equally in clinical trials.

Survival effects

Cox models were used to determine hazard rates for age groups. As shown in Table 3, survival, was poorer for younger patients than for patients >35. In unadjusted analyses, the hazard ratios for diagnosis at age ≤35 years were 2.11 (1.32–3.39) for overall survival and 1.92 (1.35–2.73) for recurrence free survival. After adjustment for major confounding factors, the calculated hazard ratios were 2.97 (1.70–5.18) for overall survival and 2.11 (1.42–3.13) for recurrence free survival.
Table 3

Overall and recurrence free survival by age at diagnosis

 

Cases ≤35 years at diagnosis

Cases >35 years at diagnosis

Overall survival

 Unadjusted HR (95% CI)

2.11 (1.32–3.39)

1.00

 N

119

1,097

 N events

21

95

 Multi-adjusted HR (95% CI)a

2.97 (1.70–5.18)

1.00

 Nb

110

1,001

 N events

19

84

Recurrence free survival

 Unadjusted HR (95% CI)

1.92 (1.35–2.73)

1.00

 Nb

117

1,081

 N events

37

194

 Multi-adjusted HR (95% CI)a

2.11 (1.42–3.13)

1.00

 Nb

108

987

 N events

34

164

HR hazard ratio, CI confidence interval

aAdjusted for T stage, N stage, grading, risk for recurrence, lymphovascular invasion, endocrine responsiveness, year of diagnosis, adjuvant treatment, and guideline compliance

bMissing data on adjustment factors

Discussion

Breast cancer in young women ≤35 years of age is still associated with a poor prognosis [1, 2]. Several studies have shown that there are significantly higher rates of locoregional recurrence and lower rates of overall survival in young breast cancer patients [3]. This seems to be a problem of the primarily more aggressive presentation of early onset tumors [7]. Primary breast cancers in young women are more often poorly differentiated, and are more often associated with lymphovascular infiltration.

In contrast to data from the literature, we could not confirm that young breast cancer patients have a greater tumor size at diagnosis. In our collective, most patients had T1-tumors. This may be a result of better diagnostic modalities in recent years, especially with growing use of MRI in unclear cases. Additionally, there has been a great effort to offer regular screening controls to young patients with a family history of breast cancer comprising patients with a genetic predisposition. For these high-risk women, regular controls with ultrasound, mammography, and MRI are available. The combination of these screening modalities is necessary to detect primary breast cancer, as well as recurrences in young high-risk women. Robinson et al. showed that younger patients are significantly less likely to have their second primary tumor detected by routine follow-up mammography, compared to older patients. On the other hand, with the use of MRI, there exists a very sensitive diagnostic tool to detect breast cancer especially in young high-risk patients at an early stage [15].

Besides the presentation of different histopathological features of early onset breast cancers, according to literature they seem to have a different molecular subtype compared to tumors from older patients. Looking at molecular subtypes, Ihemelandu et al. [13] demonstrated, that a basal cell like showed an increased association with clinicopathological variables portending a more aggressive clinical course. These could not been confirmed within our own study. Regarding adjuvant therapy, it may be necessary to further classify the group of young breast cancer patients to predict which patient would profit from a more aggressive therapy. Bartelink et al. for example, showed that especially in young women with breast conserving therapy a radiation boost to the tumor bed can reduce in-breast recurrence largely compared to older patients with breast conserving treatment [16].

The most important result of our data concerns the outcome after multi adjustment. When data is adjusted for tumor characteristics and for treatment modalities, there is still a far better prognosis for the elderly breast cancer patients. This fact is not dependent on tumor size, grading, axillary lymph node involvement, or lymphovascular infiltration, and not on the histological type, as all of these factors were adjusted for. After adjustment, early onset breast cancer patients are facing an overall survival hazard ratio of 2.97. This leads to the assumption that there are further parameters within young breast cancer patients which define prognosis for overall and recurrence free survival.

In conclusion, it should be noted that in spite of progress regarding adjuvant treatment and screening modalities in the recent years, young breast cancer patients still have a poorer prognosis compared to older women with breast cancer. This is also true for adjusted models including tumor characteristics and therapy.

One can speculate that different gene expression profiles explain the differences between the young and the elderly, or there are other factors defining the biological aggressiveness of the tumor. Maybe it can be explained by the tumor microenvironment. Taken together it is only speculation presently. Further research on the identification of subgroups of young breast cancer patients with a tailored adjuvant therapy for them is necessary.

Conflict of interest statement

None.

Copyright information

© Springer-Verlag 2010