Breast cancer diagnosed during pregnancy or 1 to 2 years after birth often occurs at a late stage. Little is known about tumor characteristics in the high-risk period shortly after a childbirth. We here explore whether stage of disease differs according to timing of births. Results are based on 22,351 Norwegian breast cancer patients of parity 0 to 5, ages 20 to 74 years. The proportion of stage II to IV tumors was considerably higher among parous than nulliparous women at age <30 years (52.7% versus 36.8%, P = 0.009), but similar or lower in other age groups (Pinteraction = 0.029). In general, the largest proportion of stage II to IV tumors was found among women diagnosed during pregnancy or <2 years after birth. However, among women with late-age births (first or second birth ≥30 years, third birth ≥35 years), as well as women with an early second birth (<25 years), the proportion with advanced disease was rather similar or even higher among those diagnosed 2 to 6 years after birth (49.3-56.0%). The association between clinical stage and time since birth reached statistical significance among women with a late first or second birth and among all triparous women (P ≤ 0.032). The subgroups with a high proportion of advanced disease 2 to 6 years after birth corresponded quite well to those previously found to have the most pronounced transient increase in risk after birth. Thus, pregnancy hormones may have a progressive effect on breast cancer tumors in addition to a possible promoting effect. A potential effect of prolactin is discussed. (Cancer Epidemiol Biomarkers Prev 2006;(15)1:65–9)

Pregnancy hormones have been suggested to act as promotors for breast tumors and may explain the observed transient increase in risk of breast cancer shortly after a childbirth (1). It is possible that the strong hormonal exposure during pregnancy also has a progressive effect by accelerating growth of existing tumors, possibly after a promoting effect that triggers transformation from premalignant to malignant breast cells. If so, one might expect a higher proportion of advanced disease among women diagnosed with breast cancer shortly after birth. Breast cancer diagnosed during pregnancy or 1 to 2 years after birth often occurs at a late stage but it is not clear whether this pattern relates to diagnostic delay or accelerated growth owing to increased vascularity, hormonal exposure, or suppressed immune system during pregnancy (2).

The observed transient increase in risk in breast cancer after a term birth, although mainly examined in relation to the first birth, reaches a maximum after ∼5 years (1). In our recent study (3), we observed a transient increase in risk also after subsequent births. The timing of the adverse effect differed somewhat according to the order of the birth and the effect was most pronounced after childbirths at age 30 years or older. Little is known about tumor characteristics in the high-risk period after birth. In the present study, we, therefore, focus on the breast cancer cases in our study cohort and examine whether clinical stage of disease at time of diagnosis differs according to time since a childbirth in subgroups defined by parity and maternal age at birth. We also compare stage of disease in parous and nulliparous women by age.

Our study was based on information for 22,351 Norwegian breast cancer patients ages 20 to 74 years, of parity 0 to 5 at time of diagnosis. The cancer cases considered were incident cases of breast cancer diagnosed in Norway during the period 1955 through 1999, and were part of a large, population-based prospective study on associations between reproductive factors and cancer risk (3). The population at risk comprised Norwegian women born in 1925 to 1979. The oldest birth cohorts (born in 1925-1934) were 21 to 30 years in 1955 when entering the risk set, whereas younger cohorts were considered to be at risk from the age of 20 years. At closing date of study at the end of 1999, the oldest women were 74 years old. Information on date of birth for a woman and all her children was obtained from the Norwegian Population Registry, whereas information on cancer diagnosis was obtained from the Cancer Registry of Norway. Linking of data from different national registers was possible due to a unique personal identification number.

Only cases classified as primary malignant neoplasm of the breast (International Classification of Diseases, seventh revision, code 170, 98.4% histologically verified), with complete information on clinical stage (99.4%), were included. In the data from the Norwegian Cancer Registry, clinical stage at date of diagnosis reflects malignancy and extension as defined by node involvement and metastases. Tumors confined to the breast, with no regional lymph node involvement and no direct extension to the skin or chest wall, are recorded as stage I tumors. A stage II tumor comprises cases with axillary lymph node involvement. A stage III tumor is one with extension to the skin or chest wall, with or without lymph node involvement, but with no distant metastasis, whereas a stage IV tumor are those with distant metastases.

Statistical Analyses

χ2 Tests were applied to examine associations between clinical stage of the disease at time of diagnoses and time (years) since the most recent full-term birth within subgroups defined by parity and maternal age at birth. Diagnoses during pregnancy, defined as starting 9 months before date of birth, were also considered. In addition, associations between clinical stage and parity, as well as age at diagnosis, were examined. To provide more detailed information about the categories of time since birth or age that have the largest effect on the overall test for association, we also give the cell-specific contributions to the χ2 test. We, therefore, calculated the difference between the observed and expected cell counts, given no association, relative to the square root of the expected value. Absolute values of this quantity exceeding ∼1.96 or 1.64 for a specific category correspond to contributions at the 5% and 10% level, respectively. A negative value will indicate that we observe fewer cases than expected, whereas a positive value will indicate that we observe more cases than expected.

Logistic regression analysis was applied to test whether the association between late-stage tumors and time since birth differed by parity or maternal age at birth, and whether the association between late-stage tumors and age differed by parity (interaction tests).

Of the 22,351 cancer cases, a total of 12,731 (57.0%) were stage I cancer, 7,909 (35.4%) stage II, 513 (2.3%) stage III, and 1,198 (5.4%) stage IV. Nulliparous women had a significantly higher proportion of stage III and IV cancer than parous women and there was also some variation in the proportion of stage IV cancers according to number of births (Table 1). However, no significant difference according to parity was seen when combining late-stage tumors (i.e., stage II to IV cancers; Table 1).

Table 1.

Clinical stage of breast tumor by parity

Total no.Clinical stage (%*)
Pheterogeneity†
IIIIIIIV
No. births      <0.001 
0 3,269 55.7 33.3 3.3 7.7 0.11‡ 
(O − E) / √E§  −0.98 −1.99 3.93 5.73  
≥1 19,082 57.2 35.7 2.1 5.0  
(O − E) / √E§  0.40 0.82 −1.62 −2.37  
Among parous women      0.030 
    1 3,506 56.9 35.0 2.4 5.8 0.43‡ 
    2 8,466 57.8 35.7 2.0 4.5  
    3 5,015 56.9 36.1 2.2 4.8  
    4 1,672 55.4 36.9 2.0 5.6  
    5 423 56.7 35.0 0.9 7.3  
Total no.Clinical stage (%*)
Pheterogeneity†
IIIIIIIV
No. births      <0.001 
0 3,269 55.7 33.3 3.3 7.7 0.11‡ 
(O − E) / √E§  −0.98 −1.99 3.93 5.73  
≥1 19,082 57.2 35.7 2.1 5.0  
(O − E) / √E§  0.40 0.82 −1.62 −2.37  
Among parous women      0.030 
    1 3,506 56.9 35.0 2.4 5.8 0.43‡ 
    2 8,466 57.8 35.7 2.0 4.5  
    3 5,015 56.9 36.1 2.2 4.8  
    4 1,672 55.4 36.9 2.0 5.6  
    5 423 56.7 35.0 0.9 7.3  

Abbreviations: O, observed number; E, expected number, given no association.

*

Row percentage.

†

χ2 Test for association between clinical stage and parity.

‡

Clinical stage dichotomized (I versus II-IV).

§

Cell-specific contribution to the χ2 test with absolute value ≥1.96 corresponding to P ≤ 0.05.

View Large

The proportion of stage II to IV cancer changed significantly with age at diagnosis both among nulliparous and parous women (Table 2), but the age patterns differed (Pinteraction = 0.099 and 0.029 with 5 and 10 years age categories, respectively). In women diagnosed with breast cancer before the age of 30 years, a considerably larger proportion of advanced disease was found among parous than nulliparous women (52.7% versus 36.8%, P = 0.009). At age 30 to 34 years, nulliparous women had a somewhat higher proportion of late-stage tumors (P = 0.29; Table 2), whereas quite similar proportions were seen at age 35 to 44 years (P ≥ 0.89; Table 2). At older ages, nulliparous women tended to be diagnosed with a more advanced disease (P = 0.019 in women ages 45 to 49 years, P ≥ 0.086 otherwise; Table 2), except for women diagnosed at age 65 years or older (P = 0.51; Table 2). The large proportion of late-stage tumors in parous women below 30 years was consistently observed across subgroups of number of births (51.6%, 52.8%, 50.0%, and 71.4% in women with 1, 2, 3, and ≥4 births). Among women diagnosed at age 30 to 39 years, those with many children also had the highest proportion of late-stage tumors (46.3%, 44.4%, 47.8%, and 54.3% in women with 1, 2, 3, and ≥4 births).

Table 2.

Clinical stage of breast tumor in nulliparous and parous women by age at diagnosis

Nulliparous women
Parous women
Total no.Stage II-IV*
Total no.Stage II-IV*
%(O − E) / √E%(O − E) / √E
Age (y) at diagnosis       
    <25 28 32.1 −0.96 14 57.1 0.63 
    25-29 78 38.5 −0.78 174 52.3 1.81 
    30-34 174 50.6 1.24 745 46.2 1.11 
    35-39 288 45.8 0.38 1,778 46.3 1.71 
    40-44 478 45.6 0.42 2,975 45.5 1.46 
    45-49 559 48.7 1.54 3,789 43.4 −0.19 
    50-54 519 44.5 0.07 3,226 43.3 −0.31 
    55-59 415 44.1 −0.07 2,562 41.6 −1.36 
    60-64 366 44.3 −0.02 2,036 39.5 −2.83 
    ≥65 364 34.1 −2.94 1,783 35.9 −5.18 
Total no. 3,269   19,082   
Pheterogeneity† 0.0021   <0.001   
Nulliparous women
Parous women
Total no.Stage II-IV*
Total no.Stage II-IV*
%(O − E) / √E%(O − E) / √E
Age (y) at diagnosis       
    <25 28 32.1 −0.96 14 57.1 0.63 
    25-29 78 38.5 −0.78 174 52.3 1.81 
    30-34 174 50.6 1.24 745 46.2 1.11 
    35-39 288 45.8 0.38 1,778 46.3 1.71 
    40-44 478 45.6 0.42 2,975 45.5 1.46 
    45-49 559 48.7 1.54 3,789 43.4 −0.19 
    50-54 519 44.5 0.07 3,226 43.3 −0.31 
    55-59 415 44.1 −0.07 2,562 41.6 −1.36 
    60-64 366 44.3 −0.02 2,036 39.5 −2.83 
    ≥65 364 34.1 −2.94 1,783 35.9 −5.18 
Total no. 3,269   19,082   
Pheterogeneity† 0.0021   <0.001   
*

Row percentage (%) and cell-specific contribution to the χ2 test, with absolute value ≥1.96 corresponding to P ≤ 0.05.

†

χ2 Test for association between clinical stage and age at diagnosis within subgroups of parity.

View Large

Clinical Stage by Time Since Most Recent Childbirth (First to Third)

Among uniparous women with a first birth at age <25 or 25 to 29 years, the highest proportion of late-stage tumors was found among those diagnosed during pregnancy or within 2 years after birth (50.0% and 59.1%, respectively). However, no significant overall association between clinical stage and time since the birth was found (Table 3). Among the oldest uniparous women (≥30 years at date of birth), the highest proportion of advanced disease was found among those diagnosed 2 to 6 years after birth (56.0%), and this particular category also made an independent significant contribution to the overall test for association (Table 3). No significant difference across subgroups defined by maternal age was found (Pinteraction = 0.70, changed to 0.27 when testing for difference below and above age 30 years at date of first birth). A more detailed categorization showed that the highest proportion of stage II to IV tumors in uniparous women occurred among those diagnosed <1 year after a delivery at age 25 to 29 years (66.7%), closely followed by those diagnosed <1 year after a delivery at age ≥30 years (61.5%). However, the number of cases was low (9 and 13 cases, respectively).

Table 3.

Clinical stage of breast tumor by time since the most recent birth in subgroups of parity and maternal age at birth

Total no.Stage II-IV*
Total no.Stage II-IV*
Total no.Stage II-IV*
%(O − E) / √E%(O − E) / √E%(O − E) / √E
Uniparous women by          
    Age at first birth (y) <25   25-29   ≥30   
    Years since first birth          
        <2† 8 50.0 0.40 22 59.1 1.14 45 46.7 0.25 
        2-6 23 43.5 0.13 82 43.9 0.10 150 56.0 2.16 
        7-11 56 44.6 0.33 80 51.3 1.11 208 46.2 0.41 
        12-16 101 39.6 −0.34 144 42.4 −0.14 200 45.5 0.27 
        17-21 190 43.2 0.29 201 42.6 −0.08 213 42.3 −0.44 
        ≥22 821 41.4 −0.17 568 41.5 −0.57 386 38.9 −1.59 
    Total 1,199 41.8  1,097 43.1  1,202 44.3  
    Pheterogeneity‡ 0.98   0.41   0.017   
Biparous women by          
    Age at second birth (y) <25   25-29   ≥30   
    Years since second birth          
        <2† 4 25.0 −0.54 40 52.5 1.08 129 51.2 1.44 
        2-6 41 53.7 1.13 194 41.2 −0.06 499 49.3 2.19 
        7-11 82 35.4 −0.95 310 43.5 0.56 642 45.2 0.89 
        12-16 156 36.5 −1.08 509 40.7 −0.28 641 43.1 0.07 
        17-21 265 48.3 1.53 662 42.6 0.45 580 44.3 0.53 
        ≥22 872 41.5 −0.30 1,622 40.6 −0.52 1,226 37.4 −2.91 
    Total 1,420 42.2  3,337 41.5  3,717 42.9  
    Pheterogeneity‡ 0.062   0.62   <0.001   
Triparous women by          
    Age at third birth (y) <30   30-34   ≥35   
    Years since third birth          
        <2† 23 60.9 1.62 59 71.2 3.11 53 54.7 0.91 
        2-6 87 40.2 0.10 184 47.8 0.74 198 55.1 1.84 
        7-11 130 40.0 0.08 313 45.4 0.31 247 43.7 −0.56 
        12-16 231 46.3 1.65 385 45.7 0.45 196 50.5 0.90 
        17-21 350 34.9 −1.39 347 43.5 0.19 161 44.7 −0.27 
        ≥22 903 38.9 −0.30 763 40.2 −1.63 389 40.4 −1.68 
    Total 1,724 39.5  2,051 44.2  1,244 46.1  
    Pheterogeneity‡ 0.032   <0.001   0.009   
Total no.Stage II-IV*
Total no.Stage II-IV*
Total no.Stage II-IV*
%(O − E) / √E%(O − E) / √E%(O − E) / √E
Uniparous women by          
    Age at first birth (y) <25   25-29   ≥30   
    Years since first birth          
        <2† 8 50.0 0.40 22 59.1 1.14 45 46.7 0.25 
        2-6 23 43.5 0.13 82 43.9 0.10 150 56.0 2.16 
        7-11 56 44.6 0.33 80 51.3 1.11 208 46.2 0.41 
        12-16 101 39.6 −0.34 144 42.4 −0.14 200 45.5 0.27 
        17-21 190 43.2 0.29 201 42.6 −0.08 213 42.3 −0.44 
        ≥22 821 41.4 −0.17 568 41.5 −0.57 386 38.9 −1.59 
    Total 1,199 41.8  1,097 43.1  1,202 44.3  
    Pheterogeneity‡ 0.98   0.41   0.017   
Biparous women by          
    Age at second birth (y) <25   25-29   ≥30   
    Years since second birth          
        <2† 4 25.0 −0.54 40 52.5 1.08 129 51.2 1.44 
        2-6 41 53.7 1.13 194 41.2 −0.06 499 49.3 2.19 
        7-11 82 35.4 −0.95 310 43.5 0.56 642 45.2 0.89 
        12-16 156 36.5 −1.08 509 40.7 −0.28 641 43.1 0.07 
        17-21 265 48.3 1.53 662 42.6 0.45 580 44.3 0.53 
        ≥22 872 41.5 −0.30 1,622 40.6 −0.52 1,226 37.4 −2.91 
    Total 1,420 42.2  3,337 41.5  3,717 42.9  
    Pheterogeneity‡ 0.062   0.62   <0.001   
Triparous women by          
    Age at third birth (y) <30   30-34   ≥35   
    Years since third birth          
        <2† 23 60.9 1.62 59 71.2 3.11 53 54.7 0.91 
        2-6 87 40.2 0.10 184 47.8 0.74 198 55.1 1.84 
        7-11 130 40.0 0.08 313 45.4 0.31 247 43.7 −0.56 
        12-16 231 46.3 1.65 385 45.7 0.45 196 50.5 0.90 
        17-21 350 34.9 −1.39 347 43.5 0.19 161 44.7 −0.27 
        ≥22 903 38.9 −0.30 763 40.2 −1.63 389 40.4 −1.68 
    Total 1,724 39.5  2,051 44.2  1,244 46.1  
    Pheterogeneity‡ 0.032   <0.001   0.009   
*

Row percentage (%, categories with ≥50% of late stage tumors marked in bold), and cell-specific contribution to the χ2 test, with absolute value ≥1.96 corresponding to P ≤ 0.05.

†

Includes diagnoses during pregnancy.

‡

χ2 test for association between clinical stage and time since birth within subgroups of parity and maternal age at birth.

View Large

A somewhat different pattern appeared for biparous women (Table 3). Among the youngest mothers (<25 years at date of second birth), the highest proportion of late-stage tumors was seen among those diagnosed 2 to 6 years after birth (53.7%), and the association with time since birth almost reached statistical significance. In the two oldest age groups, the highest proportion of advanced disease was found among those diagnosed during pregnancy or within 2 years after birth (52.5% and 51.2%, respectively). Among women with a second birth at age 30 to 34 years, the proportion of late-stage tumors dropped to ∼40% in the subsequent time period, and no significant association with time since the birth was found (Table 3). Among the oldest mothers (≥30 years at date of second birth), however, almost 50% of those diagnosed 2 to 6 years after birth also appeared with an advanced disease, and this particular category made an independent significant contribution to the overall test for association (Table 3). Despite apparent differences, the test for heterogeneity across subgroups of maternal age did not reach statistical significance (Pinteraction = 0.075, changed to 0.11 when testing for difference below and above age 30 years at date of second birth). The highest proportion of advanced disease (66.7%) was seen among the 15 women diagnosed during the second pregnancy who were 30 years or older at term birth.

Among triparous women, a significant association between clinical stage and time since third birth was found in all categories of maternal age (Table 3). Among women with a third birth before age 30 years or at age 30 to 34 years, the highest proportion of late-stage tumors was found among those diagnosed during pregnancy or within 2 years after birth (60.9% and 71.2%, respectively). Among women with a third birth at age 35 years or older, however, the highest proportion of advanced disease was found among those diagnosed 2 to 6 years after birth (55.1%). This category almost made an independent significant contribution to the overall test results (Table 3). No significant difference across subgroups of maternal age was found (Pinteraction = 0.28, changed to 0.20 when testing for difference below and above age 35 years at date of third birth). With a more detailed categorization, the highest proportion of stage II to IV tumors in triparous women was seen among the 21 women diagnosed <1 year after a third birth at age 30 to 34 years (85.7%).

In the present study, we have examined associations between clinical stage of breast cancer tumors and time interval since most recent childbirth based on information from nationwide registers with compulsory registrations. In most subgroups of parity and maternal age at birth, we found that women diagnosed during pregnancy or within 2 years after birth had the highest proportion of stage II to IV tumors. Among women with late-age births, however, as well as women with an early second birth, the proportion with advanced disease was rather similar or even higher among those diagnosed 2 to 6 years after birth. This time period overlaps with the high-risk period after birth and these particular subgroups also correspond quite well to those with the most pronounced transient increase in risk as observed in our recent study (3). These findings indicate that pregnancy hormones or other factors during pregnancy affect tumor growth and, thus, have a progressive effect on breast tumors in addition to a possible promoting effect that triggers growth of premalignant breast cells (1).

Consistent with our findings, two previous studies (4, 5) found a larger proportion of node-positive tumors, and also larger tumors, in women with <5 or 6 years since their most recent childbirth than in those with a longer time interval since a term birth. Both of these studies, however, focused on the poor prognosis among women with a very recent birth. Another Nordic register–based study (6) found that the transient increase in risk after first and second birth was most pronounced for late-stage tumors, as defined by tumor size, nodal status, and histologic grading, respectively. These findings were interpreted as being consistent with a progressive effect of pregnancy hormones. In an additional study based on the same data (7), parity and age at first birth were similarly associated with carcinoma in situ lesions and invasive breast cancer, apparently contradictory to the hypothesis of a promoting effect. However, carcinoma in situ comprises epithelial cells with morphologic features of malignancy and these cells may already have reached a phase in which they are susceptible to hormonal exposure.

A selection bias or delay in diagnosis owing to difficulties in discovering breast tumors in young women in general, as well as in women with a recent childbirth, may partly explain the larger proportion of late-stage tumors shortly after birth (2). However, such an effect is limited in time (2, 8) and can probably not fully account for the high proportion of late-stage disease as long as 2 to 6 years after a full-term birth. Our results were quite similar when redefining the two first categories to <3 and 4 to 6 years after birth. On the basis of comparisons between observed and predicted number of cancer cases, Wohlfahrt et al. (6) concluded that a delayed diagnosis or surgery due to pregnancy could not explain the entire excess risk of late-stage breast cancer that persisted as long as 10 years after birth in their study. A larger proportion of advanced disease in women diagnosed some years after a term birth may be related to a general more rapid tumor progression in younger than older women (9), possibly reflecting normal physiology at time of initiation (10). However, the present observation of a larger proportion of advanced disease among parous than nulliparous women at a young age, in agreement with results from another recent study (11), indicates that pregnancies contribute to an advanced stage at diagnosis among younger women.

We found that the proportion of late-stage tumors was particularly large immediately after a third birth before age 35 years. An effect of earlier pregnancies may have contributed to this unfavorable pattern. Among women with a third birth at age 35 years or older, the largest proportion of late-stage tumors was found in women diagnosed 2 to 6 years after the birth, overlapping with the high-risk period in that subgroup (3). Previously, we found that an early first birth did not prevent an adverse effect of subsequent births at an older age (3), apparently contradictory to the hypothesis that differentiation of breast cells after the first pregnancy makes the breast tissue less susceptible to hormonal or other carcinogenic stimuli (1). However, breast cells that are in a very early phase of a carcinogenic development at time of the first pregnancy may eventually undergo a malignant transformation, and possibly further development, through the repeated exposure to pregnancy hormones of subsequent pregnancies.

A promoting effect of pregnancy estrogens has been hypothesized to explain the transient increase in risk of breast cancer after a full-term birth (12). In a recent study (13), however, the short-term adverse effect of first pregnancy was present only for progesterone receptor–negative tumors, independent of estrogen receptor status. A less consistent pattern has been found in relation to time since last birth and estrogen/progesterone receptor status (14). The proportion of estrogen receptor–positive tumors increases with age (15, 16), and the observation that low parity and late age at first birth are more likely to be associated with estrogen receptor–positive tumors (14, 17) may reflect a general age pattern since overall associations with these reproductive factors are often studied in older women. Pregnancy-associated breast cancer (i.e., cancer diagnosed during or within 2 years after birth) tends to be estrogen receptor negative (2), as tumors in young women in general (18). However, a large proportion of estrogen receptor–negative tumors has been found to be prolactin receptor positive (19). Prolactin receptor positivity has also been found to be more common in breast tumors, as well as benign lesions, than in normal breast tissue (20, 21). Furthermore, experimental and epidemiologic research has indicated that prolactin can have a promoting (21, 22) as well as a progressive effect (21, 23-26), and also act as a local growth factor (21, 23, 25, 27). The elevated level of prolactin during pregnancy and lactation (28, 29) may thus be the factor of importance with respect to the short-term adverse effect of a childbirth. Consistent with a hypothesis of a promoting effect of prolactin, especially after late-age births (1, 3), the plasma prolactin level has been found to be elevated in healthy women whose first birth occurred after age 35 years (30), and has been found to increase with age in premenopausal women who were diagnosed with breast cancer at a median of 5 years after sampling (31). The pregnancy-related exposure to prolactin may also accelerate growth of existing tumors and thus explain the present observation of a larger proportion of late-stage tumors in women diagnosed some years after a term birth.

Prolactin has previously been hypothesized to be important both in the initiation and promotion steps of breast tumorigenesis (32-34); it may also explain associations with some reproductive factors (23, 34, 35). Inconsistent findings in relation to female breast cancer risk (36-39) may be due to difficulties with sampling, owing to daily and monthly variations in prolactin levels (35, 36, 40). Serum prolactin levels are also considerably lower than breast fluid prolactin levels (41). Prolactin has recently been suggested to be of importance also for breast cancer risk in males (42). A potential effect of prolactin may depend on estrogen and progesterone receptor status (20, 24, 43), although some experimental studies have indicated an independent effect (21, 44). Recent reports (45-49) have focused on the role of antiprolactinemic drugs, in addition to use of antiestrogens in the treatment of breast cancer. Both treatment regimens have thus far been assumed to act by blocking hormone-induced cell proliferation and/or by increasing apoptosis (45, 47-52). Some studies (44, 53), however, suggest that antiestrogen treatment may act through a down-regulation of prolactin receptors.

To further explore a potential role of prolactin, it would be of interest to analyze the joint effects of prolactin, estrogens, and progesterons on breast cancer risk, as well as tumor growth characteristics. Information on a potential role of prolactin can also be obtained from larger epidemiologic studies examining the short-term effect of a pregnancy according to whether the women have breast-fed or not, and also by the length of the lactation period. A recent study (54) found a 2-fold increase in risk of carcinoma in situ lesions in women below 50 years who had breast-fed at least 24 months. Of particular interest in view of potential biological mechanisms is also the finding that overall associations with lactation and duration of breast-feeding seem to be unrelated to estrogen and progesterone receptor status of the tumor (14). Information on breast-feeding was not available in our register-based study, but breast-feeding is, and has been, quite common in Norway. In certain subgroups defined by parity and maternal age, we found that the proportion of advanced disease was particularly high among women diagnosed within the first year after a term birth.

Grant support: The Norwegian Cancer Society.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
Pathak DR. Dual effect of first full term pregnancy on breast cancer risk: empirical evidence and postulated underlying biology.
Cancer Causes Control
 
2002
;
13
:
295
–8.
2
Woo JC, Yu T, Hurd TC. Breast cancer in pregnancy. A literature review.
Arch Surg
 
2003
;
138
:
91
–8.
3
Albrektsen G, Heuch I, Hansen S, Kvåle G. Breast cancer risk by age at birth, time since birth and time intervals between births: exploring interaction effects.
Br J Cancer
 
2005
;
92
:
167
–75.
4
Kroman N, Wohlfahrt J, Andersen KW, Mouridsen HT, Westergaard T, Melbye M. Time since childbirth and prognosis in primary breast cancer: population based study.
BMJ
 
1997
;
315
:
851
–5.
5
Phillips K-A, Milne RL, Friedlander ML, et al. Prognosis of premenopausal breast cancer and childbirth prior to diagnosis.
J Clin Oncol
 
2004
;
22
:
699
–705.
6
Wohlfahrt J, Andersen PK, Mouridsen HT, Melbye M. Risk of late-stage breast cancer after a childbirth.
Am J Epidemiol
 
2001
;
153
:
1079
–84.
7
Wohlfahrt J, Rank F, Kroman N, Melbye M. A comparison of reproductive risk factors for CIS lesions and invasive breast cancer.
Int J Cancer
 
2004
;
108
:
750
–3.
8
Saunders CM, Baum M. Breast cancer and pregnancy: a review.
J R Soc Med
 
1993
;
86
:
162
–5.
9
Tabár L, Duffy SW, Vitak B, Chen H-H, Prevost TC. The natural history of breast carcinoma. What have we learned from screening?
Cancer
 
1999
;
86
:
449
–62.
10
Olsson H. Reproductive events, occurring in adolescence at the time of development of reproductive organs and at the time of tumour initiation, have a bearing on growth characteristics and reproductive hormone regulation in normal and tumour tissue investigated decades later—a hypothesis.
Med Hypotheses
 
1989
;
28
:
93
–7.
11
Largent JA, Ziogas A, Anton-Culver H. Effect of reproductive factors on stage, grade and hormone receptor status in early onset breast cancer.
Breast Cancer Res
 
2005
;
7
:
R541
–54.
12
Henderson BE, Bernstein L. The international variation in breast cancer rates: an epidemiological assessment.
Breast Cancer Res Treat
 
1991
;
18
:
S11
–7.
13
Colditz GA, Rosner BA, Chen WY, Holmes MD, Hankinson SE. Risk factors for breast cancer according to estrogen and progesterone receptor status.
J Natl Cancer Inst
 
2004
;
96
:
218
–28.
14
Ursin G, Bernstein L, Lord SJ, et al. Reproductive factors and subtypes of breast cancer defined by hormone receptor and histology.
Br J Cancer
 
2005
;
93
:
364
–71.
15
Yasui Y, Potter JD. The shape of age-incidence curves of female breast cancer by hormone-receptor status.
Cancer Causes Control
 
1999
;
10
:
431
–7.
16
Anderson WF, Althuis MD, Brinton LA, Devesa SS. Is male breast cancer similar or different than female breast cancer?
Breast Cancer Res Treat
 
2004
;
83
:
77
–86.
17
Althuis MD, Fergenbaum JH, Garcia-Closas M, Brinton LA, Madigan MP, Sherman ME. Etiology of hormone receptor-defined breast cancer: a systematic review of the literature.
Cancer Epidemiol Biomarkers Prev
 
2004
;
13
:
1558
–68.
18
Sidoni A, Cavaliere A, Bellezza G, Scheibel M, Bucciarelli E. Breast cancer in young women: clinicopathological features and biological specificity.
Breast
 
2003
;
12
:
247
–50.
19
Scawn R, Shousha S. Morphologic spectrum of estrogen receptor-negative breast carcinoma.
Arch Pathol Lab Med
 
2002
;
126
:
325
–30.
20
Gill S, Peston D, Vonderhaar BK, Shousha S. Expression of prolactin receptors in normal, benign, and malignant breast tissue: an immunohistological analysis.
J Clin Pathol
 
2001
;
54
:
956
–60.
21
Goffin V, Binart N, Touraine P, Kelly PA. Prolactin: The new biology of an old hormone.
Annu Rev Physiol
 
2002
;
64
:
47
–67.
22
Goffin V, Bernichtein S, Kayser C, Kelly PA. Development of new prolactin analogs acting as pure prolactin receptor antagonists.
Pituitary
 
2003
;
6
:
89
–95.
23
Clevenger CV, Furth PA, Hankinson SE, Schuler LA. The role of prolactin in mammary carcinoma.
Endocr Rev
 
2003
;
24
:
1
–27.
24
Ingram DM, Roberts A, Nottage EM. Host factors and breast cancer growth characteristics.
Eur J Cancer
 
1992
;
28A
:
1153
–61.
25
Bhatavdekar JM, Patel DD, Shah NG, et al. Prolactin as a local growth promoter in patients with breast cancer: GCRI experience.
Eur J Surg Oncol
 
2000
;
26
:
540
–7.
26
Mujagic Z, Mujagic H. Importance of serum prolactin determination in metastatic breast cancer patients.
Croat Med J
 
2004
;
45
:
176
–80.
27
Ben-Jonathan N, Liby K, McFarland M, Zinger M. Prolactin as an autocrine/paracrine growth factor in human cancer.
Trends Endocrinol Metab
 
2002
;
13
:
245
–50.
28
Lotfy M. Prolactin level after prolonged lactation: contraceptive evaluation.
Contracept Deliv Syst
 
1984
;
5
:
63
–6.
29
Mucci LA, Lagiou P, Tamimi RM, Hsieh C-C, Adami H-O, Trichopoulos D. Pregnancy estriol, estradiol, progesterone and prolactin in relation to birth weight and other birth size variables (United States).
Cancer Causes Control
 
2003
;
14
:
311
–8.
30
Hill P, Wynder EL, Kumar H, Helman P, Rona G, Kuno K. Prolactin levels in populations at risk for breast cancer.
Cancer Res
 
1976
;
36
:
4102
–6.
31
Kwa HG, Cleton F, Wang DY, et al. A prospective study of plasma prolactin levels and subsequent risk of breast cancer.
Int J Cancer
 
1981
;
28
:
673
–6.
32
Welsch CW, Nagasawa H. Prolactin and murine mammary tumourigenesis: a review.
Cancer Res
 
1977
;
37
:
951
–63.
33
Bulbrook RD, Wang DY, Hayward JL, Kwa HG, Cleton F. Plasma prolactin levels and age in a female population: relation to breast cancer.
Int J Cancer
 
1981
;
28
:
43
–5.
34
Pike MC, Krailo MD, Henderson BE, Casagrande JT, Hoel DG. “Hormonal” risk factors, “breast tissue age” and the age-incidence of breast cancer.
Nature
 
1983
;
303
:
767
–70.
35
Kwa HG, Cleton F, Bulbrook RD, Wang DY, Hayward JL. Plasma prolactin levels and breast cancer: relation to parity, weight and height, and age at first birth.
Int J Cancer
 
1981
;
28
:
31
–4.
36
Kwa HG, Cleton F, de Jong-Bakker M, Bulbrook RD, Hayward JL, Wang DY. Plasma prolactin and its relationship to risk factors in human breast cancer.
Int J Cancer
 
1976
;
17
:
441
–7.
37
Ingram DM, Nottage EM, Roberts AN. Prolactin and breast cancer risk.
Med J Aust
 
1990
;
153
:
469
–73.
38
Wang DY, De Stavola BL, Bulbrook RD, et al. Relationship of blood prolactin levels and the risk of subsequent breast cancer.
Int J Epidemiol
 
1992
;
21
:
214
–21.
39
Hankinson SE, Willett WC, Michaud DS, et al. Plasma prolactin levels and subsequent risk of breast cancer in postmenopausal women.
J Natl Cancer Inst
 
1999
;
91
:
629
–34.
40
Wang DY, Sturzaker HE, Kwa HG, Verhofstad F, Hayward JL, Bulbrook RD. Nyctohemeral changes in plasma prolactin levels and their relationship to breast cancer risk.
Int J Cancer
 
1984
;
33
:
629
–32.
41
Rose DP. Hormones and growth factors in nipple aspirates from normal women and benign breast disease patients.
Cancer Detect Prev
 
1992
;
16
:
43
–51.
42
Okada K, Kajiwara S, Tanaka H, Sakamoto G. Synchronous bilateral noninvasive ductal carcinoma of the male breast: a case report.
Breast Cancer
 
2003
;
10
:
163
–6.
43
Tworoger SS, Eliassen AH, Rosner B, Sluss P, Hankinson SE. Plasma prolactin concentrations and risk of postmenopausal breast cancer.
Cancer Res
 
2004
;
64
:
6814
–9.
44
Vonderhaar BK. Prolactin involvement in breast cancer.
Endocr Relat Cancer
 
1999
;
6
:
389
–404.
45
Goffin V, Touraine P, Pichard C, Bernichtein S, Kelly PA. Should prolactin be reconsidered as a therapeutic target in human breast cancer?
Mol Cell Endocrinol
 
1999
;
151
:
79
–87.
46
Bernichtein S, Kayser C, Dillner K, et al. Development of pure prolactin receptor antagonists.
J Biol Chem
 
2003
;
278
:
35988
–99.
47
Lissoni P, Vaghi M, Villa S, et al. Antiprolactinemic approach in the treatment of metastatic breast cancer: a phase II study of polyneuroendocrine therapy with LHRH-analogue, tamoxifen and the long-acting antiprolactinemic drug cabergoline.
Anticancer Res
 
2003
;
23
:
733
–6.
48
Perks CM, Keith AJ, Goodhew KL, Savage PB, Winters ZE, Holly JMP. Prolactin acts as a potent survival factor for human breast cancer cell lines.
Br J Cancer
 
2004
;
91
:
305
–11.
49
Porta RP, Franco C, Cosmi EV, Montruccoli G, Cavazzana AO. Pregnancy-associated breast cancer.
Breast J
 
2004
;
10
:
169
.
50
Hawkins RA, Arends MJ, Ritchie AA, Langdon S, Miller WR. Tamoxifen increases apoptosis but does not influence markers of proliferation in an MCF-7 xenograft model of breast cancer.
Breast
 
2000
;
9
:
96
–106.
51
Cameron DA, Ritchie AA, Miller WR. The relative importance of proliferation and cell death in breast cancer growth and response to tamoxifen.
Eur J Cancer
 
2001
;
37
:
1545
–53.
52
Miller WR. Biological rationale for endocrine therapy in breast cancer.
Best Pract Res Clin Endocrinol Metab
 
2004
;
18
:
1
–32.
53
de Castillo B, Cawthorn S, Moppett J, Shere M, Norman M. Expression of prolactin receptor mRNA in oestrogen receptor positive breast cancers pre- and post-tamoxifen therapy.
Eur J Surg Oncol
 
2004
;
30
:
515
–9.
54
Meeske K, Press M, Patel A, Bernstein L. Impact of reproductive factors and lactation on breast carcinoma in situ risk.
Int J Cancer
 
2004
;
110
:
102
–9.
American Association for Cancer Research
2006