Elsevier

The Breast

Volume 30, December 2016, Pages 222-227
The Breast

Original article
Association with pregnancy increases the risk of local recurrence but does not impact overall survival in breast cancer: A case–control study of 87 cases

https://doi.org/10.1016/j.breast.2015.09.006Get rights and content

Abstract

Pregnancy-associated breast cancer (PABC) constitutes 7% of all BCs in young women. The prognosis of PABC remains controversial. In this study, we evaluated the impact of the association of pregnancy with BC on the rates of overall survival (OS), disease free survival (DFS), and distant and local recurrence-free survival.

We conducted a retrospective unicenter case–control study. We enrolled PABC patients treated at our institution between 1992 and 2009. For each case, 2 BC controls were matched for age and year of diagnosis. Univariate and multivariate analyses were performed to assess the parameters associated with prognosis.

Eighty-seven PABC patients were enrolled and matched with 174 controls. The univariate analysis did not reveal any significant differences in OS, DFS or distant recurrence rates between the 2 groups. Pregnancy associated status, a tumor larger than T2 and neoadjuvant chemotherapy as the primary treatment were significantly associated with an increased risk of local relapse. The multivariate analysis showed that the pregnancy associated status and the tumor size were strong prognostic factors of local recurrence. Pregnancy associated status negates the prognostic value of tumor size, as both T0–T2 and T3–T4 PABC patients have the same poor prognosis as control BC patients with T3–T4 tumors. Interestingly, although PABC patients have more locally advanced tumors, they did not have a higher rate of radical surgery than the control BC patients.

Pregnancy associated status is a strong prognostic factor of local relapse in BC. In PABC patients, when possible, radical surgery should be the preferred first treatment step.

Introduction

Pregnancy-associated cancers are typically defined in the literature as cancers occurring during pregnancy or within the year following delivery [1]. They are not uncommon; they occur in 1/10,000 to 1/3000 pregnancies [1], [2], [3], [4]. The cancer that is most frequently associated with pregnancy is breast cancer (BC), with pregnancy-associated BC (PABC) constituting approximately 7% of all BCs in women less than 45 years of age [1], [5]. The incidence of PABC has increased in recent decades in western countries and is expected to continue to grow as women delay childbearing in these countries [1], [6]. The prognosis of PABC remains controversial. Some large retrospective studies have reported that PABC was associated with an increased risk of specific death or recurrence regardless of the patient's age, the tumor size or the pathological characteristics [2], [7], [8], [9], [10], [11], [12], [13], [14], [15]. PABC patients were previously described to have pejorative prognostic factors, such as young age, more locally advanced tumors, and more frequent negative hormone receptor status [14], [16], [17], [18]. After adjustment for age and tumor characteristics, some authors did not find a worse prognosis for PABC patients compared to control BC patients [3], [4], [16], [19], [20]. Furthermore, PABC patients were frequently not offered the recommended treatment strategy, thus resulting in a bias in the interpretation of survival data. We evaluated the prognosis of PABC compared with that of control BC. We examined the factors of therapeutic management and adjusted the data for the currently used prognostic factors (age, tumor stage, hormonal receptor status, human epidermal growth factor receptor 2 [HER2] expression level and proliferation measured as the Ki67 rate). Therefore, we conducted a retrospective case–control study to determine the impact of the pregnancy associated status in our BC population. We matched cases and controls for age and year of diagnosis excluding other known prognosis factors (such as tumor size, node involvement, hormonal or HER2 status) as we expected that these differences could be take into account in the overall assessment of the prognosis of this young population of BC patients.

Section snippets

Patient selection

From the prospective database of all BC patients treated at Centre René Huguenin-Institut Curie (Saint-Cloud, France), we retrospectively selected female patients having first unilateral BC without distant metastasis at the time of the diagnosis who were treated between 1st July 1992 and 30th September 2009. PABC was defined according to the literature as BC diagnosed during pregnancy or in the year following delivery. The medical reports of patients suspected of having PABC based on the

Population characteristics and treatments

Two-hundred sixty-one patients were enrolled in the study, including 87 cases and 174 matched controls. The median follow-up period was 113 months (range 1.8–255.3). No significant differences were observed for the duration of follow-up between the 2 groups (PABC group: median follow-up 111 months, control group: 113 months, p = 0.85). The pathological characteristics of the patients' tumors are summarized in Table 1. The main histological type in both groups was invasive ductal carcinoma.

Discussion

PABC constitutes 7% of BCs in young women, and its incidence is expected to increase [1], [6]. Its prognosis and the underlying therapeutic strategy for affected patients are currently subjects of debate. In this case–control study, we found that PABC patients have worse prognostic factors than controls matched for age at diagnosis, as T3–T4 tumors were more than twice as frequent in PABC patients than controls, and nearly half of the PABC tumors were hormone-negative vs. one third of the

Conclusion

Association of BC with pregnancy is strongly associated with an increased risk of local recurrence, and pregnancy association negates the prognostic value of tumor size. Indeed, small tumors in PABC patients have the same poor prognosis as more advanced tumors in control patients, although no significant difference in prognosis was observed between the PABC and control patients with T3–T4 tumors. In PABC patients, when possible, radical surgery as the first-line treatment rather than

Conflict of interest

The authors declare that they have no conflict of interest and no funding source for this study.

References (25)

  • W.B. Moreira et al.

    Prognosis for patients diagnosed with pregnancy-associated breast cancer: a paired case–control study

    São Paulo Med J Rev Paul Med

    (mai 2010)
  • A.O. Rodriguez et al.

    Evidence of poorer survival in pregnancy-associated breast cancer

    Obstet Gynecol

    (juill 2008)
  • Cited by (20)

    • Chemosensitivity, tumor infiltrating lymphocytes (TILs), and survival of postpartum PABC patients treated by neoadjuvant chemotherapy

      2018, Breast
      Citation Excerpt :

      PABC's pathogenic pathway is probably different from that of non-PABC [2,5–7]. It is usually associated with pejorative prognostic factors such as young age, locally advanced tumors, or hormone receptor-negative tumors [4,8,9]. However, so far, studies on PABC have scarcely distinguished breast cancers diagnosed during pregnancy (PREG-PABC) from those diagnosed during the first post-partum year (post-partum PABC; designed as PP-PABC).

    • Propensity score to evaluate prognosis in pregnancy-associated breast cancer: Analysis from a French cancer network

      2018, Breast
      Citation Excerpt :

      Furthermore, the maternal prognosis of pregnancy associated breast cancer, and especially the impact of pregnancy, is highly debated. Some authors report that pregnancy itself negatively influences the prognosis by specific involvement of the pro-inflammatory micro-environment of mammary gland, while others suggest that the prognosis is similar in pregnant and non-pregnant patients [5,7,8]. So far, few studies have compared breast cancer in young patients with and without pregnancy [5,9].

    View all citing articles on Scopus
    View full text