BACKGROUND: In young women, a breast cancer diagnosis after childbirth increases the risk for metastasis and death. Studies in rodents suggest that post-weaning mammary gland involution contributes to the poor prognosis of postpartum breast cancers. However, this association has not been investigated in humans, mainly because of missing information on the patient's lactation status at diagnosis.
PATIENTS AND METHODS: Clinicopathological data of 1180 young women with primary invasive breast cancer, diagnosed within 2 years postpartum (PP-BC), during pregnancy (Pr-BC), or nulliparous (NP-BC), were collected. For PP-BC patients, breastfeeding history was retrieved to differentiate breast cancers identified during lactation (PP-BCDL) from those diagnosed post-weaning (PP-BCPW). Differences in prognostic parameters, first site of distant metastasis, and risks for metastasis and death were determined between patient groups.
RESULTS: Cox proportional hazard models pointed to a twofold increased the risk of metastasis and death in PP-BCPW patients compared with PP-BCDL (hazard ratio [HR] 2.1 [PDRS = 0.021] and 2.9 [POS = 0.004]), Pr-BC (HR 2.1 [PDRS<0.001] and 2.3 [POS<0.001]) and NP-BC (HR 2.1 [PDRS<0.001] and 2.0 [POS<0.001]) patients. Prognosis was poorest for PP-BCPW patients who did not breastfeed or only for ≤ 3 months before diagnosis. This could not fully be attributed to differences in standard prognostic characteristics. In addition, PP-BCPW tumours showed a 3- to 8-fold increased risk to metastasise to the liver, yet this did not correlate with the poor outcome of this patient cohort.
CONCLUSIONS: Breast cancer diagnosed shortly after weaning specifically adds to the poor prognosis in women diagnosed with PP-BC. Apart from the importance of an increased awareness, these data show that detailed lactation data need to be registered when breast cancer outcome in young women is investigated.