P2-01-04: The Post-Partum Diagnosis of Pregnancy Associated Breast Cancer Confers an Increased Risk for Metastasis without Increased Incidence of Poorer Prognosis Biologic Subtype.

Abstract

Abstract Background: All women who complete a childbirth have an increased risk for the subsequent breast cancer regardless of age at their first birth. In younger women, this period truncates 10 years later and pregnancy then becomes protective, whereas in older mothers the risk persists. Moreover, breast cancers diagnosed subsequent to a recent childbirth have an increased risk for metastasis and death. However much of the existing data is confounded by inclusion of cases diagnosed during pregnancy. Hypothesis: Our preclinical models of pregnancy associated breast cancer (PABC) has identified the “involution hypothesis”, where the normal physiologic post-partum breast involution is the pregnancy-related event promotional of breast cancer growth, invasion and metastasis. We have performed a retrospective cohort study, restricted to postpartum PABC and controls, to fully characterize post-partum PABC by biologic subtype and known predictors for metastasis. Also we identify if the specific period of post-partum PABC has a higher risk for metastasis then non-PABC, as predicted by our animal model studies. Methods: We performed a retrospective cohort of 527 cases of breast cancer diagnosed at age 45 or younger from 1981–2011. Pregnancy status was identified from medical records and defined as nulliparous (n=107), PABC (up to 5 years post-partum of last childbirth, n=114), and Later Parous (>10 years post-partum, n=239). Clinicopathologic characteristics were obtained from pathology records. Outcomes data obtained through the University of Colorado Tumor Registrar. Hazard ratios of risk of metastatic recurrence of nulliparous and PABC groups were analyzed using a Cox Proportional Hazards Model and time to metastatic recurrence was analyzed using standard Kaplan Meier curves. Results: Compared to nulliparous cases, PABC cases were more frequently Stage IV(11.4% v. 4.9%), T3 tumor size(17.3% v. 13.1%), poorly differentiated(56.6% v. 49.0%), lobular histology(4.5% v.1.0%), with lymphovascular invasion(30.4% v. 24.5%) and with involved lymphnodes(57.6% v. 48.6%). No differences were seen between Luminal A, Luminal B, Her 2 and Triple Negative subtypes. Average follow-up was 3.4 years for nulliparous, 3.5 for PABC and 4.6 for Later Parous. Survival analysis revealed a cumulative 5-year metastatic recurrence free probability of 80.6% for nulliparous, 68.5% for PABC and 80.5% for Later Parous. Compared to nulliparous, PABC had greater than a three times higher risk for subsequent metastatic recurrence (HR 3.29, 95% CI: 1.25−8.63). No significant difference in the risk of metastatic recurrence for Later Parous compared to nulliparous was observed (HR: 1.079, 95% CI: 0.46−2.55). Conclusion: Post-partum PABC is characterized by higher stage at diagnosis but without enrichment for poorer prognosis biologic subtypes. Post-Partum PABC has an increased risk for metastasis within five years of diagnosis, with an overall 3.29 times higher risk of a metastatic recurrence. These data support our hypothesis that the post-partum window, and the associated role of naturally occurring involution, may be promotional of tumor growth, invasion and metastasis as previously identified in our animal models. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P2-01-04.