Fig. 2. Association of 163 proteins in the targeted panel with incident breast cancer.

We used Cox proportional hazard models to investigate the impact of cardiometabolic (CAM) and immune-oncology (IMONC)-related proteins on the risk to develop breast cancer within 3 years. Triangles pointing up indicate that increased protein levels result in increased risk for breast cancer and vice versa. Significantly associated proteins (P < 0.05) are shown above or below the blue dotted line and are labelled. No proteins were found to be significantly associated with incident breast cancer risk after accounting for multiple testing (i.e., P < 0.05/163), as indicated by the red dotted line.