In breast cancer CA 15.3 is considered the tumour marker of choice. CA 15.3 is directly related to the disease extent and to hormone status (estrogen receptor ER+/ ER-, progesterone receptor PR+/PR-). This study was designed to assess the impact of disease extent, hormone receptor and HER2-status, and circulating blood volume on the area-under the ROC-curve of CA 15.3 to separate FDG PET positive from negative findings.
Patients, methods: We retrospectively evaluated 379 FDG PET/CT examinations performed in 80 patients with breast cancer. Blood volumes were derived using the formulas by Nadler and multiplied by their corresponding CA 15.3 measurement.
Results: ROC-curve analysis revealed an AUC of 0.695 (p = 0.0001) for CA 15.3 to separate FDG PET positive from negative findings. AUC measurements to separate normal scan findings from loco-regional disease and metastatic disease were 0.527 (p = 0.587) and 0.732 (p = 0.0001), respectively. AUC measurements for CA 15.3 to separate positive from negative FDG PET findings, in ER+ and ER- patients, were respectively 0.772 (p = 0.0001) and 0.596 (p = 0.143). AUC measurements for CA 15.3 to separate positive from negative FDG PET findings, in PR+ and PR- patients, were respectively 0.675 (p = 0.0001) and 0.694 (p = 0.0001). In HER2-positive and -negative patients, the AUC measurements were respectively 0.594 (p = 0.178) and 0.701 (p = 0.0001) to separate positive from negative FDG PET findings.
Conclusion: The AUC for CA 15.3 measurements to separate FDG PET positive from negative findings in breast cancer patients with suspected recurrence proved to be directly related to the extent of the recurrent disease and hormone receptor status and inversely related to HER2-status. Correcting CA 15.3 measurements for blood volumes did not impact the AUC.
Keywords: CA 15.3; FDG PET; ROC-curve; breast cancer; recurrence.