Calcifications on mammogram do not correlate with tumor size after neoadjuvant chemotherapy

Calcifications can be indicative of malignancy, but calcifications also can be a byproduct of necrotic tissue as cancer cells die. Current treatment regimens require excision of calcifications. The objective of this study was to examine the correlation between the extent of calcification on mammography and actual tumor size after neoadjuvant chemotherapy (NAC) as well as magnetic resonance imaging (MRI) for comparison.

We retrospectively reviewed all patients at the University of California, San Diego, who underwent NAC for breast cancer between 2007 and 2013. Pearson correlation coefficients were computed between breast imaging and pathological measurements.

There were 136 patients total. Average age was 51 years. Fifty-three patients had calcifications on imaging (calc+); 83 did not (calc-). In the calc- group, extent of disease measured by mammogram (MMG) and MRI correlated moderately well with pathological tumor size (0.46 and 0.48, p = not significant). In the calc+ group, MRI was more likely to correlate with pathology than MMG (0.55 vs. -0.12, p = 0.01). Twenty-five calc+ patients had increased calcification after NAC; six of these had complete pathologic response. MRI correlated better with tumor size on pathology in patients with anti-HER2neu-based regimens than in patients with cytotoxic chemotherapy-alone regimens (0.88 vs. 0.4, p = 0.0001). MRI also is more accurate at predicting pathological tumor size in patients with triple negative disease (p = 0.002).

Magnetic resonance imaging correlated well while MMG calcification measurements correlated poorly with tumor size on final pathology. Extent of calcifications on diagnostic mammography may not be accurate in preoperative evaluation of breast cancers after NAC.