Mammographically Occult Breast Cancers

Abstract

A small percentage of breast cancers are not visible on mammography. Since mammographically occult malignancies may be more difficult to diagnose, we hypothesized that the lack of visualization would cause a delay in detection, more aggressive surgical and adjuvant therapy, and poorer outcome. Patients with mammographically occult malignancies were compared to patients with cancers visible on mammogram. The significance of mammographic visibility for treatment and local and distant recurrence rates were evaluated. Ninety-one of the 813 (11%) cancers were mammographically occult. Patients with mammographically occult malignancies were significantly younger, of lower body weight, and had fewer pregnancies than patients with cancers visible on mammography: age, body weight, and parity were statistically significant (p < 0.001) in stepwise logistic regression. Ductal carcinoma in situ was significantly more frequently diagnosed in patients with mammographically visible malignancies (14% versus 4%, p = 0.0163) and nodal involvement was significantly more frequent in patients with mammographically occult malignancies (35% versus 24%, p = 0.0391). Diagnostic delays exceeding 3 months were experienced by 24% of patients with mammographically occult malignancies compared to 13% of patients with tumor visible on mammography (p < 0.0001). Adjuvant chemotherapy was given to 63% of patients with occult malignancies compared to 41% of patients with mammographically visible cancers (p = 0.0027). The use of breast-conserving therapy and adjuvant radiation and tamoxifen were comparable. Survival free of local recurrence and distant metastases for the 403 patients followed for 5 years or more was not related to mammographic visibility.

Screening ultrasonography revealed 15% of mammographically occult breast cancers

“…Tarter et al [5] reported that 11% (91 of 813) of breast cancer in their series could not be found with mammography. They also reported that false-negative results are associated with lower age, weight, and parity.…”

Mammographically occult breast cancers detected with AI-based diagnosis supporting software: clinical and histopathologic characteristics

“…This is in line with the finding of McKinney et al, in which the distribution of cancer type was not significantly different between AI-detected and radiologist-detected cancers [ 21 ]. In previous studies, invasive carcinomas were more likely to be mammographically occult compared with DCIS [ 5 , 28 ], and most of the cancers detected only by an AI software were invasive cancers [ 21 ]. Another study also showed that the sensitivity of an AI software was higher in invasive cancers than in DCIS [ 10 ].…”