A large number of in situ malignancies and calcification patterns were found among women who had prior false positives on mammography screening in a new study from Spain. The findings suggest the false positives may have progressed to cancer and reinforces the importance of encouraging women with false-positive results to undergo regular screening, according to the researchers.
Most women with abnormalities detected on mammography screening will not be diagnosed with cancer after being recalled for further assessment with additional imaging or biopsy, the study authors wrote.
"However, women with false-positive results have been reported to be at higher risk for cancer detection in subsequent screening rounds," lead author Laia Domingo, PhD, from the department of epidemiology and evaluation at Hospital del Mar Medical Research Insitute in Barcelona, and colleagues wrote in their study to be published in Cancer Epidemiology (October 2013, Vol. 37:5, pp. 660-665). "Although this association seems consistent from a statistical point of view, a clear explanation is lacking, and few studies have specifically evaluated this relationship."
The high risk of breast cancer after a false-positive result may be partly explained by the false-negative hypothesis (i.e., cancers missed after further assessments in the screening that are diagnosed at the next screening), the authors noted. Another explanation is women with benign breast disease have a greater risk of developing invasive breast carcinoma.
To date, few studies have evaluated information on tumor characteristics and radiological appearance, and none have distinguished between a false positive that was followed with invasive procedures versus those that were assessed with additional imaging, according to the researchers.
The current study compared women's personal characteristics, tumor features, and radiological appearance of lesions among patients with and without a previous false-positive result. The study also assessed the agreement between the location of the false-positive lesion and that of the subsequent malignant tumor.
From 1996 to 2007, 111,098 women between the ages of 45 and 69 participated in four population-based breast cancer screening programs in Spain, and 1,281 cancers were detected. All cancers detected in subsequent screenings (n = 703) were included, and the researchers explored the occurrence of previous false-positive results.
The researchers identified false positives requiring additional imaging or invasive procedures. Differences in tumor features (invasiveness, tumor size, and lymph node status), as well as radiological appearance, were assessed by the Chi-square test, and agreement between the location of cancer and prior suspicious lesions were assessed by Cohen's kappa coefficient. A multivariate analysis was performed to evaluate the effect of previous screening results and age on the odds of presenting with an in situ carcinoma.
Among the 703 cancers detected in subsequent screenings, 148 women (21.1%) had a previous false-positive result, according to the authors. Of these, 105 were by additional imaging and 43 by invasive procedures.
Women with prior false-positive results requiring invasive assessment and women with prior false positives requiring additional imaging had a higher proportion of in situ carcinomas and microcalcifications compared with women with negative tests.
The proportion of in situ carcinomas was even higher in women older than 60, and ipsilateral cancer was observed in 65.7% of cases with prior cytology or biopsy.
| Breast pathology recurrence by type of follow-up | |||
| Prior false positives requiring additional imaging | Negative tests | False positives requiring invasive assessment | |
| In situ carcinomas | 12.9% | 15.3% | 31.7% |
| Microcalcifications | 9.5% | 20.2% | 37.2% |
The results reinforce the importance of encouraging women with false-positives results to attend regular screening, because they could still develop cancer at a later date.
"Our data reveal that the percentage of in situ cancers is higher among women with a previous invasive assessment that excluded a malignancy, especially among women older than 60 years," Domingo and colleagues wrote. "Cancers detected after an invasive assessment were also associated with a higher percentage of microcalcifications, and two out of three cancers were detected in the same breast where the first suspicious lesion was assessed."
The exclusion of cancer after an invasive procedure does not indicate the absence of a benign abnormality, the researchers noted. Given a high index of suspicion, most biopsies and cytological analyses have a final diagnosis of a benign breast lesion.
Regarding radiological appearance, cancers with prior additional imaging were more likely to present as masses than cancers with prior negative assessments, they added. Calcifications and distortions, which are the two radiological patterns with the highest positive predictive value, accounted for a small proportion of tumors with prior additional imaging.
Limitations
Although the data were drawn from a project including more than 240,000 screening tests, the sample size was relatively small, since the occurrence of cancer after a false-positive result is fairly infrequent.
Also, some important variables were not routinely collected by breast cancer screening program databases. Information on breast density was unavailable, and consequently its potential as a confounder could not be tested, the authors noted.
In addition, interval cancers were not included in the study, and some of the lesions identified as false positives could also progress to an interval cancer.
Future studies should concentrate on analysis of interval cancers, which could aid comprehension of cancers that appear shortly after a false-positive assessment, the researchers wrote.
"Our results provide new insights into the association between a false-positive result and subsequent cancer detection within a screening program, which can probably be explained by a combination of factors, including false-negative results and progression of benign lesions," they concluded.
