This is the first known study in Southeast Asia to incorporate radiological and histological characteristics to develop a predictive model for malignant upgrading of ADH diagnosed on core needle biopsy.
The upgrade rate in our study was 28.9%, in the lower range of reported studies of 7 to 87% [1, 2, 15-20]. This could be a result of improved breast imaging modalities, and a more stringent review of our screening mammograms by multiple dedicated breast radiologists. All lesions after biopsy are also discussed at a multidisciplinary meeting with radiologists, pathologists and surgeons.
Our BSS participation rate increased from 10% in 2005 to 39.6% in 2010. Performance indicators, with the except of recall rates, specificity, interval cancer rate (for first screen), generally improved over the years and are comparable with organized breast screening programmes in other developed countries [24, 25].
We found that the strongest factors predictive of upgrade to malignancy were presence of mass on sonography, mammographic parenchymal density and number of foci of ADH. Multivariate modelling was used to identify a subgroup of women with ADH on core biopsy who have low risk (2.1%) of harboring concomitant malignancy, in whom surgical excision might be avoided. Approximately one eighth of our study population met the low-risk criteria, defined as (1) absence of mass on ultrasound, (2) mammographic parenchymal density of scattered areas of fibroglandular density, (3) two or less foci of ADH on core biopsy.
Similar to previous reports, our study also shows a higher upgrade rate in patients with higher mammographic parenchymal density . Mammographic percent density, computed as the proportion of the parenchymal area occupied by radiologically dense breast tissue, has been related to screening sensitivity and specificity and is one of the strongest established risk factors for the development of DCIS  and invasive breast cancer . Mammographic density is a function of genetic factors , is higher in nulliparous women, and is inversely associated with age and body mass index (BMI) [13, 16]. Asian women have a lower average BMI compared to their western counterparts and more frequently have dense breasts on mammography [27-30]. The increased mammographic parenchymal density could account for the higher upgrading rates of CNB-diagnosed atypical ductal hyperplasia [31-35] as well as the younger peak age of breast cancer (10 years younger) [36, 37] seen in Asia compared to Western countries.
Microcalcification with or without a mass has been reported to be the most common finding on screening mammograms for both ADH (58%, 88%) and DCIS (68%, 98%) [38-40]. Ninety percent of non-palpable DCIS and 20% of non-palpable infiltrating carcinomas were identified by their microcalcifications . The microcalcification morphology and distribution on imaging have been predictive of upgrading in several studies. In particular, calcifications which were clustered and had a segmental or linear branching distribution had a significantly higher association with malignancy. A subgroup with fine, rounded calcifications without density or mass has been identified as having a low risk of malignancy . In our study, we demonstrated that patients with diffuse calcifications were more likely to have their core biopsy diagnosis upgraded on subsequent excision biopsy. Inherent in diffuse calcifications is the increased difficulty of taking an adequate representative sampling compared to clustered calcification. Hence, in patients with diffuse calcifications on their mammogram, it is recommended that sufficient and representative samples are taken for core biopsies to increase the adequacy of biopsy sampling. Likewise, for patients with ADH and diffuse calcifications, where the risk of false negative is higher, surgical excision needs to be considered if the adequacy of the sampling is of concern.
Although the breast screening programme is designed for women who are clinically asymptomatic, 12% (11/90) of the screening mammograms revealed a mass, of which 64% (n=7) were clinically palpable. Out of the 11 mammograms harbouring a mass lesion, 5 lesions were upgraded to in situ or invasive carcinoma on excision biopsy. This underscores the importance of regular breast self-examination in addition to regular mammographic screening in detecting breast cancer in its early stages, similar to previous reports . The vast majority of women diagnosed with ADH on CNB were however asymptomatic, with anomalous findings picked up only with mammographic screening. This highlights the complementary roles that breast self-examination and mammographic screening play. This is in contrast to the Canadian study that showed that annual mammography screening had no effect on breast cancer mortality beyond that of breast physical examinations . This may be attributed to a difference in our culture and population screening habits, with a much to be improved percentage uptake of mammogram screening in Singapore.
Some studies have reported that the presence of a mass on imaging as a significant predictor of invasion [44, 45]. The presence of a mass on imaging likely represents a more aggressive lesion with higher potential of local invasiveness: breaking through the basement membrane of the breast ducts and infiltration of adjacent tissues. In our study, the presence of a mass on either mammogram or ultrasound was associated with increased risk of upgrading to DCIS or invasive cancer, with presence of mass on ultrasound being an independent predictor of upgrade to in-situ or invasive malignancy. As ADH is a microscopic lesion, it is rare for pure ADH to present as a mass, and therefore one should have a high suspicion of a false sampling, should a biopsy of a breast mass yield only ADH.
The histological criteria found to be independently associated with cancer upgrade in our study was the number of foci of ADH seen on core biopsy. Similar to previous studies [17, 18], our study also demonstrates that ADH involving 3 or more foci in CNB is an independent predictor of upgrading on excision, with upgrading found in 13 (65.0%) out of 20 such cases, in contrast with 13 (18.6%) out of 70 cases with ADH confined to 1 or 2 foci. These findings suggest that ADH confined to 1 or 2 foci in adequately sampled CNB specimens is predictive of a good outcome and that a greater number of foci of involvement strongly correlates with concomitant malignancy.
Some studies have purported that micropapillary histologic subtype predicted the presence of DCIS [17, 46, 47]. However, the architectural pattern of ADH relating to upgrading risk was not observed in our study (p=0.80).
Although there is no universal agreement about the number of cores necessary to achieve an accurate histological diagnosis, most studies have demonstrated that increasing the number of core biopsies can decrease the risk of malignancy underestimation [38-41, 48]. Similarly, our study shows decreased upgrade rates with larger samples via larger needle gauge size and the use of vacuum device. However these did not reach significance on multivariate analysis. It is probable that this lack of independent correlation may be related to the size of the mammographic or pathologic lesion biopsied. One would also expect higher rates of underestimation when sampling is less adequate, which would be more likely in the case of larger target lesions. It also must be emphasized that, as shown also in the present study, even with use of vacuum assistance with an 11-gauge needle and complete removal of the mammographic lesion, there is still a 20% risk of malignant upgrade with the diagnosis of ADH.
Prior studies have investigated whether the absence of residual microcalcifications after biopsy would obviate the need for surgical excision and the consensus is mixed [18, 21, 49, 50]. In the study by Bonnett et al, residual calcifications were present in all cases showing carcinoma at open excision  . However, residual calcifications had low specificity because they were present in 71% of cases having negative surgical excisions. In our study, the presence of residual calcifications associated with ADH lesions is a strong predictor of upgrading on subsequent excision, with a sensitivity of 0.96 in our patient population. In the absence of residual calcifications post biopsy, 20% (n=7) of CNB-diagnosed ADH were still upgraded to in-situ or invasive cancer after surgical biopsy.
For women found to have a low risk of upgrade, short term radiologic follow up should be adequate to detect progression. This is supported by the fact that all but one of the upgraded lesions were in- situ disease. The only case upgraded to invasive carcinoma was a case of mucinous carcinoma for which CNB revealed a mucocele like lesion with ADH. Mucinous carcinoma usually carries a favourable prognosis with a low recurrence and metastatic rate. This was also demonstrated by Menen et al.  , where 125 women with low-risk ADH (using Nguyen et al.’s criteria ) were observed without surgical excision. With a median follow-up of 3 years (and chemoprevention use by 23% of women), only seven breast cancers occurred (5.6%). Index site and ipsilateral cancer rates were comparable to the group that underwent excision while contralateral breast cancers only occurred in the surgical group.
Although there were predictive tools proposed before, they have been validated in different populations [21,43]. We have developed a predictive tool based on radiological and histological variables to predict risk of malignant upgrading in core needle biopsy diagnosed ADH in a multiracial heterogeneous South-east Asian population, as shown in Figure 1. The variables included differ from the previous predictive tools and include histopathology in addition to imaging. Our results show that in patients whose risk score is less than or equal to -10, there is a 2.1% or less chance of malignant upgrading. This is equivalent to the risk of a BIRADS 3 lesion which can be conservatively managed with short interval radiological surveillance. With further validation on a larger cohort, this nomogram has potential to be used as a adjunct to aid the surgeon and the patient to make an informed decision regarding open excision biopsy following a diagnosing of ADH on core needle biopsy.
In conclusion, we have retrospectively studied and examined the upgrade rate of core biopsy diagnosed ADH on excision biopsy in Singaporean women. The strongest factors predictive of upgrade to malignancy were presence of mass on sonography, mammographic parenchymal density and the number of foci of ADH. We have also developed a nomogram, to help identify a group of women who are at low risk (<5%) of upgrade to cancer, and may be candidates for avoiding open excision surgery, following a core-needle biopsy diagnosis of ADH.