Multicentric/multifocal breast cancer

The role of SLNB in multicentric and multifocal disease in breast cancer is not yet clearly defined. Multicentricity is described as the presence of more than one cancer spots in the breast, occurring in separate quadrants or at a distance of more than 2–5 cm from each other [Reference Gump3]. In contrast to that definition, the multifocal disease is referred to the presence of several foci in the same quadrant of the breast.

It occurs in approximately 10–63% [Reference Gump3,Reference Vlastos, Rubio and Mirza4] of the cases and may result in inaccurate lymph node staging and high false-negative rates after an SLNB [Reference Veronesi, Paganelli and Viale5,Reference Hsueh, Turner and Glass6]. The most trials addressing SLNB have excluded women with multicentric or multifocal lesions. Therefore, this topic is not clearly evaluated and a recommendation is difficult to give. Most experts and panelists have been considered multicentricity/multifocality as a relative contraindication because of the suspicion that several tumor spots might drain to more than one dominant lymphatic pathway and sentinel lymph node (SLN). However, this point of view begins to change. At the beginning of SLNB era, this biopsy technique was used by injecting either radiolabeled colloid or blue dye peritumorly, that is directly near the tumor lesion [Reference Krag, Weaver, Alex and Fairbank7,Reference Giuliano, Kirgan, Guenther and Morton8]. Recently some trials have evaluated different sites of injection like intradermal [Reference McMasters, Wong and Martin9,Reference Martin, Derossis and Fey10], subdermal [Reference Veronesi, Paganelli and Viale5] or subareolar [Reference Kern and Rosenberg11,Reference Klimberg, Rubio and Henry12]. It has been shown that these injection sites were associated even with greater success and a comparable false negative rate to that associated with the standard injection site [Reference Kern13]. These results support the theory that all quadrants of the breast drain into the same lymph node. Assuming the idea that the whole breast drains to the same SLN, one could suggest that SLNB can even be performed in multicentric and multifocal cancer. This theory is supported by several authors [Reference Klimberg, Rubio and Henry12,Reference Tuttle, Colbert and Christensen14–Reference Borgstein, Meijer, Pijpers and van Diest16]. Additionally, some investigators were able to show that the test performance of SLNB in multicentric and multifocal disease is nearly equivalent to that for unifocal breast cancer [Reference Kumar, Jana and Heiba17–Reference Tousimis, Van Zee and Fey19]. Identification rates ranged from 90% to 97% [Reference Knauer, Konstantiniuk and Haid20] and false negative rates from 0% to 8% [Reference Tousimis, Van Zee and Fey19]. Knauer et al. [Reference Knauer, Konstantiniuk and Haid20] reported on a multi-institutional trial in which the SNB-feasibility and accuracy was evaluated in 142 patients with multicentric breast cancer. Compared to patients with unicentric cancer it has been shown a higher rate of SN metastases, whereas there was no difference in sensitivity (96.0%), negative predictive value (93.3%) and overall accuracy (97.3%) [Reference Knauer, Konstantiniuk and Haid20].

Nevertheless, it should be pointed out that the above-mentioned studies were only small and non-randomized series, so that further evaluations of this procedure have to be done. Therefore, we finally conclude with the German Society of Senology and point out that there is not sufficient evidence to recommend the procedure of SLNB in multicentric disease outside of randomized trials for routine clinical practice [Reference Kuehn, Bembenek and Decker21]. However, the multifocal breast cancer should not represent a contraindication according to the leading professional societies and published trials [Reference Lyman, Giuliano and Somerfield2,Reference Kuehn, Bembenek and Decker21,Reference Schwartz, Giuliano and Veronesi22].

Prior surgery – previous biopsy

The impact of prior surgery of the breast and the axilla on accurate lymphatic mapping has not been evaluated very well either. On one hand, there are the oncologic biopsy procedures like incisional, excisional or core biopsy, which seem not to affect the success of SLNB according to some limited data [Reference Haigh, Hansen, Qi and Giuliano23]. On the other hand, there are the non-oncologic and oncologic procedures like reduction and augmentation mammoplasty or breast reduction that are not sufficiently examined. Neither the breast surgery nor the axillary surgery (i.e. mammoplasty through an axillary incision) is fully evaluated.

Concerning the prior breast surgery with the aim of a diagnostic biopsy, growing data suggest that the success of SLNB is not affected [Reference Haigh, Hansen, Qi and Giuliano23–Reference McMasters, Tuttle and Carlson27]. Nevertheless, it has to be emphasized that many previous trials on SLNB have excluded women with excisional biopsy [Reference Albertini, Lyman and Cox28] and/or previous axillary surgery [Reference Krag, Weaver and Ashikaga29]. Additionally, some authors reported on a higher failure rate after an excisional biopsy [Reference Krag, Weaver and Ashikaga29–Reference Feldman, Krag and McNally31], whereas others found promising results. Miner et al. [Reference Miner, Shriver, Jaques, Maniscalco-Theberge and Krag32] were able to clarify that the type of previously performed diagnostic biopsy or the location of the primary lesion did not influence the localization of the SLN. In another study it has been shown that excision done before SLNB did not interfere with the subsequent identification rate of the SLN [Reference Wong, Edwards and Chao33]. It could be shown that neither the excision volume nor the time interval between the diagnostic biopsy and the lymph node mapping did play a role in affecting the success of SLNB [Reference Haigh, Hansen, Qi and Giuliano23]. So, taking into account the available data, a prior biopsy seems not to influence the successful lymphatic mapping.

The other group of prior breast surgery with non-oncologic and oncologic aims has been less examined. The effect on the identification rates of the SLN in women who have undergone a non-oncologic breast surgery, for example reduction mammoplasty, is not clearly known. Probably, the most important point for the correct identification of the SLN is the presence of an intact lymphatic pathway between the tumor and the axilla. If the tumor is situated in the upper quadrant, it is likely that, for example a reduction procedure does not affect the lymphatic drainage. Therefore, a sentinel procedure could be done, especially when prior surgery was performed more than 6–12 month previously. In the re-operative oncologic setting, it could be speculated that the intact lymphatic pathway seems to be as essential as in the non-oncologic setting for a successful identification of the sentinel node. Up to date, there are no randomized clinical trials (RCT) concerning this special situation.

The impact of the axillary surgery also remains uncertain. In a retrospective trial it was pointed out that SLNB after prior axillary surgery, like ALND or SLNB, is associated with a higher rate of failure, of 25% [Reference Port, Fey and Gemignani34]. Although there have been only 32 cases in this study that have had an attempted sentinel biopsy after prior axillary surgery, this result indicates a negative effect of repeated SLNB. On the other hand, Intra et al. [Reference Intra, Trifiro and Viale35] published data from a single institution trial that show feasibility and efficacy of SLNB in the re-operative setting. The authors demonstrated an identification rate of 87% in patients with 10 or less axillary lymph nodes previously removed [Reference Intra, Trifiro and Viale35]. However, in patients with more than 10 lymph nodes excised, the rate decreased to 44%. The recently published results of Port et al. [36] support these data. The authors were able to show in an unicenter trial that the success of re-operative SLNB was inversely related to the number of nodes removed previously and was more likely to be successful after a previous SLNB than a previous ALND (74% vs. 38%) [36].

Taken together, these results suggest that SLNB should not be performed after previous axillary surgery, neither in the oncologic setting nor in the non-oncologic setting. However, in a selected group of patients this procedure could be taken into account, especially after prior SLN procedure. SLNB can be performed in patients who have had a previous excisional biopsy and even in patients after a limited non-oncologic breast surgery [Reference Schwartz, Giuliano and Veronesi22]. In summery, the application of SLNB after extensive previous breast surgery has a diminished identification rate and should therefore be contraindicated.

Large and locally advanced breast cancer

Most experts recommend the procedure of SLNB in T1 and T2 tumors [Reference Lyman, Giuliano and Somerfield2]. Although there is only one randomized and controlled published trial in which SLNB was compared with ALND for the effects on long-term survival [Reference Veronesi, Paganelli and Viale37], there is a strong evidence for the high quality of test performance of SLNB in patients with early breast cancer, given by the numerous amount of test performance trials [Reference Lyman, Giuliano and Somerfield2]. Nevertheless, it should be emphasized that most validation trials included only patients with T1 and T2 tumors. A recently published meta-analysis and three other additional systematic reviews have underlined the advantage and quality of this procedure in this special group of women [Reference Lyman, Giuliano and Somerfield2,Reference Kim, Giuliano and Lyman38–Reference Fraile, Rull and Julian40]. As stated in another representative multicenter setting, the sensitivity of SLNB is about 90% and the false negative rate is about 10% [Reference Tafra, Lannin and Swanson25,Reference Krag, Weaver and Ashikaga29] for early breast cancer. However, for the locally advanced and larger T3 and T4 tumors, there are also a few reports that describe SLNB in these patients. Some authors did reveal that the identification rate and the sensitivity of SLNB in patients with smaller tumors were equivalent to those with breast tumors larger than 4 cm [Reference Jakub, Pendas and Reintgen41]. Chung et al. [Reference Chung, Ye and Giuliano42] reported on an acceptable accuracy rate in patients with larger tumors (⩾5 cm). In this trial, the SLN status accurately predicted the regional nodal status in 98% (40 of 41) of cases [Reference Chung, Ye and Giuliano42]. Wong et al. [Reference Wong, Chao and Edwards43] published data of an analysis that was performed to determine whether tumor size affects the accuracy of SLNB. The SLN identification rate, false negative rate and overall accuracy of SLNB were not significantly different among tumor stages T1, T2 and T3 [Reference Wong, Chao and Edwards43]. Another study did reveal similar results. The SLN was identified in 99% of cases. SLNB false-negative rate was 2% [Reference Bedrosian, Reynolds and Mick44]. Although these data [Reference Jakub, Pendas and Reintgen41–Reference Bedrosian, Reynolds and Mick44] seem to support the statement that tumor size does not harm the sensitivity, we conclude with the recommendation that SLNB should not be used routinely for this group of patients because of the small amount of RCT. In a selected group of patients with advanced breast cancer, SLNB has also been used after neoadjuvant chemotherapy [Reference Schwartz, Giuliano and Veronesi45]. It could be shown an identification rate between 85% and 94% [Reference Mamounas, Brown and Anderson46–Reference Breslin, Cohen and Sahin48] and a false negative rate of 12% [Reference Mamounas, Brown and Anderson46]. These data seem to indicate a potent role for SLNB in this setting. Nevertheless, there have to be more trials including large numbers of patients to clarify this indication of SLNB [Reference Kaufmann, von Minckwitz and Smith49]. In the meantime, SLNB after neoadjuvant chemotherapy should be restricted to the hands of experts with adequate experience [Reference Kaufmann, Hortobagyi and Goldhirsch50].