Surgical Management of the Axilla in Clinically Node-Positive Breast Cancer Patients Converting to Clinical Node Negativity through Neoadjuvant Chemotherapy: Current Status, Knowledge Gaps, and Rationale for the EUBREAST-03 AXSANA Study

Simple Summary Currently, it is unclear which kind of axillary staging surgery breast cancer patients with lymph node metastasis should receive after neoadjuvant chemotherapy. For decades, these patients have been treated with a full axillary lymph node dissection, even if they converted to clinical node negativity. However, the removal of a large number of lymph nodes during the procedure can increase arm morbidity and impact quality of life. Therefore, several studies investigated less radical surgical strategies in this setting, such as sentinel lymph node biopsy or targeted axillary dissection, i.e., removal of a previously marked node combined with sentinel node removal. In this review, we summarize current evidence on the different surgical techniques and compare national and international recommendations. We show that many questions regarding oncological safety of different surgery types and the optimal marking technique remain unanswered and present the multinational prospective cohort study AXSANA that will address these open issues. Abstract In the last two decades, surgical methods for axillary staging in breast cancer patients have become less extensive, and full axillary lymph node dissection (ALND) is confined to selected patients. In initially node-positive patients undergoing neoadjuvant chemotherapy, however, the optimal management remains unclear. Current guidelines vary widely, endorsing different strategies. We performed a literature review on axillary staging strategies and their place in international recommendations. This overview defines knowledge gaps associated with specific procedures, summarizes currently ongoing clinical trials that address these unsolved issues, and provides the rationale for further research. While some guidelines have already implemented surgical de-escalation, replacing ALND with, e.g., sentinel lymph node biopsy (SLNB) or targeted axillary dissection (TAD) in cN+ patients converting to clinical node negativity, others recommend ALND. Numerous techniques are in use for tagging lymph node metastasis, but many questions regarding the marking technique, i.e., the optimal time for marker placement and the number of marked nodes, remain unanswered. The optimal number of SLNs to be excised also remains a matter of debate. Data on oncological safety and quality of life following different staging procedures are lacking. These results provide the rationale for the multinational prospective cohort study AXSANA initiated by EUBREAST, which started enrollment in June 2020 and aims at recruiting 3000 patients in 20 countries (NCT04373655; Funded by AGO-B, Claudia von Schilling Foundation for Breast Cancer Research, AWOgyn, EndoMag, Mammotome, and MeritMedical).


Introduction
In breast cancer patients, the optimal surgical management of the axilla has been controversially discussed over the last two decades. For many years, axillary lymph node dissection (ALND) has been considered as the standard of care. The rationale behind this was firstly to assess the pathological lymph node status (diagnostic value, "staging") and secondly to provide locoregional control (therapeutic value). Due to its high morbidity, this approach has gradually been abandoned in favor of sentinel lymph node biopsy (SLNB), a less invasive procedure for axillary staging of clinically node-negative patients, over the past two decades. In recent years, SLNB has also become standard of care for patients with clinically unsuspicious nodes at time of diagnosis who have completed neoadjuvant chemotherapy. The detection rate and accuracy of SLNB are excellent in this setting, and axillary recurrence rates are negligible [1][2][3].
Nonetheless, in patients with clinically apparent axillary lymph node metastases (cN+) at time of diagnosis who achieve complete clinical response in the axilla (ycN0) after neoadjuvant chemotherapy (NACT), it is unclear which axillary surgical staging strategy should be offered. This uncertainty is expressed in the heterogeneity of recommendations endorsed by different national and international societies, which range from SLNB to targeted axillary dissection (TAD) or ALND (Tables 1 and 2). Some societies do not consider SLNB standard of care in this setting because of the relatively high false negative rates (FNRs) reported in large prospective trials and confirmed in a meta-analysis [1,[4][5][6].
In the SENTINA and the ACOSOG Z1071 trials, FNRs were 12% and 14% respectively, and thus higher than the arbitrarily chosen but widely accepted cut-off value of 10% (Table 3). So far, however, only limited data regarding the oncologic outcome following SLNB alone are available in this setting [3].

Type of Surgery Description
Axillary lymph node dissection (ALND) Systematic removal of lymph nodes from the axilla, usually level I and II, sometimes including also level III Sentinel lymph node biopsy (SLNB) Identification and removal of the sentinel lymph node, usually using radioactive tracer (Technetium-99) or blue dye Targeted lymph node biopsy (TLNB) Selective removal of metastatic lymph node(s) marked before neoadjuvant therapy Targeted axillary dissection (TAD) Combination of TLNB and SLNB Marking positive lymph nodes at the time of diagnosis and prior to neoadjuvant chemotherapy with clips, coils, radioactive seeds, or other markers may improve the FNR of de-escalated surgical staging procedures [7,8]. The best marking technique, however, has not been unanimously identified yet. Importantly, data comparing recurrence rates and surgical morbidity among SLNB, ALND, and TAD are so far not available.
As a result, the German S3 guidelines updated in 2020 recommend ALND after NACT in cN+ patients, as do Austrian and Scandinavian guidelines. As a contrast, high-impact networks such as European Society of Medical Oncology (ESMO) in Europe and National Comprehensive Cancer Network (NCCN) in the USA recommend SLNB, provided that dual tracers are used and a minimum of three sentinel nodes are removed [9,10]. In countries such as Italy, Denmark, Russia, and Hungary, SLNB or TAD are accepted as first choice for axillary staging in this group of patients. The German Breast Committee of the Working Group for Gynaecological Oncology (AGO Breast Committee) endorses both TAD and ALND as recommended strategies [11].
Unanswered questions include the role of axillary imaging for selection of patients who might safely be offered surgical de-escalation [12]. Further, the necessity of regional therapy, e.g., radiotherapy following axillary staging in patients with pathological complete response on SLNB or TAD, is still a matter of debate.  cN1 patients should be clipped and ycN0 patients should be managed by TAD, with omission of ALND in ypN0 if the following criteria are fulfilled: (1) SLNB performed using dual traced, (2) clipped node removed, and (3) more than 2 removed nodes Not provided Russia (Association of Oncologists of Russia) [25] It is recommended to mark the tumor before starting neoadjuvant therapy to enable visualization during subsequent surgical treatment.
If it is impossible to perform SLNB or if a metastatic focus in the SLN is detected, it is recommended to perform ALND.

III, B I, A
Spain (Spanish Society of Medical Oncology) [26] ALND is recommended. In selected cN+ cases, in which positive axillary node has been marked prior to NACT, the identification and recovery of >2 negative SLNs (including the marked node) with a double tracer technique (Tc99 and methylene blue) may avoid ALND.
I, A II, C Table 3. Studies on sentinel node biopsy after neoadjuvant therapy in initially node-positive patients.  2 Only in patients with at least 2 sentinel nodes removed (pre-defined study criterion); in case of only one sentinel node removed, the false negative rate was 29.3% [7].

Number of Patients Preoperative Axillary Assessment
In order to shed light on this much debated topic, the European Breast Cancer Research Association of Surgical Trialists (EUBREAST) has initiated AXSANA (AXillary Surgery After NeoAdjuvant Treatment), a multinational prospective cohort study (NCT04373655) which enrolls cN+ patients undergoing NACT who convert to ycN0. The aim of AXSANA is to assess the impact of different surgical staging procedures in the axilla on the oncologic outcome and on health-related quality of life.

Which Marking Technique Is Optimal?
So far, several methods for marking of target lymph node(s) have been developed, usually based on techniques that are already in use for the localization of non-palpable breast lesions (Table 4). Interestingly, there are notable regional differences regarding the use of various techniques. The same method may be the technique of choice in one country, while being completely unknown in another.  To date, the largest amount of data has been published on clip-based targeted axillary dissection (TAD). Unless intraoperative ultrasound is used, this strategy requires a preoperative localization step, performed either by the use of a wire or by placing another marker (e.g., magnetic or radioactive seed, radar marker or radiofrequency identification [RFID] tag) into the clipped area that will allow identification during surgery. Still, the success of target lymph node (TLN) removal depends on the ultrasound visibility of the clip inserted before NACT.
The study that brought international attention to the technique was a retrospective analysis of a prospective database at the M.D. Anderson Cancer Center (Table 5) [8]. Nearly all patients in this study underwent ultrasound-guided placement of a radioactive iodine-125 seed into the previously clipped node prior to surgery. This strategy offers more flexibility than wire localization, which was used in two patients only, because the seed can be inserted several days before surgery, whereas the wire placement is usually scheduled for the morning before the operation or (rarely) the day before. The study by Caudle et al. showed that the FNR of TAD can be as low as 2.0% [8]. However, since only patients with successful preoperative localization were included in this analysis, it is unclear whether the clip could not be visualized in some patients, making preoperative localization impossible.   The results of the prospective German multicenter SENTA register study were presented at the ESMO conference in 2019 and are now available as a full publication [28,29]. In this study, the TLN was clipped before NACT in 473 patients. A Tumark Vision clip (SOMATEX®, Berlin, Germany) was used in the majority of cases (71%), followed by an O-Twist clip (BIP, Türkenfeld, Germany) (12%). Clip types used in the remaining patients have not been described. In 50 out of 473 patients, a targeted lymph node biopsy (TLNB) was not attempted either because the clip was not visible upon preoperative ultrasound or TLNB was for some reason not planned by the investigator. The TLN could be successfully removed in 329 of 423 patients, mostly after wire localization, resulting in an overall removal rate of 78% among patients in whom a TLNB/TAD was attempted. The TLN detection rates were higher in case a Tumark Vision or a O-Twist clip was used, compared to other clips (79.7%, 78.4%, 69.4%, respectively). Interestingly, triple negativity and high Ki67 index were associated with TLNB failure (i.e., TLN not removed during TLNB attempt). In 63% of patients, the SLN and the TLN were identical. In the subgroup of 278 patients who received ALND, TLNB alone and TAD resulted in FNRs of 7.2% and 4.3%, respectively.
Another marking method is the injection of a small volume (0.1-0.5 mL) of a carbon solution into the suspicious lymph node. At surgery, the TLN is identified visually by its dark staining. The results of TATTOO, the largest prospective trial on carbon solution-based TAD, have been presented recently [47]. In 94% of 118 included patients, the marked node could be detected intraoperatively. In 60% of cases, SLN and TLN were identical. In 5 (4.5%) patients, unintentional tattooing of the skin was observed, but it remains unclear whether this effect is permanent. In 61 cN+ patients converting to ycN0, completion ALND was performed. In this subgroup, three out of 33 patients with residual axillary disease had not been correctly identified by TAD, resulting in a FNR of 9.1%. A similar FNR of 8.3% was reported from the only other study on FNR in carbon solution-based TAD [50].
The MARI technique ("Marking the Axillary lymph node with Radioactive Iodine seeds") was reported from the Netherlands in 2015 [44]. It implies that a titaniumencapsuled radioactive seed is inserted into the suspicious lymph node before NACT under ultrasound guidance. In contrast to technetium-99-based localization of non-palpable breast lesions before primary surgery, the longer half-life (59 days) of iodine-125 allows for use in the neoadjuvant setting. Donker et al. evaluated the MARI technique in 100 patients scheduled for NACT [44]. Marking of the biopsy-proven positive lymph node and the breast tumor using identical seeds was conducted at one procedure. In three patients, the seed could not be properly positioned in the lymph node, resulting in a removal rate of the biopsied node of 97%. No relevant loss of signal was observed after a median NACT duration of 17 weeks, and all seeds could be removed successfully. Importantly, no SLNB was performed in this study, and thus the reported FNR of 7% refers to the TLNB alone.
Two aspects have been critically discussed after the introduction of the MARI technique. First, the procedure requires complex safety regulations and cannot be performed in some countries due to national radiation protection laws. Secondly, both the localization of iodine-125 and technetium-99 used for SLN mapping require a gamma probe so interference might potentially occur in case of down-scatter of technetium-99m into the energy spectrum of iodine-125. The first results of the RISAS trial (NCT02800317), presented by Simons et al. at the San Antonio Breast Cancer Symposium in 2020 [53], could, however, show that the combination of the MARI technique and SLNB is feasible: TAD was successfully performed in 223 out of 227 (98%) patients. The trial was designed to test the non-inferiority of TAD over ALND, with FNR as the main endpoint. Non-inferiority was assumed if the upper bound of a 95% confidence interval around the FNR was lower than 6.25%. Although the FNR was low (3.5%), the prespecified primary endpoint was not met (95% confidence interval 1.38-7.16). The authors reported, however, that 4 out of 5 patients with a false-negative result (i.e., residual lymph node metastasis in the ALND specimen despite tumor-free TAD) were found among the first 10 enrolled patients of participating institutions, suggesting a learning curve that should be considered when planning future studies.
Another technique to mark a biopsy-proven lymph node metastasis is based on the placement of a magnetic seed. So far, the only magnetic seed for axillary node marking evaluated in studies is the MagSeed ® (Endomag, London, United Kingdom). The seed is detected during surgery by using a magnetic probe called Sentimag ® (Endomag, London, United Kingdom) surgical guidance platform. This method has already provided evidence for the successful resection of non-palpable breast lesions [54]. Its use for TAD was restricted by the initial requirement to remove seeds within 30 days after placement, but recently, both the Food and Drug Administration (FDA, 2018) and the Conformité Européenne (CE) marking (2020) approved long-term implantation in any soft tissue, thus allowing use in the neoadjuvant setting. First data on the magnetic seed-based TAD have been recently published by Thill et al. [45]. This small study showed a detection rate of 100%. Since magnetic seeds can lead to large magnetic resonance imaging (MRI) artifacts, it should be used with caution in patients in whom MRI-based assessment of response during NACT is planned.
Other possible techniques are radar reflector-localization clips [55] and radiofrequency identification (RFID) tags [56,57], that have provided promising results for non-palpable lesion localization, while data on lymph node marking are limited. Table 4 shows potential advantages and disadvantages of various methods and Table 5 gives an overview of current evidence.

How Many Nodes Should Be Marked?
Since TAD is a relatively new technique that still needs to be validated and standardized, various institutions follow different strategies. So far, there is no consensus on the number of nodes that should be marked in patients with more than one suspicious lymph node on imaging. While most studies report the marking of a single node, it is unclear which lymph node was selected for biopsy and marking in these cases. Options vary between the largest, the "most suspicious", or the best accessible lymph node [8,47]. Other authors report marking of all suspicious nodes [42,48]. Both strategies offer their own advantages and disadvantages (Table 6).
Regarding the accuracy of lymph node assessment, marking only one axillary lymph node in case of patients with several suspicious or biopsy-proven nodes might increase the FNR since the marked lymph node might be free of tumor while other previously suspicious but not marked lymph nodes may still contain tumor residuals and remain in situ. A heterogenous axillary response to NACT is common and can be diagnosed in up to 74% of patients with hormone receptor-positive HER2-negative disease, followed by 29% in triple-negative and 25% in HER2-positive tumors [58]. Lim et al. conducted a small study with a complex design to explore FNRs of different TLNB strategies in the same set of patients [42]. All suspicious nodes of a patient were marked with different types of clips to aid individual node identification (e.g., UltraCor Twirl, HydroMARK, and UltraClip).
Altogether, 21 nodes in 14 patients were clipped. After NACT, all patients received TLNB and ALND. The first clipped node alone had a FNR of 7.1%, which sank to 0% when the second clipped node was taken into account. No SLNB was performed, and it remains unclear how this might have influenced the FNR. Concerning arm morbidity, marking all suspicious nodes will necessarily result in removal of more lymph nodes. Natsiopoulos et al. conducted carbon solution-based TAD (Spot ® , GI Supply, Inc., Mechanicsburg, PA, USA) in 75 patients [48], marking each biopsy-proven or strongly suspicious node (median 2, range 1-5). At surgery, 2-10 nodes (median 4) were retrieved. How this relatively high number of removed nodes may affect quality of life and arm morbidity remains to be clarified. Importantly, a clinically relevant balance between high accuracy (more extensive staging) and low morbidity (less extensive staging) must be found.

When Should Lymph Nodes Be Marked?
To date, there is no consensus on the optimal timepoint of marker placement. In most studies, lymph node metastases were confirmed by fine needle aspiration or core biopsy. Such procedures, however, are not standard in some countries, and not all guidelines recommend routine minimally invasive biopsy in case of suspicious findings. Especially in case of large tumors and multiple highly suspicious nodes, clinicians may find it sufficient to perform biopsy of the breast tumor only.
If a minimally invasive biopsy is performed, the marker may be inserted into the same lymph node(s) at the same session or at a second session upon confirmation of metastasis by pathology/cytology. Data on this aspect lack detail in available studies but most authors report the placement of the marker into the "previously proven" node [44,48]. In contrast, marking of the suspicious lymph node immediately after the biopsy is reported by others [31,52]. Table 7 provides an overview of advantages and disadvantages of both strategies.

What to Do in Case of a "Lost Marker"?
A major concern expressed by clinicians trying to implement TAD at their institutions is the uncertainty of how to deal with patients in whom the marker has not been retrieved at surgery. Since a TAD/TLNB cannot be successfully performed in these patients, completion ALND is an obvious choice. Still, in some patients, the marker will not be found in the ALND specimen either. The pivotal study by Caudle et al. included 208 patients whose metastatic nodes were clipped prior to NACT [8]. In five of these patients, the clip could neither be identified in the surgical specimen nor upon radiography of the axilla, suggesting clip dislodgement. In another study, the clip could not be retrieved in two out of 73 patients, but no further details were reported [30]. To date, the management of patients with lost clips is unclear. In case of other markers, several additional concerns need to be addressed. Since leaving radioactive seeds behind results in a major radiation regulation breech that must be reported to regulation authorities, seed explantation should be attempted whenever possible unless it would jeopardize patient's well-being [59]. In case of magnetic seeds, no radioactivity is involved, but leaving a magnetic marker in the axilla can result in large MRI artifacts and thus compromise imaging assessment during post-treatment surveillance. Minor MRI artifacts are also possible in case of unsuccessful radar marker or RFID tag retrieval.
While lost clips can only be removed using imaging-guidance (usually radiography, ultrasound, or computer tomography), radioactive and magnetic seeds as well as radar markers and RFID tags can be identified using a special probe and carbon ink-marked nodes can only be visualized intraoperatively during dissection of the axilla, possibly resulting in different radicality of retrieval procedure and individual risk faced by the patient.

Is TAD Safe for All Patients?
It is yet unclear whether all cN+ patients can safely omit completion ALND in case of negative axillary staging. So far, while none of the studies on TAD have reported oncological outcome or health-related quality of life, there is limited information available on SLNB alone in this setting. Kahler-Ribeiro-Fontana et al. have recently reported on long-term outcomes following SLNB after NACT [3]. Among 222 clinically node-positive patients, 123 had no residual axillary disease at surgery. Of these, only two patients (1.6%) developed axillary recurrence after 3.6 and 5.5 years from surgery and were alive without disease at the last follow-up. Among all cN1/2 patients, no significant differences in 5-year and 10-year overall survival were found between patients who received ALND and those in whom ALND was omitted.
Most TAD validation studies include heterogeneous groups of patients. In the pivotal study by Caudle et al., nearly all patients had cT1-cT3 tumors, but 28% of patients presented with at least four abnormal nodes on ultrasound [8]. Similarly, in the largest study on carbon solution-based TAD, 25% of patients had ≥4 and 4% had ≥10 suspicious nodes before NACT [47]. 124 out of 423 (29%) patients enrolled in the SENTA trial had at least three abnormal nodes [29]. Intuitively, the more nodes appear suspicious on initial ultrasound, the more probable it is for the TAD to miss residual disease, especially in case only one node was marked, thus resulting in a higher FNR [60].
Other factors that can improve identification of patients who do not benefit from completion ALND are breast response to therapy and tumor subtype. A multivariable analysis of 13,396 patients showed that pathological complete response (pCR) n the breast was the most important predictor of pCR in the axilla (odds ratio 20.37 for yT0) [61]. Other studies reported a strong association between response in the breast and in the axilla, particularly in patients with HER2-positive and triple-negative breast cancer [62][63][64], implying that tumor subtype should probably also be implemented into a potential decision algorithm. Since patients with HER2-positive and triple-negative cancer achieving radiological breast response have the highest probability of reaching axillary pCR, they are probably most likely to benefit from de-escalation of surgical treatment, such as TAD.

Beyond Surgical Therapy: Which Fields Should Be Irradiated after TAD?
Another controversial issue is the target volume for nodal irradiation. While some guidelines exclude levels I and II from irradiation in case of a negative TAD, others suggest to cover all regions if they have not been assessed with a therapeutic intent. More specifically, it remains unclear whether patients with an extensive axillary tumor burden who achieve pCR after NACT benefit from additional regional treatment.
Koolen et al. from the Netherlands Cancer Institute proposed a hypothetical algorithm based on initial nodal status and response to NACT [65]. In this algorithm, patients receive not only ultrasound but also PET-CT prior to NACT. Those with 1-3 positive nodes are recommended no further axillary therapy in case of axillary pCR, defined as a negative MARI procedure (i.e., resection of a radioactive seed-marked TLN) [44]. In case of non-pCR, axillary radiotherapy is performed. For patients with ≥4 positive nodes before NACT, axillary radiotherapy is always indicated but patients with non-pCR are also recommended an ALND. In a cohort of 100 patients treated by TLNB (MARI) and ALND, the proposed algorithm would have led to omission of ALND in 74%, and some patients potentially risk under-(3%) or over-treatment (17%) [65]. Whether such tailored strategies might be an acceptable compromise between high oncological safety on one side and lower arm morbidity on the other remains to be clarified in future trials. In this context, the results of the NSABP-B51 trial (NCT01872975) are expected to be published in 2023. This phase III clinical trial is designed to test whether regional nodal irradiation (RNI) improves the recurrence-free interval rate in women with cN1 breast cancer before NACT who become pathologically node-negative at the time of surgery. Patients who undergo mastectomy are randomly assigned to observation or radiotherapy to the chest wall and undissected axilla, internal mammary nodes, and ipsilateral supraclavicular fossa, whereas women who undergo breast-conserving surgery are randomized to adjuvant whole breast irradiation vs. whole breast and regional nodal irradiation. In the ALLIANCE A011202 (NCT01901094) trial, women with node-positive status before NACT receive SLNB at the time of surgery. Patients with axillary residual disease are randomized to completion ALND and RNI vs. RNI alone.

The AXSANA Study: Which Axillary Strategy Is Optimal in the cN+ → ycN0 Setting?
AXSANA, initiated by EUBREAST (http://axsana.eubreast.com; accessed on 27 March 2021, Figure 1), is a large, prospective, non-interventional cohort study aiming to evaluate the role of axillary treatment in cN+ patients undergoing NACT. With a target accrual of 3000 patients, the study is expected to be able to resolve several open issues. Patients with clinically positive nodal status at time of diagnosis who are scheduled to receive axillary surgery after NACT can be enrolled. Axillary staging procedures and treatment modalities are chosen at the discretion of the treating physicians and according to national and institutional guidelines. Inclusion and exclusion criteria are presented in Table 8. Follow-up is annually during the first five years after surgery. Arm morbidity and quality of life are evaluated at baseline and after 1, 3, and 5 years, using four val-idated questionnaires (EORTC QLQ-C 30, EORTC QLQ BR 23, Lymph-ICF, and Sense of Coherence). Financial support has been provided by the AGO-B study group, the AWOgyn (German Working Group for Reconstructive Surgery in Oncology-Gynecology), Claudia von Schilling Foundation for Breast Cancer Research, EndoMag, Mammotome, and MeritMedical. AXSANA is further supported by the North-Eastern German Society of Gynecological Oncology (NOGGO) and the German Breast Group (GBG).  The first AXSANA study site was opened in Germany in June 2020, and recruitment began the same month. Currently, there are 328 patients enrolled. Twenty countries are participating in the study, most of which are in the process of applying for ethical approval. Ten countries have at least one study site open (Figure 2). The study will hopefully address several unanswered issues, such as:

•
Which staging technique should be recommended to cN+ patients converting to ycN0? • Is imaging helpful in identifying patients most likely to achieve pCR in the axilla? If yes, which method should be recommended? • Should cN1 and cN2/3 patients be offered different surgical strategies for axillary staging? • Which lymph node marking technique offers highest rates of successful TAD/TLNB? Due to high complexity and discordant recommendations, a randomized trial comparing different techniques seems hardly feasible and therefore would not clarify currently open issues within a reasonable timeframe. In the AXSANA study, patients are treated at physicians' discretion. To allow comparisons between different cohorts (SLNB, TLNB, TAD, ALND), detailed data regarding clinical and pathological parameters are obtained.
Other currently ongoing studies investigating axillary management in the neoadjuvant setting in cN+ patients are presented in Table 9. Abbreviations: ALND-axillary lymph node dissection, ART-axillary radiation therapy, DFS-disease-free survival, iDFS-invasive disease-free survival, TAD-targeted axillary dissection, TLN-target lymph node.

Conclusions
In view of continuously improving primary systemic treatments with increasing response rates, there is an urgent need to adapt and de-escalate strategies for axillary surgery, since it is strongly associated with postoperative morbidity. While SLNB has successfully replaced ALND as a staging procedure in primary surgery and after NACT for patients with an initial cN0 status, there is an ongoing debate on appropriate axillary staging for patients who convert from cN+ to ycN0.
This review revealed heterogeneous guideline recommendations and practice throughout the international community. This observation is explained by the lack of evidence concerning minimally invasive staging procedures like SLNB or TAD and their association with oncological outcomes, arm morbidity, and quality of life. The review also identified a multitude of unresolved issues regarding indications, surgical staging procedures, and technical aspects of lymph node marking. As a consequence, the European Breast Cancer Research Association of Surgical Trialists (EUBREAST) initiated a prospective cohort study that will allow the assessment of clinically relevant parameters in different axillary staging procedures after NACT, and of many open issues that have been highlighted here. AXSANA is open for all countries provided that patients receive treatment according to current international standards (http://axsana.eubreast.com, accessed on 27 March 2021). Funding: The funding of the AXSANA study has been described in detail in the manuscript.