The concept of moving systemic treatment from after surgery to before surgery was first developed from a hypothesis that the earlier disseminated single tumour cells were being killed the less likely was development of metastasis at a later stage. However, trials conducted in the 1980s and 1990s have not confirmed this concept but found a similar outcome of patients treated with chemotherapy before and after surgery. Thereafter, neoadjuvant chemotherapy was mainly used to obtain a better operability of large operable or even inoperable locally advanced tumours. Today, this approach is considered as a model to learn not only of the biology of breast cancer in general but also of the treatment sensitivity of an individual patient's tumour (Fig. 1). Tumour tissue obtained by core biopsy before treatment allows biological characterization of the tumour and development of an individual treatment plan. Early assessment after the first couple of cycles allows not only the testing for functional predictors but also the opportunity to change treatment according to early response. After completing all neoadjuvant treatment, histologic examination of the surgically removed breast and axillary tissue allows precise assessment of tumour response and characterization of residual, treatment-resistant disease.
These arguments for neoadjuvant therapy have led to its broader use in earlier disease with smaller tumours. However, adjuvant treatment remains the first option of care in most parts of the world. In the following, we will discuss recent findings that might support the use of neoadjuvant chemotherapy as a standard of care in some (mainly more aggressive) subtypes of breast cancer.
Prognostic impact of pathologic complete response
Recent results from a meta-analysis of seven German neoadjuvant studies of 6377 patients revealed that pathological complete response (pCR) is prognostic for disease-free and overall survival for more aggressive subtypes of breast cancer [1▪▪]. When defining subtypes using oestrogen and progesterone receptor status, human epidermal growth factor receptor 2 (HER2) status and tumour grade , patients with Luminal B (HER2-negative), HER2-positive (nonluminal) and triple-negative breast cancers, which achieved a pCR, showed a far better outcome than those without a pCR. However, in patients with low risk, hormone receptor positive, HER2-negative (Luminal A-like) breast cancers, prognosis was not different for patients with and without a pCR. Interestingly, also in patients with triple-positive (hormone receptor positive and HER2-positive) cancer, prognosis was not different with or without pCR. This has to be of course confirmed by other data sets. However, at least in the Luminal A-like tumours, pCR cannot be used as a surrogate marker for survival.
Comparing patients with pCR of the various subtypes, it could be shown that aggressive subtypes in which pCR showed a prognostic impact had a better outcome than less aggressive subtypes (Fig. 2) [1▪▪]. We can conclude that neoadjuvant treatment can be used as a kind of triage. Patients with more aggressive subtypes having achieved a pCR can be informed that their prognosis has turned from poor to highly favourable.
POTENTIAL NEW NEOADJUVANT PATHWAY FOR DRUG APPROVAL
The American Food and Drug Administration (FDA) is currently considering a new pathway for conditional approval of new agents in a neoadjuvant indication with pCR as an endpoint. A global meta-analysis project is currently being undertaken not only to reproduce the above-reported findings but also to show that a change in pCR rate by, for example, a better treatment, is directly associated with a change in survival time in the same direction. This is a prerequisite for confirming pCR as a valid surrogate endpoint for survival. The literature on this association is quite controversial, as many neoadjuvant chemotherapy trials have shown differences in pCR rates between various regimens, but this was predominantly not associated with a better survival [3,4], or even the opposite was found, that is no difference in pCR rates but improvements in survival .
PATHOLOGICAL COMPLETE RESPONSE AS A SURROGATE MARKER FOR SURVIVAL
However, this association might be detectable for patients with HER2-positive disease, when anti-HER2 targeted agents are given in addition to neoadjuvant chemotherapy. In the NOAH study [6▪▪], in which patients receiving neoadjuvant chemotherapy with or without trastuzumab were compared, it was demonstrated that the pCR rate could be increased by absolute 20% with the combined treatment. In addition, a significant improvement in event-free survival was reported. This trial therefore led to the approval of trastuzumab for neoadjuvant treatment by the European Medical Agency in early 2012 (Fig. 3a) [6▪▪,7,8▪,9]. Several trials  have tested the tyrosine kinase inhibitor lapatinib head-to-head with trastuzumab. In all trials, there was a trend towards a lower pCR rate for the combination of chemotherapy with lapatinib. The largest of these trials, the GeparQuinto study , even demonstrated a significant inferior pCR rate for patients receiving 24 weeks of lapatinib and anthracycline–taxane containing chemotherapy. These results are in line with the closure of one arm of the ALTTO trial (NCT00490139), which used lapatinib as the only adjuvant anti-HER2 treatment on the basis of the results of an early interim efficacy analysis (Fig. 3b) . Another pair of trials, in which a correlation of the effect of neoadjuvant treatment and the outcome of adjuvant treatment will become possible, is related to an antibody prohibiting the dimerization of the HER2 and the HER3 receptor, pertuzumab. The NeoSphere study [8▪] clearly demonstrated a higher pCR rate for patients receiving docetaxel chemotherapy with trastuzumab and pertuzumab. The Aphinity (BIG 4-11) (NCT01358877) study is currently enrolling an adjuvant trial to test the addition of pertuzumab to various chemotherapy–trastuzumab combinations (Fig. 3c) [8▪].
The other subtype in which a strong association is suspected between response to neoadjuvant therapy and overall outcome is triple-negative breast cancer. It is the subgroup with the largest difference in outcome when comparing patients with and without a pCR [1▪▪]. However, due to the high heterogeneity of the disease, it appears much more challenging to develop targeted agents that can add to the effect of chemotherapy. One compound that raised high interest due to promising results in pretreated metastatic breast cancer was iniparib . Initially considered to be a PARP inhibitor, it was combined with other DNA damaging agents such as carboplatin. However, a subsequent phase III study  failed to show significant results regarding the primary endpoints prompting a re-examination of the mode of action and revealing that in fact the compound was not inhibiting PARP. Only one neoadjuvant trial with iniparib was started at that time. Results on this NeoPARP study were published recently showing no change in pCR when iniparib was given at two different schedules together with paclitaxel to patients with early breast cancer . Therefore, it might be a better approach to target the host instead of this heterogenic subtype. In fact, the GeparQuinto trial showed a significant improvement of pCR when bevacizumab was added to chemotherapy in patients with HER2-negative breast cancer . Subgroup analysis suggested that this effect was derived mainly from patients with triple-negative breast cancer; pCR rates increased from 27.9% with chemotherapy alone to 39.3% with chemotherapy and bevacizumab . The sustainability of this result will be shown by the Beatrice study (NCT00528567) that examined various chemotherapy regimens with or without bevacizumab as an adjuvant treatment for patients with triple-negative breast cancer (Fig. 3d) . However, a first announcement by the sponsor claimed that the predefined endpoint invasive disease-free survival was not met (http://www.roche.com/hy12e02.pdf [Accessed 19 September 2012]).
ROLE OF THE EXTENT OF RESIDUAL DISEASE AFTER NEOADJUVANT CHEMOTHERAPY
Extent and characterization of residual disease after neoadjuvant chemotherapy raises increasingly more interest. The above-mentioned meta-analysis [1▪▪] showed that the extent of residual disease after neoadjuvant chemotherapy in the breast as well as in the nodes is associated with survival of patients. Worst prognosis with a median survival of around 5 years was observed for patients with posttreatment persisting inflammatory breast cancer, as well as large (>10) involvement of axillary nodes. Further biological characterization of residual disease was initiated by examining Ki-67 on the residual tumour. We determined Ki-67 on 1150 patients from the GeparTrio study for disease-free and overall survival . Patients were subdivided into four groups (pCR, Ki-67 0–15%, Ki-67 15.1–35%, Ki-67 >35%) having significantly different disease-free and overall survival prospects (Fig. 4) . In fact, posttreatment Ki-67 measurements were prognostically more relevant than pretreatment measurements or changes from before to after treatment. When analysing patients separately according to the hormone receptor status of their tumour, it became obvious that in hormone receptor positive disease, patients with low or moderate posttreatment Ki-67 levels had a comparable outcome with patients with a pCR, and only patients with high Ki-67 levels showed a very high risk of relapse. In hormone receptor negative disease, however, patients with a pCR did better than those with low or moderate Ki-67 levels, and again, patients with highly proliferating tumours had a most unfavourable outcome.
NEW CONCEPTS FOR POSTNEOADJUVANT SYSTEMIC TREATMENT
Developing treatment strategies for those patients with an unfavourable future outcome due to residual disease after neoadjuvant chemotherapy appears to be a major task for the future. The NATAN study has recruited (NCT00512993) over 800 patients with residual disease after anthracycline–taxane based chemotherapy and randomized them to 5 years treatment with zoledronic acid or observation. However, as 80% of the patients had hormone receptor positive disease and the patients were not otherwise selected for a high relapse risk, event rate was much lower than expected and a report of the results had to be postponed. Due to the unfavourable outcome of patients without a pCR after neoadjuvant treatment with chemotherapy and trastuzumab, patients with HER2-positive tumours might be better candidates for such a postneoadjuvant approach. A registration trial is currently under development, with the aim to compare postsurgical treatment with trastuzumab with postsurgical treatment with trastuzumab emtansine (T-DM1), a fusion molecule of trastuzumab and an antimicrotubule agent derived from maytansine.
Response-guided neoadjuvant chemotherapy
The concept of response-guided neoadjuvant chemotherapy was explored by the GeparTrio study [15,16]. Patients received two cycles of docetaxel, doxorubicin, cyclophosphamide (TAC). Response was assessed predominantly by sonography, and patients without an early response randomized to either four further cycles of TAC or four cycles of vinorelbine/capecitabine. Patients with an early response were considered sensitive to this type of chemotherapy and were randomized to a further four or six cycles of TAC. Analysis of pCR did not show any significant differences between the randomized groups. However, after a median of 5 years of follow-up, patients treated with the experimental treatments showed a better disease-free and overall survival than patients treated with conventional six cycles of TAC . An analysis by subgroup was performed to reveal why pCR did not predict these long-term results. It could be shown that the treatment effects were restricted to patients with hormone receptor positive tumours in which pCR was not prognostic (Fig. 5a) . However, in patients with HER2-positive (nonluminal) and triple-negative breast cancers, no difference between response-guided and conventional treatment was observed (Fig. 5b) . As these are the subgroups, wherein pCR shows predominantly its prognostic impact, the forecast of the surrogate marker was correct. It therefore appears crucial to assess the relationship between the prognostic impact of pCR and the treatment effect on pCR and survival separately by subtypes.
Furthermore, the response-guided approach should be further examined in patients with hormone receptor positive disease. However, the findings do not recommend aggressive chemotherapy to all kind of Luminal A tumours, as the GeparTrio study excluded patients with a low risk profile. Those patients included, despite a Luminal A-type feature of the tumour, had other clinical risk factors qualifying them for anthracycline–taxane based chemotherapy.
RISK OF LOCOREGIONAL RELAPSE AFTER NEOADJUVANT THERAPIES
The database on the German neoadjuvant studies was further analysed regarding locoregional tumour relapses , as early neoadjuvant trials showed a higher relapse risk for this approach. It was demonstrated that in general, patients with a pCR showed a lower locoregional relapse risk. These patients are therefore candidates to further reduce extent of locoregional treatment. It was further shown that patients treated by breast cancer conserving techniques had a lower locoregional relapse risk. This was also the case for patients with locally advanced tumours before the start of treatment. Even patients with inflammatory breast cancer showed a lower locoregional relapse risk when treated with breast cancer conservation than those treated with mastectomy (Fig. 1) [6▪▪,7,8▪,9]. As this is not a randomized comparison, it can only state that in case the investigator and the patient had chosen this nonstate-of-the-art approach (probably due to an extensive clinical response), it did not take patients at a high locoregional relapse risk.
The model of neoadjuvant chemotherapy has been improved over the last decade and is now successfully used to increase our knowledge on not only the pathophysiology of the disease but also on the activity of conventional and new treatment approaches. Due to this evolving evidence, a trend towards a more frequent use of this approach can be observed. Registration by authorities of new treatments in the neoadjuvant but as well in the postneoadjuvant setting further motivate to use this approach more frequently, as it will be the only option for patients to get early access to new and promising treatment options. The response-guided approach might be an option to further improve treatment, especially in hormone receptor positive tumours.
Conflicts of interest
Gv.M. has received funds and honoraria by GSK, Novartis, Roche and Sanofi-Aventis. M.U. and S.L. declared no conflicts of interests.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
▪ of special interest
▪▪ of outstanding interest
Additional references related to this topic can also be found in the Current World Literature section in this issue (pp. 88–89).
1▪▪. von Minckwitz G, Untch M, Blohmer JU, et al. Definition and impact of pathological complete response on prognosis after neoadjuvant chemotherapy in various intrinsic breast cancer subtypes. J Clin Oncol
A large meta-analysis exploring the impact of pCR in various breast cancer subtypes as well as which definition best separates patients with a poor and favourable outcome.
2. Goldhirsch A, Wood WC, Coates AS, et al. Strategies for subtypes: dealing with the diversity of breast cancer: highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer. Ann Oncol
3. Rastogi P, Anderson SJ, Bear HD, et al. Preoperative chemotherapy: updates of National Surgical Adjuvant Breast and Bowel Project Protocols B-18 and B-27. J Clin Oncol
4. Kaufmann M, Eiermann W, Schütte M, et al. Long term results form the neoadjuvant GeparDuo trial comparing a dose-dense regimen of adriamycin and docetaxel (ADoc) with a standard regimen of adriamycin /cyclophosphamide followed by docetaxel (AC–Doc) in patients (pts) with operable breast cancer (BC). J Clin Oncol
2010; 28 (15s):76s(Suppl; abstr 537).
5. von Minckwitz G, Blohmer JU, Costa SD, et al. Neoadjuvant chemotherapy adapted by interim response improves overall survival of primary breast cancer patients: results of the GeparTrio trial. Cancer Res
2011; 71 (Suppl 24):103s.
6▪▪. Gianni L, Eiermann W, Semiglazov V, et al. Neoadjuvant chemotherapy with trastuzumab followed by adjuvant trastuzumab versus neoadjuvant chemotherapy alone, in patients with HER2-positive locally advanced breast cancer (the NOAH trial): a randomised controlled superiority trial with a parallel HER2-negative cohort. Lancet
A landmark study leading to registration of trastuzumab for use in combination with neoadjuvant anthracycline–taxane based chemotherapy in Europe.
7. Untch M, Loibl S, Bischoff J, et al. Lapatinib versus trastuzumab in combination with neoadjuvant anthracycline-taxane-based chemotherapy (GeparQuinto, GBG 44): a randomised phase 3 trial. Lancet Oncol
8▪. Gianni L, Pienkowski T, Im YH, et al. Efficacy and safety of neoadjuvant pertuzumab and trastuzumab in women with locally advanced, inflammatory, or early HER2-positive breast cancer (NeoSphere): a randomised multicentre, open-label, phase 2 trial. Lancet Oncol
The first study using a double HER2 blockage together with a full-dose anthracycline–taxane based neoadjuvant chemotherapy.
9. von Minckwitz G, Eidtmann H, Rezai M, et al. Neoadjuvant chemotherapy and bevacizumab for HER2-negative breast cancer. N Engl J Med
10. O'Shaughnessy J, Osborne C, Pippen JE, et al. Iniparib plus chemotherapy in metastatic triple-negative breast cancer. N Engl J Med
11. O'Shaughnessy J, Schwartzberg MA, Danso HS, et al.
A randomized phase III study of iniparib (BSI-201) in combination with gemcitabine/carboplatin (G/C) in metastatic triple-negative breast cancer (TNBC). J Clin Oncol 2011; 29(Suppl):abstr 1007.
12. Llombard A, Lluch A, Villanueva C, et al.
SOLTI NeoPARP: a phase II, randomized study of two schedules of iniparib plus paclitaxel and paclitaxel alone as neoadjuvant therapy in patients with triple-negative breast cancer (TNBC). J Clin Oncol 2012; 30(Suppl):abstr 1011.
13. Gerber B, Eidtman H, Rezai M, et al.
Neoadjuvant bevacizumab and anthracycline–taxane-based chemotherapry in 686 triple-negative primary breast cancers: secondary endpoint analysis of the GeparQuinto study (GBG 44). J Clin Oncol 2011; 29(Suppl):abstr 1006.
14. von Minckwitz G, Müller B, Blohmer JU, et al.
Prognostic and predictive impact of Ki-67 before and after neoadjuvant chemotherapy on PCR and survival: results of the GeparTrio trial. J Clin Oncol 2012; 30(Suppl):abstr 1023.
15. von Minckwitz G, Kummel S, Vogel P, et al. Intensified neoadjuvant chemotherapy in early-responding breast cancer: phase III randomized GeparTrio study. J Natl Cancer Inst
16. von Minckwitz G, Kümmel S, Vogel P, et al. Neoadjuvant vinorelbine-capecitabine versus docetaxel-doxorubicin-cyclophosphamide in early nonresponsive breast cancer: phase III randomized GeparTrio trial. J Natl Cancer Inst
17. von Minckwitz G, Kaufmann M, Kümmel S, et al. Local recurrence risk after neoadjuvant chemotherapy. Pooled analysis on 5477 breast cancer patients. Cancer Res
2011; 71 (Suppl 24):142s.