Abstract

Neoadjuvant chemotherapy has several appealing potential benefits compared with classic adjuvant chemotherapy. Of these, the only proven benefit is to facilitate the surgical approach, either by converting an inoperable cancer to one that is operable, or by converting a patient who is felt to be a candidate for mastectomy to one who might be treated successfully with breast conserving therapy. Randomized trials comparing neoadjuvant chemotherapy with postoperative chemotherapy have failed to demonstrate prolongation of overall survival. The benefits of monitoring apparent response during neoadjuvant chemotherapy have not been proven. Conduct of phase II drug development trials in the neoadjuvant setting may be advantageous compared with performing such trials in the metastatic setting. However, such trials raise concerns that are not unavoidable but need to be addressed.

Potential Benefits From Neoadjuvant Chemotherapy

Multiple prospective randomized clinical trials involving thousands of women have clearly demonstrated that early systemic treatment of micrometastatic breast cancer improves survival (1,2). These data led several investigators to assess whether adjuvant systemic therapy might be even more effective if delivered before surgery.

There are several hypothetical reasons that preoperative chemotherapy (“neo-adjuvant chemotherapy,” NACT) might be even more beneficial than classic, postsurgical adjuvant chemotherapy (ACT) (Table 1). First, NACT might be expected to result in better overall survival (OS) than postsurgical ACT. The presumed mechanism of improved OS for classic ACT is avoidance of development of genetic heterogeneity, and associated resistance, with progression of the cancer. Thus, one might presume that avoiding the postoperative delay in initiation of therapy might even further improve OS. Alternatively, surgery itself might promote local or distant relapse due to release of massive amounts of wound-healing cytokines, and therefore NACT might serve to eliminate micrometastatic deposits before exposure to this cytokine storm.

Table 1.

Possible benefits of neoadjuvant chemotherapy and neoadjuvant trials

Potential benefit Proven with high levels of evidence? 
Prolongation of overall survival No 
Optimize surgical approach Yes 
Modify chemotherapeutic regimen No 
Add further chemotherapy No 
Phase II testing of new drugs, new regimens, or compare regimens Partially 
Identify new biomarkers of response or toxicity No 
Potential benefit Proven with high levels of evidence? 
Prolongation of overall survival No 
Optimize surgical approach Yes 
Modify chemotherapeutic regimen No 
Add further chemotherapy No 
Phase II testing of new drugs, new regimens, or compare regimens Partially 
Identify new biomarkers of response or toxicity No 

Second, one might apply NACT to improve and/or optimize the surgical approach. For example, patients with locally advanced “inoperable” cancer might become surgical candidates. Likewise, NACT might convert patients who were initially felt to require mastectomy because of unacceptable tumor size to candidates for breast conserving surgery.

Third, one might monitor apparent response, or lack of it, to adjust the specific regimen. This approach might be taken early during application of one regimen to direct a change to a different regimen, or one might use failure to achieve an optimal response (such as pathological complete response [pCR]) to elect to administer additional chemo- or other types of therapies.

Finally, the preoperative setting might be preferable to the metastatic setting to perform phase II trials or to identify biomarkers that might permit individualization of the therapy. Prospective randomized clinical trials in the postoperative setting require thousands of patients and often nearly a decade to complete. Classically, phase II trials are performed in the metastatic setting, which may be less than optimal compared with conduct of such trials in early stage disease. Patients with metastatic disease may be more likely to have emergent resistance and may be more susceptible to toxicities related to heavy pretreatment and tumor progression. Moreover, in the metastatic setting, determination of response as an endpoint may be more difficult due to lack of measurable disease.

Data Regarding Each of These Potential Reasons

Improvement in OS

Several prospective randomized trials that compared NACT versus classic postoperative ACT were conducted in the 1970s and 1980s. These trials were conducted in patients with locally advanced breast cancer or in patients with potentially operable breast cancer (3). The trial designs differ widely, with some being confounded by adaptive approaches to the patient based on apparent response or not. However, in the purist, and largest, trial (the National Surgical Adjuvant Breast and Bowel Project [NSABP] B18), patients were randomly assigned to neoadjuvant versus classical adjuvant doxorubicin and cyclophosphamide (4). These investigators failed to observe any hint of improvement in OS in this large and properly powered trial (Figure 1) (4). In general, the results of other studies, both in patients with operable and inoperable cancers, support these observations (5).

Facilitation of Surgical Approach

In his original papers, Halsted (6) noted that the OS of patients who had a series of clinical features, namely peau d’orange, inflammatory changes, fixation to the chest wall, or skin nodules or ulceration, was so poor that surgery was contraindicated. These “five grave findings” were subsequently encoded as reasons for “inoperability” by Haagensen (7). With the apparent success of inducing responses in the metastatic setting and in reducing mortality by administration of classical ACT, several groups began to investigate the use of preoperative chemotherapy to “down-stage” locally advanced breast cancers and make these patients “operable.” Indeed, these studies have demonstrated that approximately 75%–90% of patients experience sufficiently favorable clinical responses that their surgeons feel comfortable proceeding with mastectomies that would have, in the past, been considered contraindicated (5).

In the 1980s, results from several prospective randomized clinical trials demonstrated that, for patients with primary tumors that could be assessed, breast conserving therapy (BCT) with adjuvant radiation was just as effective in regards to long term OS as mastectomy (8,9). However, only approximately 60% of patients with stage I–III disease are considered candidates for BCT. There are several absolute and relative indications for mastectomy, most of which involve the concern that the risk of in-breast recurrence is unacceptably high. Although the latter criterion is relatively arbitrary, most experts recommend mastectomy if this risk is greater than 10%–15% over the succeeding 5–10 years.

Regardless, the main contraindications to BCT include multifocal lesions, widespread microcalcifications related to in situ disease, or such a large tumor size that complete excision with negative margins results in unacceptable cosmetic outcome or, frankly, requires a mastectomy. Although NACT will not change the indication for mastectomy for patients with either of the first two, it is possible that it might decrease tumor size and permit wide excision with acceptable cosmetics. Indeed, in NSABP B-18, surgeons were asked to provide their assessment of appropriate operative approach before the patient’s random assignment. If the patient was assigned to surgery first, that approach was taken, while if the patient was assigned to NACT, the surgeon was permitted to change surgical strategy if it was felt that such a change was appropriate. Reflecting responses in the breast, approximately 15% fewer mastectomies were performed in the NACT (4). It should be pointed out that the in-breast recurrence rate in this arm is slightly, although not statistically, increased in the NACT arm, raising concerns that perhaps the tumors do not shrink concentrically (4). Nonetheless, this indication is an acceptable reason to recommend NACT, as opposed to initial surgery followed by classic ACT, especially for patients who are felt to be likely to respond to the chemotherapy and who do not have any of the other absolute contraindications to BCT. However, the number of women who benefit from this approach will be small, since many patients have other contraindications to BCT in addition to tumor size, or have tumors that do not respond sufficiently to NACT to permit BCT.

Use of Response to Adjust Therapeutic Regimen

In the classic adjuvant setting, one does not have any method of monitoring the success of the therapy, and therefore patients are treated for a full course of the chosen regimen unless toxicities prevent doing so. Determination of whether a chosen regimen is effective or futile might permit either “in-course” adjustment or administration of additional therapy after the initial regimen is complete.

In-course Adjustment

The concept of changing from an ineffective therapy to an effective therapy is of course quite appealing. However, in-course adjustment may be inappropriate for two reasons: 1) Failure to document an apparent response during treatment does not mean that it is ineffective against systemic micrometastases; and 2) Clinical and/or radiographic evaluation of response is insufficiently accurate.

Changing therapy in mid-course may result in under-treatment. Prospective randomized clinical trials in the standard adjuvant setting have suggested that administration of four to eight cycles of chemotherapy is optimal compared with shorter durations (10). Therefore, if the indication of lack of response is not absolutely accurate, one might be discontinuing therapy prematurely. Although several trials have demonstrated that a pCR at the end of therapy is associated with more favorable prognosis in some patients (11), neither they nor any other study has suggested that a lack of apparent response in the breast indicates that the chemotherapy is not active against distant metastases (12,13).

In this regard, efforts to determine ongoing response by physical examination or radiographic evaluation, even using highly sensitive measures such as magnetic resonance imaging, have been fraught with low sensitivity and specificity for response (14,15). Investigators have reported that use of serial measures of proliferation by Ki67 immunohistochemistry may accurately reflect response to preoperative endocrine therapy (16,17). However, these results have not been translated to NACT, and the reproducibility of Ki67 by immunohistochemistry is insufficient to use it to make routine clinical decisions in the community (18,19).

Additional Treatment After Standard Therapy

Since patients who fail to achieve pCR have a worse prognosis, in theory such patients might benefit from additional therapy, chemo or otherwise (11). There are two concerns inherent in this theory. First, although one is inclined to offer further therapy to patients with a bad prognosis, these patients may have fundamentally chemoresistant disease, and more chemotherapy may not be of any additional value. Furthermore, although those with pCR do have a better prognosis, the cure rate for such patients is far from 100%. It is possible that these patients, with relatively chemosensitive disease, are actually the patients who might benefit from further therapy.

Stated in a different way, the presence of a poorer prognosis is not an indication for further treatment. Rather, one must couple evidence of poor prognosis with equally important evidence that further chemotherapy will improve it. No trial has demonstrated that patients who fail to achieve pCR will benefit from alternative, additional therapy, nor are there any trials that have assessed the benefit of additional therapy for those who experience a pCR but still have a high risk of subsequent recurrence. In this regard, the results of NSABP B-27 did demonstrate that addition of a taxane after neoadjuvant doxorubicin/cyclophosphamide resulted in superior pCR rates, but only marginal improvement in disease-free survival. This trial was under-powered to determine if the benefit was equally distributed across the entire study group or was confined to those who did not have a pCR, or for that matter, to those who did (20,21).

Rather than giving more cytotoxic therapy, it is more likely that addition of targeted therapy will be beneficial. Indeed, patients with estrogen receptor-rich tumors clearly benefit from subsequent adjuvant endocrine therapy. Similar results have been demonstrated for anti-HER2 therapies for patients with HER2—rich cancers. However, these benefits are accrued regardless of response to prior chemotherapy. It would be of great value to identify those patients whose residual risk after NACT is so favorable that, even if subsequent therapy is active, it would not be needed. However, such data do not exist.

In summary, although patients who experience pCR have a better prognosis, they are still at substantial risk for recurrence and death, and they might be more likely to benefit from further administration from more of the same or a different type of chemotherapy. However, this theory is unproven and deserves further study. In the meantime, one cannot invoke the use of apparent lack of benefit during or after NACT as a reason to alter or expand the ACT regimen.

Use of NACT as a Phase II/Drug Development Platform

As noted, early stage breast cancer might be a more advantageous setting to test activity of new drugs or novel mechanisms of delivery of previously established chemotherapeutic agents, or to compare different regimens. Using the NACT setting must be considered “phase II” in nature, using pCR as a surrogate of more clinically meaningful endpoints, such as disease-free survival or OS. Indeed, the United States Food and Drug Administration (FDA) has recently announced that accelerated approval for a novel agent might be awarded based on improvement of pCR in the neoadjuvant setting (http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM305501.pdf). However, inherent in this consideration is that the trial must be sufficiently powered to provide meaningful data regarding the more clinically relevant endpoints of disease-free survival (DFS) and OS, and that crossover cannot be permitted.

The design and conduct of such trials raise several concerns that need to be addressed. First, one must be cognizant that the majority of patients undergoing NACT will be cured by standard primary surgery and radiation and adjuvant systemic therapy. Therefore, one must be wary of severe toxicities, even if unusual, related to new drugs that will, by definition, only have undergone evaluation in small phase I trials, usually in patients with limited lifespans. Applying such relatively untested agents in a potentially curable population, as opposed to the metastatic population, is worrisome.

For the same reason, one must be certain that the regimen the patient will receive is not inferior to standard ACT. Thus, the control arm needs to contain proven, standard therapy, either before or after the investigational part of the trial. This concern raises several subsequent issues. For example, one might design a trial of a new drug versus placebo for some period of time to assess response independent of other therapies. Some investigators have recommended so-called “window of opportunity” studies, administering only a short trial of therapy (1–2 weeks) between the diagnostic and therapeutic biopsies, and using surrogate biomarker changes (such as reduction of Ki67) as an indication of benefit (18). However, it has not been demonstrated that such an approach is satisfactory to determine activity of novel agents, and the success rate of accrual to these window of opportunity studies has been, at times, difficult (22).

It is unethical to withhold standard adjuvant systemic therapy for the several months one needs to determine clinical or radiographic (as opposed to molecular) response to a single agent phase II drug. Therefore, investigators have designed trials that incorporate randomized phase II designs, in which all patients receive standard chemotherapy and are randomly assigned to concurrent new drug or not (23). For agents that are synergistic or additive to chemotherapy, this trial design is satisfactory. However, if the new drug is antagonistic with chemotherapy, the patient is disadvantaged by having participated in the trial.

A third concern to the neoadjuvant Phase II trial approach is whether the drug in question will induce the desired endpoint (such as pCR), even if it is active. For example, neoadjuvant endocrine treatment, such as tamoxifen or the aromatase inhibitors, rarely induces pCR, and yet it is one of the most effective strategies in all of cancer.

Indeed, hypothetically, tamoxifen provides a case study of incorporation of new drugs into phase II trial designs in the standard neoadjuvant setting. If it were to be given simultaneously with chemotherapy, its proven antagonistic effect with chemotherapy (24) would be damaging to patients. Moreover, its lack of induction of pCR would result in its being considered a less than active drug. Thus, if tamoxifen were studied in a neoadjuvant trial on a backbone of chemotherapy, both the patient and the drug would be harmed.

Biomarker Testing

A final potential benefit of the neoadjuvant setting might be to discover important biomarkers that could be used in larger, definitive adjuvant trials, or even in standard clinical settings, to personalize patient care. Several investigators have reported that chemotherapy may be more or less effective in different intrinsic subtypes, and these observations have been confirmed in the neoadjuvant setting (11,25). Unfortunately, thus far, no important biomarkers have been identified in neoadjuvant trials for targeted therapies, such as anti-HER2 agents or bevacizumab. Such studies are incorporated into ongoing trials, raising hope that future studies and treatments can be give more efficiently.

Conclusion

In conclusion, there are several appealing potential reasons to recommend NACT. However, only optimization of the planned surgical approach has been demonstrated with high levels of clinical utility. No evidence suggests that NACT, compared with standard postoperative ACT, improves OS. Although NACT might provide an opportunity to take an adaptive approach during or after chemotherapy, the data to support doing so are nearly nonexistent. Using the neoadjuvant setting to test new drugs is seductive, but should be carefully considered to be certain that the patient is not harmed due to unexpected toxicities or antagonism with standard, life-saving therapy, and that potentially active drugs are not discarded due to false negative observations as a result of inappropriate endpoints.

Funding

Supported in part by Fashion Footwear Charitable Foundation of New York/QVC Presents Shoes on Sale™ (DFH).

References

1.
Early Breast Cancer Trialists’ Collaborative Group
.
Adjuvant chemotherapy in oestrogen-receptor-poor breast cancer: patient-level meta-analysis of randomised trials
.
Lancet
 .
2008
;
371
(9606)
:
29
40
.
2.
Early Breast Cancer Trialists’ Collaborative Group T
.
Comparisons between different polychemotherapy regimens for early breast cancer: meta-analyses of long-term outcome among 100,000 women in 123 randomized trials
.
Lancet
 .
2012
;
379
(9814)
:
432
444
.
3.
Legler
CM
Shapiro
CL
Harris
JR
et al
Primary chemotherapy of resectable breast cancer
.
Breast J
 .
1995
;
1
(
1
):
42
51
.
4.
Fisher
B
Bryant
J
Wolmark
N
et al
Effect of preoperative chemotherapy on the outcome of women with operable breast cancer
.
J Clin Oncol
 .
1998
;
16
(
8
):
2672
2685
.
5.
Redden
MH
Fuhrman
GM
.
Neoadjuvant chemotherapy in the treatment of breast cancer
.
Surg Clin North Am
 .
2013
;
93
(
2
):
493
499
.
6.
Halsted
WS. I
.
The Results of Operations for the Cure of Cancer of the Breast Performed at the Johns Hopkins Hospital from June, 1889, to January, 1894
.
Ann Surg
 .
1894
;
20
(
5
):
497
555
.
7.
Haagensen
CD
.
Diseases of the Breast
 .
2
nd ed.
Philadelphia, PA
:
W.B. Saunders
;
1971
.
8.
Fisher
B
Anderson
S
Bryant
J
et al
Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer
.
N Engl J Med
 .
2002
;
347
(
16
):
1233
1241
.
9.
Veronesi
U
Cascinelli
N
Mariani
L
et al
Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer
.
N Engl J Med
 .
2002
;
347
(
16
):
1227
1232
.
10.
Early Breast Cancer Trialists’ Collaborative Group
.
Polychemotherapy for early breast cancer: an overview of the randomised trials
.
Lancet
 .
1998
;
352
(
9132
):
930
42
.
11.
von Minckwitz
G
Untch
M
Blohmer
JU
et al
Definition and impact of pathologic complete response on prognosis after neoadjuvant chemotherapy in various intrinsic breast cancer subtypes
.
J Clin Oncol
 .
2012
;
30
(
15
):
1796
1804
.
12.
von Minckwitz
G
.
Neoadjuvant chemotherapy in breast cancer-insights from the German experience
.
Breast Cancer
 .
2012
;
19
(
4
):
282
288
.
13.
von Minckwitz
G
Untch
M
Loibl
S
.
Update on neoadjuvant/preoperative therapy of breast cancer: experiences from the German Breast Group
.
Curr Opin Obstet Gynecol
 .
2013
;
25
(
1
):
66
73
.
14.
Schott
AF
Roubidoux
MA
Helvie
MA
et al
Clinical and radiologic assessments to predict breast cancer pathologic complete response to neoadjuvant chemotherapy
.
Breast Cancer Res Treat
 .
2005
;
92
(
3
):
231
238
.
15.
Marinovich
ML
Houssami
N
Macaskill
P
et al
Meta-analysis of magnetic resonance imaging in detecting residual breast cancer after neoadjuvant therapy
.
J Natl Cancer Inst
 .
2013
;
105
(
5
):
321
333
.
16.
Dowsett
M
Smith
IE
Ebbs
SR
et al
Prognostic value of Ki67 expression after short-term presurgical endocrine therapy for primary breast cancer
.
J Natl Cancer Inst
 .
2007
;
99
(
2
):
167
170
.
17.
Ellis
MJ
Tao
Y
Luo
J
et al
Outcome prediction for estrogen receptor-positive breast cancer based on postneoadjuvant endocrine therapy tumor characteristics
.
J Natl Cancer Inst
 .
2008
;
100
(
19
):
1380
1388
.
18.
Dowsett
M
Nielsen
TO
A’Hern
R
et al
Assessment of Ki67 in breast cancer: recommendations from the International Ki67 in Breast Cancer working group
.
J Natl Cancer Inst
 .
2011
;
103
(
22
):
1656
1664
.
19.
Polley
MY
Leung
SC
McShane
LM
et al.  
An international ki67 reproducibility study
.
J Natl Cancer Inst
 
2013
;
105
(24)
:
1897
906
.
20.
Bear
HD
Anderson
S
Brown
A
et al
The effect on tumor response of adding sequential preoperative docetaxel to preoperative doxorubicin and cyclophosphamide: preliminary results from National Surgical Adjuvant Breast and Bowel Project Protocol B-27
.
J Clin Oncol
 .
2003
;
21
(
22
):
4165
4174
.
21.
Bear
HD
Anderson
S
Smith
RE
et al
Sequential preoperative or postoperative docetaxel added to preoperative doxorubicin plus cyclophosphamide for operable breast cancer: National Surgical Adjuvant Breast and Bowel Project Protocol B-27
.
J Clin Oncol
 .
2006
;
24
(
13
):
2019
2027
.
22.
Stearns
V
Coop
A
Singh
B
et al
A pilot surrogate end point biomarker trial of perillyl alcohol in breast neoplasia
.
Clin Cancer Res
 .
2004
;
10
(
22
):
7583
7591
.
23.
Barker
AD
Sigman
CC
Kelloff
GJ
et al
I-SPY 2: an adaptive breast cancer trial design in the setting of neoadjuvant chemotherapy
.
Clin Pharmacol Ther
 .
2009
;
86
(
1
):
97
100
.
24.
Albain
KS
Barlow
WE
Ravdin
PM
et al
Adjuvant chemotherapy and timing of tamoxifen in postmenopausal patients with endocrine-responsive, node-positive breast cancer: a phase 3, open-label, randomised controlled trial
.
Lancet
 .
2009
;
374
(
9707
):
2055
2063
.
25.
Hayes
DF
.
Targeting adjuvant chemotherapy: a good idea that needs to be proven!
J Clin Oncol
 .
2012
;
30
(
12
):
1264
1267
.