Factors associated with surgical management following neoadjuvant therapy in patients with primary HER2-positive breast cancer: results from the NeoALTTO phase III trial

Factors associated with surgical management followingneoadjuvant therapy in patients with primary HER2-positive breast cancer: results from the NeoALTTOphase III trialC. Criscitiello1,†, H. A. Azim Jr2,†*, D. Agbor-tarh3, E. de Azambuja2, M. Piccart2, J. Baselga4,H. Eidtmann5, S. Di Cosimo6, I. Bradbury3 & I. T. Rubio71Division of Medical Oncology, Department of Medicine, European Institute of Oncology, Milan, Italy; 2Department of Medicine, BrEAST Data Centre, Institut JulesBordet, Université Libre de Bruxelles, Brussels, Belgium; 3Frontier Science, Kincraig, Kingussie, Scotland, UK; 4Massachusetts General Hospital, Harvard MedicalSchool, Boston, USA; 5University Hospital Kiel, Kiel, Germany; 6Department of Medical Oncology, Istituto Nazionale Tumori, Milan, Italy; 7Breast Surgical Oncology Unit,Universitario Vall D’Hebron, Barcelona, Spain

Received 17 December 2012; revised 30 January 2013; accepted 18 February 2013

Background: The NeoALTTO trial showed that dual HER2 blockade nearly doubles the rate of pathologic completeresponse (pCR) in patients with primary HER2-positive breast cancer. However, this did not translate into a higher rateof breast-conserving surgery (BCS).Patients and methods: In NeoALTTO, patients with HER2-positive breast cancer were randomly assigned to eithertrastuzumab, lapatinib or their combination with paclitaxel before surgery with pCR as the primary end point. Weinvestigated the association between the surgery type and clinicopathological factors and response to treatment,adjusting for the treatment arm.Results: Four hundred and twenty-nine patients were subjected to breast surgery. Two hundred and forty-two (56%)and 187 (44%) patients underwent mastectomy and BCS, respectively. In a logistic regression model, negativeestrogen receptor (ER), multicentricity and the presence of a palpable mass before surgery were significantly associatedwith a low chance of BCS. Conversely, patients with small tumors and those eligible for BCS at diagnosis weremanaged more with BCS, independent of the treatment arm. Radiological response was not associated with thesurgical decision.Conclusions: Tumor characteristics before neoadjuvant therapy play a main role in deciding the type of surgery callingfor a clear consensus on the role of BCS in patients responding to neoadjuvant therapy.Key words: breast cancer, neoadjuvant, HER2, surgery, breast conservation, trastuzumab

introductionData from randomized trials have shown no difference insurvival between breast cancer patients receiving adjuvant orneoadjuvant primary systemic therapy [1, 2]. While to date theconventional sequence is to administer treatment in thepostoperative setting, the neoadjuvant approach has beenendorsed by several groups and experts for a wide variety ofreasons [3, 4]. Down-staging of the tumor resulting in anincrease in breast conservation rates remains one of the mostimportant advantages of the neoadjuvant approach [5].However, even in patients who are eligible for breast

conservation at diagnosis, the neoadjuvant setting can still offerpotential advantages, such as determining the sensitivity of thetumor to systemic therapy and possible elimination ofmicrometastases [6].Pathological complete response (pCR) to chemotherapy has

been identified as a reliable surrogate marker for long-termsurvival [7–9], particularly in patients with HER2 amplification[9]. The addition of the HER2-targeted agent, trastuzumab, tochemotherapy has been shown to be associated with a higherrate of pCR compared with chemotherapy alone [10–12]. Thishigh pCR rate further translated into an improvement indisease-free survival [10]. This has resulted in using theneoadjuvant setting to gain relatively fast insights into theefficacy of HER2-targeted agents in the primary setting.Recently, four trials namely the NeoAdjuvant Lapatinib and/

or Trastuzumab Treatment Optimization (NeoALTTO) [13],the neoadjuvant study of pertuzumab and herceptin in an early†Co-first authors.

*Correspondence to: Dr H. A. Azim Jr., Department of Medicine, BrEAST Data Centre,Institut Jules Bordet. Tel: +32-2-541-3854; Fax: +32-2-541-3477; E-mail: [email protected]

original

articles

original articles Annals of Oncology 24: 1980–1985, 2013doi:10.1093/annonc/mdt129Published online 7 April 2013

© The Author 2013. Published by Oxford University Press on behalf of the European Society for Medical Oncology.All rights reserved. For permissions, please email: [email protected].

at Grand V

alley State Univ on N

ovember 1, 2013

http://annonc.oxfordjournals.org/D

ownloaded from

at G

rand Valley State U

niv on Novem

ber 1, 2013http://annonc.oxfordjournals.org/

Dow

nloaded from

at Grand V


ovember 1, 2013


ownloaded from

at G

rand Valley State U

niv on Novem


Dow

nloaded from

at Grand V


ovember 1, 2013


ownloaded from

at G

rand Valley State U

niv on Novem


Dow

nloaded from

http://annonc.oxfordjournals.org/






regimen evaluation [14], the National Surgical Adjuvant Breastand Bowel Project-B41 [15] and the trastuzumab pluspertuzumab in neoadjuvant HER2-positive breast cancer(TRYPHAENA) [16] reported on the value of dual HER2blockade in primary HER2-positive breast cancer. Thesestudies consistently showed that the addition of dual HER2blockade to chemotherapy significantly improves pCR ratescompared with the same regimen in combination withtrastuzumab alone.In the NeoALTTO trial [13], although dual HER2 blockade

with lapatinib and trastuzumab in combination with weeklypaclitaxel nearly doubled the rate of pCR, compared withpaclitaxel combined with either drug alone, this high pCR ratedid not translate into a higher rate of breast-conserving surgery(BCS), which was around 40% across the three treatment arms.This observation was rather surprising. Hence, we conductedthis sub-study to investigate the different factors that may haveaffected the choice of surgical modality in patients enrolled inthe NeoALTTO trial.

patients and methodsNeoALTTO is a randomized, phase III, multicenter, international trial inwhich patients with primary HER2-positive breast cancer, and tumor size>2 cm were randomizly assigned to either lapatinib, trastuzumab or theircombination for 6 weeks followed by the addition of paclitaxel for 12 moreweeks before breast surgery [13]. All patients were evaluated according towhether or not they were eligible for BCS before initiation of neoadjuvanttherapy (will be referred to as ‘planned surgery at diagnosis’). All patientswere prospectively subjected to clinical examinations at baseline, weeks 2, 4,6 and weekly afterwards until surgery. As per protocol, all patients werealso subjected to either mammography or echography at baseline, week 6and after completion of neoadjuvant therapy (i.e. before surgicalintervention). This information was prospectively collected on the casereport form.

Following surgery, all patients received three cycles of 5-fluorouracil,epirubicin 100 mg/m2 and cyclophosphamide (i.e. FEC100) followed by thecontinuation of the same HER2-targeted therapy that was administered inthe neoadjuvant setting to complete a total duration of 1 year. Further

information on the trial including eligibility criteria, statistical analysis andthe results is described in detail elsewhere [13].

The primary end point of the main study was the rate of pCR defined asthe absence of invasive tumor cells in the breast at the time of surgery [17].The ethics committees of all participating sites approved the study and aninformed consent was obtained from all patients at study entry. This sub-study was approved by the NeoALTTO executive committee.

The randomization was stratified according to four factors: clinicaltumor size (T2 versus ≥T3), clinical nodal status (N0-1 versus ≥N2),estrogen receptor (ER) and/or progesterone receptor (PgR) (positive versusnegative) and whether a patient was a candidate for BCS (yes versus no).The main analysis indicated that all factors were evenly distributed acrossthe three treatment arms, except for the tumor size, in which more ≥T3tumors were randomly assigned to the trastuzumab only arm (53%)compared with the lapatinib only arm (38.3%) and the combination arm(35.5%).

In the current analysis, we investigated the association between the typeof surgery and the different clinicopathological factors, which were asfollows: age, histological subtype, histological grade, clinical tumor size, ERstatus and tumor multicentricity/multifocality. This is in addition to theresponse to treatment (both radiological and clinical) and the country

where the treatment was administered. Countries were classified asdeveloped or developing; the former being defined as high-incomecountries as per the classification of the Organization for Economic Co-operation and Development (OECD) (www.oecd.org).

We evaluated the association between the type of actual surgerycarried out and the clinicopathological factors, and preoperative responseto treatment using Mann–Whitney or Kruskal–Wallis, as appropriate. Alogistic regression model relating the probability of BCS to the candidatepredictors and the treatment arms was carried out in order to identifywhether any of the clinicopathological factors was associated with BCSand whether adjustment for these predictors changes the apparentrelationship between the treatment arm and BCS. We fitted the modelincluding all factors with and without the type of planned surgery atdiagnosis, adjusting for the treatment arm. Finally, we evaluated thefactors that could be related to the type of planned surgery at diagnosis.All the statistical analyses were carried out by Frontier Science(Scotland).

resultsA total of 455 patients in 23 countries were enrolled in theNeoALTTO trial. Of those, 26 patients did not undergo breastsurgery and hence, were excluded from this analysis.Table 1 shows the association and distribution of the

different clinicopathological factors, response to treatment(clinical and radiological) and the type of actual surgerycarried out. BCS was less common in patients treated indeveloping countries (P < 0.0001), initially planned formastectomy at the time of diagnosis (P < 0.0001), hadmultifocal/multicentric tumors (P = 0.007), unknownhistological grade (P = 0.02), ER-negative tumors (P = 0.005),tumors >5 cm before initiation of neoadjuvant therapy(P < 0.0001) and a residual palpable mass at the pre-surgeryvisit (P = 0.0001). No association between the type of surgeryand age (P = 0.38), histological subtype (P = 0.95), and pre-operative radiological response (P = 0.86) was observed.Strikingly, out of 68 and 199 patients showing a complete andpartial radiological response before surgery, 52.9% and 55.3%were subjected to mastectomy, respectively.Table 2 shows the logistic regression model taking into

account all possible confounding factors, including the type ofplanned surgery at diagnosis. As expected, we found thatpatients who were initially planned for BCS were nearly fivetimes more likely to undergo BCS compared with patients whowere initially planned for mastectomy or were consideredineligible for resection at diagnosis [OR: 5.46 (95% CI, 3.17–9.41), P < 0.0001]. Moreover, patients who had a tumor >5 cmat baseline were more likely to be offered mastectomy [OR:1.97 (95% CI, 1.20–3.23), P = 0.0076]. On the other hand,patients with multicentric/multifocal tumors [OR: 0.21 (95%CI, 0.09–0.48), P = 0.0002], ER-negative tumors [OR: 0.50(95% CI, 0.31–0.81), P = 0.005] and those with a palpable massbefore surgery [OR: 0.41 (95% CI, 0.23–0.71), P = 0.0069] wereless likely managed with BCS. A trend of a higher rate of BCSwas observed in patients from developed countries [OR: 1.62(95% CI, 0.98–2.67), P = 0.059]. We did not observe any effectof age, histological grade, radiological response or completionof therapy on the decision of surgery. These associations wereobserved independent of the treatment arm (P = 0.41).

Annals of Oncology original articles

Volume 24 | No. 8 | August 2013 doi:10.1093/annonc/mdt129 |

www.oecd.org

In order to understand the effect of the individual factorsindependent of the type of planned surgery before startingneoadjuvant therapy, we repeated the same model excludingplanned surgery (Table 3). In addition to the formerlyobserved associations, we found that patients in developed

countries were clearly more likely to be managed with BCS[OR: 2.27 (95% CI, 1.43–3.61), P = 0.0004] independent of thetreatment arm (P = 0.385).Finally, we evaluated the association between the different

baseline characteristics and the type of planned surgery before

Table 1. Distribution of pre-surgery covariates and their association with actual surgery

Breast-conservingsurgery (BCS), n (%)

Mastectomy, n (%) P value

Geographic regionsa <0.0001Developing countries 63 (32.1) 133 (67.9)Developed countries 124 (53.2) 109 (46.8)

Age 0.3820–<40 years 34 (41.5) 48 (58.5)40–<50 years 52 (39.4) 80 (60.6)50–<60 years 58 (44.3) 73 (55.7)

≥60 years 43 (51.2) 41 (48.8)Histologic grade 0.021 5 (45.5) 6 (54.5)2–3 166 (46.4) 192 (53.6)Unknown 16 (26.7) 44 (73.3)

Histologic type 0.95Lobular 0 17 (100)Ductal 179 (45.5) 214 (54.5)Others 8 (42.1) 11 (57.9)

Hormone receptor status 0.0056Positive (ER and/or PgR) 108 (50.2) 107 (49.8)Negative 79 (36.9) 135 (63.1)

Tumor multicentricity/multifocality 0.0073No 175 (45.9) 206 (54.1)yes 12 (25.0) 36 (75.0)

Planned surgery at diagnosis <0.0001Not conservative 92 (30.6) 209 (69.4)Conservative 95 (74.2) 33 (25.8)

Tumor size at baseline (cm) <0.0001≤5 130 (52.8) 116 (47.2)>5 56 (30.9) 125 (69.1)

Clinical lymph node status at baseline 0.07N0/1 166 (45.4) 200 (54.6)N2 or Nx 21 (33.3) 42 (66.7)

Tumor measurements by mammogram/ultrasound at pre-surgery visit 0.86Complete response 32 (47.1) 36 (52.9)Partial response 89 (44.7) 110 (55.3)Stable disease 38 (42.2) 52 (57.8)Progressive disease 6 (35.3) 11 (64.7)Not measured 22 (40.0) 33 (60.0)

Tumor measurements by breast palpation at pre-surgery visit 0.0001Complete disappearance 90 (54.2) 76 (45.8)Residual palpable disease 67 (32.8) 137 (67.2)Not measured 30 (50.8) 29 (49.2)

Neoadjuvant treatment completion 0.2No 55 (48.7) 58 (51.3)Yes 132 (41.8) 184 (58.2)

Randomization arm 0.628Arm A (lapatinib alone) 66 (46.2) 77 (53.8)Arm B (trastuzumab alone) 58 (40.6) 85 (59.4)Arm C (lapatinib + trastuzumab) 63 (44.1) 80 (55.9)

aDeveloping countries defined include low, low-middle and upper-middle income economies; developed countries include: high-income economies andhigh-income Organization for Economic Co-operation and Development (OECD) members.

original articles Annals of Oncology

| Criscitiello et al. Volume 24 | No. 8 | August 2013

starting neoadjuvant therapy in an adjusted logistic regressionmodel (Table 4). We found that independent of the differencesin age, tumor stage and pathological features, BCS was aboutfour times more likely to be offered in developed countriesthan developing countries [OR: 4.79 (95% CI, 2.96–7.75),P < 0.0001]. As expected, smaller tumors (≤5 cm) and limitednodal involvement were more likely planned to receive BCS[OR: 2.77 (95% CI, 1.76–4.38), P < 0.0001] and [OR: 3.35 (95%CI, 1.55–7.26), P = 0.0022], respectively. No association wasobserved between the type of planned surgery and either age,hormone receptor status or tumor multicentricity.

discussionIn this study, we evaluated the factors associated with surgicalmanagement following neoadjuvant therapy in patientsenrolled in the NeoALTTO trial. We found that the type ofplanned surgery at diagnosis, tumor multicentricity, ER status,tumor size at baseline and the presence of residual tumor onbreast palpation were the main determinants of the type of thesurgery that was carried out following neoadjuvant therapy.The type of planned surgery itself was influenced by obviousfactors such as tumor size, but also by the geographic region.Strikingly, the radiological response to treatment appeared tohave played no role in the surgical decision. This highlights anegative attitude that may deny a large fraction of women thechance of preserving their breast, with no clinical reasons thatjustify this decision.For nearly three decades, we have sufficient evidence

supporting that a BCS followed by radiotherapy is as effectiveas mastectomy in terms of local control, disease-free andoverall survival in newly diagnosed breast cancer patients[18, 19]. However, evidence is less clear regarding the role of

BCS following neoadjuvant therapy. Nevertheless, severalstudies point out that patients managed with BCS afterneoadjuvant therapy have an acceptably low rate of loco-regional recurrences [20–25]. Such an approach has beenfurther endorsed by several groups as well [3, 4] and iscurrently widely accepted, in the absence of classiccontraindications to BCS. Accordingly, improvements inneoadjuvant therapies are expected to have a positive impacton breast conservation rates. However, this did not appear tobe the case in our study.Patients with ER-negative tumors were less likely to have a

BCS compared with patients with ER-positive disease. This wasobserved independent of other important factors like tumorstage, tumor multicentricity, response to treatment or even thetype of planned surgery at diagnosis. This is a major concernas ER-negative tumors are highly responsive to chemotherapyin the neoadjuvant setting. Interestingly, across the fourneoadjuvant trials testing the role of dual HER2 blockade, thepCR rates were significantly higher in the ER-negative patientpopulation, despite being all HER2-positive. Hence, suchattitude would deny the most suitable patient population forthe neoadjuvant approach the chance to preserve their breast,which is so far the main advantage for preoperative therapy.Indeed, it is not clear why patients with ER-negative tumorswere less considered for BCS. It appears that the knownaggressive nature of ER-negative tumors contributed to suchdecision, which underscores that BCS is perceived as aninferior option to mastectomy in this patient population, aconception that is not supported by evidence and, in ouropinion, should be challenged.At diagnosis, the tumor size appeared to influence the type

of surgery that the patient is eligible to undergo, which isexpected. However, we noticed that patients treated in

Table 2. Logistic regression model showing the factors associated with performing breast-conserving surgery including the type of planned surgery atdiagnosis

Comparison by categories Odds ratio (95% CI) P value

Geographic regions* Developed versus Developing countries 1.62 (0.98–2.67) 0.059Age categorical 20–<40 versus ≥60 0.88 (0.41–1.88) 0.283

40–<50 versus ≥60 0.57 (0.28–1.13)50–<60 versus ≥60 0.93 (0.47–1.85)

Histologic grade Grade 1 versus Grade (2 or 3) 0.86 (0.19–3.87) 0.309Grade X versus Grade (2 or 3) 0.58 (0.28–1.17)

Hormone receptor status Negative versus Positive 0.50 (0.31–0.81) 0.0049

Multicentricity/multifocality Yes versus No 0.21 (0.09–0.48) 0.0002Clinical tumor size ≤5 versus >5 cm 1.97 (1.20–3.23) 0.0076Clinical nodal stage N0 or N1 versus N2 or Nx 0.84 (0.42–1.70) 0.633Planned surgery at diagnosis Conservative versus not conservative 5.46 (3.17–9.41) <0.0001Clinical response not measured versus complete response 0.54 (0.24–1.23) 0.0069

No complete response versus complete response 0.41 (0.23–0.71)Radiological response Complete response versus stable disease 1.18 (0.52–2.65) 0.735

Not measured versus stable disease 1.08 (0.45–2.60)Partial response versus stable disease 1.16 (0.63–2.13)Progressive disease versus stable disease 0.49 (0.13–1.79)

Neoadjuvant treatment completion No versus yes 1.22 (0.68–2.19) 0.499Treatment arm Arm A versus Arm C 1.31 (0.74–2.31) 0.41

Arm B versus Arm C 1.50 (0.81–2.77)

Arm A, lapatinib alone arm; Arm B, trastuzumab alone arm; Arm C, combined lapatinib and trastuzumab arm.



developing countries were more likely to be assigned tomastectomy independent of the presence of advancedtumor stage at diagnosis. This could be attributed to the lackof optimal radiotherapy services in some of these countries[26]. Another possible reason could be the fact that breastcancer patients are less likely to be managed withinmultidisciplinary teams in the developing world [27, 28].Hence, the lack of communication between the medical andsurgical teams in such cases could considerably affect thesurgical decision.The introduction of BCS and later sentinel lymph node

biopsy has resulted in tremendous improvement in the qualityof life of millions of women, who were spared aggressive

mutilating surgeries [29]. While we have limited long-termdata on the outcome of BCS after neoadjuvant therapy, webelieve that modern breast cancer surgery should orientateits strategy focusing mainly on post-treatment outcomesrather than the baseline tumor characteristics. We believe thatour results call for a clear consensus on the role of BCS inpatients responding to neoadjuvant therapy. This issuehighlights the importance of multidisciplinary discussions andthe need for mutual collaboration between the medical andsurgical teams, particularly in the neoadjuvant setting. Thiswill translate the progress in neoadjuvant therapies intoimproved breast conservation rates, which would positivelyimpact patient’s quality of life and overall patient care.

Table 3. Logistic regression model showing the factors associated with performing breast-conserving surgery excluding the type of planned surgery atdiagnosis

Comparison by categories Odds ratio (OR, 95% CI) P value

Geographic regions* Developed versus Developing countries 2.27 (1.43–3.61) 0.0005Age categorical 20–<40 versus ≥60 0.94 (0.46–1.95) 0.318

40–<50 versus ≥60 0.64 (0.33–1.23)50–<60 versus ≥60 1.04 (0.54–2.01)


Hormone receptor status Negative versus Positive 0.53 (0.34–0.84) 0.0063

Multicentricity/multifocality Yes versus No 0.19 (0.09–0.43) <0.0001Clinical tumor size ≤5 versus >5 cm 2.38 (1.49–3.81) 0.0003Clinical nodal stage N0 or N1 versus N2 or Nx 1.04 (0.53–2.01) 0.915Clinical response not measured versus complete response 0.52 (0.25–1.12) 0.0004

No complete response versus complete response 0.34 (0.20–0.58)Radiological response Complete response versus stable disease 1.00 (0.46–2.17) 0.916

Not measured versus stable disease 0.97 (0.43–2.18)Partial response versus stable disease 1.09 (0.62–1.93)Progressive disease versus stable disease 0.63 (0.19–2.08)

Neoadjuvant treatment completion No versus yes 1.12 (0.65–1.94) 0.678Treatment arm Arm A versus Arm C 1.39 (0.81–2.37) 0.385

Arm B versus Arm C 1.42 (0.80–2.54)

Arm A, lapatinib only arm; Arm B, trastuzumab only arm; Arm C, combined lapatinib and trastuzumab arm.

Table 4. Logistic regression model showing the factors associated with breast-conserving surgery as the type of planned surgery before starting neoadjuvanttherapy

Comparison by categories Odds ratio (95% CI) P value

Geographic regionsa Developed versus Developing countries 4.13 (2.48–6.89) <0.0001Age categorical 20–<40 versus ≥60 0.92 (0.47–1.80) 0.859

40–<50 versus ≥60 1.14 (0.63–2.06)50–<60 versus ≥60 0.93 (0.51–1.70)


Histologic type Ductal versus others 0.74 (0.28–1.91) 0.447Lobular versus others 0.37 (0.08–1.74)

Hormone receptor status Negative versus Positive 0.81 (0.53–1.22) 0.306Multicentricity/multifocality Yes versus No 1.16 (0.59–2.28) 0.658Clinical tumor size ≤5 versus >5 cm 2.77 (1.76–4.38) <0.0001Clinical lymph node status N0/1 versus N2 or Nx 3.35 (1.55–7.26) 0.0022

aDeveloping countries defined include low, low-middle and upper-middle income economies; developed countries include high-income economies andhigh-income Organization for Economic Co-operation and Development (OECD) members.

original articles Annals of Oncology

| Criscitiello et al. Volume 24 | No. 8 | August 2013

acknowledgementsThis work was presented as an oral presentation at theEuropean Society for Medical Oncology (ESMO) annualmeeting (Abstract # 247O_PR), September 28–October 2,2012, Vienna, Austria.

fundingCC acknowledges a TRANSBIG fellowship. HAA Jracknowledges a fellowship grant from the European Society forMedical Oncology (no grant numbers).

disclosureGlaxoSmithKline distributed the study drugs and providedfinancial support to the NeoALLTO trial, but imposed norestriction on data analysis, decision to publish or preparationof this manuscript. Frontier Science received funding fromGlaxoSmithKline and Roche. EdA has served on an advisoryboard and received a travel grant from GlaxoSmithKline, andhas been a speaker for Roche. MJP has received honorariafrom Roche-Genentech. JB has received honoraria from Roche.HE has been a speaker and received travel grants fromGlaxoSmithKline. Serena Di Cosimo has been a speaker forGlaxoSmithKline. All the remaining authors havedeclared no conflicts of interest.

references1. Bear HD, Anderson S, Smith RE et al. Sequential preoperative or postoperative

docetaxel added to preoperative doxorubicin plus cyclophosphamide for operablebreast cancer:National Surgical Adjuvant Breast and Bowel Project Protocol B-27.J Clin Oncol 2006; 24: 2019–2027.

2. Fisher B, Brown A, Mamounas E et al. Effect of preoperative chemotherapy onlocal-regional disease in women with operable breast cancer: findings fromNational Surgical Adjuvant Breast and Bowel Project B-18. J Clin Oncol 1997;15: 2483–2493.

3. Kaufmann M, von Minckwitz G, Mamounas EP et al. Recommendations from aninternational consensus conference on the current status and future ofneoadjuvant systemic therapy in primary breast cancer. Ann Surg Oncol 2012;19: 1508–1516.

4. Berruti A, Generali D, Kaufmann M et al. International expert consensus onprimary systemic therapy in the management of early breast cancer: highlights ofthe Fourth Symposium on Primary Systemic Therapy in the Management ofOperable Breast Cancer, Cremona, Italy (2010). J Natl Cancer Inst Monogr2011; 2011: 147–151.

5. Mieog JS, van der Hage JA, van de Velde CJ. Neoadjuvant chemotherapy foroperable breast cancer. Br J Surg 2007; 94: 1189–1200.

6. Fisher B, Saffer E, Rudock C et al. Effect of local or systemic treatment prior toprimary tumor removal on the production and response to a serum growth-stimulating factor in mice. Cancer Res 1989; 49: 2002–2004.

7. Kong X, Moran MS, Zhang N et al. Meta-analysis confirms achieving pathologicalcomplete response after neoadjuvant chemotherapy predicts favourable prognosisfor breast cancer patients. Eur J Cancer 2011; 47: 2084–2090.

8. Liedtke C, Mazouni C, Hess KR et al. Response to neoadjuvant therapy and long-term survival in patients with triple-negative breast cancer. J Clin Oncol 2008;26: 1275–1281.

9. von Minckwitz G, Untch M, Blohmer JU et al. Definition and impact of pathologiccomplete response on prognosis after neoadjuvant chemotherapy in variousintrinsic breast cancer subtypes. J Clin Oncol 2012; 30: 1796–1804.

10. Gianni L, Eiermann W, Semiglazov V et al. Neoadjuvant chemotherapy withtrastuzumab followed by adjuvant trastuzumab versus neoadjuvant chemotherapy

alone, in patients with HER2-positive locally advanced breast cancer (the NOAHtrial): a randomised controlled superiority trial with a parallel HER2-negativecohort. Lancet 2010; 375: 377–384.

11. Untch M, Rezai M, Loibl S et al. Neoadjuvant treatment with trastuzumab inHER2-positive breast cancer: results from the GeparQuattro study. J Clin Oncol2010; 28: 2024–2031.

12. Buzdar AU, Ibrahim NK, Francis D et al. Significantly higher pathologic completeremission rate after neoadjuvant therapy with trastuzumab, paclitaxel, andepirubicin chemotherapy: results of a randomized trial in human epidermalgrowth factor receptor 2-positive operable breast cancer. J Clin Oncol 2005; 23:3676–3685.

13. Baselga J, Bradbury I, Eidtmann H et al. Lapatinib with trastuzumab for HER2-positive early breast cancer (NeoALTTO): a randomised, open-label, multicentre,phase 3 trial. Lancet 2012; 379: 633–640.

14. Gianni L, Pienkowski T, Im YH et al. Efficacy and safety of neoadjuvantpertuzumab and trastuzumab in women with locally advanced, inflammatory, orearly HER2-positive breast cancer (NeoSphere): a randomised multicentre, open-label, phase 2 trial. Lancet Oncol 2012; 13: 25–32.

15. Robidoux A, Tang G, Rastogi P et al. Evaluation of lapatinib as a component ofneoadjuvant therapy for HER2+ operable breast cancer: NSABP protocol B-41.J Clin Oncol 2012; 30(Suppl): abstr LBA506.

16. Schneeweiss A, Chia S, Hickish T et al. Neoadjuvant pertuzumab andtrastuzumab concurrent or sequential with an anthracycline-containing orconcurrent with an anthracycline-free standard regimen: a randomized phase IIstudy (TRYPHAENA). Cancer Res 2011; S5–S6.

17. Fisher ER, Wang J, Bryant J et al. Pathobiology of preoperative chemotherapy:findings from the National Surgical Adjuvant Breast and Bowel (NSABP) protocolB-18. Cancer 2002; 95: 681–695.

18. Fisher B, Anderson S, Bryant J et al. Twenty-year follow-up of a randomizedtrial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiationfor the treatment of invasive breast cancer. N Engl J Med 2002; 347:1233–1241.

19. Veronesi U, Cascinelli N, Mariani L et al. Twenty-year follow-up of a randomizedstudy comparing breast-conserving surgery with radical mastectomy for earlybreast cancer. N Engl J Med 2002; 347: 1227–1232.

20. Beriwal S, Schwartz GF, Komarnicky L et al. Breast-conserving therapyafter neoadjuvant chemotherapy: long-term results. Breast J 2006; 12:159–164.

21. Chen AM, Meric-Bernstam F, Hunt KK et al. Breast conservation afterneoadjuvant chemotherapy: the MD Anderson cancer center experience. J ClinOncol 2004; 22: 2303–2312.

22. Min SY, Lee SJ, Shin KH et al. Locoregional recurrence of breast cancerin patients treated with breast conservation surgery and radiotherapy followingneoadjuvant chemotherapy. Int J Radiat Oncol Biol Phys 2011; 81: e697–e705.

23. Cebrecos I, Cordoba O, Deu J et al. Can we predict local recurrence in breastconserving surgery after neoadjuvant chemotherapy? Eur J Surg Oncol 2010;36: 528–534.

24. Peintinger F, Symmans WF, Gonzalez-Angulo AM et al. The safety of breast-conserving surgery in patients who achieve a complete pathologic response afterneoadjuvant chemotherapy. Cancer 2006; 107: 1248–1254.

25. Mauri D, Pavlidis N, Ioannidis JP. Neoadjuvant versus adjuvant systemictreatment in breast cancer: a meta-analysis. J Natl Cancer Inst 2005; 97:188–194.

26. Anderson BO, Cazap E, El Saghir NS et al. Optimisation of breast cancermanagement in low-resource and middle-resource countries: executive summaryof the Breast Health Global Initiative consensus, 2010. Lancet Oncol 2011; 12:387–398.

27. El Saghir NS, El-Asmar N, Hajj C et al. Survey of utilization of multidisciplinarymanagement tumor boards in Arab countries. Breast 2011; 20(Suppl 2):S70–S74.

28. Saini KS, Taylor C, Ramirez AJ et al. Role of the multidisciplinary team in breastcancer management: results from a large international survey involving 39countries. Ann Oncol 2012; 23: 853–859.

29. Youssef OZ, Azim HA Jr. Understanding the factors associated with the surgicalmanagement of early breast cancer. Gland Surg 2013; 2: 4–6.



Documents

Factors associated with surgical management following neoadjuvant therapy in patients with primary HER2-positive breast cancer: results from the NeoALTTO phase III trial