reviews Annals of Oncology

12. Egger M, Davey Smith G, Schneider M et al. Bias in meta-analysis detected by a 18. Capotorto AM, Pavesi L, Pedrazzoli P et al. Randomized, controlled, multicenter
simple, graphical test. Br Med J 1997; 315: 629–634. phase III trial of standard-dose fluorouracil-epirubicin-cyclophosphamide (FEC),
13. Pfreundschuh M, Trumper L, Kloess M et al. Two-weekly or 3-weekly CHOP compared with time-intensive FEC (FEC-G) and mitoxantrone-methotrexate-
chemotherapy with or without etoposide for the treatment of elderly patients with mitomycin C (MMM-G) in metastatic breast carcinoma. J Chemother 2003; 15:
aggressive lymphomas: results of the NHL-B2 trial of the DSHNHL. Blood 2004; 184–91.
104: 634–641. 19. Diehl V, Franklin J, Pfreundschuh M et al. Standard and increased-dose BEACOPP
14. Osby E, Hagberg H, Kvaloy S et al. CHOP is superior to CNOP in elderly patients chemotherapy compared with COPP-ABVD for advanced Hodgkin’s disease. N
with aggressive lymphoma while outcome is unaffected by filgrastim treatment: Engl J Med 2003; 348: 2386–2395.
results of a Nordic Lymphoma Group randomized trial. Blood 2003; 101: 20. Henderson IC, Berry DA, Demetri GD et al. Improved outcomes from
3840–3848. adding sequential Paclitaxel but not from escalating Doxorubicin dose in
15. Bonomi P, Kim K, Fairclough D et al. Comparison of survival and quality of life in an adjuvant chemotherapy regimen for patients with node-positive
advanced non-small-cell lung cancer patients treated with two dose levels of primary breast cancer. J Clin Oncol 2003; 21: 976–83.
paclitaxel combined with cisplatin versus etoposide with cisplatin: results of 21. Sledge GW, Neuberg D, Bernardo P et al. Phase III trial of doxorubicin,
an Eastern Cooperative Oncology Group trial. J Clin Oncol 2000; 18: paclitaxel, and the combination of doxorubicin and paclitaxel as front-line
623–631. chemotherapy for metastatic breast cancer: an intergroup trial (E1193) [see
16. Burnell M, Levine MN, Chapman JA et al. Cyclophosphamide, epirubicin, and comment]. J Clin Oncol 2003; 21: 588–592.
fluorouracil versus dose-dense epirubicin and cyclophosphamide followed by 22. Maurel J, Lopez-Pousa A, de Las Penas R et al. Efficacy of sequential high-dose
Paclitaxel versus Doxorubicin and cyclophosphamide followed by Paclitaxel in node- doxorubicin and ifosfamide compared with standard-dose doxorubicin in patients
positive or high-risk node-negative breast cancer. J Clin Oncol 2010; 28: 77–82. with advanced soft tissue sarcoma: an open-label randomized phase II study of the
17. Burnell MJI, Levine MN, Chapman JA et al. A phase III adjuvant trial of sequenced Spanish group for research on sarcomas. J Clin Oncol 2009; 27: 1893–1898.
EC + filgrastim + epoetin-alpha followed by paclitaxel compared to sequenced AC 23. Moebus V, Jackisch C, Lueck HJ et al. Intense dose-dense sequential
followed by paclitaxel compared to CEF in women with node-positive or high-risk chemotherapy with epirubicin, paclitaxel, and cyclophosphamide compared with
node-negative breast cancer (NCIC CTG MA.21). J Clin Oncol 2007; 25: 550. conventionally scheduled chemotherapy in high-risk primary breast cancer: mature
(ASCO 2007 Meeting Abstract 550). results of an AGO phase III study. J Clin Oncol 2010; 28: 2874–2880.

Annals of Oncology 24: 2484–2492, 2013

doi:10.1093/annonc/mdt239
Published online 12 July 2013

Neoadjuvant treatment of borderline resectable

and non-resectable pancreatic cancer
V. Heinemann*, M. Haas & S. Boeck
Department of Medical Oncology and Comprehensive Cancer Center, Ludwig-Maximilians-University of Munich, Munich, Germany

Received 2 April 2013; revised 20 May 2013; accepted 21 May 2013

Neoadjuvant therapy is increasingly becoming a valid treatment option for patients with locally advanced pancreatic
cancer (LAPC). In borderline resectable disease, neoadjuvant therapy is employed to improve the probability of margin-
clear resections. In non-metastatic, non-resectable pancreatic cancer, treatment primarily aims to induce disease control,
but may achieve conversion to surgical resectability in some patients. Several treatment modalities including
chemotherapy, chemoradiotherapy (CRT) or the sequential use of both have been investigated in numerous, mostly small
and non-randomized studies. Nevertheless, there is a consistent finding that neoadjuvant therapy can induce resectability
in up to 30%–40% of LAPC patients. Once resection has been achieved, overall survival appears to be comparable to
that observed for primarily resectable patients. Thus, patient selection evolves as an important aspect of neoadjuvant
therapy; retrospective analyses identified induction chemotherapy as an appropriate tool to define LAPC patients who
may benefit most from subsequent treatment with CRT. The clinical importance of induction chemotherapy may further
increase once highly active protocols such as the FOLFIRINOX or the gemcitabine plus nab-paclitaxel regimen are
introduced into novel multimodality treatment concepts.
Key words: chemoradiotherapy, chemotherapy, neoadjuvant, pancreatic cancer

*Correspondence to: Prof. Volker Heinemann, Department of Medical Oncology and

Comprehensive Cancer Center, Ludwig-Maximilians-University of Munich, Marchioninistr.
15, D-81377 Munich, Germany. Tel: +49-89-7095-2208; Fax: +49-89-7095-2257;
E-mail: volker.heinemann@med.uni-muenchen.de

© The Author 2013. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For permissions, please email: journals.permissions@oup.com.
Annals of Oncology reviews
introduction down-staging is needed to achieve R0 resection in borderline
resectable PC [7].
At first diagnosis, only 10%–20% of pancreatic cancer (PC)
patients present with primarily resectable disease, and locally
definition according to the NCCN guidelines
advanced, non-metastatic PC (LAPC) is observed in up to 30%
(version 2.2012)
of patients [1]. Median overall survival (OS) of primarily
resectable patients currently is in the order of 20–24 months, The definition of borderline resectable tumours includes the
while survival is only 9–13 months in LAPC. Several autopsy following characteristics: (i) no distant metastases; (ii) venous
studies suggest that 8%–15% of PC patients die with locally involvement of the superior mesenteric vein (SMV) or portal
advanced disease and without metastatic spread [2]. LAPC was vein demonstrating tumour abutment with impingement and
previously defined as non-metastatic, but surgically unresectable narrowing of the lumen, encasement of the SMV/portal vein,
disease. More recently, this entity was further subdivided into but without encasement of the nearby arteries, or short-segment
borderline resectable and non-resectable disease. Compared venous occlusion resulting from either tumour thrombus or
with primarily resectable PC, borderline resectable disease is encasement, but with suitable vessel proximal and distal to the
characterized by a higher risk of margin-positive resection and area of vessel involvement, allowing for safe resection and
accordingly is associated with a higher risk of recurrence. As a reconstruction; (iii) gastroduodenal artery encasement up to the
result, the prognosis of borderline resectable patients is expected hepatic artery with either short-segment encasement or direct
to be significantly better than that of non-resectable patients, abutment of the hepatic artery, without extension to the celiac
but is significantly worse than that of resectable patients [3]. axis; (iv) tumour abutment of the superior mesenteric artery
The present review therefore asks the clinically relevant question (SMA) not to exceed >180° of the circumference of the vessel
if the prognosis of borderline resectable and also non-resectable wall [4]. By contrast, non-metastatic pancreatic tumours are
PC patients can be improved with the use of novel neoadjuvant considered non-resectable if the following characteristics are
treatment approaches. fulfilled: (i) >180° SMA- or celiac artery encasement; (ii)
unreconstructible SMV/portal vein occlusion; (iii) aortic
invasion or encasement; (iv) metastases to lymph nodes beyond
the field of resection [4].
diagnostic workup of LAPC patients
The diagnosis of LAPC requires dedicated, high-quality imaging MD Anderson classifications of borderline
implemented by pancreatic protocol computerized tomography resectable PC
(CT) [4]. A recent consensus statement proposed 3D imaging of In addition to the purely radiological criteria of the NCCN
the pancreas and the associated mesenteric vessels by guidelines, Katz et al. [6] initially included more patient-related
multidetector CT as state of the art modality [5]. In selected factors and proposed a subgrouping of borderline resectable
cases, additional information may be required from endoscopic patients into three distinct subtypes: Group A was defined by
ultrasound (EUS) or endoscopic retrograde radiological criteria including tumour abutment of the visceral
cholangiopancreatography. Tumour resectability should be arteries or short-segment occlusion of the SMV. In group B,
determined by multidisciplinary tumour boards. They will also patients had findings suggestive, but not diagnostic of
estimate the probability to achieve tumour-free resection metastasis, and group C was characterised by medical
margins and can accordingly decide on the necessity to perform comorbidities and marginal performance status (supplementary
neoadjuvant therapy. Decision-making must also account for Table S1, available at Annals of Oncology online). The
patient-related parameters such as performance status, attribution of patients to the MD Anderson subsets was
comorbidity and compliance [6]. Before the start of therapeutic specifically important with regard to the duration of
interventions, a histological confirmation of pancreatic neoadjuvant treatment, which was markedly shorter in group C
adenocarcinoma is mandatory [4, 5]. This is preferably done by compared with groups A or B (4 versus 16 versus 15 weeks,
EUS-guided fine-needle aspiration (FNA) biopsy, but can also P = 0.002). More recently, also the MD Anderson group has
be achieved by CT-guided core needle biopsy. The additional focused on a primarily radiographic-based definition of
use of staging laparoscopy to exclude peritoneal metastasis in borderline resectable PC; however, a uniformly accepted set of
LAPC patients has been suggested by some authors, but is not criteria that define patients with borderline resectable disease
generally accepted [5]. still does not exist [8].

outcome of borderline resectable patients

definition of borderline resectable and
Evidence regarding outcome of borderline resectable PC
non-resectable PC patients is limited. Katz reported on 160 patients that had a
Katz et al. assessed 2454 PC patients and classified 160 patients median OS of 18 months and a 5-year survival rate of 18%.
(7%) as borderline resectable [6]. The difficulty to provide a Among this patient group, 41% underwent surgical tumour
clear definition of borderline resectable PC is characterized by resection and had an OS of 40 months. In non-resected patients,
its denomination as ‘the imprecise continuum between median OS was much shorter and reached only 13 months [6].
radiologically and technically resectable and unresectable Survival data with regard to the MD Anderson groups A, B and
disease’ [6]. While different terminologies have been used to C are also summarized in supplementary Table S1, available at
describe this entity, the underlying premise is that tumour Annals of Oncology online.

Volume 24 | No. 10 | October 2013 doi:10.1093/annonc/mdt239 | 

reviews Annals of Oncology

strategies to apply neoadjuvant therapy range of 9–12 months [10–13]. Gemcitabine-based combination
chemotherapy induced higher response rates (26% and 27%)
Due to the lack of randomized studies, the optimal performance than single-agent gemcitabine (4% and 15%). In addition,
of neoadjuvant therapy remains a matter of debate. Patients response rates obtained in LAPC were at least as high as those
with borderline resectable tumours principally have the choice observed in metastatic disease when identical regimens were
of upfront surgery. However, the risk of margin-positive applied (Table 1). This observation is in accordance with results
resection is substantial in this group of patients and is related to of a meta-analysis obtained by Gillen et al. [9] in non-resectable
an unfavourable outcome. Accordingly, there is a general LAPC patients: they reported that neoadjuvant treatment with
assumption that neoadjuvant therapy is specifically beneficial in combination chemotherapy resulted in a higher resection rate
borderline resectable tumours and improves the fraction of compared with single-agent chemotherapy (33% versus 27%).
resectable tumours. In their evaluation of 160 patients with
borderline resectable PC, Katz et al. [6] reported that 78%
completed preoperative therapy and restaging, and 41%
chemoradiation in LAPC
underwent pancreatectomy. In initially non-resectable disease, The use of CRT alone in LAPC is based on limited evidence.
treatment primarily aims to achieve tumour control. Also in this One randomized study demonstrated the superiority of 5-
subgroup, a minority of patients may benefit from conversion fluorouracil (5-FU)-based CRT compared with best supportive
treatment and may become eligible for surgery. Regular care [14]. A meta-analysis of two randomized studies using
evaluation of secondary resectability therefore remains an outdated radiation regimens indicated that CRT prolonged
important task specifically in responding patients. At present survival compared with radiation alone [15–17].
time, there is hardly any evidence to demonstrate the differential Most studies used 5-FU as reference chemotherapy in
benefit from neoadjuvant therapy in borderline resectable combination with radiation doses of 50–60 Gy. Also the
versus non-resectable PC since most studies evaluated LAPC as combination of gemcitabine with radiation has been
one group. As local tumour reduction together with systemic investigated in multiple studies. Owing to the increased toxic
tumour control remain primary goals in borderline resectable effect of this regimen, a reduction of either chemotherapy- or
and unresectable patients, common treatment strategies may be radiation dose became necessary and so far prevented the
employed initially in both entities. definition of one standard regimen. A recent meta-analysis (three
The parameter of choice to judge the efficacy of conversion randomized studies and one retrospective comparative study)
therapy is its ability to induce tumour response and to prevent suggested that the combination of radiation with gemcitabine
tumour progression. Most studies investigating neoadjuvant might be more effective than the combination with 5-FU [18].
therapy in PC have used chemoradiotherapy (CRT) as a Gillen recently evaluated 111 trials including 4394 PC
concurrent application of systemic chemotherapy and radiation patients [9]. Neoadjuvant therapy involved chemotherapy in
[9]. Owing to the development of more active regimens, also 96% and radiation therapy in 94% of studies. For non-resectable
induction treatment with chemotherapy alone is in a process of patients, the estimated overall response rate was 35%. Given a
evaluation. stable disease rate of 42%, disease control was obtained in 77%
of patients. Among patients with initially non-resectable
tumours, surgical exploration was carried out in 47%. The
chemotherapy alone in LAPC overall resection rate after neoadjuvant therapy was 33%, of
Most of the randomized trials investigating gemcitabine-based which 79% were carried out as R0 resections. By contrast,
combination chemotherapy in advanced PC not only included primarily unresectable patients, who were not resected, had a
patients with metastatic disease, but also evaluated 20%–30% of median OS of 10.2 months, while the 33% resected patients
LAPC patients. In four randomized trials, median survival of survived for a median of 20.5 months [9]. In summary, this
LAPC patients treated with chemotherapy alone was in the analysis demonstrates that neoadjuvant therapy predominantly

Table 1. Efficacy of gemcitabine-based chemotherapy in LAPC versus metastatic disease (selected studies)

References Regimen LAPC Metastatic disease

n ORR (%) PFS/TTP (months) OS (months) n ORR (%) PFS/TTP (months) OS (months)
Louvet [10] Gem + Oxaliplatin 51 27.4 7.4 10.3 106 26.4 4.6 8.5
Gem 47 14.9 5.3 10.3 109 18.3 3.0 6.7
Rocha Lima [11] Gem + Irinotecan 27 25.9 7.7 9.8 148 14.9 3.0 5.4
Gem 24 4.2 3.9 11.7 145 4.8 2.9 5.9
Poplin [12] Gem 86 na 5.4 9.2 737 na 2.7 5.4
Gem FDR
Gem + Oxaliplatin
Kindler [13] Gem + Bevacizumab 93 na na 9.9 509 na na 5.7
Gem + Placebo

LAPC, locally advanced pancreatic cancer; ORR, objective response rate; PFS, progression-free survival; TTP, time to progression; OS, overall survival;
Gem, gemcitabine; FDR, fixed-dose rate.

 | Heinemann et al. Volume 24 | No. 10 | October 2013

Annals of Oncology reviews
applied as CRT can induce remissions in about one-third of Table 2. Randomized comparison of upfront CRT followed by
LAPC patients. Tumour resection was achieved in one-third of chemotherapy versus chemotherapy alone in LAPC
LAPC patients and was associated with markedly prolonged
survival which was comparable to that reported for initially References Regimen n ORR PFS OS
resectable PC patients. Comparable data were also described by (%) (months) (months)
other systematic reviews and meta-analyses [19, 20]. Using Chauffert [22] CRTa → Gem 59 na na 8.6
more restrictive study selection criteria, Laurence et al. [20] Gem 60 13.0*
determined an overall resectability rate of 40% after application
Loehrer [23] CRTb → Gem 34 6 6.0 11.1**
of neoadjuvant CRT to patients with initially non-resectable PC.
Gem 37 5 6.7 9.2
However, these authors also state that the quality of the
currently available data is poor and that it is presently not LAPC, locally advanced pancreatic cancer; Gem, gemcitabine; CRT,
possible to draw firm conclusions with regard to the benefit chemoradiation; ORR, objective response rate; PFS, progression-free
from neoadjuvant CRT. survival; OS, overall survival.
a
CRT: 60 Gy plus concurrent chemotherapy with 5-fluorouracil/cisplatin.
b
CRT: 50.4 Gy plus concurrent chemotherapy with gemcitabine.
*P = 0.03; **P = 0.017.
upfront chemoradiation followed by gemcitabine
versus gemcitabine alone in LAPC notion that locoregional tumour control expectedly has a small
The concept of upfront CRT is based on (i) improvement of impact on outcome.
tumour control, (ii) reduction of metastatic potential of the
primary tumour and (iii) achievement of resectability. In their
systematic review on the treatment of LAPC, Huguet et al. [21] sequence of chemotherapy followed by
concluded that upfront application of CRT was not superior to chemoradiation in LAPC
exclusive chemotherapy in terms of OS, but increased About one-third of LAPC patients develop overt metastasis
treatment-related toxic effects. Further data provided by two during the initial phase of treatment [25]. These patients are
randomized studies comparing upfront CRT followed by unlikely to benefit from locoregional treatment such as upfront
gemcitabine to gemcitabine alone are inconclusive (Table 2): CRT and rather require systemic chemotherapy. This is
The FFCD/SFRO study randomly assigned patients either to 5- specifically true for CRT regimens where chemotherapy is
FU/cisplatin-based CRT followed by gemcitabine or to mainly used as a radiosensitizing component and has low
gemcitabine alone [22]. Owing to low recruitment, this study systemic anticancer therapy. Primary application of systemic
was stopped prematurely with 119 patients included. OS was chemotherapy therefore aims at tumour reduction and selection
markedly shorter in the CRT arm compared with the of patients without rapid development of distant metastasis.
gemcitabine arm (8.6 versus 13 months, P = 0.03). The rather Two retrospective studies suggested that induction
poor result obtained in the CRT arm was explained by its chemotherapy followed by CRT in selected patients may be an
significantly greater treatment-related toxic effect resulting in a optimal neoadjuvant approach for LAPC (Table 3). Huguet
reduction of gemcitabine application compared with the et al. [25] evaluated LAPC patients and investigated whether
chemotherapy-alone arm. In the maintenance phase, the initial chemotherapy could be used to select patients who would
median total dose of gemcitabine was only 46% (6.845 mg) in benefit from subsequent CRT. LAPC patients were analysed
the CRT arm compared with the control arm (15.000 mg). The after an initial chemotherapy of at least 3 months; 29% of
results of this study may be interpreted to the extent that patients developed metastasis after three months of
upfront application of intensive CRT is counterproductive when chemotherapy and therefore were not eligible for CRT. Among
it results in a major attenuation of subsequent systemic therapy. the remaining patients with non-metastatic disease, those who
The ECOG E4201 study compared gemcitabine-based CRT received subsequent CRT had a markedly better outcome than
followed by gemcitabine to treatment with gemcitabine alone those who continued on chemotherapy alone (OS: 15.0 versus
[23]. After enrolment of 74 patients, also this study was 11.7 months, P = 0.0009). These data are in line with another
terminated early due to low accrual. No differences were retrospective analysis by Krishnan et al. [26]. The authors
obtained with regard to response rate (6% versus 5%) and analysed 323 consecutive LAPC patients that were either treated
median progression-free survival (PFS, 6.0 versus 6.7 months). with CRT as initial therapy (n = 247) or had received induction
However, the rate of stable disease at 3 months was greater (68% chemotherapy with gemcitabine for a median duration of 2.5
versus 35%) as was OS (11.1 versus 9.2 months, P = 0.017). months before CRT (n = 76). Induction chemotherapy was
Owing to the limited patient number, there was a wide associated with a significant improvement of PFS (6.4 versus 4.2
overlapping of confidence intervals around the median OS; the months, P < 0.001) and OS (11.9 versus 8.5 months, P < 0.001).
results should therefore be interpreted with some caution [24]. Interestingly, the use of induction chemotherapy did not lead to
Nevertheless, this small study might indicate that the addition a significant difference in the pattern of recurrence [26]. In
of CRT to systemic therapy may have improved survival support of this rationale, Katz et al. [6] proposed a treatment
through an improvement of local control, while distant algorithm including the sequential use of induction
metastasis and with it PFS were not affected. On the other hand, chemotherapy followed by CRT in non-progressing patients.
the survival benefit induced by the addition of CRT to Recently, data from the first randomized phase II trial
gemcitabine treatment was small (1.9 months) and points to the investigating a sequential approach of induction chemotherapy

Volume 24 | No. 10 | October 2013 doi:10.1093/annonc/mdt239 | 

reviews Annals of Oncology

Table 3. Sequential application of chemotherapy followed by CRT in three investigating 122 borderline resectable patients that received
LAPC studies different preoperative treatment protocols 84 patients had a
stable disease, 15 patients a partial response and 23 progressive
Reference Regimen Type of n PFS OS disease by RECIST. Although only one patient had his disease
study (months) (months) down-staged to resectable status by imaging criteria, 85
Huguet [25] Gem-based CT Retrospective 56 7.4 11.7 patients were able to undergo pancreatectomy after surgical
Gem-based CT analysisa 72 10.8 15.0 exploration (with 81 patients achieving a R0 resection) [28].
→ CRT The authors therefore concluded that each patient with
borderline resectable disease should undergo surgical
Krishnan [26] CRT Retrospective 247 4.2 8.5
exploration after neoadjuvant treatment in the absence of
Gem-based CT analysis 76 6.4 11.9
metastases.
→ CRT

Mukherjee Gem/Cap → Randomized 38 10.4 13.4 surgery in initially non-resectable disease

[27] Gem-based phase II
The evaluation of surgical resectability after induction
CRT
chemotherapy and/or CRT in patients with non-resectable
Gem/Cap → 36 12.0 15.2
disease may be a different one compared with the more
Cap-based
aggressive approach for borderline resectable patients discussed
CRT
above. An evaluation of surgical resectability (based on imaging
modalities) should be carried out e.g. after induction
LAPC, locally advanced pancreatic cancer; Gem, gemcitabine; PFS, chemotherapy and also after the completion of an eventual
progression-free survival; OS, overall survival; CT, chemotherapy; CRT,
subsequent CRT (ideally within the setting of a
chemoradiation; Cap, capecitabine.
a
multidisciplinary tumour board and with the goal of a R0
Retrospective evaluation of 181 LAPC patients. After 3 months of
resection). There is currently no evidence available to support a
chemotherapy, 53 patients (29%) developed metastatic disease and were
‘routine’ surgical exploration for the assessment of resectability
excluded from the analysis.
in this patient population [21, 24].

followed by CRT became available: within the SCALOP study, innovative strategies to improve outcome
114 LAPC patients received a gemcitabine/capecitabine regimen in LAPC
as induction treatment, and subsequently 74 non-progressing
patients were randomly allocated to a gemcitabine- versus a Given the genetic complexity of PC, it is difficult to assume that
capecitabine-based CRT (50.4 Gy) approach [27]. Although improvement of locoregional treatment alone will make a major
this was a small phase II study, the obtained data suggested contribution to prolong survival [24]. Therefore, several new
that a capecitabine-based CRT regimen might be preferable to avenues are investigated to improve treatment efficacy in LAPC.
a gemcitabine-based regimen in the context of consolidation One of them is the exploration of new and active chemotherapy
chemoradiotherapy after a course of induction chemotherapy: regimens such as the FOLFIRINOX multidrug combination or
patients in the capecitabine-based CRT arm not only had a the use of nab-paclitaxel combined with gemcitabine. Another
longer PFS and OS (Table 3), but also a trend for a better option is to move into the direction of personalized medicine
quality of life and a lower rate of adverse events [27]. The and to improve molecular patient selection.
efficacy data from SCALOP are in concordance with the
results from the two previous retrospective LAPC analyses FOLFIRINOX: a new strategy to improve treatment
(Table 3); however, they stand in contrast to a meta-analysis efficacy in LAPC
which suggested that the combination of radiation with Conroy and et al. recently reported the high efficacy of the
gemcitabine might be more effective than the combination FOLFIRINOX regimen [29]. In a randomized phase III trial,
with 5-FU [18]. exclusively carried out in metastatic PC, FOLFIRINOX was
compared with single-agent gemcitabine and demonstrated its
the role of surgery after neoadjuvant superiority with regard to objective response rate (32% versus
treatment in LAPC 9%, P < 0.001), PFS (HR 0.47, P < 0.001) and OS (HR 0.57,
P < 0.001). This phase III study essentially confirmed the data of
surgery in initially borderline resectable disease a previous phase II study which had included both, LAPC and
In PC patients who receive neoadjuvant treatment for a tumour metastatic PC (Table 4). Interestingly, the response rate among
that initially was classified as borderline resectable the role of 11 LAPC patients was nearly identical to that observed in 35
subsequent surgery is essential to achieve a long-lasting disease- metastatic patients (27% versus 26%) [30].
free interval or even cure. In this context, the role of imaging as Preliminary data on the efficacy of (neoadjuvant)
a basis for the decision to proceed to surgery after neoadjuvant FOLFIRINOX in LAPC are presently available from several
treatment has to be discussed critically. There is increasing retrospective and registry analyses (summarized in Table 4).
evidence that response of borderline resectable PC to Hosein et al. [31] reported a retrospective analysis of 18
neoadjuvant therapy is not necessarily reflected by standard patients with unresectable or borderline resectable LAPC. R0
radiographic indicators. In a recent single-centre study resection after neoadjuvant FOLFIRINOX was achieved in 5

 | Heinemann et al. Volume 24 | No. 10 | October 2013

Annals of Oncology reviews
Table 4. Efficacy of FOLFIRINOX in LAPC based on objective response rate

References Study design Stage of disease n CR (%) PR (%) SD (%) PD (%) ORR (%) DCR (%)
Conroy [30] Phase II LAPC 11 na na na na 27 na
Metastatic 35 26

Gunturu [32] Retrospective LAPC 16 6 44 44 0 50 94

Metastatic 17 0 47 35 12 47 82

Faris [33] Retrospective LAPC 12 0 42 58 0 42 100

Metastatic 17 0 35 41 24 35 76

Peddi [34] Registry LAPC 18 6 28 50 17 34 84

Metastatic 22 0 18 46 36 18 64

Marthey [35] Prospective database LAPC 53 0 30 53 17 30 83

LAPC, locally advanced pancreatic cancer; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease, ORR, objective response
rate; DCR, disease control rate.

of 18 (28%) of patients. Gunturu [32] reported on 16 LAPC nab-paclitaxel plus gemcitabine

patients that achieved an overall response rate of 50% and a Von Hoff et al. [37] reported a phase I-II study evaluating the
disease control rate of 94%. This group modified the combination of gemcitabine plus albumin-bound paclitaxel
FOLFIRINOX regimen so that only 17% (6 of 35) of patients (nab-paclitaxel). This study was limited to patients with
received full dose FOLFIRINOX at the first cycle. The metastatic disease. However, the response rate of 48% and the
authors concluded that despite routine dose modifications, median OS of 12.2 months obtained in 44 patients at the MTD
the response to treatment was not significantly different from level was encouraging and points to the substantial antitumour
the historical control group previously reported by Conroy. activity of this combination. It is hypothesized that the
Faris et al. evaluated 12 LAPC patients and observed a antitumour effect of nab-paclitaxel is mediated via depletion of
response rate of 42% [33]. Peddi and et al. [34] evaluated 23 the peritumoural stroma and a subsequently improved transport
borderline and non-resectable PC patients from a of chemotherapeutic agents such as gemcitabine to the tumour.
FOLFIRINOX registry (n = 61). The regimen was modified in The first publication of the phase III MPACT trial (gemcitabine
about half of the patients beginning with the first cycle. versus gemcitabine plus nab-paclitaxel in metastatic PC) at the
Deletion of the 5-FU bolus and dose reduction of irinotecan ASCO-GI meeting 2013 recently reported not only a significant
were the two most common modifications applied. In spite of increase in median OS for the combination (6.7 versus 8.5
the dose reductions, FOLFIRINOX remained effective months, P = 0.000015), but also a significant improvement in
inducing either a partial remission or stable disease in the PFS (3.7 versus 5.5 months) and in the overall response rate (7%
majority of patients (Table 4). Prophylactic growth factor versus 23%) [38]. Nab-Paclitaxel thus may serve as an active
support was given in 67% of patients beginning with the first drug for neoadjuvant treatment options in LAPC, e.g.
cycle, another 10% of patients received G-CSF in subsequent also within a sequential approach combining gemcitabine/
cycles. Given this support, grade 3–4 neutropenia was limited nab-paclitaxel (thereby inducing a stroma depletion) followed
to 20% and only 3 of 61 patients (5%) developed neutropenic by FOLFIRINOX [39, 40].
fever [34].The latest report on the role of conventional
FOLFIRINOX in LAPC was a prospective database analysis
by Marthey et al.: preliminary data from the first 53 enrolled patient selection based on molecular parameters
patients showed a response rate of 30% with a corresponding of the tumour
disease control rate of 83%. Sequential CRT was applied in In an autopsy series the loss of Smad4/DPC4 expression was
62% of patients and 32% underwent subsequent tumour observed in only 22% of LAPC patients compared with 73% of
resection [35]. It is of interest to note that throughout the patients with metastatic disease [2]. These data may relate to
various studies comparable response rates were obtained in previous reports stating that Samd4 inactivation was associated
LAPC and metastatic disease. These preliminary analyses, with poorer survival in surgical series of PC [41, 42]. Crane
therefore, support the substantial activity of FOLFIRINOX reported a phase II study where 69 LAPC patients received
also in LAPC and give rise to the expectation that response initial chemotherapy with gemcitabine/oxaliplatin plus
rates in the range of 25%–40% can be achieved by intensive cetuximab followed by CRT plus cetuximab. Also in this study,
chemotherapy alone [36]. Prospective studies are required to Smad4 expression significantly correlated with a local rather
test the efficacy and tolerability of FOLFIRINOX in LAPC than distant dominant pattern of disease progression [43]. In
patients, also with regard to the use of more liberal inclusion contrast, a retrospective translational study from a large
criteria and with an optimal use of supportive therapy. surgical series (n = 471) showed that Smad4 expression was

Volume 24 | No. 10 | October 2013 doi:10.1093/annonc/mdt239 | 

reviews Annals of Oncology

Figure 1. Proposed treatment algorithm for LAPC patients outside clinical trials. In patients with borderline resectable PC, neoadjuvant treatment of about 2
months (with either chemotherapy or CRT) could be carried out in order to improve the R0 resection rate; subsequent surgical exploration is recommended for
all patients without evidence for distant metastasis. In patients with non-resectable, non-metastatic PC induction chemotherapy for about 2–3 months is
recommended; in patients with disease control, subsequent CRT should be delivered in order to improve local tumour response. A regular evaluation of
surgical resectability based on radiographic findings is recommended after induction chemotherapy and after CRT. PC, pancreatic cancer; CT, chemotherapy,
CRT, chemoradiotherapy.

not correlated with recurrence pattern but was predictive for tumour control, but with increasing antitumour activity also
benefit from adjuvant chemotherapy [44]. Interestingly, the conversion to resectability will move into the focus. The efficacy
chemokine receptor CXCR4 was a strong independent of neoadjuvant treatment as a conversion therapy may be
prognostic biomarker associated with distant metastatic judged based on its ability to induce radiographic tumour
recurrence in this investigation. As inconclusive data on responses (mainly for non-resectable patients) or to enable a
Smad4 have been reported form mainly retrospective studies subsequent R0 resection (for borderline resectable PC),
[45], the need for prospective translational investigations respectively. A clear superiority of single treatment modalities
(ideally conducted within prospective clinical trials) becomes such as neoadjuvant CRT or chemotherapy cannot be defined at
evident. This specifically holds true for subentities like LAPC, present. If chemotherapy alone is used, combination
where separate biomarker studies still are very rare [46]. chemotherapy appears to achieve higher response rates than
However, if a predictor of early metastatic spread could be single-agent chemotherapy. New options arise with the
identified, patients without the predictor should be those to development of the FOLFIRINOX regimen, while several cohort
obtain the greatest benefit from intensified locoregional studies support the activity of FOLFIRINOX also in LAPC, this
treatment, while patients at a higher risk of metastasis should needs to be verified in prospective clinical studies.
rather receive systemic therapy [24]. CRT was shown to be better than best supportive care, and
CRT was also superior to radiation alone. However, there is
insufficient evidence to show that upfront CRT is superior to
summary chemotherapy alone. Initial application of CRT may delay or
reduce necessary systemic therapy. Modern, hypofractionated
High-level evidence from sufficiently powered randomized trials radiation regimens may possibly overcome this problem.
carried out in neoadjuvant treatment of LAPC is largely On the other hand, initial induction chemotherapy not only
missing. In addition, conclusions from systematic reviews and prevents systemic progression, but may also be used to select for
meta-analyses are limited due to the lack of consistent non-progressive patients that benefit most from subsequent
definitions of unresectable versus borderline resectable disease radiation therapy. While the initial treatment approach may be
between different studies. The available data therefore need to be similar for borderline resectable and unresectable patients,
viewed with caution, and firm conclusions can only be drawn future studies carried out in LAPC should stratify between these
once more mature evidence is provided. Nevertheless, the two subgroups to account for different prognostic implications.
following summary may be helpful for treatment of LAPC in
daily clinical practice (Figure 1): in borderline resectable
patients, neoadjuvant therapy is intended to increase the
probability of tumour-free resection margins. The primary goal
disclosure
of treatment in non-resectable patients is achievement of The authors have declared no conflicts of interest.

 | Heinemann et al. Volume 24 | No. 10 | October 2013

Annals of Oncology reviews
references 21. Huguet F, Girard N, Guerche CS et al. Chemoradiotherapy in the management of
locally advanced pancreatic carcinoma: a qualitative systematic review. J Clin
1. Hidalgo M. Pancreatic cancer. N Engl J Med 2010; 362: 1605–1617. Oncol 2009; 27: 2269–2277.
2. Iacobuzio-Donahue CA, Fu B, Yachida S et al. DPC4 gene status of the primary 22. Chauffert B, Mornex F, Bonnetain F et al. Phase III trials comparing intensive
carcinoma correlates with patterns of failure in patients with pancreatic cancer. J induction chemoradiotherapy (60 Gy, infusional 5-FU and intermittent cisplatin)
Clin Oncol 2009; 27: 1806–1813. followed by maintenance gemcitabine with gemcitabine alone for locally advanced
3. Takahashi S, Kinoshita T, Konishi M et al. Borderline resectable pancreatic cancer: unresectable pancreatic cancer. Definitive results of the 2000–01 FFCD/SFRO
rationale for multidisciplinary treatment. J Hepatobiliary Pancreat Sci 2011; 18: study. Ann Oncol 2008; 19: 1592–1599.
567–574. 23. Loehrer PJ, Sr, Feng Y, Cardenes H et al. Gemcitabine alone versus
4. NCCN guidelines ‘Pancreatic adenocarcinoma’, version 2.2012. gemcitabine plus radiotherapy in patients with locally advanced pancreatic
5. Callery MP, Chang KJ, Fishman EK et al. Pretreatment assessment of resectable cancer: an Eastern Cooperative Oncology Group trial. J Clin Oncol 2011; 29:
and borderline resectable pancreatic cancer: expert consensus statement. Ann 4105–4112.
Surg Oncol 2009; 16: 1727–1733. 24. Philip PA. Locally advanced pancreatic cancer: where should we go from here?
6. Katz MH, Pisters PW, Evans DB et al. Borderline resectable pancreatic cancer: the J Clin Oncol 2011; 29: 4066–4068.
importance of this emerging stage of disease. J Am Coll Surg 2008; 206: 25. Huguet F, André T, Hammel P et al. Impact of chemoradiotherapy after disease
833–846; discussion 846–848. control with chemotherapy in locally advanced pancreatic adenocarcinoma in
7. Abrams RA, Lowy AM, O’Reilly EM et al. Comnined modality treatment of GERCOR phase II and III studies. J Clin Oncol 2007; 25: 326–331.
resectable and borderline resectable pancreas cancer: expert consensus 26. Krishnan S, Rana V, Janjan NA et al. Induction chemotherapy selects patients with
statement. Ann Surg Oncol 2009; 16: 1751–1756. locally advanced, unresectable pancreatic cancer for optimal benefit from
8. Katz MH, Marsh R, Herman JM et al. Borderline resectable pancreatic cancer: consolidative chemoradiation therapy. Cancer 2007; 110: 47–55.
need for standardization and methods for optimal clinical trial design. Ann Surg 27. Mukherjee S, Hurt CN, Bridgewater J et al. Gemcitabine-based or
Oncol 2013 June 21 [epub ahead of print], doi: 10.1007/s11864-013-0244-6. capecitabine-based chemoradiotherapy for locally advanced pancreatic cancer
9. Gillen S, Schuster T, Meyer zum Büschenfelde C et al. Preoperative/neoadjuvant (SCALOP): a multicentre, randomised, phase 2 trial. Lancet Oncol 2013; 14:
therapy in pancreatic cancer: a systemic review and meta-analysis of response 317–326.
and resection percentages. PLoS Medicine 2010; 7: e1000267. 28. Katz MH, Fleming JB, Bhosale P et al. Response of borderline resectable
10. Louvet C, Labianca R, Hammel P et al. Gemcitabine in combination with oxaliplatin pancreatic cancer to neoadjuvant therapy is not reflected by radiographic
compared with gemcitabine alone in locally advanced or metastatic pancreatic indicators. Cancer 2012; 118: 5749–5756.
cancer: results of a GERCOR and GISCAD phase III trial. J Clin Oncol 2005; 23: 29. Conroy T, Desseigne F, Ychou M et al. FOLFIRINOX versus gemcitabine for
3509–3516. metastatic pancreatic cancer. N Engl J Med 2011; 364: 1817–1825.
11. Rocha Lima CM, Green MR, Rotche R et al. Irinotecan plus gemcitabine results in 30. Conroy T, Paillot B, Francois E et al. Irinotecan plus oxaliplatin and leucovorin-
no survival advantage compared with gemcitabine monotherapy in patients with modulated fluorouracil in advanced pancreatic cancer—a Groupe Tumeurs
locally advanced or metastatic pancreatic cancer despite increased tumor Digestives of the Federation Nationale des Centres de Lutte Contre le Cancer study.
response rate. J Clin Oncol 2004; 22: 3776–3783. J Clin Oncol 2005; 23: 1228–1236.
12. Poplin E, Feng Y, Berlin J et al. Phase III, randomized study of gemcitabine and 31. Hosein PJ, Macintyre J, Kawamura C et al. A retrospective study of neoadjuvant
oxaliplatin versus gemcitabine (fixed-dose rate infusion) compared with FOLFIRINOX in unresectable or borderline-resectable locally advanced pancreatic
gemcitabine (30-minute infusion) in patients with pancreatic carcinoma E6201: adenocarcinoma. BMC Cancer 2012; 12: 199.
A trial of the Eastern Cooperative Oncology Group. J Clin Oncol 2009; 32. Gunturu KS, Yao X, Cong X et al. FOLFIRINOX for locally advanced and metastatic
27: 3778–3785. pancreatic cancer: single institution retrospective review of efficacy and toxicity.
13. Kindler HL, Niedzwiecki D, Hollis D et al. Gemcitabine plus bevacizumab compared Med Oncol 2013; 30: 361.
with gemcitabine plus placebo in patients with advanced pancreatic cancer: phase 33. Faris JE, Hong TS, McDermott S et al. FOLFIRINOX in locally advanced or
III trial of the Cancer and Leukemia Group B (CALGB 80303). J Clin Oncol 2010; metastatic pancreatic cancer. J Clin Oncol 2012; 30(suppl 4): abstract #313.
28: 3617–3622. 34. Peddi PF, Lubner S, McWilliams R et al. Multi-institutional experience with
14. Shinchi H, Takao S, Noma H et al. Length and quality of survival after external- FOLFIRINOX in pancreatic adenocarcinoma. JOP 2012; 13: 497–501.
beam radiotherapy with concurrent continuous 5-fluorouracil infusion for locally 35. Marthey L, Sa-Cunha A, Blanc JF et al. FOLFIRINOX for locally advanced
unresectable pancreatic cancer. Int J Radiat Oncol Biol Phys 2002; 53: 146–150. pancreatic adenocarcinoma. results of an AGEO multicentric prospective study.
15. Moertel CG, Frytak S, Hahn RG et al Therapy of locally unresectable pancreatic Ann Oncol 2012; 23(suppl 9): ix238, abstract #716P.
carcinoma: a randomized comparison of high dose (6000 rads) radiation alone, 36. Conroy T, Gavoille C, Samalin E et al. The role of the FOLFIRINOX regimen for
moderate dose radiation (4000 rads + 5-fluorouracil), and high dose radiation + 5- advanced pancreatic cancer. Curr Oncol Rep 2013; 15: 182–189.
fluorouracil: The Gastrointestinal Tumor Study Group. Cancer 1981; 48: 37. Von Hoff DD, Ramanathan RK, Borad MJ et al. Gemcitabine plus nab-paclitaxel
1705–1710. is an active regimen in patients with advanced pancreatic cancer: A phase I/II trial.
16. Cohen SJ, Dobelbower R, Lipsitz S et al. A randomized phase III study of J Clin Oncol 2011; 29: 4548–4554.
radiotherapy alone or with 5-fluorouracil and mitomycin-C in patients with locally 38. Von Hoff DD, Ervin TJ, Arena FP et al. Randomized phase III study of weekly nab-
advanced adenocarcinoma of the pancreas. Eastern Cooperative Oncology Group paclitaxel plus gemcitabine versus gemcitabine alone in patients with metastatic
study E8282. Int J Radiat Oncol Biol Phys 2005; 62: 1345–1350. adenocarcinoma of the pancreas (MPACT). J Clin Oncol 2012; 30(suppl 34):
17. Sultana A, Smith CT, Cunningham D et al. Meta-analyses of chemotherapy for abstract #LBA148.
locally advanced and metastatic pancreatic cancer. J Clin Oncol 2007; 25: 39. Ramanathan RK, Lee P, Leach JW et al. Phase II study of induction therapy with
2607–2615. gemcitabine and nab-paclitaxel followed by consolidation with mFOLFIRINOX in
18. Zhu CP, Shi J, Chen YX et al. Gemcitabine in the chemoradiotherapy for locally patients with metastatic pancreatic cancer. J Clin Oncol 2012; 30(suppl 34):
advanced pancreatic cancer: a meta-analysis. Radiother Oncol 2011; 99: abstract #233.
108–113. 40. Mondo EL, Noel MS, Katz AW et al. Unresectable locally advanced pancreatic
19. Morganti AG, Massaccesi M, La Torre G et al. A systematic review of resectability cancer: treatment with neoadjuvant leucovorin, fluorouracil, irinotecan, and
and survival after concurrent chemoradiation in primarily unresectable pancreatic oxaliplatin and assessment of surgical resectability. J Clin Oncol 2013; 31:
cancer. Ann Surg Oncol 2010; 17: 194–205. e37–e39.
20. Laurence JM, Tran PD, Morarji K et al. A systematic review and meta-analysis of 41. Tascilar M, Skinner HG, Rosty C et al. The SMAD4 protein and prognosis of
survival and surgical outcomes following neoadjuvant chemoradiotherapy for pancreatic ductal adenocarcinoma. Clin Cancer Res 2001; 7:
pancreatic cancer. J Gastrointest Surg 2011; 15: 2059–2069. 4115–4121.

Volume 24 | No. 10 | October 2013 doi:10.1093/annonc/mdt239 | 

reviews Annals of Oncology

42. Biankin AV, Morey AL, Lee CS et al. DPC4/Smad4 expression and outcome in chemotherapy in resected pancreatic adenocarcinoma. Ann Oncol 2012; 23:
pancreatic ductal adenocarcinoma. J Clin Oncol 2002; 20: 4531–4542. 2327–2335.
43. Crane CH, Varadhachary GR, Yordy JS et al. Phase II trial of cetuximab, gemcitabine, 45. Winter JM, Tang LH, Klimstra DS et al. Failure patterns in resected pancreas
and oxaliplatin followed by chemoradiation with cetuximab for locally advanced (T4) adenocarcinoma: lack of predicted benefit to SMAD4 expression. Ann Surg 2013
pancreatic adenocarcinoma: correlation of Smad4(Dpc4) immunostaining with January 28 [epub ahead of print], doi: 10.1097/SLA.0b013e31827fe9ce.
pattern of disease progression. J Clin Oncol 2011; 22: 3037–3043. 46. Bidard FC, Huguet F, Louvet C et al. Circulating tumor cells in locally advanced
44. Bachet JB, Maréchal R, Demetter P et al. Contribution of CXCR4 and SMAD4 pancreatic adenocarcinoma: the ancillary CirCe 07 study to the LAP 07 trial. Ann
in predicting disease progression pattern and benefit from adjuvant Oncol 2013; 24: 2057–2061.

Annals of Oncology 24: 2492–2500, 2013

doi:10.1093/annonc/mdt217
Published online 4 July 2013

Emerging approaches for treating HER2-positive

metastatic breast cancer beyond trastuzumab
W. J. Gradishar*
Northwestern University Feinberg School of Medicine, Robert H. Lurie Comprehensive Cancer Center, Chicago, USA

Received 30 November 2012; revised 3 May 2013; accepted 6 May 2013

Because metastatic breast cancer (MBC) is incurable in most cases, the goals of treatment are improvement in quality of
life, management of symptoms, and prolonged survival. The human epidermal growth factor receptor 2 (HER2) is
overexpressed in up to 30% of breast tumors, and before the development of HER-targeted therapy, HER2 positivity was
predictive of poorer clinical outcomes. Trastuzumab and pertuzumab (anti-HER2 monoclonal antibodies), lapatinib (a
small molecule inhibitor of HER2 and the epidermal growth factor receptor [EGFR]) are approved for treating HER2-
positive MBC in the United States. Although trastuzumab plus chemotherapy is currently regarded as the first-line
standard of care for HER2-positive MBC, it is not without shortcomings; these include its association with certain adverse
events (e.g. cardiotoxic effect) and development of resistance. A number of investigational agents that target HER2 and
other members of that receptor family are in clinical development for patients with HER2-positive MBC whose disease
has progressed on trastuzumab. In addition, in an effort to overcome treatment resistance, clinical trials are evaluating
combination therapy (investigational HER-targeted agents with trastuzumab or lapatinib). This review discusses recently
completed and ongoing phase II and III clinical trials of investigational HER-targeted agents in the setting of trastuzumab-
progressive, HER2-positive MBC.
Key words: HER inhibitors, HER2-positive, metastatic breast cancer, monoclonal antibodies, neoadjuvant therapy,
trastuzumab

introduction incurable, and the goals of treatment are to optimize quality of

life, manage symptoms, and prolong survival.
In 2013, it is estimated that there will be 39 620 breast cancer- Approximately 25%–30% of breast tumors overexpress
related deaths among women in the United States [1] and human epidermal growth factor receptor 2 (HER2, a
88 886 in Europe [2]. Systemic treatment modalities for breast transmembrane tyrosine kinase [TK] receptor within the HER
cancer management include cytotoxic chemotherapy, endocrine family), a tumor characteristic that tends to occur in younger
therapy, biologic therapy, or combinations of these modalities, patients and, before the advent of HER2-directed therapy,
with selection based on a number of factors (including patient predicted a poor clinical outcome [4, 5]. This review focuses on
and tumor characteristics, stage of disease, and patient the current status of investigational anti-HER2 agents actively
preference) [3]. In most cases, metastatic breast cancer (MBC) is being evaluated in phase II or III clinical trials, specifically in
the setting of trastuzumab-pretreated/progressive MBC, and
includes an overview of accumulating neoadjuvant data for
*Correspondence to: Dr William J. Gradishar, Northwestern University Feinberg School of
Medicine, Department of Medicine, Division of Hematology Oncology, Robert H. Lurie earlier use of anti-HER2 combinations. Relevant clinical trials
Comprehensive Cancer Center, 676 North St. Clair St, Suite 850, Chicago, IL 60611, were selected for inclusion based on searches of PubMed, key
USA. Tel: +1-312-695-4541; Fax: +1-312-695-6189; E-mail: w-gradishar@ oncology congresses, and the ClinicalTrials.gov registry.
northwestern.edu

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