ORIGINAL ARTICLE

Prognostic Factors of Survival After Neoadjuvant Treatment and

Resection for Initially Unresectable Pancreatic Cancer
Ulla Klaiber, MD,  Eva S. Schnaidt, MD,  Ulf Hinz, MSc,  Matthias M. Gaida, MD,y Ulrike Heger, MD, 
Thomas Hank, MD,  Oliver Strobel, MD,  John P. Neoptolemos, MD,  André L. Mihaljevic, MD, 
Markus W. Büchler, MD,  and Thilo Hackert, MD 
Downloaded from https://journals.lww.com/annalsofsurgery by 01UGrXh3ipqzR4DKqW7bOJtSxsRVOheLV9OzOeHq2POX2t1GrKUD6m1aBQdlm0lX7bZxxCFBpuxAM3exuxdXARVaPXROhrEIELxLuiE5dAoxEXL1EbBXOmtN5qOe2K0cfOG+czRLAQJzYcB+sBMdZA== on 03/16/2021

Objective: To evaluate the impact of clinical and pathological parameters,
including resection margin (R) status, on survival in patients undergoing
pancreatic surgery after neoadjuvant treatment for initially unresectable
pancreatic ductal adenocarcinoma (PDAC).
Background: Prognostic factors are well documented for patients with
resectable PDAC, but have not been described in detail for patients with
initially unresectable PDAC undergoing resection after neoadjuvant therapy.
Methods: Prospectively collected data of consecutive patients with initially
unresectable pancreatic cancer treated by neoadjuvant treatment and resection
were analyzed. The R status was categorized as R0 (tumor-free margin
>1 mm), R1 1 mm (tumor-free margin 1 mm), and R1 direct (microscopic
tumor infiltration at margin). Clinicopathological characteristics and out-
comes were compared among these groups and tested for survival prediction.
Results: Between January, 2006 and February, 2017, 280 patients with
borderline resectable (n ¼ 18), locally advanced (n ¼ 190), or oligometastatic
(n ¼ 72) disease underwent tumor resection after neoadjuvant treatment.
Median overall survival from the time of surgery was 25.1 months for R0 (n ¼
82), 15.3 months for R1 1 mm (n ¼ 99), and 16.1 months for R1 direct (n ¼
From the Department of General, Visceral and Transplantation Surgery, Univer-
sity of Heidelberg, Heidelberg, Germany; and yInstitute of Pathology, Univer-
sity of Heidelberg, Heidelberg, Germany.
Authors’ contributions: Ulla Klaiber has made substantial contributions to the
conception and design of the study, acquisition of data, and analysis and
interpretation of data. Eva S. Schnaidt has made substantial contributions to the
conception and design of the study, acquisition of data, and interpretation of
data. Ulf Hinz has made substantial contributions to the conception and design
of the study, analysis, and interpretation of data. Matthias M. Gaida has made
substantial contributions to the conception and design of the study, acquisition
of data, and interpretation of data. Ulrike Heger has made substantial contri-
butions to the conception and design of the study, acquisition of data, and
interpretation of data. Thomas Hank has made substantial contributions to the
conception and design of the study, acquisition of data, and interpretation of
data. John P. Neoptolemos has made substantial contributions to the concep-
tion and design of the study, and interpretation of data. André L. Mihaljevic has
made substantial contributions to the conception and design of the study, and
interpretation of data. Oliver Strobel has made substantial contributions to the
conception and design of the study, and interpretation of data. Markus W.
Büchler has made substantial contributions to the conception and design of the
study, and interpretation of data. Thilo Hackert has made substantial contri-
butions to the conception and design of the study, acquisition of data, and
analysis and interpretation of data. All authors have participated in drafting the
article or revising it critically for important intellectual content and all authors
have given final approval of this version to be published.
Funding: The resources and the facilities available at the Department of Surgery,
University of Heidelberg, were used to conduct this study. No additional
funding source was used.
All authors declare no financial interests and no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions of
this article on the journal’s Web site (www.annalsofsurgery.com).
Reprints: Thilo Hackert, MD, Department of General, Visceral and Transplanta-
tion Surgery, University of Heidelberg, Im Neuenheimer Feld 110, 69120
Heidelberg, Germany. E-mail: Thilo.Hackert@med.uni-heidelberg.de.
Copyright ß 2019 Wolters Kluwer Health, Inc. All rights reserved.
ISSN: 0003-4932/19/27301-0154
DOI: 10.1097/SLA.0000000000003270
99), with 3-year overall survival rates of 35.0%, 20.7%, and 18.5%, respec-
tively (P ¼ 0.0076). The median duration of the neoadjuvant treatment period
was 5.1 months. In multivariable analysis, preoperative CA 19–9 levels,
lymph node status, metastasis category, and vascular involvement were all
significant prognostic factors for overall survival. The R status was not an
independent prognostic factor.
Conclusions: In patients undergoing resection after neoadjuvant therapy for
initially unresectable PDAC, preoperative CA 19–9 levels, lymph node
involvement, metastasis category, and vascular involvement, but not the R
status, were independent prognostic factors of overall survival.
Keywords: locally advanced pancreatic cancer, neoadjuvant therapy,
pancreatic surgery, prognostic factors
(Ann Surg 2021;273:154–162)
P ancreatic ductal adenocarcinoma (PDAC) is currently the third
most frequent cause of cancer-related death in western countries,
and is predicted to become the second leading cause of death from
cancer within the next 10 years.1,2 Complete tumor resection remains
the only potentially curative option for pancreatic cancer, but only
15% to 20% of patients qualify for immediate surgery at the time of
diagnosis.3 In the majority of patients, distant metastases are found at
diagnosis, or the pancreatic tumor is unresectable due to locally
advanced spread into surrounding major vessels, ruling out safe
reconstruction. In these patients with locally advanced or oligome-
tastatic disease, neoadjuvant treatment increases the prospect of
achieving resectability. Combination chemotherapy with folinic acid,
5-fluorouracil, irinotecan and oxaliplatin (FOLFIRINOX), or chemo-
radiotherapy protocols may result in successful resection in up to
60% of patients with locally advanced PDAC and higher rates of
margin-negative resections in borderline resectable tumors, with a
significant survival benefit.4,5 Therefore, neoadjuvant chemotherapy
with or without chemoradiotherapy is an emerging concept in
pancreatic cancer surgery.
Clinical and pathological prognostic factors are well docu-
mented in patients with resectable PDAC, but are poorly described in
initially unresectable pancreatic cancer after neoadjuvant therapy.6,7
Achieving tumor-free resection margins is a key objective in pan-
creatic cancer surgery.8,9 In part due to variations in the definition of
resection margins (R), the reported R status rates and associated
survival times differ widely in the literature, with R1 rates ranging
from 0% to 87%.10,11 Standardized histopathological work-up of all
relevant resection margins, together with the concept of the circum-
ferential resection margin (CRM), allows better evaluation of the
surgical treatment and its associated prognostic impact, and also
better interstudy comparability.12,13 The R status has been shown to
be an important independent survival predictor in patients with
primarily resectable PDAC, especially when the revised R0 defini-
tion with 1 mm clearance is used.14–20 The prognostic significance of
the R status, and also other clinicopathological variables in patients

154 | www.annalsofsurgery.com Annals of Surgery  Volume 273, Number 1, January 2021

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.

Annals of Surgery  Volume 273, Number 1, January 2021
with initially unresectable PDAC undergoing resection after neo-
adjuvant treatment is uncertain. To investigate this was the aim of the
present study.
METHODS
Study Design and Patient Selection
A STROBE-compliant cohort study of patients with initially
unresectable PDAC was undertaken using the prospectively main-
tained pancreatic database at the Department of Surgery, University
of Heidelberg, Germany.21 The primary objective was to identify
prognostic factors, including the R status, in patients with borderline
resectable and locally advanced PDAC after neoadjuvant therapy and
resection. Patients with oligometastatic disease in whom the metas-
tases could be completely resected along with the primary were also
included. Informed consent was obtained from all patients, and the
study was approved by the institutional ethics committee on May 10,
2017 (approval no. S-226/2017). All consecutive patients undergoing
resection between January 1, 2006, and February 15, 2017 were
included. The census date of 2006 was chosen due to changes in the
pathological work-up and variations in pathological reporting12,13.
Patients with periampullary carcinoma, neuroendocrine tumor, ana-
plastic carcinoma, carcinoma in situ, and patients with other syn-
chronous metastatic tumors were excluded. Patients with moderate to
large metastases that could not be resected were also excluded, as
were patients with macroscopic (R2) or indeterminable (Rx) residual
tumor. Unresectability was defined according to the consensus
statement of the International Study Group of Pancreatic Surgery
(ISGPS), which is based primarily on the recommendations of the
National Comprehensive Cancer Network (NCCN).3 Depending on
the extent of vascular involvement of the tumor and the presence/
absence of distant metastases on cross-sectional imaging at diagno-
sis, tumors were categorized as borderline resectable, locally
advanced, or metastatic. Patients with resectable tumors were not
included in this study.
Neoadjuvant and Adjuvant Treatment
Neoadjuvant treatment was performed in all patients included
in this study. In the absence of standardized evidence, we used a
variety of different regimens involving combination chemotherapy
with or without chemoradiotherapy.22 Adjuvant treatment was
administered depending on the extent of neoadjuvant treatment,
the patient’s treatment tolerance, and tumor response, and in accor-
dance with the recommendations of the National Center of Tumor
Disease (NCT), University of Heidelberg, Germany.
Surgical Procedures
After neoadjuvant therapy, surgical exploration was under-
taken once the serum carbohydrate antigen (CA) 19–9 levels had
decreased considerably and if cross-sectional imaging revealed a
tumor response or stable disease according to the RECIST criteria
v1.1.23 The artery-first approach was routinely used to achieve
radical resection of the tumor surrounding the superior mesenteric
artery (SMA).24 The pancreatic transection margin and any suspi-
cious tissue surrounding critical vascular structures were assessed
during surgery by frozen-section microscopy. The triangle operation
was routinely performed to clear lymphatic tissue between the SMA,
celiac axis, and hepatic portal vein/superior mesenteric vein (HPV/
SMV).25 Resection was extended to take in nearby organs whenever
indicated to achieve complete tumor removal. Resection was also
performed for oligometastatic disease with simultaneous resection of
the metastases in selected patients with a good performance status
(WHO ¼ 0). Intraoperative radiotherapy (IORT) was used in patients
with large tumors (maximum 11.5 cm), based on preoperative
ß 2019 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.
Prognostic Neoadjuvant Pancreatic Cancer
assessment by the surgery, radiology, and oncology teams and on
patient choice with informed consent.
Pathological Reporting
A standardized protocol for pathological analysis including
multicolor staining of the different margins of the resected specimen,
axial slicing, and rigorous assessment of the circumferential soft
tissue margins was introduced in January, 2006 at the Institute of
Pathology, University of Heidelberg, Germany.12 This included the
pancreatic transection margin, the bile duct margin, the vascular
margins, the stomach/duodenum margins, and the circumferential
soft tissue margins (medial, anterior surface, superior, posterior). The
R0 and R1 status were classified according to the 8th edition of the
Cancer Staging Manual of the American Joint Committee on Cancer
(AJCC)26 as follows: R0 ¼ tumor-free margin >1 mm (also called
‘‘R0 CRM-’’), and R1 1 mm ¼ tumor-free margin 1 mm (also
called ‘‘R0 CRMþ’’). We also included R1 direct ¼ tumor on
ink ¼ microscopic tumor infiltration at the resection margin, as
recommended by the Union for International Cancer Control (UICC,
8th edition). Tumor histology and staging was also reported accord-
ing to the 8th edition of the UICC TNM classification system.27
Infiltration and resection of the celiac trunk, hepatic artery, SMA,
HPV/SMV, or inferior vena cava were summarized as arterial/venous
infiltration and resection as applicable. Tumor grade was not reported
due to difficulty of interpretation after neoadjuvant treatment.
Outcomes and Follow-up
Patient clinical characteristics including the American Society
of Anesthesiologists (ASA) score, body mass index (BMI), disease-
specific information including preoperative CA 19–9 serum levels,
reasons for primary unresectability, the type and duration of neo-
adjuvant treatment, intraoperative findings, and histopathological
details were extracted from the pancreatic database. Importantly,
none of the patients had obstructive jaundice at the time of measure-
ment of CA 19–9 levels. Postoperative surgical complications
comprised pancreatic fistula, hemorrhage, chyle leak, bile leak,
delayed gastric emptying, intra-abdominal fluid collection, abscess,
surgical site infection, and burst abdomen, according to consensus
definitions of the ISGPS, as appropriate. Postoperative medical
(nonsurgical) complications were defined as those involving the
cardiovascular, pulmonary, hepatic, and renal systems.
Structured postoperative follow-up included cross-sectional
imaging of the abdomen and thorax and evaluation of CA 19–9 and
carcinoembryonic antigen (CEA) serum levels every 3 months for the
first 2 years after operation, and subsequently every 6 months.
Patients were followed up until their most recent oncologic exami-
nation or until death. If patients underwent follow-up at other
institutions, the attending specialists and general practitioners were
contacted for the results. Follow-up data included adjuvant treatment,
time and pattern of tumor recurrence, and the date of and reason
for death.
Statistical Analysis
Quantitative variables are presented as median and dispersion
as interquartile range (IQR) or 95% confidence interval (CI). The
nonparametric Kruskal-Wallis test was used to compare continuous
parameters. For categorical parameters, absolute and relative fre-
quencies were calculated and compared using the chi-square test or
Fisher exact test, as appropriate. Overall survival was defined as the
time from the date of pancreatic resection to either death from any
cause or last follow-up. Disease-free survival was defined as the time
from the date of operation to either disease recurrence or last follow-
up. Patients who died within 90 days or were lost to follow-up within
90 days after surgery were excluded from analysis of disease-free
www.annalsofsurgery.com | 155
Klaiber et al Annals of Surgery  Volume 273, Number 1, January 2021

Assessed for eligibility:

Patients undergoing resection following neoadjuvant treatment
for locally advanced or oligometastatic PDAC between
01/2006 and 02/2017
(n = 320)

Not eligible:
(n = 24)
Periampullary cancer n=9
Anaplastic carcinoma n=2
Resectable tumor at diagnosis n=9
Other synchronous, metastatic cancer n=1
Irreversible electroporation (no resection) n=2
Other reason n=1

Met inclusion criteria:

(n = 296)

Excluded:
(n = 16)
Macroscopic residual tumor (R2) n=4
Indeterminable resection status (Rx) n = 11
Metastases not resected n=1

Included in analysis:
(n = 280)

R0 n = 82
R1 ≤1 mm n = 99
R1 direct n = 99

FIGURE 1. Study profile.

survival. Survival was estimated using the Kaplan-Meier method.
Patients alive at the last follow-up were censored and marked in the
figures (j). Median survival times and 3-year survival rates are
presented. The log-rank test was used to compare survival curves.
Univariable associations among patient characteristics, tumor and
resection characteristics, and overall and disease-free survival were
assessed by proportional hazards regression (Cox model) analysis.
All significant parameters were further examined by multivariable
Cox proportional-hazard regression analysis. Missing data were rare,
so no imputation was performed. To account for potential bias by the
presence of distant metastases, sensitivity analysis excluding patients
with metastatic disease (M1) was performed. Two-sided P values
were used throughout, and significance set at P < 0.05. Data
were analyzed using SAS software (Release 9.4, SAS Institute,
Inc., Cary, NC).
RESULTS
Between January 1, 2006 and February 15, 2017, 296 patients
with initially unresectable PDAC underwent tumor resection after
neoadjuvant treatment. Sixteen (5.4%) patients with macroscopic
(R2) residual tumor (n ¼ 4), unresected residual metastases (n ¼ 1),
or indeterminable (Rx) residual tumor (n ¼ 11) were excluded from
analysis (Fig. 1). The final study population consisted of 280 patients
with 142 (50.7%) men and 138 (49.3%) women. The median (IQR)
age was 61 (53–66) years. R0 was found in 82 (29.3%) patients, R1
1 mm in 99 (35.4%) patients, and R1 direct in 99 (35.4%) patients.
No significant differences were found among the R status groups
with respect to age, sex, BMI, or ASA stage (Table 1). Eleven
patients who were lost to follow-up within 90 days after surgery (5
patients with R1 1 mm and 6 with R1 direct) and 17 patients who
died within 90 days after surgery (4 patients with R0, 5 with R1
1 mm, and 8 with R1 direct) were excluded from disease-free
survival analyses.
Before the start of neoadjuvant treatment, unresectability was
evaluated on the basis of cross-sectional imaging findings in 178
(63.6%) patients and surgical exploration in 102 (36.4%) patients.
At initial diagnosis, the cancer was borderline resectable in 18
(6.4%) patients, locally advanced in 190 (67.9%) patients, and 72
(25.7%) patients had oligometastatic disease. The types of neo-
adjuvant chemotherapy were not significantly different among
patients with R0, R1 1 mm, and R1 direct resections. Neoadjuvant
chemoradiotherapy was performed in 125 (44.6%) out of 280
patients and was also not significantly different between the 3
resection margin groups. The duration of the neoadjuvant treatment

156 | www.annalsofsurgery.com ß 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.

Annals of Surgery  Volume 273, Number 1, January 2021 Prognostic Neoadjuvant Pancreatic Cancer

TABLE 1. Patient Characteristics

Parameter R0 (n ¼ 82) R1 1 mm (n ¼ 99) R1 Direct (n ¼ 99) P
Age, yrs 61 (54-66) 58 (52–66) 62 (54–68) 0.3524
Sex 0.1740
Male 35 (42.7) 51 (51.5) 56 (56.6)
Female 47 (57.3) 48 (48.5) 43 (43.4)
BMI, kg/m2 (8 missing values) 22.9 (20.4–25.8) 24.1 (21.6–26.5) 23.7 (21.5–25.3) 0.2094
ASA classification (4 missing values) 0.6348
I 2 (2.5) 2 (2.0) 2 (2.1)
II 45 (56.3) 53 (53.5) 46 (47.4)
III 32 (40.0) 44 (44.4) 49 (50.5)
IV 1 (1.3) 0 0
Reason for primary unresectability (3 missing values) 0.1468
Locally unresectable 53 (65.4) 73 (75.3) 79 (79.8)
Locally unresectable and distant metastases 5 (6.2) 5 (5.2) 7 (7.1)
Distant metastases 23 (28.4) 19 (19.6) 13 (13.1)
Reason for local unresectability 0.1268
Artery involvement 31 (37.8) 39 (39.4) 38 (38.4)
Artery plus vein involvement 16 (19.5) 24 (24.2) 33 (33.3)
Vein involvement 9 (11.0) 14 (14.1) 15 (15.2)
Unknown 3 (3.7) 3 (3.0) 0
Neoadjuvant chemotherapy (1 missing value) 0.9846
FOLFIRINOX (only) 27 (32.9) 34 (34.7) 31 (31.3)
Gemcitabine alone 26 (31.7) 30 (30.6) 29 (29.3)
Gemcitabine and paclitaxel 5 (6.1) 4 (4.1) 5 (5.1)
Others/multiple 24 (29.3) 30 (30.6) 34 (34.3)
FOLFIRINOX as part of any chemotherapy regime (1 missing value) 36 (43.9) 43 (43.9) 43 (43.4) 0.9973
Neoadjuvant chemoradiotherapy 39 (47.6) 42 (42.4) 44 (44.4) 0.7861
Neoadjuvant treatment period, days 165.5 (123–249) 148.0 (112–224) 151.0 (116–218) 0.2473
Preoperative CA 19–9, U/mL (4 missing values) 23.9 (11.8–50.5) 49.5 (14.8–240.0) 68.4 (20.9–270.7) 0.0005
Preoperative CEA, g/L (7 missing values) 1.6 (0.9–2.6) 2.1 (1.1–4.0) 2.4 (1.3–4.1) 0.0277
Surgical procedure 0.0150
Distal pancreatectomy 24 (29.3) 22 (22.2) 21 (21.2)
Partial pancreatoduodenectomy 43 (52.4) 42 (42.4) 36 (36.4)
Total pancreatectomy 15 (18.3) 35 (35.4) 42 (42.4)
Vascular resection 28 (34.2) 57 (57.6) 64 (64.7) 0.0001
Intraoperative radiotherapy (IORT) 17 (20.7) 19 (19.2) 25 (25.3) 0.5648
Patients with 1 surgical complication 43 (52.4) 51 (51.5) 50 (50.5) 0.9668
Patients with 1 medical complication 23 (28.1) 26 (26.3) 29 (29.3) 0.8921
30-d mortality 1 (1.2) 3 (3.0) 2 (2.0) 0.8768
90-d mortality 4 (4.9) 5 (5.1) 8 (8.1) 0.6978
Adjuvant chemotherapy given (out of 252 patients) 41/74 (55.4) 51/89 (57.3) 53/89 (59.6) 0.8746
Data are medians (with interquartile ranges) or numbers of patients (with percentages).

Celiac axis, hepatic artery, superior mesenteric artery/vein, portal vein, inferior vena cava.

period did not significantly differ among the groups, with a median
(IQR) of 148.0 (112–224) days in the R1 1 mm group, 151.0 (116–
218) days in the R1 direct group, and 165.5 (123–249) days in the
R0 group.
Preoperative serum CA 19–9 and CEA levels differed signifi-
cantly among the 3 groups, with the lowest values in the R0 group
(Table 1). All patients underwent oncological pancreatic resection.
Depending on the location and extend of the tumor, partial pan-
creatoduodenectomy (n ¼ 121), distal pancreatectomy (n ¼ 67), or
total pancreatectomy (n ¼ 92) was performed, with proportionally
more total pancreatectomies associated with R1 1 mm (35.4%) and
R1 direct (42.4%) resections than with R0 (18.3%) resections. This
tended to mirror the proportion of vascular resections, which were
57.6% in R1 1 mm and 64.7% in R1 direct resections, compared
with 34.2% in R0 resections. Of 280 patients, 111 (39.6%) underwent
only vein resection, 7 of 280 (2.5%) underwent only artery resection,
and 31 of 280 (11.1%) underwent artery and vein resection com-
bined. In all, 144 patients were deemed to have unresectable tumors
preoperatively due to artery involvement, but did not receive an
arterial resection despite complete resection of the cancer. The R
status in this patient subgroup was R0 in 43 of 144 patients (29.9%),
R1 1 mm in 46 patients (31.9%), and R1 direct in 55 patients
(38.2%). There were 45 (16.1%) patients with oligometastatic dis-
ease at the time of surgery who underwent resection of metastases
along with the primary, involving the peritoneum in 13 patients, the
liver in 27, distant lymph nodes in 4 patients, and the lung, spleen,
adrenal gland, and os pubis in 1 patient each. The distribution of
oligometastatic disease did not differ among the 3 resection margin
groups (Table 2). IORT was administered to 61 (21.8%) patients, but
was not associated with significant differences among the resection
margin groups. Postoperative surgical and medical morbidity was
comparable among the 3 resection margin groups (Table 1).
There was no significant difference in tumor location among
the 3 resection margin groups, but R0 tumors were associated with
smaller maximum tumor diameter and lower ypT stages than those in
the positive resection margin groups (Table 2). A complete patho-
logical response (ypT0) was found in 14 (17.3%) patients of the R0
group, but in no patient in the R1 1 mm or R1 direct group. The
median (IQR) number of lymph nodes analyzed was 24 (16–34) and
was comparable among the resection margin groups. R0 margin

ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 157

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.

Klaiber et al Annals of Surgery  Volume 273, Number 1, January 2021

TABLE 2. Histopathological Findings

Parameter R0 (n ¼ 82) R1 1 mm (n ¼ 99) R1 Direct (n ¼ 99) P
Tumor localization in pancreas 0.7479
Body 12 (14.6) 10 (10.1) 14 (14.1)
Body and tail 5 (6.1) 7 (7.1) 5 (5.1)
Head 50 (61.0) 54 (54.6) 56 (56.6)
Head and body 3 (3.7) 9 (9.1) 10 (10.1)
Head and body and tail 3 (3.7) 8 (8.1) 7 (7.1)
Tail 9 (11.0) 11 (11.1) 7 (7.1)
Maximum tumor diameter, cm (1 missing value) 2.5 (1.5–3.5) 3.5 (2.5–4.5) 3.5 (2.5–4.5) <0.0001
ypT stage (1 missing value) <0.0001
ypT0 14 (17.3) 0 (0) 0 (0)
ypT1 15 (18.5) 11 (11.1) 11 (11.1)
ypT2 33 (40.7) 50 (50.5) 49 (49.5)
ypT3 15 (18.5) 21 (21.2) 30 (30.3)
ypT4 4 (4.9) 17 (17.2) 9 (9.1)
Lymph node involvement <0.0001
ypN0 65 (79.3) 35 (35.4) 36 (36.4)
ypN1 15 (18.3) 37 (37.4) 38 (38.4)
ypN2 2 (2.4) 27 (27.3) 25 (25.3)
Number of examined lymph nodes 21 (14–30) 26 (17–37) 24 (18–34) 0.0678
Number of positive lymph nodes 1 (1–2) 3 (1–6) 3 (1–5) 0.0406
ypM stage (3 missing values) 0.4714
ypM0 68 (84.0) 78 (80.4) 86 (86.9)
ypM1 13 (16.1) 19 (19.6) 13 (13.1)
Arterial/venous infiltration 19 (23.2) 47 (47.5) 52 (52.5) <0.0001
Data are medians (with interquartile ranges) or numbers of patients (with percentages).

Celiac trunk, hepatic artery, superior mesenteric artery/vein, portal vein, inferior vena cava.

tumors, however, were associated with fewer involved lymph nodes
and lower ypN stages than positive resection margin tumors. There
was no lymph node involvement (ypN0) associated with 65 (79.3%)
R0 tumors, compared with 35 (35.4%) R1 1 mm and 36 (36.4%) R1
direct resections. There was infiltration of the celiac trunk, hepatic
artery, SMA, SMV, HPV, or inferior vena cava in 19 (23.2%) patients
with R0 resections, 47 (47.5%) patients with R1 1 mm resections,
and 52 (52.5%) patients with R1 direct resections.
After pancreatic resection, 145 (57.5%) of 252 patients had
adjuvant chemotherapy: 41 (55.4%) of 74 patients in the R0 group,
51 (57.3%) of 89 patients in the R1 1 mm group, and 53 (59.6%) of
89 patients in the R1 direct group. Median (95% CI) overall survival
without adjuvant chemotherapy was 12.8 (8.8–20.6) months, com-
pared with 24.1 (19.8–29.6) months with adjuvant chemotherapy,
and the 3-year survival rates (95% CI) were 20.0% (11.4–30.3)
versus 31.9% (23.7–40.3) (P ¼ 0.0225). Median (95% CI) disease-
free survival without adjuvant chemotherapy was 6.2 (4.6–7.6)
months, compared with 11.5 (9.7–13.6) months with adjuvant
chemotherapy, and the 3-year survival rates (95% CI) were 13.0%
(6.4–21.9) versus 12.0% (7.0–18.5) (P ¼ 0.0071) (Supplementary
Fig. S2, http://links.lww.com/SLA/B623). There were no significant
differences in 30-day and 90-day mortality between the 3 resection
margin groups (Table 1). After a median (IQR) follow-up of 18 (11–
36) months, 76 (27.1%) of 280 patients were still alive. The median
(95% CI) overall survival of all 280 patients was 19.0 (14.6–22.3)
months, with a 3-year overall survival rate (95% CI) of 24.7% (19.1–
30.6). The median (95% CI) overall survival was 25.1 (16.2–30.0)
months for the R0 group, 15.3 (11.0–23.5) months for patients with
R1 1 mm tumors, and 16.1 (9.8–22.3) months for the R1 direct
group. The 3-year overall survival rates (95% CI) were 35.0% (24.3–
45.8), 20.7% (12.0–31.0), and 18.5% (10.5–28.4), respectively (P ¼
0.0076) (Fig. 2A). The median (95% CI) disease-free survival of all
252 patients included in this analysis was 9.0 (7.9–10.3) months,
with a 3-year disease-free survival rate (95% CI) of 11.6%
(7.7–16.4). The median (95% CI) disease-free survival was 9.8
(7.6–16.9) months for patients with R0 tumors, 8.9 (7.4–10.5)
months for patients with R1 1 mm tumors, and 8.3 (6.2–10.4)
months for patients with R1 direct tumors. The 3-year disease-free
survival rates (95% CI) were 19.6% (11.3–29.6), 9.8% (4.1–18.6),
and 5.9% (2.0–13.0), respectively (P ¼ 0.0250) (Fig. 2B).
R1 1 mm and R1 direct resections were significant negative
prognostic factors for overall and disease-free survival in univariable
analysis, but not in multivariable analysis (Tables 3 and 4). Preoper-
ative serum CA 19–9 levels 100 U/mL, lymph node involvement,
ypM1 stage, and vascular infiltration were each independently
associated with a poor prognosis for overall survival in multivariable
analysis (Table 3, Fig. 3). Preoperative serum CA 19–9 and CEA
levels, ypT4 stage, lymph node involvement, distant metastases at
initial diagnosis, and multiple tumor localizations within the pan-
creas were associated with reduced disease-free survival (Table 4,
Supplementary Fig. S1, http://links.lww.com/SLA/B623). Sensitiv-
ity analysis excluding patients with M1 disease did not substantially
alter the results for independent prognostic factors of overall and
disease-free survival (Supplementary Table S1, http://links.lww.com/
SLA/B623).
DISCUSSION
Neoadjuvant treatment may enable a higher proportion of
patients with initially unresectable tumors to have a successful
resection.4,28–30 R0 resections with a clearance of >1 mm from
the resection margin were achieved in 29% of this group of patients,
which is comparable with the rate for patients with resectable PDAC
and surgery upfront.14–16,18 Recent studies suggest that an even
greater rate of R0 resection—80% or more—might be achieved
after neoadjuvant treatment.5,31,32 R0 resections may have been
overestimated in some of these studies, where stratification into
R0 >1 mm and R1 1 mm may not have been performed. Stratifi-
cation of R status based on margin clearance R0 >1 mm and margin

158 | www.annalsofsurgery.com ß 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.

Annals of Surgery  Volume 273, Number 1, January 2021 Prognostic Neoadjuvant Pancreatic Cancer

FIGURE 2. Survival after resection. (A) Overall survival; (B) disease-free survival. Survival data are calculated starting from the date of
the operation. To estimate survival from the beginning of neoadjuvant treatment, a median of 5.1 months has to be added. Patients
alive at the last follow-up are censored (I).

involvement based on R1 1 mm and R1 direct has been shown to significantly lower preoperative serum CA 19–9 and CEA levels,
have a clear impact on survival not only in the adjuvant set- smaller tumor size, lower ypT and ypN stages, and less extended
ting,14,15,17–20 but also in the neoadjuvant setting, as confirmed in resections, which may be attributable to a good response of tumors to
the present study. In this study, R0 resections were associated with the neoadjuvant treatment.31

TABLE 3. Univariable and Multivariable Cox Regression Analysis of Prognostic Factors for Overall Survival
Variables Category Hazard Ratio 95% CI P
Univariable analysis
Sex Male vs female 0.93 0.70–1.22 0.5893
Age, yrs 70 vs <70 1.43 0.99–2.06 0.0577
BMI, kg/m2 30 vs <30 0.71 0.40–1.25 0.2317
ASA classification III–IV vs I–II 1.18 0.90–1.56 0.2376
Neoadjuvant treatment FOLFIRINOX vs no FOLFIRINOX 1.03 0.77–1.37 0.8672
Reason for primary unresectability Distant metastases vs local tumor spread 1.13 0.83–1.56 0.4360
Basis for diagnosis before neoadjuvant treatment Surgical exploration vs radiology 0.93 0.70–1.24 0.6369
Neoadjuvant chemoradiotherapy Yes vs no 1.02 0.77–1.34 0.9071
Preoperative CA 19–9 level, U/mL 37 to <100 vs <37 1.10 0.76–1.59 0.6109
100 vs <37 2.16 1.56–2.99 <0.0001
Preoperative CEA level, g/L 2.5 vs <2.5 1.66 1.25–2.22 0.0005
Type of surgery TP vs DP 1.57 1.08–2.28 0.0183
PD vs DP 0.94 0.66–1.34 0.7318
Localization of tumor Body vs head 0.92 0.59–1.42 0.7034
Tail vs head 1.01 0.61–1.65 0.9800
Multiple segmentsy vs head only 1.45 1.02–2.07 0.0384
T stage ypT2 vs ypT0/1 1.67 1.12–2.49 0.0117
ypT3 vs ypT0/1 2.33 1.50–3.62 0.0002
ypT4 vs ypT0/1 3.44 2.02–5.87 <0.0001
N stage ypN1 vs ypN0 1.98 1.45–2.70 <0.0001
ypN2 vs ypN0 2.79 1.89–4.12 <0.0001
M category ypM1 vs. ypM0 1.88 1.30–2.70 0.0007
R status R1 1 mm vs R0 1.48 1.05–2.10 0.0267
R1 direct vs R0 1.70 1.21–2.40 0.0025
Arterial/venous infiltration Yes vs no 1.97 1.49–2.61 <0.0001
Multivariable analysis
Preoperative CA 19–9 level, U/mL 100 vs <100 1.58 1.15–2.17 0.0046
N stage ypN1 vs ypN0 1.72 1.23–2.40 0.0013
ypN2 vs ypN0 2.25 1.51–3.36 <0.0001
M category ypM1 vs ypM0 2.04 1.39–3.01 0.0003
Arterial/venous infiltration Yes vs no 1.87 1.39–2.52 <0.0001

Celiac trunk, hepatic artery, superior mesenteric artery/vein, portal vein, inferior vena cava.
yMultiple segments of pancreas incl. pancreatic body.
DP indicates distal pancreatectomy; PD, partial pancreatoduodenectomy; TP, total pancreatectomy.

ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 159

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.

Klaiber et al Annals of Surgery  Volume 273, Number 1, January 2021

TABLE 4. Univariable and Multivariable Cox Regression Analysis of Prognostic Factors for Disease-free Survival
Variables Category Hazard Ratio 95% CI P
Univariable analysis
Sex Male vs female 0.89 0.68–1.17 0.3957
Age, yrs 70 vs <70 1.20 0.84–1.73 0.3212
BMI, kg/m2 30 vs <30 0.83 0.50–1.38 0.4703
ASA classification III–IV vs I–II 0.95 0.72–1.26 0.7350
Neoadjuvant treatment FOLFIRINOX vs no FOLFIRINOX 1.36 1.03–1.79 0.0287
Reason for primary unresectability Distant metastases vs local unresectability 1.69 1.25–2.30 0.0007
Basis for diagnosis before neoadjuvant treatment Surgical exploration vs radiology 0.86 0.65–1.14 0.2891
Neoadjuvant radiochemotherapy Yes vs no 0.71 0.54–0.94 0.0174
Preoperative CA 19–9 level, U/mL 37 to <100 vs <37 1.38 0.97–1.97 0.0714
100 vs <37 2.25 1.62–3.10 <0.0001
Preoperative CEA level, g/L 2.5 vs <2.5 1.67 1.26–2.22 0.0004
Type of surgery TP vs DP 1.01 0.71–1.45 0.9534
PD vs DP 0.69 0.50–0.97 0.0328
Localization of tumor Body vs head 1.50 1.01–2.23 0.0471
Tail vs head 1.25 0.79–1.97 0.3363
Multipley segments vs head only 1.56 1.09–2.24 0.0155
T stage ypT2 vs ypT0/1 1.67 1.14–2.45 0.0088
ypT3 vs ypT0/1 1.91 1.24–2.96 0.0037
ypT4 vs ypT0/1 2.95 1.67–5.22 0.0002
N stage ypN1 vs ypN0 1.70 1.24–2.34 0.0009
ypN2 vs ypN0 2.31 1.59–3.36 <0.0001
M category ypM1 vs ypM0 2.28 1.60–3.24 <0.0001
R status R1 1 mm vs R0 1.45 1.03–2.04 0.0356
R1 direct vs R0 1.56 1.11–2.18 0.0104
Arterial/venous infiltration Yes vs no 1.39 1.05–1.83 0.0196
Multivariable analysis
Preoperative CA 19–9 level, U/mL 100 vs <37 1.80 1.26–2.56 0.0011
37 to <100 vs <37 1.38 0.95–1.98 0.0875
Preoperative CEA level, g/L 2.5 vs <2.5 1.45 1.06–1.98 0.0213
N stage ypN1 vs ypN0 1.43 1.02–2.01 0.0377
ypN2 vs ypN0 2.02 1.36–3.01 0.0005
Reason for primary unresectability Distant metastases vs local tumor spread 1.85 1.35–2.54 0.0002
Localization of the tumor Multipley segments vs single 1.70 1.26–2.31 0.0006
T stage ypT4 vs ypT0/1/2/3 1.79 1.06–3.02 0.0290

Celiac trunk, hepatic artery, superior mesenteric artery/vein, portal vein, inferior vena cava.
yMultiple segments of pancreas incl. pancreatic body.
DP indicates distal pancreatectomy; PD, partial pancreatoduodenectomy; TP, total pancreatectomy.

With a median overall survival of 19.0 months and a 3-year
overall survival rate of 24.7% after operation, survival was consid-
erably longer in the present study than in patients undergoing
exploration alone.4,16,33,34 Taking into consideration the median
treatment period of 5.1 months for neoadjuvant therapy before
surgery, overall survival from the time of diagnosis was even longer.
Both disease-free survival rates and overall survival rates were more
favorable after R0 resection than after R1 1 mm or R1 direct
resection in this study. Moreover, our data support the emerging
inclusion of M1 resections in selected patients.35– 37 Despite the more
challenging surgery in this series, the overall surgical morbidity, 30-
day mortality, and 90-day mortality were acceptable, at 51.4%, 2.1%,
and 6.1%, respectively.
Preoperative serum CA 19 – 9 levels, lymph node involve-
ment, presence of distant metastases, and arterial/venous infiltra-
tion of major vessels were all independent overall survival
predictors. For disease-free survival the independent determinants
were preoperative CA 19 – 9 levels, preoperative CEA levels, lymph
node involvement, initial unresectability due to distant metastases,
tumor involvement of multiple pancreatic segments, and higher
ypT stage. These findings are consistent with previous studies
showing that lower preoperative CA 19 – 9 levels are associated
with longer overall and disease-free survival in both the
neoadjuvant and the adjuvant setting.30,38 – 41 As in upfront surgery,
the extent of lymph node involvement was a strong prognostic
factor.6,42 – 45 A French prospective multicenter study including 147
patients recently indicated that the predictive survival value of
lymph node involvement may outweigh the potential survival
impact of the R status.46 Positive lymph nodes were not indepen-
dently associated with survival in a recent study from the Massa-
chusetts General Hospital including 110 patients with borderline
resectable and locally advanced PDAC undergoing resection after
neoadjuvant FOLFIRINOX.30 The discrepancy between the results
of the 2 studies and our study may be explained by differences in
study populations and in modeling, especially regarding the extent
of detailed pathological evaluation. In our study, we included
patients with oligometastatic disease, but there were comparable
proportions of M1 in the R0, R1 1 mm, and R1 direct tumor
resection groups; in fact, there were fewer patients with M1 disease
in the R1 direct group. In support of this notion, sensitivity analysis
excluding patients with M1 disease did not substantially alter the
results concerning independent prognostic factors of overall and
disease-free survival. Pancreatic cancer is now regarded as a
systemic disease from a very early stage, but from a clinical
standpoint, overt metastases at presentation tend to be associated
with more advanced tumor characteristics.47,48

160 | www.annalsofsurgery.com ß 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.

Annals of Surgery  Volume 273, Number 1, January 2021 Prognostic Neoadjuvant Pancreatic Cancer

FIGURE 3. Prognostic factors for overall postoperative survival. (A) Influence of preoperative serum CA 19–9 levels; (B) influence of
lymph node status; (C) influence of M category; (D) influence of vascular involvement (ie, celiac trunk, hepatic artery, superior
mesenteric artery/vein, portal vein, inferior vena cava). Survival data are calculated starting from the date of the operation. To
estimate survival from the beginning of neoadjuvant treatment, a median of 5.1 months has to be added. Patients alive at the last
follow-up are censored (I).

One of the limitations of this study is the varied use of
neoadjuvant treatments, pointing to the need for randomized con-
trolled trials addressing this issue. In addition, a considerable number
of patients were evaluated as having unresectable tumors at other
centers and were referred to our department only after the com-
mencement of neoadjuvant treatment at the original site. However,
the study population represents the daily practice observed in
national and international centers, so external validity of the study
results is given.
CONCLUSIONS
In summary, this single-center study presents the largest
analysis to date of the prognostic impact of the R status and other
potentially predictive factors in patients with initially unresectable
pancreatic cancer after neoadjuvant treatment and resection. The data
provide strong evidence that survival rates after R0, R1 1 mm, and
R1 direct resections are better than those observed after exploration
without resection. Although R status is important, with pragmatic
differences between R0, R1 1 mm, and R1 direct, this factor is not
an independent determinant of survival in this setting. Preoperative
CA 19–9 levels, lymph node involvement, presence of distant
metastases, and arterial/venous infiltration of major vessels were
all found to be independently associated with overall survival after
neoadjuvant therapy and should be taken into consideration when
planning prospective clinical trials and interstudy comparisons.
REFERENCES
1. American Cancer Society. Cancer Facts and Figures 2019. Atlanta: American
Cancer Society; 2019.
2. Rahib L, Smith BD, Aizenberg R, et al. Projecting cancer incidence and deaths
to 2030: the unexpected burden of thyroid, liver, and pancreas cancers in the
United States. Cancer Res. 2014;74:2913–2921.
3. Bockhorn M, Uzunoglu FG, Adham M, et al. Borderline resectable pancreatic
cancer: a consensus statement by the International Study Group of Pancreatic
Surgery (ISGPS). Surgery. 2014;155:977–988.
4. Hackert T, Sachsenmaier M, Hinz U, et al. Locally advanced pancreatic
cancer: neoadjuvant therapy with folfirinox results in resectability in 60% of
the patients. Ann Surg. 2016;264:457–463.
5. Murphy JE, Wo JY, Ryan DP, et al. Total neoadjuvant therapy with FOLFIR-
INOX followed by individualized chemoradiotherapy for borderline resect-
able pancreatic adenocarcinoma: a phase 2 clinical trial. JAMA Oncol.
2018;4:963–969.

ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 161

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.

Klaiber et al
6. Epstein JD, Kozak G, Fong ZV, et al. Microscopic lymphovascular invasion is
an independent predictor of survival in resected pancreatic ductal adenocar-
cinoma. J Surg Oncol. 2017;116:658–664.
7. Winter JM, Brennan MF, Tang LH, et al. Survival after resection of pancreatic
adenocarcinoma: results from a single institution over three decades. Ann Surg
Oncol. 2012;19:169–175.
8. Nitschke P, Volk A, Welsch T, et al. Impact of intraoperative re-resection to
achieve R0 status on survival in patients with pancreatic cancer: a single-
center experience with 483 patients. Ann Surg. 2017;265:1219–1225.
9. Wagner M, Redaelli C, Lietz M, et al. Curative resection is the single most
important factor determining outcome in patients with pancreatic adenocarci-
noma. Br J Surg. 2004;91:586–594.
10. Butturini G, Stocken DD, Wente MN, et al. Influence of resection margins and
treatment on survival in patients with pancreatic cancer: meta-analysis of
randomized controlled trials. Arch Surg. 2008;143:75–83.
11. Demir IE, Jager C, Schlitter AM, et al. R0 versus R1 resection matters after
pancreaticoduodenectomy, and less after distal or total pancreatectomy for
pancreatic cancer. Ann Surg. 2018;268:1058–1068.
12. Esposito I, Kleeff J, Bergmann F, et al. Most pancreatic cancer resections are
R1 resections. Ann Surg Oncol. 2008;15:1651–1660.
13. Verbeke CS, Leitch D, Menon KV, et al. Redefining the R1 resection in
pancreatic cancer. Br J Surg. 2006;93:1232–1237.
14. Hank T, Hinz U, Tarantino I, et al. Validation of at least 1 mm as cut-off for
resection margins for pancreatic adenocarcinoma of the body and tail. Br J
Surg. 2018;105:1171–1181.
15. Strobel O, Hank T, Hinz U, et al. Pancreatic cancer surgery: the new R-status
counts. Ann Surg. 2017;265:565–573.
16. Konstantinidis IT, Warshaw AL, Allen JN, et al. Pancreatic ductal adenocar-
cinoma: is there a survival difference for R1 resections versus locally advanced
unresectable tumors? What is a ‘‘true’’ R0 resection? Ann Surg.
2013;257:731–736.
17. van Roessel S, Kasumova GG, Tabatabaie O, et al. Pathological margin
clearance and survival after pancreaticoduodenectomy in a US and European
pancreatic center. Ann Surg Oncol. 2018;25:1760–1767.
18. Ghaneh P, Kleeff J, Halloran CM, et al. The impact of positive resection
margins on survival and recurrence following resection and adjuvant chemo-
therapy for pancreatic ductal adenocarcinoma. Ann Surg October. 2017 [Epub
ahead of print].
19. Ocuin LM, Miller-Ocuin JL, Zenati MS, et al. A margin distance analysis of
the impact of adjuvant chemoradiation on survival after pancreatoduodenec-
tomy for pancreatic adenocarcinoma. J Gastrointest Oncol. 2017;8:696–704.
20. Neoptolemos JP, Palmer DH, Ghaneh P, et al. Comparison of adjuvant
gemcitabine and capecitabine with gemcitabine monotherapy in patients with
resected pancreatic cancer (ESPAC-4): a multicentre, open-label, randomised,
phase 3 trial. Lancet. 2017;389:1011–1024.
21. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of
Observational Studies in Epidemiology (STROBE) statement: guidelines for
reporting observational studies. Lancet. 2007;370:1453–1457.
22. Gillen S, Schuster T, Meyer zum Büschenfelde C, et al. Preoperative/neo-
adjuvant therapy in pancreatic cancer: a systematic review and meta-analysis
of response and resection percentages. PLoS Med. 2010;7:e1000267.
23. Schwartz LH, Litiere S, de Vries E, et al. RECIST 1.1-Update and clarifica-
tion: from the RECIST committee. Eur J Cancer. 2016;62:132–137.
24. Weitz J, Rahbari N, Koch M, et al. The ‘‘artery first’’ approach for resection of
pancreatic head cancer. J Am Coll Surg. 2010;210:e1–e4.
25. Hackert T, Strobel O, Michalski CW, et al. The TRIANGLE operation: radical
surgery after neoadjuvant treatment for advanced pancreatic cancer: a single
arm observational study. HPB (Oxford). 2017;19:1001–1007.
26. Amin MB, Edge SB, Greene F, et al., eds. AJCC Cancer Staging Manual. 8th
edn., New York: Springer; 2017.
27. Brierley JD, Gospodarowicz MK, Wittekind C, eds. TNM Classification of
Malignant Tumours. 8th edn., Oxford: Wiley Blackwell; 2017.
162 | www.annalsofsurgery.com
Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery  Volume 273, Number 1, January 2021
28. Rombouts SJ, Walma MS, Vogel JA, et al. Systematic review of resection rates
and clinical outcomes after FOLFIRINOX-based treatment in patients with
locally advanced pancreatic cancer. Ann Surg Oncol. 2016;23:4352–4360.
29. Wo JY, Niemierko A, Ryan DP, et al. Tolerability and long-term outcomes of
dose-painted neoadjuvant chemoradiation to regions of vessel involvement in
borderline or locally advanced pancreatic cancer. Am J Clin Oncol.
2018;41:656–661.
30. Michelakos T, Pergolini I, Castillo CF, et al. Predictors of resectability and
survival in patients with borderline and locally advanced pancreatic cancer
who underwent neoadjuvant treatment with FOLFIRINOX. Ann Surg.
2018;268:e95–e96.
31. Schorn S, Demir IE, Reyes CM, et al. The impact of neoadjuvant therapy on
the histopathological features of pancreatic ductal adenocarcinoma: a system-
atic review and meta-analysis. Cancer Treat Rev. 2017;55:96–106.
32. Chatzizacharias NA, Tsai S, Griffin M, et al. Locally advanced pancreas
cancer: staging and goals of therapy. Surgery. 2018;163:1053–1062.
33. Strobel O, Berens V, Hinz U, et al. Resection after neoadjuvant therapy
for locally advanced, ‘‘unresectable’’ pancreatic cancer. Surgery. 2012;152:
S33–42.
34. van der Geest LGM, Lemmens V, de Hingh I, et al. Nationwide outcomes in
patients undergoing surgical exploration without resection for pancreatic
cancer. Br J Surg. 2017;104:1568–1577.
35. Hackert T, Niesen W, Hinz U, et al. Radical surgery of oligometastatic
pancreatic cancer. Eur J Surg Oncol. 2017;43:358–363.
36. Shrikhande SV, Kleeff J, Reiser C, et al. Pancreatic resection for M1 pancreatic
ductal adenocarcinoma. Ann Surg Oncol. 2007;14:118–127.
37. Tachezy M, Gebauer F, Janot M, et al. Synchronous resections of hepatic
oligometastatic pancreatic cancer: disputing a principle in a time of safe
pancreatic operations in a retrospective multicenter analysis. Surgery.
2016;160:136–144.
38. Hartwig W, Strobel O, Hinz U, et al. CA19-9 in potentially resectable
pancreatic cancer: perspective to adjust surgical and perioperative therapy.
Ann Surg Oncol. 2013;20:2188–2196.
39. Hata S, Sakamoto Y, Yamamoto Y, et al. Prognostic impact of postoperative
serum CA 19-9 levels in patients with resectable pancreatic cancer. Ann Surg
Oncol. 2012;19:636–641.
40. Boone BA, Steve J, Zenati MS, et al. Serum CA 19-9 response to neoadjuvant
therapy is associated with outcome in pancreatic adenocarcinoma. Ann Surg
Oncol. 2014;21:4351–4358.
41. Combs SE, Habermehl D, Kessel KA, et al. Prognostic impact of CA 19-9 on
outcome after neoadjuvant chemoradiation in patients with locally advanced
pancreatic cancer. Ann Surg Oncol. 2014;21:2801–2807.
42. Pawlik TM, Gleisner AL, Cameron JL, et al. Prognostic relevance of lymph
node ratio following pancreaticoduodenectomy for pancreatic cancer. Surgery.
2007;141:610–618.
43. Ferrone CR, Pieretti-Vanmarcke R, Bloom JP, et al. Pancreatic ductal adeno-
carcinoma: long-term survival does not equal cure. Surgery. 2012;152:S43–49.
44. Strobel O, Hinz U, Gluth A, et al. Pancreatic adenocarcinoma: number of
positive nodes allows to distinguish several N categories. Ann Surg.
2015;261:961–969.
45. Malleo G, Maggino L, Ferrone CR, et al. Number of examined lymph nodes
and nodal status assessment in distal pancreatectomy for body/tail ductal
adenocarcinoma. Ann Surg. 2018. doi: 10.1097/SLA.0000000000002781
[Epub ahead of print].
46. Delpero JR, Jeune F, Bachellier P, et al. Prognostic value of resection margin
involvement after pancreaticoduodenectomy for ductal adenocarcinoma:
updates from a French prospective multicenter study. Ann Surg.
2017;266:787–796.
47. O’Dwyer PJ. Locally advanced pancreatic cancer: maybe not so local. Lancet
Oncol. 2016;17:694–695.
48. Tuveson DA, Neoptolemos JP. Understanding metastasis in pancreatic cancer:
a call for new clinical approaches. Cell. 2012;148:21–23.
ß 2019 Wolters Kluwer Health, Inc. All rights reserved.