Journal of Geriatric Oncology 9 (2018) 373–381

Contents lists available at ScienceDirect

Journal of Geriatric Oncology

An onco-geriatric approach to select older patients for optimal

treatments of pancreatic adenocarcinoma
Elisabeth Castel-Kremer a,⁎,1, Solene De Talhouet b,1, Anne-Laure Charlois c, Emmanuelle Graillot b,
Xavier Chopin-Laly d, Mustapha Adham d, Brigitte Comte a, Catherine Lombard-Bohas b,
Thomas Walter b, Gilles Boschetti b,c
a
Department of Geriatrics, Hospices Civils de Lyon and University Claude Bernard Lyon 1, France
b
Department of Digestive Oncology, Edouard Herriot University Hospital, Lyon, France
c
Department of Gastroenterology, Lyon-Sud Hospital, Lyon, France
d
Department of Digestive Surgery, Edouard Herriot University Hospital, Lyon, France

a r t i c l e i n f o a b s t r a c t

Article history: Introduction: Pancreatic adenocarcinoma affects mainly older patients. Surgery is indicated for localized tumors
Received 10 January 2017 while chemotherapy alone is proposed in advanced or metastatic tumors.
Received in revised form 23 January 2018 Objective: To evaluate the feasibility of standard of care oncologic treatments in this population, the accuracy of
Accepted 13 March 2018 the geriatric evaluation to predict the ability of patients to tolerate the recommended treatments and to identify
Available online 9 June 2018
specific geriatric prognosis factors.
Methods: We included, between 2007 and 2014, all consecutive patients over 70 years of age with a pathologi-
Keywords:
Pancreatic cancer
cally diagnosed pancreatic cancer. The patients underwent a comprehensive geriatric assessment before
Elderly therapeutic decision in a multidisciplinary team meeting. We analyzed factors independently associated with
Geriatric assessment all-cause mortality with Cox survival analysis.
Surgery Results: Seventy-three patients (median age = 77.9 years) were prospectively included. Among them, 42 patients
Prognostic factors underwent surgery whereas the 31 other patients not eligible for surgical treatment received chemotherapy (n =
22) or best supportive care alone (n = 9). Almost 62% of operated patients received adjuvant chemotherapy. In the
non-surgical group, a mean of 9 cycles of palliative chemotherapy per patients were administrated. Median overall
survival was 21.3 months in the surgical group and 6.1 months in the palliative group (p = 0.0001). Most of onco-
logic parameters were found to be independent survival predictors. Age was not associated with the survival, but a
significant impact of Lawton's Instrumental Activities of Daily Living (IADL) impairment (IADLb4) (HR = 5.0, p =
0.047), Cumulative Index Rating Scale-Geriatric (CIRS-G) ≥2 (HR = 19, p = 0.035) and weight loss N10% (HR =
4.6, p = 0.03) on survival was detected. Surgery was the only factor independently predictive of survival in mul-
tivariate analysis (p b 0.001).
Conclusion: Almost 90% of selected older pancreatic patients with cancer (64 out of 73 patients) may benefit from
the same standard treatments as younger patients. IADL impairment of patients, CIRS-G ≥2, and weight loss N10%
constitute survival prognostic factors which should be added to the oncological criteria in the therapeutic decision-
making process.
© 2018 Elsevier Ltd. All rights reserved.

1. Introduction
The incidence of pancreatic adenocarcinoma varies between 7 and
11/100,000 persons in developed countries and has risen continuously
over the past 40 years [1]. The prognosis of pancreatic cancer is very
poor, being the 4th leading cause of cancer deaths in Europe with the
⁎ Corresponding author at: Department of Geriatrics, Hospices Civils de Lyon, Edouard
Herriot University Hospital, 5, Place d'Arsonval, 69437 Lyon, France.
E-mail address: elisabeth.castel-kremer@chu-lyon.fr (E. Castel-Kremer).
1
The 2 first authors contributed equally to this work.
mortality rate is quite similar to the incidence rate [2]. Ninety percent
of pancreatic cancer are exocrine ductal adenocarcinoma and the
median age at diagnosis is 72 years old [3]. Risk factors for pancreatic
cancer are not well established, except tobacco, diabetes or other factors
related to dietary habits, but none of them are specific to the older adult
population [4]. Although familial pancreatic cancer is involved in 5 to 10%
of cases, the impact of genetic predisposition appears restricted to youn-
ger patients [5]. The diagnosis of pancreatic cancer is mostly established
at an advanced stage because of a late clinical expression of the disease
and lack of screening tests for pancreatic cancer. Thus, about 90% of pa-
tients have an advanced pancreatic cancer at the time of diagnosis and

https://doi.org/10.1016/j.jgo.2018.03.007
1879-4068/© 2018 Elsevier Ltd. All rights reserved.
374 E. Castel-Kremer et al. / Journal of Geriatric Oncology 9 (2018) 373–381
only 20% of the patients are diagnosed at a stage where the tumor is
resectable [6]. Despite recent advances in chemotherapy, pancreatic
resection followed by 6 months of gemcitabine or 5-fluorouracil
adjuvant chemotherapy remains the only potential curative option for
patients with pancreatic cancer [7,8]. The 5-year survival rate at any
stage of the disease is about 5% and in case of surgery with adjuvant
chemotherapy, it is about 15–20%. Since 1997, gemcitabine monother-
apy was established as the standard of care in advanced or metastatic
disease [9] but recently, 2 randomized controlled studies introduced
new standard treatments in this context. The first study published in
2011, compared gemcitabine vs a combination of 5 Fluoruracile,
Oxaliplatine and Irinotecan, FOLFIRINOX [10]. Because only patients
under 75 years were included in the study and given the toxicity of this
combined schedule of chemotherapies, most oncological teams continue
to reserve this treatment for young patients in good general condition
(under 70–75 years of age with performance status 0 or 1). In 2013,
Von Hoff et al. showed a significant improvement of the overall survival
in patients treated by gemcitabine + nab-paclitaxel compared to those
who received gemcitabine alone [11]. In this study, there was no age
limit for patient inclusion but the median age was ultimately young
(around 63 years old). Therefore, there remains a lack of data in older
patients with pancreatic cancer with gemcitabine + nab-paclitaxel or
FOLFIRINOX chemotherapy regimens.
The implementation of a comprehensive geriatric assessment (CGA)
is now recommended during the care of older adult patients with cancer
by the SIOG (International Society of Geriatric Oncology) [12]. Indeed,
the benefit of an onco-geriatric approach is demonstrated to improve
quality of life and to optimize specific treatments (surgery, chemother-
apy, etc.) [13]. Regarding pancreatic cancer, many authors studied the
outcome of older patients who underwent surgery [14] or who received
chemotherapies. Most of them concluded that standard of care oncologic
treatments can be proposed to geriatric population but subject to a
precise check-up, to detect specific frailties [15–18]. The data concerning
the specific impact of geriatric criteria in term of treatment tolerance and
overall survival remain scarce.
Here, we report a single-center experience of a systematic onco-
geriatric evaluation and its integration in the entire process of treatment
decision for older patients affected by a local, advanced, or metastatic
pancreatic cancer. We evaluated i) the feasibility of standard of care
Table 1
Content of the comprehensive geriatric assessment and scoring rules.
Geriatric conditions
- Autonomy and functional status
- Nutrition
- Comorbidities
- Cognitive and psychological status
- Polypharmacy
- Social conditions
oncologic treatments in this population, ii) the adequacy between the
proposed treatments after the multidisciplinary onco-geriatric evalua-
tion and those truly realized and iii) the respective weight of prognosis
factors, whether oncological, general, or geriatric.
2. Patients & Methods
2.1. Patients
From March 2007 to September 2014, all consecutive patients from
our single tertiary center older than 70 years, with histological diagnosis
of pancreatic adenocarcinoma were enrolled in a prospective cohort. All
patients underwent a geriatric assessment. At baseline, a complete med-
ical history including co-morbidities, circumstances of diagnosis, and an
evaluation of performance status (PS) according to the Eastern Coopera-
tive Oncology Group (ECOG) scale was obtained. In addition to a physical
examination, blood samples were also collected for measurement of
routine laboratory parameters including serum chemistry analysis,
serum albumin level, liver function tests, full blood cell count, specific
tumor marker (Ca19-9), lactate-déshydrogenase (LDH), alkaline phos-
phatase (PAL), and neutrophil/lymphocyte ratio. The extension of the
pancreatic disease was assessed for each patient with a thoraco-
abdominal computed tomography (CT), a liver ultrasonography, and if
necessary a pancreatic magnetic resonance imaging (MRI), and/or a
pancreatic endoscopic ultrasonography. Histological diagnosis was
performed after examination of operative specimens in case of surgery
or biopsies in case of metastatic or advanced pancreatic cancer. Pathol-
ogy reports described the tumor differentiation, the tumor size and local-
ization, lymph nodes and margin status (R0 = complete resection, R1 =
microscopic residual tumor, R2 = macroscopic residual tumor). Finally,
the tumor stage (TNM) was defined according to the American Joint
Cancer Committee (AJCC) criteria [19]. For each patient, a personal plan
of care was developed in a multidisciplinary onco-geriatric meeting
composed of oncologists, geriatricians, surgeons, radiotherapists, and
nurses, following the French guidelines for pancreatic cancer treatment
[20]. All patients gave written informed consent for participation to the
study which was previously approved by the local Ethics Committee of
the University of Lyon.
Tools and/or scales
- Katz Activities of Daily Living (ADLs) [21]
→ ADL disability if score ≥ 1
- Lawton's Instrumental Activities of Daily Living (IADLs) [22]
→ ADL disability if score ≥ 4
- Walking without help
- Loss of weight
→ Significant if ≥10% of body weight loss
- Mini Nutritional Assessment (MNA) [23]
→ Denutrition = score b 17.0
→ Risk of denutrition = score between 17.0 and 23.5
→ Correct nutritional status = score N 23.5
- Serum albumin level
- Cumulative Index Rating Scale-Geriatric (CIRS-G) (14 items) [24]
Each item is scored from 0 to 4 as follows:
→ 0 = no problem
→ 1 = minor problem;
→ 2 = morbidity or moderate discomfort, requiring treatment
→ 3 = severe problem difficult to control
→ 4 = extremely severe problem, requiring immediate treatment or severe functional impairment.
- Mini-Mental State Examination (MMSE) [25]
→ Alteration of cognitive function if score b 24/30
- Geriatric Depression Scale (GDS) [26]
→ Depression if score N 5
- Number of different medicines
→ Significant if number of medicines ≥5 [27]
- Social environment questionnaire
 E. Castel-Kremer et al. / Journal of Geriatric Oncology 9 (2018) 373–381 375

2.2. Geriatric Evaluation
The geriatric evaluation was performed at baseline by 2 senior
geriatricians and a nurse experienced in geriatric oncology. For the
comprehensive geriatric assessment (CGA) different validated tests
and questionnaires were used, on various aspects of the patient's
functions: autonomy, functional and nutritional status, cognitive func-
tions, psychological state, comorbid medical conditions, pain evaluation
and social support. The scales and instruments used for the geriatric
assessment are summarized in Table 1.
Autonomy and functional status were measured with: the Katz
Activities of Daily Living (ADLs) evaluating the necessary abilities for
basic living (bathing, transfers, dressing, continence, toileting and
feeding) and the Lawton's Instrumental Activities of Daily Living (IADLs),
evaluating the necessary skills for living independently in society
(using the phone, medication management, housework, linen washing,
finances' management, and cooking) [21,22]. The evaluation of the
nutritional status was made by measuring the loss of weight, serum
albumin levels at the beginning of the study and by using the Mini
Nutritional Assessment (MNA) [23]. Comorbidities were evaluated
using the Cumulative Index Rating Scale-Geriatric (CIRS-G) based on
the presence or absence of 14 items [24]. The global cognitive state (i.
e. orientation, memory, calculation, language and praxis) was assessed
with the Mini-Mental State Examination (MMSE) [25]. The Geriatric De-
pression Scale (GDS) was used in our study to evaluate the psychological
state [26]. In addition to the medical history, a precise collection and de-
scription of all the medicines taken by patients was performed during
the geriatric evaluation. The number of 5 concomitant medicines was
considered, in our study, as significantly associated with an increase of
side effects and a risk of abnormally high death rate [27]. Finally, social
geriatric conditions were globally estimated by questioning patients
about the help provided by the family circle, the material and financial
support already organized (nurse, home help, meals on wheels, etc.).
2.3. Surgical Procedure
All resections were performed by an experienced senior surgeon
(MA). All patients received prophylactic antibiotics preoperatively
(cefotaxime) and proton pump inhibitors postoperatively. Octreotide
administration was not used routinely. Pancreatic resections were
performed mainly via laparotomy with exploration of the abdominal
cavity ensuring the absence of contraindications. Pylorus preserving
duodenopancreatectomy (DP), total pancreatectomy, and left
splenopancreatectomy were performed according to our procedures
previously described [28]. Grading of complications was performed ac-
cording to Clavien-Dindo classification [29]. We categorized grade I & II

Table 2
Demographic, oncological and geriatric baseline characteristics of patients and comparison between surgical and non-surgical groups. P-value is calculated by comparing surgery and no-
surgery (*) groups.

Variables All patients Surgery* No surgery* P-value

N = 73 N = 42 N = 31

- Sex (male/female) 46/27 26/16 20/11 0.82

- Age (mean, range) 77.9 (70–93) 77.1 (70–87) 78.9 (72–93) 0.055
b80 yrs. (n, %) 49 (67) 32 (76) 17 (55)
N80 yrs. (n, %) 24 (33) 10 (24) 14 (45)
- ECOG Performance Status (n, %) b0.0001
0 8 (11) 7 (17) 1 (3)
1 40 (55) 30 (71) 10 (33)
2 20 (27) 5 (12) 15 (48)
3 5 (7) 0 (0) 5 (16)
- Weight loss N10% (n, %) 41 (56) 21 (50) 20 (67) 0.159
- Tumor localization (n, %) 0.3
Head 54 (74) 33 (79) 21 (68)
Body + tail 19 (26) 9 (21) 10 (32)
- TNM classification (n, %)
T1 5 (7) 5 (12) 0 (0) b0.001
T2 12 (16) 7 (17) 5 (16)
T3 39 (53) 28 (66) 11 (36)
T4 16 (23) 2 (5) 14 (45)
Tx 1 (1) 0 (0) 1 (3)
N0 10 (14) 9 (21) 1 (3) b0.001
N1 38 (52) 31 (74) 7 (23)
Nx 25 (34) 2 (5) 23 (74)
M0 55 (75) 41 (98) 14 (45) b0.001
M1 15 (21) 1 (2) 14 (45)
Mx 3 (4) 0 (0) 3 (10)
- ADL (n, %) 0.022
0 64 (88) 40 (95) 24 (77)
≥1 9 (12) 2 (5) 7 (23)
- IADL (n, %) 0.007
0–3 58 (79) 38 (90) 20 (65)
≥4 15 (21) 4 (10) 11 (35)
- MMSE (n, %) 0.49
0–23 16 (22) 8 (19) 8 (26)
≥24 57 (78) 34 (81) 23 (74)
- GDS (n,%) 0.31
0 26 (36) 17 (40) 9 (29)
≥1 47 (64) 25 (60) 22 (71)
- CIRS-G (number of grade 3 or 4 comorbidities) 0.071
0–1 63 (86) 39 (93) 24 (77)
≥2 10 (14) 3 (7) 7 (23)
- Walking without help (n, %) 61 (84) 38 (90) 23 (74) 0.064
- Albumin g/Lb35 (n, %) 26 (36) 12 (28) 14 (46) 0.136

ECOG: Eastern Cooperative Oncology Group, TNM: Tumor Nodes Metastasis classification of malignant tumors, ADLs: Katz Activities of Daily Living, IADLs: Lawton's Instrumental Activities
of Daily Living, MMSE: Mini-Mental State Examination, GDS: Geriatric Depression Scale, CIRS-G: Cumulative Index Rating Scale-Geriatric, n: number.
376 E. Castel-Kremer et al. / Journal of Geriatric Oncology 9 (2018) 373–381
as “mild complication” and grade III to V as “severe complication”. Sur-
gical mortality has been defined as any death, regardless of the cause,
occurring within 30 days after surgery in or out of hospital. The feasibil-
ity of treatment was evaluated by i) the number of patients with poten-
tially resectable disease according to TNM who ultimately underwent
surgery, ii) the morbidity and mortality after surgery and iii) the fre-
quency of adjuvant chemotherapy delivered.
2.4. Chemotherapy Treatment Schedule
The chemotherapy treatment was chosen according to the French
Guidelines and was, in majority, gemcitabine. Indeed, for adjuvant
therapy, gemcitabine (1000 mg/m2) was administrated on days 1, 8,
and 15, which corresponds to a cycle of chemotherapy. Each cycle of
gemcitabine was repeated every 28 days for a total duration of 6 months
(i.e. 6 cycles) [7]. For locally advanced and/or metastatic patients,
gemcitabine was also administered at the same dose and regimen and
was repeated every 28 days [9] until disease progression or intolerance.
Fig. 1. Flow chart of the cohort of patients with pancreatic cancer.
For a few patients, in a very good physical condition, FOLFIRINOX was
used as the first line chemotherapy for metastatic disease (i.e. oxaliplatin
85 mg/m2, irinotecan 180 mg/m2, leucovorin 400 mg/m2, and fluorouracil
was given as a bolus of 400 mg/m2 followed by a 46 h of continuous
intravenous infusion of 2400 mg/m2 given every 2 weeks) [10]. Because
the vast majority of patients were included before the results of Von
Hoff et al. [11], none of them received the gemcitabine + nab-paclitaxel
chemotherapy regimen. Treatment was prematurely stopped in case of
recurrence or progression of the disease, bad tolerance of the drug or
refusal by the patient.
Treatment was delivered in a multidisciplinary medical oncology
day care unit after adequate clinical exam and blood test performed
before each injection (neutrophils were N1.5 billion/L and platelets
N100 billion/L). Below these limits, treatment was delayed by 1 week.
Chemotherapy doses were reduced if needed, according to the
following guidelines: for patients experiencing excessive (grade 4
hematological side effects or grade 3 non hematological side effects)
“general” toxicities (e.g., asthenia, hematological toxicity, nausea and
 E. Castel-Kremer et al. / Journal of Geriatric Oncology 9 (2018) 373–381 377

vomiting), drug doses were reduced to 80% and then to 50%, if a subse-
quent dose reduction was needed. Side effects were collected before
each infusion according to the National Cancer Institute – Common
Terminology for Adverse Events (NCI-CTAE) The use of granulocyte-
colony stimulating factor (G-CSF) and erythropoietin was possible for
secondary prevention of neutropenia and anemia, respectively, in case
of metastatic disease. G-CSF was systematically used in the prophylaxis
of neutropenia when FOLFIRINOX was administrated.
During chemotherapy, all patients were assessed every 2 to 3
months with a clinical examination and a CT scan. After chemotherapy
withdrawal (because of either progression, toxicity, or end of the adju-
vant therapy) patients were followed-up every 3 months with a clinical
examination and imaging studies (CT scan and/or ultrasonography).
In case of early recurrence after adjuvant gemcitabine or disease
progression after a first-line palliative chemotherapy, a second-line
chemotherapy may be considered. If the clinical state of patients
allowed it, they were treated with FOLFOX for the majority, or GEMOX
in case of contraindication to 5-fluorouracil. FOLFOX consists of an
association of oxaliplatin 85 mg/m2, leucovorin 400 mg/m2, and fluoro-
uracil given as a bolus of 400 mg/m2 followed by a 46 h of continuous
intravenous infusion of 2400 mg/m2. GEMOX is an association of
gemcitabine 1000 mg/m2 as a 30-mn intravenous infusion followed
by oxaliplatin 100 mg/m2 as a 2-h intravenous infusion [30]. FOLFOX
or GEMOX were administrated every 2 weeks and were continued
until the progression of the disease or a severe toxicity was detected.
In case of the use of FOLFIRINOX as a first line chemotherapy, eligible
patients could receive gemcitabine alone as a second line treatment.
Neoplasm. At the time of diagnosis, 66% of patients have a normal or
slightly impaired performance status (PS = 0 or 1) and only 5 patients
(7%) have a low performance status (PS = 3).
Three quarters of tumors (n = 54) were localized in the pancreatic
head which mirrors the predominantly localization of pancreatic adeno-
carcinoma in the general population [6]. According to the radiological or
pathological evaluation, 75% of tumors were classified as T3 or T4 and N
+ in N50% of cases. The lymph node status could not be determined in
25 patients given the absence of surgery and no clear evidence of
tumor invasion provided by medical imaging. At the time of diagnosis,
about 20% of patients had a metastatic disease with a majority of hepatic
invasion.
Regarding the geriatric assessment, the majority of patients were
considered to have a no impairment in ADL and IADL, as shown at
baseline by 88 and 79% of patients having respectively an ADL = 0
and an IADL between 0 and 3. An alteration of the cognitive function
was diagnosed in N20% of patients (n = 16). A median weight loss of
6.6 kg with a maximum of 19 kg was reported in the study. About a
quarter of patients were considered undernourished (MNA b17.0) as
opposed to a quarter of patients who seemed to have a correct
nutritional status with an MNA N23.5 points. Thirty-six percent of
patients had a serum albumin level lower than 35 g/L. On average,
patients in our study took 4.8 medicines with a maximum of 11 and a
minimum of no medicine. Looking at the evaluation of comorbidities,
the median CIRS-G was about 8 points (range: 1–22) and 50% of
patients scored 3 or 4 (severe or extremely severe respectively) in one
or more pathologies, according to the CIRS-G.

2.5. Statistical Analysis

3.2. Treatment
Categorical variables were described as number and percentage;
continuous variables were described as mean and range or standard Forty-two, out of the 73 patients, underwent surgery (57%) and the
deviation (SD). Fischer exact test was used to compare categorical others were treated by best supportive care alone (n = 9) or palliative
variables, and Mann-Whitney U test was used to compare continuous chemotherapy (n = 22) (Fig. 1). Only 2 and 3 patients with a tumor
variables. A survival analysis was performed to assess the impact of classified T2 and T3, respectively, without metastatic disease were
the following factors on survival: gender, age, TNM, diabetes, weight contraindicated for surgery after the multidisciplinary oncology meeting
loss, ADL, IADL, MMS, GDS, number of severe comorbidities. Survival
was calculated from the date of diagnosis to the date of death or the Table 3
last follow-up news. Survival distributions were estimated using the Type of procedure, pathological results and outcome of patients who underwent curative
Kaplan-Meier method, and were compared using the Log-rank test. surgery.
Univariate Cox proportional hazard model was fitted. The proportional
N = 42
hazards' assumption was tested using Schoenfeld residuals before
Type of surgery, n (%)
testing each variable in univariate analysis. All the significant variables
- Duodenopancreatectomy 26 (62)
identified by the univariate analysis (p b 0.1) were used to build - Total pancreatectomy 10 (24)
multivariate models to assess the effect of each potential confounding - Left splenopancreatectomy 6 (14)
factor on survival. Statistical analysis was performed using the Statistical Resection margin, n (%)
Package for the Social Sciences Inc. (SPSS) (Chicago, IL, USA) and Stata - R0 30 (71)
- R1 12 (29)
11.0 (StataCorp. 2009. Stata Statistical Software: Release 11. College - R2 0 (0)
Station, TX: StataCorp LP, USA). All tests were 2-tailed, a p-value b0.05 Morbidity according to Clavien-Dindo classification, n (%)
was considered statistically significant. - No complication 23 (55)
- Mild complications (Grade I-II) 12 (29)
- Severe complications (Grade III-IV) 7 (16)
3. Results
Surgical mortality, n (%) 0 (0)
Median lenght of stay in hospital, days (range) 21 (14–31)
3.1. Patients' Characteristics Adjuvant chemotherapy with gemcitabine, n (%)
- Proposed 33 (79)
Between March 2007 and September 2014, 73 consecutive patients -≥80% realized 17 (40)
-b80% realized 9 (21)
aged at least 70 years and admitted to our surgical, geriatric or oncology
Evolution during the post-surgical follow-up, n (%)
departments with pancreatic adenocarcinoma met our inclusion - No recurrence 27 (64)
criteria. Patients' characteristics are summarized in Table 2. The median - Recurrence 15 (36)
age at diagnosis is 78 (range 70–93) years and patients aged 80 years Treatment after post-surgical recurrence, n (%)
- Best supportive care 10 (24)
and older represent 33% of our cohort. At the time of diagnosis the
- Second line chemotherapy
vast majority of patients (n = 67) exhibit one or more disease-related - GEMOX 3 (7)
symptoms: body-weight loss ≥10% (n = 47), jaundice (n = 33), pain - FOLFOX 2 (5)
(n = 26) or de novo diabetes (n = 15). Pancreatic cancer was diagnosed - Third line chemotherapy 2 (5)
fortuitously in 9 patients (12%) and in 10 out of the 73 patients (14%) in N: number, R: resection boundaries, GEMOX: gemcitabine and oxaliplatin chemotherapy
a context of a tight follow-up for an Intraductal Papillary Mucinous regimen, FOLFOX: fluorouracil, leucovorin and oxaliplatin chemotherapy regimen.
378 E. Castel-Kremer et al. / Journal of Geriatric Oncology 9 (2018) 373–381

because of low PS in 2 cases, comorbidities in 2 other cases and refusal of
the patient in 1 case. Other patients with tumor classified T4 and/or
metastatic disease received palliative treatment. Patients who underwent
curative surgery were significantly younger, with better PS compared to
those who only benefited from palliative treatments (Table 2). In addition
to tumor-related parameters, the surgical and non-surgical group differ
also in term of ADL and IADL which are significantly better in the surgical
group (p = 0.022 and p = 0.007, respectively). Twenty-six (62%)
duodenopancreatectomy were performed whereas a total pancreatec-
tomy or a left splenopancreatectomy were done in 10 and 6 cases,
respectively. Only one patient in our study had a tumor considered not
resectable during surgery. Pathological results and outcome of surgical
patients were summarized in Table 3. The median length of stay in
hospital for surgery was 21 (range = 14–31) days. Adjuvant chemother-
apy was suggested, after the multidisciplinary oncogeriatric meeting, to
33 out of 42 surgical patients (79%). Twenty-six patients received
ultimately an adjuvant treatment with gemcitabine (62%) and among
them, 17 patients (65%) got at least 80% of the planned treatment (i.e.
N4 out of 6 planned cycles of gemcitabine). The 7 other patients finally
did not receive the initially planned adjuvant chemotherapy mainly due
to refusal by the patient or a rapid deterioration of the patient's general
condition. During the follow-up of surgical patients (médian follow-up
20.4 (7.3–27.1) months), 15 out of 42 exhibited a proven recurrence of
the adenocarcinoma.

Fig. 2. Kaplan–Meier curves representing the overall survival of patients with pancreatic adenocarcinoma according to the type of treatment (surgery or not) (A), age (≥ or b80 years old)
(B) and IADL (≥ or b4) (C).
 E. Castel-Kremer et al. / Journal of Geriatric Oncology 9 (2018) 373–381 379

Table 4 4. Discussion
Univariate analysis of overall survival (n = 73).

Variables Hazard ratio P-value The interest in the study of the treatments of pancreatic cancer in the
(95% CI) older adults is increasing. Tumor resection is the only curative option
- Treatment (surgery vs no surgery) 0.2 (0.09–0.4) b0.001 but remains a major surgery with a non-negligible morbi-mortality.
- Sex (male vs female) 1.9 (0.7–5) 0.2 This surgery ideally requires high volume centers as well as a high
- Age (≥80 vs b80 y.o.) 2.8 (0.2–3.1) 0.5 level of qualification of the surgical teams [31]. To date, N15 studies
- ECOG performance status (≥2 vs b2) 10.7 (3.0–38.3) b0.001
have addressed the question of the feasibility of such surgery especially
- Weight loss (N10% vs ≤10) 4.6 (1.2–18.6) 0.03
- TNM classification in patients older than 80 years [32–34]. Most have concluded that age
T2 vs T0 7 (1.4–36.3) 0.04 should not be considered as a limiting factor for surgery, although 2
T3 vs T0 4.7 (1.07–21.1) 0.006 recent studies, including a meta-analysis, have found a higher morbidity
T4 vs T0 10 (1.9–52) 0.4 and mortality in octogenarians compared with other younger patients
N1 vs N0 1.4 (0.5–4.0) 0.006
Nx vs N0 4.6 (1.5–13)) 0.005
[35,36]. In our study, overall survival in the surgical group was not
M1 vs M0 3.4 (1.4–8) 0.008 significantly different in patients aged less than or over 80 years,
Mx vs M0 1.4 (0.19–10.9) 0.7 confirming the absence of impact of age on survival of carefully selected
- ADL (b1 vs ≥1) 3 (1.2–12) 0.2 patients. All authors agree that an exhaustive preoperative evaluation is
- IADL (b4 vs ≥4) 5.0 (1.02–24.5) 0.047
an indispensable prerequisite for all surgical treatment projects, partic-
- MMSE (≥24 vs b24) 2 (0.7–5.4) 0.1
- GDS (≥1 vs b1) 1.5 (0.7–3)) 0.15 ularly in this population of older subjects. However, most of the studies
- CIRS-G (number of grade 3 or 4 comorbidities) 19 (2.1–76) 0.035 evaluate these patients only with conventional tools (American Society
(≥2 vs b2) of Anesthesiologists (ASA) score, nutritional and performance status
- Walking (with help vs without) 0.2 (0.08–48) 0.57 tools) and few of them included a specific geriatric evaluation. These re-
N: number, CI: confidence interval, y.o.: years old, ECOG: Eastern Cooperative Oncology sults are consistent with those of a recent study by the SIOG surgical task
Group, TNM: Tumor Nodes Metastasis classification of malignant tumors, ADLs: Katz Ac- force showing that b7% of surgeons use of a Comprehensive Geriatric
tivities of Daily Living, IADLs: Lawton's Instrumental Activities of Daily Living, MMSE:
Assessement (CGA) in daily practice prior to tumor surgery [37]. In
Mini-Mental State Examination, GDS: Geriatric Depression Scale, CIRS-G: Cumulative
Index Rating Scale-Geriatric. our study, we systematicaly performed a CGA to optimize the treatment
strategy (surgery or palliative care) confirming the feasability of this
approach. However, we acknowledge that this requires to perfectly
coordinate the surgical, oncologic, and geriatric teams, and also to hire
dedicated geriatricians. Thus, the use of geriatric screening tools such
Twenty-two patients, ineligible for surgery, were treated with first- as the G8 scale by surgeons and oncologists may be a promising alterna-
line palliative chemotherapy gemcitabine (n = 19), FOLFIRINOX (n = tive in a busy practice to identify patients requiring a full CGA. This
2) or GEMOX in only one case. In our cohort, 87 cycles of palliative complete geriatric evaluation should be performed by geriatricians but
chemotherapy were administrated thus corresponding to a median of only in patients where the geriatric syndrom makes the decision of
9 cycles per patient (range = 1–12). Because of disease progression, 8 treatment difficult. These geriatric screening tools should be evaluated
patients received a second line of chemotherapy with GEMOX (n = 5) in the context of pancreatic cancer [38,39].
or FOLFOX (n = 3), other patients received best supportive care. About 71% of patients not eligible for surgery received a first-line
Tolerance profile of the chemotherapy was relatively good with a palliative chemotherapy and nearly 40% of them received a second-
need of withdrawal caused by side-effects for only 7 patients (4 in a line chemotherapy. A median of 9 cycles of chemotherapy was adminis-
palliative context and 3 in an adjuvant purpose) exclusively due to trated per patient. As observed in the surgically-treated patients, no
hematotoxicity (4 grade 3 thrombopenia and 3 grade 3 neutropenia). impact of age (±80 years old) was found on the overall survival of
Chemotherapy was mainly stopped because of patient's poor general non-surgical patients. Altogether, these results highlight the possibility
condition and/or disease progression. of offering standard of care palliative treatment in older patients subject
to careful selection. Although in our study conducted between 2007 and
2014, only two patients received palliative chemotherapy with
3.3. Overall Survival and Prognostic Factors FOLFIRINOX which may limit the applicability of our results in current
practice, the introduction of the new treatment standards with
The mean duration of follow-up was of 1.1 (0.1–4.8) years per patient. combination chemotherapy (i.e. gemcitabine + nab-paclitaxel and
The median overall survival in the whole cohort was 13.8 (1–58) months. FOLFIRINOX) will doubtless increase the interest for a CGA [10,11].
Crude analysis demonstrated that surgery was associated with a signifi- Recently, combination of gemcitabine and capecitabine has been pro-
cant decreased mortality (overall survival 21.3 vs 6.1 months (p = posed as the new standard of care following pancreatic resection [40].
0.0001) in the surgical group vs non-surgical group, respectively). Thus, Indeed, these more effective treatments are also more toxic, making
at 12 months, 75% of surgical patients were still alive against 25% in the the detection of frailties of older adults all the more important to intro-
non-surgical group (Fig. 2A). It can be noted that the effect of surgery duce these drugs only when there is a real clinical benefit for the patient.
on survival is independent of patient's age. Thus the 1 year survival was The interest of a CGA and of a multidisciplinary therapeutic decision-
comparable between patients over or under 80 years of age in the surgical making (surgeons, oncologists and geriatricians) is also highlighted in
group (70% (CI 33–89%) vs 77% (CI 57–88%), respectively) as well as in the our study by the adequacy between the treatment proposed during
non-surgical group (22% (CI 5–88%) vs 28% (CI 33–89%), respectively) the multidisciplinary onco-geriatric meeting and the one that was
(Fig. 2B). Results from the Univariate analysis are summarized in finally realized. Thus, all the patients, with the exception of one, for
Table 4. This analysis showed, beyond oncological parameters, an impact whom a surgery was proposed, could benefit from a resection with
of IADL and CIRS-G on survival with a risk of death significantly higher if curative purposes. In addition, almost 80% of patients with an indication
IADL was b4 (HR = 5.0 (CI 1.02–24.5), p = 0.047) (Fig. 2C) or CIRS-G ≥2 of adjuvant chemotherapy received the treatment and 70% of scheduled
(HR = 19 (CI 2.1–76), p = 0.035). In addition, nutrition parameters chemotherapy was actually administered. Moreover, in a palliative
significantly impacted the survival with an increased risk of death if the situation, only 4 patients ceased chemotherapy due to adverse events
weight loss was N10% (HR = 4.6 (CI 1.2–18.6)), p = 0.03). Results from or poor tolerance, thus demonstrating the adequation of the proposed
the Multivariate analysis failed to demonstrate independent predictors treatment to the clinical situation of the patients. However, we ac-
of survival except for surgery (p b 0.001). knowledge that the design of our study does not accurately quantify
380 E. Castel-Kremer et al. / Journal of Geriatric Oncology 9 (2018) 373–381
the impact of the CGA in terms of treatment decision and even more
impact on survival which is a significant limitation to our work.
Concerning the survival, one year after diagnosis, 75% of surgically
treated patients were alive vs 25% of those who were treated with
palliative chemotherapy or best supportive care alone. These survival
results are comparable to those found in studies dedicated to older
adults [32–35] but better than those estimated in the general popula-
tion [41]. However we acknowledge that our patients came from a
tertiary center and that only those patients, for whom exhaustive
investigations for diagnosis (imaging and biopsy) could be done, were
prospectively included. Nevertheless, despite this selection of patients,
our study represents real life management of a cohort of patients aged
about 78 years on average suffering from pancreatic cancer, with co-
morbidities, polymedication, and geriatric syndroms (weight loss,
signs of depression, decreased MMS and autonomy). This last point is
critical because we show in our study that, besides the classical factors
directly dependent on oncological criteria (TNM stage, PS, surgery or
not), IADL, CIRS-G, and weight loss were significantly associated with
survival. These results, in line with those reported in a recent publica-
tion in the context of colorectal cancer [42], highlighted the importance
of geriatric evaluation, particularly impairment in IADL, comorbidities,
and nutrition status in predicting survival and treatment outcomes.
However, because of the small number of patients in our study, we
failed to find in multivariate analysis a factor, other than surgery,
independently associated with survival.
In conclusion, we reported here a single-center experience of a
multidisciplinary care of elderly patients suffering from pancreatic
adenocarcinoma, encompassing a CGA. This overall evaluation of
patients is feasible in practice and useful in terms of predicting possible
treatments in these patients. Age should no longer be a limiting criteria
to propose standard of care oncologic treatments for older adults, but
impairment in IADL, CIRS-G, and weight loss as prognostic factors for
survival should be taken into consideration as an addition to the onco-
logical criteria in the therapeutic decision-making process.
Disclosures/Author Contributions
Guarantor of the Article
Gilles Boschetti, MD, PhD.
Financial Support
None.
Potential Competing Interests
None.
Specific Author Contributions
ECK and SDT: acquisition of data, statistical analysis, drafting of the
manuscript.
ALC: analysis and interpretation of data.
EG: critical revision of the manuscript.
XCL and MA: surgical care and critical revision of the manuscript.
BC, CLB and TW: critical revision of the manuscript for important
intellectual content.
GB: study concept and design, drafting of the manuscript and study
supervision.
References
[1] Malvezzi M, Bertuccio P, Levi F, La Vecchia C, Negri E. European cancer mortality pre-
dictions for the year 2014. Ann Oncol 2014;25:1650–6.
[2] Bertuccio P, Rosato V, Andreano A, Ferraroni M, Decarli A, Edefonti V, et al. Dietary
patterns and gastric cancer risk: a systematic review and meta-analysis. Ann Oncol
2013;24:1450–8.
[3] Ilic M, Ilic I. Epidemiology of pancreatic cancer. World J Gastroenterol 2016;22:
9694–705.
[4] Maisonneuve P, Lowenfels AB. Risk factors for pancreatic cancer: a summary review
of meta-analytical studies. Int J Epidemiol 2015;44:186–98.
[5] Yeo TP. Demographics, epidemiology, and inheritance of pancreatic ductal
adenocarcinoma. Semin Oncol 2015;42:8–18.
[6] Ducreux M, Cuhna AS, Caramella C, Hollebecque A, Burtin P, Goéré D, et al. Cancer of
the pancreas: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-
up. Ann Oncol 2015;26(Suppl. 5):56–68.
[7] Oettle H, Post S, Neuhaus P, Gellert K, Langrehr J, Ridwelski K, et al. Adjuvant
chemotherapy with gemcitabine vs observation in patients undergoing curative-
intent resection of pancreatic cancer: a randomized controlled trial. JAMA 2007;
297:267–77.
[8] Neoptolemos JP, Stocken DD, Bassi C, Ghaneh P, Cunningham D, Goldstein D, et al.
Adjuvant chemotherapy with fluorouracil plus folinic acid vs gemcitabine following
pancreatic cancer resection: a randomized controlled trial. JAMA 2010;304:
1073–81.
[9] Burris HA, Moore MJ, Andersen J, Green MR, Rothenberg ML, Modiano MR, et al. Im-
provements in survival and clinical benefit with gemcitabine as first-line therapy for
patients with advanced pancreas cancer: a randomized trial. J Clin Oncol 1997;15:
2403–13.
[10] Conroy T, Desseigne F, Ychou M, Bouché O, Guimbaud R, Bécouarn Y, et al.
FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med
2011;364:1817–25.
[11] Von Hoff DD, Ervin T, Arena FP, Chiorean EG, Infante J, Moore M, et al. Increased sur-
vival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med 2013;
369:1691–703.
[12] Wildiers H, Heeren P, Puts M, Topinkova E, Janssen-Heijnen ML, Extermann M, et al.
International Society of Geriatric Oncology Consensus on Geriatric Assessment in
older patients with cancer. J Clin Oncol 2014;32:1–17.
[13] Puts MT, Santos B, Hardt J, Monette J, Girre V, Atenafu EG, et al. An update on a sys-
tematic review of the use of geriatric assessment for older adults in oncology. Ann
Oncol 2014;25:307–15.
[14] Audisio RA, Pope D, Ramesh HS, Gennari R, Van Leeuwen BL, West C, et al. Shall we
operate ? Preoperative assessment in elderly patients (PACE) can help. A SIOG sur-
gical task force prospective study. Crit Rev Oncol Hematol 2008;65:156–63.
[15] Oliveira-Cunha M, Malde DJ, Aldouri A, Morris-Stiff G, Menon KV, Malvern Smith A.
Result of pancreatic surgery in the elderly: is age a barrier? HBP (Oxf) 2013;15:
24–30.
[16] Frakes JM, Strom T, Springett GM, Hoffe SE, Balducci L, Hodul P, et al. Resected pan-
creatic cancer outcomes in the elderly. J Geriatr Oncol 2015;6:127–32.
[17] Adham M, Bredt LC, Robert M, Perinel J, Lombard-Bohas C, Ponchon T, et al. Pancre-
atic resection in elderly: should it be denied? Langenbecks Arch Surg 2014;399:
449–59.
[18] Nagrial AM, Chang DK, Nguyen NQ, Johns AL, Chantrill LA, Humphris JL, et al. Adju-
vant chemotherapy in elderly patients with pancreatic cancer. Br J Cancer 2014;110:
313–9.
[19] McIntyre CA, Winter JM. Diagnostic evaluation and staging of pancreatic ductal
adenocarcinoma. Semin Oncol 2015;42:19–27.
[20] Thesaurus national de Cancerologie digestive. In: André T, editor. cancer du pan-
créas; 2010www.tncd.org.
[21] Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged,
the index of ADL: a standardized measure of biological and psychosocial functions.
JAMA 1963;185:914–9.
[22] Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumen-
tal activities of daily living. Gerontologist 1969;9:179–86.
[23] Guigoz Y, Vellas B, Garry P. Mini nutritional assessment: a practical assessment tool for
grading the nutritional state of elderly patients. Fact Res Gerontol 1994(Suppl. 2):
15–59.
[24] Linn BS, Linn MW, Gurel L. Cumulative illness rating scale. J Am Geriatr Soc 1968;16:
622–6.
[25] Folstein MF, Folstein SE, PR McHugh. “Mini-mental state”. A practical method for
grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:
189–98.
[26] Rinaldi P, Mecocci P, Benedetti C, Ercolani S, Bregnocchi M, Menculini G, et al.
Validation of the five-item geriatric depression scale in elderly subjects in three
different settings. J Am Geriatr Soc 2003;51:694–8.
[27] Monegat M, Sermet C, Perronnin M, Rococo E. Polypharmacy: definitions, measure-
ment and stakes involved, review of the literature and measurement tests.
Questions d'économie de la santé 2014;204:1–8.
[28] Adham M, Singhirunnusorn J. Surgical technique and results of total mesopancreas
excision (TMpE) in pancreatic tumors. Eur J Surg Oncol 2012;38:340–5.
[29] Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new
proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann
Surg Aug 2004;240:205–13.
[30] Afchain P, Chibaudel B, Lledo G, Selle F, Bengrine-Lefevre L, Nguyen S, et al. First-line
simplified GEMOX (S-GemOx) versus classical GEMOX in metastatic pancreatic can-
cer (MPA): results of a GERCOR randomized phase II study. Bull Cancer 2009;96:
E18-2.
[31] Derogar M, Blomberg J, Sadr-Azodi O. Hospital teaching status and volume related to
mortality after pancreatic surgery in a national cohort. Br J Surg 2015;102:548–57.
[32] Di Carlo V, Balzano G, Zerb A, Villa E. Pancreatic cancer resection in elderly patients.
Br J Surg 1998;85:607–10.
 E. Castel-Kremer et al. / Journal of Geriatric Oncology 9 (2018) 373–381
[33] Hatzaras I, Schmidt C, Klemanski D, Muscarella P, Melvin WS, Ellison EC, et al.
Pancreatic resection in the octogenarian: a safe option for pancreatic malignancy. J
Am Coll Surg 2011;212:373–7.
[34] Beltrame V, Gruppo M, Pastorelli D, Pedrazzoli S, Merigliano S, Sperti C. Outcome of
pancreaticoduodenectomy in octogenarians: single institution's experience and
review of the literature. J Visc Surg 2015;152:279–84.
[35] Lee DY, Schwartz JA, Wexelman B, Kirchoff D, Yang KC, Attiyeh F. Outcomes of
pancreaticoduodenectomy for pancreatic malignancy in octogenarians: an
American College of surgeons national Surgical Quality Improvement Program
analysis. Am J Surg 2014;207:540–8.
[36] Casadei R, Ricci C, Lazzarini E, et al. Pancreatic resection in Patients 80 years or older.
A meta-analysis and systematic review. Pancreas 2014;43:1208–18.
[37] Ghignone F, van Leeuwen BL, Montroni I, Huisman MG, Somasundar P, Cheung KL,
et al. The assessment and management of older cancer patients: a SIOG surgical
task force survey on surgeons' attitudes. Eur J Surg Oncol 2016;42:297–302.
[38] Decoster L, Van Puyvelde K, Mohile S, Wedding U, Basso U, Colloca G, et al. Screening
tools for multidimensional health problems warranting a geriatric assessment in
381
older cancer patients: an update on SIOG recommendations. Ann Oncol 2015;26:
288–300.
[39] Kenis C, Decoster L, Van Puyvelde K, De Grève J, Conings G, Milisen K, et al.
Performance of two geriatric screening tools in older patients with cancer. J Clin
Oncol 2014;32:19–26.
[40] Neoptolemos JP, Palmer DH, Ghaneh P, Psarelli EE, Valle JW, Halloran CM, et al. Com-
parison 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;389:1011–24.
[41] American Cancer Society. Cancer facts & figures. www.cancer.org/research/
cancerfactsfigures/cancerfactsfigures/cancer-facts-figures-2016; 2016.
[42] Aparicio T, Gargot D, Teillet L, Maillard E, Genet D, Cretin J, et al. Geriatric factors anal-
yses from FFCD 2001-02 phase III study of first-line chemotherapy for elderly meta-
static colorectal cancer patients. Eur J Cancer 2016 [pii: S0959-8049(16)32475-3].