Sage Journals: Discover world-class research

Abstract

BACKGROUND:

Pancreatic adenocarcinoma (PDAC) is one of the deadliest human malignancies. Although surgery is currently the only effective treatment for PDAC, most patients survive less than 20 months after tumor resection.

OBJECTIVE:

The primary goal was to investigate alterations in KRAS, TP53, SMAD4 and CDKN2A/p16 in tumors from patients with exceptionally long survival after surgery.

METHODS:

Tumors from 15 patients with PDAC that survived more than 55 months after surgery (“LS”) were analyzed for KRAS, TP53, IDH1, NRAS and BRAF using next-generation sequencing. SMAD4 and CDKN2A/p16 was tested using immunohistochemistry. MGMT promoter methylation was investigated.

RESULTS:

Tumors from “LS” have a lower prevalence of KRAS and TP53 mutations and had more frequently SMAD4 retained expression, if compared with that of patients died within 24 months from surgery. The survival of patients with wild-type KRAS and TP53 tumors was more than twice longer than that of patients bearing KRAS and TP53 mutations (90.2 vs. 41.1 months). Patients with KRAS wild-type tumors and that retained SMAD4 expression had a survival twice longer than cases with alterations in both genes (83.8 vs. 36.7 months). Eleven tumors (39.3%) showed MGMT methylation.

CONCLUSIONS:

Our data indicate that absence of KRAS, TP53 and SMAD4 genetic alterations may identify a subset of pancreatic carcinomas with better outcome.

Keywords

Pancreatic adenocarcinoma KRAS TP53 SMAD4 mutation

1. Introduction

Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest solid human malignancies [39], with overall 5-year survival well under 5% across all stages. Such a dismal prognosis is due to diagnosis at late stages, together with the lack of effective systemic therapies to control the disease [28]. Less than 20% of patients are amenable to surgery. Of those who undergo surgery, approximately 20% survive 5 years and only 10% are still alive after 10 years [11].

Surgical resection of PDACs offers the best hope to improve survival, but despite advances in pancreatic surgery, resected patients have a median survival $<$ 20 months [28, 38, 54].

The factors responsible for long-term survival of patients with PDAC are poorly understood [11]. Numerous studies focusing on 5-year and 10-year survivors have shown that conventional clinicopathological parameters such as early stage of disease, lack of lymph node spread and negative surgical margins predict a more favorable prognosis [1, 18, 19, 26, 47]. However, these same studies also showed that positive resection margins or tumor metastases to lymph nodes did not preclude long-term survival. In fact, 20–40% of patients who survive more than 5 years after surgery had nodal disease and/or positive margins. These findings suggest that pathologic staging is not the sole determinant of a relatively long survival in patients with PDAC. The longer survival in this subset of PDAC patients may depend on distinct genetic, epigenetic, or other biologic factors, such as changes in the tumor microenvironment or enhanced immune response. The possibility of identifying those molecular parameters that can predict survival in the pre-operative setting would be most helpful for patient management and clinical decision making.

Researches in the field have uncovered genetic aberrations that occur during PDAC development and progression. In most cases, PDAC is initiated by oncogenic mutant KRAS [17]. Whole-exome sequencing of 24 PDACs was reported in 2008 and the exon analysis of 20,661 genes indicated that PDACs contain an average of 63 genomic alterations, most of which are point mutations [29]. High throughput next generation sequencing, including a large study by Bailey et al. [4], have shown that four main genes (referred to as “mountain genes” for their relevance) are altered in PDAC: KRAS, TP53, SMAD4, CDKN2A/p16 [21, 51, 58]. However, it is still under debate if there are molecular markers capable of predicting prognosis of PDAC patients [11, 45]. At present, medical societies exclude a need for personalized, molecularly targeted treatment of PDAC patients [15].

In the database of the Surgery Unit (Bologna Italy) we have identified 15 patients out of total 342 operated for PDAC who survived longer than 4 years after surgery (4.4%).

The aim of the present study is to define the molecular status for the KRAS, TP53, CDKN2A and SMAD4 “mountain genes” and of other potentially relevant markers such as IDH1, microsatellite instability (MSI) and MGMT (O6-methylguanine methyl-transferase) promoter methylation in these 15 PDAC patients with an unusually long survival.

2. Material and methods

2.1 Case selection

In this study, 15 patients with PDAC and a survival rate higher than 55 months from surgery (“long-survivor group” – LS) were analyzed.

All histologic diagnoses were reviewed by one of the authors (AF). Diagnosis, pTNM and stage evaluation were performed according to the World Health Classification (WHO) and the American Joint Committee on Cancer (AJCC) criteria, respectively [8, 16].

All samples were formalin fixed-paraffin embedded (FFPE) and obtained from resection specimens. They were analyzed for KRAS (exons 2-3-4) and TP53 (exons 5-6-7-8) mutational status. In KRAS wild-type samples, NRAS (exons 2-3-4) and BRAF (exon 15) genes were analyzed. Microsatellite instability (MSI) and MGMT promoter methylation status were also determined. Protein expression of p16 and SMAD4 was evaluated by immunohistochemistry.

A group of 15 patients with PDAC that died from the disease within 24 months from surgery (“non-long-survivor group” – NLS) were also analyzed: this group was matched for age, sex, stage and margin status with the “LS” group.

Table 1
Primers used for mutational analysis

Gene (exon)	Forward sequence (5’-3’)	Reverse sequence (5’-3’)
KRAS (2)	ggCCTgCTgAAAATgACTgAA	TgTATCAAAgAATggTCCTgCAC
KRAS (3)	TgTTTCTCCCTTCTCAggATTC	TTTAAACCCACCTATAATggTgAA
KRAS (4)	AAgAgTTAAggACTCTgAAgATgTACC	TTATgATTTTgCAgAAAACAgATC
NRAS (2)	TggAAggTCACACTAgggTTTT	TggATTgTCAgTgCgCTTTTC
NRAS (3)	CAgAAAACAAgTggTTATAgATggTg	TAATATCCgCAAATgACTTgCTA
TP53 (5)	TCTgTCTCCTTCCTCTTCCTACA	AACCAgCCCTgTCgTCTCT
TP53 (6)	CAggCCTCTgATTCCTCACT	CAgTTgCAAACCAgACCTCA
TP53 (7)	CTTgggCCTgTgTTATCTCC	gTgTgCAgggTggCAAgT
TP53 (8)	gCCTCTTgCTTCTCTTTTCC	gCTTCTTgTCCTgCTTgCTT
BRAF (15)	TgCTTgCTCTgATAggAAAATgA	TgCTTgCTCTgATAggAAAATgA
IDH1 (4)	gAAACAAATgTggAAATCACCA	TCACATTATTgCCAACATgACT

All information regarding the human material was managed using anonymous numerical codes. All samples were handled in compliance with the Helsinki Declaration (http://www.wma.net/en/30publications/10policies/b3/). The study was approved by the internal review board committee (Azienda USL of Bologna, number 15035).

2.2 Mutational analysis (KRAS, NRAS, BRAF, IDH1, TP53)

In order to assess the molecular status of the genes analyzed, DNA was extracted from FFPE material scraped under microscope guidance from three 10- $\mu$ m thick sections, using the MasterPure DNA FFPE extraction kit (Roche Diagnostic, Manheim, Germany). Scraped areas were selected by a pathologist (AF) after evaluation of the percentage of neoplastic cells.

After careful review of all histology sections, FFPE blocks with the largest representative areas of viable tumor cells were selected for molecular and immunohistochemical analyses (see section below). The fraction of mutated neoplastic cells was calculated normalizing the mutated allele fraction for the proportion of neoplastic cells in the area subjected to molecular analysis, as previously described [55].

Sequencing was performed using the 454 GS-Junior next generation sequencer (Roche Diagnostic, Mann- heim, Germany), as previously described [13] (http://www.454.com/). Primer sequences are reported in Table 1. Sequencing results were analyzed using Amplicon Variant Analyzer (AVA) software (Roche Diagnostic, Mannheim, Germany). Only nucleotide variations observed in both strands were considered for mutational calls. Mutations were checked using COSMIC (Catalogue of Somatic Mutations in Cancer) database (http://cancer.sanger.ac.uk/cosmic) and “IARC TP53 Database” (http://p53.iarc.fr/).

2.3 MGMT promoter methylation analysis

Methylation analysis of MGMT promoter region was performed according to previously published protocols [42].

2.4 Microsatellite instability

MSI analysis was performed investigating 12 microsatellite regions, as previously described [41].

Table 2
Antibody used for immunohistochemistry analysis

Antibody	Clone	Dilution
CDKN2A/P16	E6H4	VentanaPrediluted
SMAD4	B8	1:350

2.5 Immunohistochemistry

FFPE blocks with the largest representative areas of viable tumor cells were selected for immunohistochemical analysis (and molecular analysis, see section above). Immunohistochemistry (IHC) was performed on two consecutive 3 $\mu$ m-thick sections using the VentanaBenchmark immunostainer (Ventana-Roche, Mannheim, Germany) for p16 protein and the UltraVisionQuanto Detection System (Thermo Fisher Scientific, MA, USA) for SMAD4 detection (Table 2). Immunohistochemical evaluation was performed as previously described [45].

CDKN2A/p16 is an oncosuppressor gene that is inactivated by mutation, loss of heterozygosity, epigenetic changes [8, 50]. Its inactivation, resulting in the lack of protein expression, is a very common event in pancreatic cancers [8, 15]. There is a statistical correlation between lack of p16 immunoreactivity and gene inactivation [49, 50]. Even if this correlation is not perfect, IHC is routinely used because it is the most cost effective way to infer gene inactivation, regardless the underlying molecular mechanism(s) [8]. The expression of p16 was scored based on the percentage neoplastic cells showing strong nuclear immunoreactivity. Tumors with $<$ 10% of positive neoplastic cells were classified as “negative”, tumors with $\geqslant$ 10% positive neoplastic cells were classified as “positive” [45]. Samples with a negative SMAD4 immunostaining or with a “1+” score were considered as “SMAD4-negative”, while specimens scored as “2+” or “3+” were considered as “SMAD4-positive”, according to what previously reported [45].

2.6 Statistical analysis

Categorical variables were compared using the Fisher exact test. Continuous variables were compared using the unpaired t-test or Mann-Whitney test. A p-value $<$ 0.05 was considered as statistically significant.

Kaplan Meier method and Cox regression analyses were used to evaluate the mutational status as predictor of survival in the study population in the multiple regression analysis. Statistical analyses were performed using GraphPad Prism 6.01 (GraphPad Software) and Stata/SE 14.2 statistical software.

3. Results

3.1 “Long-survivor” group characterization

The group was composed of 15 patients, 8 males (53.3%) and 7 females (46.7%), aged from 45 to 78 years (mean 62.3 ys). Follow-up time ranged from 55 to 133 months (mean 100.4 ms, median 105.0 ms) (Table 3). All patients but one (died from disease with lung metastases) were alive at the time of the last follow-up. Clinicopathological features and molecular status for each cohort patients are reported in Supplementary Table 1. Pancreatic ductal adenocarcinomas were classified as follows: 10 classic PDAC (two of them associated with Intraductal Papillary Mucinous Neoplasia – IPMN), and five morphological variants of PDAC (Supplementary Table 1). These latter were: one medullary carcinoma associated with IPMN, one colloid carcinoma, one undifferentiated carcinomas with osteoclast-like giant cells, one PDAC Large-gland type and one PDAC Large-duct type.

Table 3
Clinical and molecular data of the samples analyzed compared with data reported in the literature

Features	Long survivor (LS)	Non-long survivor (NLS)	p	Reported frequency of genetic alterations
Mean age (range)	62.3 ys (45–78)	66.0 ys (44–81)	0.3387	/
Sex			0.7152	/
Male	8	6
Female	7	9
Mean follow-up (range)	100.4 ms (55–133)	9.7 ms (4–23)	$<$ 0.0001	/
pT			0.3406	/
T1	2	0
T2	1	1
T3	12	14
pN			0.0092	/
N0	12	4
N1	3	11
Stage			0.2241	/
I	3	0
II	12	15
Resection margins			1.000	/
R0	14	14
R1	1	1
KRAS mutated samples	9/15 (60.0%)	13/15 (86.7%)	0.2148	90% [4, 58]
NRAS mutated samples ${}^{\text{a}}$	0/6	0/2	1.000	0.4% [21]
TP53 mutatedsamples	5/15 (33.3%)	7/15 (46.7%)	0.7104	50–75% [51, 58]
BRAF mutated samples ${}^{\text{a}}$	0/6	0/2	1.000	1.3% [21]
KRAS/TP53 mutated samples	3/15 (20.0%)	6/15 (40.0%)	0.4270	/
IDH1 mutated samples ${}^{\text{b}}$	0/9	0/12	1.000	0.3% [21]
MGMT methylated samples ${}^{\text{c}}$	6/14 (42.9%)	5/14 (35.7%)	1.000	12% [46]
MSI samples ${}^{\text{d}}$	0/9	0/7	1.000	0.3% [32]
P16
Mean % of positive cells (range)	8.3 (0–70)	16.1 (0–90)	0.7650
Cases with % positive cells $<$ 10%	10/15 (66.7%)	10/14 (71.4%)	1.000	90% ${}^{\text{f}}$ [21, 58]
SMAD4 ${}^{\text{e}}$
Neg/1+	2/13 (15.4%)	7/14 (50.0%)	0.1032	55% ${}^{\text{f}}$ [51]
2+/3+	11/13 (84.6%)	7/14 (50.0%)

${}^{\text{a}}$ NRAS and BRAF analysis was performed only in those cases with KRAS WT; ${}^{\text{b}}$ IDH1 analysis was performed in 21 samples; ${}^{\text{c}}$ In one “LS” case and in one “NLS” the MGMT promoter methylation status could not be determined; ${}^{\text{d}}$ MSI analysis was performed in 21 samples; ${}^{\text{e}}$ In two “LS” cases and one “NLS” case SMAD4 expression could not be assessed; ${}^{\text{f}}$ Loss of expression.

Twelve patients had a stage II disease (9 patients IIA and 3 patients IIB) and 3 a stage I disease (one patient IB, and 2 patients IA). The vast majority (12 of 15, 80.0%) were pT3 and had no lymph nodal involvement at presentation (Table 3). No distant metastases were observed at presentation in our cohort. Fourteen patients had negative resection margins (R0 disease), whereas 1 patient had positive margins (R1 disease) (Table 3).

In 9 patients (60.0%) a KRAS mutation was identified (Table 3). All the KRAS mutations were in exon 2. In the 6 patients wild-type for KRAS gene, no mutations in NRAS (exons 2-3-4) or in BRAF (exon 15) genes were detected (Table 3). Five patients (33.3%) harbored a TP53 mutation (Table 3). In 3 of these 5 patients also a KRAS mutation was observed (Table 3). No single nucleotide polymorphisms (SNPs) were detected in the amplicons analyzed.

To address tumor heterogeneity, we estimated the percentage of mutated neoplastic cells for KRAS or TP53 (see materials and methods): the mutated neoplastic cell range was 43–100% (mean 68.9%) and 71.4–88.9% (mean 81.5%), respectively.

Six of 14 evaluable samples (40.0%) were MGMT-methylated. In no one of the analyzed samples was detected MSI or mutation in IDH1 gene (Table 3).

In ten samples, the percentage of neoplastic cells positive for p16 were $<$ 10%; the percentage of neoplastic cells positive for p16 ranged from 0 to 70% (mean 8.3%).

Two samples were negative or scored as “1+” for SMAD4 staining, 11 were “2+” or “3+” and in two samples the staining was not evaluable (Table 3).

3.2 Comparison between “Long-survivors” and “Non-long survivors”

Clinicopathological and molecular features of the “NLS” group are described in Table 3 and Supplementary Table 1.

Clinicopathological features. The 15 “NLS” were matched with “LS” for age, sex, stage and margin positivity (Table 3). “LS” group showed less frequent lymph nodal involvement (20.0% vs 73.3%, $p=$ 0.0092). Pancreatic ductal adenocarcinomas were classified as follows: 14 classic PDAC (one of them associated with IPMN), and one PDAC large-gland type (Supplementary Table 1).

As expected, the survival time from surgery was significantly higher in the “LS” group (88.1 ms vs 9.7 ms, $p<$ 0.0001) (Table 3).

Molecular status. The “LS” cases had a lower frequency of KRAS mutations (60.0% vs 86.7%), although the difference did not reach statistical significance ( $p=$ 0.2148, Fisher’s exact test). All KRAS mutations in the “LS” group were in exon 2. In the “NLS” group we identified KRAS mutations also in exon 3 (one case, p.Q61H) and in exon 4 (one case, p.E143K). To the best of our knowledge, the p.E143K mutation has never been reported in pancreatic cancer, but it has been previously described in colorectal carcinoma [44]. The “LS” cases had a lower frequency of TP53 mutations (33.3% vs 46.7%), although the difference was not statistically significant ( $p=$ 0.7104). No single nucleotide polymorphisms (SNPs) were detected in the amplicons analyzed.

To address tumor heterogeneity, we estimated the percentage of mutated neoplastic cells for KRAS or TP53 (see materials and methods): the mutated neoplastic cell range was 30–100% (mean 70.2%) and 30–100% (mean 64.9%), respectively.

No significant differences were observed in MGMT promoter methylation status between “LS” (42.9%) and “NLS” (33.3%) ( $p=$ 0.7104). MSI and IDH1 mutations were not detected in either the “LS” or in the “NLS” group. No differences were observed in p16 expression between “LS” and “NLS” ( $p=$ 1.000). The “LS” cases showed a higher percentage (86.4%) of SMAD4 positive samples if compared to “NLS” (50.0%), but the difference did not reach statistical significance ( $p=$ 0.1032).

3.3 Overall survival according to the status of molecular markers

To evaluate if molecular markers predicted prognosis, we evaluated patient survival according to the molecular status of their tumors from the time surgery for all cases (15 “LS” and 15 “NLS”).

Patients with KRAS-WT carcinomas had a survival 1.5-fold higher than patients with tumors KRAS-mutated (77.1 ms vs 47.1 ms, respectively; $p=$ 0.0577 Mann Whitney test; $p=$ 0.1138 Log-rank test; $p=$ 0.0663 Breslow-Wilcoxon test) (Table 4, Fig. 1A). Subjects with TP53-WT tumors had a mean survival higher than those with TP53-mutated tumors (62.0 ms vs 44.7 ms, respectively; $p=$ 0.4204 Mann Whitney test; $p=$ 0.0129 Log-rank test; $p=$ 0.0102 Breslow-Wilcoxon test) (Table 4, Fig. 1B).

Table 4
Survival analysis according to the status of molecular markers

Markers	OS mutated (range)	OS WT (range)	p
KRAS	47.1 ms (4–133)	77.1 ms (13–126)	0.0663
TP53	44.7 ms (4–133)	62.0 ms (13–120)	0.0102
	OS methylated	OS unmethylated	p
MGMT	59.0 ms (4–126)	53.1 ms (4–133)	0.8443
	OS IHC negative	OS IHC positive	p
p16 (positive if $\geqslant$ 10%)	53.3 ms	55.4 ms	0.8662
SMAD4 (neg/1+ vs 2+/3+)	35.2 ms	64.2 ms	0.1716
Combined markers	OS (ms)		p
KRASwt/TP53wt vs KRASmut/TP53mut	90.2 vs 41.1		0.0717
KRASwt/SMAD4pos vs KRASmut/SMAD4neg	83.8 vs 36.8		0.0526
TP53wt/SMAD4pos vs TP53mut/SMAD4neg	65.1 vs 12.3		0.1144

OS: Overall Survival; IHC: immunohistochemistry; WT: Wild-Type; mut: mutated; ms: months; pos: positive; neg: negative.

Figure 1.

Survival of 30 PDAC patients (15 long surviving and 15 controls) according to the status of molecular markers (Kaplan-Meyer curves). A) KRAS mutated vs. wild type cases. B) TP53 mutated vs. wild type cases. C) SMAD4 loss-of-expression cases vs. cases with retained expression. D) $p16$ loss-of-expression cases vs. cases with retained expression.

SMAD4 positive cases (scores 2+ and 3+) had a mean survival nearly double than that of SMAD4 negative ones (scores 0 and 1+) (56.4 ms vs 30.2 ms), but the difference did not reach statistical significance ( $p=$ 0.1171, Mann Whitney test; $p=$ 0.1461 Log-rank test; $p=$ 0.1130 Breslow-Wilcoxon test) (Table 4, Fig. 1C). No difference in survival was observed between p16 positive (mean: 55.4 ms) and negative cases (53.3 ms) ( $p=$ 0.8662, Mann Whitney test; $p=$ 0.3765 Log-rank test; $p=$ 0.2951 Breslow-Wilcoxon test) (Table 4, Fig. 1D).

No difference in mean survival was also observed between MGMT promoter-methylated samples and those that did not show promoter methylation (59.9 ms vs 53.1 ms; $p=$ 0.8439; Mann Whitney test; $p=$ 0.8262 Log-rank test; $p=$ 0.8443 Breslow-Wilcoxon test) (Table 4).

Figure 2.

Survival of 30 PDAC patients (15 long surviving and 15 controls) according to the pattern of combined molecular alterations (Kaplan-Meyer curves). A) KRAS and TP53 mutated cases vs. cases wild-type for both. B) Mean survival of cases according to KRAS and TP53status. C) KRAS mutated and SMAD4 negative cases vs. cases KRAS wild-type and SMAD4-positive. D) Mean survival according to KRAS and SMAD4 status. E) TP53 mutated and SMAD4 negative cases vs. cases TP53 wild-type and SMAD4-positive. F) Mean survival according to TP53 and SMAD4 status. KR+TP53: KRAS and TP53 mutated cases; KR+SMAD4: KRAS mutated and SMAD4negative cases; KR: cases mutated only for KRAS; TP53: cases mutated only for TP53; WT/WT: cases wild-type for both gene sets analyzed (KRAS and TP53; KRAS and SMAD4).

Intriguingly, when we compared the survival of cases that were KRAS and TP53 mutated with that of cases that were wild type for both, we observed that the survival of the latter group was more than twice longer (90.2 ms vs 41.1 ms, $p=$ 0.0894 Mann Whitney test; $p=$ 0.0825, Log-rank Mantel-Cox test, $p=$ 0.0717 Breslow-Wilcoxon Test) (Table 4, Fig. 2). The difference was also higher if the median survival time was considered (105.0 ms in WT/WT patients, 12.0 ms in KRAS+TP53 mutated patients). No differences were observed in tumor stage between the two groups (KRAS-WT/TP53-WT and KRAS+TP53 mutated). Patients with tumors harboring only KRAS (but no TP53) mutations had a mean survival time of 51.4 months, while patients withTP53 (but no KRAS)-mutated tumors had a mean survival time of 55.3 months (Table 4, Fig. 2).

Patients with PDAC KRAS-WT and with positive expression of SMAD4 had a survival time twice longer than subjects with a PDAC KRAS-mutated and with negative expression of SMAD4 (83.8 months vs 36.7 months; $p=$ 0.1181 Mann Whitney test; $p=$ 0.1051 Log-rank Mantel-Cox test, $p=$ 0.0526 Breslow-Wilcoxon Test) (Table 4 and Fig. 2).

Patients with TP53-WT PDAC and positive expression of SMAD4 had a survival time five-times longer than subjects with a TP53-mutated PDAC and negative expression of SMAD4 (65.1 months vs 12.3 months; $p=$ 0.1181 Mann Whitney test; $p=$ 0.0626 Log-rank Mantel-Cox test, $p=$ 0.1144 Breslow-Wilcoxon Test) (Fig. 2).

The “LS” and the “NLS” group differed with regards to the presence of lymph node metastases that was statistically lower in the “LS” group (see above). For this reason, to further investigate the relationship of KRAS and TP53 mutations, and of SMAD4 loss of expression with survival we performed multivariate analysis. The hazard ratio (HR) for poor survival was 2.72 (95% confidence interval – CI: 0.33–22.2; $p=$ 0.35) in patients without lymph node metastases (N0) and KRAS or TP53 mutation; the HR increased to 3.82 (95% CI: 0.44–33.0; $p=$ 0.22) in patients without lymph node metastases and KRAS and TP53 mutation. The patients with lymph node metastases (N1) and without mutations in KRAS and TP53 had a HR for poor survival of 3.88 (95% CI: 1.29–11.63; $p=$ 0.015); the HR increased 10.56 (95% CI: 1.20–93.08; $p=$ 0.034) in patients with lymph node metastases and KRAS or TP53 mutations; the HR further increased to 14.84 (95% CI: 1.55–141.73; $p=$ 0.019) in patients with lymph node metastases and both KRAS and TP53 mutations. The presence of KRAS mutation and SMAD4 loss of expression or TP53 mutation and SMAD4 loss of expression did not affect the HR for patient survival, regardless of lymph node status (data not shown).

Too few cases lacked gene alterations in KRAS and TP53 and SMAD4 (triple negative cases KRAS-WT and TP53-WT and SMAD4-positive; $n=$ 4: 1 in “NLS” group, and 3 in the “LS” group) or had molecular alterations in KRAS and TP53 and SMAD4 (triple positive cases KRAS-mutated and TP53-mutated and SMAD4-negative; $n=$ 2: 2 in “NLS” group, and 0 in the “LS” group) for statistical analysis.

4. Discussion

Alterations of KRAS, TP53, SMAD4 and CDKN2A/ p16 define the molecular profile of PDAC, but their prognostic and clinical implications are still controversial [5, 6, 22, 48, 52]. One of the reasons is the dismal prognosis of most patients with PDAC: the few operable patients often die from disease within a few months after surgery. The background of genetic alterations in the few patients that survive years after surgery is not well documented. In the present study, we have analyzed the status of the KRAS, TP53, p16 and SMAD4 in tumors from long survivors of PDAC (“LS” patients). We also investigated the molecular status of IDH1, microsatellite instability and MGMT promoter methylation, all of which have been reported with low prevalence in PDAC, reasoning that their prevalence may be higher in cases associated with an unusually long survival.

We showed that tumors from long PDAC survivors have the same general alterations found in those of patients that follow the usually dismal course, similar to what reported by Dal Molin et al. [11].

However, we found some important differences after comparison with a PDAC control group (“NLS” patients) from our institution and with the data reported in large series in the literature. Tumors of long PDAC survivors from our group had a remarkably lower prevalence of KRAS mutation and of SMAD4 loss of expression. We also found that absence of KRAS and TP53 mutations, and that absence of KRAS mutations combined with positive SMAD4 expression correlated with longer survival. We have observed that “LS” patients have a low frequency of KRAS mutations (60.0%) when compared with “NLS” patients (86.7%) and with data reported in literature (90%). Also TP53 mutations are less frequent in tumors from “LS” patients when compared to “NLS” PDAC patient controls and data from the literature, and the survival of cases with TP53 mutation was higher than that of the TP53-WT cases. Interestingly, the coexistence of KRAS and TP53 mutations was correlated with survival. Patients without KRAS and TP53 mutations had a survival rate more than twice longer when compared with subjects carrying tumors with KRAS and TP53 mutations. Furthermore, multivariate analysis showed that patients without KRAS and TP53 mutations had a lower risk of dying of disease compared with patients with KRAS and TP53 mutated tumors, regardless of lymph node metastases. The risk difference was statically significant in patients with lymph node involvement. These observations appears in line with experimental data showing how oncogenic KRAS species are not only involved with tumor development and progression [14], but also promote the development of adenocarcinoma from pancreatic ductal cells when TP53 is inactivated by mutation [3].

Regarding the CDKN2A/p16 status, no major differences were observed in p16 immunohistochemical reactivity between tumors from “LS” and “NLS” patients. On the other hand, SMAD4 was less frequently lost in tumors from “LS” (15.4%) compared with “NLS” patients and data from the literature (50% and 55% respectively). Moreover, absence of alterations in KRAS and SMAD4 seem to identify a subset of patients with a twice longer survival compared with cases harboring both KRAS mutation and SMAD4 loss of expression.

This data indicates that absence of KRAS and TP53 mutations and a preserved SMAD4 expression may identify patients with a favorable prognosis. Similar results have been recently observed in lung adenocarcinomas, where KRAS and TP53 mutations identify patients with worse clinical outcome [36].

We have also investigated whether MGMT promoter methylation, IDH1 mutations or microsatellite instability may be enriched in tumors from “LS” patients.

MGMT promoter hypermethylation confers sensitivity to alkylating agents (e.g. temozolomide) in patients with gliomas [27]. Contrary to what observed by Ueki et al. [53], but in accordance with Peng et al. [46], we observed that the MGMT promoter can be hypermethylated in PDAC. The higher percentage of MGMT promoter-methylated PDAC found in our cohort compared to that reported by Peng et al. may be due to the higher sensitivity of the method adopted in our study. This epigenetic modification has been observed both in the “LS” and “NLS” groups, but MGMT promoter methylation did not influence survival. Nevertheless, MGMT promoter methylation may be considered as a predictive marker for therapy with temozolamide as in the case of gliomas. The treatment of advanced pancreatic cancer with temozolamide has been attempted in a phase II study in 1998 [40], but no clinical response was seen in the 15 patients subjected to treatment. It is worth noting that, in the study, the treatment was performed without selecting patients according to MGMT promoter methylation status. Efficacy of alkylating agents has been demonstrated in neuroendocrine forms of pancreatic cancer [20].

Somatic mutations in the isocitrate dehydrogenase 1 gene (IDH1) have been described in acute myelogenous leukemia, cartilaginous tumors and gliomas [10], and have been linked to epigenetic deregulation that contributes to tumorigenesis [31, 35]. Few studies have investigated IDH1 in PDAC. In our series, no exon 4 IDH1 mutations were identified, a result compatible with the very low prevalence of mutations found in other studies [7, 9].

The risk of developing PDAC is increased in the Lynch syndrome caused by germ-line mutations in mismatch repair genes [25]. Tumors with mismatch repair defects typically feature a high degree of microsatellite instability (MSI) [30, 33]. The MSI-High phenotype is frequent in colorectal cancers, mainly in mucinous types arising in the right side of the colon and has been associated with a relatively favorable prognosis. Most pancreatic adenocarcinomas have a MSI-stable or sometimes a MSI-Low phenotype [2, 23, 24, 32, 37]. Only the few pancreatic adenocarcinomas occurring in the setting of Lynch syndrome and/or exhibiting poorly differentiated medullary histopathology display the MSI-High phenotype [2, 32, 37, 57, 59]. All tumors analyzed from our “LS” patients were MSI-stable, as were those from “NLS” controls. None of our cases was associated with Lynch syndrome or showed poorly differentiated medullary features. Thus, our data do not support a correlation between the MSI-High phenotype and prognosis, at variance with what reported by Nakata and colleagues [43].

Various clinical factors such as tumor size and differentiation, lymphatic and/or vascular invasion, tumor stage, involvement of resection margins influence outcome of PDAC patients. However, these parameters cannot be evaluated (or can only be partially defined) before surgery is performed and the tumor removed. Alterations in KRAS, TP53 and SMAD4 can easily be assessed in the specimens used to establish a pre-operative diagnosis of adenocarcinoma (e.g. EUS-FNA and cytology samples) [13]. For mutation detection, we used a highly sensitive NGS protocol that requires a limited amount of input DNA and outperforms Sanger sequencing or commercial kits [12, 13]. In addition, NGS allows to evaluate the proportion of mutated alleles, an important piece of information to define a mutation as sub-clonal. Genetic heterogeneity has been reported in several tumors as well as in pancreatic cancers [34, 56]. Considering that pre-operative samples contain only a small part of the lesion, molecular heterogeneity is a well-known general limitation for the analysis of limited pre-operative samples. In our work, we addressed this problem by estimating the proportion of neoplastic cells carrying specific mutations (KRAS, TP53) and did not find genetic heterogeneity a relevant issue.

Applying a highly sensitive NGS protocol to pre-surgical samples may provide important information that the surgeon can integrate with the endoscopic evaluation and the pathological diagnosis to establish the best clinical management for each patient.

This cohort of patients is limited. Even if “LS” patients and a “NLS” control group are not clearly identified by specific pattern of mutations, the absence of KRAS, TP53 and SMAD4 alterations did identify in our study a subset of tumors with more favorable outcome. Considering how long survival of patients with pancreatic adenocarcinoma is exceptional, the relatively low prevalence of KRAS, TP53 and SMAD4 alterations in tumors from these patients definitely deserves additional investigation.

Footnotes

Acknowledgments

This work was supported in part by an Italian Government-Ministero Della Salute Grant No. RF-2011-02350857 to G.T.

Conflict of interest

The authors have nothing to disclose.

Supplementary data

The supplementary files are available to download from http://dx.doi.org/10.3233/CBM-170464.

References

Adham

Jaeck

Le Borgne

Oussoultzouglou

Chenard-Neu

M.P.

Mosnier

J.F.

Scoazec

J.Y.

Mornex

and Partensky

, Long-term survival (5–20 years) after pancreatectomy for pancreatic ductal adenocarcinoma: a series of 30 patients collected from 3 institutions, Pancreas 37 (2008), 352–357.

Alexander

Watanabe

T.T.

Rashid

and Hamilton

S.R.

, Histopathological identification of colon cancer with microsatellite instability, The American Journal of Pathology 158 (2001), 527–535.

Bailey

J.M.

Hendley

A.M.

Lafaro

K.J.

Pruski

M.A.

Jones

N.C.

Alsina

Younes

Maitra

McAllister

Iacobuzio-Donahue

C.A.

and Leach

S.D.

, p53 mutations cooperate with oncogenic Kras to promote adenocarcinoma from pancreatic ductal cells, Oncogene 35 (2016), 4282–4288.

Bailey

Chang

D.K.

Nones

Johns

A.L.

Patch

A.M.

Gingras

M.C.

Miller

D.K.

Christ

A.N.

Bruxner

T.J.

Quinn

M.C.

Nourse

Murtaugh

L.C.

Harliwong

Idrisoglu

Manning

Nourbakhsh

Wani

Fink

Holmes

Chin

Anderson

M.J.

Kazakoff

Leonard

Newell

Waddell

Wood

Wilson

P.J.

Cloonan

Kassahn

K.S.

Taylor

Quek

Robertson

Pantano

Mincarelli

Sanchez

L.N.

Evers

Pinese

Cowley

M.J.

Jones

M.D.

Colvin

E.K.

Nagrial

A.M.

Humphrey

E.S.

Chantrill

L.A.

Mawson

Humphris

Chou

Pajic

Scarlett

C.J.

Pinho

A.V.

Giry-Laterriere

Rooman

Samra

J.S.

Kench

J.G.

Lovell

J.A.

Merrett

N.D.

Toon

C.W.

Epari

Nguyen

N.Q.

Barbour

Zeps

Moran-Jones

Jamieson

N.B.

Graham

J.S.

Duthie

Oien

Hair

Grutzmann

Maitra

Iacobuzio-Donahue

C.A.

Wolfgang

C.L.

Morgan

R.A.

Lawlor

R.T.

Corbo

Bassi

Rusev

Capelli

Salvia

Tortora

Mukhopadhyay

Petersen

G.M.

Munzy

D.M.

Fisher

W.E.

Karim

S.A.

Eshleman

J.R.

Hruban

R.H.

Pilarsky

Morton

J.P.

Sansom

O.J.

Scarpa

Musgrove

E.A.

Bailey

U.M.

Hofmann

Sutherland

R.L.

Wheeler

D.A.

Gill

A.J.

Gibbs

R.A.

Pearson

J.V.

Biankin

A.V.

and Grimmond

S.M.

, Genomic analyses identify molecular subtypes of pancreatic cancer, Nature 531 (2016), 47–52.

Biankin

A.V.

Morey

A.L.

Lee

C.S.

Kench

J.G.

Biankin

S.A.

Hook

H.C.

Head

D.R.

Hugh

T.B.

Sutherland

R.L.

and Henshall

S.M.

, DPC4/Smad4 expression and outcome in pancreatic ductal adenocarcinoma, Journal of Clinical Oncology: Official Journal of the American Society of Clinical Oncology 20 (2002), 4531–4542.

Blackford

Serrano

O.K.

Wolfgang

C.L.

Parmigiani

Jones

Zhang

Parsons

D.W.

Lin

J.C.

Leary

R.J.

Eshleman

J.R.

Goggins

Jaffee

E.M.

Iacobuzio-Donahue

C.A.

Maitra

Cameron

J.L.

Olino

Schulick

Winter

Herman

J.M.

Laheru

Klein

A.P.

Vogelstein

Kinzler

K.W.

Velculescu

V.E.

and Hruban

R.H.

, SMAD4 gene mutations are associated with poor prognosis in pancreatic cancer, Clinical Cancer Research: An Official Journal of the American Association for Cancer Research 15 (2009), 4674–4679.

Borger

D.R.

Tanabe

K.K.

Fan

K.C.

Lopez

H.U.

Fantin

V.R.

Straley

K.S.

Schenkein

D.P.

Hezel

A.F.

Ancukiewicz

Liebman

H.M.

Kwak

E.L.

Clark

J.W.

Ryan

D.P.

Deshpande

Dias-Santagata

Ellisen

L.W.

Zhu

A.X.

and Iafrate

A.J.

, Frequent mutation of isocitrate dehydrogenase (IDH)1 and IDH2 in cholangiocarcinoma identified through broad-based tumor genotyping, The Oncologist 17 (2012), 72–79.

Bosman

F.T.

Carneiro

Hruban

R.H.

and Theise

N.D.

, WHO Classification of Tumours of the Digestive System, Fourth ed., IARC, Lyon, France, 2010.

Brody

J.R.

Yabar

C.S.

Zarei

Bender

Matrisian

L.M.

Rahib

Heartwell

Mason

Yeo

C.J.

Peiper

S.C.

Jiang

Varieur

Madhavan

Petricoin

, 3rd Fortuna

Curtis

Wang

Z.X.

Pishvaian

M.J.

and Winter

J.M.

, Identification of a novel metabolic-related mutation (IDH1) in metastatic pancreatic cancer, Cancer Biology & Therapy, 2016.

10.

Cohen

A.L.

Holmen

S.L.

and Colman

, IDH1 and IDH2 mutations in gliomas, Current Neurology and Neuroscience Reports 13 (2013), 345.

11.

Dal Molin

Zhang

de Wilde

R.F.

Ottenhof

N.A.

Rezaee

Wolfgang

C.L.

Blackford

Vogelstein

Kinzler

K.W.

Papadopoulos

Hruban

R.H.

Maitra

and Wood

L.D.

, Very Long-term Survival Following Resection for Pancreatic Cancer Is Not Explained by Commonly Mutated Genes: Results of Whole-Exome Sequencing Analysis, Clinical Cancer Research: An Official Journal of the American Association for Cancer Research 21 (2015), 1944–1950.

12.

de Biase

de Luca

Gragnano

Visani

Bellevicine

Malapelle

Tallini

and Troncone

, Fully automated PCR detection of KRAS mutations on pancreatic endoscopic ultrasound fine-needle aspirates, Journal of Clinical Pathology, 2016.

13.

de Biase

Visani

Baccarini

Polifemo

A.M.

Maimone

Fornelli

Giuliani

Zanini

Fabbri

Pession

and Tallini

, Next generation sequencing improves the accuracy of KRAS mutation analysis in endoscopic ultrasound fine needle aspiration pancreatic lesions, PloS One 9 (2014), e87651.

14.

di Magliano

M.P.

and Logsdon

C.D.

, Roles for KRAS in pancreatic tumor development and progression, Gastroenterology 144 (2013), 1220–1229.

15.

Ducreux

Cuhna

A.S.

Caramella

Hollebecque

Burtin

Goere

Seufferlein

Haustermans

Van Laethem

J.L.

Conroy

and Arnold

, Cancer of the pancreas: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up, Annals of Oncology: Official Journal of the European Society for Medical Oncology 26(Suppl 5) (2015), v56–68.

16.

Edge

S.B.

Byrd

D.R.

Compton

C.C.

Fritz

A.G.

Greene

F.L.

and Trotti

, AJCC cancer staging manual Springer, New York, NY, 2010.

17.

Eser

Schnieke

Schneider

and Saur

, Oncogenic KRAS signalling in pancreatic cancer, British Journal of Cancer 111 (2014), 817–822.

18.

Ferrone

C.R.

Brennan

M.F.

Gonen

Coit

D.G.

Fong

Chung

Tang

Klimstra

and Allen

P.J.

, Pancreatic adenocarcinoma: the actual 5-year survivors, Journal of Gastrointestinal Surgery: Official Journal of the Society for Surgery of the Alimentary Tract 12 (2008), 701–706.

19.

Ferrone

C.R.

Pieretti-Vanmarcke

Bloom

J.P.

Zheng

Szymonifka

Wargo

J.A.

Thayer

S.P.

Lauwers

G.Y.

Deshpande

Mino-Kenudson

Fernandez-del Castillo

Lillemoe

K.D.

and Warshaw

A.L.

, Pancreatic ductal adenocarcinoma: long-term survival does not equal cure, Surgery 152 (2012), S43–49.

20.

Fine

R.L.

Gulati

A.P.

Krantz

B.A.

Moss

R.A.

Schreibman

Tsushima

D.A.

Mowatt

K.B.

Dinnen

R.D.

Mao

Stevens

P.D.

Schrope

Allendorf

Lee

J.A.

Sherman

W.H.

and Chabot

J.A.

, Capecitabine and temozolomide (CAPTEM) for metastatic, well-differentiated neuroendocrine cancers: The Pancreas Center at Columbia University experience, Cancer Chemotherapy and Pharmacology 71 (2013), 663–670.

21.

Forbes

S.A.

Beare

Boutselakis

Bamford

Bindal

Tate

Cole

C.G.

Ward

Dawson

Ponting

Stefancsik

Harsha

Kok

C.Y.

Jia

Jubb

Sondka

Thompson

and Campbell

P.J.

, COSMIC: somatic cancer genetics at high-resolution, Nucleic Acids Research 45 (2017), D777–D783.

22.

Gerdes

Ramaswamy

Ziegler

Lang

S.A.

Kersting

Baumann

Wild

Moll

Rothmund

and Bartsch

D.K.

, p16INK4a is a prognostic marker in resected ductal pancreatic cancer: an analysis of p16INK4a, p53, MDM2, an Rb, Annals of Surgery 235 (2002), 51–59.

23.

Ghimenti

Tannergard

Wahlberg

Liu

Giulianotti

P.G.

Mosca

Fornaciari

Bevilacqua

Lindblom

and Caligo

M.A.

, Microsatellite instability and mismatch repair gene inactivation in sporadic pancreatic and colon tumours, British Journal of Cancer 80 (1999), 11–16.

24.

Goggins

Offerhaus

G.J.

Hilgers

Griffin

C.A.

Shekher

Tang

Sohn

T.A.

Yeo

C.J.

Kern

S.E.

and Hruban

R.H.

, Pancreatic adenocarcinomas with DNA replication errors (RER+) are associated with wild-type K-ras and characteristic histopathology. Poor differentiation, a syncytial growth pattern, and pushing borders suggest RER+, The American Journal of Pathology 152 (1998), 1501–1507.

25.

Grover

and Syngal

, Hereditary pancreatic cancer, Gastroenterology 139 (2010), 1076–1080, 1080 e1071-1072.

26.

Han

S.S.

Jang

J.Y.

Kim

S.W.

Kim

W.H.

Lee

K.U.

and Park

Y.H.

, Analysis of long-term survivors after surgical resection for pancreatic cancer, Pancreas 32 (2006), 271–275.

27.

Hegi

M.E.

Diserens

A.C.

Gorlia

Hamou

M.F.

de Tribolet

Weller

Kros

J.M.

Hainfellner

J.A.

Mason

Mariani

Bromberg

J.E.

Hau

Mirimanoff

R.O.

Cairncross

J.G.

Janzer

R.C.

and Stupp

, MGMT gene silencing and benefit from temozolomide in glioblastoma, The New England Journal of Medicine 352 (2005), 997–1003.

28.

Hidalgo

, Pancreatic cancer, The New England Journal of Medicine 362 (2010), 1605–1617.

29.

Jones

Zhang

Parsons

D.W.

Lin

J.C.

Leary

R.J.

Angenendt

Mankoo

Carter

Kamiyama

Jimeno

Hong

S.M.

Lin

M.T.

Calhoun

E.S.

Kamiyama

Walter

Nikolskaya

Nikolsky

Hartigan

Smith

D.R.

Hidalgo

Leach

S.D.

Klein

A.P.

Jaffee

E.M.

Goggins

Maitra

Iacobuzio-Donahue

Eshleman

J.R.

Kern

S.E.

Hruban

R.H.

Karchin

Papadopoulos

Parmigiani

Vogelstein

Velculescu

V.E.

and Kinzler

K.W.

, Core signaling pathways in human pancreatic cancers revealed by global genomic analyses, Science 321 (2008), 1801–1806.

30.

Kastrinos

Mukherjee

Tayob

Wang

Sparr

Raymond

V.M.

Bandipalliam

Stoffel

E.M.

Gruber

S.B.

and Syngal

, Risk of pancreatic cancer in families with Lynch syndrome, Jama 302 (2009), 1790–1795.

31.

King

Selak

M.A.

and Gottlieb

, Succinate dehydrogenase and fumarate hydratase: linking mitochondrial dysfunction and cancer, Oncogene 25 (2006), 4675–4682.

32.

Laghi

Beghelli

Spinelli

Bianchi

Basso

Di Caro

Brecht

Celesti

Turri

Bersani

Schumacher

Rocken

Grantzdorffer

Roncalli

Zerbi

Neuhaus

Bassi

Montorsi

Scarpa

and Malesci

, Irrelevance of microsatellite instability in the epidemiology of sporadic pancreatic ductal adenocarcinoma, PloS One 7 (2012), e46002.

33.

Laghi

Bianchi

and Malesci

, Differences and evolution of the methods for the assessment of microsatellite instability, Oncogene 27 (2008), 6313–6321.

34.

Laghi

Orbetegli

Bianchi

Zerbi

Di Carlo

Boland

C.R.

and Malesci

, Common occurrence of multiple K-RAS mutations in pancreatic cancers with associated precursor lesions and in biliary cancers, Oncogene 21 (2002), 4301–4306.

35.

Rajeshkumar

N.V.

Maitra

and Dang

C.V.

, Conceptual framework for cutting the pancreatic cancer fuel supply, Clinical Cancer Research: An Official Journal of the American Association for Cancer Research 18 (2012), 4285–4290.

36.

Lococo

Gandolfi

Rossi

Pinto

Rapicetta

Cavazza

Cesario

Galeone

Paci

and Ciarrocchi

, Deep Sequencing Analysis Reveals That KRAS Mutation Is a Marker of Poor Prognosis in Patients with Pulmonary Sarcomatoid Carcinoma, Journal of Thoracic Oncology: Official Publication of the International Association for the Study of Lung Cancer 11 (2016), 1282–1292.

37.

Maple

J.T.

Smyrk

T.C.

Boardman

L.A.

Johnson

R.A.

Thibodeau

S.N.

and Chari

S.T.

, Defective DNA mismatch repair in long-term (> or =3 years) survivors with pancreatic cancer, Pancreatology 5 (2005), 220–227; discussion 227–228.

38.

McDowell

B.D.

Chapman

C.G.

Smith

B.J.

Button

A.M.

Chrischilles

E.A.

and Mezhir

J.J.

, Pancreatectomy predicts improved survival for pancreatic adenocarcinoma: results of an instrumental variable analysis, Annals of Surgery 261 (2015), 740–745.

39.

Miller

K.D.

Siegel

R.L.

Lin

C.C.

Mariotto

A.B.

Kramer

J.L.

Rowland

J.H.

Stein

K.D.

Alteri

and Jemal

, Cancer treatment and survivorship statistics, 2016, CA: A Cancer Journal for Clinicians 66 (2016), 271–289.

40.

Moore

M.J.

Feld

Hedley

Oza

and Siu

L.L.

, A phase II study of temozolomide in advanced untreated pancreatic cancer, Investigational New Drugs 16 (1998), 77–79.

41.

Morandi

de Biase

Visani

Monzoni

Tosi

Brulatti

Turchetti

Baccarini

Tallini

and Pession

, T([20]) repeat in the 3’-untranslated region of the MT1X gene: a marker with high sensitivity and specificity to detect microsatellite instability in colorectal cancer, International Journal of Colorectal Disease 27 (2012), 647–656.

42.

Morandi

Franceschi

de Biase

Marucci

Tosoni

Ermani

Pession

Tallini

and Brandes

, Promoter methylation analysis of O6-methylguanine-DNA methyltransferase in glioblastoma: detection by locked nucleic acid based quantitative PCR using an imprinted gene (SNURF) as a reference, BMC Cancer 10 (2010), 48.

43.

Nakata

Wang

Y.Q.

Yashiro

Nishioka

Tanaka

Ohira

Ishikawa

Nishino

and Hirakawa

, Prognostic value of microsatellite instability in resectable pancreatic cancer, Clinical Cancer Research: An Official Journal of the American Association for Cancer Research 8 (2002), 2536–2540.

44.

Naser

W.M.

Shawarby

M.A.

Al-Tamimi

D.M.

Seth

Al-Quorain

Nemer

A.M.

and Albagha

O.M.

, Novel KRAS gene mutations in sporadic colorectal cancer, PloS One 9 (2014), e113350.

45.

Oshima

Okano

Muraki

Haba

Maeba

Suzuki

and Yachida

, Immunohistochemically detected expression of 3 major genes (CDKN2A/p16, TP53, and SMAD4/DPC4) strongly predicts survival in patients with resectable pancreatic cancer, Annals of Surgery 258 (2013), 336–346.

46.

Peng

D.F.

Kanai

Sawada

Ushijima

Hiraoka

Kitazawa

and Hirohashi

, DNA methylation of multiple tumor-related genes in association with overexpression of DNA methyltransferase 1 (DNMT1) during multistage carcinogenesis of the pancreas, Carcinogenesis 27 (2006), 1160–1168.

47.

Riall

T.S.

Cameron

J.L.

Lillemoe

K.D.

Winter

J.M.

Campbell

K.A.

Hruban

R.H.

Chang

and Yeo

C.J.

, Resected periampullary adenocarcinoma: 5-year survivors and their 6- to 10-year follow-up, Surgery 140 (2006), 764–772.

48.

Salek

Minarikova

Benesova

Nosek

Strnad

Zavoral

and Minarik

, Mutation status of K-ras, p53 and allelic losses at 9p and 18q are not prognostic markers in patients with pancreatic cancer, Anticancer Research 29 (2009), 1803–1810.

49.

Smigiel

Sasiadek

Krecicki

Ramsey

Jagielski

and Blin

, Inactivation of the cyclin-dependent kinase inhibitor 2A (CDKN2A) gene in squamous cell carcinoma of the larynx, Molecular Carcinogenesis 39 (2004), 147–154.

50.

Tadokoro

Shigihara

Ikeda

Takase

and Suyama

, Two distinct pathways of p16 gene inactivation in gallbladder cancer, World Journal of Gastroenterology 13 (2007), 6396–6403.

51.

Takai

and Yachida

, Genomic alterations in pancreatic cancer and their relevance to therapy, World Journal of Gastrointestinal Oncology 7 (2015), 250–258.

52.

Tascilar

Skinner

H.G.

Rosty

Sohn

Wilentz

R.E.

Offerhaus

G.J.

Adsay

Abrams

R.A.

Cameron

J.L.

Kern

S.E.

Yeo

C.J.

Hruban

R.H.

and Goggins

, The SMAD4 protein and prognosis of pancreatic ductal adenocarcinoma, Clinical Cancer Research: An Official Journal of the American Association for Cancer Research 7 (2001), 4115–4121.

53.

Ueki

Toyota

Sohn

Yeo

C.J.

Issa

J.P.

Hruban

R.H.

and Goggins

, Hypermethylation of multiple genes in pancreatic adenocarcinoma, Cancer Research 60 (2000), 1835–1839.

54.

Vincent

Herman

Schulick

Hruban

R.H.

and Goggins

, Pancreatic cancer, Lancet 378 (2011), 607–620.

55.

Visani

Acquaviva

Marucci

Paccapelo

Mura

Franceschi

Grifoni

Pession

Tallini

Brandes

A.A.

and de Biase

, Non-canonical IDH1 and IDH2 mutations: a clonal and relevant event in an Italian cohort of gliomas classified according to the 2016 World Health Organization (WHO) criteria, Journal of Neuro-Oncology, 2017.

56.

Visani

de Biase

Baccarini

Fabbri

Polifemo

A.M.

Zanini

Pession

and Tallini

, Multiple KRAS mutations in pancreatic adenocarcinoma: molecular features of neoplastic clones indicate the selection of divergent populations of tumor cells, International Journal of Surgical Pathology 21 (2013), 546–552.

57.

Wilentz

R.E.

Goggins

Redston

Marcus

V.A.

Adsay

N.V.

Sohn

T.A.

Kadkol

S.S.

Yeo

C.J.

Choti

Zahurak

Johnson

Tascilar

Offerhaus

G.J.

Hruban

R.H.

and Kern

S.E.

, Genetic, immunohistochemical, and clinical features of medullary carcinoma of the pancreas: A newly described and characterized entity, The American Journal of Pathology 156 (2000), 1641–1651.

58.

Witkiewicz

A.K.

McMillan

E.A.

Balaji

Baek

Lin

W.C.

Mansour

Mollaee

Wagner

K.U.

Koduru

Yopp

Choti

M.A.

Yeo

C.J.

McCue

White

M.A.

and Knudsen

E.S.

, Whole-exome sequencing of pancreatic cancer defines genetic diversity and therapeutic targets, Nature Communications 6 (2015), 6744.

59.

Wolfgang

C.L.

Herman

J.M.

Laheru

D.A.

Klein

A.P.

Erdek

M.A.

Fishman

E.K.

and Hruban

R.H.

, Recent progress in pancreatic cancer, CA: A Cancer Journal for Clinicians 63 (2013), 318–348.

Long-term survivors of pancreatic adenocarcinoma show low rates of genetic alterations in KRAS,TP53 and SMAD4

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Material and methods

2.1 Case selection

Table 1 Primers used for mutational analysis

2.3 MGMT promoter methylation analysis

2.4 Microsatellite instability

Table 2 Antibody used for immunohistochemistry analysis

2.6 Statistical analysis

3. Results

3.1 “Long-survivor” group characterization

Table 3 Clinical and molecular data of the samples analyzed compared with data reported in the literature

3.3 Overall survival according to the status of molecular markers

Table 4 Survival analysis according to the status of molecular markers

Footnotes

Acknowledgments

Conflict of interest

Supplementary data

References

Table 1
Primers used for mutational analysis

Table 2
Antibody used for immunohistochemistry analysis

Table 3
Clinical and molecular data of the samples analyzed compared with data reported in the literature

Table 4
Survival analysis according to the status of molecular markers