Digestive and Liver Disease 46 (2014) 97 104

Contents lists available at ScienceDirect

Digestive and Liver Disease

j ourna l ho mepage: www.elsev ier .c

Review Article

Small bowel adenocarcinoma: Epidemiology, risktreatment

Thomas Aparicioa,, Aziz Zaananb,c, Magali Svrcekd, Pierre LauSylvain M f g h

a Gastroenterol 3, Sorbb Gastroenterolc UMR-S775, INd Anatomopathe Digestive Surgf Hepato-Gastrg Hepato-Gastrh Oncology Uni

a r t i c l

Article history:Received 7 January 2013Accepted 29 April 2013Available online 21 June 2013

Key words:CarcinogenesiChemotherapyLynch syndromPrognostic facRare tumourSmall intestine

Small bowel adenocarcinomas are rare tumours, but their incidence is increasing. Their most commonprimary location is the duodenum. The few studies that have collected data regarding small bowel ade-nocarcinoma are not homogeneous and are widely spread over time. Even though these tumours aremost often sporadic, some predisposing diseases have been identied, among which Crohns disease andgenetic syndromes. Early diagnosis of small bowel adenocarcinoma remains difcult despite signicant

1. Epidem

1.1. Inciden

Despite the length face area, s5% of gastroNational Cacers in the 22.7 cases/m

Correspontive, Hpital AvTel.: +33 1 48

E-mail add

1590-8658 http://dx.doi.os

etor

adenocarcinoma

radiological and endoscopic progress. After surgical resection the main prognostic factor is node inva-sion; in this case, adjuvant chemotherapy can be expected to be benecial, although this has not beenestablished by randomised trials. For metastatic disease, platinum-based chemotherapy seems to be themost effective treatment. Targeted therapies have not yet been evaluated in this type of cancer.

2013 The Authors. Published by Elsevier Ltd on behalf of Editrice Gastroenterologica Italiana S.r.l.

iology

ce

the fact that the small intestine makes up 75% ofof the digestive tract and 90% of its mucosal sur-mall bowel cancer is rare, accounting for less thanintestinal cancers [1]. According to the United Statesncer Database, the incidence of all small bowel can-USA rose from 11.8 cases/million persons in 1973 toillion persons in 2004 [2]. Similarly, in France, their

ding author at: Service de Gastroentrologie et Cancrologie Diges-icenne, HUPSSD, APHP, 125 rue de Stalingrad, 93000 Bobigny, France.

95 54 31; fax: +33 1 48 95 54 39.ress: thomas.aparicio@avc.aphp.fr (T. Aparicio).

incidence also rose over the 19762001 period [3]. Four histologicaltypes of cancer predominate in the small bowel: adenocarcinomas,neuroendocrine tumours, gastrointestinal stromal tumours andlymphomas.

Small bowel adenocarcinoma (SBA) accounts for around 40%of all cancers of the small bowel [24]; similarly, neuroendocrinetumours have roughly the same incidence. In the USA, the inci-dence of SBA has been estimated to be of 5300 new cases, witharound 1100 deaths per year [5]. The median age at diagnosisis in the sixth decade of life. According to the EUROCARE data,the estimated number of annual new cases of SBA in Europe is3600 [6]. The estimated incidence rate is 5.7 cases per millionpersons. In France the estimated incidence of SBA for the period19892001 was 0.31/100,000 for men and 0.23/100,000 for women.According to the data from the Burgundy cancer registry, thenumber of new cases in France can be estimated to be 200 peryear [3].

The duodenum is the most frequently involved segment, with5582% of cases, followed by the jejunum (1125%) and ileum(717%) [2,3,710]. The increasing incidence of SBA is mainly dueto the increase in duodenal tumours [11].

2013 The Authors. Published by Elsevier Ltd on behalf of Editrice Gastroenterologica Italiana S.r.l. rg/10.1016/j.dld.2013.04.013

Open access under CC BY-NC-ND license.

Open access under CC BY-NC-ND license.anfredi , Christophe Locher , Pauline Afchainogy and Digestive Oncology Unit, Avicenne Hospital, HUPSSD, APHP, Universit Paris 1ogy and Digestive Oncology Unit, Georges Pompidou Hospital, APHP, Paris, FranceSERM, Paris, Franceology Unit, Saint Antoine Hospital, APHP, Paris Franceery Unit, Purpan Hospital, Toulouse, France

oenterology Unit, Pontchaillou Hospital, Rennes, Franceoenterology Unit, Meaux Hospital, Meaux, Francet, Saint Antoine Hospital, APHP, Paris, France

e i n f o a b s t r a c tom/ lo cate /d ld

factors, diagnosis and

rent-Puigc, Nicolas Carreree,

onne Paris Cit, Bobigny, France

98 T. Aparicio et al. / Digestive and Liver Disease 46 (2014) 97 104

2. Etiopathogenetic factors

2.1. Environmental factors

In contrSBA are contion [12] anrisk of SBA.among the meat or smhigher intak

The marrectal adenoIn the smalxenobioticsowing to thintestine cobacteria. Thbut is still produces xdeconjugatis a potentiof the smalenzymes, inthem againis no clear duodenum

2.2. Carcino

The bioloof patients.have been s

The adencated mutatin a loss of ras a result. Ting events igene mutat1/15 (8%) [carcinogeneertheless, sof -CATENthe -CATENmutation. Ain variable 12/61 (20%)deletions abilisation omicrosatell

Other abReduced m(38%) casesdetected in(42%) [21],SBA cases. cases [23]. Mlial growthreceptor (Erespectivelya treatmenmutation hof cases [2KRAS mutatassessed byOverman et

[24]. Similarly, in the AGEO study, HER2 expression was observedin 2/51 (3.9%) cases, in both cases in the ileum [21].

Inactivation of the DNA mismatch repair (MMR) gene was foundin around 15% of colorectal cancers [27]. A decient MMR (dMMR)

caus(usuae of ter ints [28anginund SBA (18f exp

ts un(23%servncy oanceMR

ctal csion

MSH% offolloand LH1 a dM

due towftenmougeneumo37 tuin 16tumo(MAour

on w% of

ctal c

en toon cacarcied inay isncy ories, cer. synd

largeent

ed in

netic

Fami is atienta you

loca7 (4.5y locr in 1ast to colorectal cancer, studies on the pathogenesis ofstrained by the rarity of the disease. Alcohol consump-d smoking [13] have been associated with an increased

Other studies have reported an increased risk of SBAhighest consumers of sugar, rened carbohydrates, redoked food, while a reduced risk was observed withes of coffee, sh, fruit, and vegetables [14,15].

ked difference between the incidences of SBA and colo-carcinoma suggests different exposures to carcinogens.l bowel, the contact time between intestinal cells and

or dietary carcinogens is shorter than in the colon,e shorter transit time. In addition, the proximal smallntains low concentrations of aerophilic Gram-positivee density of the microbiota increases in the distal ileum,much lower than in the colon, where the microbiotaenobiotic transformation during which bile salts areed and dehydroxylated to form desoxycholic acid, whichal tumour promoter [16]. Moreover, the epithelial cellsl bowel are equipped with a wide range of microsomalcluding the benzopyrene hydroxylase, that may protectst food-derived carcinogens [17]. Nevertheless, thereexplanation for the differing cancer incidences in theand jejuno-ileum.

genesis in SBA

gy of SBA has been investigated only in a small number The main genes involved in colorectal carcinogenesistudied also in SBA (Table 1).omatous polyposis coli (APC) gene has acquired a trun-ion in up to 80% of sporadic colorectal cancers, resultingegulation of -CATENIN that accumulates in the nucleushis mutation is considered to be one of the main trigger-n colorectal carcinogenesis. The prevalence of the APCion in SBA was reported to be rather low: 0/21 [18],19] and 3/17 (18%) [20]. These data suggest that thesis of SBA differs from colorectal carcinogenesis. Nev-

everal studies have reported the nuclear accumulationIN, most probably due to a gain-of-function mutation inIN gene rather than a loss of regulation due to the APCbnormal nuclear expression of -CATENIN was foundproportions of the cases analysed: 10/21 (48%) [18],

[21] and 10/20 (50%) [22]. In another study [28], largend insertions in the -CATENIN gene, resulting in sta-f the aberrant -CATENIN, were found exclusively inite stable carcinomas.normal protein expressions have been reported in SBA.embrane expression of E-CADHERIN was found in 8/21

of SBA [18]. Overexpression of the p53 protein was the nuclei of 5/21 (24%) [18], 4/15 (27%) [19], 26/62

and 14/27 (52%) [23] tumours in different series ofA loss of SMAD4 expression was found in 5/27 (18%)oreover, abnormal expression of the vascular endothe-

factor-A (VEGF-A) and the epidermal growth factorGFR) was found in 50/54 (92%) and 36/54 (66%) cases,, suggesting that this type of cancer could benet from

t targeting the EGFR and VEGF pathways [24]. A KRASas been described in 12/21 (57%) [25] and 21/49 (43%)1]. A specic study of 78 duodenal tumours found aion in 34% of the cases [26]. HER2 expression has been

immunohistochemistry in 2 studies: in the study of al., only one out of 54 (1.7%) tumours expressed HER2

can begenes the caspromopatienable, rwas foof 89 16/89 a loss opatien10/43 was obfrequerectal cthat dMcoloreexpresLoss ofand 26erally MSH2 and Mstudy, mainlya trendmore oileal tuing to of 37 tin 22/found (MSI) stable CIN tummutatiand 22colore[30].

Takcommrectal observpathwfrequeSBA setal canLynch Only aassessminvolv

2.3. Ge

2.3.1. FAP

FAP pacer at cancerFAP, 5primarof Vateed either by a germline mutation of one of the 4 MMRlly MSH2 or MLH1, and more rarely MSH6 or PMS2) inLynch syndrome, or by hypermethylation of the MLH1

sporadic tumours, especially those occurring in elderly]. In SBA, the frequency of the dMMR phenotype is vari-g from 5% to 35% of cases. Sporadic MMR deciencyin only 1/21 (5%) cases of SBA [18]. In another seriespatients, the frequency of the dMMR phenotype was%). In this latter study, immunohistochemistry revealedression of MLH1 in 7/16 dMMR tumours. Amongst the

der 60 years of age, the dMMR phenotype was found in) and loss of expression of the MLH1 and MSH2 proteinsed with the same frequency [29]. Also, in this study thef dMMR appeared to be greater in SBA than in colo-r, as well as being higher in young patients, suggesting

is more often due to Lynch syndrome in SBA than inarcinoma. In another series of 54 SBA patients, a loss ofof one of the MMR proteins occurred in 35% of patients.2, MSH6, PMS2 and MLH1 was observed in 6%, 11%, 24%

patients, respectively. Loss of the MMR protein gen-wed 2 patterns: the loss of the combination of eitherMSH6 (when the MSH2 gene was affected) or PMS2(when the MLH1 gene was affected) [24]. In the AGEOMR phenotype was observed in 14/61 patients (23%),

to a loss of MLH1 expression [21]. In this latter study,ards a more frequent dMMR phenotype was observed

in duodenal (9/32) or jejunal (5/18) tumours than inrs (0/13) (p = 0.07). The characterisation of SBA accord-tic and epigenetic alterations was performed in a seriesurs [30]. A chromosomal instability (CIN) was detectedmours (59%). A high level of DNA methylation was% of the CIN tumours, in 44% of microsatellite instableurs and in 44% of microsatellite and chromosomal-

CS) tumours. A KRAS mutation was observed in 55% ofs, 0% of MSI tumours and 10% of MACS tumours. A BRAFas detected in 6% of CIN tumours, 22% of MSI tumours

MACS tumours. These ndings suggest that SBA andancer could belong to different molecular subgroups

gether, these ndings suggest the existence of somercinogenesis pathways shared between SBA and colo-nogenesis. Nevertheless, the APC mutation is less often

SBA than in colorectal cancer, even though the Wnt involved through -CATENIN alteration. Moreover, thef the dMMR phenotype appears to differ in the differentbut it is generally slightly more frequent than in colorec-Nevertheless, a bias towards an over-representation ofrome patients could be suspected in tertiary care series.

study based on an unselected cohort will allow for the of the frequencies of the various biological alterations

SBA carcinogenesis.

predisposition

lial adenomatous polyposis (FAP) consequence of a germinal mutation of the APC gene.s are exposed to a very high incidence of colorectal can-ng age, and SBA is the second most common primarytion. In a pooled registry study of 1255 patients with%) had an upper digestive tract adenocarcinoma. Theation was the duodenum in 29 cases (50%), the ampulla0 (18%), the stomach in 7 (12%), the jejunum in 5 (8.5%),

T. Aparicio et al. / Digestive and Liver Disease 46 (2014) 97 104 99

Table 1Molecular changes in small bowel adenocarcinoma.

Reference Number ofpatients

Abnormal P53 Abnormal-CATENIN

HER2 over-expression

APC mutation KRAS mutation dMMR phenotype

Wheeler et aArai et al. [1Blaker et al. Aparicio et aSvrcek et al.Overman et Blaker et al. Planck et al.

and the ileurisks for ducompared t132681; ptively [32]. FAP patientmain cause a coloproct

2.3.2. LynchLynch sy

match repaneoplasia, squently ovaVarious levpatients wistudy, the rhas been esyndrome [tion and 10Nevertheles1% accordinSo far, it hapatients fosystematicapresence of

2.3.3. PeutzThe Peu

der due to to hamartoof 520 (95%syndrome pfrom the intlesion.

2.4. Other p

2.4.1. CrohnCrohns

of the digesinvolved. Thact with cecarcinogenedisease hasrange fromThe SBA arto sporadic patients (focumulativedisease andon the exteyears old, id

ls, an2.07f Cro6%); SEER

pattien

trea6].

Coeliliac ds imat c

k of ts witish s7 lym

disephomish r

dise].

prectiveber

in syndme posinnt in

gnos

inica

clinimptof 21hen merobstmoul. [18] 21 24% 48% 9] 15 27% [20] 17 l. [21] 63 42% 20% 3.9%

[23] 27 52% 7.4% al. [24] 54 1.7% [25] 21 24% [29] 89

m in 1 case (1.7%) [31]. In another study, the relativeodenal adenocarcinoma or ampulloma in a FAP patiento those in the general population were 330 (95% CI,

< 0.001), and 123 (95% CI, 33316; p < 0.001), respec-Even though the risk of duodenal adenocarcinoma in a

remained less than 5%, this cancer is nevertheless theof cancer-related death in patients who have undergoneectomy [33,34].

syndromendrome is caused by a germline mutation of a DNA mis-ir gene, which exposes the patient to various types ofuch as colorectal and endometrial cancers; he less fre-rian, urothelial, gastric, biliary tract cancers and SBA.els of increased risk for SBA have been reported forth Lynch syndrome. According to data from a Dutchelative risk of SBA for a patient with Lynch syndromestimated to range from 25 in the early phases of the35] to 291 (95% CI, 71681) in case of an MLH1 muta-3 (95% CI, 14729) in case of an MSH2 mutation [36].s, the lifetime cumulative risk remains low: 0.6% andg to Finnish and French registries, respectively [37,38].s not been recommended to screen Lynch syndromer SBA. However, analysis of the MMR phenotype islly recommended in SBA, because it could reveal the

Lynch syndrome [39,40].

Jeghers syndrometzJeghers syndrome is an autosomal dominant disor-the STK11 suppressor gene mutation that predisposesmatous gastrointestinal tract polyposis. A relative risk

CI, 2201306) for SBA was observed in PeutzJeghersatients [41]. The adenocarcinoma probably originatesra-epithelial neoplasia observed in the hamartomatous

redisposing conditions

s diseasedisease induces chronic inammation in every segmenttive tract, and the distal ileum is the most frequentlye chronic inammation releases cytokines that inter-

ll surface receptors and target genes that can promotesis [16]. The increased relative risk of SBA in Crohns

contro(OR = 1lence olow (1.in the diseasethat palongedSBA [4

2.4.2. Coe

inducecells ththe rispatienIn a Bricer (10coeliacof lyma Swedcoeliac10 [49

TheprospeSeptemdiseaseLynch syndropredisfreque

3. Dia

3.1. Cl

TheThe systudy opain wof an eBowel ileal tu been estimated in several population-based studies to 17 to 41 compared to the general population [42,43].ises in an inamed small bowel segment. In contrastSBA, in Crohns disease, this cancer appears in youngerurth decade of life), and mainly in the ileal segment. The

risk is estimated to be 0.2% after 10 years of Crohns 2.2% after 25 years [44]. Another estimation, basednsive SEER database and restricted to patients over 65entied 923 cases of small bowel cancer and 142,273

p = 0.06) [8(after 1988(28%), surg(18%), ultranosis was mpatients halymph-nodin another synchronou0% 5%8% 53% 18% 12% 43% 14% 7% 35%10% 57% 18%

d conrmed the increased risk of SBA in Crohns disease; 95% CI, 6.0720.80; p < 0.001). In this study, the preva-hns disease in patients with small bowel cancer wasnevertheless many cases of SBA could have been missed

database as the median age of onset of SBA in Crohnsients is less than 65 years [45]. Another study suggeststs who have undergone a small bowel resection or pro-tment with salicylate have a lower risk of developing

ac diseaseisease is characterised by a lymphocytic inltrate thatmunological disruption and damage to the epithelialan include premalignant changes, and could increaseboth SBA and small bowel lymphoma. A cohort of 235th coeliac disease showed an 8% prevalence of SBA [47].urvey study that included 395 cases of small bowel can-phomas, 175 SBAs and 79 neuroendocrine tumours),

ase was found in 13% of cases of SBA and 39% of casesa; primary location of SBA was usually jejunal [48]. Inegistry study, the relative risk of SBA in patients withase versus the general population was estimated to be

liminary results of the French NADEGE cohort thatly included 127 patients with SBA from March 2009 to2010, revealed a genetic syndrome or a predisposing20% of the patients: Crohns disease (8.6%), FAP (3%),rome (3%), coeliac disease (1.5%) and PeutzJeghers(0.8%) (7). These preliminary results indicate that ag disease or genetic syndrome is considerably more

SBA than in colorectal cancer.

is

l presentation

cal presentation and diagnosis of SBA is usually delayed.ms are initially rather non-specic. In a single-centre7 patients with SBA, 66% of the patients had abdominaldiagnosed. SBA was usually diagnosed in the contextgency involving an occlusion (40%) or bleeding (24%).ruction was mainly observed in cases of jejunal andr, and less frequently in duodenal tumours (47% vs 34%;

]. In this study, diagnoses in the more recent period) were obtained by upper gastrointestinal endoscopyery (26%), small bowel barium transit (22%), CT scansound (3%) or physical examination (3%). The diag-ainly obtained at advanced stages, when 35% of the

d synchronous metastases and 39% had tumours withe invasion [8]. A similar stage distribution was foundstudy of 129 tumours in which 38% of the patients hads metastases and 38% had lymph-node invasion [50]. In

100 T. Aparicio et al. / Digestive and Liver Disease 46 (2014) 97 104

Crohns disease, the diagnosis is mainly obtained postoperativelyafter resection of an obstructed small bowel segment [44].

3.2. Diagnosis

For SBA,50% [51], anIt should beupper and systematicaclysis, MR eallow for anmake early8595% for 9096% [51suspected oupper and lunderwentaccuracy in(19 of 22), 9MR enteroc3 yielded fa

Capsule explorationperformed explore obsbowel tumo[56,57]. Doof small boconvenientbiopsy or prever, enterovideocapsu

3.3. Other r

The Frenrecommendassess distanal endoscogenetic dis(CEA) and cespecially imarkers are

In a conease, a fullenteroscannstenosis) to

An assaybiopsy are r

Two diffrst identilite loci, anmismatch rniques.

4. Prognos

SBA carrsurvival (OSOS is correlstage I (incfrom 14% toand 35% foprognosis agastric canc

Table 2TNM classication of small intestine adenocarcinomas.

Primary tumour (T)

Tx Primary tumour cannot be assessedNo evidence of primary tumourCarcinoma in situTumour invades the lamina propria,muscularis mucosa or submucosa.Tumour invades the lamina propria ormuscularis mucosaTumour invades the submucosaTumour invades the muscularis propriaTumour invades 2 cm or less into thesubserosa or into thenon-peritonealized perimusculartissue (mesentery orretroperitoneuma)Tumour perforates the visceralperitoneum or directly invades other

al lym

t metaNo distant metastasisDistant metastasis

bin et al. [82].: The non-peritonealized perimuscular tissue for the jejunum and ileum ishe mesentery. For the duodenum, it is part of the retroperitoneum in areasere is no serosa.

otentially curative treatment [50]. The frequency of locally-ed unresectable cancer is not reported in most of the studies,er, in the preliminary report of the NADEGE cohort, a locally-ed cancer occurred in 5% of the cases [7].ph-node invasion is the main prognostic factor for local,50]; moreover, the number of lymph nodes assessed andmber of positive lymph nodes are of prognostic value. InII patients, a positive number of invaded lymph nodes 3worse 5-year disease-free survival (DFS) rate than patients2 invaded lymph nodes (37% vs 57%) [62]. For jejuno-ilealrs, when 10 or more lymph nodes were examined, the OScreased non-signicantly in stage I (73.2% vs 55.6%, NS) andantly in stage II (61.8% vs 32.9%, p < 0.001). Multivariate anal-entied advanced age, advanced stage, ileal location, thery of

T. Aparicio et al. / Digestive and Liver Disease 46 (2014) 97 104 101

signicant predictors of poor OS [63]. Thus, a curative resectionat an early stage (stages I and II) should systematically include aregional lymphadenectomy.

Several studies suggest that a duodenal primary tumour has aworse progOther factorage, pT4 turesection mof 10% [9,6cers after a to Lynch sya prolonged

In metasapy, a retrostatus and afactors for p

5. Treatme

5.1. Localis

Completregional lymposterior inand resectio

In the cresection ias bowel omultidiscipresectable, surgical proval betweenevaluate th

For duoformed [8] or for an inAdditionallhepatic lymof the rightduodenal redenal tumoadjacent orated with pas R1 or R2[69].

For jejunresection aperformed.ileocoecal otion of the i

To date,to the lack oare those frthe variousadminister rospective tend to be tcurrently-u

In a revireceived raduodenal pstudy of 48intent, chemavailable doin R1 or R2

reatmg to t

eral rthera0,71ents uate

sin wit

ts re adjueme

the higed/we OSencen ana

an in 2005

reped Sjuvantion thera.tncdhemoativers wtherarospeted bation

begi

etast

y fewed foall, l, these raective studies suggest that chemotherapy prolongs OS in

ts with advanced SBA [8,74,75], but there is no agreedne regimen owing to the lack of randomised trials. A ret-tive comparison of OS, according to whether palliativetherapy was prescribed, showed a signicant increase ofl in treated patient (12 months vs 2 months, p = 0.02) [8].ther series from the registry of British Columbia, patientsceived chemotherapy (n = 16) had an OS of 15.6 months,

patients who did not (n = 21) had an OS of 7.7 months [74].w studies compared a specic regimen of chemotherapy torotocols. A retrospective study of 44 patients suggests thatnosis than a jejunal or ileal primary tumour [8,9,62].s for poor prognosis have also been reported: advancedmour stage, poorly differentiated tumour, positiveargins, lymphovascular invasion and a lymph node ratio4,65]. One study has reported 12/74 (16%) second can-

curative resection, 5 of which could have correspondedndrome. This high frequency of second cancers justies

follow-up after SBA treatment [66].tatic or locally-advanced SBA treated with chemother-spective study found that impaired WHO performancen above-normal value of CEA or CA 19.9 were prognosticoor survival [60].

nt

ed cancer

e resection (R0) of the primary tumour with loco-ph node resection is mandatory. In the context of

vasion, pre-operative treatment should be considered,n reconsidered after 23 months of chemotherapy.ontext of unresectable metastases, primary tumours not recommended except in an emergency suchbstruction, perforation or uncontrolled bleeding. If alinary evaluation concludes that the metastases areresection of the metastases can consist of either 1 or 2cedures, possibly with chemotherapy during the inter-

the procedures. However, more data are required toe value of metastasis resection in SBA.denal tumours, a Whipple resection should be per-for a tumour in the second segment of the duodenumltrating tumour in the proximal or distal duodenum.y, resection of the peri-duodenal, peri-pancreatic, andph nodes should also be performed, as well as resection

side of the coeliac and superior mesenteric arteries. Asection alone could be performed for a proximal duo-ur or a distal duodenal tumour with no inltration ofgans [67], despite the fact that this procedure is associ-oor prognosis [68]. An R0 resection is to be preferred,

resections are strongly associated with poor prognosis

al and ileal tumours, an R0 resection with lymph nodend jejuno-jejunal or ileo-ileal anastomosis should be

If the last ileal loop or Bauhins valve are involved, anr right hemicolectomy should be performed with liga-leocolic artery so as to allow for lymph node resection.

no standard adjuvant regimen has been dened duef randomised controlled trials. The only data availableom retrospectives studies. The inability to control for

prognostic factors inuencing the original decision toadjuvant therapy has been a major limitation of ret-studies, as the patients who receive adjuvant therapyhose at greater risk of disease recurrence on basis of thesed prognostic factors.ew of the US National Cancer Database, 11% of patientsdiotherapy with or without chemotherapy mainly forrimary tumours [9]. Nevertheless, in a retrospective

duodenal adenocarcinomas resected with a curativeoradiotherapy did not improve survival [68]. The data

not establish a clear recommendation for radiotherapyresection or locally-advanced duodenal cancer.

Fig. 1. Taccordin

Sevchemo[8,64,7of patiinadeq

In atreatedpatienuse ofimprovnot inwith a[invadimprovof evidSBA, ashown24% in

Theadvancthe admendachemo(wwwvant ccumultumouchemo

A ppromoobservshould

5.2. M

Verapy usare smOveralresponretrosppatienfrontlirospecchemosurvivaIn anowho rewhile Only feother pent for localised small bowel adenocarcinoma after an R0 resectionhe 2013 French guidelines (www.tncd.org).

etrospective studies have found no benet in adjuvantpy after potentially curative surgical resections of SBA

]. These negative results may be due to the small numbertreated, a bias in selecting patients for treatment or an

chemotherapy regimen.gle-centre, retrospective study including 54 patientsh an R0 resection between 1990 and 2008, 30 (56%)ceived adjuvant therapy. In multivariate analysis, thevant therapy was found to be associated with annt in the DFS (HR 0.27; 95% CI 0.070.98, p = 0.05), butOS (HR 0.47; 95% CI 0.131.62, p = 0.23). In patientsh risk of recurrence (dened as a lymph node ratioithout metastasis] 10%), adjuvant therapy appeared to

(p = 0.04), but not DFS (p = 0.15) [65]. Despite the lack supporting the delivery of adjuvant chemotherapy forlysis in the USA of the National Cancer Database hascrease in the use of chemotherapy from 8% in 1985 to

[2].orted efcacy of uoropyrimidine and oxaliplatin inBA [60,72], and the efcacy of the same regimen int chemotherapy of CRC have led to the French recom-of an adjuvant uoropyrimidine and oxaliplatin-basedpy after a curative R0 resection of a stage-III SBA.org, last updated in 2013). For stage-II tumours, adju-therapy is optional for pT4. Nevertheless, given the

evidence of poor prognosis for poorly-differentiatedith a resected lymph node number

102 T. Aparicio et al. / Digestive and Liver Disease 46 (2014) 97 104

Table 4Studies evaluating chemotherapy in advanced small bowel adenocarcinoma.

Reference Regimen Number of patients Response rate (%) Progressionfree survival

Overall survival(months)

Crawley et al. [73] 5FU 8 37 Locher et al. [78] 5FU + cisplatin 20 21 Gibson et al. [76] 5FU + doxorubicin + MMC 38 18 Zaanan et al. [60] FOLFOX 48 34

LV5FU2 10 0 LV5FU2 + cisplatin 19 30 FOLFIRI 16 9

Overman et al. [79] 5FU + cisplatin 29 41 5FU without cisplatin 41 17

Overman et al. [72] Capecitabine + oxaliplatin 30 52 Zaanan et al. [81] FOLFIRI (second line) 28 20

gemcitabine and irinotecan-based chemotherapy gives betterresults than 5FU monotherapy [75]. A retrospective, single-centrestudy of 80 patients with metastatic SBA suggested that platinum-based chemotherapy gave a higher ORR than other chemotherapyregimens (46% vs 16%; p = 0.01) and longer median progression-freesurvival (PFS) (8.7 vs 3.9 months; p 0.01), although the medianOS was not signicantly different (14.8 vs 12.0 months; p = 0.10)[79]. However, this study did not address the respective risk-benetratios of the different platinum salts (cisplatin, oxaliplatin, etc.),even though the clinical efcacy and antitumour mechanisms ofthese drugs are very different [80]. A prospective phase-II trialof capecitapatients winomas, wit11.3 monthrospective (n = 48), FOsecutive paLV5FU2, FOand 4.8 mo13.5, 17.8, 1ysis, the WCEA (p = 0.0variables sipatients treanalysis shoPFS (p < 0.00

Fig. 2. TreatmFrench guideli

series, the was investiinum salts 20% and thand OS werthe FOLFIRIment in pasalt-based

So far, nbased chemrecommend

.tncdrall, , but

med

nces

gut Ael. Ca

moriates: chals ofage Clignanericanal A, Sicianmar

oma. Ovre J, patien520ricio lE (NAaja Bel: p4;101

we JR,sion oall bo9;86:on YWean inbine plus oxaliplatin has given interesting results inth advanced small-bowel and ampullary adenocarci-h an ORR of 50%, a median time to progression ofs, and a median OS of 20.4 months [72]. Another ret-multicentre study evaluated LV5FU2 (n = 10), FOLFOXLFIRI (n = 19) and LV5FU2-cisplatin (n = 16) in 93 con-tients. The median PFS times in patients treated withLFOX, FOLFIRI and LV5FU2-cisplatin were 7.7, 6.9, 6.0nths, respectively, while the median OS times were0.6 and 9.3 months, respectively. In a multivariate anal-HO performance status (p < 0.0001) and the elevated2) and CA 19.9 (p = 0.03) serum levels were the onlygnicantly associated with poor OS. In the subgroup ofated with platinum-based chemotherapy, multivariatewed that LV5FU2-cisplatin was associated with poorer01) and OS (p = 0.02) than FOLFOX [60]. From the same

(wwwOve

cancerwith a

Refere

[1] Neubow

[2] BiliStaAnn

[3] LepmaAm

[4] JemClin

[5] Kumcin

[6] Faiof 199

[7] ApaGr

[8] Dabbow200

[9] Homissm199

[10] MoKorent for metastatic small bowel adenocarcinoma according to the 2013nes (www.tncd.org).

and Clini[11] Chow JS,

of malignEpidemio

[12] Wu AH, Yintestina

[13] Kaerlev Lintake ormulti-cen

[14] Chow WHCancer Ca

[15] Negri E, Bsmall int

[16] Schottenpathogen2009;19:(months)

7.8 13.08.0 14.05.0 8.06.9 17.87.7 13.56.0 9.64.8 10.68.7 14.83.9 12.0

11.3 20.03.2 10.5

effectiveness of FOLFIRI in second-line chemotherapygated in 28 patients who had been treated with plat-in rst line therapy: the objective response rate wase disease control rate was 52%, while the median PFSe 3.2 and 10.5 months, respectively. This suggests that

regimen exhibits modest activity as a second-line treat-tients with advanced SBA after failure of a platinumrst-line chemotherapy [81].

o data is available for targeted therapy. The oxaliplatin-otherapy seems to be the best choice, and it ised as a rst line treatment by the French guidelines.org, last updated in 2013) (Fig. 2).advanced SBA had a worse prognosis than colorectal

a better prognosis than gastric or pancreatic cancer,ian OS exceeding 12 months.

I, Jacobson JS, Suh S, et al. The epidemiology of cancer of the smallncer Epidemiology, Biomarkers & Prevention 1998;7:24351.

KY, Bentrem DJ, Wayne JD, et al. Small bowel cancer in the Unitedanges in epidemiology, treatment, and survival over the last 20 years.

Surgery 2009;249:6371., Bouvier AM, Manfredi S, et al. Incidence and management of primaryt small bowel cancers: a well-dened French population study. The

Journal of Gastroenterology 2006;101:282632.iegel R, Ward E, et al. Cancer statistics, 2008. CA: A Cancer Journal for

s 2008;58:7196. S, Ciesielski TE, Fogarasi MC. Management of small bowel adenocar-ncology (Williston Park) 2002;16:13649.Trama A, De Angelis R, et al. Incidence, prevalence and survivalts with rare epithelial digestive cancers diagnosed in Europe in02. European Journal of Cancer 2012;48:141724.T, Manfredi S, Coriat R, et al. CA.D Nationale ADEnocarcinome intestinDEGE): faisabilit et tat davancement; 2011 www.jfhod.comS, Suki D, Pro B, Bonnen M, Ajani J. Adenocarcinoma of the smallresentation, prognostic factors, and outcome of 217 patients. Cancer:51826.

Karnell LH, Menck HR, et al. The American College of Surgeons Com-n Cancer and the American Cancer Society. Adenocarcinoma of the

wel: review of the National Cancer Data Base, 19851995. Cancer2693706., Rha SY, Shin SJ, et al. Adenocarcinoma of the small bowel at a singlestitute: management and prognosticators. Journal of Cancer Research

cal Oncology 2010;136:38794.

Chen CC, Ahsan H, et al. A population-based study of the incidenceant small bowel tumours: SEER, 19731990. International Journal oflogy 1996;25:7228.u MC, Smoking Mack TM. alcohol use, dietary factors and risk of smalll adenocarcinoma. International Journal of Cancer 1997;70:5127., Teglbjaerg PS, Sabroe S, et al. Is there an association between alcohol

smoking and small bowel adenocarcinoma? Results from a Europeanter case-control study. Cancer Causes Control 2000;11:7917., Linet MS, McLaughlin JK, et al. Risk factors for small intestine cancer.uses Control 1993;4:1639.osetti C, La Vecchia C, et al. Risk factors for adenocarcinoma of the

estine. International Journal of Cancer 1999;82:1714.feld D, Beebe-Dimmer JL, Vigneau FD. The epidemiology andesis of neoplasia in the small intestine. Annals of Epidemiology5869.

T. Aparicio et al. / Digestive and Liver Disease 46 (2014) 97 104 103

[17] Delaunoit T, Neczyporenko F, Limburg PJ, et al. Pathogenesis and risk factors ofsmall bowel adenocarcinoma: a colorectal cancer sibling. The American Journalof Gastroenterology 2005;100:70310.

[18] Wheeler JM, Warren BF, Mortensen NJ, et al. An insight into the genetic pathwayof adenocarcinoma of the small intestine. Gut 2002;50:21823.

[19] Arai M, Sadenocar1997;70:

[20] Blaker H,small inteSMAD4 g

[21] Aparicio Taccording

[22] Breuhahnnot pointing divergThe Journ

[23] Svrcek McarcinomPathology

[24] OvermancharacterCancer 20

[25] Blaker H,MAPK andJournal o

[26] Fu T, Guzentiationresected

[27] Zaanan Acancer: frOncology

[28] Aparicio Tnotype isand Liver

[29] Planck Msion of 2003;97:

[30] Warth A, adenocar2011;24:

[31] Jagelmanadenoma

[32] Offerhausnal cance19802.

[33] Bulow S,nomatouDisease 1

[34] Vasen HFdenal ade

[35] Watson Pcancer. C

[36] Vasen HFnonpolypology 199

[37] Aarnio MhereditarJournal o

[38] BonadonamutationAmerican

[39] Babba T, HNPCC sy

[40] Schulmanbowel c2005;128

[41] Giardiellocancer in144753.

[42] Bernsteinmatory b

[43] Jess T, Lotory boweGastroen

[44] Palascak-patients wnovo. In

[45] Shaukat AnocarcinoBiomarke

[46] Piton G, adenocarof Gastro

[47] Swinson 1983;1:1

[48] Howdle PD, Jalal PK, Holmes GK, et al. Primary small-bowel malignancy inthe UK and its association with coeliac disease. Quarterly Journal of Medicine2003;96:34553.

[49] Askling J, Linet M, Gridley G, et al. Cancer incidence in a population-based cohortof individuals hospitalized with celiac disease or dermatitis herpetiformis. Gas-

enteroamontlysis urgerdiaf M

of miologykley Jnocarogra

rton Kensio3;41:eul F,trast-gy 200sselli luatiotman

wirelh obscnaziotrointes. Gaithi

eo capcure g6;101s A, Mel ma

es 200nan Ael ad0;21:rmancinomurgicarmanmall

end rholl Mged anfdanace wiurgerrmannocarrieres

casesgery 1lamannagem0;179n TA,tors in8;2:7aeen uodeawal rimar7;14:

TJ, Yearcino6;30:rman

oxaliater. wley mall brouraykowerienhe Roman ctiveniew ofson Min, and5;10:asu Da of thimizu S, Imai Y, et al. Mutations of the Ki-ras, p53 and APC genes incinomas of the human small intestine. International Journal of Cancer3905.

von Herbay A, Penzel R, et al. Genetics of adenocarcinomas of thestine: frequent deletions at chromosome 18q and mutations of the

ene. Oncogene 2002;21:15864., Svrcek M, laforest A, et al. Small bowel adenocarcinoma phenotype

to the primary localisation. Annals of Oncology 2012;23, ix234. K, Singh S, Schirmacher P, et al. Large-scale N-terminal deletions but

mutations stabilize beta-catenin in small bowel carcinomas, suggest-ent molecular pathways of small and large intestinal carcinogenesis.al of Pathology 2008;215:3007., Jourdan F, Sebbagh N, et al. Immunohistochemical analysis of adeno-a of the small intestine: a tissue microarray study. Journal of Clinical

2003;56:898903. MJ, Pozadzides J, Kopetz S, et al. Immunophenotype and molecularisation of adenocarcinoma of the small intestine. British Journal of10;102:14450.

Helmchen B, Bonisch A, et al. Mutational activation of the RAS-RAF- the Wnt pathway in small intestinal adenocarcinomas. Scandinavian

f Gastroenterology 2004;39:74853.zetta AA, Jeschke J, et al. KRAS G>A mutation favors poor tumor differ-

but may not be associated with prognosis in patients with curativelyduodenal adenocarcinoma. International Journal of Cancer 2012., Meunier K, Sangar F, et al. Microsatellite instability in colorectalom molecular oncogenic mechanisms to clinical implications. Cellular

(Dordrecht) 2011;34:15576., Schischmanoff O, Poupardin C, et al. Decient mismatch repair phe-

a prognostic factor for colorectal cancer in elderly patients. Digestive Disease 2012., Ericson K, Piotrowska Z, et al. Microsatellite instability and expres-MLH1 and MSH2 in carcinomas of the small intestine. Cancer15517.Kloor M, Schirmacher P, et al. Genetics and epigenetics of small bowelcinoma: the interactions of CIN, MSI, and CIMP. Modern Pathology56470.

DG, DeCosse JJ, Bussey HJ. Upper gastrointestinal cancer in familialtous polyposis. Lancet 1988;1:114951.

GJ, Giardiello FM, Krush AJ, et al. The risk of upper gastrointesti-r in familial adenomatous polyposis. Gastroenterology 1992;102:

Alm T, Fausa O, et al. Duodenal adenomatosis in familial ade-s polyposis. DAF Project Group. International Journal of Colorectal995;10:436., Bulow S, Myrhoj T, et al. Decision analysis in the management of duo-nomatosis in familial adenomatous polyposis. Gut 1997;40:7169., Lynch HT. Extracolonic cancer in hereditary nonpolyposis colorectalancer 1993;71:67785., Wijnen JT, Menko FH, et al. Cancer risk in families with hereditaryosis colorectal cancer diagnosed by mutation analysis. Gastroenter-6;110:10207., Mecklin JP, Aaltonen LA, et al. Life-time risk of different cancers iny non-polyposis colorectal cancer (HNPCC) syndrome. Internationalf Cancer 1995;64:4303.

V, Bonaiti B, Olschwang S, et al. Cancer risks associated with germlines in MLH1, MSH2, and MSH6 genes in Lynch syndrome. Journal of the

Medical Association 2011;305:230410.Schischmanoff O, Lagorce C, et al. Small bowel carcinoma revealingndrome. Gastroentrologie Clinique et Biologique 2010;34:3258.n K, Brasch FE, Kunstmann E, et al. HNPCC-associated small

ancer: clinical and molecular characteristics. Gastroenterology:5909.

FM, Brensinger JD, Tersmette AC, et al. Very high risk of familial PeutzJeghers syndrome. Gastroenterology 2000;119:

CN, Blanchard JF, Kliewer E, et al. Cancer risk in patients with inam-owel disease: a population-based study. Cancer 2001;91:85462.ftus Jr EV, Velayos FS, et al. Risk of intestinal cancer in inamma-l disease: a population-based study from olmsted county, Minnesota.

terology 2006;130:103946.Juif V, Bouvier AM, Cosnes J, et al. Small bowel adenocarcinoma inith Crohns disease compared with small bowel adenocarcinoma de

ammatory Bowel Diseases 2005;11:82832., Virnig DJ, Howard D, et al. Crohns disease and small bowel ade-ma: a population-based casecontrol study. Cancer Epidemiology,rs & Prevention 2011;20:11203.Cosnes J, Monnet E, et al. Risk factors associated with small bowelcinoma in Crohns disease: a case-control study. The American Journalenterology 2008;103:17306.CM, Slavin G, Coles EC, et al. Coeliac disease and malignancy. Lancet115.

tro[50] Tal

anaof S

[51] BoutionRad

[52] BucadeTom

[53] Hodim200

[54] Pillconolo

[55] Maeva

[56] Haringwit

[57] Pengascas

[58] Hadvidobs200

[59] Rosbowenc

[60] Zaabow201

[61] Ovecarof S

[62] Ovein sand

[63] Nicsta

[64] Halrienof S

[65] Oveade

[66] Vey100Sur

[67] Kakma200

[68] Sohfac199

[69] Bakin d

[70] Agrin p200

[71] Wunoc200

[72] Oveandof V

[73] Craa suo

[74] Czaexpof t

[75] Fishefferev

[76] Gibbic200

[77] Jigynomlogy 2002;123:142835.i MS, Goetz LH, Rao S, et al. Primary cancers of the small bowel:of prognostic factors and results of surgical management. Archivesy 2002;137:56470., Jaff A, Soyer P, et al. Small-bowel diseases: prospective evalua-

ulti-detector row helical CT enteroclysis in 107 consecutive patients. 2004;233:33844.

A, Siegelman SS, Jones B, et al. The accuracy of CT staging of small bowelcinoma: CT/pathologic correlation. Journal of Computer Assistedphy 1997;21:98691.M, Fishman EK. The current status of multidetector row CT and three-nal imaging of the small bowel. Radiologic Clinics of North America199212.

Penigaud M, Milot LS, et al. Possible small-bowel neoplasms:enhanced and water-enhanced multidetector CT enteroclysis. Radi-6;241:796801.

G, Polettini E, Casciani E, et al. Small-bowel neoplasms: prospectiven of MR enteroclysis. Radiology 2009;251:74350.n D, Schmidt H, Bolz G, et al. A prospective two-center study compar-ess capsule endoscopy with intraoperative enteroscopy in patientsure GI bleeding. Gastrointestinal Endoscopy 2005;61:82632.

M, Santucci R, Rondonotti E, et al. Outcome of patients with obscureestinal bleeding after capsule endoscopy: report of 100 consecutivestroenterology 2004;126:64353.M, Heine GD, Jacobs MA, et al. A prospective study comparingsule endoscopy with double-balloon enteroscopy in patients withastrointestinal bleeding. The American Journal of Gastroenterology:527.ehdizadeh S, Tokar J, et al. Double balloon enteroscopy detects smallss lesions missed by capsule endoscopy. Digestive Diseases and Sci-8;53:21403., Costes L, Gauthier M, et al. Chemotherapy of advanced small-enocarcinoma: a multicenter AGEO study. Annals of Oncology

178693. MJ, Hu CY, Kopetz S, et al. A population-based comparison of adeno-a of the large and small intestine: insights into a rare disease. Annalsl Oncology 2012;19:143945.

MJ, Hu CY, Wolff RA, et al. Prognostic value of lymph node evaluationbowel adenocarcinoma: analysis of the surveillance, epidemiology,esults database. Cancer 2010;116:537482.B, Ahuja V, Conway WC, et al. Small bowel adenocarcinoma: under-d undertreated. Annals of Surgical Oncology 2010;17:272832.rson TR, McWilliams RR, Donohue JH, et al. A single-institution expe-th 491 cases of small bowel adenocarcinoma. The American Journaly 2010;199:797803.

MJ, Kopetz S, Lin E, et al. Is there a role for adjuvant therapy in resectedcinoma of the small intestine. Acta Oncologica 2010;49:4749.

M, Baillet P, Hay JM, et al. Factors inuencing long-term survival in of small intestine primary adenocarcinoma. The American Journal of997;173:2379.os IG, Bathe OF, Franceschi D, et al. Extent of resection in theent of duodenal adenocarcinoma. The American Journal of Surgery:3741.

Lillemoe KD, Cameron JL, et al. Adenocarcinoma of the duodenum:uencing long-term survival. Journal of Gastrointestinal Surgery987.FG, Murr MM, Sarr MG, et al. What prognostic factors are importantnal adenocarcinoma. Archives of Surgery 2000;135:63541.S, McCarron EC, Gibbs JF, et al. Surgical management and outcomey adenocarcinoma of the small bowel. Annals of Surgical Oncology22639.h CN, Chao TC, et al. Prognostic factors of primary small bowel ade-ma: univariate and multivariate analysis. World Journal of Surgery3918.

MJ, Varadhachary GR, Kopetz S, et al. Phase II study of capecitabineplatin for advanced adenocarcinoma of the small bowel and ampullaJournal of Clinical Oncology 2009;27:2598603.C, Ross P, Norman A, Hill A, et al. The Royal Marsden experience ofowel adenocarcinoma treated with protracted venous infusion 5-cil. British Journal of Cancer 1998;78:50810.ski P, Hui D. Chemotherapy in small bowel adenocarcinoma: 10-yearce of the British Columbia Cancer Agency. Clinical Oncology: A Journalyal College of Radiologists 2007;19:1439.PN, Pond GR, Moore MJ, et al. Natural history and chemotherapyess for advanced adenocarcinoma of the small bowel: a retrospective

113 cases. American Journal of Clinical Oncology 2006;29:22531.K, Holcroft CA, Kvols LK, et al. Phase II study of 5-uorouracil, doxoru-

mitomycin C for metastatic small bowel adenocarcinoma. Oncologist1327., Bedikian AY, Stroehlein JR. Chemotherapy for primary adenocarci-he small bowel. Cancer 1984;53:235.

104 T. Aparicio et al. / Digestive and Liver Disease 46 (2014) 97 104

[78] Locher C, Malka D, Boige V, et al. Combination chemotherapy in advanced smallbowel adenocarcinoma. Oncology 2005;69:2904.

[79] Overman MJ, Kopetz S, Wen S, et al. Chemotherapy with 5-uorouracil and aplatinum compound improves outcomes in metastatic small bowel adenocar-cinoma. Cancer 2008;113:203845.

[80] Kelland L. The resurgence of platinum-based cancer chemotherapy. NatureReviews Cancer 2007;7:57384.

[81] Zaanan A, Gauthier M, Malka D, et al. Second-line chemotherapy withuorouracil, leucovorin, and irinotecan (FOLFIRI regimen) in patients withadvanced small bowel adenocarcinoma after failure of rst-line platinum-based chemotherapy: a multicenter AGEO study. Cancer 2011;117:14228.

[82] Sobin LH, Gospodarowicz MK, Wittekind C, editors. TNM classication of malig-nant tumours. 7th ed. Wiley Blackwell; 2009. p. 1539.

Small bowel adenocarcinoma: Epidemiology, risk factors, diagnosis and treatment1 Epidemiology1.1 Incidence

2 Etiopathogenetic factors2.1 Environmental factors2.2 Carcinogenesis in SBA2.3 Genetic predisposition2.3.1 Familial adenomatous polyposis (FAP)2.3.2 Lynch syndrome2.3.3 PeutzJeghers syndrome

2.4 Other predisposing conditions2.4.1 Crohn's disease2.4.2 Coeliac disease

3 Diagnosis3.1 Clinical presentation3.2 Diagnosis3.3 Other recommended investigations after SBA diagnosis

4 Prognosis5 Treatment5.1 Localised cancer5.2 Metastatic SBA

References