UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
Hereditary & familial colorectal cancer
Identification, characteristics, surveillance
Kallenberg, F.G.J.
Publication date
2017
Document Version
Other version
License
Other
Link to publication
Citation for published version (APA):
Kallenberg, F. G. J. (2017). Hereditary & familial colorectal cancer: Identification,
characteristics, surveillance.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
6
F.G.J. Kallenberg, A. Latchford, N.C. Lips, C.M. Aalfs, B.A.J. Bastiaansen, S.K. Clark, E. Dekker
Diseases of the Colon & Rectum 2017; In press
duodenal adenomas In patIents
wIth multIple colorectal adenomas
wIthout GermlIne apc or
mutyh mutatIons
6
A B S T R A C T
Background
Patients with genetic adenomatous polyposis syndromes have an increased risk for duodenal cancer and clear surveillance recommendations exist for this group. However, limited data are available on the duodenal phenotype of patients with multiple colorectal adenomas (10-99) without a germline APC or MUTYH mutation.
Objective
We aimed to assess the frequency, extent and progression of duodenal adenomas in patients with multiple colorectal adenomas without a germline APC or MUTYH mutation.
Design
Historical cohort study.
Settings
This study was undertaken at two polyposis registries: the Academic Medical Center, the Netherlands and St. Mark’s Hospital, UK.
Patients
We collected data on all patients with 10-99 colorectal adenomas and absent APC and MUTYH mutations, who underwent ≥1 esophagogastroduodenoscopy.
Main outcome measures
The frequency, extent and progression of duodenal adenomas. Demographic and endoscopic data were collected, described and compared between patients with and without duodenal adenomas.
Results
Eighty-three patients were identified of which eight (9.6%) had duodenal adenomas, detected at a median of 58 years (range 45-75). Duodenal adenomas were detected in 6/8 patients at first esophagogastroduodenoscopy. At diagnosis, all eight patients had Spigelman stage
6
I or II disease. Two of five patients with duodenal adenomas that underwent follow-up esophagogastroduodenoscopies, increased to stage III disease. The other three remained stable. No one developed duodenal cancer. No differences in demographic and endoscopic data were found between patients with and without duodenal adenomas.
Limitations
The retrospective design, selection bias and a small sample size.
Conclusion
Duodenal adenomas are found in a minority of patients with multiple colorectal adenomas without a germline APC or MUTYH mutation, at an average age of 58 years and at diagnosis, disease severity is mild. These results are a first step in unraveling the duodenal phenotype of these patients, which is needed to provide appropriate upper gastrointestinal screening and surveillance recommendations.
6
I N T R O d u C T I O N
Patients with 10 to 99 synchronous or metachronous colorectal adenomas are at risk of having a hereditary adenomatous polyposis syndrome: attenuated familial adenomatous polyposis (AFAP) or MUTYH-associated polyposis (MAP).1 AFAP is an autosomal dominantly
inherited disorder caused by a germline mutation in the APC gene and MAP is the result of biallelic germline mutations in the MUTYH gene.2, 3 As these patients have an increased risk
of developing colorectal cancer (CRC) due to the large number of adenomas, colonoscopic surveillance recommendations are made.4-7
Patients with these syndromes are also prone to develop extra-colonic manifestations, of which duodenal adenomas and cancers are the most frequent.8, 9 The lifetime risk of duodenal cancer
is estimated at 4% to 12%, at an average age of approximately 60 years.4, 5, 9-11 Guidelines therefore
provide upper gastrointestinal screening, surveillance and treatment recommendations from the age of 25 or 30, with a frequency based on the Spigelman staging system (Table 1).4-6
However, in only a minority (9-17%) of patients with 10 to 99 colorectal adenomas, a germline mutation in APC or MUTYH is identified.1 This genetically unexplained syndrome is referred
to as ‘Multiple ColoRectal Adenomas without germline APC or MUTYH mutations (MCRA)’.12
Only a small number of studies has reported on the colorectal phenotype of MCRA patients, but even less is known about the upper gastrointestinal findings in these patients. Only three studies have described the occurrence of duodenal adenomas in this patient group.12-14 Two of
these studies found duodenal adenomas in 32% and 29% of the patients respectively, whereas
Table 1. Modified Spigelman staging system and recommended EGD surveillance intervals 16, 17
Factor
Score
1 Point 2 Points 3 Points
No. of polyps 1-4 5-20 >20
Polyp size, mm 1-4 5-10 >10
Histology Tubular Tubulovillous Villous
Dysplasia Low-grade - High-grade
Total points Spigelman stage Recommended follow-up interval
0 1-4 5-6 7-8 9-12 0 I II III IV 5 years 5 years 3 years 1 year, consider therapy 6-12 months, consider therapy
6
the other reported upper gastro-intestinal findings in none of the patients. However, all study cohorts were small, ranging between nine and 32 patients, and no data were available regarding the extent and progression of duodenal disease.
To date, these patients have been considered to have a phenotype comparable to AFAP and MAP. Therefore, upper gastrointestinal screening and surveillance recommendations are similar for patients with and without identified genetic mutations.4, 5, 7 However, these might be
distinct conditions for which different recommendations could apply. To clarify the duodenal phenotype of MCRA patients, the aim of this study was to elucidate the frequency, extent and progression of duodenal adenomas in a large cohort of Dutch and British MCRA patients.
M A T E R I A L S A N d M E T h O d S
Study design
This was a two-center, historical cohort study from prospectively collected data held at two polyposis registries: the Academic Medical Center (AMC) (Amsterdam, The Netherlands) and St. Mark’s Hospital (SMH) (London, UK).
Study population
Patients visiting the outpatient clinic for polyposis in one of the centers between 1994 and 2016 were identified from both registry databases. We included patients that fulfilled all of the following criteria:
1. Cumulative number of at least 10 colorectal adenomas, but less than 100 adenomas at time of diagnosis;
2. At least 25 years of age at the time of first diagnosis of 10 or more cumulative adenomas; 3. Genetically confirmed absence of a pathogenic APC or biallelic MUTYH mutation.
Patients were excluded if they had not undergone an esophagogastroduodenoscopy (EGD) or if they also fulfilled the WHO criteria for serrated polyposis syndrome.15 In case of uncertainty
regarding the age at MCRA diagnosis or number of colorectal adenomas at baseline, we discussed these patients in an expert panel and excluded them if they were most likely to have >100 adenomas at baseline or >10 adenomas before the age of 25.
Data collection
Demographic data were retrieved from medical records. This included data on gender, age, genetic test details, medical history and family history of CRC in first-degree relatives.
6
Colonoscopy data
Colonoscopy reports were used to collect data regarding the indication for colonoscopy and the location and number of colorectal adenomas. The presence of hyperplastic and sessile serrated polyps was also reported. To describe the distribution of colonic adenomas, the colon was defined as proximal or distal to the splenic flexure. Patients were classified into groups estimating the number of colorectal adenomas, as numbers were often too large for an exact calculation: 10 to 30, 31 to 60, 61 to 99 or >100 cumulative colorectal adenomas. Narrow band imaging, chromo-endoscopy, dye spray and butyl scopolamine bromide were used at the discretion of the gastroenterologist.
EGD data
EGD reports were used to collect data regarding the indication for endoscopy and the location, number and size of duodenal, ampullary and gastric adenomas and fundic gland polyps. The duodenum was divided into D1 to D4. The adenoma size was defined as the largest adenoma endoscopically measured. If the endoscopic size was not reported, the size in the pathology report was used in case of an en bloc resection. Endoscopies were performed using a forward-viewing endoscope and/or a side-forward-viewing endoscope and biopsies and polypectomies were performed at the discretion of the endoscopist. The ampulla of Vater (visualized using a forward- and or a side-viewing endoscope) was only biopsied if it macroscopically appeared dysplastic. EGDs were considered incomplete if the duodenum was visualized up to only D1 and were then excluded. EGD reports without duodenal findings were only included if performed within five years of the MCRA diagnosis date.
Staging and follow-up of duodenal disease
All patients with duodenal adenomas were staged according to the modified Spigelman staging system (Table 1).16, 17 Changes in Spigelman scores over time were described as well as
therapeutic duodenal interventions if performed. As patients were not part of a prospective study protocol, the interval of surveillance was determined by the judgment of the treating gastroenterologist or by the Spigelman staging system.
Histopathology
Histopathology data were retrieved from medical files. Histologically confirmed adenomas were reported as tubular, tubulovillous or villous adenomas with low or high-grade dysplasia, according to the Vienna classification.18 The histopathology of adenomas detected at
colonoscopy was verified in the pathology report, except for patients that had too numerous polyps for endoscopic removal or biopsy. In these cases the polyp histology was assumed
6
by sampling and was later pathologically confirmed in the colectomy specimen if available. Patients with duodenal adenomas had at least one histologically confirmed duodenal adenoma. Furthermore, the above mentioned other gastrointestinal findings (such as fundic gland polyps and sessile serrated polyps) were pathologically confirmed in most cases.
Statistical analysis
We compared demographic and endoscopic data between patients with and without duodenal adenomas using the Fisher’s exact test or Mann-Whitney U test statistics, depending on the type of variable. For analyses purposes, endoscopies and surgeries performed within six months were considered one procedure, as progression in such a short time was unlikely. Age at MCRA diagnosis was defined as the age when a cumulative number of 10 or more colorectal adenomas were detected. Data were combined from the two centers for the purposes of analysis. In all comparisons, p-values smaller than 0.05 were considered to indicate statistically significant differences. All statistical analyses were performed using SPSS statistical software version 23 (IBM SPSS, Chicago, IL, USA).
Ethics
This study was approved by the institutional review board of both centers and it was carried out in accordance with the ethical principles as described in the Helsinki Declaration.
R E S u L T S
A total of 133 patients who fulfilled all inclusion criteria were identified from both databases. Of these patients, 19 (17 AMC, 2 SMH) were excluded, as they also fulfilled the WHO criteria for serrated polyposis syndrome, and 25 had not undergone an EGD at time of inclusion (including one patient with an incomplete EGD). Five patients were discussed in the expert panel, because of uncertainties about the number of adenomas or age at MCRA diagnosis, and all were considered to have FAP and were then excluded. Therefore, 83 MCRA patients (54 from the AMC, 29 from SMH) were included, of whom 60 (72.3%) were male. Median age at first detection of cumulative 10 or more colorectal adenomas was 58.7 (range 27.8-83.6) years. Of these patients, who were all genetically evaluated from 2003 onwards, 78 (94.0%) had full assessment of MUTYH and APC by full sequencing and multiplex ligation-dependent probe amplification (MLPA; to identify duplications and large deletions of the APC and MUTYH genes) and five (6.0%) had full sequencing and MLPA of the APC gene and assessment for the common MUTYH mutations. A total of two patients underwent panel testing that was
6
negative in both. For this purpose, NGS polyposis panel testing was performed, which included the following genes: APC, MUTYH, BMPR1A, SMAD4, STK11, MLH1, MSH2, MSH6, PMS2, PTEN,
POLE, POLD1, ACVRL1, AXIN2, ENG, NTHL1, TSC1 and TSC2. Furthermore, a total of 21 patients
had POLE and POLD1 analyzed in addition to APC and MUTYH testing and a germline mutation in any of the Lynch syndrome associated mismatch repair genes was excluded in four patients. Of all 83 patients, 21 (25.3%) had at least one first-degree relative with CRC; reliable family history data on MCRA (i.e. exact number and type of colorectal polyps) was not available, but none of the patients had a known relative with a clinical diagnosis of FAP. Extra-intestinal cancers were diagnosed in 20 of 83 patients. No notable similarities in specific cancer types were seen amongst these patients, except for Hodgkin lymphoma in three patients. Regarding extra-intestinal manifestations that occur frequently in patients with FAP, no desmoid tumors or osteomas were detected, but one patient was diagnosed with thyroid cancer, one had multiple lipomas and three had an adrenal incidentaloma, of which one was resected and pathology showed adenoma. Patients had not routinely been screened for these extra-intestinal manifestations. None of the patients received chemoprevention to manage their polyposis.
Colonoscopy findings
Indications for baseline colonoscopies varied: gastrointestinal symptoms (n=37; 44.6%), CRC population screening (n=22; 26.5%), family history of CRC or polyposis (n=9; 10.8%), abdominal imaging showing colorectal polyps (n=4; 4.8%) and unknown in nine (10.8%). In two patients (2.4%), duodenal adenomas were detected before a colonoscopy was performed and colonoscopy was performed thereafter to screen for colorectal adenomas.
Of all 83 patients, 69 (83.1%) were diagnosed with 10 or more colonic adenomas at their first colonoscopy. The other 14 patients were diagnosed with MCRA after a median of 2.9 (0.5-12.6) years of surveillance. At the time of diagnosis, 62 patients (74.7%) had between 10 and 30 colorectal adenomas (Table 2). The adenomas were located both in the left and right hemi-colon in 70 patients (84.3%).
In total 71 (85.5%) patients underwent surveillance colonoscopies or sigmoidoscopies with a median follow-up of 3.9 (range 0.7-15.4) years. Of those patients, 17 developed an increasing number of polyps; 14 patients increased to a cumulative number of 31-60 polyps, two patients to 61-99 adenomas and one patient to more than 100 adenomas. Of all 83 patients, 16 (19.3%) were diagnosed with CRC at a median age of 58.3 (range 35.8-72.7) years, all at baseline colonoscopy. Two patients had a second, synchronous CRC and one patient developed a local recurrence after three years. Other colorectal lesions were detected in 52 (62.6%) patients, including hyperplastic and sessile serrated polyps and one patient with a caecal granular cell tumor.
6
A total of 36 (43.4%) patients underwent colorectal surgery. Indications for surgery were CRC in 16, polyp burden/size in 17 and a suspicion of CRC in three patients.
EGD findings
Baseline EGDs were performed at a median age of 60.0 years (range 28.3-83.7). The most common indication for these EGDs was screening for duodenal adenomas (n=67; 80.7%). Other indications included gastrointestinal symptoms in 13 (15.7%), abnormalities on abdominal imaging in one, and was unknown in two patients. Eight patients (9.6%; six from AMC, two from SMH) were diagnosed with duodenal adenomas (Figure 1). Median age at detection of duodenal adenomas was 58.4 years (range 44.9-74.8). In six of eight patients, duodenal adenomas were diagnosed at baseline EGD, in the remaining two patients this was diagnosed 2.2 and 2.3 years after a normal baseline EGD, at the age of 64.6 and 59.7 years respectively. In patients without duodenal adenomas, 26 of 75 patients (34.7%) had at least one follow-up EGD, with a median total interval of 4.8 years (range 0.8-10.6). Other upper gastrointestinal findings detected at baseline or follow-up EGD included fundic gland polyps in 13 patients (15.7%), of whom three had duodenal adenomas. One male patient was diagnosed at the age of 71 with gastric cancer
Table 2. Patient characteristics and colonoscopy findings at time of MCRA diagnosis
Characteristics (n=83)
Median age at MCRA1 diagnosis (range) 58.7 (27.8-83.6)
Male sex, n (%) 60 (72.3)
First-degree relatives with CRC2, n (%) 21 (25.3)
Cum. no. of colorectal adenomas at MCRA diagnosis, n (%)
10 - 30 31 - 60 61 - 99 62 (74.7) 16 (19.3) 5 (6.0)
Colonic adenoma distribution at MCRA diagnosis, n (%)
Right colon Left colon Both-sided Unknown 2 (2.4) 2 (2.4) 70 (84.3) 9 (10.9) CRC2, n (%) 16 (19.3)
Patients with other colorectal findings at MCRA diagnosis, n (%)
Hyperplastic polyps Sessile serrated polyps
Hyperplastic and sessile serrated polyps None
19 (22.9) 6 (7.2) 10 (12.0) 48 (57.8)
6
which was located in the antrum, in the absence of any other upper gastrointestinal findings. The ampulla was visualized, either with forward- or side-viewing endoscopy, and described in 39 patients (47.0%) and was never abnormal.
Six of eight patients with duodenal adenomas had colonoscopic follow-up ranging between 1.5 and 8.0 years, of whom one patient demonstrated an increase in the number of adenomas (from 10-30 to 31-60).
Staging and location of duodenal disease at first detection
At the time of first diagnosis of duodenal adenomas, two patients had a Spigelman stage I, five stage II and one patient stage I or II (unclear as the size of the duodenal adenoma was unknown) (Table 3). Four of those eight patients (50%) had one duodenal adenoma and the maximum number was three adenomas, all with low-grade dysplasia and histology was tubular in six and tubulovillous in two. The median size of the adenomas was 7 (range 1-10) mm and all were located in D2. The ampulla was described in seven patients and was normal in all.
Staging and location of duodenal disease at follow-up
After the first diagnosis of duodenal adenomas, five out of eight (63%) patients with duodenal adenomas had at least one surveillance EGD (Table 4). The median follow-up of these five patients was 2.9 years (range 1.1-8.3). Progression in Spigelman stage occurred in two of five patients at the age of 46 and 51 years respectively, 1.1 and 2.4 years after diagnosis of duodenal adenomas. Spigelman stage changed from II to III and from I to III, respectively. These increases were due to an increased number and size of adenomas in one, and an increased size and development of tubulovillous features in the other patient. One of them was subsequently down-staged from stage III to 0 by endoscopic polypectomy and APC ablation. Three of five patients who received EGD follow-up had a stable Spigelman stage II, 3.0, 2.8 and 1.1 years
Figure 1. Endoscopic image of two duodenal
6
after duodenal adenomas were first detected. Two of these three patients had undergone endoscopic polypectomies, which did not change Spigelman staging. No spontaneous down-staging occurred in any patient. None of patients developed duodenal cancer, high-grade dysplasia or villous duodenal adenomas.
Comparison between patients with and without duodenal adenomas
Between the group diagnosed with duodenal adenomas and the group without duodenal adenomas, there were no significant differences in demographic data and endoscopic findings (Table 5).
d I S C u S S I O N
Limited data are available on the duodenal phenotype of patients with 10 to 99 colorectal adenomas without a germline APC or MUTYH mutation. Therefore, in this historical cohort study, we reported on the frequency, extent and progression of duodenal adenomas in patients with MCRA. Our findings show that duodenal adenomas are detected in nearly 10% of our
Table 3. Details of duodenal adenomas at first detection in eight patients with duodenal adenomas
Characteristics (n=8)
Median number of duodenal adenomas (range) 1.5 (1 - 3)
Median size of largest adenoma, mm (range)1 7 (1-10)
Most advanced histology, n (%)
Tubular adenoma Tubulovillous adenoma Villous adenoma 6 (75) 2 (25)
-Most advanced dysplasia, n (%)
Low-grade High-grade 8 (100) -Location, n (%) D1 D2 D3 -8 (100) -Spigelman stage, n (%)2 Stage I Stage II Stage III Stage IV 2 (29) 5 (71)
6
Ta b le 4 . E G D fo llo w -u p d et ai ls o f e ig ht M C R A p at ie nt s w it h du o de na l a de no ma s Pa t. A g e at fi rs t EG D (y rs ) Sp ig el m an s ta g e a n d s co re R ea so n fo r ad va n ci n g st ag e A t fi rs t EG D A t EG D 2 (f o llow -u p 2) A t EG D 3 (f o llow -u p 2) A t EG D 4 (f o llow -u p 2) A t EG D 5 (f o llow -u p 2) A t EG D 6 (f o llow -u p 2) 1 48 .6 II, 1+ 2+ 2+ 1 1 III , 2 +2 +2 +1 (2 .4 y rs ) II, 2 +2 +1 +1 (3 .4 y rs ) III , 2 +2 +2 +1 (6 .2 y rs ) III , 2 +3 +2 +1 (7 .3 y rs ) III , 2 +3 +2 +1 (8 .3 y rs ) N um b er a nd s iz e 2 74 .8 II 3, 1 +2 +1 +1 N /A 3 62 .4 0 II, 2 +1 /2 +1 +1 (2 .2 y rs ) ? 4, 2 +2 +? +? (3 .0 y rs ) II 5, 1 +2 +1 +1 (4 .6 y rs ) II, 1+ 2+ 1+ 1 (5 .2 y rs ) N /A 4 44 .9 I, 1+ 1+ 1+ 1 III 3, 1 +3 +2 +1 (1 .1 yr s) 0 (2 .2 y rs ) ? 6, 1 +1 +? +? (3 .2 y rs ) 0 (4 .2 y rs ) Si ze a nd h is to lo gy 5 61 .1 II 7, 1 +2 +1 +1 II, 1+ 2+ 1+ 1 (1 .3 y rs ) II 8, 1 +2 +1 +1 (2 .8 y rs ) N /A 6 54 .6 0 0 (1 .1 yr s) 0 (2 .7 y rs) I, 1+ 1+ 1+ 1 (5 .0 y rs ) N /A 7 57 .2 II, 1+ 2+ 2+ 1 1+ 2+ 1+ 1 (1 .1 yr s) N /A 8 45 .3 I o r II 9, 1 +? +1 +1 N /A 1 S co re c o ns is ts o f S pi ge lm an s ta gi ng b as ed p o in ts fo r a d en o m a n um b er + s iz e + h is to lo gy + d ys pl as ia ; 2Fo llo w -u p f ro m b as el in e E G D ; 3 D o w n-st ag ed to S ta ge 0 a ft er re se ct io n o f an a de no m a; 4 L ar ge st tw o re se ct ed , n o h is to lo gy a va ila bl e; 5 O ne p o ly p p ar ti al ly re se ct ed ; 6 D o w n-st ag ed to S ta ge 0 a ft er re se ct io n o f r ec ur re nt a de no m a, n o h is to lo gy a va ila bl e; 7 O ne o f t hr ee a d en o m as r es ec te d; 8 O ne a de no m a re se ct ed ; 9 D o w n-st ag ed to S ta ge 0 a ft er re se ct io n o f a n ad en o m a w it h un kn o w n si ze6
Ta b le 5 . D iff er en ce s b et w ee n M C R A p at ie nt s w it h an d w it ho ut d uo d en al a d en o m as Pa ti en t ch ar ac te ris ti cs ( n =8 3) Pa ti en ts w it h d uo d en al ad en o m as ( n =8 ) Pa ti en ts w it h o ut d uo d en al ad en o m as ( n =7 5) P-va lu e M al e g en d er , n ( % ) 5 (6 2. 5) 55 ( 73 .3 ) 0. 68 1 M ed ia n a g e at b as el in e EG D ( ra n g e) 55 .9 (4 4. 9-74 .8 ) 60 .5 ( 28 .3 -8 3. 7) 0. 28 2 M ed ia n a g e at d uo d en al a d en o m a d et ec ti o n v s at la st E G D fo r th o se w it h o ut d uo d ena l a d en o m as ( ra n g e) 58 .4 (4 4. 9-74 .8 ) 61 .1 (3 4. 2-83 .7 ) 0. 27 2 M ed ia n a g e at M C R A d ia g n o si s (r an g e) 56 .3 (4 4. 3-72 .7 ) 58 .7 ( 27 .8 -8 3. 6) 0. 64 2 Cu m ul at iv e n um b er o f c o lo re ct al a d en o m as a t d ia g n o si s o f d uo d en al ad en o m as v s at la st E G D fo r th o se w it h o ut d uo d en al a d en o m as , n ( % ) 10 -3 0 31 -6 0 61 -9 9 5 (6 2. 5) 3 3 (3 7. 5) 0 44 ( 58 .7 ) 25 ( 33 .3 ) 6 (8 .0 ) 4 1. 0 0 1 Fu n d ic g la n d p o ly p s at d ia g n o si s o f d uo d en al ad en o m as v s at la st E G D fo r th o se w it h o ut d uo de n al a d en o m as , n ( % ) 3 (3 7. 5) 10 ( 13 .3 ) 0. 11 1 C R C fa m ily h is to ry , n ( % ) 1 ( 12 .5 ) 20 ( 26 .7 ) 0. 67 1 C R C , n ( % ) 3 (3 7. 5) 13 ( 17 .3 ) 0. 18 1 1 F is he r’ s e xa ct te st w it h 2 -s id ed P -v al ue ; 2 M an n-W hi tn ey U te st w it h 2 -s id ed P -v al ue ; 3In cl ud in g o ne p at ie nt w it h 0 -5 c o lo re ct al p o ly ps a t t he ti m e o f d uo de na l a de no m a d et ec ti o n, w ho fu lfi lle d M C R A c ri te ri a ( ≥1 0 a de no m as ) 8. 3 ye ar s th er ea ft er ; 4 in cl ud in g o ne p at ie nt w it h >1 0 0 p o ly ps6
patients with MCRA, and these are most commonly diagnosed in their late fifties. At the time of detection, duodenal disease is relatively mild, with all patients having Spigelman stage I or II disease, and a maximum number of three adenomas. Two of five patients who underwent follow-up EGDs increased to a stage III disease, after 1.1 years and 2.4 years respectively; the other three remained stable, although two of them underwent duodenal polypectomies, thereby influencing the Spigelman stage. None of the patients developed duodenal cancer or high-grade dysplastic lesions and no differences in demographic and endoscopic data were found between MCRA patients with and without duodenal adenomas.
This is the largest study to evaluate the frequency of duodenal adenomas in MCRA patients and the first to elucidate their extent and progression after follow-up. All patients with duodenal adenomas had histologically proven duodenal adenomas, which is unique for this kind of studies. And finally, by applying clear inclusion criteria, we created a homogeneous cohort consisting of Dutch and British MCRA patients, thereby increasing the validity of our results. However, several limitations need to be acknowledged. Although APC and MUTYH testing was done in all patients, full sequencing of MUTYH was not done in all patients and other potential genetic causes of adenomatous polyposis, such as polymerase proofreading associated polyposis caused by POLE and POLD1 mutations, were not excluded equally in all patients, potentially limiting the homogeneity of patients. However, these mutations are very rare and are likely to be the cause of only a small percentage of MCRA cases.19 Similarly, the contribution
of unclassified or unrecognized mutations in polyposis-associated genes, mutations in unidentified polyposis-associated genes and mosaicism is unknown.20 Interestingly, based on
our findings, another, yet unknown inherited cause might be considered for MCRA, as a quarter of our patients had at least one first degree relative with CRC, which is more than the expected 10% in the general population.21 Second, due to the retrospective design there was no uniform
screening and surveillance protocol. As a result the age of onset of duodenal adenomas was unknown in most patients. Therefore, it is possible that duodenal adenomas in our cohort were present long before their detection. Since all patients had mild duodenal disease with a Spigelman stage I or II at the time of detection, the age of first development does not seem of much clinical importance. EGDs had not yet been performed in a fifth of MCRA patients at the time of this study and follow-up EGDs were performed in a third of those without duodenal adenomas both of which could influence the results. Furthermore, we only included patients who had been referred to our polyposis outpatient clinic, which could have led to referral bias. Patients with a borderline number of approximately 10 colorectal adenomas might not have been referred, which could influence our description of the colonic and duodenal phenotype of MCRA patients. Moreover, due to differences in historical guidelines between the UK and the Netherlands, SMH genetically tests only those with 10 synchronous or 20 metachronous adenomas, whereas AMC tests patients with 10 synchronous or metachronous adenomas. And
6
finally, endoscopies had been performed by several endoscopists from two different centers, which creates bias. However, both are expert centers with the involved endoscopists being experienced in the management of polyposis syndromes.
The frequency of duodenal adenomas was 9.6% in our cohort. However, in some cases patients were referred to our clinic because of the duodenal adenomas, possibly leading to an overestimation of the frequency of duodenal adenomas. Nevertheless, the frequency of duodenal adenomas in our study is still lower than was reported in two previous retrospective studies, in which nearly a third of patients were affected. One of these studies, by Lefevre et al., reported a frequency of 29%.13 This was based on a small number of 32 MCRA patients and at least
seven of them (22%) had more than 100 colorectal adenomas, indicating that these patients had clinical FAP rather than an attenuated phenotype. Further details about these patients, such as age at time of duodenal adenoma detection, were not described. Tieu et al reported that six of 19 (32%) MCRA patients had duodenal adenomas.12 Those with an abnormal EGD (duodenal
adenomas, fundic gland polyps or both) had a relatively large median number of 54.9 colorectal polyps compared to 36.3 for those with a normal EGD and lesions were detected around the age of 45. In our study most patients had 10-30 polyps, including those with duodenal adenomas, and duodenal adenomas were detected at a median age of 58.4. Although speculative, this could indicate that a larger number of colorectal polyps correlates to developing duodenal polyps and/ or to the development of duodenal adenomas earlier in life (although the older age of duodenal adenoma detection in our study could also reflect the later age at the time of the baseline EGD). This theory is strengthened by the Thirlwell study, in which no upper gastrointestinal findings were seen in 25 patients who were diagnosed with MCRA at a relatively old median age of 61 years and with a low mean polyp number of 24.14 Whether these frequencies differ from the duodenal
adenoma prevalence in MAP and genetically proven AFAP remains unclear, as frequencies in these syndromes are not well known and vary between 17% and 34% and 12% and 25% respectively.3, 8, 9, 11
Nevertheless, the duodenal adenoma frequency in our study is increased compared to that in the general population (0.1-0.3%) and thus likely to be related to colonic polyposis.22
A point of interest is whether the colonic phenotype in patients with duodenal adenomas behaves the same during follow-up as those without. This information could help to identify in whom we should perform EGD and in whom we could safely refrain. If one were to assume that those with duodenal adenomas had a ‘real’ polyposis syndrome, then one might expect that those are the ones that will accumulate more polyps over time, whereas those with sporadic adenomas do not. Our data show that no differences exist in colonic phenotype between the two groups and only one patient with duodenal adenomas showed an increase in the number of colorectal adenomas over follow-up. However, other patients might have shown a small increase as well, but within a colorectal adenoma group (i.e. from 10 to 15 adenomas and thus still in the 10-30 adenoma group). Until more is known, we cannot differ between sporadic
6
adenomas and polyposis syndromes and it thus remains unknown if a 70 year old patient with 10 metachronous adenomas should be treated similarly to a 30 year old patient with a comparable number of adenomas.
Most guidelines now recommend that MCRA patients are managed in a similar way to genetically proven AFAP and MAP patients. Therefore, it is advised to start upper gastrointestinal screening at the age of 25.4, 5, 7 Based on our data, we think it is justified to deviate from these
recommendations. We suggest upper gastrointestinal screening at the time of MCRA diagnosis, irrespective of the age; If at the age of 65 or older and no duodenal adenomas are identified, further EGDs might not be required. Based on FAP data, the 10-year duodenal cancer risk would then be virtually nil.23 If MCRA is diagnosed before the age of 65, we suggest performing
an EGD at diagnosis and if no duodenal adenomas are identified, again around the age of 45 (as our results show that duodenal adenomas are identified from the age of 45 and in all cases in a mild form), then every 5 years until the age of 65 years. Unfortunately, our follow-up data are insufficient to describe the natural course of duodenal adenomas, including the risk of duodenal cancer, and therefore we suggest to apply the Spigelman classification system in case duodenal adenomas are identified, as is done in classical FAP.16
Future multicenter prospective long-term cohort studies in MCRA patients, who have preferentially undergone multigene testing, are needed in order to further develop appropriate evidence-based upper gastrointestinal screening and surveillance guidelines, and results should be compared with patients with an attenuated phenotype with a germline mutation. These studies could also aim, just like in classical FAP, to identify risk factors for the development of duodenal adenomas (including familial occurrence of MCRA) in a search to optimize surveillance strategies. All studies should use similar inclusion criteria, they need to be performed by expert endoscopists and decisions should be made on whether to include patients with a synchronous, rather than a metachronous number of 10 adenomas, and whether having a family history of MCRA or CRC should also be incorporated in the inclusion criteria. Thereafter, a next step would then be to determine if and how often first-degree relatives of MCRA patients should be screened.
In conclusion, duodenal adenomas are found in only a minority of MCRA patients at an age of around 60 and disease severity is mild at that time. Therefore, until more and prospective data are available, we suggest initiating upper gastrointestinal screening at MCRA diagnosis, with surveillance strategies according to the Spigelman classification in case duodenal adenomas are identified and if these are not present, at least 5 yearly surveillance. Surveillance can be terminated at the age of 65, in case no duodenal adenomas are found. These results are a first step in unraveling the phenotype of MCRA, a condition that is most likely to be encountered more frequently, due to the widespread implementation of population screening, the awareness of quality in endoscopy and the advanced endoscopic imaging techniques.
6
R E F E R E N C E S
1. Grover S, Kastrinos F, Steyerberg EW, et al. Prevalence and phenotypes of APC and MUTYH mutations in patients with multiple colorectal adenomas. Jama. 2012;308:485-492.
2. Sieber OM, Lipton L, Crabtree M, et al. Multiple colorectal adenomas, classic adenomatous poly-posis, and germ-line mutations in MYH. N Engl J
Med. 2003;348:791-799.
3. Lynch HT, Smyrk T, McGinn T, et al. Attenuated familial adenomatous polyposis (AFAP). A pheno-typically and genopheno-typically distinctive variant of FAP. Cancer. 1995;76:2427-2433.
4. Syngal S, Brand RE, Church JM, Giardiello FM, Hampel HL, Burt RW. ACG clinical guideline: Genetic testing and management of hereditary gastrointestinal cancer syndromes. Am J
Gastro-enterol. 2015;110:223-262; quiz 263.
5. Stoffel EM, Mangu PB, Gruber SB, et al. Hereditary colorectal cancer syndromes: American Society of Clinical Oncology Clinical Practice Guideline endorsement of the familial risk-colorectal cancer: European Society for Medical Oncology Clinical Practice Guidelines. J Clin Oncol. 2015;33:209-217. 6. Dutch Society For Clinical Genetics. CBO
Guide-line Hereditary Colorectal Cancer 2015. . 7. Balmana J, Balaguer F, Cervantes A, Arnold D.
Familial risk-colorectal cancer: ESMO Clinical Prac-tice Guidelines. Ann Oncol. 2013;24 Suppl 6:vi73-80. 8. Knudsen AL, Bulow S, Tomlinson I, Moslein G,
Heinimann K, Christensen IJ. Attenuated familial adenomatous polyposis: results from an inter-national collaborative study. Colorectal Dis. 2010;12:e243-249.
9. Walton SJ, Kallenberg FG, Clark SK, Dekker E, Latchford A. Frequency and Features of Duodenal Adenomas in Patients With MUTYH-Associated Polyposis. Clin Gastroenterol
Hepatol. 2016;14:986-992.
10. Jasperson KW, Tuohy TM, Neklason DW, Burt RW. Hereditary and familial colon cancer.
Gastroen-terology. 2010;138:2044-2058.
11. Vogt S, Jones N, Christian D, et al. Expanded extracolonic tumor spectrum in MUTYH-associ-ated polyposis. Gastroenterology. 2009;137:1976-1985 e1971-1910.
12. Tieu AH, Edelstein D, Axilbund J, et al. Clinical
Characteristics of Multiple Colorectal Adenoma Patients Without Germline APC or MYH Muta-tions. J Clin Gastroenterol. 2016;50:584-588. 13. Lefevre JH, Parc Y, Svrcek M, et al. APC, MYH, and
the correlation genotype-phenotype in colorectal polyposis. Ann Surg Oncol. 2009;16:871-877. 14. Thirlwell C, Howarth KM, Segditsas S, et al.
Inves-tigation of pathogenic mechanisms in multiple colorectal adenoma patients without germline APC or MYH/MUTYH mutations. Br J Cancer. 2007;96:1729-1734.
15. Snover DC AD, Burt RW, Odze RD. . Serrated polyps of the colon and rectum and serrated (“hyper-plastic”) polyposis.WHO Classification of tumours of the digestive system. . Springer-Verlag; . 2010. 16. Spigelman AD, Williams CB, Talbot IC, Domizio
P, Phillips RK. Upper gastrointestinal cancer in patients with familial adenomatous polyposis.
Lancet. 1989;2:783-785.
17. Saurin JC, Gutknecht C, Napoleon B, et al. Surveil-lance of duodenal adenomas in familial adeno-matous polyposis reveals high cumulative risk of advanced disease. J Clin Oncol. 2004;22:493-498. 18. Schlemper RJ, Riddell RH, Kato Y, et al. The
Vienna classification of gastrointestinal epithelial neoplasia. Gut. 2000;47:251-255.
19. Spier I, Holzapfel S, Altmuller J, et al. Frequency and phenotypic spectrum of germline mutations in POLE and seven other polymerase genes in 266 patients with colorectal adenomas and carci-nomas. Int J Cancer. 2015;137:320-331.
20. Mongin C, Coulet F, Lefevre JH, et al. Unex-plained polyposis: a challenge for geneticists, pathologists and gastroenterologists. Clin
Genet. 2012;81:38-46.
21. de Jong AE, Vasen HF. The frequency of a positive family history for colorectal cancer: a popula-tion-based study in the Netherlands. Neth J Med. 2006;64:367-370.
22. Culver EL, McIntyre AS. Sporadic duodenal polyps: classification, investigation, and manage-ment. Endoscopy. 2011;43:144-155.
23. Groves CJ, Saunders BP, Spigelman AD, Phillips RK. Duodenal cancer in patients with familial adenomatous polyposis (FAP): results of a 10 year prospective study. Gut. 2002;50:636-641.
