Scandinavian Journal of Gastroenterology. 2014; 49: 506–510
CASE REPORT
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
Gastric carcinoid in the absence of atrophic body gastritis and with low Ki67 index: a clinical challenge
EDITH LAHNER1, EMANUELA PILOZZI2, GIANLUCA ESPOSITO1, GLORIA GALLI1 & BRUNO ANNIBALE1 1
Department of Digestive and Liver Disease, Sant’Andrea Hospital, II Medical School, Sapienza University, Rome, Italy, and 2Department of Pathology, Sant’Andrea Hospital, II Medical School, Sapienza University, Rome, Italy
Abstract Background. Gastric carcinoids (GCs) represent 23% of all digestive neuroendocrine tumors (NETs). They can be distinguished into three types: type I (in the presence of atrophic body gastritis, ABG), type II (in the presence of Zollinger-Ellison/multiple endocrine neoplasia type I syndrome), type III (sporadic carcinoids, without any background pathology). Aim. To describe a case of undetermined type of GCs in an Italian referral center for NETs and its prevalence among GCs during a 6-year period. Results. In a case series of 16 GCs seen at our unit between 2007 and 2012, 14 (83.3%) patients had type I carcinoid and 1 patient (6.2%) had type III carcinoid. One patient did not accomplish to the actual classification criteria. This patient had a well-differentiated carcinoid with low Ki67, but multiple gastric biopsies performed at 3-year follow-up gastroscopies excluded the presence of ABG. The patient had fundic cystic polyps, suggesting long-term use of proton pump inhibitors, possibly associated with GCs. Conclusions. This case shows that a GC may occur in the absence of ABG and with low Ki67 index, making classification according to actual criteria difficult. Further studies are needed to better understand the occurrence of this particular type of GCs.
Key Words: gastric carcinoid, immunohistochemistry, type I gastric carcinoid, type III gastric carcinoid, undetermined type
Background Gastric carcinoids (GCs) have always been considered rare tumors, but their incidence has been increasing over the last 30 years, probably due to the progress in diagnostic tools available; literature reports an incidence of 3–4/1,000,000 persons/year, but recent data describe GCs as accounting for 23% of all digestive neuroendocrine tumors (NETs) [1,2]. According to the ENETS guidelines, they are classified into three types: type I GCs are associated with atrophic body gastritis (ABG), type II involve patients with Zollinger-Ellison/multiple endocrine neoplasia type I (MEN-I) syndrome, while type III GCs are sporadic (no association with any background pathology), show a more aggressive behavior, and include NET G3 tumors according to WHO 2010 [3,4].
The major pathogenetic factor for type I GC is hypergastrinemia due to ABG, since gastrin acts as a growth factor for enterochromaffin-like (ECL) cells; thus, in ABG, ECL cells are chronically induced to proliferate, giving rise to a multistep process passing from hyperplasia to dysplasia and then carcinoid [5]. Recently, some papers have suggested that long-term use of proton pump inhibitors (PPIs) may be another possible pathogenetic factor for GCs; in fact, these drugs induce chronic hypergastrinemia, and may subsequently cause ECL cells proliferation and carcinoids that are different from type I because they arise without ABG [6,7]. Beyond this hypothesis, cases escaping the classifications available to describe GCs have been poorly reported so far; here we present the case of a woman, observed in a tertiary referral center for ABG and NETs, affected by a undetermined type GC, since
Correspondence: Bruno Annibale, Department of Digestive and Liver Disease, Sant’Andrea Hospital, Via di Grottarossa 1035–1039, 00189 Rome, Italy. E-mail: [email protected]
(Received 31 October 2013; revised 3 December 2013; accepted 7 December 2013) ISSN 0036-5521 print/ISSN 1502-7708 online 2014 Informa Healthcare DOI: 10.3109/00365521.2013.878381
Gastric carcinoid without atrophic gastritis and low Ki67 it was a well-differentiated carcinoid with low proliferative index (expressed as Ki67) but no evidence of ABG; the patient reported previous use of PPIs.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
Methods We retrospectively evaluated the clinical charts of all the new cases of GCs seen between 2007 and 2012 in our unit, a tertiary referral center for ABG and NETs. Median endoscopic follow-up of the study population was calculated as the time interval between the first and the last endoscopy performed by each patient. Esophagogastroduodenoscopies (EGDs) were performed under sedation with Olympus video gastroscope GIF-Q165. During each EGD all polyps found were resected, and a complete bioptic gastric mapping was performed, according to a previously described protocol [8]: biopsies were collected in gastric body (four samples), fundus (four samples), and antrum (two samples), using a standard oval fenestrated biopsy forceps (Olympus Optical Co., Ltd., Tokyo, Japan). In case of GCs presenting as polyps >1.5 cm of diameter, suspected incomplete resection, or carcinoids other than type I, endoscopic ultrasonography was performed after sedation by intravenous propofol using a Pentax EG-3630UR endoscope, to assess the polyp complete removal and to exclude lymph nodal involvement [3]. The specimens were formalin-fixed and routinely processed. Five-micrometer-thick mucosal gastric sections were stained with hematoxylin-eosin for routine examination and Giemsa staining for Helicobacter pylori (Hp). Immunohistochemistry (IHC) was performed using a monoclonal antibody anti-Chromogranin A (CgA) (Clone DAK-A3; Dako, Glostrup, Denmark), Ki67 (Clone MIB1, Dako, Glostrup, Denmark), and a rabbit polyclonal antibody anti-Gastrin (Dako Code A 0568), and visualized by Envision-Flex (Dako) in a Dako Autostainer instrument according to the manufacturer’s instructions. Serological assays were performed in every patient, evaluating: fasting serum gastrinemia (nv < 40 pg/ml), CgA (nv < 98 ng/ml), anti-parietal cells antibodies (PCAs), pepsinogen I (nv > 25 ng/ml), and IgG Hp antibodies (nv < 21 UI/ml). Gastrin and CgA were measured by radioimmunoassay (RIA), pepsinogen I was assessed by an RIA commercial kit (Pepsik; Sorin, Saluggia, Italy), PCA was evaluated on serum using a solid-phase immunosorbent assay commercial kit (Autostat, Cogent Diagnostic Ltd, Edinburgh, UK), while Hp antibodies IgG were measured by an ELISA commercial kit (GAP test IgG; Biorad, Milan, Italy). Anemia pattern was also assessed, evaluating
507
hemoglobin, mean corpuscolar value (MCV), ferritin, and vitamin B12 values. ABG diagnosis was based on the presence of hypergastrinemia and atrophy of the body mucosa as confirmed by histological evaluation. According to the updated Sydney System, gastric body atrophy was defined as focal or complete replacement of oxyntic glands by metaplastic pyloric or intestinal glands; this variable was graded on a four-grade scale represented by absence of replacement (score 0), replacement to a mild degree (score 1), moderate degree (score 2), or severe degree (score 3), as previously reported [9]. GC diagnosis was performed when ECL cells’ proliferation was >500 mm [10] classified according to WHO 2010 criteria [4] as follows: NET G1 if they had Ki67 £2%, NET G2 if Ki67 was 3–20%, and NET G3 in the case of Ki67 >20%. In the case of patients referred from other hospitals to our unit (that is a tertiary referral center for ABG and NETs), they were considered in this report only if carcinoid diagnosis had been performed within the previous 12 months; moreover, in these cases, a histological revision by our expert pathologists was performed. Hp status was considered as positive when a positive Hp immunoglobulin G titer was detected and/or bacteria were revealed at histology. MEN-I was excluded by means of a questionnaire regarding symptoms, personal and family history, and through appropriate laboratory tests [11,12]. Type I GCs were treated with endoscopic management, as described elsewhere [8,13]. Type III GCs were accurately staged and treated with total gastrectomy, according to the ENETS Guidelines [3]. Results Sixteen GCs were observed in our unit in this 6-year period (Figure 1): nine patients were new diagnosis while seven were referred from other hospitals. Median endoscopic follow-up time of the series was 19 months (range 2–48). Type I GCs were 14 (87.6%) and all of them had serological alterations related to ABG. Type III GC was found in one (6.2%) patient who underwent total gastrectomy, and histological report described a well-differentiated NET with Ki67 of 60% (T2N0M0). In this case series, type II GCs were not detected, but one (6.2%) patient presented an undetermined type of GC, as she had a histological diagnosis of GC not responding to ENETS classification criteria for GCs. This patient was a 48-year-old woman with hyperthyroidism treated with tapazole, hypertension treated with sartans, and mild iron deficiency anemia due to gynecological causes. A first EGD (for dyspepsia and
508
E. Lahner et al.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
Gastric carcinoids n = 16 Age: 66 (31–78) years Female gender: 14 (87.5%) Fasting gastrin: 775 (37–5726) pg/ml Pepsinogen l: 8 (2–52) µg Chromogranin A: 241 (42–474) ng/ml
Type l n = 14 (87.6%) Age: 66 (31–78) years Female gender: 12 (85.7%) Fasting gastrin: 775 (37–5726) pg/ml Pepsinogen l: 6 (2–13) µg Chromogranin A: 241 (42–474) ng/ml
Type lll n = 1 (6.2%) Age: 78 years Female Fasting gastrin: 47pg/ml Pepsinogen l: 45 µg Chromogranin A: 58 ng/ml
Undetermined type n = 1 (6.2%) Age: 48 years Female Fasting gastrin: 37pg/ml Pepsinogen l: 36 µg Chromogranin A: 57 ng/ml
Figure 1. Demographics and biochemical characteristics of patients with gastric carcinoid seen at our unit between 2007 and 2012 (continuous variables are expressed as median and range).
gastroesophageal reflux disease) had shown three subcentimetric polyps of gastric body (histological report not available); she had then started PPIs. After 8 years she underwent a new EGD, showing small polyps in gastric fundus (no biopsies performed); a pedunculated, 10-mm erythematous polyp under the cardia was resected, and diagnosed as well-differentiated carcinoid (Ki67: 1%). The patient was referred to us after 7 months; histological revision confirmed an NET G1 tumor (Figure 2), supported by the negativity of A
B
C
D
Ki67 immunostaining (Figure 3). PPIs were stopped 1 month before serological testing which were all within the normal range, and before EGD, showing small multiple cystic polyps of gastric fundus; histological evaluation of multiple gastric biopsies diagnosed a chronic corpus gastritis without any sign of active inflammation, atrophy, pyloric or intestinal metaplasia, Hp negative. Strict follow-up was planned; after 6 months, EGD confirmed the cystic polyps, and bioptic mapping was negative for ABG but showed focal simple ECL cells hyperplasia. CT scan was normal. Total follow-up rated for 36 months, without any GC recurrence, and bioptic mapping remained negative for ABG. Thus, this carcinoid does not fit with actual ENETS classification criteria for GCs and may be classified either as a type I-like carcinoid without hypergastrinemia and ABG or as type III-like carcinoid with low Ki67 index. Discussion
Figure 2. Hematoxylin-eosin and immunohistochemical staining for Chromogranin A (CgA) (20X). A: Fragment of the gastric carcinoid resected at the first EGD. B–C. Representative fundus/ corpus biopsies of two different EGDs, with no evidence of both atrophy and metaplasia. Focal chronic inflammatory infiltrate was observed in one biopsy in the deep portion of lamina propria. D: CgA immunostaining in gastric bioptic mapping showed focal simple ECL hyperplasia.
This study describes a case series of GCs diagnosed in a referral center for NETs during a 6-year period. As previously reported [2,3], type I is the most frequent type of GCs, representing 83.3% of this population. As indicated in Figure 1, these patients, being affected by ABG, were mostly women (87.5%) with serological alterations related to ABG. One case of type III GC was also found, and surgically treated; as expected [3], type III represented the less-frequent GC (6.2% of the study population). The peculiarity of this case series is, however, the occurrence of one GC (6.2% of the series), not responding to the ENETS
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
Gastric carcinoid without atrophic gastritis and low Ki67
Figure 3. Ki67 immunostaining (10): The left side shows negative immunostaining in gastric carcinoid fragment; the right side (arrow) shows epithelial cells as positive control.
classification criteria available. It was in fact represented by a well-differentiated tumor with low Ki67 (NET G1) and small diameter; although these features were suggestive for a type I GC, hypergastrinemia and ABG were excluded, thus rather suggesting a benign type III GC. The possibility of a diagnostic mistake was reasonably avoided, repeating endoscopic biopsies and gastric atrophy serological markers during a 3-year follow-up; moreover, both previous and present Hp infections were accurately excluded through both Giemsa staining and IgG antibodies [9], and G cells hyperplasia was excluded through specific IHC. This case did not respond to type II classification criteria, as MEN-I syndrome was not present, and did not show features compatible with type III carcinoids. Genetic profiling was not performed, as beyond MEN-I alterations (and these carcinoids did not respond to type II criteria) no specific abnormalities have been described in GCs so far. The matter that classification available is not able to perfectly define all the GC patients emerges from these data, and as this undetermined type carcinoid represent 6% of the population observed in a 6-year period in our center, this can be considered a new clinical challenge. For what concerns hypotheses of pathogenesis, the patient had used PPIs for years, as confirmed by the presence of fundic polyps; hypergastrinemia, a consequence of long-term use of these drugs, might have contributed to GC development, as gastrin acts as a growth factor on ECL cells [7,14]. This hypothesis seems to be supported by the presence of simple ECL hyperplasia, and no new GCs recurring after stopping PPIs at 3-year follow-up. In the literature, three cases of GCs related to PPIs have been recently reported
509
[6,7]. One case was characterized by a poorly differentiated GC (thus with histological features different from our cases) in the presence of ECL cells hyperplasia but without ABG. The other two cases were more similar to the undetermined type GC we describe; in fact they were NET G1 carcinoids in the absence of ABG or hypergastrinemia, one endoscopically resected and the other disappearing after PPIs discontinuation. The authors concluded suggesting long-term use of PPIs as a pathogenetic factor for the occurrence of these carcinoids. However, these reports have some differences from our data: In one patient, follow-up was not reported, and the exclusion of ABG was evaluated only with one gastric bioptic mapping; moreover, the presence of Hp was investigated only by gastric biopsies, and previous infection was not excluded through IgG antibodies. Instead, in the other case follow-up is described, but active Hp infection was found and treated but not considered in the discussion. These data may in our opinion represent a bias, limiting the possibility to support PPIs as a possible pathogenetic factor for GC occurrence. The inappropriate use of PPIs is an emergent problem in clinical practice; among the risks of a long-term use [15–17], it is well known that hypergastrinemia and ECL cells growth may be further complications [18]. A recent randomized trial has evaluated the changes in exocrine and endocrine gastric cells [19] in patients with gastroesophageal reflux disease treated with standard dose of PPIs versus laparoscopic fundoplicatio; Fiocca et al. thus observed an increase in ECL cells proliferation in the arm of PPIs (hyperplasia), but did not diagnose any GC after 5-year use. Data in the literature about the use of PPIs for more than 5 years are scanty, in terms of effects on ECL cells; thus, the cause–effect relationship between these drugs and the development of GCs in the absence of ABG remain a hypothesis that still awaits confirmation by long-term longitudinal studies. However, we are not able to exclude that our patient actually had a non-aggressive type III GC with low proliferative index, which develops completely independently from ECL hyperplasia and eventual PPI use [20]. In conclusion, this case shows that a GC may occur in the absence of ABG and with low Ki67 index, making classification according to actual criteria difficult. Further studies are needed to better understand the occurrence of this particular type of GCs.
Declaration of interests: This study was funded by research grants from “Sapienza” University of Rome (2010–2011).
510
E. Lahner et al.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
References [1] Modlin IM, Oberg K, Chung DC, et al. Gastroenteropancreatic endocrine tumours. Lancet Oncol 2008;9:61–72. [2] Niederle MB, Hackl M, Kaserer K, Niederle B. Gastroenteropancreatic neuroendocrine tumours: the current incidence and staging based on the WHO and European Neuroendocrine Tumour Society classification: an analysis based on prospectively collected parameters. Endocr Relat Cancer 2010;17:909–18. [3] Delle Fave G, Kwekkeboom DJ, Van Cutsem E, et al. ENETS Consensus Guidelines for the management of patients with gastroduodenal neoplasms. Neuroendocrinology 2012;95:74–87. [4] Rindi G, Arnold R, Bosman FT, et al. Nomenclature and classification of neuroendocrineneoplasms of the digestive system. In: Bosman FT, et al, editors. WHO Classification of Tumours of the Digestive System. Lyon: IARC press; 2010. pp13-4 [5] Delle Fave G, Capurso G, Milione M, Panzuto F. Endocrine tumours of the stomach. Best Pract Res Clin Gastroenterol 2005;19:659–73. [6] Jianu CS, Fossmark R, Viset T, et al. Gastric carcinoids after long-term use of proton pump inhibitor. Aliment Pharmacol Ther 2012;36:644–9. [7] Jianu CS, Lange OJ, Viset T, et al. Gastric neuroendocrine carcinoma after long-term use of proton pump inhibitor. Scand J Gastroenterol 2012;47:64–7. [8] Annibale B, Azzoni C, Corleto VD, et al. Atrophic body gastritis patients with enterochromaffin-like cell dysplasia are at increased risk for the development of type I gastric carcinoid. Eur J Gastroenterol Hepatol 2001;13:1449–56. [9] Annibale B, Marignani M, Azzoni C, et al. Atrophic body gastritis: distinct features associated with Helicobacter pylori infection. Helicobacter 1997;2:57–64. [10] Solcia E, Bordi C, Creutzfeldt W, et al. Histopathological classification of nonantral gastric endocrine growths in man. Digestion 1998;41:185–200. [11] Berna MJ, Annibale B, Marignani M, et al. A prospective study of gastric carcinoids and enterochromaffin-like cell
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
changes in multiple endocrine neoplasia type 1 and Zollinger-Ellison syndrome: identification of risk factors. J Clin Endocrinol Metab 2008;93:1582–91. Bordi C, Corleto VD, Azzoni C, et al. The antral mucosa as a new site for endocrine tumors in multiple endocrine neoplasia type 1 and Zollinger-Ellison syndromes. J Clin Endocrinol Metab 2001;86:2236–42. Merola E, Sbrozzi Vanni A, Panzuto F, et al.Type I. Gastric Carcinoids: A Prospective Study on Endoscopic Management and Recurrence Rate. Neuroendocrinology 2011;95: 207–13. Koop H, Klein M, Arnold R. Serum gastrin levels during long-term omeprazole treatment. Aliment Pharmacol Ther 1990;4:131–8. Ngamruengphong S, Leontiadis GI, Radhi S, Dentino A, Nugent K. Proton pump inhibitors and risk of fracture: a systematic review and meta-analysis of observational studies. Am J Gastroenterol 2011;106:1209–18. Kim YG, Graham DY, Jang BI. Proton pump inhibitor use and recurrent Clostridium difficile-associated disease: a casecontrol analysis matched by propensity score. J Clin Gastroenterol 2012;46:397–400. Schmidt M, Johansen MB, Robertson DJ, et al. Concomitant use of clopidogrel and proton pump inhibitors is not associated with major adverse cardiovascular events following coronary stent implantation. Aliment Pharmacol Ther 2012;35:165–74. Burkitt MD, Varro A, Pritchard DM. Importance of gastrin in the pathogenesis and treatment of gastric tumors. World J Gastroenterol 2009;15:1–16. Fiocca R, Mastracci L, Attwood SE, et al. Gastric exocrine and endocrine cell morphology under prolonged acid inhibition therapy: results of a 5-year follow-up in the LOTUS trial. Aliment Pharmacol Ther 2012;36: 959–71. Rindi G, Azioni C, La Rosa S, Klersy C, Parlotti D, Rappel S, et al.L cell tumor and poorly differentiated endocrine carcinoma of the stomach: prognostic evaluation by pathological analysis. Gastroenterology 1999;116: 532–41.
CASE REPORT
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
Gastric carcinoid in the absence of atrophic body gastritis and with low Ki67 index: a clinical challenge
EDITH LAHNER1, EMANUELA PILOZZI2, GIANLUCA ESPOSITO1, GLORIA GALLI1 & BRUNO ANNIBALE1 1
Department of Digestive and Liver Disease, Sant’Andrea Hospital, II Medical School, Sapienza University, Rome, Italy, and 2Department of Pathology, Sant’Andrea Hospital, II Medical School, Sapienza University, Rome, Italy
Abstract Background. Gastric carcinoids (GCs) represent 23% of all digestive neuroendocrine tumors (NETs). They can be distinguished into three types: type I (in the presence of atrophic body gastritis, ABG), type II (in the presence of Zollinger-Ellison/multiple endocrine neoplasia type I syndrome), type III (sporadic carcinoids, without any background pathology). Aim. To describe a case of undetermined type of GCs in an Italian referral center for NETs and its prevalence among GCs during a 6-year period. Results. In a case series of 16 GCs seen at our unit between 2007 and 2012, 14 (83.3%) patients had type I carcinoid and 1 patient (6.2%) had type III carcinoid. One patient did not accomplish to the actual classification criteria. This patient had a well-differentiated carcinoid with low Ki67, but multiple gastric biopsies performed at 3-year follow-up gastroscopies excluded the presence of ABG. The patient had fundic cystic polyps, suggesting long-term use of proton pump inhibitors, possibly associated with GCs. Conclusions. This case shows that a GC may occur in the absence of ABG and with low Ki67 index, making classification according to actual criteria difficult. Further studies are needed to better understand the occurrence of this particular type of GCs.
Key Words: gastric carcinoid, immunohistochemistry, type I gastric carcinoid, type III gastric carcinoid, undetermined type
Background Gastric carcinoids (GCs) have always been considered rare tumors, but their incidence has been increasing over the last 30 years, probably due to the progress in diagnostic tools available; literature reports an incidence of 3–4/1,000,000 persons/year, but recent data describe GCs as accounting for 23% of all digestive neuroendocrine tumors (NETs) [1,2]. According to the ENETS guidelines, they are classified into three types: type I GCs are associated with atrophic body gastritis (ABG), type II involve patients with Zollinger-Ellison/multiple endocrine neoplasia type I (MEN-I) syndrome, while type III GCs are sporadic (no association with any background pathology), show a more aggressive behavior, and include NET G3 tumors according to WHO 2010 [3,4].
The major pathogenetic factor for type I GC is hypergastrinemia due to ABG, since gastrin acts as a growth factor for enterochromaffin-like (ECL) cells; thus, in ABG, ECL cells are chronically induced to proliferate, giving rise to a multistep process passing from hyperplasia to dysplasia and then carcinoid [5]. Recently, some papers have suggested that long-term use of proton pump inhibitors (PPIs) may be another possible pathogenetic factor for GCs; in fact, these drugs induce chronic hypergastrinemia, and may subsequently cause ECL cells proliferation and carcinoids that are different from type I because they arise without ABG [6,7]. Beyond this hypothesis, cases escaping the classifications available to describe GCs have been poorly reported so far; here we present the case of a woman, observed in a tertiary referral center for ABG and NETs, affected by a undetermined type GC, since
Correspondence: Bruno Annibale, Department of Digestive and Liver Disease, Sant’Andrea Hospital, Via di Grottarossa 1035–1039, 00189 Rome, Italy. E-mail: [email protected]
(Received 31 October 2013; revised 3 December 2013; accepted 7 December 2013) ISSN 0036-5521 print/ISSN 1502-7708 online 2014 Informa Healthcare DOI: 10.3109/00365521.2013.878381
Gastric carcinoid without atrophic gastritis and low Ki67 it was a well-differentiated carcinoid with low proliferative index (expressed as Ki67) but no evidence of ABG; the patient reported previous use of PPIs.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
Methods We retrospectively evaluated the clinical charts of all the new cases of GCs seen between 2007 and 2012 in our unit, a tertiary referral center for ABG and NETs. Median endoscopic follow-up of the study population was calculated as the time interval between the first and the last endoscopy performed by each patient. Esophagogastroduodenoscopies (EGDs) were performed under sedation with Olympus video gastroscope GIF-Q165. During each EGD all polyps found were resected, and a complete bioptic gastric mapping was performed, according to a previously described protocol [8]: biopsies were collected in gastric body (four samples), fundus (four samples), and antrum (two samples), using a standard oval fenestrated biopsy forceps (Olympus Optical Co., Ltd., Tokyo, Japan). In case of GCs presenting as polyps >1.5 cm of diameter, suspected incomplete resection, or carcinoids other than type I, endoscopic ultrasonography was performed after sedation by intravenous propofol using a Pentax EG-3630UR endoscope, to assess the polyp complete removal and to exclude lymph nodal involvement [3]. The specimens were formalin-fixed and routinely processed. Five-micrometer-thick mucosal gastric sections were stained with hematoxylin-eosin for routine examination and Giemsa staining for Helicobacter pylori (Hp). Immunohistochemistry (IHC) was performed using a monoclonal antibody anti-Chromogranin A (CgA) (Clone DAK-A3; Dako, Glostrup, Denmark), Ki67 (Clone MIB1, Dako, Glostrup, Denmark), and a rabbit polyclonal antibody anti-Gastrin (Dako Code A 0568), and visualized by Envision-Flex (Dako) in a Dako Autostainer instrument according to the manufacturer’s instructions. Serological assays were performed in every patient, evaluating: fasting serum gastrinemia (nv < 40 pg/ml), CgA (nv < 98 ng/ml), anti-parietal cells antibodies (PCAs), pepsinogen I (nv > 25 ng/ml), and IgG Hp antibodies (nv < 21 UI/ml). Gastrin and CgA were measured by radioimmunoassay (RIA), pepsinogen I was assessed by an RIA commercial kit (Pepsik; Sorin, Saluggia, Italy), PCA was evaluated on serum using a solid-phase immunosorbent assay commercial kit (Autostat, Cogent Diagnostic Ltd, Edinburgh, UK), while Hp antibodies IgG were measured by an ELISA commercial kit (GAP test IgG; Biorad, Milan, Italy). Anemia pattern was also assessed, evaluating
507
hemoglobin, mean corpuscolar value (MCV), ferritin, and vitamin B12 values. ABG diagnosis was based on the presence of hypergastrinemia and atrophy of the body mucosa as confirmed by histological evaluation. According to the updated Sydney System, gastric body atrophy was defined as focal or complete replacement of oxyntic glands by metaplastic pyloric or intestinal glands; this variable was graded on a four-grade scale represented by absence of replacement (score 0), replacement to a mild degree (score 1), moderate degree (score 2), or severe degree (score 3), as previously reported [9]. GC diagnosis was performed when ECL cells’ proliferation was >500 mm [10] classified according to WHO 2010 criteria [4] as follows: NET G1 if they had Ki67 £2%, NET G2 if Ki67 was 3–20%, and NET G3 in the case of Ki67 >20%. In the case of patients referred from other hospitals to our unit (that is a tertiary referral center for ABG and NETs), they were considered in this report only if carcinoid diagnosis had been performed within the previous 12 months; moreover, in these cases, a histological revision by our expert pathologists was performed. Hp status was considered as positive when a positive Hp immunoglobulin G titer was detected and/or bacteria were revealed at histology. MEN-I was excluded by means of a questionnaire regarding symptoms, personal and family history, and through appropriate laboratory tests [11,12]. Type I GCs were treated with endoscopic management, as described elsewhere [8,13]. Type III GCs were accurately staged and treated with total gastrectomy, according to the ENETS Guidelines [3]. Results Sixteen GCs were observed in our unit in this 6-year period (Figure 1): nine patients were new diagnosis while seven were referred from other hospitals. Median endoscopic follow-up time of the series was 19 months (range 2–48). Type I GCs were 14 (87.6%) and all of them had serological alterations related to ABG. Type III GC was found in one (6.2%) patient who underwent total gastrectomy, and histological report described a well-differentiated NET with Ki67 of 60% (T2N0M0). In this case series, type II GCs were not detected, but one (6.2%) patient presented an undetermined type of GC, as she had a histological diagnosis of GC not responding to ENETS classification criteria for GCs. This patient was a 48-year-old woman with hyperthyroidism treated with tapazole, hypertension treated with sartans, and mild iron deficiency anemia due to gynecological causes. A first EGD (for dyspepsia and
508
E. Lahner et al.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
Gastric carcinoids n = 16 Age: 66 (31–78) years Female gender: 14 (87.5%) Fasting gastrin: 775 (37–5726) pg/ml Pepsinogen l: 8 (2–52) µg Chromogranin A: 241 (42–474) ng/ml
Type l n = 14 (87.6%) Age: 66 (31–78) years Female gender: 12 (85.7%) Fasting gastrin: 775 (37–5726) pg/ml Pepsinogen l: 6 (2–13) µg Chromogranin A: 241 (42–474) ng/ml
Type lll n = 1 (6.2%) Age: 78 years Female Fasting gastrin: 47pg/ml Pepsinogen l: 45 µg Chromogranin A: 58 ng/ml
Undetermined type n = 1 (6.2%) Age: 48 years Female Fasting gastrin: 37pg/ml Pepsinogen l: 36 µg Chromogranin A: 57 ng/ml
Figure 1. Demographics and biochemical characteristics of patients with gastric carcinoid seen at our unit between 2007 and 2012 (continuous variables are expressed as median and range).
gastroesophageal reflux disease) had shown three subcentimetric polyps of gastric body (histological report not available); she had then started PPIs. After 8 years she underwent a new EGD, showing small polyps in gastric fundus (no biopsies performed); a pedunculated, 10-mm erythematous polyp under the cardia was resected, and diagnosed as well-differentiated carcinoid (Ki67: 1%). The patient was referred to us after 7 months; histological revision confirmed an NET G1 tumor (Figure 2), supported by the negativity of A
B
C
D
Ki67 immunostaining (Figure 3). PPIs were stopped 1 month before serological testing which were all within the normal range, and before EGD, showing small multiple cystic polyps of gastric fundus; histological evaluation of multiple gastric biopsies diagnosed a chronic corpus gastritis without any sign of active inflammation, atrophy, pyloric or intestinal metaplasia, Hp negative. Strict follow-up was planned; after 6 months, EGD confirmed the cystic polyps, and bioptic mapping was negative for ABG but showed focal simple ECL cells hyperplasia. CT scan was normal. Total follow-up rated for 36 months, without any GC recurrence, and bioptic mapping remained negative for ABG. Thus, this carcinoid does not fit with actual ENETS classification criteria for GCs and may be classified either as a type I-like carcinoid without hypergastrinemia and ABG or as type III-like carcinoid with low Ki67 index. Discussion
Figure 2. Hematoxylin-eosin and immunohistochemical staining for Chromogranin A (CgA) (20X). A: Fragment of the gastric carcinoid resected at the first EGD. B–C. Representative fundus/ corpus biopsies of two different EGDs, with no evidence of both atrophy and metaplasia. Focal chronic inflammatory infiltrate was observed in one biopsy in the deep portion of lamina propria. D: CgA immunostaining in gastric bioptic mapping showed focal simple ECL hyperplasia.
This study describes a case series of GCs diagnosed in a referral center for NETs during a 6-year period. As previously reported [2,3], type I is the most frequent type of GCs, representing 83.3% of this population. As indicated in Figure 1, these patients, being affected by ABG, were mostly women (87.5%) with serological alterations related to ABG. One case of type III GC was also found, and surgically treated; as expected [3], type III represented the less-frequent GC (6.2% of the study population). The peculiarity of this case series is, however, the occurrence of one GC (6.2% of the series), not responding to the ENETS
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
Gastric carcinoid without atrophic gastritis and low Ki67
Figure 3. Ki67 immunostaining (10): The left side shows negative immunostaining in gastric carcinoid fragment; the right side (arrow) shows epithelial cells as positive control.
classification criteria available. It was in fact represented by a well-differentiated tumor with low Ki67 (NET G1) and small diameter; although these features were suggestive for a type I GC, hypergastrinemia and ABG were excluded, thus rather suggesting a benign type III GC. The possibility of a diagnostic mistake was reasonably avoided, repeating endoscopic biopsies and gastric atrophy serological markers during a 3-year follow-up; moreover, both previous and present Hp infections were accurately excluded through both Giemsa staining and IgG antibodies [9], and G cells hyperplasia was excluded through specific IHC. This case did not respond to type II classification criteria, as MEN-I syndrome was not present, and did not show features compatible with type III carcinoids. Genetic profiling was not performed, as beyond MEN-I alterations (and these carcinoids did not respond to type II criteria) no specific abnormalities have been described in GCs so far. The matter that classification available is not able to perfectly define all the GC patients emerges from these data, and as this undetermined type carcinoid represent 6% of the population observed in a 6-year period in our center, this can be considered a new clinical challenge. For what concerns hypotheses of pathogenesis, the patient had used PPIs for years, as confirmed by the presence of fundic polyps; hypergastrinemia, a consequence of long-term use of these drugs, might have contributed to GC development, as gastrin acts as a growth factor on ECL cells [7,14]. This hypothesis seems to be supported by the presence of simple ECL hyperplasia, and no new GCs recurring after stopping PPIs at 3-year follow-up. In the literature, three cases of GCs related to PPIs have been recently reported
509
[6,7]. One case was characterized by a poorly differentiated GC (thus with histological features different from our cases) in the presence of ECL cells hyperplasia but without ABG. The other two cases were more similar to the undetermined type GC we describe; in fact they were NET G1 carcinoids in the absence of ABG or hypergastrinemia, one endoscopically resected and the other disappearing after PPIs discontinuation. The authors concluded suggesting long-term use of PPIs as a pathogenetic factor for the occurrence of these carcinoids. However, these reports have some differences from our data: In one patient, follow-up was not reported, and the exclusion of ABG was evaluated only with one gastric bioptic mapping; moreover, the presence of Hp was investigated only by gastric biopsies, and previous infection was not excluded through IgG antibodies. Instead, in the other case follow-up is described, but active Hp infection was found and treated but not considered in the discussion. These data may in our opinion represent a bias, limiting the possibility to support PPIs as a possible pathogenetic factor for GC occurrence. The inappropriate use of PPIs is an emergent problem in clinical practice; among the risks of a long-term use [15–17], it is well known that hypergastrinemia and ECL cells growth may be further complications [18]. A recent randomized trial has evaluated the changes in exocrine and endocrine gastric cells [19] in patients with gastroesophageal reflux disease treated with standard dose of PPIs versus laparoscopic fundoplicatio; Fiocca et al. thus observed an increase in ECL cells proliferation in the arm of PPIs (hyperplasia), but did not diagnose any GC after 5-year use. Data in the literature about the use of PPIs for more than 5 years are scanty, in terms of effects on ECL cells; thus, the cause–effect relationship between these drugs and the development of GCs in the absence of ABG remain a hypothesis that still awaits confirmation by long-term longitudinal studies. However, we are not able to exclude that our patient actually had a non-aggressive type III GC with low proliferative index, which develops completely independently from ECL hyperplasia and eventual PPI use [20]. In conclusion, this case shows that a GC may occur in the absence of ABG and with low Ki67 index, making classification according to actual criteria difficult. Further studies are needed to better understand the occurrence of this particular type of GCs.
Declaration of interests: This study was funded by research grants from “Sapienza” University of Rome (2010–2011).
510
E. Lahner et al.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
References [1] Modlin IM, Oberg K, Chung DC, et al. Gastroenteropancreatic endocrine tumours. Lancet Oncol 2008;9:61–72. [2] Niederle MB, Hackl M, Kaserer K, Niederle B. Gastroenteropancreatic neuroendocrine tumours: the current incidence and staging based on the WHO and European Neuroendocrine Tumour Society classification: an analysis based on prospectively collected parameters. Endocr Relat Cancer 2010;17:909–18. [3] Delle Fave G, Kwekkeboom DJ, Van Cutsem E, et al. ENETS Consensus Guidelines for the management of patients with gastroduodenal neoplasms. Neuroendocrinology 2012;95:74–87. [4] Rindi G, Arnold R, Bosman FT, et al. Nomenclature and classification of neuroendocrineneoplasms of the digestive system. In: Bosman FT, et al, editors. WHO Classification of Tumours of the Digestive System. Lyon: IARC press; 2010. pp13-4 [5] Delle Fave G, Capurso G, Milione M, Panzuto F. Endocrine tumours of the stomach. Best Pract Res Clin Gastroenterol 2005;19:659–73. [6] Jianu CS, Fossmark R, Viset T, et al. Gastric carcinoids after long-term use of proton pump inhibitor. Aliment Pharmacol Ther 2012;36:644–9. [7] Jianu CS, Lange OJ, Viset T, et al. Gastric neuroendocrine carcinoma after long-term use of proton pump inhibitor. Scand J Gastroenterol 2012;47:64–7. [8] Annibale B, Azzoni C, Corleto VD, et al. Atrophic body gastritis patients with enterochromaffin-like cell dysplasia are at increased risk for the development of type I gastric carcinoid. Eur J Gastroenterol Hepatol 2001;13:1449–56. [9] Annibale B, Marignani M, Azzoni C, et al. Atrophic body gastritis: distinct features associated with Helicobacter pylori infection. Helicobacter 1997;2:57–64. [10] Solcia E, Bordi C, Creutzfeldt W, et al. Histopathological classification of nonantral gastric endocrine growths in man. Digestion 1998;41:185–200. [11] Berna MJ, Annibale B, Marignani M, et al. A prospective study of gastric carcinoids and enterochromaffin-like cell
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
changes in multiple endocrine neoplasia type 1 and Zollinger-Ellison syndrome: identification of risk factors. J Clin Endocrinol Metab 2008;93:1582–91. Bordi C, Corleto VD, Azzoni C, et al. The antral mucosa as a new site for endocrine tumors in multiple endocrine neoplasia type 1 and Zollinger-Ellison syndromes. J Clin Endocrinol Metab 2001;86:2236–42. Merola E, Sbrozzi Vanni A, Panzuto F, et al.Type I. Gastric Carcinoids: A Prospective Study on Endoscopic Management and Recurrence Rate. Neuroendocrinology 2011;95: 207–13. Koop H, Klein M, Arnold R. Serum gastrin levels during long-term omeprazole treatment. Aliment Pharmacol Ther 1990;4:131–8. Ngamruengphong S, Leontiadis GI, Radhi S, Dentino A, Nugent K. Proton pump inhibitors and risk of fracture: a systematic review and meta-analysis of observational studies. Am J Gastroenterol 2011;106:1209–18. Kim YG, Graham DY, Jang BI. Proton pump inhibitor use and recurrent Clostridium difficile-associated disease: a casecontrol analysis matched by propensity score. J Clin Gastroenterol 2012;46:397–400. Schmidt M, Johansen MB, Robertson DJ, et al. Concomitant use of clopidogrel and proton pump inhibitors is not associated with major adverse cardiovascular events following coronary stent implantation. Aliment Pharmacol Ther 2012;35:165–74. Burkitt MD, Varro A, Pritchard DM. Importance of gastrin in the pathogenesis and treatment of gastric tumors. World J Gastroenterol 2009;15:1–16. Fiocca R, Mastracci L, Attwood SE, et al. Gastric exocrine and endocrine cell morphology under prolonged acid inhibition therapy: results of a 5-year follow-up in the LOTUS trial. Aliment Pharmacol Ther 2012;36: 959–71. Rindi G, Azioni C, La Rosa S, Klersy C, Parlotti D, Rappel S, et al.L cell tumor and poorly differentiated endocrine carcinoma of the stomach: prognostic evaluation by pathological analysis. Gastroenterology 1999;116: 532–41.
