The inflammatory fibroid polyp (IFP) is a benign lesion occurring in the digestive tract, mostly in the stomach and small bowel, composed of fibrovascular tissue infiltrated by inflammatory cells including eosinophils and mastocytes. Its pathogenesis has been controversial (reactive versus neoplastic). The recent finding of mutations in platelet-derived growth factor receptor α (PDGFRA) in most gastric and small intestinal IFPs supported their neoplastic etiology, moreover helping in their differential diagnosis. In the only gallbladder IFP reported so far, the diagnosis was based on morphologic and immunohistochemical grounds, which in current standards would probably be considered not fully conclusive. Conversely, the gallbladder IFP we report shows typical pathologic features supported by a PDGFRA mutation, similar to its usual gastric and small intestinal counterparts, constituting the first report of an unequivocal IFP at gallbladder level. Thus, IFPs must be considered in the differential diagnosis of gallbladder mesenchymal masses, and genetic analysis of PDGFRA is a helpful tool for this purpose.
The inflammatory fibroid polyp (IFP) is a localized lesion of the gastrointestinal tract constituted by a proliferation of spindle, stellate, or epithelioid mesenchymal cells infiltrated by a variable number of inflammatory cells, including lymphocytes, eosinophils, and mast cells, with a vascularized stroma with collagen fibers. IFPs are usually centered in the submucosa, and often bulge into the visceral lumen, assuming a polypoid shape.1,2 Inflammatory fibroid polyp was first reported by Vanek3 as “submucosal granuloma with eosinophilic infiltration.” Subsequently, IFPs were variously interpreted and/or named, using terms such as eosinophilic granuloma, gastric fibroma with eosinophilic infiltration, granuloma with eosinophils, and inflammatory pseudotumor;1 inflammatory fibroid polyp, the term now widely accepted for these lesions, was coined by Helwig and Ranier4 in 1953. The pathogenesis of IFPs has been debated for a long time, with some authors sustaining their reactive nature, and others in favor of a neoplastic origin. The occurrence of familial IFPs5 and the relatively recent demonstration of platelet-derived growth factor receptor α (PDGFRA) mutations in most IFPs1,6 strongly support their neoplastic nature. Inflammatory fibroid polyps typically arise in the gastrointestinal tract (ie, stomach, small intestine, and colon);2 only 1 case of gallbladder IFP has been reported so far,7 with a diagnosis based exclusively on relatively weak morphologic and immunohistochemical grounds, a level of evidence which, although reasonable for the knowledge about IFPs available when the article was published, would probably be considered insufficient to reliably support a diagnosis of IFP in such an unusual site as gallbladder using current standards. Here we report an unequivocal gallbladder IFP, demonstrating for the first time in this anatomic location a mutation of PDGFRA in this type of tumor.
A 72-year-old woman was admitted for symptomatic cholelithiasis; she was otherwise in good general health. The patient was also affected by asymptomatic mitral valve stenosis and had no history of neoplasm. Gallbladder stones were diagnosed 25 years before, causing no symptoms until 3 months before admission, when the patient suffered from right upper abdominal pain and biliary colic. Transabdominal ultrasonography showed a thin-walled gallbladder containing many mobile and small congregated gallstones. No other abnormal abdominal findings were identified at ultrasonography. Laparoscopic cholecystectomy was performed.
MATERIALS AND METHODS
Histology and Immunohistochemistry
The gallbladder was fixed in 4% buffered formalin, processed in the usual manner, and paraffin embedded; 4-μm sections were stained with hematoxylin-eosin. For immunohistochemistry the following antibodies were used: CD34 (1:50; Novocastra, Newcastle, United Kingdom), CD117 and S100 (rabbit polyclonal), and smooth muscle actin (SMA), calponin, and anaplastic lymphoma kinase-1 (ALK-1) (mouse monoclonal) 1:400, 1:800, 1:100, 1:50, and undiluted, respectively; DAKO, Glostrup, Denmark); DOG1 (rabbit polyclonal, 1:100; Spring Bioscience, Pleasanton, California); and PDGFRA (rabbit polyclonal, 1:400; sc-338, Santa Cruz Biotechnology, Inc, Santa Cruz, California). Antigen retrieval was performed for DOG1, S100, calponin, ALK-1, and PDGFRA (10 minutes in 0.01 M citrate buffer, pH 6, microwave at 750 W) and for SMA (10 minutes in 0.01 M Tris 0.001 M EDTA buffer, pH 8, microwave at 750 W). Specific preimmune sera or isotype-specific unrelated primary antibodies were used for the staining controls. Hydrogen peroxide, serum biotinylated immunoglobulins, and avidin-biotin complexes were used according to the manufacturer's instructions; after induction of the color reaction with freshly made diaminobenzidine solution (Dakopatts, Glostrup, Denmark), slides were counterstained with hematoxylin.
Three 10-μm slides were cut from paraffin-embedded tissues, treated twice with xylene, and then washed with ethanol. DNA was extracted by using the QIAamp tissue kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol. Pathologic areas, microdissected on slides, contained at least 70% disease-specific tissue (macroresection of the lesion) to minimize contamination of normal cells. Mutational analysis was also performed in the cystic lymph node of Lund.
PDGFRA Mutational Analysis
PDGFRA gene (exons 12, 14, and 18) was amplified by using the same primers and polymerase chain reaction (PCR) conditions described elsewhere.8 Briefly, DNA (100–200 ng) of normal (lymph node) and pathologic tissues was amplified in a mixture containing 1× PCR buffer (20 mM Tris, pH 8.3; 50 mM KCl; 1.5 mM MgCl2), deoxynucleotide triphosphates (200 mM each), primers (20 pM each), and 0.5 U Platinum Taq DNA polymerase (Invitrogen, Milan, Italy) in a final volume of 25 μl. Polymerase chain reaction conditions were as follows: an initial denaturation of 95°C for 8 minutes, followed by 35 cycles of 95°C for 40 seconds, 55°C for 40 seconds, and 72°C for 40 seconds. After visualization onto agarose gel, PCR products were treated with ExoSAP-IT (USB Corp, Cleveland, Ohio) by following the manufacturer's protocol, amplified with BigDye Terminator version 3.1 cycle sequencing kit (Applied Biosystems, Milan, Italy) by using forward and reverse primers, and sequenced with an ABI PRISM 3100-Avant Genetic Analyzer (Applied Biosystems). Water was used as a negative control.
The resected gallbladder measured 8.5 cm. In the fundus there was a single, well-delimited intramural nodule measuring 3 mm. Many small pigment gallstones were found in the lumen. The cystic lymph node of Lund was present at the gallbladder neck.
Histologic sections of the 3-mm nodule showed a tumor composed of an admixture of fibroblast-like spindle or epithelioid to polygonal mesenchymal cells and inflammatory cells, including eosinophilic granulocytes and mast cells, with no atypia and no mitotic figures, interspersed in a myxoid extracellular matrix with tiny collagen fibers, with variously represented blood vessels (Figure 1, A and B). Tumor mesenchymal cells were strongly immunopositive for CD34 (Figure 1, C) and negative for CD117 (which highlighted the infiltrating mast cells) (Figure 1, D); additionally, PDGFRA immunoreaction did not show any significant staining, and stainings with DOG1, S100, calponin, SMA, and ALK-1 were all negative (not shown). The cystic lymph node of Lund, present at the gallbladder neck, showed nonspecific reactive lymphoadenitis.
PDGFRA Mutational Analysis
By PCR direct sequencing, we found that the investigated gallbladder intramural lesion harbored a heterozygous deletion of 3 nucleotides (GAC) at codon 842 in exon 18 (del D842) of PDGFRA (Figure 2, A). The cystic lymph node of Lund, present at the gallbladder neck, was also analyzed, revealing a wild-type PDGFRA mutation (Figure 2, B).
We describe an unequivocal IFP found in the gallbladder of a 72-year-old woman, which was resected due to symptomatic cholelithiasis.
Inflammatory fibroid polyps are mesenchymal lesions occurring throughout the digestive tract, especially in the stomach and small intestine;2 to our knowledge, only 1 case has been reported so far in the gallbladder.7 Notably, the diagnosis of this single case, although reasonable for the knowledge on IFPs available at the time the article was published, probably cannot be considered fully conclusive nowadays, being based only on morphology and on immunostaining for CD34 that was only focal or restricted around large blood vessels. In fact, the morphologic gamut of IFPs, with different possible assortments of their components (ie, mesenchymal cells, various types of inflammatory cells, vessels, and collagen fibers), overlaps with that of other inflammatory and/or spindle cell tumors (as the controversial story of the interpretation and denomination of this entity indirectly proves). Moreover, CD34, besides having been detected only focally in the abovementioned report, is itself a relatively nonspecific immunohistochemical marker, especially when dealing with mesenchymal lesions.9 These caveats deserve particular attention when diagnosing an IFP in such an unusual site as the gallbladder.
For these reasons—and because the only possible diagnostic immunohistochemical tool other than CD34 for diagnosing an IFP, namely, PDGFRA (which, in the case we report, resulted in a focal and faint staining of uncertain specificity), has been often deemed unreliable,9,10 and has been reported, moreover, not to stain a fraction as high as 40% of IFPs11—we investigated the status of the PDGFRA gene. In fact, mutations in this gene were found in most small intestinal and gastric IFPs,1,6 thus gaining diagnostic value for these tumors.
The gallbladder lesion we report revealed the presence of a deletion of 3 nucleotides (GAC) at codon 842 in exon 18 (del D842) of PDGFRA, an activating mutation previously described in IFP.12 The low rate of mutated versus wild-type DNA likely depends on the large normal cell component present in IFP,1 which “dilutes” the genotypically altered fibroblast-like cells.6
With regard to the differential diagnosis of the investigated mass, the finding of a PDGFRA mutation by itself excludes all but one mesenchymal lesion, which can exceptionally occur in the gallbladder: gastrointestinal stromal tumor (GIST).13 However, the following observations are noteworthy: (1) the morphology of the present lesion, with its heavy leukocytic infiltrate, is rather unusual for a GIST; (2) PDGFRA-driven GISTs are typically composed, at least in part, of cells with a fully developed epithelioid morphology,9 while only a relatively small part of the mesenchymal cells of the tumor we describe show a vaguely epithelioid aspect; and finally, (3) the investigated mass does not stain with either CD117 or DOG1, both of which show positivity in most GISTs (including most PDGFRA-driven cases).8,9,14
The absence of PDGFRA mutation in the cystic lymph node of Lund in the studied specimen ruled out the possibility of a germline mutation.
Recently, “small bowel” and “gastric” phenotypes of IFP have been proposed depending on the PDGFRA mutational pattern, with exon 12 mutations mostly found in the small bowel and exon 18 mutations prevailing in the stomach.11 Accordingly, the reported gallbladder tumor would belong to the “gastric” type of IFPs; however, considering the specific PDGFRA mutation instead of the exon harboring it, it is noteworthy that the only IFP with a PDGFRA del D842 reported before the one described in this study was actually found in the small bowel, namely in the duodenum.12 Thus, in a perspective of possible site-specific PDGFRA mutations of IFPs, it seems conceivable that the present gallbladder case is more akin to a comparatively unusual intestinal-type of IFP bearing the PDGFRA exon 18 mutation del D842 than to the “gastric” IFP family sharing a PDGFRA exon 18 mutation in a broad sense.
PDGFRA-expressing mesenchymal cells have been described juxtaposed to the epithelial cell layer in the gallbladder of mouse embryos, and their diminished presence, detected in mice haploinsufficient for forkhead box f1 gene, seems to contribute to the defective formation of this viscus.15 This finding parallels the observation of clusters of PDGFRA-positive mesenchymal cells found along the basement membranes of intestinal villi of mouse embryos, and the analogous scattered cells found in mature mouse gut.16 In both cases, these PDGFRA-positive cells could be the normal counterpart of IFPs (and PDGFRA-driven GISTs), as happens with Cajal cells for KIT-driven GISTs. Other proposed origins of IFPs include dendritic cells17 and CD34+ perivascular cells;18 however, the ubiquity throughout the human body of such cellular types less easily explains the restriction of IFPs to the gastrointestinal tract than does the presence of the abovementioned PDGFRA-positive mesenchymal cells.
In conclusion, the intramural gallbladder lesion we investigated showed morphologic, immunohistochemical, and genotypic features fully consistent with an IFP, and constitutes the first unequivocal reported case of IFP at gallbladder level. Inflammatory fibroid polyp must be therefore considered in the differential diagnosis of mesenchymal lesions of the gallbladder, and PDGFRA analysis is a helpful tool for this purpose.
The authors have no relevant financial interest in the products or companies described in this article.