Inflammatory fibroid polyp: A series of 29 cases and a systematic review of the literature

Abstract

An inflammatory fibroid polyp (IFP) of the gastrointestinal tract is a localized, benign mesenchymal lesion consisting of spindle-shaped stromal cells, eosinophilic granulocytes, and some lymphocytes and plasma cells. The discovery of a frequent mutation of the platelet-derived growth factor receptor A (PDGFRA) gene was the first hint of a gene-regulating process in IFPs. The aim of this study was to investigate the interaction of inflammatory processes and the role of mutation and expression of the PDGFRA gene in the development of IFPs for the first time. We used immunohistochemistry to analyze the composition of inflammatory cells and next generation sequencing (NGS) to provide a broad overview of gene mutations.

We report on 29 cases of IFP. The mean age, gender differences, and localization were compatible with the literature. Spindle cell histomorphology was present in 79% of cases showing a typical onion skin-like perivascular arrangement and significantly high CD34 positivity (p = 0.002, Fisher’s exact test). Eosinophilic granulocytes were present in an average density of 60 ± 49/high power field (HPF) (range: 15–200), and there was a significantly higher rate of IFPs larger than 2 cm in size (p = 0.018, Wilcoxon test). All but one cases could be analyzed by NGS. Mutations were observed in 17 cases (60.7%), including 13 (46.4%) mutations in the PDGFRA gene. Among the gastric lesions, mutations were found in exon 18 of the PDGFRA gene with amino acid exchange (Asp842Val) for eight out of 10 cases and in exon 12 in two cases. All three cases in the small intestine revealed mutation of the PDGFRA gene in exon 12. We found no PDGFRA mutation in our colonic cases. PDGFRA expression was significantly correlated with mutations of the same gene (p = 0.005, Fisher’s exact test) and especially with mutations in exon 12 of the same gene (p < 0.001, Fisher’s exact test). Interestingly, three of our cases (10.3%) without mutation or expression of the PDGFRA gene revealed an unusually high concentration of IgG-positive plasma cells (average: 140 ± 26/HPF, range: 110–160) and IgG4-positive plasma cells (average: 87 ± 21/HPF, range: 60–100). For comparison, an IgG4/IgG ratio of more than 0.4 is commonly observed in IgG4-related diseases. Our molecular results were in accordance with 113 genetically analyzed cases published to date. There was a correlation between the IFP site and mutation variants of the PDGFRA gene. IFPs were localized in the stomach in 49.1% of cases, in the small intestine in 47.3%, and in the colon in 3.6%. Exon 12 of the PDGFRA gene was mutated in 41.1% of cases and primarily occurred in the small intestine (82.6%). Exon 18 was mutated in 22.3% of cases and primarily occurred in the stomach (80.0%). The mutated codon interval 566–571 in exon 12 and codon 842 in exon 18 were compatible, as observed in a gastrointestinal stromal tumor. Conclusively, the correlation between mutation and expression of the PDGFRA gene points to different pathways in IFPs. Additionally, our data hint at a morphological but not genetic overlap between IFPs and IgG4-related pseudotumors.