Validation of a fluorescence in situ hybridization method using peptide nucleic acid probes for detection of Helicobacter pylori clarithromycin resistance in gastric biopsy specimens

Abstract

Here, we evaluated a previously established peptide nucleic acid-fluorescence in situ hybridization (PNA-FISH) method as a new diagnostic test for Helicobacter pylori clarithromycin resistance detection in paraffin-embedded gastric biopsy specimens. Both a retrospective study and a prospective cohort study were conducted to evaluate the specificity and sensitivity of a PNA-FISH method to determine H. pylori clarithromycin resistance. In the retrospective study (n = 30 patients), full agreement between PNA-FISH and PCR-sequencing was observed. Compared to the reference method (culture followed by Etest), the specificity and sensitivity of PNA-FISH were 90.9% (95% confidence interval [CI], 57.1% to 99.5%) and 84.2% (95% CI, 59.5% to 95.8%), respectively. In the prospective cohort (n = 93 patients), 21 cases were positive by culture. For the patients harboring clarithromycin-resistant H. pylori, the method showed sensitivity of 80.0% (95% CI, 29.9% to 98.9%) and specificity of 93.8% (95% CI, 67.7% to 99.7%). These values likely represent underestimations, as some of the discrepant results corresponded to patients infected by more than one strain. PNA-FISH appears to be a simple, quick, and accurate method for detecting H. pylori clarithromycin resistance in paraffin-embedded biopsy specimens. It is also the only one of the methods assessed here that allows direct and specific visualization of this microorganism within the biopsy specimens, a characteristic that allowed the observation that cells of different H. pylori strains can subsist in very close proximity in the stomach.

Fig 1

Detection of H. pylori strains in gastric biopsy histological slides by PNA-FISH. (A) Visualization of a resistant strain in the green channel. (B) Visualization of the same microscopic field in the red channel, where no H. pylori-sensitive strains are present. (C) Visualization of a susceptible strain in the red channel. (D) Visualization of the microscopic field presented in panel C in the green channel. The very faint cells that are picked up in the green channel are the result of bleed through, but the signal intensity can be easily differentiated from that of a true positive signal. (E and F) Presence of a mixed infection within the same biopsy specimen. (E) H. pylori cells that contain the point mutations associated with clarithromycin resistance and that can be seen only in the green channel (solid arrows). (F) H. pylori cells in which no point mutations are present (wild type) and that can be seen only in the red channel (dashed arrows). As can be observed in panels E and F, cells of different H. pylori strains can subsist in very close proximity in the stomach (in locations less than 5 μm apart). They also appear interspersed in the biopsy specimen and are not segregated according to species. Images were obtained with equal exposure times. (Original magnification, ×600).

Fig 2

Correspondence between the molecular results (PCR-sequencing and PNA-FISH) for biopsy specimens that were negative by culture in the prospective study. The figure key presents the percentage of cases associated with each correspondence, as well as the number of cases.