Diagnostic value of immunohistochemistry and polymerase chain reaction for microsatellite instability in colorectal cancer
Objective To compare the consistency and reliability of immunohistochemistry(IHC)and polymerase chain re-action(PCR)methods in diagnosing microsatellite instability(MSI)in patients with colorectal cancer.Methods A retro-spective analysis was conducted on clinical data and pathological specimens of 135 patients with colorectal cancer who un-derwent metastatic resection or biopsy in the Department of General Surgery and Gastroenterology at Hunan Aerospace Hospital from January 17,2019 to December 22,2021.Tumor tissue and normal colon mucosal tissue at least 5 cm away from the tumor boundary were collected from each patient,and fixed,embed,and sliced.IHC and PCR were used to per-form MSI detection on tissue samples.Results Among the 135 patients,125 had normal mismatch repair function(pM-MR)and 10 had missing mismatch repair function(dMMR).Compared with pMMR patients,dMMR patients had a lar-ger tumor volume,were more likely to occur in the right half of the colon,and were more prone to mucinous adenocarci-noma or poorly differentiated cancer.Among dMMR patients,4 cases showed PMS2 deficiency and 3 cases showed patch-y PMS2 expression.PCR detection revealed that 8 patients belonged to the highly unstable microsatellite group(MSI-H),1 patient belonged to the low instability microsatellite group(MSI-L),and 1 patient belonged to the stable microsat-ellite group(MSS).Among patients with PMS2 deficiency or weak expression,62.5％(5/8)were MSI-H.The Kappa value of the consistency analysis between IHC and PCR detection results was 0.943,with P＜0.001.Conclusions IHC and PCR have high consistency and reliability in diagnosing MSI status in colorectal cancer patients.However,when IHC testing is negative or weakly positive,it is recommended to perform PCR testing to avoid missed diagnosis.The comprehensive application of these two methods can help improve the accuracy of MSI detection in colorectal cancer pa-tients.
NETL
NSTL
万方数据
