Novel Biomarkers Help Predict Recurrence of Clostridioides difficile Infections


Investigators use a large PCR-only data set to confirm that PCR CT can predict recurrence-free survival, using Cox proportional hazards regression.

Novel biomarkers can help prevent recurrence of Clostridioides difficile infections (CDI), using accurate risk stratification and improving underutilization of therapies, including bezlotoxumab, fecal transplant, and fidaxomicin, according to the results of a study published in Infection and Immunity.

Investigators included a biorepository of 257 individuals who were hospitalized with 24 featured collected at diagnosis. These included 17 plasma cytokines, total/neutralizing anti-toxin B immunoglobulin G, stool toxins, and PCR cycle threshold (CT), which is a proxy for stool organism burden.

They separated the data based on individuals who developed recurrent infection, died, or survived without recurrence. Additionally, they found that lower age, PCR CT, interleukin (IL)-6, IL-8, L-17A and higher IL-16, endothelial growth factor, and CCL-5 were the best univariate class septation for recurrent infection compared with recurrence-free survival and death.

The best set of predictors for recurrent infection was selected by Bayesian model for inclusion in a final Bayesian logistic regression model, according investigators.

They used a large PCR-only data set to confirm that PCR CT can predict recurrence-free survival, using Cox proportional hazards regression. The features were endothelial growth factor with a 0.192 probability, eotaxin with a 0.066 probability, hepatocyte growth factor with a 0.058 probability, IL-4 with a 0.047 probability, IL-6 with a 0.764 probability, IL-8 with a 0.111 probability, IL-10, and PCR CT with a 0.711 probability.

The top 5 features were used to train the final logistic regression model using all 257 observations, with 32 recurrent CDIs.

Investigators found that the accuracy of the final model was 0.88. Also, they found that, among 1660 cased with PCR-only data, CT was associated with recurrence-free survival.

Additionally, they found that certain biomarkers with CDI severity were important to protect recurrence.The PCR CT and markers of type 2 immunity, including endothelial growth factor and eotaxin, were positive predictors of recurrence. Negative predictors included type 17 immune makers, including IL-6 and IL-8.

However, investigators found that no single feature was a statistically significant predictor when other features were present.

In addition to the serum biomarkers, including IL-6, endothelial growth factor, and IL-8, the PCR CT could be critical for underperforming clinical models of CDI recurrence.

For the PCR CT analysis, there were data on 1660 hospitalized cases of CDI occurring between November 2013 and April 2021 among 1412 individuals, with 250 suffering a recurrent episode within 180 days of the previous infection.

The PCR CT measurements ranged from 17.7 to 37 cycles.

According to the univariate Cox regression, the hazard ration for PCR CT was 0.95, which was 5% relatively lower risk of recurrent infection for each standardized single unit increase, according to investigators.

The AUROC was 0.60 for recurrent CDI within 90 days.

The PCR CT achieved a sensitivity of 0.70, a negative predictive value of 0.89, a specificity of 0.43, and a predictive value of 0.18, according to investigators.


Madden GR, Rigo I, Boone R, Abhyankar MM, et al. Novel Biomarkers, including tcdB PCR cycle threshold, for predicting recurrent Clostridioides difficile. Infect Immun. 2023;e0009223. doi:10.1128/iai.00092-23

Related Videos
Image credit: |
Human brain digital illustration. Electrical activity, flashes, and lightning on a blue background. | Image Credit: Siarhei -
3D illustration of heart
Model of a heart
Man experiencing heart pain
3D illustration of heart
Patient with diabetes testing blood sugar levels
© 2023 MJH Life Sciences

All rights reserved.