Thiopurine medications (azathioprine & mercaptopurine) are widely used to treat pediatric patients with inflammatory bowel disease (IBD). 6-thioguanine (6-TG), a thiopurine metabolite, can be measured to guide dosing for optimal efficacy and to avoid serious toxicity. However, the 6-TG assay is expensive and turn-around time is slow. Thiopurines cause leukopenia and red blood cell macrocytosis, 2 markers that are used for informal monitoring, but have not been formally validated. Mean corpuscular volume (MCV) to white blood cell (WBC) ratio may serve as an inexpensive surrogate for 6-TG. The goal of our study was to investigate MCV/WBC ratio as a predictor of clinical response in pediatric IBD.Methods:
We performed an IRB approved retrospective cohort study of pediatric IBD patients (2–24 years). We evaluated laboratory results and Physician Global Assessment (PGA) while on thiopurine medications. PGA scores were prospectively assessed according to standardized guidelines defining disease activity as quiescent, mild, or moderate/severe. Bivariate and ordinal logistic regression analyses were conducted to determine associations between disease activity and 6-TG or MCV/WBC ratio when available. The primary outcome was the association between MCV/WBC ratio and PGA among patients with PGA scores within 60 days of lab collection. Association with 6-TG was a secondary outcome. Analyses were limited to one sample per patient. We planned a pre-specified post hoc analysis of patients on thiopurine monotherapy (no other concomitant therapies).Results:
We identified 101 patients with quiescent (73%; n = 74), mild (18%; n = 18) and moderate/severe (9%; n = 9) disease. 65% (n = 66) were on thiopurine monotherapy. 6-TG was assessed in 42% (n = 42) of patients, 8 of whom (19%) had moderate/severe disease. Only one patient with moderate/severe disease did not have 6-TG measured (2%; P = 0.01). MCV/WBC ratio was associated with PGA (odds ratio 0.90, 95% confidence interval [CI] = 0.81 to 0.996, P = 0.042; n = 99). Higher MCV/WBC ratio was seen with better-controlled disease. MCV/WBC ratio predicted quiescent disease with an area under receiver operator characteristic (ROC) curve of 0.63 (95% CI, 0.49–0.75). Higher MCV/WBC ratio was weakly associated with lower erythrocyte sedimentation rate (r = −0.18, P = 0.07; n = 98), but not associated with C-reactive protein (r = −0.17, P = 0.22; n = 57). 6-TG was weakly associated with PGA (odds ratio 1.01, 95% CI, 0.999–1.01, P = 0.088; n = 42); higher 6 TG was seen with more active disease. However, 6-TG poorly predicted quiescent disease with an area under ROC curve of 0.39 (95% CI, 0.20–0.58). 6-TG was also weakly associated with MCV/WBC ratio (r = 0.26, P = 0.099; n = 42). Among patients receiving thiopurine monotherapy, increased MCV/WBC ratio better predicted quiescent disease with area under ROC 0.64 (95% CI, 0.49–0.79; n = 65), and also correlated better with higher 6-TG levels (r = 0.32, P = 0.051; n = 36).Conclusions:
In a retrospective study, we found higher MCV/WBC ratio is associated with improved clinical disease activity in pediatric IBD patients on thiopurine therapy, particularly among those on monotherapy. MCV/WBC ratio outperformed 6-TG in predicting disease activity, though 6-TG was disproportionately measured more commonly in sicker patients. MCV/WBC ratio appears to provide an easy, low-cost alternative to thiopurine metabolite monitoring. Validation in a larger prospective study is warranted.