Corticosteroid reduction by addition of cetirizine and montelukast in biopsy-proven minimal-change nephrotic syndrome concomitant with allergic disorders

Recent reports suggest helper T-cell abnormalities in minimal-change nephrotic syndrome (MCNS), which often complicate allergic disorders that show a similar helper T-cell profile with Th2/Th17 predominance. However, the effect of anti-allergy therapy on MCNS remains unknown. This retrospective study included 51 patients with biopsy-proven MCNS recruited between November 2012 and October 2015, with follow-up through November 2017. We analyzed relapse and temporal daily corticosteroid dose with and without co-administration of histamine H1 receptor antagonist, cetirizine, and cysteinyl-leukotriene receptor antagonist, montelukast, as well as between baseline and after follow-up. Thirteen patients were treated with cetirizine and montelukast in addition to conventional therapy, whereas 38 patients were treated by conventional therapy only, consisting of corticosteroids and immunosuppressants. To adjust for baseline clinical characteristics, a 1:1 propensity score–matched model was applied. The clinical characteristics of the two groups after matching were similar at baseline. The treatment group showed a significant reduction in the lowest daily dose of oral prednisolone throughout the entire treatment course after the study compared to that of baseline (p < 0.025), which was not observed in the control group (p = 0.37), and showed significantly higher percentage of patients establishing corticosteroid-free state for the first time throughout the entire treatment course by addition of cetirizine and montelukast compared to the control group (p < 0.025). The study shows, for the first time, the steroid sparing effect of cetirizine and montelukast in addition to conventional treatment in MCNS patients with concomitant allergies.


Treatment course
To maintain disease control, oral corticosteroids, mainly prednisolone, were used, and very few patients received methylprednisolone or betamethasone. All patients were followed up in our outpatient clinic at 1-to 3-month intervals. In most patients, the daily dose of oral prednisolone was tapered by 2.5 to 5 mg at every visit until reaching ~20 mg/day and by 0.5 to 2.5 mg at every visit until reaching ~10 mg/day. Subsequently, prednisolone was tapered more slowly and cautiously to reach the lowest dose possible, depending on patient conditions. Some patients had been prescribed with immunosuppressants, such as cyclosporin A, bredinin, or tacrolimus, while using corticosteroids for further disease control based on the decision of their primary nephrologist. No immunosuppressants were added during the observation period.

Data collection
We collected the data by reviewing medical records retrospectively. Collected data included onset age, sex, status at time of presentation to our hospital (remission, relapse, or onset), laboratory data (serum total protein, albumin, urea nitrogen, Cre, IgG, IgA, IgM, IgE, eosinophils, urinary protein) on disease onset, disease duration at inclusion from onset, number of past relapses, follow-up period after inclusion, concomitant allergic disorders (rhinitis, atopic dermatitis, sinusitis, drug allergy, food allergy, asthma, urticaria), other immunosuppressants (cyclosporin A, bredinin, or tacrolimus), and concomitant hypertension, dyslipidemia, diabetes mellitus, and hyperuricemia that required 5 medication, infection, irregular corticosteroid reduction, or new onset malignancy. Relapse was considered as proteinuria level 1.0 g/g Cre (or g/day) that required an increased corticosteroid dose for disease control. Four milligrams of methylprednisolone and 0.8 mg of betamethasone were calculated to be equivalent to 5 mg of prednisolone according to a commonly used corticosteroid conversion rate.

Statistical analyses
Baseline patient characteristics were presented as number (percentage) for categorical variables and median (interquartile range) for continuous variables. Categorical and continuous variables were compared using Fisher's exact test and Mann-Whitney U test, respectively. Wilcoxon signed-rank test was used for comparison before and after treatment. Cox regression analysis and logistic regression analysis was used for possible risk factors of relapse. Univariate or multivariate losgistic regression analysis was used for factors associated with the presence of allergic disorders.
Relapse free survival between patients with and without anti-allergic treatment was assessed with Kaplan-Meier and log-rank analyses. To account for the confoundment arising from different clinical characteristics of patients treated with (vs. without) cetirizine and montelukast, we performed a propensity score-matched analysis as our primary analysis. Propensity scores for the likelihood of presence versus absence of cetirizine and montelukast treatment were calculated by logistic regression using ten variables at baseline (i.e., the date of patient inclusion) ; age, sex, eGFR, urinary protein, 6 number of past relapses per year, follow-up length, other immunosuppressants, concomitant allergic disorders, minimum prednisolone dose, and prednisolone dose at baseline. We then matched patients with the anti-allergic treatment with comparable patients without the treatment using a 1:1 nearestneighbor matching without replacement. Differences between variables were examined by calculating standardized differences, and values <0.3 were considered acceptable for the matching. We also conducted propensity score method using two variables, serum IgE and MCNS aggravating factors (infection, irregular corticosteroid reduction, or new onset malignancy), to focus on possible clinically relevant MCNS relapse. For co-primary outcomes, we splitted alpha into two in this study. The reported p values are two sided and reported as significant at <0.025 for the two primary outcomes.
Other outcomes were considered significant at p value of <0.05. The patients were followed-up until November 2017. All statistical analyses were performed with EZR version 1.35 (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). 1

Clinical characteristics
The characteristics of the two groups (control and treatment groups) are summarized in Table   S1 upon allocation (i.e., the baseline). There were no differences in age, sex, and laboratory data including IgE and eosinophil levels at disease onset. There were no differences in eGFR, urinary protein, disease duration, age, or use of other immunosuppressants within the study duration between the two groups. The numbers of past relapses and past relapses per year were significantly higher in the treatment group. The follow-up period was significantly higher in the control group because patients in the treatment group tended to satisfy the inclusion criterion (the prescription of cetirizine and montelukast) later in the inclusion window period. The prevalence of allergic disorders was significantly higher in the treatment group. Among the allergic diseases, rhinitis, atopic dermatitis, and sinusitis were significantly more prevalent in the treatment group. For medication-requiring concomitant diseases, dyslipidemia was significantly more prevalent in the control group, but no difference was seen for hypertension, diabetes mellitus, or hyperuricemia. The minimum prednisolone dose or prescribed prednisolone dose at baseline was similar between the two groups.

Supplementary Description
Possible molecular mechanism of contribution of anti-allergy treatment in MCNS disease stability.
IL-13, categorized as a Th2 cytokine, is reported to have a role in MCNS pathophysiology. Xue et al. 2,3 showed that LTE4 induces IL-13 and other Th2 proinflammatory cytokines in human Th2 cells. Yap human podocytes with IL-13 and showed loss of cell processes or projections as well as downregulation of slit diaphragm proteins CD2AP, synaptopodin, or ZO-1, which were significantly 9 reversed by montelukast treatment. This suggests that montelukast has a direct effect in restoring podocyte dysregulation by IL-13. Therefore, montelukast may have a dual role in protecting patients from MCNS by inhibiting T cells from producing Th2 cytokines such as IL-13 and protecting podocytes from dysregulation caused by IL-13.
No study thus far has reported on the effects of histamine H1 receptor blockade for MCNS patients. Although histamine plays a pivotal role in allergies and its action on T lymphocytes is not elucidated enough because of highly complicated regulations by different cell types, 9,10 its role on Th2 cytokine IL-13 has been described. Elliot et al. 11 showed that Th2 cells treated with histamine increased IL-13 production through H1 and H2 receptors. According to Okamoto et al., 12 H1 receptor antagonist blocks Th2 cytokine production. H1 receptor antagonist cetirizine may have therapeutic effects in MCNS patients due to its immunomodulating properties, as stated above. Values are presented as median ± standard deviation.