Impact of Chemotherapy on Urinary Biomarkers and Non-Invasive Urodynamics in Children

Primary outcomes

1. Proportion of non-normal niUDS patterns between groups

   Time frame: Immediately after niUDS done at the first visit

   a. Patterns will be scored by blinded reviewers b. Will calculate an Inter-rater Kappa correlation co-efficient i. Utilize consensus review for discrepancies between reviewers and between the patient's 1st/2nd niUDS for the final assignment ii. Will calculate an intra-test reliability (compare correlation between 1st and 2nd patter) c. Chi-square test to compare proportions of non-normal vs. normal pattern type i. Group 1a vs. 2a ii. Group 1b vs. 2b d. Sample Size Calculation: Based on published data of niUDS patterns in children with normal bladder function, we expect 15% of those children in Group 2 to have non-normal niUDS patterns. We hypothesize that 40% of children in Group 1 to have non-normal niUDS patterns. 38 subjects in each gender sub-group provide 80% power to detect this 25% difference with an one-sided chi-square test with an alpha of 0.05.

2. Proportion of Elevated Post-Void Residual (PVR) between groups

   Time frame: 13 months

   a. Will use an "average PVR" between the 1st/2nd niUDS for the analysis b. Chi-square test to compare proportions of elevated PVR i. Group 1a vs. 2a ii. Group 1b vs. 2b c. Sample Size Calculation: Based on published data of PVR in children with normal bladder function, we expect 1.5% of those children in Group 2 to have elevated PVR. We hypothesize that 25% of children in Group 1 to have an elevated PVR. 21 subjects in each gender sub-group provide 80% power to detect this 23.5% difference with an one-sided chi-square test with an alpha of 0.05.

3. Compare Peak Flow Rates (Qmax) between groups

   Time frame: 13 momths

   a. Will use the average of the two Qmax between the 1st/2nd flow rate for the analysis b. Two group t-test for comparing continuous variables (Qmax) i. Group 1a vs. 2a ii. Group 1b vs. 2b c. Sample Size Calculation: 38 subjects in Group 1a and 2a provide 80% power to detect an effect size of 0.7 (mean difference divided by the common standard deviation) using a 2-sided two-group test with an alpha of 0.05. if the outcome is not normally distributed, transformation will be carried out so the distribution will be approximately normal. This power calculation also applies to the comparison between Group 1b vs 2b

4. Compare Average Flow Rates (Qavg) for age/gender

   Time frame: 13 months

   a. Will use the average of the two Qavg between the 1st/2nd flow rate for the analysis b. Two group ttest for comparing continuous variables (Qavg) i. Group 1a vs. 2a ii. Group 1b vs. 2b c. See the power calculation in 3c.

5. Compare Urinary Biomarker Levels between groups

   Time frame: 13 months

   • Compare urinary biomarker levels between groups i. ATP, BDNF, NGF ii. Two-group t-test for comparing continuous variables

   • Group 1 vs. 2
   • Sub-analysis

   • Group 1a vs. 2a
   • Group 1b vs. 2b The sample size calculation in 1d can apply to each biomarker. Based on published data in other populations with BD, for NGF, ATP and BDNF, we hypothesize to detect a difference of these urinary biomarkers 2x higher in Group 2 (Elevated DVSS scores) vs. Group 1 (Normal DVSS scores).

   A mixed effects model will be used to estimate and compare a biomarker expression between the two groups by incorporating the individual measurements on each subject into the model, where gender and group assignment will be the predictor. Interaction term between group and gender will be considered to study the effect modification by gender on a biomarker if possible.