**3. Results**

#### *3.1. Primary Outcomes*

Participant demographics are tabulated in Table 1.


**Table 1.** Demographics by groups.

For the primary analysis, the JOHD group had significantly longer T1ρ relaxation times in the caudate, putamen, globus pallidus, and thalamus compared to the control group (Table 2), indicating significant neuronal damage in these areas in patients with JOHD. However, there were no significant group differences regarding T1ρ relaxation times in the hippocampus and cerebellum, indicating no difference in neuronal damage in these areas.


**Table 2.** Regional T1ρ relaxation times—juvenile-onset Huntington's disease (JOHD) participants.

#### *3.2. Secondary Outcomes*

Higher T1ρ relaxation times were associated with lower volumes in the caudate (t = −2.46, *p* = 0.039), putamen (t = −5.63, *p* = 0.0006), and globus pallidus (t = −2.32, *p* = 0.0491). There was a negative relationship between T1ρ relaxation times and volume of the thalamus, but the results did not reach statistical significance (t = −1.92, *p* = 0.0912). Next, we demonstrated significant positive relationships between CAG repeat length and T1ρ relaxation times in the caudate (t = 3.02, *p* = 0.018), putamen (t = 3.73, *p* = 0.006), globus pallidus (t = 7.88, *p* < 0.0001), and thalamus (t = 2.68, *p* = 0.026) (Figure 1A–D). The positive relationship indicates that the higher CAG repeat length is associated with increased neuronal damage in these brain regions.

**Figure 1.** CAG repeat length significantly predicts T1ρ relaxation times in the (**A**) caudate, (**B**) putamen, (**C**) globus pallidus, and (**D**) thalamus. Results show raw data points, the fitted regression line of the model, and 95% confidence interval. CAG: Cytosine-adenine-guanine.

While the relationship between CAG repeat length and T1ρ relaxation times in the caudate and thalamus were statistically significant, there seemed to be some outlying data that could have influenced the results. Given the small sample size of patients, we performed unplanned follow-up analyses to account for the potential influence of outliers on the results. Specifically, we repeated the analyses using robust linear mixed effects regression using the "robust" package in R. The relationship between CAG repeat length and T1ρ relaxation times in the caudate (t = 2.91, *p* = 0.021) and thalamus (t = 3.18, *p* = 0.012) remained significant after accounting for potential outliers.

Next, we assessed the relationship between T1ρ relaxation times and disease burden scores. The disease burden scores significantly predicted T1ρ relaxation times in the caudate (t = 3.97, *p* = 0.003) and thalamus (t = 3.07, *p* = 0.012), but not in the globus pallidus (t = 1.97, *p* = 0.081) or putamen (t = 1.95, *p* = 0.08). We also investigated the relationship between regional T1ρ relaxation times and motor function. We found that higher mean T1ρ relaxation times in the caudate (t = 3.55, *p* = 0.006), putamen (t = 3.26, *p* = 0.011), globus pallidus (t = 3.4, *p* = 0.008), and thalamus (t = 3.29, *p* = 0.042) were positively related to increased UHDRS scores (Figure 2A–D).

**Figure 2.** T1ρ relaxation times in the (**A**) caudate, (**B**) putamen, (**C**) globus pallidus, and (**D**) thalamus. Significantly predicted total motor scores as measured by the UHDRS. Results show raw data points, the fitted regression line of the model, and 95% confidence interval. GP: Globus pallidus; UHDRS: Unified Huntington's Disease Rating Scale.

Mean T1ρ relaxation times in the caudate (t = 3.08, *p* = 0.013) and putamen (t = 3.51, *p* = 0.007) were also directly proportional to the JOHDRS score, but not in the globus pallidus (t = 1.79, *p* = 0.131) or thalamus (t = 1.09, *p* = 0.357).

Lastly, we identified significant positive relationships between disease duration and T1ρ relaxation times in the caudate (t = 3.58, *p* = 0.008), putamen (t = 3.83, *p* = 0.01), and globus pallidus (t = 3.49, *p* = 0.007), but not in the thalamus (t = 1.09, *p* = 0.304). This suggests that T1ρ relaxation times may be able to track the disease course over time.
