Study selection and patient population
A total of 6782 articles were retrieved from the search after removing duplicates from the various databases. These articles underwent initial screening based on title and abstract, resulting in 98 articles that underwent a full-text review. Following a rigorous selection process, 41 studies with 4667 patients and a median follow-up of 18.8 months were included in the analysis [23-63]. Out of the 41 articles, 22 were randomized controlled trials (RCTs), and 19 were cohort studies. Seventeen articles reported the efficacy and safety of LLT therapy compared to control (25, 28-33, 37, 39, 41, 46, 49, 53, 55-57, 63), while the remaining 24 assessed LLT pre- and post-treatment (23, 24, 26, 27, 34-36, 38, 40, 42-45, 47, 48, 50-52, 54, 58-62). The PRISMA flow diagram is depicted in Figure S1, and the key characteristics of the included studies are presented in Table 1. The mean age of patients was 12.08 ± 2.4 and 46% were females. Out of 41 studies, 7 (16.7%) used low intensity LLT, 9 (21.4%) low to moderate, 10 (23.8%) moderate, 7 (16.7%) moderate to high, and 7 (16.7%) high intensity or combination therapy (Table 1).
Efficacy of LLT on lipidic profile and lipoproteins
The pooled analysis showed that the group commenced on LLT had higher baseline TC with a WMD of 0.25 mmol/L (9.7 mg/dl) (95%CI: 0.02 to 0.49, p = 0.04) and LDL-C of 0.23 mmol/L (8.9 mg/dl) (95% CI: 0.05 to 0.41, p = 0.01) compared to control (Figure S2). TG and HDL did not differ between the groups (p>0.05 for all, Figure S3), as did Lp (a), apoB, and apoA (p>0.05 for all, Figure S4). At median follow-up of 18.8 months, the group on LLT had significantly higher mean TC reduction -1.75 mmol/L (67.7 mg/dL) (95%CI: -1.95 to -1.55; p<0.001), LDL-C reduction -1.84 mmol/L (71.2 mg/dL) (-2.13 to -1.54; p<0.001), increased mean HDL-C by 0.08 mmol/L (3.1 mg/dL) (0.01 to 0.15; p=0.03) and reduced TG -0.11 mmol/L (9.74 mg/dL) (-0.16 to -0.07; p<0.001), compared to control (Figure S5 & S6). In addition, the LLT significantly reduced apoB (-0.32 g/L; p<0.001), while apoA and Lp(a) were not affected significantly compared to control (p>0.05, for both). The same results were achieved the inverse analysis we applied (Figure S7, S8). A summary of efficacity of LLT is shown on Figure 1.
In the subgroup analysis according to different types of LLT, using an inverse analysis model, we observed s significantly higher mean reduction of TC with statins combined with ezetimibe (WMD -2.73 [-105.6 mg/dl], 95%CI, -3.12 to -2.34 mmol/L)) followed by PCSK9is (-2.27 [-87.8 mg/dl], -2.39 to -2.15 mmol/L), statins in monotherapy (-2.21 [-85.5 mg/dl], -2.45 to -1.98 mmol/L), and only ezetimibe (-1.50 [-58 mg/dl], -1.68 to -1.32 mmol/L; test for overall effect: p<0.001; Figure S9). According to the above order of treatment, a mean reduction of LDL-C was similar: statins plus ezetimibe (WMD -2.48 [-95.9 mg/dl], 95%CI, -3.13 to -1.83 mmol/L), statin monotherapy (-2.16 [-83.5 mg/dl], -2.33 to -1.99 mmol/L], PCSK9is (WMD -2.03 [-78.5 mg/dl], -2.39 to -2.15 mmol/L] and ezetimibe monotherapy (-1.50 [-58 mg/dl], -1.71 to -1.29 mmol/L; test for overall effect: p<0.001, Figure 2). LLT also significantly affected TG, from -0.07 mmol/L (-6.2 mg/dl) for ezetimibe, through -0.15 mmol/L (-13.3 mg/dl) for statin in monotherapy, to -0.35 mmol/L (-31 mg/dl) for statins and ezetimibe in combination; no significant effect was observed for PCSK9 inhibitors (Figure S10). The overall test for subgroup differences was not statistically significant for HDL-C (p=0.76, respectively; with only statins that showed significant increase of HDL-C by 0.05 mmol/L [2 mg/dl], Figure S11).
LDL-C goal achievement using different types of LLT
At median follow-up of 18.8 months the pooled LDL-C was reduced for 33.44% (-2.14 [--2.30, - 1.98], p<0.001) and failed to reach the goal (LDL-C <3.5 mmol/L) for all investigated heFH children by 12.6% (95%CI, 12.4 – 12.9%) (Table 1). Among the 3522 children treated with LLT (1145 were on the dietary interventions), 1362 (38.7%) achieved the LDL-C goal, while 843 (23.9%) fell short by up to 10%, 377 (10.7%) experienced moderate failure (additional 10-20% overall LDL-C reduction was required), and 940 (26.7%) failed by more than 20% to reach the LDL-C target (Figure S12A).
When comparing different regions, only Sweden and Greece achieved the LDL-C goal, followed by the Netherlands, Norway, Poland, USA, UK, France, Spain, Belgium, and Austria (with the following additional required LDL-C reduction to be on the goal: 2.2%, 3.4%, 3.5%, 8.9%, 10.2%, 11.2%, 11.2%, 15%, 19.4%, respectively). Australia, Canada, Italy, Finland, and Japan experienced more than a 20% mean LDL-C reduction failure (Figure 3). Similarly, when comparing different types of LLT, the heFH children that were administered moderate to high dose of statins, high dose of statins, high-dose statins combined with ezetimibe, and PCSK9 inhibitors most often achieved the LDL-C goal while groups using only statins (low, moderate, low to moderate and moderate) and ezetimibe as monotherapy did not reach the treatment goal (Figure S12B).
Safety and tolerability of LLT in children
All parameters related to endocrine function, including cortisol, DHEA-S, FSH, LH, estradiol in girls, and testosterone in boys, were unaffected by LLT therapy (p>0.05 for all; Figure S13). Similarly, demographic indicators such as height, weight, body mass index (BMI), and surface area of the children increased comparably in both the LLT and control groups (p>0.05 for all; Figure S14). At the end of the follow-up period, no significant differences were observed in changes from baseline for ALT, AST, and CK (p>0.05 for all; Figure S15). A summary of LLT safety profile for demographic, endocrine, and liver/muscle enzyme parameters is provided in Figure 4.
The prevalence of treatment-emergent adverse events (TEAEs) was 4.6%, with a discontinuation rate of only 0.8%. Apart from headaches (10.2%), the prevalence of other adverse events was below 8.0%, including gastrointestinal effects, myalgia, sleep disorders, influenza-like disease, skin reactions, temporary increases in ALT, AST, CK, as well as ALT >3 x ULN, AST >3 x ULN, and CK >3 x ULN (8.41%, 3.18%, 2.54%, 5.62%, 3.71%, 1.89%, 1.50%, 0.52%, 0.83%, 0.51%, 0.67%, respectively; Figure 5A). Finally, the adverse events were not reported significantly higher when compared to the control group (Figure 5B, Figures S16-S19).
Risk of bias assessment
The assessment of risk of bias in the included studies using RoB2 for RCTs and NOS for cohort studies showed that most studies had moderate to high quality level in defining objectives and the main outcomes (Tables S4, S5).