BACKGROUND: There is a critical need for effective treatment of the core symptoms of autism spectrum disorder (ASD). The purinergic antagonist suramin may improve core symptoms through restoration of normal mitochondrial function and reduction of neuro-inflammation via its known antagonism of P2X and P2Y receptors. Nonclinical studies in fragile X knockout mice and the maternal immune activation model support these hypotheses. METHODS: We conducted a 14 week, randomized, double-blind, placebo-controlled proof -of-concept study (N = 52) to test the efficacy and safety of suramin intravenous infusions in boys aged 4-15 years with moderate to severe ASD. The study had 3 treatment arms: 10 mg/kg suramin, 20 mg/kg suramin, and placebo given at baseline, week 4, and week 8. The Aberrant Behavior Checklist of Core Symptoms (ABC-Core) (subscales 2, 3, and 5) was the primary endpoint and the Clinical Global Impressions-Improvement (CGI-I) was a secondary endpoint. RESULTS: Forty-four subjects completed the study. The 10 mg/kg suramin group showed a greater, but statistically non-significant, numeric improvement (- 12.5 ± 3.18 [mean ± SE]) vs. placebo (- 8.9 ± 2.86) in ABC-Core at Week 14. The 20 mg/kg suramin group did not show improvement over placebo. In exploratory analyses, the 10 mg/kg arm showed greater ABC Core differences from placebo in younger subjects and among those with less severe symptoms. In CGI-I, the 10 mg/kg arm showed a statistically significant improvement from baseline (2.8 ± 0.30 [mean ± SE]) compared to placebo (1.7 ± 0.27) (p = 0.016). The 20 mg/kg arm had a 2.0 ± 0.28 improvement in CGI-I, which was not statistically significant compared to placebo (p = 0.65). CONCLUSION: Suramin was generally safe and well tolerated over 14 weeks; most adverse events were mild to moderate in severity. Trial Registration Registered with the South African Health Authority, registration number DOH-27-0419-6116. CLINICALTRIALS: Gov registration ID is NCT06058962, last update posted 2023-09-28.

While tremendously useful, automated techniques for tuning the precision of floating-point programs face important scalability challenges. We present Blame Analysis, a novel dynamic approach that speeds up precision tuning. Blame Analysis performs floating-point instructions using different levels of accuracy for their operands. The analysis determines the precision of all operands such that a given precision is achieved in the final result of the program. Our evaluation on ten scientific programs shows that Blame Analysis is successful in lowering operand precision. As it executes the program only once, the analysis is particularly useful when targeting reductions in execution time. In such case, the analysis needs to be combined with search-based tools such as Precimonious. Our experiments show that combining Blame Analysis with Precimonious leads to obtaining better results with significant reduction in analysis time: the optimized programs execute faster (in three cases, we observe as high as 39.9% program speedup) and the combined analysis time is 9× faster on average, and up to 38× faster than Precimonious alone.