Fentanyl, a widely used synthetic opioid, can be fatal even at low exposure, and in recent years the illegal abuse of fentanyl and its analogues has become a serious global concern. The scientists, led by a team from the Shanghai Doping Analysis Institute, Shanghai University of Sports and the Faculty of Environmental Science and Engineering, University of Fudan, report on the development of a machine learning platform called Fentanyl Hunter, which can screen opioid metabolites in biological and environmental samples.
In tests, the platform identified 27 unknown metabolites of fentanyl compounds. in vitro Two samples: human hepatocytes and patient urine samples. The screening platform also discovered biomarkers for opioids and their derivatives in over 250 human and environmental samples from eight countries.
Dr. Changzi Si, the lead author of Shanghai University of Sports, and colleagues, reported on their development Advances in scienceIn a paper entitled “Machine Learning and Multilayer Molecular Network Assisted Fentanyl Compound Screening Hunt,” he wrote, “This study introduces a rapid, accurate and comprehensive screening platform for fentanyl and its transformed products.
Fentanyl and many of its derivatives have permeated communities around the world, and new fentanyl-type compounds continue to emerge, avoiding detection. The authors cited figures showing that fentanyl abuse accounted for around 75,000 deaths in the United States in 2023, accounting for nearly 70% of all drug overdose deaths. “Comprehensive monitoring of fentanyl metabolites is essential to assess substance abuse, toxicity and metabolism, with important applications that prevent overdose and provide key forensic and toxicological evidence,” the investigator further noted.
She and her colleagues have also synthesized more than 1,400 fentanyl analogs that have similar or even greater potency to fentanyl. “They are gradually becoming more and more prevalent in the drug market.” These previously unknown fentanyl are designed to avoid analytical detection, and have challenges with global regulation and control. The number of previously unknown fentanyl analogs is also constantly increasing,” the team admitted. “However, most have low detection rates. As structural modifications become more complicated, rapid screening for fentanyl is becoming more and more difficult.”
Over the past decade, high-resolution mass spectrometry (HRMS) and non-target analysis (NTA) have demonstrated the ability to identify fentanyl with high-throughput and confident identification, the team says. However, they said, “Current NTA platforms that rely on MS and spectral libraries are insufficient to detect unique fentanyl compounds and their metabolites.”
Changzhi shi and colleagues have developed a fentanyl hunter to identify new urgent fentanyl analogs via metabolites and address the detection challenges. The screening platform explained that using machine learning classifiers and multilayer molecular networks, selecting and annotating fentanyl compounds using mass spectrometry (MS). “It includes a machine learning model for screening fentanyl compounds (Fentanyl_finder) and a multilayer molecular network assisted structure annotation tool (Fentanyl_ID). Machine learning classification was trained on a 772 fentanyl spectrum and a multilayer network.
![Figure 1. Construction of fentanyl hunter methods and construction of applications for comprehensive profiling and annotation of the fentanyl family. The workflow primarily involves the training phase of the MS functional classifier for (i) fentanyl. (ii) Import sample MS data for fentanyl feature filters, (iii) fentanyl annotation, (iv) identify the structure of adjacent fentanyl assisted by multilayer networks, and (V) identify the assignment of trust level. [Shi et al., Sci. Adv. 11, eadw2799]](https://www.genengnews.com/wp-content/uploads/2025/09/low-res-10.jpeg)
The team screened fentanyl metabolites using fentanyl hunters in vitro and in vivo. They looked into it first in vitro Biotransformation of four derivatives (fentanyl, remifentanil, sufentanil, and alfentanil) in fragments of human liver cells. In the fentanyl hunter, eight already recorded metabolites and 27 previously unknown metabolites were discovered. In urine samples from fentanyl users, the platform also successfully detected metabolites, including two previously not described metabolites. in vitro test.
Shi et al. We then looked at the wastewater from the treatment plants to fentanyl hunters to confirm the control of norfentanyl as the main metabolite of the wastewater. The platform also discovered fentanyl, sufentanyl, norfentanyl, or remifentanil in sewage sludge, surface water, seawater, and 250 samples from eight countries: France, Luxembourg, the United States, China, Israel, Indonesia, Australia, and Argentina.
Their findings suggest that “…we need to validate our methods and carefully examine and carefully reassess the risks of fentanyl abuse and environmental pollution.” Furthermore, “… highlights the urgent need for strengthened regulations in the Fentanyl family, particularly with regard to variants and metabolites… The platform may serve as a robust tool for the detection and regulation of fentanyl variants and metabolites in public health, forensic medicine, environmental surveillance and law enforcement applications.”
