Commentary on Zhu <i>et al</i> .: Nitazenes—Tracking a moving target from linked United States databases

Nitazenes have added complexity to recreational drug markets since 2019. Integrating complementary national databases enables clearer insight into temporal and geographic detection patterns, while generating lessons relevant to stakeholders worldwide. 2-Benzylbenzimidazoles, colloquially called nitazenes, were virtually unknown to most toxicologists before 2019. Yet, two members of this family of synthetic opioids (etonitazene and clonitazene) had been controlled internationally for almost 6 decades [1]. It was only from 2019 onward, with the emergence of isotonitazene on the recreational drug market, that a whole new group of nitazene analogues took off. In their study, Zhu and colleagues [2] assessed how nitazenes emerged in the United States (US), both temporally and geographically, in the period 2019 to 2024. To do so, they compiled two complementary databases: seizure data from the US Drug Enforcement Administration's National Forensic Laboratory Information System (NFLIS), and biospecimen toxicology data from the Center for Forensic Science Research and Education (CFSRE). The strength of combining both databases lies not only in better temporal and geographical coverage, but also in bridging seizure data and toxicology findings. Several interesting observations emerged from this first large-scale compilation of nitazene analogue detections in the United States. First, a clear rise in both the number of cases involving nitazene analogues as well as in the diversity of nitazene analogues could be observed in both datasets. From only two nitazene analogues in 2019 (primarily isotonitazene), the panel rapidly diversified, with some 20 different nitazene analogues (excluding metabolites) being detected in 2024. Second, both datasets nicely displayed the profile typically seen with new psychoactive substances, with the decline of one substance being accompanied by an increase of another substance, with different substances ‘peaking’ over time [3]. Interestingly, the 5 years covered by this study illustrate how rapidly a field may develop, with ‘first-generation’ nitazene analogues (typically differing in the alkoxy tail and/or the presence of a 5-nitro group) transitioning to ‘second-generation’ (encompassing logical variations, e.g. N-pyrrolidino metonitazene) and to more structurally distinct ‘third-generation’ analogues (e.g. methylenedioxynitazene), as we previously suggested [4]. Third, intriguingly, although one may argue that the decline of isotonitazene may have been driven by its scheduling, this observation was less apparent for its ‘successors’, metonitazene and protonitazene. Only time can tell what the impact of further scheduling in the United States [5] will be, as well as how the field will react to the generic scheduling of nitazene analogues in China as of July 2025 [6]. Will ‘ban-evading’ nitazenes emerge? Will the production shift to other countries, or will the field move toward other high-potency opioids (e.g. the ‘orphine’ class) [7, 8]? Fourth, some interesting differences could be observed between the NFLIS and CFSRE databases, with some nitazene analogues being present in seizures, but not in intoxications and vice versa. This may at least partially be related to some obvious differences between both databases, for example, CFSRE testing of biospecimens also includes metabolites. There are also some intrinsic caveats, for example, the NFLIS data may be impacted by capacity and non-standardized testing practices and does not cover all nitazene analogues (e.g. N-desethyl metonitazene). Moreover, the CFSRE has a different (smaller) geographical coverage. Although the aforementioned caveats could be conceived as a limitation, they also clearly demonstrate the added value of aligning both datasets, as the limitations of one approach may be complemented by the other. Last, some apparent discrepancies between both datasets may also point to compounds that have only very recently emerged (i.e. readily present in seizures, but not (yet) identified in intoxication cases, such as e.g. N,N-dimethylamino etonitazene), and/or to compounds that are less likely to result in fatalities (e.g. compounds with relatively lower μ-opioid receptor activation potential, such as ethyleneoxynitazene [9]). Several lessons can be learnt from the work by Zhu and colleagues [2]. A first lesson is that combining complementary databases offers synergy. Efforts like this, also beyond the borders of the United States, may facilitate a timely recognition of—and response to—emerging drug threats. Although differences between states were observed, as well as differences with substances notified by the European Union Drugs Agency, history has taught us that, certainly for the most potent substances, it is just a matter of time before seizures and intoxications spread globally [3]. A second lesson is that further temporal refinement, if possible in the future, may allow insight into whether seizures precede intoxications (as one might expect). This is relevant, because it implies that rapid communication to toxicology laboratories, together with, for example, available pharmacological data [10], may allow these laboratories to anticipate new intoxication patterns. Importantly, this information may also feed harm reduction efforts, aiming to alert people who use drugs about emerging substances with high opioid potency that may therefore pose a particularly high risk of overdose. Christophe P. Stove: Conceptualization (equal); writing—original draft (equal); writing—review and editing (equal). Marthe M. Vandeputte: Conceptualization (equal); writing—original draft (equal); writing—review and editing (equal). None. None.