A comprehensive screening method using UHPLC-HRMS/MS to detect 90 NPSs in OF has been successfully validated, allowing its application to 21 pretested real samples. The method was designed to detect a broad panel of NPSs, including analogs of the most recent emerging classes such as “orphines” and last generation SCs. In particular, the non-fentanyl synthetic opioids have been replacing the fentanyl analogues on the illicit market since 2020, with a constant increase in popularity among users, which is currently alerting the international agencies and governments to a new wave of opioid crises [22,23]. In this regard, the present method is the first multi-class screening method allowing the detection of brorphine in OF. α-PHiP [24] and MDPHP [25] were also included due to their increasing prevalence on the Italian illicit market and their association with severe psychostimulant effects and intoxications, as reported in the literature.

OF was selected as the target matrix due to its non-invasive collection, rapid sampling, and ability to provide reliable information on recent drug intake. Its use is also supported by Italian regulations, which have officially adopted OF for roadside toxicological testing [16]. Moreover, OF analyses have already proved to be suitable for drug detection in clinical contexts or epidemiological studies on NPS [12,15]. A key advantage of the proposed method is its ability to detect multiple chemical classes simultaneously with a simple “dilute and shoot” approach [18]. This reduces sample handling, saves time, and allows high-throughput screening in clinical and forensic laboratories. Compared with more labor-intensive extraction techniques reported in the literature, such as microextraction on packed sorbent (MEPS) [26], liquid–liquid extraction [27], or parallel artificial liquid membrane extraction (PALME) [28], this approach provides several practical benefits. These techniques usually involve several preparation steps, longer processing times, and sometimes higher solvent consumption or specific optimization for certain compounds. In contrast, the dilute and shoot approach reduces sample handling, shortens preparation time, lowers the risk of analyte loss, and requires smaller volumes of OF [27]. It also allows high-throughput analysis while maintaining good sensitivity and reproducibility. The dilute and shoot approach for NPS detection in OF was applied by Malaca et al. in a quantitative LC-MS/MS assay to detect 13 classic drugs of abuse and NPSs. Despite the limited number of substances, the method showed satisfactory validation parameters [18]. An expanded panel of NPSs was included in a validated assay to detect 77 NPSs and 24 classic drugs of abuse in OF through LC-MS/MS, which used the dilute and shoot approach with M3^® solvent (Comedical, Trento, Italy) [17]. In this case, the M3^® buffer stabilized the analytes, increasing the cost of analysis. Furthermore, the absence of interferences from the matrix proved that the approach could sufficiently prepare the samples for the instrumental analysis. The use of HRMS combined with chromatographic separation provides high analytical specificity. HRMS allows the discrimination of structurally related compounds, including positional isomers and isobaric species, which are common among NPSs [17]. We obtained the separation of structural isomers, such as α-PHP and α-PHiP, positional isomers such as 2-CMC and 3-CMC, cis 3-methyl norfentanyl and trans 3-methyl norfentanyl as shown in Supplementary Materials. Accurate mass measurements, isotopic pattern analysis, and informative MS/MS fragmentation spectra all contribute to confident identification, bypassing the analogues’ co-elution-related issues. Moreover, the combination of Full MS and ddMS² acquisition allows both untargeted screening and confirmatory analysis in a single workflow, enhancing overall robustness. Sensitivity was another critical factor in method development. Method sensitivity was sufficient to reliably detect 40 out of 90 target analytes at 1 ng/mL, while all compounds were detectable at higher concentrations (ranged from 5 to 100 ng/mL), reflecting uneven performance across different chemical classes. In general, substances characterized by higher LOI values, such as SCRAs, are typically consumed at higher doses, partially compensating for the reduced analytical sensitivity, which remains adequate for practical screening since typical NPS concentrations in OF are generally higher than the method’s LOI for most substances [29]. The reduced sensitivity for such substances could be attributed to the poor ionization efficiency at the applied source conditions, which represents the best compromise to detect the extensive panel included in the method. Nonetheless, caution is warranted when interpreting negative results for these compounds or in cases of low-dose exposure. In contrast, opioids showed low LOI values (1 ng/mL) for most of the compounds, reflecting their high pharmacological potency. The optimal dilution factor of 1:2 was selected after testing several ratios (1:2, 1:3, 1:5, and 1:10) to balance matrix effects and analyte detectability. Higher dilutions reduced matrix interference but decreased signal intensity, while lower dilutions increased background noise. This finding emphasizes the importance of carefully optimizing sample preparation for qualitative screening. Carryover tests confirmed that residual analytes did not affect subsequent injections, supporting the reliability of the method for analyzing high-concentration samples. Application to real OF samples confirmed the robustness and practical applicability of the method. The procedure reliably detected multiple NPSs in samples with varying composition, providing correct results in 90.5% of the cases. Out of the 21 samples, 10 were negative, with five containing no substances and five positives for compounds not included in the panel, the partial detection of a single analyte, and one false positive. The reliability rate of the method was 93%, matching the proposed acceptance criteria of Lemyre et al. [30]. These rare discrepancies could be attributed to degradation of the analytes, in case of false negatives, or unexpected interference from the matrix due to rare endogenous analytes. Despite these limitations, NPSs from multiple chemical classes including SCRAs, SCs, and opioids were reliably detected, confirming the method’s effectiveness in real biological samples and its practical applicability in forensic and NPS monitoring contexts. Another important limitation is the coelution of some positional isomers, such as 3-MMC and 4-MMC and 3-CMC and 4-CMC, whose exact discrimination requires a specific method during confirmatory step analysis. Nonetheless, certain limitations should be acknowledged. As a qualitative screening method, the approach is not intended for accurate quantification, and differences in sensitivity were observed among compounds belonging to different chemical classes. Furthermore, the method’s performance depends on chromatographic and dilution parameters, which may require adjustment when incorporating new NPSs into the panel. Continuous updating of the target list will be necessary to maintain the relevance of the method in the rapidly evolving NPS market. For each new substance, retention time and characteristic fragmentation spectra must be determined, and the method must be partially revalidated to confirm reliable detection under the established analytical conditions [31]. The UHPLC-HRMS/MS method developed allows incorporation of emerging NPSs without changing the overall sample preparation or chromatographic setup, ensuring the method remains practical and applicable for forensic, clinical, and roadside monitoring.