LC–MS/MS Tracks Drugs in Australian Wastewater

Drug and alcohol use is a major problem in Australia, both in terms of cost to society and impact on health. A recent national survey found that nearly half of Australians aged 14 and up have tried an illegal drug at some point in their lives. Sadly, over 1600 people died from drug-related causes in 2023 alone. On top of that,¹ Australians spent an estimated $11.5 billion on drugs like cocaine, heroin, MDMA, and methamphetamine in just one year.¹ Making things even more complicated, new synthetic drugs keep popping up that scientists know very little about, which makes them especially risky.

To get a better picture of what's really happening with drug use, researchers pull data from lots of different sources: hospital emergency rooms, used needle testing, surveys, police records, drug seizures, on-site drug checking services, and even wastewater testing (yes, testing sewage to see what drugs people have been using). Each method has its blind spots, but combining them gives a much clearer, more complete picture than relying on just one.

Certain times of year—like holidays, festivals, concerts, and big sporting events—tend to see spikes in drug use, since people are more likely to party and let loose during these periods. One way to keep tabs on what's actually being used (and how safe it is) is through drug checking services, where people can bring in a sample of their drugs to get it tested for free. These services not only tell people what's actually in their drugs, but also offer harm-reduction advice, like how to avoid overdosing.

A recent study (which inspired a recently published paper in the journal Drug Testing and Analysis¹) looked at two of these methods—wastewater testing and drug checking—side by side, focusing specifically on the New Year holiday period in several cities and regions across Australia. Wastewater testing showed the bigger picture of drug use across the population, while drug checking offered more specific details about what exact substances people were using and how pure they were. By comparing this data with existing national reports, the researchers aimed to show how using both methods together—rather than just one—gives a much more complete understanding of drug use during high-risk periods like holidays, and how the drug market shifts during those times.

LCGC International spoke to two of the authors of the aforementioned paper, Richard Bade and Malcolm McLeod, about the research and their findings.

For quantifying trace-level drug metabolites in wastewater (often ng/L concentrations), why is liquid chromatography-tandem mass spectrometry (LC-MS/MS) typically preferred over gas chromatography-mass spectrometry (GC-MS), and in what circumstances would GC-MS still be the better choice?

RB: LC-MS/MS is typically preferred because many drug residues and metabolites in wastewater are polar, non-volatile, and thermally labile, making them unsuitable for direct GC-MS analysis. GC-MS is still useful for volatile or semi-volatile compounds, compounds that can be readily derivatised, and as a complementary tool for confirmation or broader screening studies.

Wastewater is a complex matrix full of interfering compounds. Walk me through how you would design a solid-phase extraction (SPE) protocol to isolate illicit drug residues before chromatographic analysis, and what factors influence sorbent choice.

RB: Our initial SPE method was adapted from a forensic toxicology method, but we have subsequently optimised it for wastewater applications by evaluating a range of SPE sorbents. Sorbent selection is driven primarily by the target analytes. If the aim is to monitor a specific drug class, such as cathinones, opioids, benzodiazepines or cannabinoids, there are sorbents designed to maximise recovery for those compounds. However, for wastewater-based epidemiology we often want to capture a broad range of traditional and novel psychoactive substances, so no single sorbent is ideal for every compound. In these cases, the challenge is to balance recovery with broad class coverage.

When designing a new SPE method, I would first consider which compounds are of interest and their physicochemical properties, then evaluate several broad-spectrum sorbents to determine which provides the best overall performance. I would also optimise factors such as sample pH, wash and elution conditions, and the concentration factor. For example, in our laboratory we typically concentrate samples from 100 mL down to 100 μL prior to LC-MS/MS analysis, but this can be adjusted depending on instrument sensitivity and the expected concentrations in wastewater.

How would you detect and correct for ion suppression or enhancement caused by the wastewater matrix in an LC-MS/MS assay, and why does this matter for accurately back-calculating community-level drug consumption?

RB: We use isotopically labelled internal standards to correct for ion suppression and enhancement caused by the wastewater matrix. However, isotopically labelled standards are not available for all compounds, particularly many novel psychoactive substances (NPS). In those cases, we select surrogate internal standards and conduct recovery and matrix-effect experiments to determine which surrogate provides the best correction.

Ion suppression or enhancement directly affects the measured analyte concentration, causing under- or overestimation of drug loads in wastewater. Because community drug consumption estimates are calculated from these concentrations, uncorrected matrix effects can lead to biased consumption estimates and misleading conclusions about drug use trends or differences between populations.

NPS often include structural isomers or analogues with very similar physicochemical properties. What column chemistry and gradient strategies would you use to achieve baseline separation of closely related NPS?

RB: There are indeed many structural isomers among NPS. To achieve baseline separation, I would evaluate different stationary phases, as column chemistry is often the most important factor governing selectivity between closely related compounds. Columns such as biphenyl, phenyl-hexyl, C18 and C8 can provide different interaction mechanisms and should be assessed during method development.

Gradient optimisation is also important. For closely eluting isomers, a shallower gradient or an isocratic hold around the elution window can improve resolution by allowing more time for chromatographic separation.

For example, in our conventional NPS LC-MS/MS analysis we use a biphenyl column, which works well to separate several structural isomers, including the methylmethcathinone isomers. However, 2’- and 3’- fluoro-2-oxo-PCE still co-eluted so we needed a different column. As explained in the paper, to allow complete separation of the 2’- and 3’- fluoro-2-oxo-PCE isomers we employed a C8 column. This separation was likely governed primarily by differences in hydrophobic and steric interactions between the isomers and the stationary phase, resulting in sufficient chromatographic resolution.

MM: From the drug checking perspective usually employ techniques like Fourier-transform infrared (FTIR) and nuclear magnetic resonance (NMR)that provide structural information and can be more discriminating than LC-MS/MS in some instances. For the case in point, we identified 2’-fluoro-2-oxo-PCE in the drug checking service as the FTIR showed differences to the isomer 3’-fluoro-2-oxo-PCE (fluorexetamine). In this case NMR was pivotal in confirming the substitution pattern on the aromatic ring.

A narrative account of the identification of 2’-fluoro-2-oxo-PCE is available:

https://theconversation.com/an-entirely-new-illicit-drug-has-been-discovered-by-australian-chemists-heres-how-they-did-it-192855

In drug checking services like CanTEST, results need to be delivered quickly and confidently to clients. How does the balance between speed, sensitivity, and specificity differ between screening techniques (e.g., FTIR, immunoassay) and confirmatory chromatographic methods like GC-MS or LC-quadrupole time-of-flight (QTOF)?

MM: There are many different models adopted for drug checking around the world. Some services test on site with the clients in attendance, and others use a drop-off or even mail-in model of testing where samples are collected and analysed in a laboratory with the results reported back to clients later.

At CanTEST we have adopted an on-site testing model and use LC-PDA on site for quantitative and qualitative analysis using RT and UV spectrum matching to identify substances. This includes the quantification of a small collection of the most expected drugs with an injection-to-injection time of just over 6 min. The PLC complements FTIR, test strips and in rare cases reagent testing.

The CanTEST service is supported by laboratory analysis for quality assurance that routinely includes low- and high-resolution liquid chromatography tandem mass spectrometry (LC-MS/MS). In some cases, we use NMR for drug identification and quantitative NMR (qNMR) for quantification. These laboratory results typically are not reported back to clients but do help us ensure the methods at the service are functioning well.

When converting drug concentrations detected in wastewater influent into population-level consumption estimates, what role does chromatographic quantitation accuracy play, and how do you account for degradation of the target analyte during transit through the sewer system?

RB: Accurate chromatographic quantification is essential because any error in the measured wastewater concentration is directly transferred to the final estimate of community drug consumption. To account for sewer degradation, we have been exploring the use of laboratory bioreactors to investigate the in-sewer stability of target analytes. These studies help determine whether compounds are sufficiently stable to serve as wastewater biomarkers. Unfortunately, there are limited stability data available for many NPS, highlighting an important knowledge gap that we are hoping to address in future work.

What key validation parameters (LOD, LOQ, recovery, precision, matrix-matched calibration) would you prioritize when developing a new multi-residue chromatographic method to simultaneously screen for stimulants, opioids, and their metabolites?

RB: When developing a new method to simultaneously screen for stimulants, opioids, and their metabolites, I would prioritise recovery, precision, and matrix effects before LOD/LOQ. A very low detection limit is not particularly useful if analyte recovery is poor or results are not reproducible. Therefore, I would first ensure accurate and precise quantification using appropriate recovery studies and matrix-matched calibration, then optimise sensitivity to achieve suitable LODs and LOQs.

New psychoactive substances frequently lack reference standards or established retention time libraries. How would you approach tentative identification of an unknown compound using high-resolution mass spectrometry coupled with chromatography (e.g., LC-QTOF or GC-QTOF) in the absence of a certified standard?

MM: In the drug checking context sample is typically not limited and can be analysed or extracted and analysed readily by techniques that include NMR in addition to high-resolution GC-MS and LC-MS/MS. Using this approach, it is usually possible to characterize NPS and establish compound identity. Given the health implications of drug checking, rapid compound identification is essential. Structure elucidation can be achieved on a rapid timescale of hours or days, while obtaining reference materials in Australia typically takes months.

I provide some links below to open access articles.

https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/dta.3637

https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/dta.3910

https://www.sciencedirect.com/science/article/pii/S2468170926000202

This study compares wastewater analysis with drug checking data to identify consumption trends. What analytical or chromatographic discrepancies might arise between these two data sources (e.g., differences in what's detected, timing, or population coverage), and how would you reconcile them?

MM: This is discussed at some length in the article. Some of the main differences in the current study are that drug checking is limited to one geographic location and this was distinct from the areas surveyed by wastewater analysis. Drug checking service clients are a small group of people motivated and able to attend the service and may not reflect the wider population of drug users. Drug checking also has access to larger quantities of the samples presented and is more likely to detect and report on relatively rare drug presentations that may be below limits of detection in wastewater due to this rarity. The differences provide some limitations in comparing data but also some strengths in providing complementary analytical pictures of drug use.

RB: While each approach has its own limitations, they are highly complementary. Together, they help bridge the gap between individual- and community-level insights, providing a more complete picture of changing drug consumption trends.

References

1. Usman, M.; McLeod, M.; Nadarajan, D. et al. Understanding Consumption of Illicit Drugs and New Psychoactive Substances in Australia by Triangulating Drug Checking and Wastewater-Based Epidemiology. Drug Test Anal. 2026. DOI: 10.1002/dta.70120