A paradigm shift in environmental monitoring – The time for non-targeted analysis (NTA) is now

As residents of a global society, we are exposed to a plethora of chemicals through the air we breathe, water we drink, food we consume, products we purchase and use, and choices we make. Many of us are unaware of the chemical soup surrounding us daily. It is beneficial for all of us that government agencies and institutions exist with a shared purpose in protecting human health from environmental chemical exposure.

Most environmental monitoring for chemical exposure is based on targeted analysis for a relatively small suite of well-studied chemicals using robust and validated analytical methods. While targeted methods serve their purpose very well, they lack breadth in overall “chemical space coverage”. In other words, targeted analysis works very well for chemicals for which it is designed to monitor, yet completely misses all other chemicals. One example of how we all are exposed to chemicals, both known and unknown, is through drinking water consumption. For over a decade, environmental forensic investigations of chemical pollution in water have made great strides by implementing non-targeted analysis (NTA) for compound discovery. NTA techniques look for what chemicals are in a sample, rather than a targeted subset of chemicals. There are plentiful examples of formerly unknown compounds, first discovered and communicated through NTA investigations, for which action was ultimately taken based on this discovery. A prime example was the detection of hexafluoropropylene oxide dimer acid (HFPO-DA, i.e., GenX) in the drinking water supply in Wilmington, NC USA (Strynar et al., 2015).

Despite the strengths of NTA, there are complications when a new chemical is discovered in the environment. One hurdle for analysts is the need for authentic and validated chemical standards to confirm chemical identity and to quantify concentrations. For many compounds, commercial sources of newly identified chemical standards often do not yet exist, requiring chemical synthesis. Critically, authenticated standards are unquestionably required to quantitatively study the fate, transport, and toxicology of newly discovered chemicals. In the interim, though, we can relax the need of an authentic standard for 1. absolute chemical confirmation and 2. chemical quantitation, and rely more heavily on semi-quantitation for concentration estimates of non-targeted data, rather than delay progress. There are often enough reliable data upon which informed interventions can be made while awaiting this additional information. For example, if a compound is detected in a waste stream, and is being emitted into a drinking water supply, one does not necessarily need to know with 100 % confidence the exact identity and concentration of the chemical to take steps to reduce emissions. Making these decisions and evaluating their impacts (e.g., intervention effectiveness) can be made with less than 100 % certainty – not that new chemicals standards will become unnecessary for confirmation and quantitation, but that “perfection can be the enemy of the good” in some situations. Many state-of-the-art environmental laboratories possess the instruments that can detect and identify chemicals with a very high degree of accuracy. Such instruments measure molecular weights of chemicals and molecular fragments to the fourth decimal place (e.g. high-resolution mass spectrometers (HRMS) with quadrupole time-of-flight and orbital trap detectors), making compound elucidation and formula prediction attainable and robust. Additional instrumentation can measure other chemical and physical properties (e.g., collision cross section (CCS) via high resolution ion mobility) to further support a compound’s identification or reduce the list of suspect chemicals requiring follow-up. When analysts consider each chemical piece of information as weight of evidence, the addition of CCS measurements can dramatically strengthen compound identification confidence. Overall, contemporary scientists now have a variety of analytical instrumentation at their disposal, which was uncommon and out of reach for most scientists just a little more than a decade ago. Now findings from environmental forensics studies often uncover chemicals not addressed in required lists of target analytes. Making intervention decisions that do not consider NTA data paints an incomplete picture of the issues being addressed. For example, on the Cape Fear River, the highest concentration compound detected (PFMOAA) was not on contemporary targeted analysis methods (Strynar et al., 2015), yet is now used to monitor for breakthrough of water treatment via granular activated charcoal.

The work of many global researchers has relied on the use of HRMS as the epicenter of many investigations for more than a decade. As a community, we have grown tremendously, and through shared experiences have advanced the science of NTA and environmental forensics along the way (Peter et al., 2021). We have made considerable progress in reporting confidence in our findings to the greater scientific community (Schymanski et al., 2015; Charbonnet et al., 2022). Some opponents of NTA techniques (and even some proponents) may argue it is not mature enough for widescale adoption, or that there is not a single uniform NTA protocol. Most will agree, though, that there is ample analytical equipment and scientific knowledge to perform NTA studies, and there have been many examples of successful application of NTA techniques. There also exists an uncomfortable space whereby NTA investigators employ methods and applications that are specific to individual laboratories, and there is truly no one standardized way to conduct NTA studies. However, there are countless examples to point to wherein discovery of compounds, such as 6PPDQ and its impact on Pacific Northwest coho salmon (Peter et al., 2018); perfluorinated ionic fluids in lithium-ion batteries (Guelfo et al, 2024); chloronitramide in drinking water (Fairey et al., 2024) and the detection of new psychoactive substances (NPS) (Klingberg et al., 2022) would not have been possible without application of NTA techniques. Additionally, in the interim after new compounds are detected, determining semi-quantitative estimates of concentrations to bridge the gap between contaminant discovery and risk characterization is useful and possible (McCord et al., 2022).

An important topic for discussion is when are there enough data to say we have a new compound present in the environment and that action is warranted? At what point in the life of a scientific advancement is an application mature enough for universal adoption? When are we willing to live with some uncertainty in using NTA data, rather than delaying action? I contend that NTA has now reached the proper maturity for general adoption as a front-line analytical technique rather than as an add-on method. It is mature enough to be a standard technique in any assessment of waste streams, biota samples, and environmental media. While the application of NTA in such studies does not ensure complete, chemical coverage, it does advance the degree of coverage dramatically. Analysts will continue to conduct targeted analysis with predefined validated analytical methods; that should not change. However, NTA should also be the norm rather than the exception. As scientists, we should aspire to provide useful, actionable information for protecting human health and the environment even if we don’t yet have all the answers.

We owe it to the billions of impacted people living in the chemical milieu around us to make scientifically sound judgements about exposures, hazards, and risk. It starts with a holistic understanding of all the chemicals around us. Numerous investigations include NTA methods on a regular basis, but many scientific investigations still lack an NTA component. As a multidisciplinary international community of researchers within the broad field of public and environmental health sciences, we can lead in the direction and application of the most impactful environmental study methods possible, and non-targeted analysis rises to this level of significance.

Data availability

No data was used for the research described in the article.