Evaluating Blood Microsampling for Large-Scale Chemical Exposome Studies

Understanding the chemical exposome—cumulative environmental exposures through air, food, and water—and its impact on health requires large-scale, longitudinal studies. Traditional blood sampling methods, while biologically informative, are costly, invasive, and limit repeated measurements necessary to capture intra- and interindividual exposure variability. Emerging blood microsampling technologies offer a promising alternative by enabling minimally invasive, user-friendly collection, facilitating repeated sampling and broader population recruitment. However, potential chemical contamination from the devices themselves (“chemical background”) poses a challenge for sensitive nontargeted exposomics analyses.

A recent study conducted in the research group of Professor Jonathan Martin at Stockholm University (Sweden) critically reviewed commercially available microsamplers, their use in multiomics research, and experimentally evaluated their chemical backgrounds. LCGC International spoke to Solveig Thiele, lead author of the resulting paper (1), about their work.

What advantages does high-resolution mass spectrometry (HRMS) offer for studying the chemical exposome compared to low-resolution mass spectrometric techniques?

While low-resolution mass spectrometry is a very useful tool in chemical exposomics because of its high sensitivity, it is only useful when you know what you want to look for, and if you also have the pure authentic standards available—this is the traditional approach to targeted analysis. HRMS instruments can also be quite sensitive, and due to their higher spectral resolution can reduce chemical noise, separate isobaric compounds and deliver an exact mass, isotope patterns and HRMS fragmentation spectra for unknown compounds. This opens the door to molecular discovery, making it possible to accurately predict chemical formulas and structures for thousands of detected features in untargeted data acquisitions. It is always exciting to process my data and look at the interesting list of chemicals in my samples that I had not expected to be there, or that nobody else has ever seen before!

How does HRMS enable both targeted and nontargeted analysis in exposomics, and why is this dual capability important?

Importantly, using HRMS does not mean that we give up on targeted analysis, most of the projects in our lab use combined targeted/nontargeted workflows, and the joint datasets are highly complementary. For those hundreds of priority chemicals that we know are toxic and are likely in our samples, target analysis enables us to report quantitative data, which is important for risk assessment and certain health studies. Including a large panel of target analytes in a nontargeted method also allows us to define the chemical space that we can detect, and allows us to benchmark the overall method sensitivity and linearity. After validating the methods, in our workflows it is only a matter of running calibration curves and adding a mix of internal standards before sample preparation. We use comprehensive tandem HRMS acquisition strategies in our lab, which detect most target analytes within the untargeted acquisition (1). All precursor ions (MS1) are measured in HRMS, and all collision-induced fragments (MS2) are also measured in parallel by HRMS, theoretically providing a paired MS2 spectrum for all precursor ions detected in MS1 in all samples. This data-independent acquisition (DIA) strategy, supported by powerful spectral deconvolution steps, enables us to annotate the structure of many detected features, which is very important for chemical exposomics where we are looking for potentially toxic substances that may only be present at low concentrations.

Given the vast number of chemicals in the global environment, how can HRMS support the detection and identification of previously unknown or unmonitored compounds in human blood?

As already noted, using HRMS enables nontargeted analysis and molecular discovery, which is the key to finding and identifying previously unknown compounds in human blood and environmental samples. Nontargeted analysis generates feature lists as the output, with each feature potentially representing a unique chemical with respective spectral information (MS1 and MS2) and retention time. These features can be explored in high-throughput with various tools. First and foremost, this includes spectral library matching, where the precursor mass and MS2 spectrum of a feature are compared to reference library spectra and scored according to the level of agreement between the measured and reference spectra. However, we know that there are at least 350,000 man-made substances in global commerce (2), and most are not included in spectral databases today. For this reason, other tools like in-silico structure predictions and molecular networking can be very useful to annotate unknown compounds, and we can leverage long lists of relevant chemical structures (such as PubChemLite) to constrain the predictions.

What are some of the major analytical challenges associated with using HRMS for nontargeted exposome-wide studies, especially when dealing with complex biological matrices like plasma?

The major analytical challenges when looking for trace-level environmental analytes in complex biological matrices, like plasma or whole blood, are molecular interferences and method sensitivity. Especially in blood, where our analytes are generally 1000-fold lower in concentration than endogenous or dietary compounds (3), high sensitivity is crucial and might require sample pre-treatment to remove the major small molecule interferences—phospholipids. Previous work in our lab has demonstrated that by carefully removing phospholipids from blood plasma, we reduce the matrix effects and enable larger injection volumes onto the chromatographic column (4,5), which have a dual benefit to method sensitivity, and differentiates our methods from traditional metabolomics. Besides high sensitivity, nontargeted analysis with HRMS also requires high-quality spectra to get as many good-quality library matches as possible. By removing complex mixtures of major phospholipids from the samples, the MS2 spectra of other small molecules should be cleaner and easier to deconvolute, thus improving spectral matching and accuracy of in-silico structure predictions.

How can chemical background contamination from microsampling devices interfere with HRMS-based analysis, and what strategies can minimize such interference?

Chemicals present in any sampling equipment, including microsampling devices, can interfere with the HRMS analysis of the chemical exposome by migrating into the sample during sampling, shipment or storage. Since exposomics focuses on many trace-level environmental contaminants that may also be currently used in plastics or other commercial products, background contamination from sampling devices can create false positive results, or may mask the true low levels present in the sample. These problems lead to inaccuracy, or a loss of sensitivity. Apart from that, a major chemical background signal can also create competition for ionization in the ion source of the mass spectrometer, broadly reducing sensitivity for all coeluting sample ions. Chemical background from the sampler cannot be fully avoided, but simulated sampling studies can help to choose the cleanest devices or can inform the choice of solvents used in extraction to reduce leaching of chemicals from the sampler, as we have shown in our recent publication (1). In addition, field blanks should always be taken alongside real samples. Field blanks consist of analyte-free matrix, or a suitable matrix substitute, collected at the same place and time as the real samples, and also shipped, stored and processed together with them. Such blanks control for all possible types of contaminations and can be used to adjust the results.

Why is it necessary to collect repeated blood samples from the same individuals in exposomics studies, and how does HRMS help address the variability of chemical exposures over time?

Collecting repeated samples from the same individual is crucial for exposomics studies because the chemical exposome is highly dynamic over time. A previous study from our group (6) found that most environmental substances varied greatly within the same person between six visits over a period of two years. Since the environment of a person changes constantly, many of the less stable or easily excreted environmental substances will have fluctuating levels in a person over the course of any day. Therefore, measuring accurate exposure levels requires multiple samples, as any one sample may not be fully representative. Microsampling together with higher-throughput HRMS analysis holds promise to provide the number of data-points needed to fully understand the dynamics of environmental exposures and their effects on health.

In what ways can blood microsampling technologies complement HRMS-based exposomics research, particularly regarding sample throughput and population diversity?

Blood microsampling devices were conceived for layperson use outside the clinic, including at home, which should facilitate large studies and capture new segments of rural and remote populations that do not live close to clinical services (1). The sampling kits can be mailed directly to the participant, and as mentioned above, any research participant could be requested to take multiple blood microsamples over the day, or throughout a week, to capture the dynamics of environmental exposure. Sampling takes less than 15 minutes total, and can be done from the comfort of one’s home, so sampling every other day, or even multiple times in one day is likely feasible. With further method development and validation to handle very large numbers of blood microsamples, we hope the chemical exposome can be characterized and studied at large scale with greater temporal resolution than ever before.

What are the potential limitations of integrating microsampling devices into HRMS workflows, especially concerning trace-level analyte stability and quantification accuracy?

Integrating blood microsampling devices in exposomics workflows comes with some potential limitations. Aside from chemical background, other limitations may include method sensitivity and analyte stabilities (1). Microsamplers, as their name suggests, provide a small volume of sample (as low as 10 µL), which challenges method sensitivity. Analyzing trace-level analytes from these small blood volumes therefore require rigorous method development and highly sensitive instruments. Ensuring the stability of the analytes on the microsamplers during sampling, transport and storage is also crucial. More studies need to be conducted with different samplers to identify the optimal conditions for sample handling, shipment and storage, considering factors such as drying time and storage temperature, and developing standard operating procedures that participants can carefully follow.

How does the high data complexity of HRMS impact the statistical and computational approaches needed for exposome-wide association studies (ExWAS)?

Pre-processing of HRMS DIA data from a large study can take several days, even on dedicated and powerful modern computers in our facility. The resulting complex datasets, containing tens- or hundreds-of-thousands of molecular features impacts how statistical analyses in ExWAS proceeds. Since testing each feature’s association with the endpoint of interest, one at a time, leads to an explosion of false associations (Type I error), the data complexity should be reduced as far as practical in initial steps. For example, this can be achieved by grouping highly-correlated chemical exposures together to minimize the number of comparisons. Another approach that can be applied in longitudinal studies (repeated samples from the same people over time) that we have suggested (6) is to exclude many features in a dataset that have low detection frequencies and low intraclass correlation coefficients (unstable signals in repeated sampling), as this will greatly reduce the complexity of chemical exposome datasets.

What considerations should researchers keep in mind when designing large-scale exposomics studies that rely on HRMS data, especially regarding sample representativeness, quality control, and reproducibility?

Large-scale exposomics studies will inevitably include tens-of-thousands of samples, ideally including multiple samples and field blanks from each participant over time. In fact, a European-level exposome study of over 10 million people has been proposed to the European parliament (7). This means that many different laboratories will need to collaborate on the sample analysis, so having robust harmonized methods for both LC- and GC-HRMS analysis of the chemical exposome will be the key to international success. The field is not quite ready for this today, this is why our laboratory now focuses on method development for high-throughput chemical exposomics analysis. Other laboratories have the same interest, and various interlaboratory studies are now being conducted to identify sources of data variability. There is reason to be optimistic that the international community can eventually pull this off, as there are international networks for coordination (IHEN and Nexus), and a pan-European exposome infrastructure is now at the stage of being implemented (EIRENE RI).

References

1. Thiele, S.; Martin, J. W. A Critical Review and Evaluation of Blood Microsampling Devices for Exposomics. Environ. Sci. Technol. Lett. 2025, acs.estlett.5c00707. DOI: 10.1021/acs.estlett.5c00707
2. Wang, Z.; Walker, G. W.; Muir, D. C. G. et al. Toward a Global Understanding of Chemical Pollution: A First Comprehensive Analysis of National and Regional Chemical Inventories. Environ. Sci. Technol. 2020, 54 (5), 2575–2584. DOI: 10.1021/acs.est.9b06379
3. Rappaport, S. M.; Barupal, D. K.; Wishart, D. et al. The Blood Exposome and Its Role in Discovering Causes of Disease. Environ. Health Perspect. 2014, 122 (8), 769–774. DOI: 10.1289/ehp.1308015
4. Xie, H.; Sdougkou, K.; Bonnefille, B. et al. Chemical Exposomics in Human Plasma by Lipid Removal and Large-Volume Injection Gas Chromatography–High-Resolution Mass Spectrometry. Environ. Sci. Technol. 2024, 58 (40), 17592–17605. DOI: 10.1021/acs.est.4c05942
5. Sdougkou, K.; Xie, H.; Papazian, S.et al. Phospholipid Removal for Enhanced Chemical Exposomics in Human Plasma. Environ. Sci. Technol. 2023, 57 (28), 10173–10184. DOI: 10.1021/acs.est.3c00663
6. Sdougkou, K.; Papazian, S.; Bonnefille, B. et al. Longitudinal Exposomics in a Multiomic Wellness Cohort Reveals Distinctive and Dynamic Environmental Chemical Mixtures in Blood. Environ. Sci. Technol. 2024, 58 (37), 16302–16315. DOI: 10.1021/acs.est.4c05235
7. Vermeulen, R. Human Exposome Research - Potentials, Limitations and Public Policy Implications; PE 765.791; EPRS | European Parliamentary Research Service, Scientific Foresight Unit (STOA), 2025. www.europarl.europa.eu/RegData/etudes/STUD/2025/765791/EPRS_STU(2025)765791_EN.pdf (accessed 2025-11-28).