Enter the recently concluded Surveillance For Anti-Cancer Drugs Exposure Study (SurFACES), which examined the variability of handling practices for antineoplastic drugs at nine cancer centers in Canada and the U.S. In this article, we share some of the lessons we learned in addressing challenges associated with this year-long study from three perspectives: clinical or pharmaceutical, analytical laboratory, and occupational hygiene. Whether you are conducting academic research in clinical environments, working in the healthcare industry, or consulting to the healthcare industry, these lessons will be useful in informing your work practices.  THE PHARMACY/NURSING SIDE Little guidance exists on how to conduct sampling for the surveillance of chemotherapy drugs. It’s important to first consider overall surveillance objectives: is reduction of environmental contamination among them? If so, knowledge of how the drugs move through a facility is an important factor to consider.  Chemotherapy drugs are typically received in the pharmacy where they are compounded and prepared for the patient. When the drugs arrive from manufacturers, the vial surfaces are frequently contaminated with drug residues. There are multiple environmental surfaces affected by the movement of these drugs as they are stored, retrieved, moved to the biological safety cabinet for compounding, and transferred to work surfaces to confirm that the dosage is correct before being transported to the patient administration areas. While these drugs are sometimes given in pill form or as injections, very often they are infusions that are prepared according to each patient’s specific drug treatment. Infusion bags or syringes are taken from the pharmacy to the patient administration areas, where they are received by the nurses who take them to a patient’s room or the treatment bay. While our study did not address the surface contamination potential found in oral dispensaries, we anticipate similar issues in that environment (surface contamination of vials and packaging, for example) before the drugs are given to patients to take home.  Nurses conduct several preparatory tasks, including critical quality assurance and patient safety steps, before initiating an infusion. Throughout these procedures, the infusion bags containing the drugs contact numerous surfaces. The nurses’ steps vary slightly depending on the facility layout, the patient, and the specific treatment, potentially increasing the number of surfaces that come into contact with chemotherapy drugs.  We highly recommend putting a plan in place concerning what to do with the surface contamination measurement results before they are generated. This plan should include how the results will be reported, interpreted, and communicated. In our study, we defined levels that would trigger follow-up actions prior to initiating wipe sampling. For example, we defined a “cleaning criteria level”; if that level was observed, an escalating cleaning protocol was invoked along with additional wipe sampling to verify successful decontamination.  Another key question for which little guidance exists is how to report results in a way that balances the duty to warn with the duty to present the science in an appropriate context. After consulting with our clinical partners, we defined an action plan for surfaces on which residues were found to exceed a predefined level. We categorized results into three different levels: 1) below the limit of detection; 2) above the LOD, but below the level defined by our team as warranting additional cleaning; and 3) levels that were above our defined cleaning criteria level. In our study, all measurements were expressed as drug mass per cm2 of surface area sampled.  It is reasonable to interpret wipe sample results as indicative of the presence of a specific drug on the day the wipe sampling was conducted, affirming the potential for dermal exposure. In interpreting the results for stakeholders, it is helpful to understand the scope of contamination. Widespread surface contamination or the contamination of many surfaces suggests a different issue compared to when one or a subset of surfaces is found to be contaminated. An isolated event resulting in widespread contamination could indicate specific work practices that need improvement or an unplanned event such as a spill. A more formal statistical analysis of the geometric standard deviation of the wipe sample dataset is helpful in evaluating whether a work process is “out of control,” but conducting such an analysis can require advanced statistical skills when the dataset includes results that are below the LOD. Comparing which drugs were detected with the compounding records is useful for identifying potential persistent residual contamination. Documenting information about work conditions on the day of sampling (for example, which drugs were handled, the number of doses prepared and treatments completed) is critical to accurately interpreting the results. Unplanned events such as spills can significantly skew results, so it’s important to document their occurrence, location, surfaces, and the drug involved, as well as the half-life of the drugs in question. During this year-long study, we learned that taking photos of specific locations outlined in our initial planning meant that we could consistently sample the intended surfaces and sites despite vacations, staff turnover, and unanticipated absences.  ANALYTICAL LABORATORY PERSPECTIVE The laboratory that analyzes your samples will likely use one of several published liquid chromatography/tandem mass spectrometry (LC-MS/MS) analytical methods to measure chemotherapy drugs. The method we used was validated and published by several members of our team (see the resources below for more information). The sampling and analytical methods were optimized to provide the simultaneous measurement of 11 different chemotherapy drugs, trading some degree of sensitivity to maximize the number of drugs we were able to detect.  Limits of detection differ depending on the drugs and surfaces sampled, the analytical method, and even the laboratory instrument. It is important to emphasize that a high LOD (driven by low method sensitivity) could yield a “negative” result for a surface that in fact is still contaminated—at levels that may be biologically relevant. For example, the eight-hour time-weighted average provisional occupational exposure limit for cyclophosphamide, as proposed by pharmaceutical companies that make the drug, is 0.01 mg/m3. This can be converted to a mass per unit area (ng/cm2), assuming the volume of air inhaled over an eight-hour period is 10 m3, of roughly 10 ng/cm2. Thus, a lab whose analytical LOD for cyclophosphamide is 10 ng/cm2 would report a value of <LOD for anything below this level. The LOD and reporting style will vary between methods and service providers, so a clear understanding of what the results mean is crucial to interpreting them appropriately.

Sample recovery (recovery of drug from the wipe sample) is another important consideration for any wipe method. As part of our method validation, we conducted studies of sample stability and surface recovery to determine whether time and surface characteristics might constrain sample recovery and impact the laboratory analysis. We found that surface recovery (amount of drug on the surface that is removed when wiped) will affect the mass detected and should be determined before routine sampling is conducted. Surface recovery generally decreases with increasing porosity and wear of the surface. Differences in drug potency (as indicated by the dose, for example) affect drug mass; more potent drugs require less mass per dose, which means that an infusion bag with a more potent drug would have less drug mass than an infusion of a drug that is less potent. When documenting wipe sampling details, we recommend capturing this type of information for future reference. OCCUPATIONAL HYGIENE CONSIDERATIONS Chemotherapy drugs have been designated by NIOSH to be controlled to levels that are “as low as reasonably achievable.” Only USP General Chapter <800> speaks to specific contamination levels. Wipe sampling provides objective assessment of the presence of drug residues that meets the intentions set out in specific guidelines such as USP <800> and NIOSH’s guidance on preventing exposure to antineoplastic drugs in healthcare settings, and serves as an important element of a comprehensive exposure control plan. Wipe sampling can also help confirm the adequacy of workplace engineering controls and housekeeping measures as well as identify the need for supplemental health and safety training or the modification of work practices.  While guidance on where and when to sample is scarce, the location and timing should align with your objectives. The surfaces we selected for our study reflect our objective to understand how much and how variable these drug residues are. We elected to omit surfaces that are well known to be contaminated, such as patient chair armrests. When selecting where to sample, consider differences in the  physical layouts of facilities, as these differences can reflect potential differences in work practices and pathways of the drugs through the facilities. Before sampling, we recommend visiting each site, meeting with the staff leads, and walking through each area to identify the surfaces to be sampled. Select surfaces after carefully observing the staff at work to ensure that you understand how surfaces are used. We consulted with the workers compounding and administering the drugs in each facility to gather their input on surfaces to sample. This information helped supplement our assumptions derived from our previous work and the literature. Incorporating staff input indicates researchers’ responsiveness to staff concerns and promotes buy-in for when the time comes to interpret the results and propose follow-up action.  Another step we found helpful is to create a reference guide using the pictures taken to document the surfaces selected for assessment and where on each surface the samples are to be collected. A reference guide is invaluable when referring back to where you sampled in the months and years following wipe sampling. It can also be helpful and efficient to premeasure and record surface areas before sampling commences. Sampling surface areas that are the same (for example, using a 100 cm2 template, a fairly standard area for wipe sampling that approximates the surface area of the palm of one hand) will make it easier to compare contamination across surfaces. Consider prefilling the sampling form and labeling the sample vials before going into the field to save time and reduce labeling errors. Establishing a sampling schedule that aligns with the surveillance strategy requires detailed knowledge of the schedule of cancer treatment services provided at the clinic. The most representative results will be obtained when the samples are collected during normal work hours on a typical workday. If the objective is to ensure ongoing control of exposures, sampling should not be done right after surfaces have been decontaminated. To reduce the risk of exposure bias, it is also important to vary the sampling day since some drugs may only be used on certain days at some facilities. For example, specialized treatments that are administered for specific types of cancer may only be available on days when the oncologist-specialist overseeing the patient’s treatment works at that clinic, so those drugs will be compounded and administered on some days but not others. When it’s finally time for environmental health and safety contacts at the clinic to review the results of the study, hygienists should have a plan for disseminating the results and, if necessary, communicating recommendations for follow-up. This is challenging given the lack of health-based occupational exposure limits or surface wipe limits, and necessitates the development of an internal provisional risk management policy that may include some threshold level for follow-up measures. If results exceed a predefined threshold, specific exposure reduction protocols can be invoked. In our study, we defined a cleaning criteria level that was loosely based on limited published occupational exposure data related to airborne exposures during the manufacture of chemotherapy drugs. We defined thresholds requiring additional “deep cleaning” protocols that corresponded to specific cleaning instructions. For the highest level of contamination, we recommended post-cleaning wipe sampling to verify cleaning effectiveness.     HUGH DAVIES, PhD, CIH, is an associate professor in the School of Population and Public Health at the University of British Columbia in Vancouver, Canada. SUSAN ARNOLD, PhD, CIH, is an assistant professor and director of the Exposure Science and Sustainability Institute in the School of Public Health at the University of Minnesota in Minneapolis, Minn. MATTHEW JERONIMO, BSc, is the manager of the occupational and environmental health laboratory in the School of Population and Public Health at the University of British Columbia. GEORGE ASTRAKIANAKIS, PhD, is retired from the University of British Columbia School of Population and Public Health. CAROLE CHAMBERS, BSc(Pharm), MBA, FCSHP, is the pharmacy director of Cancer Services with the Alberta Health Services Pharmacy in Calgary, Alberta, Canada. Editor’s note: All of the authors of this article were involved in the Surveillance For Anti-Cancer Drugs Exposure Study. Send feedback to The Synergist.