Routine blood testing for "Forever Chemicals": implications of the 2022 NASEM PFAS report recommendations

Introduction

The National Academies of Sciences, Engineering, and Medicine (NASEM) issued their “Guidance on Perfluoroalkyl and Polyfluoroalkyl substances (PFAS) Exposure, Testing, and Clinical Follow-up Consensus Report” (hereafter referred to as the “NASEM PFAS Report”) in July 2022 [1]. As physician specialists in Medical Toxicology and Occupational and Environmental Medicine (OEM), we were grateful for the attention this document brought to the important topic of environmental health and toxicology. We acknowledge that this area of medicine is fluid, with rapidly advancing science that has changed the medical community’s understanding of the types of health conditions associated with exposure to PFAS and the strength of those associations [2]. Of particular note, the International Agency for Research on Cancer (IARC) recently labeled a representative PFAS, perfluorooctanoic acid (PFOA), to be a Group 1 human carcinogen [3]. After reviewing the NASAM PFAS Report, however, we believe that the Committee’s recommendations for clinical management of PFAS exposure, testing, and follow-up cannot be successfully implemented as outlined. We highlight seven unresolved issues that undermine the recommendations of the NASEM PFAS Report and their potential implications.

Serum PFAS testing must be standardized

The NASEM PFAS Report correctly identifies that there are no current standard methods for PFAS exposure biomonitoring. The Committee recommends that commercial laboratories providing these tests use the method outlined by the Centers for Disease Control and Prevention (CDC) in the National Health and Nutrition Examination Survey (NHANES) program, known as online solid phase extraction coupled to high performance liquid chromatography-turboionspray ionization-tandem mass spectrometry (online SPE-HPLC-TIS-MS/MS). As noted in the NASEM PFAS Report, not all laboratories use methods similar to those used by the CDC, and may not be subject to external proficiency testing programs. Prior to implementation of large-scale testing in the general population, standardization of methodology and quality processes must be clearly established and subject to federal oversight systems.

Better define who should be tested

We acknowledge the strong desire of some community members to know their personal serum PFAS concentrations, and we note that the NASEM PFAS Report recommends serum PFAS testing for those with a “history of elevated exposure.” However, we are concerned about the definition and clinical application of the terminology “elevated exposure” used in the NASEM PFAS Report. In defining those who should be tested, the authors refer to persons who “lived in areas where PFAS contamination may have occurred.” These areas are vaguely defined as “near” such facilities as commercial airports, military bases, and wastewater treatment plants among others. Without context for geographic proximity to these facilities, duration of living in these areas, or time since departing these areas, we believe many individuals would be unnecessarily recommended for serum PFAS testing. The NASEM PFAS Report acknowledges that virtually all Americans have PFAS exposure to some degree, and that concentrations obtained from serum PFAS testing cannot be attributed to specific sources. As currently written, the implication is that testing should be considered for a substantial percentage of the US population. The volume of testing implied by the NASEM PFAS Report blurs the lines between population based biosurveillance and individualized clinical testing.

Interpretation of the result – application of the German human biomonitoring commission (HBM) values

The Committee indirectly notes that a population “reference range” approach to interpretation of serum PFAS concentrations only serves to identify varying degrees of PFAS exposure and absorption among a population. This approach to toxicological health outcome risk assessment is complicated by time since exposure and toxicokinetics. Acknowledging this, the Committee chose to utilize risk-based thresholds established by the German Human Biomonitoring Commission (HBM) [4, 5]. We note some recent calls to address PFAS as a class for purposes of regulation [6], as well as the Committee’s thoughts that potency factors similar to those used for dioxins may be necessary. However, the use of summed serum PFAS concentrations for individualized health risk assessment purposes is an approach that we do not feel can be extrapolated for a clinical setting. We believe that adoption of the HBM-I values (which defined perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) concentrations below which no adverse health effects are expected), and HBM-II values (which defined concentrations above which may lead to health impairment) might have been reasonable if the Committee recommended their use specifically for PFOA and PFOS as designed with the acknowledgement that these values were not intended to be utilized if evidence of carcinogenicity was established. In contrast, the summation of PFOA, PFOS, and five other PFAS serum concentrations effectively establishes the class of chemicals as having the same potential health effects per unit of exposure/measure. The German HBM Commission established different HBM-I and HBM-II values even just between PFOA and PFOS, in each instance recommending lower values for PFOA than PFOS. Thus, even between only PFOA and PFOS the use of summed human serum concentration does not appear justified. Further, by the Committee’s own estimation, only 2% of the US population would fall below their summed 2 ng/mL threshold for no expected adverse health effects. The Committee’s conclusions could be interpreted as intending to recommend that most, if not all, US residents undergo serum PFAS testing.

Recommendations for post serum PFAS testing clinical management do not change standard primary care practice

We acknowledge the Committee’s desire to encourage primary care preventive practice for all patients, regardless of their history of PFAS exposure. We also note that the Committee made few, if any, recommendations for post-testing management that would markedly differ from standard primary care practice. An oft-quoted adage of clinical medicine is “Do not order a test that will not change your management” [7,8]. In the case of serum PFAS testing, it seems that the clinical recommendations made are much in line with standard clinical practice, regardless of serum PFAS concentration. Why perform serum PFAS testing, if clinical management will remain essentially unchanged regardless of the result? Further, we note that American Academy of Pediatrics recommendations for dyslipidemia screening in children may be “controversial” at best, [9] and the United States Preventive Services Task Force (USPSTF) noted again in 2023 that there is insufficient evidence for or against screening for dyslipidemia in children and adolescents [10]. This point is particularly important as the committee has advised performing a test without providing guidance on the clinical management of the controversial outcome. In the case of screening for thyroid disease, the USPSTF again notes that there is insufficient evidence for screening for thyroid disorder in nonpregnant, asymptomatic adults [11]. Despite this statement from the USPSTF, it is well-recognized that thyroid disorder screening occurs regularly, albeit unnecessarily, in primary care practice. So much so that the Choosing Wisely campaign felt it necessary to include a recommendation to “Avoid Thyroid Stimulating Hormone (TSH) screening in annual well-visits for asymptomatic adults, regardless of age” [12–14]. We note that there are no current recommendations in primary care practice for routine screening for testicular cancer or ulcerative colitis. At present, the USPSTF still recommends against screening for testicular cancer in men [15]. We acknowledge that the NASEM PFAS Report recommends assessing for “signs and symptoms” of testicular cancer and/or ulcerative colitis for those patients older than 15 years with a summative serum PFAS concentration of > 20 ng/mL. However, without a dedicated screening modality for these conditions, we question the utility of this focused assessment of “signs and symptoms” beyond what is occurring in general primary care practice.

Using microscopic hematuria to screen for kidney cancer is clinically not indicated

Recently, IARC has labeled PFOA as carcinogenic to humans (Group 1) whereas PFOS was classified as possibly carcinogenic to humans [3]. IARC noted, however, that there is still limited evidence in humans for the carcinogenicity of PFOA towards renal cell carcinoma (RCC) or testicular cancer. Understanding this, broadly implemented screening tests for RCC and testicular cancer screening for those with elevated serum PFAS concentrations should not yet be recommended. Nevertheless, if we infer that PFOA causes RCC, and even if the increased risk was very significant, the current limitations of RCC and kidney cancer screening via urinalysis as recommended by the NASEM PFAS Report for individuals 45 years of age and older leads to limited, if any, benefit with the potential to create iatrogenic harm.

For example, as recommended by the NASEM PFAS Report, a summative serum PFAS concentration > 20 ng/mL in a patient over age 45 years would prompt kidney cancer screening with urinalysis to detect microscopic hematuria. We acknowledge that the Committee derived this recommendation from the C8 Medical Panel, [16,17] but feel that the limitations of this screening approach must be addressed. Positive microscopic hematuria would be followed by advanced imaging diagnostics and, in some cases, surgical biopsy. The pooled prevalence of undiagnosed renal cell carcinoma, which represents about 85% of all kidney cancers, has been estimated to be 0.1% to 0.21% [18,19]. However, the true prevalence among the general population is likely much lower given that many of these studies were focused toward older adults. Further, the number of individuals with a positive screening test is likely to be very high. In a population cohort in the Netherlands of men aged 50 and over, 23.4% had microscopic hematuria [20]. In the general population, the reported prevalence of asymptomatic microhematuria ranges from 1.7% to 31.1%, although it would appear that in clinical practice, a prevalence of 4% to 5% may be realistic [21]. Several studies have estimated the positive predictive value (PPV) of microscopic hematuria for kidney cancer to be less than 1% [22–24]. Moreover, in Sugimaru’s 2001 study, only 35% of RCC patients had either macroscopic or microscopic hematuria. Phrased another way, the sensitivity of either macroscopic or microscopic hematuria to detect RCC was estimated at 35% (95% CI 27–44%) with 65% of detected RCC cases not displaying either macroscopic or microscopic hematuria. After eliminating those cases that noted macroscopic hematuria, the sensitivity of urinalysis for RCC was estimated at 18% (95% CI 11–26%). Even with overestimations of the prevalence of undiagnosed kidney cancer (0.25%) and an overestimation of sensitivity (45% if using both macroscopic and microscopic hematuria), the estimated positive predictive value (PPV) is 2% or less at best (Table 1). Thus, the use of urinalysis as a screening test for kidney cancer would result in 98 false positive tests for every two kidney cancers detected.

Table 1. Positive predictive value (PPV) and negative predictive value (NPV) of microscopic hematuria for detecting kidney cancer assuming 5% prevalence of microscopic hematuria in general population and 0.25% prevalence of undiagnosed kidney cancer in general population accounting for contribution of renal cell carcinoma and transitional cell carcinoma.

	Kidney Cancer Present	Kidney Cancer Not present	Totals	Predictive Values
Microscopic hematuria present	113	4887	5000	PPV is 2.26% (1.88, 2.71)
Microscopic hematuria not present	137	94863	95000	NPV is 99.86% (99.83, 99.88)
Total	250	99,750	100,000

If the clinician wants to rule out kidney cancer (or other causes of hematuria) following a positive hematuria screening test, the American Urological Association (AUA) offers an evidenced based consensus statement wherein follow-up evaluation is directed by risk stratification [25]. Following these recommendations, patients considered high risk for malignancy undergo both computed tomography (CT) urogram and cystoscopy [25]. In the above scenario, almost 4900 CT scans and cystoscopies would be needed to confirm the presence of 113 kidney cancers among 100,000 persons screened. Despite this, 137 kidney cancers would still be missed simply due to the poor sensitivity of the initial screening test. We also note that the use of ionizing radiation in a CT scan is not without its own risk of carcinogenicity, even though the individual dose of radiation received is relatively small [26].

The costs of PFAS testing and subsequent screening

Clinicians are expected to counsel patients on the cost of testing [7]. The Committee briefly discussed the cost of testing serum PFAS concentrations, however, the source of funding was not addressed. The NASEM PFAS Report cited a large and commonly utilized specialized laboratory which charges $600 for a human serum PFAS concentration panel. In our current healthcare model, private and/or government insurers will not routinely shoulder the cost of PFAS testing. Thus, the cost of PFAS testing will be translated directly to the patient in the majority of circumstances. In our clinical practice, we find this to be a significant limitation even further complicated in occupational medicine settings. Before the medical community can effectively implement the recommendations laid out by the Committee, the issue of cost in both occupational/environmental and general practice clinics must be clearly addressed. Further, the Committee estimated that 9% of the US population has a summative serum PFAS concentration > 20 ng/mL, or approximately ∼30 million people. The US Census Bureau estimates roughly 42% of the US population is 45 years or older; thus, the population to be screened for kidney cancer via urinalysis might be estimated at roughly 12 million people [27]. As highlighted above, while a screening urinalysis may be inexpensive, it is neither useful for ruling in, nor ruling out kidney cancer. Assuming an approximately 5% prevalence of asymptomatic microscopic hematuria in the general population, about 600,000 would screen positive for hematuria and necessitate advanced imaging. Assuming even a 2% positive predictive value (noting again, this is higher than noted in previously mentioned studies), 98 persons would undergo an unnecessary CT scan for every 2 confirmed cases of kidney cancer. In this scenario, about 600,000 of the original 12 million screened would undergo previously avoidable testing with the associated potential harms of incidental findings requiring further unnecessary interventions that lead to radiation risk and other procedural risk. In addition, to identify kidney cancer, CT scans would be contrast enhanced, which is associated with an adverse reaction risk of 1–12%, and 0.2% of these may be life threatening reactions [28]. The procedural risks notwithstanding, the expense of abdominal CT scanning would be large in this population of 600,000 that would require secondary testing after hematuria noted on initial screening urinalysis. It is nearly impossible to generate an average cost of contrast enhanced CT abdomen and pelvis within the United States that includes both professional fees and technical fees due to geographic variability in pricing. However, a reasonable rough estimate might be $525 per the New Choice Health cost comparison [29]. This would then equate to a reasonable rough estimate of cost of over $315 million for the 600,000 CT scans needed after screening urinalysis. As noted above per AUA recommendations for evaluation of microscopic hematuria, these patients would also be indicated for cystoscopy, a procedure that would incur additional cost and risk of adverse effects.

Role of physician specialists in toxicology

We endorse that national committees addressing important issues involving environmental chemical exposures should be appropriately and thoughtfully configured. We believe that the aims of such committees are best served when input from board-certified, practicing clinical specialists from relevant fields are also included. Given the nature of recommendations outlined in the NASEM PFAS Report, we believe that in addition to colleagues in the allied public health fields of exposure science and environmental health, it would have been appropriate to include medical toxicologists as committee members. Medical toxicology is the clinical specialty that provides a nexus between parameters of exposure and dose, dose-response, and the potential for adverse clinical effects. We believe that including input from medical toxicologists in the analysis of environmental exposures is essential in deriving conclusions most useful to public health, public policy and in arriving at clinically applicable recommendations.

Conclusion

We agree that PFAS contamination in water and food sources should be limited to the degree feasible. The US Environmental Protection Agency (EPA), as well as many state environmental agencies, has made efforts to control these substances in water supplies [30]. We acknowledge the potential health concerns, and encourage thoughtfully conceived, evidence based and clinically relevant medical screening services outlined by primary care professional organizations and governmental organizations such as USPSTF. However, given the limitations in interpretation of serum PFAS concentrations as described above, the lack of standardized screening, and the absence of any significant change in general clinical practice made by the NASEM PFAS Report, we cannot recommend broad implementation of its clinical guidance.

Disclosure statement

All authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in the paper.

Data sharing statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.