The Journey to “Evidence-Based” for Functional Labs

In the Functional Medicine space, we all desire to be described under the banner of “Evidence-Based.” What does that mean, and how do we get there? In this article I will attempt to address the laboratory component of this question from the perspective of a company in the middle of this ongoing journey.

The value of self-critique and continual self-examination:

There is a pattern I have witnessed in the Functional Medicine Lab industry that needs to be challenged from within. We develop tests, leverage them as much as we can, and then wait for competitors or critics to raise objections or questions. If those objections never come, we may be tempted to push forward without continuing to pursue both analytical and clinical validation. Continuous self-critique is critical!

In this blog, I want to welcome you into the ongoing process of self-critique, analytical validation, and clinical validation for the DUTCH test. For this exercise, we’ll focus on just one analyte in the DUTCH test – estradiol.

QUESTION #1a – What level of validation is necessary for standard, conventional laboratory tests?

For a standard laboratory test like serum estradiol, the literature is full of clinical utility. This research continues to grow but does not need repeating. To highlight one very useful clinical correlation, serum estradiol (E2) levels within or below the postmenopausal range may predispose a woman to decreasing bone mineral density (BMD)^{[1],[2],[3,[4]} along with other clinical changes. A lab’s job is simply to validate their analytical lab process, ensure that, among other parameters, their values are accurate and reproducible, and appropriately set their reference range. This is straightforward and a validated serum E2 test (preferably LC-MS/MS for increased accuracy) could be confidently used by providers in many situations, however, there may be a better way to assess E2.

QUESTION #1b – What level of validation is necessary for a Functional Lab like the DUTCH test?

Because reproductive hormones, such as progesterone, can vary significantly throughout the day in serum,^[5] a urine test like DUTCH potentially offers less variable results as it is a better 24-hour E2 representation. Additionally, urine allows providers the opportunity to assess estrogen metabolites. Conceptually this easy to collect urine test may be attractive, but does it actually work? If not, providers are better served by using the traditionally and widely accepted available tool, serum E2. We must evolve from claims to evidence!

• CLAIM – Urine E2 levels represent systemic levels
  • EVIDENCE – Strong serum correlation from same-day ^[6]
• CLAIM – E2 levels measured from DUTCH (four dried samples collected throughout the day) represent a full, 24-hour collection
  • EVIDENCE – Strong correlation when comparing liquid and dried urine samples⁶
  • EVIDENCE – Strong correlation to 24-hour collections from same-day collections⁶
• CLAIM – Urine estrogen measurements provide additional information (the value of estrogen metabolites will not be defended in this article) thanks to the inclusion of metabolite measurements
  • EVIDENCE – Validation data above for DUTCH E2 measurements should also be completed for all measured metabolites^[7],[8]

The above claims of serum and 24-hour urine correlation are supported by evidence now published in peer-reviewed research (see figures 1 and 2).^6,7,8 This allows providers to confidently use the DUTCH test as a viable alternative to the gold standard serum test. An additional peer-reviewed publication, including complete validation data for our entire estrogen metabolite panel, was also published recently.⁷ In the Functional Lab industry, tests should be supported by more than anecdotal evidence or an unproven clinical hypothesis. Towards the end of this article, I highlighted an example of a urine marker (HMG) that serves as an example of a test in need of more proof to support the claims of its potential clinical utility.

Fig. 1 Hormone profiles of serum progesterone versus urinary β-pregnanediol (a) and serum versus urinary estradiol (b) in one premenopausal woman’s cycle. Metabolites of subject 2. Cr, creatinine; βPg, β-pregnanediol.⁶

Fig. 2 Interclass correlations of 24-h urine collections for β-pregnanediol (a), estradiol (b), versus the 4-spot assay. Correlation coefficients reported are Spearman correlations. Cr, creatinine; βPg, 5β-pregnanediol.⁶

QUESTION #2 – What is the next level of evidence to further increase confidence in clinical utility?

What should be our collective expectations of Functional Labs? Precision Analytical has now published validation data for all hormone metabolites and organic acids.⁷ Data has also been published showing correlation between saliva (gold standard) and urine free cortisol patterns.⁸ We are proud of these peer-reviewed publications showcasing the analytical and clinical veracity of our testing, but we cannot be complacent.

Over the past few years Precision Analytical has made it a priority to publish data from our analytical and clinical validation studies in peer-reviewed literature. This is just a starting point. Continuing with the theme of focusing on our E2 measurements, what are the next steps for this clinical exploratory process? There is much more to come!

We have recently submitted data for peer review showing the DUTCH test’s clinical utility as it relates to monitoring estradiol therapy. Do E2 levels in urine scale up as the dosage of E2 patches, gels, and creams increase? How about E2 metabolite levels? Do similarly impactful products from gels and patches (creams have never been tested by research) result in similar urine levels? The affirmative answers to these questions, especially once published, increase the confidence in using this test to monitor estrogen replacement therapy (ERT). This may give readers a window into the continuous process of evidence pursuit that we believe Functional Labs should be engaged in.

This pursuit of evidence should never end, and some of these efforts do not always affirm our testing. The most difficult task asked of our industry leaders is to pursue truth regardless of where it takes us. For example, evidence implies that oral and sublingual E2 preparations do not result in urine concentrations that are clinically useful. Swallowed E2 floods the urine with metabolites that do not reflect serum or systemic exposure. Our job is not to “sell” our testing to all providers in all scenarios. The focus of our scientists and clinicians is to promote proper and optimal testing in every situation, regardless of what type of testing that may be.

The above is an exploration of our pursuit of evidence-led best practices for just one test from our Dried Urine Test for Comprehensive Hormones (DUTCH). For our publication efforts, data related to E2 and cortisol has led the charge. Subsequently, we have published the validation data for all our DUTCH hormone metabolites and organic acids,^7,8 and we have already mined some exciting clinical correlations being readied for peer-reviewed publication for many of the markers on our profiles.

Existing literature creates the basis for clinical validation for much of the DUTCH Test:

Methylmalonate (MMA), xanthurenate (xan), and pyroglutamate (pyro) are found on the DUTCH test thanks to their ability to indicate a nutrient deficiency (elevated MMA may indicate B12 deficiency, xan for B6, and pyro for glutathione deficiency). In fairness, the literature supporting the use of pyroglutamate as a glutathione deficiency marker is a little thin.^[1],[2],[3] More confidence can be found in the literature on MMA^{[4],[5],[6],[7]} and xan.^[8],[9]

­­For each of these analytes, our job is simply to create and publish basic validation data (seen here in peer-reviewed literature) to ensure confident use. We intend to publish research that highlights the clinical utility of these tests as well, but some utility is already outlined in published research. One of the reasons we have resisted adding some additional organic acids to our panels is that some lack any published research to verify that they carry any clinical significance. I have outlined one such case below:

Hydroxymethylglutarate (HMG) – a marker for CoQ10 deficiency?

Given that so many individuals explore hormone profiles due to issue with fatigue and decreased energy, a marker for CoQ10 deficiency might be a nice addition to the DUTCH test. HMG has been proposed as an organic acid that may indicate a need for additional CoQ10. Why?

• HMG sits upstream of CoQ10 in its biochemical synthesis (see graph below).
• HMG cannot move downstream towards CoQ10 if statins (which inhibit the body’s ability to move HMG downstream to the next biochemical step in cholesterol and CoQ10 synthesis) are taken.
• When statins are taken, CoQ10 deficiency is a known consequence and urinary HMG levels are thought to increase because its metabolism is chemically blocked.

Because of the known relationships above, it is presumed that urinary HMG excretion may increase if statins are taken. I have witnessed this phenomenon personally, and you can see it detailed anecdotally in this case study.^[1] This potential relationship has been used and extrapolated to HMG’s use as a marker of CoQ10 deficiency IN THE ABSENCE OF STATINS. Is this appropriate? A hormone analogy might make this easier to understand. In the presence of an aromatase inhibitor, testosterone’s conversion to estradiol is blocked. Testosterone levels increase slightly, and estradiol levels drop dramatically. That makes sense. In a healthy patient that is NOT taking an aromatase inhibitor, does an elevated testosterone mean that downstream estradiol is low? We would not draw that conclusion without research to tell us this connection holds true. 

Depending on which biochemical pathway you are examining (some show more intermediates than others), CoQ10 appears about 10 steps away from HMG. In the absence of a statin, there is no evidence that abnormal HMG has been correlated to CoQ10 deficiency. A biochemical association between HMG and CoQ10 has led to the belief that HMG acts as a surrogate marker for CoQ10 deficiency. As far as I can tell, this is a theoretical association that has yet to be confirmed.

Concurrent measurements of urine HMG and serum CoQ10 could clarify HMG’s potential role as a biomarker of nutrient deficiency. To date this has not been done. I don’t mean to pick on those using urine HMG testing. Years ago, I helped develop and commercialize this very same test, and we allowed clients to continue to buy this biological assumption without offering more proof than the same list of references other labs used.^{18,[1],[2],[3]} These references explain that statins lower CoQ10 levels, but none connected a COQ10 deficiency with elevated urinary HMG excretion, with or without statin use. This test has been offered as evidence of CoQ10 deficiency for more than 30 years, but more work needs to be done before using this marker with confidence. To the best of my knowledge, the evidence evaluating urinary HMG excretion has been reported in individuals with inborn errors of metabolism,^[4] not statin use or CoQ10 deficient individuals.

As Precision Analytical explores future biomarkers, we want to stay in the “evidence-based” lane. This doesn’t mean our testing is comprehensively validated for all clinical associations. For E2 we are off to a great start, but there is much more work to be done. Our team has been busy exploring clinical correlations with every biomarker on the DUTCH panel. In some cases, these relationships are established in the literature, and our goal is to confirm the reality of these relationships in our testing. In other cases, clinical validity may remain uncertain until truth emerges from exploring the data. Research, scientific writing, and peer-reviewed publications are tedious work and require a long-term commitment, which we continue to make.

As Functional Medicine moves forward, we should encourage provable claims to be put to the test. We intend to continue investing in research to illuminate where theoretical associations stand up to scrutiny and where they wilt under the light of empirical evidence. As an industry, we should not just follow the evidence where it goes, we must pursue it!

[1] Notelovitz M, et al. Effectiveness of Alora estradiol matrix transdermal delivery system in improving lumbar bone mineral density in healthy, postmenopausal women. Menopause. 2002; 9(5): 343-353. doi:10.1097/00042192-200209000-00007.

[2] Cauley JA. Estrogen and bone health in men and women. Steroids. 2015; 99(Pt A): 11-5. doi: 10.1016/j.steroids.2014.12.010.

[3] Karlamangla AS, et al. Bone Health during the Menopause Transition and Beyond. Obstet Gynecol Clin North Am. 2018; 45(4): 695-708. doi: 10.1016/j.ogc.2018.07.012.

[4] Hadji P, et al. Bone health in estrogen-free contraception. Osteoporos Int. 2019; 30(12): 2391-2400. doi: 10.1007/s00198-019-05103-6.

[5] Filicori M, et al. Neuroendocrine regulation of the corpus luteum in the human. Evidence for pulsatile progesterone secretion. J Clin Invest. 1984; 73(6): 1638-1647. doi: 10.1172/JCI111370.

[6] Newman M, et al. Evaluating urinary estrogen and progesterone metabolites using dried filter paper samples and gas chromatography with tandem mass spectrometry (GC-MS/MS). BMJ Chem. 2019; 13(1): 20. doi: 10.1186/s13065-019-0539-1.

[7] Newman M, Curran DA. Reliability of a dried urine test for comprehensive assessment of urine hormones and metabolites. BMC Chem. 2021; 15(1): 18. doi: 10.1186/s13065-021-00744-3.

[8] Newman M, et al. Dried urine and salivary profiling for complete assessment of cortisol and cortisol metabolites. J Clin Transl Endocrinol. 2020; 22: 100243. doi: 10.1016/j.jcte.2020.100243.

[9] Lord R. Long-term patterns of urinary pyroglutamic acid in healthy humans. Physiol Rep. 2016; 4(4): e12706. doi: 10.14814/phy2.12706.

[10] Hundemer GL, Fenves AZ. Acquired 5-oxoproline acidemia successfully treated with N-acetylcysteine. Proc (Bayl Univ Med Cent). 2017; 30(2): 169-170. doi: 10.1080/08998280.2017.11929570.

[11] Yu YM, et al. Plasma L-5-oxoproline kinetics and whole blood glutathione synthesis rates in severely burned adult humans. Am J Physiol Endocrinol Metab. 2002; 282(2): E247=258. doi: 10.1152/ajpendo.00206.2001.

[12] Kwok T, et al. Use if fasting urinary methylmalonic acid to screen for metabolic vitamin B12 deficiency in older persons. Nutrition. 2004; 20(9): 764-768. doi: 10.1016/j.nut.2004.06.001.

[13] Gultepe M, et al. Urine methylmalonic acid measurements for the assessment of cobalamin deficiency related to neuropsychiatric disorders. Clin Biochem. 2003; 356(4): 275-282. Doi: 10.1016/s0009-9120(03)00033-x.

[14] Ward MG, et al. Prevalence and Rsik Factors for Functional Vitamin B12 Deficiency in Patients with Crohn’s Disease. Inflamm Dis. 2015; 21(12): 2839-2847. doi: 10.1097/MIB.0000000000000559.

[15] Klee GG. Cobalamin and Folate Evaluation: Measurements of Methylmalonic Acid and Homocysteine vs Vitamin B12 and Folate. Clin Chem. 2000; 46(8 Pt 2): 1277-1283.

[16] Luhby AL. Vitamin B6 metabolism in users of oral contraceptive agents. I. Abnormal urinary xanthurenic acid excretion and its correction by pyridoxine. Am J Clin Nutr. 1971; 24(6): 684-693. doi: 10.1093/ajcn/24.6.684.

[17] Chen C-F, et al. Novel spectrophotometric method for the quantitation of urinary xanthurenic acid and its application in identifying individuals with hyperhomocysteinemia associated with Vitamin B6 deficiency. Biomed Res Int. 2013; 2013: 678476. Doi: 10.1155/2013/678476.

[18] Fitzgerald K, et al. Statin-induced Myopathy. Global Adv Health Med. 2012; 1(2): 32-36.

[19] Folkers K, et al. Lovastatin decreases coenzyme Q levels in humans. Proc Natl Acad Sci USA. 1990; 87(22): 8931-8934. doi: 1073/pnas.87.22.8931.

[20] Hernandez-Camacho JD, et al. Coenzyme Q10 Supplementation in Aging and Disease. Front Physiol. 2018; 9: 44. doi: 10.3389/fphys.2018.00044.

[21] Qu H, et al. Effects of Coenzyme Q10 on Statin-induced Myopathy: An Updated Meta-Analysis of Randomized Controlled Trials. J Am Heart Assoc. 2018; 7(19): e009835. Doi: 10.1161/JAHA.118.009835.

[22] Wortman SB, et al. The 3-methylglutaconic acidurias: what’s new? J Inherit Metab Dis. 2012; 35(1): 13-22. doi: 10.1007/s10545-010-9210-7.

Mark Newman

Mark Newman, MS, CEO is a recognized expert and international speaker in the field of hormone testing. Mark has spent nearly 20 years within specialty laboratories, developing and directing 24-hour urine hormone testing, organic acid testing and salivary hormone testing, gaining a fairly unique and thorough perspective on blood, urine and saliva hormone testing. This unique experience led to a vision for a revolutionary way to test hormones; Mark began his own lab, Precision Analytical, Inc., and is the creator of DUTCH (Dried Urine Test for Comprehensive Hormones). The question of how to best test hormones is what drove the creative process that initiated this lab. Blood (serum), urine, and salivary testing all have significant limitations. DUTCH has unique testing methods which bridge the gap between existing methods to create better tools for healthcare practitioners to address the needs of their patients.

Mark is committed to advancing innovations in hormone testing and helping providers find answers and treatment in their hormone testing. Mark has educated thousands of physicians on the different hormone tests available and best practices, especially in HRT/BHRT monitoring.