Medical Policy

This document addresses the use of saliva-based testing to determine drug-metabolizer status. This document does not address the collection of saliva for genetic testing (for example, Genecept^™ Assay).

Note: This document does not address genotype testing or other multi-gene testing for polymorphisms to determine drug-metabolizer status. Criteria for genetic testing for polymorphisms to determine drug-metabolizer status are found in applicable guidelines used by the plan.

Note: For a high-level overview of this document, please see “Summary for Members and Families” below.

Investigational and Not Medically Necessary:

Saliva-based testing to determine drug-metabolizer status is considered investigational and not medically necessary for all indications.

This document describes clinical studies and expert recommendations, and explains whether saliva-based testing to determine drug-metabolizer status is appropriate. The following summary does not replace the medical necessity criteria or other information in this document. The summary may not contain all of the relevant criteria or information. This summary is not medical advice. Please check with your healthcare provider for any advice about your health.

Key Information

Saliva-based testing to find out how someone’s body breaks down certain medicines is being studied but is not yet proven to help people stay healthier. These tests work by checking how a person’s body processes small amounts of certain drugs (called probes) by measuring them and their breakdown products in saliva. This helps doctors understand whether a person breaks down medicines slowly, quickly, or somewhere in between, and might help guide medicine choice or dose. The test involves giving a mix of these drugs and collecting saliva after a few hours. While saliva is easier to collect than blood or urine, this testing method still needs more research. At this time, saliva-based drug metabolism testing is not recommended for medical use.

What the Studies Show

Studies show that using a combination of test drugs (called a “cocktail”) can help measure the activity of certain liver enzymes that affect how drugs are broken down. These include enzymes like CYP1A2, CYP2C9, and others. The “Geneva cocktail” was tested in healthy adults in several countries to see if it was safe. Most people had only mild side effects, such as dizziness or stomach upset, which went away on their own. Some studies also looked at how inflammation or illness like COVID-19 might change how these enzymes work. These studies found changes in enzyme activity during illness, but were small and need to be repeated in larger groups.

Some research shows that saliva samples may offer a way to check enzyme activity, but the results are mixed. For example, in one study, saliva testing helped tell the difference between people who metabolize drugs very slowly and others, but it could not tell apart people who metabolize at normal and fast rates. Another study showed that some substances were too low in saliva to measure properly. Most studies so far are small or look at healthy people, so it is still unclear how helpful this test is for people who need medicine for health conditions.

Is this clinically appropriate?

This test is not appropriate because it has not been proven to improve health. While some early studies show that saliva testing may provide similar results to blood or urine for certain enzymes, other studies found that saliva does not work well for all parts of the test. The results are not consistent, and the number of people studied is small. Better studies are needed to know if saliva-based drug metabolism testing improves health. Unnecessary or unproven tests can lead to needless worry or treatment that does not help.

(Return to Description/Scope)

Summary

Saliva-based testing for drug-metabolizer status is considered investigational and not medically necessary. Research demonstrates that cytochrome P450 (CYP450) enzyme activity influences individual drug responses, and metabolic phenotyping using probe-drug cocktails, such as the Geneva and Basel cocktails, can measure this activity through concentrations of substrates and metabolites in biological samples. While studies show that these cocktails are generally safe and can detect isoform-specific modulation of CYP activity during conditions like inflammation or SARS-CoV-2 infection, most evidence relies on blood or plasma samples. Limited studies evaluating saliva suggest some potential correlation with urinary and plasma measures, but results are inconsistent and insufficient to guide clinical decision-making.

Although saliva sampling offers a less invasive alternative, current evidence shows significant limitations. Some probe metabolites cannot be reliably quantified in saliva, and early studies demonstrate that saliva-based ratios may distinguish poor metabolizers but cannot consistently differentiate intermediate from extensive metabolizers. Overall, more robust research is needed across diverse populations to determine whether saliva reliably reflects CYP450 enzyme activity and leads to improved health outcomes. As a result, saliva-based metabolic phenotyping remains investigational and is not supported for clinical use at this time.

Discussion

Clinicians commonly encounter differences in how individuals respond to pharmacologic therapy. These differences are due in large part to variations in the cytochrome P450 (CYP450) family of enzymes which is responsible for oxidative drug metabolism in the liver. The activity levels of specific CYP450 enzymes (CYPs) in the liver can be assessed by measuring the concentrations of their particular drug substrates and metabolites present in saliva. Subjects can be given a cocktail of drugs to ingest that are substrates (“probes”) for a number of CYPs of interest. Subjects then submit a saliva sample and metabolites of these probes by CYPs are measured. The ratio of the probe concentration to the concentration of probe metabolite is considered to be the metabolic phenotype for a specific enzyme. Using this information, a subject’s drug metabolizer status can be determined. With knowledge of a subject’s metabolizer status, a clinician can personalize the selection and dosage of drug prescribed to maximize efficacy based on the ability of the individual’s body to break down the drug. The cocktail approach to metabolic phenotyping has been widely studied as it allows for the simultaneous in vivo assessment of multiple pathways of drug metabolism. Several phenotyping cocktails have been developed including the Geneva cocktail and the Basel cocktail.

In a 2020 study by Rollason and colleagues, the authors reported on the Geneva cocktail which is comprised of a specific probe for six different CYPs (CYP1A2, CYP2B6, CYP2C9, CYP2C19, CYP2D6 and CYP3A) and one for P-gp. In this study the Geneva cocktail was evaluated in a group of 265 healthy adult volunteers in three different populations from Ethiopia, Oman, and the Czech Republic. The purpose of the study was to report on the safety of the cocktail. The Geneva cocktail is made up of a combination of substances (caffeine, bupropion, flurbiprofen, dextromethorphan, midazolam, and fexofenadine) in a single capsule and an additional tablet of omeprazole. Participants received the cocktail and were tested at 2, 4, and 6hours following administration. Subjects reported adverse events and standard blood spots were taken for evaluation. There were four reported adverse events including 1) dizziness, lack of concentration and headache, 2) numbness of both hands and feet, 3) nausea and abdominal heaviness, and 4) a localized erythematous macular nonpruritic rash on both thighs. It was concluded that the adverse events using this low-dose phenotyping cocktail were mild to moderate and all resolved spontaneously.

In two studies from 2021, Lenoir and colleagues assessed the impact of acute inflammation and SARS-CoV-2 infection on CYP450 activity using the Geneva phenotyping cocktail. These were prospective observational studies conducted in 30 and 28 subjects, respectively. Hip surgery was the acute inflammation model (Lenoir, 2021a), and the probe drug cocktail was administered orally before surgery, day 1 and day 3 post surgery and at discharge. Capillary blood samples were collected 2 hours after cocktail intake to assess metabolic ratios (MRs) of the six different CYPs. It was determined that inflammation modulated activity of the CYPs in an isoform-specific manner, with different magnitudes and kinetics for each. In the study of SARS-CoV-2 (Lenoir, 2021b), individuals received the Geneva cocktail orally during the first 72 hours of hospitalization for COVID-19 and after 3 months, with blood samples collected 2 hours after cocktail administration to assess MRs of the same six CYPS. As in the previous study, CYP activity was modulated in an isoform-specific manner by SARS-CoV-2 infection. In both studies, the authors concluded that inflammation or SARS-CoV-2 infection could have a clinically relevant impact on the pharmacokinetics of CYP substrates. However, these were small studies and confirmation of multivariable statistical modeling findings within a larger sample size is needed, allowing for possible adjustment with other covariates.

While most published studies address phenotyping using blood and plasma samples, saliva sampling is also being evaluated in metabolic phenotyping studies involving several CYP450 enzymes. In a meta-analysis of four studies in which caffeine and its metabolite paraxanthine were measured in saliva, plasma and urine of 78 subjects, Fuhr and colleagues (1994) found that the paraxanthine/caffeine ratios in saliva and plasma were highly correlated and provided a better estimate of CYP1A2 activity than urine ratios.

Urine collection over an 8- to 10-hour period following ingestion of dextromethorphan is one process for metabolic phenotyping. However, sometimes urine collection may not be desirable, for example in those with renal disease, or in children. In 1991 Hou and colleagues evaluated the use of saliva for metabolic phenotyping of CYP2D6 in 62 healthy volunteers, with 61 subjects included in the final analysis. In this study, urine collections were also taken for comparison. All participants received one dose of dextromethorphan, then a 5 to 10 ml saliva sample was collected over a 2- to 5-minute period. Saliva was collected by the initial 12 subjects at 30 and 60 minutes for 8 hours after taking dextromethorphan. The next 12 subjects collected saliva samples at 2 and 3 hours after taking dextromethorphan. The last 38 subjects collected saliva at 3 hours following ingestion of dextromethorphan. They found that salivary dextromethorphan/dextrorphan ratios showed a significant correlation with urinary ratios and enabled differentiation of CYP2D6 poor metabolizers from intermediate and extensive metabolizers. However, intermediate and extensive metabolizers could not be separated on the basis of salivary results. Further study is necessary to determine improvement in net health outcomes.

In a randomized crossover study of 16 healthy volunteers, Donzelli and colleagues (2014) sought to develop another cocktail (the Basel cocktail) based on probe drugs that are widely used in clinical practice. The authors also sought to determine whether other sampling methods such as dried blood spots or saliva samples could be used to simplify the sampling process. The cocktail consisted of caffeine, efavirenz, losartan, omeprazole, metoprolol and midazolam in different combinations. The Basel cocktail was tested for simultaneous phenotyping of CYP1A2, CYP2B6, CYP2C9, CYP2C19, CYP2D6 and CYP3A4 in plasma, dried blood spots and saliva. In dried blood spots, only caffeine, paraxanthine, omeprazole, 5-OH-omeprazole, efavirenz, and midazolam were quantified. In saliva, all analytes except for caffeine, paraxanthine, omeprazole, 5-hydroxyomeprazole, and metoprolol were too low to quantify reliably. While use of dried blood spots and saliva samples seems feasible for phenotyping of selected CYP isoforms, further study is necessary to confirm these results, particularly in differing population groups.

Human CYP450 enzymes in the liver are responsible for metabolizing approximately 50% of commonly used drugs. The activity of CYP450 enzymes is highly variable between individuals and is determined by genetic variants, endogenous mediators and environmental influences such as nutrients or interacting medications. Differences in drug metabolism can lead to challenges in optimizing dosage for an individual patient. Variations in CYP450 enzyme metabolic activity can be associated either with drug toxicity if too little is metabolized or with an insufficient pharmacological effect if too much is metabolized. Genotyping can only assess genetic factors contributing to CYP450 enzyme activity variation while phenotyping can assess the net effect of all influencing factors on CYP450 activity. Metabolic phenotyping uses specific probe drugs to obtain information about real-time activity of CYP450 drug-metabolizing enzymes in an individual.

Cytochrome P450: Refers to a family of 60 different enzymes involved in drug and toxin metabolism.

Metabolize: Refers to breaking down a drug so that it is no longer clinically active.

Metabolite: Substance formed during the process of metabolism.

The following codes for treatments and procedures applicable to this document are included below for informational purposes. Inclusion or exclusion of a procedure, diagnosis or device code(s) does not constitute or imply member coverage or provider reimbursement policy. Please refer to the member's contract benefits in effect at the time of service to determine coverage or non-coverage of these services as it applies to an individual member.

When services are Investigational and Not Medically Necessary:
For the following procedure codes; or when the code describes a procedure indicated in the Position Statement section as investigational and not medically necessary.

Peer Reviewed Publications:

1. Derungs A, Donzelli M, Berger B, et al. Effects of cytochrome p450 inhibition and induction on the phenotyping metrics of the basel cocktail: a randomized crossover study. Clin Pharmacokinet. 2016; 55(1):79-91.
2. Donzelli M, Derungs A, Serratore M-G, et al. The basel cocktail for simultaneous phenotyping of human cytochrome P450 isoforms in plasma, saliva and dried blood spots. Clin Pharmacokinet. 2014; 53(3):271-282.
3. Fuhr U, Rost KL. Simple and reliable CYP1A2 phenotyping by the paraxanthine/caffeine ratio in plasma and in saliva. Pharmacogenetics.1994; 4(3):109-116.
4. Hou ZY, Pickle LW, Meyer PS, Woosley RL. Salivary analysis for determination of dextromethorphan metabolic phenotype. Clin Pharmacol Ther. 1991; 49(4):410-419.
5. Lenoir C, Daali Y, Rollason V, et al. Impact of acute inflammation on cytochromes P450 activity assessed by the Geneva cocktail. Clin Pharmacol Ther. 2021a; 109(6):1668-1676.
6. Lenoir C, Terrier J, Gloor Y, et al. Impact of SARS-CoV-2 infection (COVID-19) on cytochromes P450 activity assessed by the Geneva cocktail. Clin Pharmacol Ther. 2021b; 110(5):1358-1367.
7. Rollason V, Mouterde M, Daali Y, et al. Safety of the Geneva cocktail, a cytochrome P450 and P-glycoprotein phenotyping cocktail, in healthy volunteers from three different geographic origins. Drug Saf. 2020; 43(11):1181-1189.

Government Agency, Medical Society, and Other Authoritative Publications:

1. Centers for Medicare and Medicaid Services (CMS). Local Coverage Determination (LCD): MolDX: Pharmacogenomics Testing.
   • LCD L38294. Available at: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=38294&ver=24&keyword=Polymorphisms&keywordType=starts&areaId=all&docType=F&contractOption=all&sortBy=relevance&bc=1. Accessed on February 13, 2026.
   • LCD L38335. Available at: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=38335&ver=15&keyword=Polymorphisms&keywordType=starts&areaId=all&docType=F&contractOption=all&sortBy=relevance&bc=1. Accessed on February 13, 2026.
   • LCD L38337. Available at: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=38337&ver=15&keyword=Polymorphisms&keywordType=starts&areaId=all&docType=F&contractOption=all&sortBy=relevance&bc=1. Accessed on February 13, 2026.
   • LCD L38435. Available at: https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=38435&ver=15&keyword=Polymorphisms&keywordType=starts&areaId=all&docType=F&contractOption=all&sortBy=relevance&bc=1. Accessed on February 13, 2026.

Saliva-based testing

Federal and State law, as well as contract language, including definitions and specific contract provisions/exclusions, take precedence over Medical Policy and must be considered first in determining eligibility for coverage. The member’s contract benefits in effect on the date that services are rendered must be used. Medical Policy, which addresses medical efficacy, should be considered before utilizing medical opinion in adjudication. Medical technology is constantly evolving, and we reserve the right to review and update Medical Policy periodically.

No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, or otherwise, without permission from the health plan.

© CPT Only – American Medical Association