Stability of Varenicline Concentration in Saliva Over 21 Days at Three Storage Temperatures

Abstract

Introduction

Varenicline is the most efficacious drug for smoking cessation; saliva varenicline concentrations can be useful for the evaluation of adherence in smoking cessation trials. Saliva is a useful noninvasive matrix for mail-in specimen collection, if stable. We investigated the stability of varenicline in saliva at different storage temperatures simulating the time it takes to mail in a sample.

Methods

We evaluated the concentrations of varenicline, nicotine, cotinine, 3′-hydroxycotinine, and 3′-hydroxycotinine/cotinine (3HC/COT) ratio in quality control saliva samples (and after repeated freezing and thawing), and in smokers’ saliva samples, stored for up to 21 days at room temperature (~25°C), 4°C, and −80°C.

Results

In saliva quality control samples, concentrations of varenicline, nicotine, cotinine, 3′-hydroxycotinine, and 3HC/COT remained unchanged and showed little within-sample variation (CV ≤ 5.5%) for up to 21 days at the three storage temperatures; they were also not altered after three thaw-freeze cycles. In smokers’ saliva, a significant main effect of storage duration, but not temperature, was observed for varenicline, cotinine, and 3′-hydroxycotinine, but not for nicotine or the 3HC/COT ratio. However, these changes were within analytical (i.e., equipment) variation resulting in little within-sample variation (CV ≤ 5.8%) for all analytes in smokers’ saliva.

Conclusions

Varenicline, the other analytes, and the 3HC/COT ratio remained stable in saliva during storage for 21 days at all temperatures tested and after repeated freezing and thawing with only minor changes in concentration over time. These findings support the potential use of mail-in approach for saliva samples in varenicline smoking cessation clinical trials.

Implications

Assessing saliva varenicline concentrations can be useful for the evaluation of adherence in smoking cessation trials. Saliva is a noninvasive matrix suitable for mail-in specimen collection. This is the first investigation of stability of varenicline in saliva. Varenicline, nicotine, cotinine, 3′-hydroxycotinine, and 3HC/COT were stable in saliva for up to 21 days at room temperature (~25°C), 4°C, and −80°C, supporting the use of a mail-in approach for saliva specimen in smoking cessation trials.

Introduction

Tobacco smoking remains a main preventable causes of morbidity and mortality worldwide.¹ Nicotine is the main psychoactive compound in cigarette smoke, mediating its reinforcing effects through neuronal α4β2 nicotinic receptors.² Varenicline, a partial agonist at the α4ß2 nicotinic receptor subtype, is the most efficacious monotherapy for smoking cessation.^3,4

Varenicline is an excellent candidate for therapeutic drug monitoring as it is readily absorbed after an oral dose (i.e., has high bioavailability), exhibits low (≤20%) protein binding, has a long elimination half-life (~24 hours), and is excreted almost exclusively unchanged in urine.⁵ Compared with biological measures, such as plasma varenicline concentrations, self-reported pill-counts overestimate varenicline adherence in smoking cessation trials.^6–8 Individuals with biologically verified adherence to varenicline are more likely to abstain from smoking during, and following, varenicline treatment than those who are not.^9,10

Saliva, relative to blood sampling, is useful for drug monitoring due to its noninvasive collection and because it does not require a phlebotomist or trained support. Of note, plasma varenicline and salivary varenicline are highly correlated (Spearman’s rho = 0.77, p < .001).¹¹

Saliva can be sampled repeatedly and collected by study participants offsite.¹² For example, salivary cotinine concentration is used for smoking status verification,^13,14 and the salivary 3′-hydroxycotinine to cotinine (i.e., 3HC/COT) ratio is used as a biomarker of nicotine metabolism rate.¹⁵ Clinical trials are increasingly utilizing mail-based approaches for salivary sample specimen collection (e.g., abstinence verification, 3HC/COT),^13,16 particularly during the COVID-19 pandemic.¹⁷ The stability of the nicotine metabolite ratio (i.e., 3HC/COT) over different temperatures and storage durations has been demonstrated in saliva and plasma.¹⁸ However, the stability of varenicline in saliva samples has not been evaluated. To gain a better understanding of how differing storage conditions may affect salivary varenicline concentrations, we investigated the stability of varenicline in saliva quality control (QC) samples stored at different temperatures for up to 21 days, simulating the time it might take to mail in a sample. Additionally, we studied the effect of repeated thawing and freezing of saliva on the stability of varenicline concentrations. Lastly, we assessed varenicline, in smoker saliva samples over these same storage conditions. In addition to varenicline, we evaluated concentrations of nicotine, cotinine, and 3′-hydroxycotinine as well as the stability of 3HC/COT in each sample, to ensure that the presence and measurement of these other analytes have no additional impact on varenicline stability.

Methods

Materials

Varenicline tartrate, nicotine, cotinine, and trans-3′-hydroxycotinine were obtained from Sigma Aldrich (St. Louis, MO). Internal standards (varenicline-d4 dihydrochloride, nicotine-d4, cotinine-d3, and trans-3′-hydroxycotinine-d3) were purchased from Toronto Research Chemicals (Toronto, ON, Canada). Saliva samples were collected using Salivette absorbent device without preparation from Sarstedt (Nümbrecht, Germany).

Study Design

Study 1

The stability of varenicline, nicotine, cotinine, and 3′-hydroxycotinine was determined in saliva quality control (QC) samples by analyzing five replicates of nonsmoker saliva spiked with varenicline (10 ng/mL), nicotine (5 ng/mL), cotinine (100 ng/mL), and 3′-hydroxycotinine (30 ng/mL), based on average salivary week 1 concentrations from a previous clinical trial.¹⁰ The concentration detected in the nonsmoker samples spiked with known amounts of NIC, COT, 3HC, and varenicline were as expected. Prepared QC samples were aliquoted into separate vials. Initial concentrations of all analytes were determined by liquid chromatography-tandem mass spectroscopy (LC–MS/MS) analysis (day 0) and the remaining aliquots were stored for 1, 3, 5, 7, and 21 days at room temperature (~25°C), 4°C, and −80°C. One aliquot per QC replicate and storage temperature (i.e., n = 5 each for each of room temperature, 4°C, and −80°C) was analyzed by LC–MS/MS (Supplementary Figure S1).

Study 2

The effect of repeated freezing and thawing of a separate set of identical QC samples was examined after three cycles. Each freeze–thaw cycle consisted of a minimum of 24 h frozen at −80°C followed by a complete thaw at room temperature. The samples were analyzed after the third freeze–thaw cycle by LC–MS/MS.

Study 3

Stability was also evaluated using previously frozen saliva samples collected from smokers enrolled in a varenicline smoking clinical trial (NCT03262662).¹⁹ Written informed consent was obtained from each participant and study procedures were approved by Institutional Review Boards at the University of Toronto and at the University at Buffalo. Five pooled (in order to obtain sufficient volume) samples were prepared by combining ten individual samples for each pool. The pooled saliva samples from smokers were aliquoted into separate vials, and the analyte concentrations were determined by LC–MS/MS. The storage temperature conditions and storage durations were as described for Study 1.

Analysis of Analytes

The concentrations of free (nonconjugated) varenicline, nicotine, cotinine, and 3′-hydroxycotinine in all salivary samples were quantified by LC–MS/MS using a previously published method.^8,11 The ratio of salivary 3HC (ng/mL)/COT (ng/mL) was derived as before.¹⁵

Statistical Analysis

The change in analyte concentrations and 3HC/COT were assessed as a function of storage duration and storage temperature. For studies 1 (QC study) and 3 (smokers’ saliva), analyte concentrations and 3HC/COT were tested with a two-way (storage duration and storage temperature) repeated-measures ANOVA followed by a Bonferroni post hoc analysis. Individual analytes, and the 3HC/COT ratio, were log transformed and tested for normality, prior to statistical analyses.

We also computed the within-sample coefficients of variation (CV) for all analytes and 3HC/COT for Study 1 (QC study) and Study 3 (smoker saliva) on nontransformed data. For CV calculations, 18 concentrations per analyte and 3HC/COT measurements were included (i.e., six storage durations [0, 1, 3, 5, 7, 21 days] for each of the three storage temperatures [25°C, 4°C, −80°C]).

For study 2, we used paired T-tests (two-tailed) for analyzing differences in analyte concentrations and 3HC/COT before and after three thaw-freeze cycles. Individual analytes, and 3HC/COT, were tested for normality and log-transformed prior to statistical analyses. Additionally, we calculated between-sample CV for all analytes and 3HC/COT on nontransformed data; 10 concentrations per analyte were included (before and after three thaw-freeze cycles times five QC samples).

Analyses were performed using GraphPad Prism version 5.00 (GraphPad Software, San Diego, CA). A value of p < .05 was regarded as statistically significant.

Results

Study 1

In QC samples, the concentration of varenicline (ng/mL) and 3HC/COT (mean ± SE) over 21 days at three temperatures are presented in Figure 1A and B, respectively. There were no main effects of storage duration or temperature on analyte concentrations or 3HC/COT: varenicline F_time (5, 60) = 2.10, p = .078 and F_temperature (2, 60) = 2.12, p = .162; nicotine F_time (5, 60) = 0.74, p = .593, and F_temperature (2, 60) = 0.55, p = .589; cotinine F_time (5, 60) = 1.30, p = .275, and F_temperature (2, 60) = 1.95, p = .185; and 3′-hydroxycotinine F_time (5, 60) = 0.80, p = .556, and F_temperature (2, 60) = 1.21, p = .332 or 3HC/COT F_time(5, 60) = 0.70, p = .629, and F_temperature (2, 60) = 0.66, p = .533. The initial concentrations (mean ± SE, n = 5) of varenicline, nicotine, cotinine, 3′-hydroxycotinine, and 3HC/COT for the five QC samples, and percent change from initial concentration (mean ± SE, n = 5), after storage for up to 21 days at the three temperatures are presented in Supplementary Table S1. The within-sample CV ranges (mean ± SE) were 2.6%–3.7% (3.2% ± 0.22%) for varenicline, 1.6%–5.5% (3.6% ± 0.74%) for nicotine, 1.2%–1.7% (1.4% ± 0.09%) for cotinine, 1.6%–2.5% (2.1% ± 0.17%) for 3′-hydroxycotinine, and 1.8–2.8% (2.2% ± 0.17%) for 3HC/COT.

Figure 1.

Varenicline (VAR) and 3HC/COT in QC saliva samples (Study 1) after storage for up to 21 days at room temperature (RT), 4°C, and −80°C, showing mean ± SE (n = 5) for non-log₁₀-transformed values. Due to low SE for varenicline and 3HC/COT, the error bars are not visible on the graph.

Study 2

There were no significant changes in concentrations of varenicline (p = .234), nicotine (p = 1.000), cotinine (p = .693), 3HC (p = .139), or 3HC/COT (p = .322) in the separate set of QC samples after three thaw-freeze cycles compared with the initial values. The between-sample CV were 2.6% for varenicline, 6.5% for nicotine, 1.1% for cotinine, 2.1% for 3′-hydroxycotinine and 2.3% for 3HC/COT.

Study 3

For the pooled smoker saliva samples, the concentrations of varenicline and 3HC/COT after storage for up to 21 days at the three temperatures are illustrated in Figure 2A–C (varenicline) and Figure 2D–F (3HC/COT). There was a significant main effect of storage duration, but not temperature, on varenicline, cotinine, and 3′-hydroxycotinine: varenicline F_time (5, 60) = 3.81, p = .005, and F_temperature (2, 60) = 0.02 p = .985; cotinine F_time (5, 60) = 3.58, p = .007, and F_temperature (2, 60) = 0.00, p = 1.000, and 3HC F_time (5,60) = 2.62, p = .033, and F_temperature (2, 60) = 0.00, p = 1.00. There were no main effects of storage duration or temperature for 3HC/COT (F_time (5, 60) = 1.92, p = .104, and F_temperature (2, 60) = 0.00, p = 1.000), and nicotine (F_time (5,60) = 0.45, p = .813 and F_temperature (2, 60) = 0.00, p = 1.00).

Figure 2.

Varenicline (VAR, ng/mL, A–C) and 3HC/COT (D–F) in five separate pooled smokers’ saliva samples (Study 3) stored for up to 21 days at room temperature (RT), 4°C, and −80°C. Mean ± SE (n = 6) 0–21 days for each pooled sample is plotted for non-log₁₀-transformed values. Due to low SE for varenicline and 3HC/COT, the error bars are not visible on graphs.

The initial concentrations (mean ± SE, n = 5) of varenicline, nicotine, cotinine, 3′-hydroxycotinine, and 3HC/COT for each of the five pooled smoker saliva samples, and percent change from initial values (mean ± SE, n = 5), after storage for up to 21 days at the three temperatures, are presented in Supplementary Table S2.

The within-sample CV ranges (mean ± SE) were 2.8%–5.8% (4.6% ± 0.50%) for varenicline, 1.0%–5.6% (2.8% ± 0.79%) for nicotine, 1.7%–3.3% (2.3% ± 0.37%) for cotinine, 1.8%–4.5% (2.5% ± 0.50%) for 3′-hydroxycotinine, and 2.0%–4.3% (2.5% ± 0.45%) for 3HC/COT.

Discussion

In this study, we tested if the salivary concentrations of varenicline, nicotine, cotinine, and 3′-hydroxycotinine, as well as 3HC/COT, were affected by storage over 21 days at room temperature (~25°C), refrigerator temperature (4°C), and freezer temperature (−80°C). To our knowledge, this is the first study examining the stability of varenicline in saliva. The concentrations of varenicline, nicotine, cotinine, 3′-hydroxycotinine, and 3HC/COT were stable after storage for up to 21 days across the three temperatures tested in the QC samples (no effect of storage time or temperature was found, all p > .05). In addition, we determined that varenicline, nicotine, cotinine, 3′-hydroxycotinine, and 3HC/COT are stable in saliva over several freeze-thaw cycles (p > .05). There was no effect of temperature, but a small significant effect of storage duration on smokers’ saliva samples for varenicline, cotinine, and 3′-hydroxycotinine; this effect was minor with an average range in CV of 2.3%–4.6% for the six measures. Additionally, on nonlogged varenicline data for study 1 (normally distributed) and study 3 (not normally distributed), using a repeated-measures ANOVA and a nonparametric test (Freidman test) respectively, there was no main effect of storage time or temperature (p > .05). This minor variation in analyte concentration is unlikely to be clinically significant, and is within the range of the methodological (LC/LC–MS) % CV. In addition, there was no consistent pattern of effect (i.e., increase or decrease) on analyte concentrations in either the QC or smokers’ saliva samples (Figures 1 and 2), further suggesting that longer storage durations may be stable.

The stability of cotinine, 3′-hydroxycotinine, and 3HC/COT has been previously studied in whole blood, plasma, and saliva samples stored at room temperature and 4°C for up to 14 days, where small but significant changes in concentrations were reported for saliva samples.¹⁸ Our findings are concordant with this study; we also report relatively small changes in concentrations of cotinine and 3′-hydroxycotinine and 3HC/COT over the storage durations tested (up to 21 days) for the three temperatures. We expand on previous findings by reporting a negligible change by storage duration or temperature on varenicline concentrations.

To further investigate the magnitude of change in concentrations over a 21-day storage duration, we computed CV for each sample across the three storage temperatures. The CV for all analytes and 3HC/COT were below 6% for QC and smokers’ saliva samples, indicative of very small changes in concentration (or 3HC/COT). These values are similar to intraday CV for varenicline in saliva (1.7%–5.8%)⁸ for the LC–MS/MS method used for analyzing all samples for this study, and in range of CVs for cotinine (8.7%), 3′-hydroxycotinine (10.7%), and 3HC/COT (8.2%) established over several years for an analogous LC–MS/MS method.¹⁸ Of note, the CVs for all analytes and 3HC/COT were slightly higher for smokers’ saliva samples compared with the saliva QC samples, but again both were within the range of LC–LC/MS analytical variation.

In conclusion, our study shows that varenicline, nicotine, cotinine, 3′-hydroxycotinine, and 3HC/COT were stable in saliva with only slight changes in concentration over 21 days at room temperature, refrigerator (4°C), and freezer (−80°C) storage conditions. Furthermore, we demonstrated that repeated freezing and thawing of saliva samples does not alter the analyte concentrations or 3HC/COT. These findings support the potential use of a mail-in approach for saliva samples in varenicline smoking cessation clinical trials or for other applications, as varenicline, 3HC/COT, and the other analytes are stable for the average duration of mailing time over a wide range of temperatures. This will reduce the necessity of placing the newly acquired saliva samples at −80°C storage immediately upon collection and shipping on dry ice, increasing the feasibility of this approach.

Supplementary Material

A Contributorship Form detailing each author’s specific involvement with this content, as well as any supplementary data, are available online at https://academic.oup.com/ntr.

Funding

This research was undertaken, in part, thanks to funding from the Canada Research Chairs program (Dr. Tyndale, the Canada Research Chair in Pharmacogenomics), the Canadian Institutes of Health Research (Foundation grant FDN-154294), the Centre for Addiction and Mental Health (CAMH) and the CAMH Foundation, and the US National Institutes of Health (1R01 206193 and 2 UL1 TR001412). Pfizer provided free study medication and placebo for the parent clinical trial (samples used in Study 3) but had no role in the design, conduct, analysis, interpretation, or dissemination of the research.

Declaration of Interests

RF Tyndale has consulted for Quinn Emanuel and Ethismos Research Inc on unrelated topics; the other authors have no conflicts of interest.

Data Availability

The data underlying this article are available in the article and in its online supplementary material.