In Reply

We appreciate the constructive feedback and opportunity to respond to the comments and suggestions (1) on our recent breath cannabinoid publication (2).

In response to the question about positive admission breath specimens, Δ⁹-tetrahydrocannabinol (THC)-positive¹ breath from frequent smoking participants I and K at admission was likely the result of cannabis smoking just before arrival on our closed research unit. Our study design involved admission of participants to a secure research unit the night before smoking (16–20 h before smoking), to ensure lack of intoxication at the time of controlled dosing. Participants I and K were daily smokers who reported smoking 2.4 and 1.2 h before admission, respectively (3). These were among the shortest times reported between last smoke and admission (3). These participants also were the only 2 with blood THC concentrations at baseline and after 30 h postdose of >5 μg/L, the current per se limit in Washington state for driving under the influence of cannabis. These data were reported in a more recent paper regarding these same participants' blood and plasma cannabinoid concentrations (3). As noted, all participants provided breath that was cannabinoid-negative 1 h before smoking. The reason that these 2 participants' breath samples were THC-positive was that they smoked just before and not after admission.

We wish to stress that breath THC, 11-nor-9-carboxy-THC (THCCOOH), and cannabinol were identified and quantified with a highly specific LC-MS/MS assay. Therefore, we respectfully disagree with the comment that our results are due to possible laboratory interferences. It is true that THCCOOH is present at higher concentrations and for longer durations in frequent smokers' blood (3), plasma (3), urine (4), and oral fluid (5) compared with occasional smokers. We have presented these data after almost a decade of researching the differences between frequent and occasional cannabis smokers. However, THC and THCCOOH are easily distinguished by LC-MS/MS and GC-MS. Breath THC-positivity did not result from THCCOOH presence in breath. We monitored THCCOOH in breath independently and identified no positive breath samples. Specificity was assessed by relative retention time with matched deuterated internal standards, precursor and fragment ions, and the ratio of 2 fragment ions. Two unique multiple reaction monitoring transitions and the ion ratio between them accurately identified each analyte. This highly specific methodology does not have the cross-reactivity problems observed with immunoassays. We also tested method specificity against high concentrations of 93 potential exogenous interferences; none interfered with the accurate quantification of our low-concentration quality control.

In response to the question about frequent smokers having higher breath THC concentrations, cannabis smokers titrate their dose during smoking. They modulate puff frequency, inhalation hold times and volume, and smoking typography to achieve their own level of cardiovascular and subjective response. Frequent smokers are often more comfortable at higher THC blood concentrations than occasional smokers due to the development of some level of tolerance. Smoking typography in frequent smokers likely results in higher THC breath concentrations. Despite controlled administration of cigarettes with the same THC concentration, participants smoke ad libitum over 10 min. The abilities to titrate dose and the rapid delivery of drug to the brain during smoking are 2 of the primary reasons that smoking is such an important route of drug administration.

We agree that the lack of a placebo control is a limitation, and almost all of our studies include a within-subject placebo smoking session; we will include a placebo group in future studies. To the suggestion made about a nonsmoker control group, it is deemed unethical to administer cannabis to non-users; we are not permitted to do so, nor do we intend to do so.

We agree that psychomotor impairment data collection is of utmost importance, especially in light of recent cannabinoid legalization and medical marijuana legislation. Psychomotor, cognitive, and subjective data from the same participants in the present study are currently in preparation.

In conclusion, we thank the authors for their input through this timely letter and their encouragement to continue this line of research. We agree that these THC breath concentrations and detection windows following controlled cannabis administration provide a solid foundation for industry improvements, potentially leading over time to a roadside cannabinoid breath test that reflects psychomotor and cognitive impairment.