Pilot Data on Salivary Oxytocin as a Biomarker of LSD Response in Patients with Major Depressive Disorder

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript presents a promising pilot study investigating salivary oxytocin as a potential biomarker of LSD response in patients with treatment-resistant Major Depressive Disorder (MDD). Conducted at Geneva University Hospitals, the study included 12 patients undergoing LSD-assisted psychotherapy (100–150 µg), with salivary oxytocin data ultimately available for 10 participants. Saliva samples were collected at baseline, 60, 90, and 180 minutes post-LSD administration, alongside self-reported ratings of subjective drug intensity.

Key findings:

• Salivary oxytocin levels showed significant changes over time, peaking at 90 minutes post-LSD.
• Subjective psychedelic intensity also varied significantly, with the highest ratings at around 180 minutes.
• These findings support the potential of oxytocin as a pharmacodynamic biomarker in this context and lay important groundwork for future research exploring predictive links with clinical outcomes.

While the study’s small sample size, lack of control for sex hormones or menstrual cycle phase, continued antidepressant use, and absence of a control group are important considerations, these are acceptable given the pilot design. Notably, the study successfully demonstrates the feasibility and clinical integration of salivary oxytocin sampling in LSD-assisted psychotherapy, offering a valuable foundation for larger, controlled investigations.

1. While the manuscript describes this as an observational pilot study, it would be more precise to frame it as an observational pilot sub-study leveraging a convenience sample within routine clinical care. Clarifying this in the methods and discussion would better contextualize the sampling approach, its inherent limitations, and the generalizability of the findings.
2. Given the established influence of sex and age on oxytocin dynamics and to some degree psychedelic response, these variables should be included as covariates in the statistical models. Even if adding these covariates reduces statistical significance, the authors could provide the results of both models. Similarly, since the LSD dose differed between participants, why was this not used as a covariate?
3. If the inclusion of multiple covariates risks overfitting due to the small sample size, a supplementary model examining sex (or age) as a primary predictor of oxytocin levels would still be valuable. Reporting such exploratory analyses could provide important preliminary insights.
4. 
   Since both subjective effects and oxytocin levels were measured over time, it would be highly informative to examine their direct relationship. Correlating perceived intensity with oxytocin concentrations could support the discussion on oxytocin as both a pharmacodynamic marker and a potential predictor of acute psychedelic effects.
5. The discussion notes that the salivary oxytocin results parallel those seen in plasma studies. A more detailed comparison of the temporal dynamics, specifically whether peak and tapering times align across sample types, would strengthen the interpretation and utility of salivary measurements.
6. 
   The statement that oxytocin levels over time “paralleled the acute subjective intensity of effects” seems inaccurate given that oxytocin peaked at 90 minutes whereas subjective intensity peaked at 180 minutes. Discussing these divergent timelines explicitly would provide a more nuanced understanding of the pharmacodynamic response.
7. The phrase “suggesting 5-HT2A-mediated oxytocin release” may overstate the data, as this study did not test receptor mediation directly. It would be more accurate to phrase this as “consistent with potential 5-HT2A involvement,” given prior literature.

Minor typos and language issues to address:

1. Line 95

"aged between 18 and 55 years, old"
This contains a misplaced comma. It should be corrected to:
"aged between 18 and 55 years old,"
or more simply:
"aged between 18 and 55 years."

1. Line 97

"There were no restriction related to ongoing antidepressant medication..."
This should use the plural form:
"There were no restrictions related to ongoing antidepressant medication..."

3. Use of units
   Throughout the manuscript, LSD doses are written as "100 ug or 150 ug". These should use the correct microgram symbol:

"100 µg or 150 µg."

4. Consistency in timepoint notation
   The manuscript alternates between "90min", "90 min", and "90 minutes post-LSD." For clarity and consistency, use a uniform style throughout (e.g., "90 min post-LSD").

1. Line 240

"the oxytocin system and its associated physiological responses..."
This could be slightly streamlined for clarity as:
"the oxytocin system and associated physiological responses..."

 

Author Response

Dear Reviewer, 

thank you for your precious comments. Please find attached our reply.

Best regards, 

Tatiana Aboulafia Brakha

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

dear colleagues some comments and suggestions to make the paper reads better

the manuscript describes the study as observational but integrates it into routine clinical care i suggest to specify whether this was a prospective or retrospective design to enhance clarity for readers

the inclusion criteria mention two or more failed antidepressant treatments of different classes a more precise definition is needed such as the duration of trials or specific classes of antidepressants to align with established criteria for treatment-resistant depression

the use of 100 or 150 µg lsd doses is noted but not justified a rationale for these specific doses particularly why two different doses were used within the same study should be explained

the study includes 12 participants with data from 10 analyzed a rationale for the sample size even for a pilot study should be provided to address statistical power considerations

participants were invited after compassionate-use approval but the recruitment process should be clarified including how patients were identified and approached to ensure transparency

the manuscript notes missing salivary samples from two participants it should elaborate on how missing data were handled in the linear mixed model beyond stating they were considered at random

initial problems with salivary sampling are mentioned briefly more details on the nature of these issues and the specific adaptations made to improve sampling reliability are needed

the pre-lsd salivary sample is noted as taken 10 minutes before administration it should specify whether this timing was consistent across all participants and its rationale relative to drug administration

the choice of salivary oxytocin is justified as non-invasive but its validity compared to plasma oxytocin in mdd patients is not discussed studies validating salivary oxytocin in this context should be referenced

the manuscript references prior studies for the ria protocol but lacks specifics key details such as the antibody specificity or calibration standards should be included to ensure reproducibility

the results mention a significant effect of time on oxytocin levels but do not provide actual values or ranges estimated marginal means and standard errors should be included in the text to support figure 1

the non-significant random intercept variance is noted but its implications are not discussed clarification on what this suggests about individual variability in oxytocin response is needed

the non-significant ar1 parameter is reported without interpretation a discussion of its implications for the temporal correlation of oxytocin measurements within participants should be included

the friedman anova shows significant changes in subjective drug intensity but no descriptive statistics are provided median or mean intensity scores at each time point should be included

the discussion notes parallel modulation of oxytocin and subjective intensity but does not explore their correlation reporting whether a statistical analysis was conducted to assess this relationship is necessary

the open-label design is a noted limitation but the feasibility of including a placebo or active control in future studies is not discussed potential control conditions for larger trials should be suggested

the manuscript acknowledges ongoing antidepressant use but does not explore specific classes (ssris snris) a discussion on potential interactions with lsd or oxytocin dynamics is warranted

the study did not control for menstrual cycle or hormonal contraception reference specific studies on sex hormone influences on oxytocin to justify why this was not controlled in a pilot study

the compassionate-use approval is mentioned but not contextualized an explanation of how this framework influenced study design or participant selection should be provided

follow-up appointments are mentioned but excluded from the primary objectives clarification on whether any preliminary follow-up data were collected that could inform future studies is needed

the choice of 60 90 and 180 minutes for salivary sampling is not explained a rationale based on prior pharmacokinetic data for lsd or oxytocin should be provided

the ria detection limit is given as 0 1 to 0 5 pg/sample clarification on whether this range was sufficient for detecting oxytocin changes in all samples is necessary

the discussion notes a peak at 90 minutes post-lsd a comparison of this timing to prior studies in healthy volunteers would highlight similarities or differences

the manuscript states samples were assayed concurrently to minimize batch effects confirmation on whether any quality control measures were used to verify assay consistency is needed

the linear mixed model assumes normality of residuals reporting whether this assumption was tested and how violations if any were addressed is important

the introduction mentions mental flexibility as a mechanism but does not define it a clear definition and link to oxytocin or lsd effects should be provided

the best resource is cited but not fully integrated an explanation of how oxytocin fits specific biomarker categories (pharmacodynamic vs predictive) per the framework is needed

the manuscript links lsd effects to 5-ht2a receptor agonism an elaboration on how this receptor mediates oxytocin release based on cited studies would be beneficial

the study includes treatment-resistant mdd patients but does not discuss potential heterogeneity in symptom profiles acknowledgment of how this might affect oxytocin dynamics is necessary

the age range and mean are provided but other demographics (education ethnicity) are absent including these to contextualize the sample is important

the use of friedman anova for subjective intensity is appropriate but not justified an explanation of why a non-parametric test was chosen over parametric alternatives should be included

the discussion compares salivary oxytocin to plasma findings but does not address potential differences in sensitivity or specificity discussing these limitations is warranted

the ethics approval date (july 22 2024) precedes the registration date (august 15 2024) an explanation of this timeline is needed to avoid confusion about study planning

the manuscript does not mention adverse events during lsd administration a statement on whether any occurred and how they were managed should be included

the therapeutic setting is described as a sober medical environment more details on how this setting might influence subjective intensity or oxytocin release should be provided

the temporary storage at -20°c before centrifugation is noted justifying this step and confirming it did not affect oxytocin stability is necessary

the discussion mentions future studies on clinical outcomes but does not hypothesize specific links between oxytocin and depressive symptom reduction proposing potential mechanisms would be beneficial

the manuscript mentions trained psychiatric nurses but does not describe their training specifications of their qualifications for managing psychedelic sessions should be included

the manuscript emphasizes the pilot nature but does not define its scope clarification on whether the goal was feasibility effect size estimation or hypothesis generation is needed

lsd is described as an oral solution but details like volume or preparation method are absent including these for reproducibility is important

the non-significant random intercept variance suggests low individual variability a discussion on whether this is unexpected given mdd heterogeneity is needed

the exclusion criteria are listed but not justified an explanation of why conditions like psychotic disorders or pregnancy were excluded in the context of lsd and oxytocin is important

the small sample size limits power acknowledgment of how this affects the detection of smaller effect sizes in oxytocin or intensity changes should be included

the lmm results are provided but model fit statistics (aic bic) are not reported including these to support the model’s appropriateness is necessary

the 0-10 scale for subjective intensity is not described in detail it should specify whether it was validated or based on prior studies

the manuscript notes oxytocin peaks at 90 minutes but does not discuss its decline by 180 minutes a comparison to expected lsd pharmacokinetics would be informative

the lack of control for menopausal status is noted a discussion on its potential impact on oxytocin given the age range (25-55 years) is warranted

the non-invasive nature of salivary sampling is highlighted but not compared to alternatives discussing why saliva was preferred over blood or urine is needed

the manuscript does not address participant compliance with salivary sampling instructions reporting any challenges or adherence rates is important

the discussion cites ketanserin’s attenuation of lsd effects but not its impact on oxytocin clarification on whether ketanserin studies support oxytocin findings would be beneficial

the lack of longitudinal oxytocin follow-up is noted proposing how repeated measurements could inform treatment response in future studies is necessary

the study’s integration into clinical routine is mentioned but not detailed an explanation of how research procedures were balanced with standard care should be provided

the lmm and friedman anova results lack effect size measures including these to quantify the magnitude of oxytocin and intensity changes is important

the introduction links oxytocin to mental flexibility but does not revisit this in the discussion an elaboration on how oxytocin might mediate flexibility in mdd is warranted

the manuscript does not mention participant dropout confirmation on whether all 12/10 participants completed the study or if any withdrew is necessary

data collection from september 2024 to february 2025 is noted an explanation of why this period was chosen and whether it affected results (seasonal effects) should be included

the discussion compares lsd to mdma but does not address other psychedelics like psilocybin discussing whether oxytocin findings might apply broadly is warranted

the medical check before administration is mentioned but not detailed specifications of which vital signs were monitored and their relevance to safety should be included

the conclusion suggests oxytocin as a potential biomarker outlining specific steps for validating oxytocin as a pharmacodynamic or predictive biomarker in future studies would be beneficial

 

Author Response

Dear Reviewer, thank you for all these comments that allow a clear improvement of the manuscript. Please find attached a detailed response to each comment (with respective changes made on the manuscript).

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

No more comments.