Wim Hof Method Breathwork and Cold Exposure for Cancer-Related Inflammation: Feasibility Pilot

Core Team

Principal Investigator: Sara Matijevic, PhD, University of Oxford, Turing Enrichment Scholar

Study Coordinator: Rafael Rolli, Independent Researcher

Co-Investigator: Jim Nasr, Independent Researcher

WHM Instructor: Jane Jones, UK Lead, Certified Advanced Wim Hof Method Instructor

AI Lead: Ashley MacWhirter, Data Engineer

Medical Safety Advisor: Dr. Owais Durrani, Memorial Hermann Health System, ER Physician

Special Advisor: Wim Hof, Creator of Wim Hof Method

Conflicts of Interest and Independence Statement

Sara Matijevic, Rafael Rolli, and Jim Nasr serve as independent researchers for this study. While the study examines the Wim Hof Method and collaborates with certified WHM instructors, the research team has no financial interests in the Wim Hof organization. The Wim Hof Method is referenced as an established wellness practice, and we acknowledge the method's origins and protocols. However, this study is independently conducted and funded. All data analysis, interpretation, and publication decisions are made independently by the research team. Raw anonymized data will be made publicly available to enable independent verification of our findings.

Abstract

We propose a 16-week observational feasibility study comprising two cohorts: a primary intervention cohort of adults living with cancer, and a healthy volunteer observational cohort. We aim to assess whether adults living with cancer can safely and consistently practice the Wim Hof Method (WHM), a combination of specific breathing techniques and cold exposure, with physician approval. The primary aims focus on feasibility metrics: recruitment rate, retention rate, adherence rate, and safety monitoring. Secondary exploratory aims include preliminary signals in inflammation markers (hs-CRP, IL-6), heart rate variability (if measurable via participant wearables), and patient-reported outcomes including fatigue, sleep quality, and mood. This is not a treatment study. We are observing people who choose to practice WHM with their doctor's approval, documenting their experience, and measuring whether this approach warrants a larger controlled trial. Timeline is 16-week intervention with 4-week follow-up survey at week 20.

Background

Many people with cancer experience high inflammation and poor recovery, including fatigue and sleep disruption. Gentle cold exposure, full cold water immersion, and controlled breathing may help lower inflammatory tone and improve autonomic balance. A seminal study published in 2022 in Nature by researchers at Karolinska Institutet (Seki et al.) called "Brown-fat-mediated tumour suppression by cold-altered global metabolism" demonstrated promising outcomes in animal models. The findings indicate that cold-induced activation of brown adipose tissue may have a tumor-suppressive effect. This study used a 4°C temperature to activate BAT in cancer-inoculated mice. These findings suggest that cold therapy could potentially have a significant impact on cancer treatment, emphasizing the need for further investigation in this area of research.

The Wim Hof Method is an established wellness practice combining breathing technique, cold exposure, and mental commitment. Peer-reviewed research includes Kox et al. (2014) in PNAS demonstrating voluntary immune system modulation, Muzik et al. (2018) in NeuroImage showing brain connectivity changes during WHM breathing, and Buijze et al. (2016) in PLoS ONE finding reduced sick days with cold shower practice.

We have reviewed many cold therapy related papers and studies and developed an extensive data analysis of those studies, which guide this proposal. These analytics, dashboards and related AI data tools are planned to be surfaced and continue to evolve on our site: https://coldtherapy.acoer.com/.

Study Classification and Ethics Oversight

This is a feasibility pilot study designed to assess whether adults living with cancer can safely and consistently practice the Wim Hof Method with physician approval. This study follows a physician-gated observational approach with robust safety monitoring. Each participant must obtain written medical clearance from their treating physician before enrollment. Dr. Durrani provides ongoing medical safety oversight throughout the study. We are consulting with ResearchHub on additional ethics review requirements and will comply with all recommended oversight procedures.

Important Disclaimer: This is not a medical treatment or clinical trial. This is a feasibility assessment of an existing wellness practice. Participation does not replace medical care and all participants remain under their physician's supervision.

Objectives

Primary Outcomes (Feasibility):

Recruitment rate: Can we recruit 15-20 participants with cancer who obtain physician approval? Target is 60% or greater of contacted eligible participants enroll.
Retention rate: Will 75% or greater of enrolled participants complete the 16-week protocol?
Adherence rate: Will participants adhere to 70% or greater of prescribed breathing and cold exposure sessions?
Safety: Are there any serious adverse events or safety concerns?

Secondary Outcomes (Exploratory):

Changes from baseline to week 16 in:

High-sensitivity C-reactive protein (hs-CRP)
Interleukin-6 (IL-6)
Heart rate variability (HRV) if measurable via participant wearables
Fatigue assessed using FACIT-Fatigue scale
Sleep quality assessed using PROMIS Sleep Disturbance short form
Mood assessed using PHQ-4
HOMA-IR calculated from fasting glucose and insulin

We will document potential confounding factors such as infections, injuries, medication changes, and cancer progression. All biomarker findings will be analyzed using effect sizes and confidence intervals with no p-value thresholds for significance given the exploratory nature and small sample size. We will contextualize results based on potential confounders and avoid overstating outcomes given the single-arm feasibility design.

Exploratory:

Optional additional analyses on wearable data, causal machine learning methods to estimate adherence effects, and subgroup pattern discovery.

Study Overview

Design: Two-cohort feasibility pilot. Primary cohort: cancer patients practicing WHM. Observational cohort: healthy volunteers practicing WHM, analyzed separately as supplementary reference data.

Length: 16-week intervention period with 4-week follow-up survey at week 20

Size: 15 to 20 participants

Setting: Instructor-guided remote sessions and home practice

Eligibility: Adults age 18 and older living with cancer, access to smartphone, written medical clearance from treating physician (GP and when applicable the participant's cancer care team) confirming approval for WHM breathing and cold exposure.

Exclusions (safety): Exclusions (safety): Serious cardiovascular disease (uncontrolled hypertension, recent myocardial infarction, severe arrhythmia, heart failure NYHA class III or IV, cardiomyopathy with ejection fraction less than 50%), history of stroke or TIA within past 6 months, epilepsy or seizure disorder, pregnancy or breastfeeding, severe psychiatric illness (active psychosis, severe mania, recent suicide attempt), cold urticaria or severe Raynaud disease, recent surgery or hospitalization within 4 weeks, active wounds or skin breakdown or ports at planned cold exposure sites, severe anemia (hemoglobin less than 10 g/dL), severe immunosuppression (absolute neutrophil count less than 500, recent transplant), COPD or severe asthma or pulmonary hypertension, active infection or fever, chemotherapy-induced neuropathy grade 2 or higher, cachexia or BMI less than 16, history of febrile neutropenia, current or recent oxaliplatin therapy (within 6 months), and others as advised by physician overseer.

Participants on active chemotherapy must have written oncologist approval in addition to GP clearance. Participants on immunotherapy or radiation must have oncologist approval. Participants with cardiac history must have cardiologist clearance if physician deems necessary.

Withdrawal criteria: Participant wishes to discontinue, physician recommends discontinuation, serious adverse event, cancer progression requiring protocol modification, or inability to adhere to minimum protocol requirements.

Intervention Summary

Breathing Technique

Protocol: Daily practice with minimum 5 days per week. Duration 10 to 15 minutes per session. Technique involves 30 to 40 cycles of controlled breathing (deep inhale, passive exhale), breath retention after final exhale (self-regulated, typically 1 to 2 minutes), recovery breath (deep inhale, hold 10 to 15 seconds), repeated for 3 to 4 rounds.

Safety guidelines: Must be performed sitting or lying down on soft surface such as bed or yoga mat. Never in water, while driving, or in any situation where losing consciousness could cause injury. Stop immediately if experiencing severe dizziness, chest pain, or distress. Modifications available for participants with respiratory conditions per physician guidance.

Tracked via WHM app: Number of breathing rounds completed, breath retention times, daily adherence.

Cold Exposure (Progressive Protocol)

Weeks 1 to 2 (Adaptation Phase): Cold finish to shower for 15 to 30 seconds. Frequency 3 to 5 days per week. Focus on getting comfortable with mild cold sensation.

Weeks 3 to 8 (Building Phase): Cold shower duration 30 to 90 seconds. Temperature gradually decrease to 10 to 16°C (50 to 60°F) if comfortable. Frequency 5 to 7 days per week.

Weeks 9 to 16 (Maintenance Phase): Cold shower duration 1 to 3 minutes. Temperature 10 to 16°C or coldest available. Frequency 5 to 7 days per week. Optional room temperature reduction to 16 to 18°C for 2 to 6 hours per day if feasible.

Tracked via daily log: Cold exposure duration in seconds or minutes, temperature if known, completion yes or no, any adverse reactions.

Cold Water Immersion (Optional)

Participants who wish to practice full cold water immersion at 4°C or below may do so with separate written medical approval specifically for CWI. Note: Full-body cold water immersion is prohibited for participants on active chemotherapy; cold showers only are permitted for this group. Requirements include completion of at least 4 weeks of breathing plus cold shower practice first, minimum 3 rounds of breathing immediately before each immersion, and WHM instructor supervision. CWI participants will be analyzed as a separate exploratory cohort and not mixed with non-CWI participants in primary feasibility or biomarker analyses.

Empirically it is proven that it is likely and practical with a simple CWI protocol to reach 4°C or colder with consistent full body exposure. Conversely it is currently scientifically unproven or complex to normalize a protocol to consistently measure impact of cold exposure in a shower or through similar periodic air exposure or other related techniques. Among others, Dr. Andrew Huberman and Dr. Susanna Soberg have expressed some of the key challenges that cold showers represent from a consistent measurement of effect versus cold water immersion.

Commitment and Support

Weekly check-in from certified WHM instructor. Access to WHM instructional videos and guided sessions via app. Gradual progression encouraged with no rushing and listening to body. Optional bi-weekly group support calls.

Measures

Primary Feasibility Outcomes:

Recruitment rate: Percentage of contacted eligible participants who enroll. Target 60% or greater.
Retention rate: Percentage of enrolled participants who complete 16 weeks. Target 75% or greater.
Adherence rate: Percentage of prescribed breathing and cold exposure sessions completed. Target 70% or greater for each.
Safety: Number and severity of adverse events. Number of participants experiencing one or more adverse events. Serious adverse events. Tolerability ratings.

Secondary Exploratory Outcomes:

Biomarkers at baseline, week 8, and week 16:

High-sensitivity C-reactive protein (hs-CRP) as inflammation marker
Interleukin-6 (IL-6) as pro-inflammatory cytokine
Fasting glucose and insulin for HOMA-IR calculation as insulin resistance marker
Optional TNF-alpha and ketones

Collection method is finger-prick at-home test kits with lab processing. Participants can opt out of biomarkers.

Wearable data (continuous monitoring):

Participants use their existing wearables such as Oura Ring, Whoop, Apple Watch, Fitbit, or similar devices. If a participant does not own a wearable we will provide one as part of the study. Data collected depends on device capability and may include heart rate variability (HRV) if device supports it, resting heart rate, sleep duration and quality metrics, activity levels such as steps and active minutes, and body temperature if available. We acknowledge measurement variability across devices but prioritize real-world feasibility over perfect standardization for this pilot. Device type will be documented as a covariate.

WHM app data:

Breathing sessions including rounds and retention times. Cold exposure logs. Practice adherence and streaks.

Patient-reported outcomes via weekly surveys:

Fatigue using FACIT-Fatigue scale (0 to 52, higher equals less fatigue)
Sleep quality using PROMIS Sleep Disturbance short form (0 to 20, higher equals worse sleep)
Mood using PHQ-4 (0 to 12, higher equals worse mood and anxiety)
Pain using 0 to 10 numeric scale
Appetite using 0 to 10 scale
Weight self-reported in kilograms

Daily practice log:

Breathing rounds completed, retention times, session done yes or no. Cold exposure duration, temperature, completion yes or no. Symptoms or any adverse effects. Energy rated 1 to 5 scale. Overall well-being rated 1 to 5 scale.

Safety

Before enrollment:

Written informed consent required. Written medical clearance from physician required (oncologist clearance required for active chemotherapy participants). Baseline health screening by Dr. Durrani via remote consultation. Baseline vital signs (blood pressure, heart rate). Baseline ECG for participants over 50 years old or with cardiac history. Review of current medications, treatments, and contraindications. Review of most recent oncology labs (no need to duplicate if already available).

During study:

Weekly check-in with WHM instructor. Bi-weekly safety review by Dr. Durrani. Participants can report adverse events 24 hours a day 7 days a week via study hotline or email. Monthly data safety review by research team. Protocol hold criteria for active chemotherapy participants: Practice must be temporarily suspended if participant develops fever above 100.4°F (38°C), active infection, significant dehydration, active mucositis, or compromised central line site. Practice may resume upon physician clearance.

Adverse event reporting:

Participants report any concerning symptoms immediately. All adverse events documented and reviewed within 48 hours. Serious adverse events defined as hospitalization or significant harm are reported to physician within 24 hours. Study paused if 2 or more serious adverse events potentially related to protocol.

Stopping rules:

Individual level: Participant discretion, physician recommendation, or serious adverse event. Study-wide level: 2 or more serious adverse events potentially related to WHM practice.

Informed consent process:

Clear explanation of risks, benefits, and alternatives. Right to withdraw at any time without penalty. Option to opt out of biomarkers while remaining in study. Option to opt out of public data sharing.

Analysis Plan

Feasibility Analysis (Primary):

Recruitment: Percentage of contacted individuals who enroll. Time to reach target enrollment. Reasons for non-enrollment.

Retention: Percentage completing 16 weeks. Dropout timing and reasons.

Adherence: Percentage of prescribed breathing sessions completed. Percentage of prescribed cold exposure sessions completed. Weekly adherence patterns. Barriers to adherence via qualitative assessment.

Safety: Number of adverse events by severity. Number of participants experiencing one or more adverse event. Serious adverse events. Tolerability ratings.

Exploratory Biomarker Analysis:

Statistical approach: Paired comparisons from baseline versus week 16 using Wilcoxon signed-rank tests which are non-parametric and appropriate for small sample size. Effect sizes calculated as Cohen d with 95% confidence intervals. Descriptive statistics including medians and interquartile ranges for all biomarkers. No p-value thresholds will be used given exploratory nature.

Outcomes: Change in hs-CRP measured in mg/L. Change in IL-6 measured in pg/mL. Change in HRV measured as RMSSD in milliseconds if sufficient data available. Change in HOMA-IR. Changes in patient-reported outcomes including fatigue, sleep, mood, and pain.

Confounding factors: Document infections, injuries, medication changes, cancer progression, and new treatments. Wearable device type as covariate. Sensitivity analyses excluding participants with major confounding events. Acknowledge limitations of uncontrolled primary cohort design. Healthy observational cohort data will be reported descriptively in a separate appendix and will not be included in primary or secondary analyses.

Machine Learning and AI Analysis (Exploratory):

Clustering and pattern discovery using unsupervised methods such as k-means to explore whether subgroups emerge that distinguish between higher and lower responders.
Time-series dynamics using change-point detection and rolling averages to help identify whether meaningful shifts in HRV or fatigue occur following protocol milestones.
Predictive modeling using small recurrent neural networks to test whether early adherence and wearable signals predict later improvements. This will be evaluated using cross-validation within the cohort with careful reporting of instability due to sample size.
Causal machine learning to move beyond association and prediction. We will explore feasibility of applying causal machine learning methods such as causal forests and double machine learning. These approaches can help estimate whether adherence to the Wim Hof protocol has an identifiable causal effect on outcomes such as fatigue, HRV, or hs-CRP after adjusting for confounders. Given the small sample size results will be treated as exploratory and hypothesis-generating, but this pilot will allow us to test pipelines for causal inference that could be scaled in future larger trials.

All statistical code in R or Python and all ML and deep learning pipelines will be shared in an open repository. Both positive and null findings will be documented.

Data, Privacy, Openness

Data storage: All data encrypted at rest and in transit. HIPAA-compliant cloud infrastructure using AWS or Azure with Business Associate Agreement. GDPR-compliant for EU participants. Random study IDs with contact information stored separately.

Access controls: Only research team members have access to identifiable data. Participants can request their own data at any time. Participants can withdraw and request data deletion.

Data ownership: Participants retain ownership of their data. By consenting to participate they grant the research team right to analyze anonymized data, right to publish aggregated results, and right to release anonymized data to public research commons. Participants can access their own data at any time, withdraw from the study and request data deletion, or opt out of public data sharing.

Blockchain-verified data provenance: To ensure data integrity we use decentralized data management infrastructure with blockchain verification. Data collected is anonymized and aggregated. Cryptographic fingerprint or hash of datasets is recorded on blockchain with immutable timestamp. Anyone can verify data authenticity by comparing published datasets against blockchain record. This prevents post-hoc data manipulation or selective reporting. Think of this as a notarized seal on our data. Once we publish results anyone can verify we did not change the data afterward. Technical implementation uses Ethereum-based or similar public blockchain, IPFS for decentralized storage, and cryptographic proofs provided with all data releases.

Peer Review on ResearchHub

Open community review. Invite 2 to 3 external reviewers with small bounties. We will publish a response-to-review note and update the page before enrollment.

Milestones

Protocol version 1 posted for review. Pre-registration and final consent text. Ethics oversight finalized including physician-gated approach plus any additional requirements. Recruitment begins with target of 15 to 20 participants over 4 to 6 weeks. All participants enrolled and baseline data collected. Week 16 intervention completion for final participant. Week 20 follow-up surveys complete. Data analysis complete. Results readout to community. Manuscript submitted and data and code released. Post-study community call to discuss next steps.

Statistical and Machine Learning Methodology

Our analysis plan combines standard statistical methods with machine learning and deep learning techniques. This approach ensures that the study produces both interpretable clinically relevant results while also testing the feasibility of more advanced approaches such as machine learning to uncover new associations.

Statistical Analysis

Primary outcomes: Feasibility metrics including recruitment rate, retention rate, adherence rate, and safety outcomes will be reported descriptively with frequencies and percentages.

Secondary outcomes: Patient-reported fatigue, sleep quality, and mood scores will be analyzed using repeated-measures approaches such as linear mixed-effects models with random intercepts to account for within-person correlation over time.

Exploratory biomarkers: Where available such as IL-6 or TNF-alpha, changes will be summarized descriptively and compared against baseline using effect sizes and confidence intervals.

Confounding factors: Documented confounders including illness, infection, and medication changes will be included as covariates in sensitivity analyses.

Optional: Additional prospective analysis on wearables or other collected data from participants.

Machine Learning

As described above in Analysis Plan section.

Open-Science Practices

All statistical code in R or Python and ML and DL pipelines will be shared in an open repository. Both positive and null findings will be documented.

Budget Outline

Data collection platform development: $5,000 for custom surveys, WHM app integration, and dashboards

Finger-prick lab kits and processing: $15,000 for hs-CRP, IL-6, glucose, and insulin for 20 participants at 3 timepoints each

AI and ML model development: $20,000 for time-series analysis, causal inference, pattern discovery, and data engineering

Blockchain and decentralized data infrastructure: $10,000 for immutable data verification and cryptographic proofs

Research portal development: $15,000 for public-facing results dashboard and data commons

Project management: $15,000 for coordination, participant communication, and documentation

Reporting, communication, and community updates: $10,000 for publications, preprints, community calls, and dissemination

Ethics consultation and compliance: $4,000 for consulting and regulatory compliance support

Contingency: $5,000 for unexpected costs and additional safety measures

Overall estimated budget: $100,000

Open-Science

After the pilot we will explore launching commons for our anonymized data, methods, and code. Option to mint an IP-NFT to package releases and license terms. If the community and the Wim Hof organization agree we can later discuss an IPT governance token for the commons. No token now. The first pilot comes first.

Protocol will be registered before enrollment begins on ResearchHub, OSF, or ClinicalTrials.gov. Analysis plan specified in advance with any deviations documented and explained. Anonymized individual-level data released 6 months post-publication. Code for all statistical and ML analyses shared on GitHub. Interactive dashboard for exploring results at coldtherapy.acoer.com. Results submitted to peer-reviewed open-access journal. Preprint posted regardless of publication outcome. Null results reported with same rigor as positive results.

Limitations and Future Directions

Limitations of This Study:

Primary cancer cohort is uncontrolled, limiting causal inference. Healthy volunteer observational cohort serves as supplementary reference data only and is not used as a scientific control group. Small sample size means exploratory biomarker findings only and underpowered for definitive conclusions. Self-selected population means people motivated to try WHM may differ from general cancer population. Heterogeneous cancer types means mixed stages and types reduce ability to detect disease-specific effects. Wearable variability means different devices may measure HRV differently, documented as covariate.

What This Study Will Tell Us:

Feasibility metrics answering can we recruit, retain, and collect quality data. Safety assessment answering are there adverse events or tolerability issues. Preliminary signals providing effect size estimates for future power calculations. Responder patterns identifying who benefits most.

This study cannot prove WHM causes changes in biomarkers due to no control group. Cannot generalize to all cancer patients due to selected sample. Cannot make clinical recommendations based on this pilot alone.

Future Directions If Feasibility Is Demonstrated:

Next study would be a larger randomized controlled trial comparing WHM versus breathwork-only versus standard care. Sample size of 60 to 120 participants. Narrower cancer population such as specific cancer type in active treatment. Standardized wearables for all participants. Longer follow-up of 6 to 12 months. Additional biomarkers such as BAT activity via imaging and immune cell phenotyping.

References

Kox M, van Eijk LT, Zwaag J, et al. Voluntary activation of the sympathetic nervous system and attenuation of the innate immune response in humans. Proc Natl Acad Sci USA. 2014;111(20):7379-7384.
Seki Y, Berglund K, Mandrup S, et al. Brown-fat-mediated tumour suppression by cold-altered global metabolism. Nature. 2022;608:421-428.
Buijze GA, Sierevelt IN, van der Heijden BC, et al. The Effect of Cold Showering on Health and Work: A Randomized Controlled Trial. PLoS One. 2016;11(9):e0161749.
Muzik O, Reilly KT, Diwadkar VA. Brain over body: A study on the willful regulation of autonomic function during cold exposure. NeuroImage. 2018;172:632-641.
Additional references available at: https://coldtherapy.acoer.com/

What We Want From Reviewers

Are the feasibility measures appropriate for a small non-hospital pilot? Is the safety screening clear enough? Is the physician-gated observational approach with medical safety oversight sufficient or do you recommend additional ethics review procedures? Is broad cancer inclusion with detailed safety exclusions acceptable for feasibility pilot or must we narrow to specific cancer types or stages? Is allowing participants to use existing wearables with device type as covariate acceptable or should we standardize despite cost and accessibility trade-offs? Should HRV be made optional or exploratory given device variability? Any critical feasibility metrics we are missing? Any simple improvements to the data dictionary, consent, or release plan?

Wim Hof Method Breathwork and Cold Exposure for Cancer-Related Inflammation: Feasibility Pilot

Wim Hof Method Breathwork and Cold Exposure for Cancer-Related Inflammation: Feasibility Pilot

Funding Progress

Funders

Peer Reviews

Nonprofit Recipient

Topics

DOI