[Gary Graziano]

I’ve been taking it for about a month and a half, along with myriad other repurposed drugs, as part of a protocol for cancer. It’s really too soon to tell what effect it is having. I have also been diagnosed with Hashimoto’s Thyroiditis. (I gather this isn’t common in men, but I guess I’m the exception.) I’ve taken some steps to reduce thyroid antibodies in addition to the LDN, and will be interested to see if the numbers improve. (I’ll be tested again in August.) Naltrexone, as prescribed for addiction, may be cheap but I have to get mine compounded from a local pharmacy. It’s not prohibitively expensive, but I wouldn’t call it cheap, either. Medicare doesn’t cover it, like almost everything else I’m doing. Everything I buy at retail adds up!

[IMA-HelenT]

Thank you for sharing your experience. I hope your August tests provide some helpful answers.

Your point about cost is also important. While naltrexone itself is an inexpensive generic medication, low-dose naltrexone often has to be specially compounded, which can make it significantly more expensive and, as you’ve experienced, it’s not always covered by insurance or Medicare.

Stories like yours highlight why promising observations deserve proper scientific investigation.

Unfortunately, as we have seen time and time again, because LDN is a generic drug, securing funding for the large randomized clinical trials needed to answer these questions can be much more difficult.

Please let us know how you get on.

[aaronaf]

About the big kahuna RCTs, why couldn’t it be feasible for individual integrative physicians handling their cancer cases to funnel their patients’ blood and other fluid tests, their X-rays, miscellaneous imaging results, and other clinical diagnostic and after-repurposed drug result data into a clearing house sponsored by and managed by the many collaborating Senior Fellows of the IMA?

Because this effort would amass a variety of case histories, it would probably approximate the scientific quality of an RCT – especially if it included lab tests that examined biomarkers that probed the veracity of the Warburg effect at every possible point of every patient’s treatment.

Of course, this kind of clinical “clearing house” would require informed consent and permissions paperwork to comply with privacy laws. These permissions might easily flow from the participating patients, because not only would their highly stratified and intensely analyzed case histories inform their own therapies, their publication in the Journal of Independent Medicine would strengthen the science underlying them, and help the entire population of cancer patients.

[IMA-HelenT]

A collaborative registry or “clearing house” where physicians contribute standardized clinical data, laboratory results, imaging, outcomes, and patient-reported experiences could become an incredibly valuable source of real-world evidence. A great idea.

Perhaps one of the biggest challenges isn’t the science, it’s building the infrastructure, securing funding, and encouraging enough physicians to contribute data consistently. But it’s certainly an idea worth exploring.

[aaronaf]

I envision only a group of IMA Senior Fellows, which already exist. No new infrastructure would be needed, only a willingness to meet on a regular basis, maybe monthly or weekly, to peruse and analyze the submitted data from patient oncologists’ routine lab work. No additional lab work would be done on patient samples, because I am pretty sure that would violate rules for further investigations on patients without going through a labyrinth of paperwork and expense.

The Senior Fellows and their respective team members would seek to isolate the biomarkers that would allow them to track them for each patient, with the purpose of scientifically evaluating their underlying hypotheses as being potentially falsifiable.

Ideally, it would be wonderful if the data proved they were correct hypotheses. For example, with regard to the Warburg effect, analysts might examine pyruvate/lactate ratios, and look for trends or patterns among the different cancer patients. Doing this might necessitate a more enlightened treatment of cancer patients by their oncologists, because the analyses would likely require probing of tumor biopsies, which is associated with heightened risks. How could this approach be done without compromising the safety of patients at the same time avoiding shunting them off into RCT investigations?

[aaronaf]

I submitted your idea of a needed new infrastructure to Ai and obtained an answer that agreed with you. Here it is below:

“Yes, that’s a reasonable reaction.

Even with “fully de-identified” secondary analysis, you still typically need new infrastructure for: (1) standardized data extraction templates for many independent practices, (2) secure transfer/storage and access control, (3) data dictionaries and harmonization (units, timing rules, field mappings), (4) quality-control pipelines, and (5) governance/IRB documentation and author/publication workflows. Without that, variations between clinics quickly make results hard to validate and publish.

Whether it’s “new infrastructure” vs “lightweight tooling on top of existing vendor/contract infrastructure” depends on how many practices, what data formats they already export, and how strict your timing/window and assay comparability rules need to be.”

[aaronaf]

I submitted my idea for a collaboration with IMA and independent oncologists to Ai. I stipulated that there be exempt status from IRB regulations.

## Proposal: De-identified Secondary Analysis of Routine Oncology Labs to Study Metabolic Trajectories and Treatment Response

### 1) Purpose

Conduct a multicenter, independent-clinic collaboration to analyze routinely collected laboratory data to evaluate whether metabolic trajectories associated with the Warburg effect (and competing metabolic models) change around treatment start and response milestones.

### 2) Study Design

– Type: Retrospective/secondary analysis of already-collected routine clinical laboratory data.

– Data source: De-identified laboratory results transmitted from participating oncology clinics.

– No additional procedures: No new blood draws, imaging, interventions, or patient contact by the IMA team.

### 3) Participating Organizations and Roles

– **Participating oncology clinics (or physicians/independent oncology practices):**

– Provide de-identified routine lab datasets and relevant metadata.

– Commit to standardized data extraction fields and timing rules defined in the protocol.

– **IMA coordinating team:**

– Pre-specifies the analysis plan, data schema, quality checks, and statistical methods.

– Receives de-identified data, performs analysis, and drafts manuscripts for peer-reviewed publication.

– **Optional independent oversight (recommended):**

– Data monitoring/steering committee to oversee protocol adherence, analysis integrity, and publication.

### 4) Primary/Secondary Questions (examples)

– Primary: Are early on-treatment changes in selected metabolic laboratory readouts associated with subsequent treatment response?

– Secondary: Do trajectories differ by regimen class, baseline metabolic status, or other prespecified clinical strata?

### 5) Data Elements (minimum set)

Define a fixed data dictionary to be used by all clinics, including:

– De-identified patient ID (non-reversible)

– Cancer diagnosis type, stage (if available), and treatment start date (or treatment cycle day 1)

– Treatment regimen identifiers (standardized categories)

– Response assessment type and milestone date (e.g., imaging date and/or clinician-documented response timepoint if available)

– Routine metabolic-relevant labs available across sites (pre-specified list; e.g., LDH/lactate/glucose/insulin ± others depending on feasibility)

– Lab collection timestamp (or best-available proxy) and units

– Specimen/analyte handling fields if already present in routine records (no new assays)

### 6) Standardization and Harmonization

– Each clinic extracts labs using the same field mapping and timing rules.

– Units are converted to a common scale using prespecified conversion rules.

– Baseline and milestone windows are defined operationally (e.g., “baseline = last measurement before treatment start”; “early on-treatment = first measurement within X days after start”; “response milestone = within Y days of response assessment date”).

– Pre-specify how to handle multiple measurements within windows (e.g., nearest-to-window-center or average).

### 7) Data De-identification and Transfer

– Clinics perform de-identification prior to transfer (remove direct identifiers and any re-identification keys).

– Transfer occurs via secure channels per IMA’s requirements (encrypted storage and access control).

– The IMA receives only de-identified datasets meeting the project’s de-identification standard.

### 8) Exempt/Waiver Publication Pathway (process)

– IMA obtains an IRB determination (often **exempt** or **waived consent**) based on:

– Secondary analysis of de-identified data

– No new human-subject interaction

– Participating clinics rely on IMA’s approved protocol where permitted and/or document local determinations as required.

### 9) Privacy, Security, and Data Governance

– Data minimization: share only fields needed for the analyses.

– Access control: limit identifiable information handling (should not be present at IMA level).

– Data retention: retain only what’s required for analysis and auditability per IRB/IQA guidance.

### 10) Statistical Analysis Plan (pre-specified)

– Pre-register core endpoints and models in the protocol.

– Use trajectory/time-window comparisons and association models (e.g., baseline vs early change; milestone-linked changes), with adjustment for available covariates.

– Predefine inclusion/exclusion criteria for missing required labs within defined windows.

– Conduct sensitivity analyses for window definitions and regimen classes.

### 11) Quality Control

– Automated checks: unit consistency, impossible values, timestamp sanity checks.

– Site-level QA: proportion of patients with sufficient data per time window.

– Outlier handling rules pre-specified.

### 12) Authorship, Sharing, and Publication

– Publication in a peer-reviewed journal under IMA leadership.

– Author list follows journal and consortium authorship guidelines.

– Results are reported aggregated; no site-level disclosure that could identify patients.

### 13) Timeline

– Month 0–2: protocol + data dictionary finalization; IRB determination initiation.

– Month 2–4: clinic onboarding; data extraction template deployment; pilot data transfer.

– Month 4–6: full data collection and QA.

– Month 6–8: analysis and manuscript drafting.

– Month 8+: submission and revisions.

### 14) Deliverables

– Approved protocol + analysis plan

– De-identified dataset specification and extraction template

– Final statistical results

– Peer-reviewed manuscript and supplementary methods

If you want, tell me the specific metabolic lab readouts you care about and the response milestone source you expect (imaging dates, clinician response documentation, etc.), and I’ll tailor the “Data Elements,” “Timing windows,” and “Analysis Plan” sections accordingly.

[aaronaf]

“Here are the practical data-quality and design checks the IMA team could run to ensure the study can credibly test the predefined hypotheses and support publication-quality conclusions: [Note: This is a revised version to the earlier draft bullets of the proposal.]