The Root Room is where we open the doors most people never walk through. Some of those doors are in research labs, trial sites, and FDA review rooms. Others are inside our own feeds and our own heads, because that is where the evidence first has to compete with the noise. Each issue, we step inside with the characters who live there.

This week we are stepping into one of those interior rooms. Not a clinical room. The kind of room every one of us is sitting in right now: the place where information arrives, where feelings move faster than facts, and where an algorithm decides what we see next. There is someone there you should meet.

Her name is Debby. She is not a villain. She is not stupid. She is a person navigating a world full of health information and very few reliable tools for evaluating it. She appears in our Root to Rx Lab community, and she represents a question that deserves a real answer.

This week, Debby saw a post in her social media feed. It said: 'Big Pharma just rushes drugs to market for profit. The 10-year timeline is a myth they use to justify the prices.'

She recognized the feeling immediately. She had heard versions of this for years, from family members, in online groups, from people she trusted. It felt like it could be true. She shared it.

What Debby could not see, but what we can see standing in The Root Room, is who else was already in there with her.

Within six hours, Echo the Algorithm had filled her feed with more like it. Videos with urgent music. Screenshots of studies she could not access. A doctor she had never heard of saying the clinical trial process was mostly theater.

Then Dr. Anecdote showed up in the comments. 'My sister's oncologist told her there was a treatment that was ready years ago but the company sat on it to protect another drug's patent.' Dr. Anecdote always has a story. The story is always vivid, specific, and unverifiable.

And Bias the Brain was there too, quieter than the others, doing what Bias does best. Bias did not argue with Debby. Bias whispered what she already wanted to be true.

These are the actual antagonists of this room. Not Debby. Echo does not care if Debby is right. Echo cares if Debby keeps watching. Dr. Anecdote does not care if Debby is informed. Dr. Anecdote cares that the next vivid story keeps her engaged. Bias the Brain does not need to convince. Bias only needs to nod.

Debby started to wonder. If the process takes 10 to 15 years, why does it sometimes seem to happen faster? If the FDA is so rigorous, why do drugs get recalled? If the system works, why do so many conditions still have no approved treatment?

Those are not stupid questions. They are the right questions. The problem is not that Debby is asking them. The problem is that she is getting answers from sources that profit from her doubt rather than sources that are accountable to evidence.

Sam the Skeptic has been watching from the doorway. He does not start by telling Debby she is wrong. He starts by handing her something.

'There is a toolkit,' Sam says. 'Ten questions we use to evaluate any health claim, including the one in the post you shared. Question 1: What would actually change your mind about the timeline claim? If the timeline is a myth, what does that claim predict? Try it on the post. Then try it on the alternative.'

Rooty the Researcher steps in next, carrying the data Sam will not pretend to know off the top of his head.

'The DiMasi study,' Rooty says. '106 randomly selected drugs from 10 pharmaceutical companies. Methodology peer-reviewed and publicly available. Funding disclosed. Question 4 of the toolkit asks who funded a claim. The post Debby shared did not disclose anything.'

Debby is not impressed. 'Of course a pharma-funded study says it takes 10 years,' she says. 'They are justifying the prices.'

Sam nods. 'That instinct is fair. That is exactly what Question 4 is for. The difference is the toolkit asks the question of every claim, not just the ones that disagree with you. The post you shared had no funding disclosure at all. That should make Question 4 louder, not quieter.'

Debby is not convinced. She is not even close to convinced. She is doing what most people do when the toolkit first arrives: holding her position and looking for the next reason the toolkit does not apply.

That is the realistic place to start. Root to Rx does not exist to convert anyone in one issue. Root to Rx exists to put the toolkit in your hands so you can dig your way out, in your own time, on your own terms. We do not throw conclusions at you. We do not throw shovels at the people throwing shovels at us. We hand you the shovel and we step back.

Debby will be back. So will Sam. So will Echo, Dr. Anecdote, and the others. The Root Room has many functions. We will explore them as a community.

Next issue we open another door. A discovery that sat ignored for 13 years, a molecule that almost never made it through the pipeline, and what happened when it finally did. In March, Debby's feed lit up with a story: a 'newly declassified CIA document' supposedly proved a cancer cure had been hidden for 60 years. The story we are about to tell is the actual answer to that claim, and it is not the one Echo or Dr. Anecdote will ever tell her.

The story above is what evidence-evaluation looks like at the dinner table. What follows is the institutional version: the statutory framework governing IND applications, the full scope of ICH-GCP E6(R3), the statistical design requirements for Phase III, the FDA review mechanics, and the regulatory history that produced these standards. This is the level of detail that lets you read a primary source rather than a press release.

The Investigational New Drug application is the legal mechanism by which a sponsor obtains FDA permission to administer an unapproved drug to humans in the United States. It is governed by Title 21 of the Code of Federal Regulations, Part 312.

The IND submission has three primary components. The first is the Investigator's Brochure, a comprehensive document summarizing all preclinical and available clinical data on the compound. It is updated with each new relevant finding throughout the life of the program and constitutes the investigator's primary reference document for understanding the drug's known profile.

The IND, the protocol, the IRB review, the DSMB, the FDA Advisory Committee, and the post-market surveillance program are an institutional implementation of four questions from the Skeptic's Toolkit:

Q4. Funding and stake. Sponsors fund their own trials. Disclosure is the audit mechanism, not the absence of bias.

Q8. Sample size and replication. Phase 3 powering is set by the smallest effect size that would change clinical practice. Two pivotal trials are typically required, not one.

Q9. Independent expertise. Advisory Committees are seated with conflict-of-interest disclosures. Their public votes are the highest-credibility review available outside the sponsor.

Q10. The cost of being wrong. 21 CFR Parts 312 and 314 define the recall authority and the post-market commitments that constrain what the system can credibly approve.

The toolkit is the thinking that produced the regulatory framework. Run any drug claim you encounter through the same four questions.

The full Skeptic's Toolkit is in Issue 003b.

The second component is the clinical protocol, the detailed plan for how the trial will be conducted. A complete protocol specifies the primary and secondary endpoints, inclusion and exclusion criteria, dosing regimen, visit schedule, procedures for handling adverse events, statistical analysis plan, and stopping rules. The stopping rules are particularly important: they define in advance the safety thresholds at which the trial would be paused or terminated, and they are reviewed by an independent Data Safety Monitoring Board, or DSMB, throughout the trial.

The third component is the chemistry, manufacturing, and controls section, or CMC. This covers how the drug is made, its composition, stability, and the controls in place to ensure that what goes into participants is what the protocol specifies. Inconsistent manufacturing is one of the more common reasons INDs receive clinical holds.

The FDA's 30-day review window is statutory, established under 21 CFR 312.40. If the FDA does not issue a clinical hold within 30 days of IND receipt, the sponsor may proceed with the clinical investigation. A clinical hold can be partial, affecting only certain aspects of the trial, or complete, halting all proposed clinical work. The most common grounds for a complete clinical hold are unreasonable risk to participants, insufficient preclinical data to support human testing, or an inadequate clinical protocol.

Clinical holds are not uncommon, and they are not the end of a program. They initiate a dialogue between the sponsor and FDA. The sponsor responds to each issue raised. FDA reviews the response. The hold is lifted when the agency is satisfied the concerns have been addressed.

An endpoint is the specific outcome that a trial is designed to measure. The primary endpoint is the single most important outcome, pre-specified in the statistical analysis plan before any trial data is unblinded. For oncology trials, primary endpoints typically include overall survival, progression-free survival, or objective response rate. For other therapeutic areas they might include reduction in symptom score, time to event, or biomarker change.

The choice of primary endpoint determines the statistical requirements for the trial. If the primary endpoint is overall survival, the trial must continue until enough death events have accumulated to achieve the pre-specified statistical power, which may require years of follow-up beyond the end of dosing. Regulators and sponsors negotiate the primary endpoint and the statistical analysis plan before the trial begins. Changing the primary endpoint after unblinding constitutes an analysis practice that can invalidate the entire dataset.

Randomization is the process by which participants are assigned to treatment groups by chance rather than by investigator or participant preference. Its purpose is to distribute known and unknown confounding variables equally across groups, so that any observed difference in outcomes between groups can be attributed to the treatment rather than to pre-existing differences between the groups.

Simple randomization assigns participants with equal probability to each group. Stratified randomization ensures balance across pre-specified variables, such as disease stage, age group, or geographic region, that are known to be associated with outcomes. Block randomization ensures that at any point in enrollment, the ratio of participants across groups remains approximately balanced.

A single-blind trial is one in which participants do not know which treatment they are receiving. A double-blind trial extends that concealment to the investigators as well. Double-blinding is the standard for trials with subjective endpoints, because investigator knowledge of treatment assignment can introduce assessment bias even when the investigator is acting in good faith.

Trials with clearly distinguishable treatments, such as surgery versus medication, use blinded outcome assessment as an alternative: a centralized committee reviews outcome data without knowing the participant's treatment assignment. This is sometimes called a blinded endpoint adjudication committee.

The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use, known as ICH, is a joint initiative of regulatory authorities and the pharmaceutical industry from the United States, European Union, Japan, and increasingly from other regions. Its purpose is to harmonize technical requirements for drug registration, so that clinical data generated in one country is acceptable to regulators in others. Without harmonization, sponsors would need to run separate trials in each jurisdiction, which would dramatically increase costs and delay patient access.

The ICH Efficacy guidelines cover clinical trial design, conduct, safety reporting, and analysis. The most fundamental of these for trial conduct is E6, Good Clinical Practice. The current version, E6(R3), was finalized in November 2023 and took effect in March 2025 in the US, EU, and Japan.

ICH-GCP E6(R3) opens with thirteen principles that govern all subsequent guidance. These are worth understanding as principles rather than rules, because they explain the intent behind the specific requirements. Among the most important:

The rights, safety, and well-being of trial participants take priority over the interests of science and society. This is the foundational principle, and it is the one most often in tension with sponsor and investigator incentives to complete trials on schedule.

The available nonclinical and clinical information on an investigational product must be adequate to support the proposed clinical trial. This is the preclinical adequacy requirement that governs when a compound can proceed to humans.

Clinical trials must be scientifically sound and described in a clear, detailed protocol. Ambiguity in the protocol is not a minor administrative problem. It is a data integrity risk that can invalidate findings.

The investigator must be qualified by education, training, and experience to conduct the trial and must provide evidence of such qualifications. The site qualification visit, in which a CRO or sponsor representative assesses the investigator's facilities, staff, and experience, is the practical implementation of this principle.

The most consequential change in E6(R3) is the formal adoption of Risk-Based Quality Management, or RBQM, as the governing approach to trial monitoring. Previous GCP required near-exhaustive source data verification at clinical sites, meaning monitors reviewed essentially every data entry against the source document. In practice, for large global trials, this approach was resource-intensive, produced diminishing returns on the data that actually mattered, and failed to catch systemic issues that were not visible at the data-point level.

RBQM reorients monitoring around a risk assessment conducted before the trial begins. Critical data is defined as data that directly affects participant safety or the integrity of the primary endpoint. Critical processes are the activities that generate that data. Monitoring resources, both on-site and centralized, are concentrated on critical data and critical processes. Non-critical data is reviewed less intensively, but is still subject to audit.

Centralized statistical monitoring, in which anomaly detection algorithms flag data patterns inconsistent with the expected distribution, becomes a formal complement to on-site monitoring under E6(R3). Sites with unusual patterns, such as atypically high response rates, suspiciously low adverse event rates, or data entry timestamps that do not match the visit schedule, are escalated for targeted review. This approach can detect data integrity issues that site-level source data verification cannot.

E6(R3) provides regulatory clarity for decentralized clinical trial elements, meaning trial activities conducted outside the traditional clinical site setting. Remote assessments via telemedicine, home nursing visits, direct-to-participant drug shipment, and electronic patient-reported outcomes are now explicitly within the GCP framework, provided they are described in the protocol and meet the same data quality standards as site-based activities.

The enrollment implications are significant. Decentralized elements can reduce participant burden by eliminating travel to a clinical site for certain visit types. For rare diseases, in which the patient population is geographically dispersed, this expands the pool of eligible participants who can realistically enroll. The FDA's 2023 guidance on decentralized clinical trials and E6(R3) together form the current regulatory framework for DCT implementation.

The GCP framework did not emerge from theoretical ethics. It was built from institutional failures, each of which revealed a gap in the existing protections for research participants.

The Nuremberg Doctors' Trial, one of the Nuremberg Military Tribunals following World War II, prosecuted 23 physicians for conducting medical experiments on concentration camp prisoners without consent. The verdict produced the Nuremberg Code, the first international statement of research ethics. Its foundational requirement: the voluntary consent of the human subject is absolutely essential. The Code also established that human experimentation is only permissible when the expected benefits justify the risks, when the experiment is conducted by qualified scientists, and when participants may withdraw at any time.

The World Medical Association adopted the Declaration of Helsinki to extend the Nuremberg principles specifically to medical research conducted by physicians. It introduced the distinction between therapeutic research, in which the participant may directly benefit, and non-therapeutic research, and it established that the interests of the individual participant must take precedence over the interests of science and society. The Declaration has been revised seven times, most recently in 2013. It remains the foundational document for physician conduct in human research.

The Tuskegee Syphilis Study ran from 1932 to 1972. The United States Public Health Service enrolled 399 Black men with latent syphilis in Alabama and followed them without treatment for 40 years, including the 25 years after penicillin was established as an effective cure. Participants were not informed of their diagnosis and were actively prevented from receiving treatment. The study was exposed by a whistleblower in 1972 and terminated. The resulting public outcry produced the National Research Act of 1974, which created the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research.

The Commission's 1979 Belmont Report articulated the three foundational principles of research ethics still in use today. Respect for persons: individuals must be treated as autonomous agents and those with diminished autonomy must be protected. Beneficence: researchers must protect participants from harm and maximize possible benefits. Justice: the burdens and benefits of research must be equitably distributed and no group should bear disproportionate risks for the benefit of others.

ICH-GCP E6 was harmonized across the US, EU, and Japan in 1996, creating for the first time a single international standard for clinical trial conduct. Data from ICH-compliant trials conducted in one country became acceptable to regulatory agencies in others, dramatically reducing duplicative testing and accelerating patient access to approved treatments.

E6(R2) in 2016 introduced the concept of risk-based monitoring and acknowledged the role of centralized statistical oversight in ensuring data integrity. E6(R3) in 2023 expanded the framework to address decentralized trials, electronic data systems, and the updated risk management approach described above. Each revision has been triggered by a combination of identified gaps in the existing framework and advances in trial technology and design.

The FDA's advisory committee system is one of the most underused public resources in health literacy. Advisory committees are panels of independent scientists, clinicians, biostatisticians, and patient advocates who review drug applications and provide non-binding recommendations to the agency. Their meetings are public, their transcripts are published, and their voting records are available online.

An advisory committee meeting for a major drug application typically runs one to two full days. The sponsor presents the clinical evidence. FDA reviewers present their independent analysis, which often identifies concerns the sponsor did not highlight. The committee asks questions, deliberates, and votes. The FDA is not bound by the committee's recommendation, but it follows it in the large majority of cases.

The value of advisory committees for an evidence-literate public is considerable. The transcript of an oncology drug advisory committee meeting contains the actual clinical data, the FDA reviewer's statistical assessment, the sponsor's response to FDA concerns, and the independent committee's evaluation. It is more detailed than any press release and more accessible than the NDA itself. For any drug that matters to you or your family, the advisory committee transcript is the primary source.

The New Drug Application, or NDA, is the formal submission requesting FDA approval to market a drug. A typical NDA contains the complete clinical trial data, all preclinical studies, manufacturing and quality control information, proposed prescribing information, and a patient labeling package. The NDA is organized into a standardized electronic format called the Common Technical Document, or CTD, which is accepted by regulatory agencies in ICH member countries, allowing a single submission package to support multiple simultaneous regional applications.

For most NDA reviews, the FDA assembles a multidisciplinary review team: a medical officer who evaluates clinical efficacy and safety, a pharmacologist who reviews preclinical data, a biostatistician who independently analyzes the trial data, a chemist who reviews manufacturing, and a clinical pharmacologist who assesses pharmacokinetics. Each reviewer produces an independent assessment. The disagreements between reviewers, when they exist, are documented and resolved through the agency's internal deliberative process.

Priority Review designation, granted for drugs addressing serious conditions with unmet medical need, reduces the FDA review target from ten months to six months. Breakthrough Therapy designation, granted when preliminary evidence shows substantial improvement over existing treatments, provides intensive FDA guidance throughout development, which typically accelerates the overall timeline. Accelerated Approval allows drugs for serious conditions to be approved based on a surrogate endpoint that is reasonably likely to predict clinical benefit, with post-marketing trials required to verify the clinical benefit.

FDA approval is not the conclusion of a drug's evidentiary development. Post-marketing surveillance, or Phase IV, continues after a drug enters commercial use. The real-world patient population is broader than the trial population: older, with more comorbidities, taking more concomitant medications, and using the drug in ways the trial did not anticipate. Rare adverse events that did not appear in a trial of 3,000 participants may emerge in a population of 300,000.

The FDA Adverse Event Reporting System, or FAERS, collects voluntary reports of suspected adverse drug reactions from healthcare providers, consumers, and manufacturers. Mandatory expedited reporting of serious unexpected adverse events applies to the manufacturer. The FDA conducts ongoing benefit-risk assessments and can require label updates, add Risk Evaluation and Mitigation Strategies, or withdraw approval if post-marketing data reveals an unacceptable risk profile.

The withdrawals that do occur, Vioxx in 2004, some forms of Avandia in 2010, are often cited as evidence that the approval process is inadequate. They can also be read as evidence that the post-marketing surveillance system eventually identifies risks the trial could not detect, and that the regulatory response to those findings is real. The question is not whether the system is perfect. It is whether it is responsive.

Next week we follow the pipeline into one of the most remarkable stories in modern medicine: the Philadelphia chromosome. In 1960, two scientists at the University of Pennsylvania looked at leukemia cells under a microscope and noticed something that did not belong. A chromosome shorter than it should be. The scientific community did not notice for 13 years.

For a visual reference of the Philadelphia chromosome and the BCR-ABL translocation, see the Wikipedia article Philadelphia chromosome and the public-domain NCI Visuals Online image library.

Issue 005 traces the path from that first observation to the compound that would eventually save 150,000 lives and change what a cancer diagnosis could mean. We will also meet Debby's next claim: that Big Pharma hides cancer cures. Sam will apply the toolkit. The answer is more complicated, and more interesting, than either side tends to admit.

Issue 005 drops Thursday, May 15.

1. 21 CFR Part 312: Investigational New Drug Application. Electronic Code of Federal Regulations. ecfr.gov/current/title-21/chapter-I/subchapter-D/part-312

2. ICH. E6(R3) Guideline for Good Clinical Practice. International Council for Harmonisation. November 2023. ich.org/page/efficacy-guidelines

3. FDA. Guidance for Industry: E6(R3) Good Clinical Practice. March 2025. fda.gov/regulatory-information/search-fda-guidance-documents

4. FDA. Guidance for Industry: Decentralized Clinical Trials for Drugs, Biological Products, and Devices. May 2023. fda.gov/regulatory-information/search-fda-guidance-documents

5. The Nuremberg Code. Reprinted in: Trials of War Criminals before the Nuremberg Military Tribunals. US Government Printing Office, 1949. hhs.gov/ohrp/regulations-and-policy/belmont-report/index.html

6. World Medical Association. Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects. 2013 revision. wma.net/policies-post/wma-declaration-of-helsinki

7. The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. 1979. hhs.gov/ohrp/regulations-and-policy/belmont-report

8. Reverby SM. Examining Tuskegee: The Infamous Syphilis Study and Its Legacy. University of North Carolina Press, 2009.

9. Friedman LM, Furberg CD, DeMets DL, Reboussin DM, Granger CB. Fundamentals of Clinical Trials. 5th ed. Springer, 2015.

10. FDA. Advisory Committees. fda.gov/advisory-committees

11. FDA. Adverse Event Reporting System (FAERS). fda.gov/drugs/surveillance/questions-and-answers-fdas-adverse-event-reporting-system-faers

12. ICH. E8(R1): General Considerations for Clinical Studies. 2021. ich.org/page/efficacy-guidelines

13. ClinicalTrials.gov. Results reporting requirements. clinicaltrials.gov/about-site/overview