8.1 Clinical Trials and Informed Consent
Key Takeaways
- Phase I pediatric oncology trials establish dose and safety, including dose-limiting toxicity; Phase II tests activity; Phase III compares an approach with standard therapy; Phase IV is post-marketing safety and real-world use.
- Children's Oncology Group (COG) is the major North American pediatric cooperative-group concept; many frontline studies randomize under clinical equipoise onto a standard-of-care backbone rather than a no-treatment control.
- Parents or legal guardians give legally effective informed consent for a minor; developmentally appropriate assent is sought from children who can participate; emergency exceptions from prospective consent are rare in oncology trials.
- Consent covers purpose, procedures, risks, benefits, alternatives, voluntariness, withdrawal without loss of standard care, and HIPAA authorization; children are vulnerable subjects and need extra protection, not a 03:00 signature without an interpreter.
- The CPHON nurse verifies eligibility, the currently approved consent version, and specimen windows; educates without coercing enrollment; documents protocol deviations versus violations; and notifies the research team of serious adverse events without inventing an IRB clock.
A 4-year-old with high-risk B-ALL is offered a Children's Oncology Group (COG) randomized study that adds an investigational antibody to a standard chemotherapy backbone. The parents ask whether the trial withholds real treatment, and the child watches you open the consent binder. Domain II.B tests whether a CPHON nurse can place that conversation in the phase system, the cooperative-group model, and consent-versus-assent—without selling enrollment.
Pediatric oncology is a research-driven specialty. Survival for many childhood cancers rose because children were treated on sequential cooperative-group protocols, not because each hospital invented a private recipe. That history does not make every trial the right choice for every family. It does mean you must speak fluently about phases, randomization, equipoise, and the difference between legally effective parental consent and a child's assent.
Trial phases: the question each study is built to answer
Phase is a scientific and regulatory label for the primary question, not a prestige ranking.
Phase I asks first about dose and safety. In children, a Phase I oncology trial typically escalates dose in small cohorts, watches for dose-limiting toxicity (DLT), and estimates a recommended Phase II dose. Tumor shrinkage may be recorded, but activity is not the powered endpoint. A toddler with multiply relapsed neuroblastoma offered a first-in-human agent is in a safety study. Teach that Phase I does not mean a guaranteed last chance that will work.
Phase II asks whether the agent or combination has activity—a response rate or another early efficacy signal—in a defined disease. Safety is still collected. A Phase II kinase-inhibitor study in relapsed ALK-positive anaplastic large-cell lymphoma is looking for tumor response, not yet for a definitive comparison with standard salvage.
Phase III compares the investigational approach with standard therapy (or the current best alternative). Pediatric Phase III trials are often randomized and sized to test event-free or overall survival. Many COG frontline studies randomize an addition or intensification onto a standard-of-care backbone, so the control arm is real treatment, not a sugar pill.
Phase IV occurs after marketing approval. These studies watch longer-term safety, rare toxicities, and real-world use. A post-approval registry of a CD19 chimeric antigen receptor T-cell product is a Phase IV-type surveillance concept, not a dose-finding study.
| Phase | Primary question | What families often mishear |
|---|---|---|
| I | Dose and safety (DLT, recommended Phase II dose) | This is the best chance because it is newest |
| II | Activity in a defined population | It already replaced standard care |
| III | Compare with standard (often randomized) | The control arm is no treatment |
| IV | Post-marketing safety and real-world use | Approval means remaining risks are zero |
Do not invent a percentage of children cured on each phase as an ONCC fact. Teach the question the phase is built to answer.
COG, randomization, and equipoise
COG is the major North American pediatric cooperative group that designs and runs most multi-institution childhood cancer trials. Other consortia exist, but CPHON items use COG as the named cooperative-group concept. A child on a COG protocol is on a shared, IRB-reviewed roadmap used across centers, not a one-off attending preference.
Randomization assigns the child to a treatment arm by chance after eligibility is confirmed and consent is obtained. It is not the physician choosing the better arm in the hallway. Clinical equipoise is the ethical premise that makes randomization acceptable: the expert community is genuinely uncertain which arm is superior. If the team already knows one arm is better, randomization is not ethical.
The standard-of-care backbone is the chemotherapy, radiation, or surgery that remains in both arms, or in the control arm, so the child is not randomized to neglect. A new antibody randomized onto ALL induction still sits on vincristine, a corticosteroid, asparaginase, and intrathecal therapy as the protocol specifies. Teach: everyone receives proven backbone treatment; the study question is whether adding or changing this piece is better.
Placebo-only control is uncommon in pediatric curative-intent oncology. If a placebo is used, it is typically added to active backbone therapy. Do not tell families that cooperative-group trials withhold chemotherapy to create a control.
Informed consent versus assent
Parents or legal guardians give legally effective informed consent for a minor. The child's signature does not replace that consent. Assent is the child's affirmative agreement after a developmentally appropriate explanation. Infants cannot assent. School-age children can often understand needles, extra blood draws, and that a computer sometimes picks the medicine. Adolescents can usually understand randomization, extra scans, and the right to say they do not want to be in the study.
Assent is sought when the child is capable of a meaningful yes or no. A child's dissent is taken seriously, especially for non-therapeutic research procedures. Document the discussion; do not coach a frightened 8-year-old into a cheerful signature because the study needs accrual.
Emergency exceptions from prospective consent are rare in pediatric oncology trials. Exception-from-informed-consent research is a narrow emergency-research pathway, not a weekend shortcut for a new leukemia admission. There is time to identify the legally authorized representative, use an interpreter, and review the current consent. Do not invent a weekend waiver because the attending wants the child on study before Monday. Do not invent IRB approval clocks either; use the currently approved form and the institution's process.
Required elements, HIPAA, and vulnerable subjects
A valid research consent discussion covers, in language the family can use:
- Purpose of the study and why this child is invited
- Procedures, including extra blood, extra visits, imaging, and what is standard versus extra for research
- Risks of the investigational piece and of the known backbone
- Benefits that may reasonably occur, without promising cure
- Alternatives, including standard care off study
- Voluntariness: saying no does not reduce access to standard treatment
- Withdrawal at any time without penalty or loss of standard care
- Confidentiality and HIPAA authorization for use of protected health information in the research record
Vulnerable subjects include children as a class, plus wards of the state, prisoners (rare but still a regulatory category), economically or educationally disadvantaged families, and critically ill children whose parents are exhausted on night 2 of induction. Vulnerability means extra protection, not automatic exclusion from the only available trial. It does mean you do not obtain consent at 03:00 from a parent who has not slept, cannot read the form, and has no interpreter.
Consent is a process, not a single signature. Re-consent is required when the IRB-approved form changes in a way that could affect willingness to continue. You must use the currently approved consent version, not last year's photocopy in the drawer.
Deviations, violations, SAEs, and the nurse's job
A protocol deviation is a departure from the written protocol that does not, by itself, meaningfully increase risk or damage scientific integrity—classic example: a pharmacokinetic sample drawn outside the window because the child was in MRI. A protocol violation is a departure that does increase risk or compromise integrity—wrong dose, missed eligibility criterion, outdated consent used for a new procedure, investigational drug given by the wrong route. Centers define reporting pathways. Document factually, notify the principal investigator or research team promptly, and do not hide the event. Do not memorize an invented IRB must-report-in-24-hours clock as an ONCC fact. Report through the study and institutional mechanism as the protocol and IRB require.
A serious adverse event (SAE) results in death, a life-threatening experience, inpatient hospitalization or prolongation, persistent or significant disability, or a congenital anomaly—or, based on medical judgment, may jeopardize the child and require intervention to prevent one of those outcomes. Unblinding and sponsor timelines belong to the research team. The bedside CPHON nurse recognizes, treats, documents, and notifies so the SAE can be reported. A new seizure on an investigational kinase inhibitor is both a clinical emergency and a reportable event.
Operational checklist:
- Verify eligibility against inclusion and exclusion (age, prior therapy, organ function, pregnancy testing in adolescents of childbearing potential) before the first investigational dose.
- Confirm the correct, currently approved consent version is signed, dated, and on the chart, with interpreter documentation if used.
- Protect specimen windows and chain of custody: labeled tubes, timed draws, processing as the lab manual states.
- Educate about visits, diaries, and who to call for fever without coercing enrollment. Saying that most families on this unit join the study is coercive. Saying that standard care is available whether you enroll or not is the professional line.
- Keep research procedures distinct in teaching: extra research blood is not the same as a clinically indicated CBC.
A 16-year-old who says they will sign whatever so chemo can start tonight still needs the alternatives conversation. Speed of induction is not a reason to skip voluntariness.
The CPHON product is a family that can state the phase question, name COG as the cooperative group when that is the study, distinguish backbone care from the investigational piece, and leave the conversation free to say no.
A toddler with multiply relapsed neuroblastoma is offered a first-in-human investigational agent after standard options. Which statement correctly describes a Phase I pediatric oncology trial?
A 9-year-old with newly diagnosed standard-risk ALL is offered a COG randomized trial that adds an agent to a standard-of-care backbone. Which consent and assent statement is correct?
During week 4 of a trial, yesterday's pharmacokinetic sample was drawn 90 minutes late, outside the protocol window, and the consent on the chart is last year's version. What is the priority nursing action?