Jefferson Office of Human Research Protection

This is the current FDA guidance (2006) on DSMBs.  There is a guidance still in draft form (Feb 2024) in the comments period.

Establishment-Operation-Clinical-Trial-Data-Monitoring-Committees_Nov-2021.pdf

FDA has only one type of study that requires a DSMB, when trials occur in emergency situations and informed consent may be waived.  They have issued a comprehensive guidance that gives their current thinking on the topic.  The guidance suggests that DMSBs be considered when trials have mortality or major morbidity endpoints; large, randomized trials with treatments intended to prolong life or reduce risk of major adverse health outcomes.  They are generally recommended for controlled trials of any size that compare rates of mortality or major morbidity.

Generally, a DSMB is not needed for trials at the early stage of product development or for trials addressing lesser outcomes unless the trial population is at elevated risk of more severe outcomes.

National Institutes of Health (NIH) requires all multicenter, phase 3 studies have DSMBs.  Some institutes in NIH require DSMBs if a study is blinded and randomized.  NIH recognizes that DSMBs are needed when:

• Large, randomized trials with mortality or major morbidity endpoints, implications for clinical practice and/or public health
• Trials for which assessment of serious toxicity requires comparison of rates
• Trials of novel, potentially high-risk treatments event if unblinded
• Highly vulnerable patient populations

Not needed for single arm trials, early phase trials, short trials of treatments to relieve common symptoms, any trial for which there is not an ethically compelling need to monitor the interim comparisons of safety or efficacy.

Jefferson Guidance G 616, section 2.3

The following research situations require the oversight of a DSMB:

• The study is intended to provide definitive information about the effectiveness and/or safety of a medical intervention
• Prior data suggests that the intervention under study has the potential to induce a potentially unacceptable toxicity
• The study is evaluating mortality or another major endpoint, such that inferiority of one treatment arm has immediate implications for research participants regarding both safety and effectiveness; or
• The primary question has been definitively answered, even if secondary questions or complete safety information have not yet been fully addressed.

If the blood draws in question are standard draws, the person doing the draw needs to be qualified to perform the procedure but does not need to be study staff (i.e. does not need to be on the DOA or trained on the protocol).

Guidance-for-Staffing-of-Personnel-on-Research-Studies.pdf (jefferson.edu)

To simplify to the easiest rule of thumb, any study involving greater than minimal risk requires written consent. Studies involving minimal risk may also use written consent, or if written consent is waived, another route of consent can be used, such as verbal or text based. The determination is always risk-based. HIPAA authorization may still apply.

Yes, they can be submitted in tandem.  The IRB can even approve a study without an IND approval, but the IRB approval letter will not be released until the IND is approved so the study cannot begin. 

If the same issue exists for more than one participant, only file one UAP and list all of the affected participants in the Subject ID space.  If they do not fit in the Subject ID space provided, they should all be listed in the text body.

While there is no official recommendation for a translation service, several studies use Language Service Consultants (LSC) - with Ruth Karpeles as our contact.  It is recommended that a certificate of translation/authentication be obtained as both documentation of translation and LSC will then use translators certified in the language pairs in question.  There is a small fee for this certificate/letter.

This is a concern primarily for validated survey instruments, which are language specific.   For studies using validated survey instruments, this is not allowable even for minimal risk studies.

Non-validated surveys or instruments generally operate on a lower evidence base or stringency.  Investigators could use these WITH CAUTION but this should be explained in the IRB submission. There is not a risk issue but there is data validity issue.