William M. Pierce, Ph.D.
Executive VP for Research and Innovation
University of Louisville
Jouett Hall, Room 202B
Louisville, KY 40292
Research Project: Baclofen with Locomotor Training: Effect on Function and Neuroplasticity in Chronic Spinal Cord Injury [“Baclofen” Study]
Principal Investigator: Dr. Susan Harkema
HHS Funding Agency: National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR)
IRB Study Number: IRB #12.0151
Research Project: Spinal Epidural Electrode Array to Facilitate Standing and Stepping in Spinal Cord Injury [“EPI ORIGINAL”]
Principal Investigator: Dr. Susan Harkema
HHS Funding Agency: National Institute of Biomedical Imaging and Bioengineering
HHS Grant Number: 1R01EB007615
IRB Study Number: IRB #066.07
Research Project: Activity-Dependent Rehabilitation Model to Improve Bone and Muscle after SCI [“NMES"]
Principal Investigator: Dr. Susan Harkema
IRB Study Number: IRB #268.07
Dear Dr. Pierce:
Thank you for your July 22, 2016 report in response to our May 24, 2016 question and concern letter to the University of Louisville (UL) regarding allegations of noncompliance with Department of Health and Human Services (HHS) regulations for the protection of human research subjects (45 CFR part 46). Based on review of your responses as well as your February 1, 2, March 24, and 28, 2016 responses, we make the following determinations:
1) The complainant alleged that the investigator failed to report promptly to the institutional review board (IRB) certain unanticipated problems involving risks to subjects or others, contrary to HHS regulations at 45 CFR 46.103(b)(5). The complainant alleged that the following were unanticipated problems that, among others, were not reported to the IRB:
a. A subject involved in the Baclofen Study complained of ongoing right knee pain since participating in the study that was so severe that she needed to sleep with the joint cushioned between two pillows. On examination, this subject had a visibly displaced leg, and an x-ray revealed a deformed tibial plateau fracture. Muscle strain or joint sprain and fractures are listed as risks in the protocol and informed consent document, therefore such an event, if it occurred, would not be an unanticipated problem and therefore would not need to be reported. See however, paragraph #3 for a determination of noncompliance related to this incident.
b. A subject in the NMES protocol complained of chest pain associated with blood pressure elevations during training sessions and was sent to the clinic for evaluation. We note the following regarding this allegation with regard to subject 094_AHI: The consent form and the protocol for this study do not list chest pain as a risk. An episode of chest pain during research training for subject 094_AHI was not reported to the IRB, appropriate institutional officials, the funding agency, or OHRP. We also note that your audit found that adverse events for the study have not been tracked or monitored by the research team. We have determined that failure to report the unanticipated problem of chest pain is contrary to the regulatory requirements at 45 CFR 46.103(a) and 45 CFR 46.103(b)(5), which require that unanticipated problems involving risks to subjects or others, among other things, must be reported to appropriate institutional officials, the IRB, OHRP and the head of the sponsoring Federal department or agency.
Corrective Action: UL is implementing corrective actions to address this determination, including education by the research compliance monitors of Dr. Harkema and the research team on which events should be reported according to local guidelines. The research compliance monitors also educated the regulatory coordinator and research nurse on which events are adverse events and which are classified as unanticipated problems involving risks to subjects or others. The protocol and consent forms related to any standing and stepping protocols will be amended to include mention of chest pain. Subjects that are currently enrolled in these active protocols will be asked to sign the revised consent form and will be provided the opportunity to ask questions.
2) The complaint alleged that the investigator initiated certain protocol changes without prior IRB review and approval, contrary to HHS regulations at 45 CFR 46.103(b)(4)(iii). These regulations require that the IRB review and approve all proposed changes in a research activity prior to initiation of such changes during the period for which IRB approval has already been given, except when necessary to eliminate apparent immediate hazards to the subjects.
a. The complainant alleged that subjects for the Baclofen study were told that after insurance paid for the first forty training sessions and benefits were exhausted, they would have to pay for the remainder of the training (approximately $15,000), even though the protocol states that no participant will be personally responsible for costs associated with routine study participation. We note the following regarding this allegation:
i. Your response stated “There were inconsistencies noted in what was covered by insurance or what was covered by the study grant . . . . but physical therapy assessments in form of locomotor training [are] not included and are not coverable via funding from the grant.”
ii. The consent form for this study stated: “Differences between the regular patient and yourself would include: . . . . Any testing, visits, or medications required beyond what is needed for regular outpatient participation will be paid for by the study grant . . . . ” [emphasis added]
iii. The Info Session Explanation for this study stated “For any testing or training a typical outpatient would be asked to [perform], we would charge your insurance. You would be responsible for any copays or charges your insurance does not cover. For any testing that a patient would not be regularly asked to perform, the research department will cover those charges.” [emphasis added]
iv. The Baclofen protocol at page 7 states “3b. Initial Start-up . . .During their initial interviews, potential participants will be informed that any evaluations or examinations associated with the study, including physician visits, laboratory tests, medication, and LT sessions, will be provided at no cost to them . . . .” [emphasis added]
We have determined that subjects may have been misinformed regarding costs of research procedures that they would be responsible for, in violation of HHS regulations at 45 CFR 46.116(b)(3), which require that subjects be informed, when appropriate, as part of the informed consent of any additional costs to them that may result from participation in the research. This is also contrary to HHS regulations at 45 CFR 46.103(b)(4)(iii) which require that the IRB review and approve all proposed changes in a research activity prior to initiation of such changes during the period for which IRB approval has already been given, except when necessary to eliminate apparent immediate hazards to the subjects.
Corrective Action: UL is implementing the following corrective actions to address this determination: If the University allows the Baclofen study to continue, in order to reduce any future miscommunication or misinformation, the investigator will be required to revise the consent form to include a list of procedures that are covered by research funding and those for which the subject or insurance will be responsible. All subjects that are currently enrolled will be asked to sign the revised consent form and will be given the opportunity to ask any questions. We note that NIDILRR discontinued funding for this study.
b. The complainant alleged a subject was enrolled in the CV EPI protocol without having completed all screening laboratory tests described in the protocol. We note that your audit of this study found “Limited information as far as source documents to verify eligibility (i.e. lab results missing or not signed off by the MD).” We also note that your audit of the NMES study found that “The MD on the study is currently approving each subject to participate in the study before all screening procedures are completed. In addition, he is not signing off on lab values. There are research meetings to discuss the subject’s enrollment into the study, but they primarily consist of Dr. Harkema, Physical Therapist, Research Nurse, and Research Coordinator . . . .” We have determined that the investigator did not conduct the study as described in the IRB-approved protocol, contrary to HHS regulations at 45 CFR 46.103(b)(4)(iii).
Corrective Action: UL is implementing the following corrective actions to address this determination: The principal investigator has received training from the research compliance monitors concerning documenting inclusion criteria for each enrolled subject. The principal investigator has modified the standard operating procedures. Eligibility will not be determined until all screening procedures have been completed. The IRB is requesting that the study physician review all research medical reports.
3) The subject involved in the Baclofen Study complained of ongoing right knee pain since participating in the study (referenced above in paragraph 1a). OHRP notes that the evidence of joint injury for subject 203_BFZ described in the September 4, 2014 research locomotor training notes did not lead to discontinuation of training sessions and was not reported to the site physician, which was a deviation from the IRB-approved study protocol. The Baclofen Protocol dated January 31, 2014, on pages 24-25 states: “5a. Foreseeable Risks and Response: Locomotor Training Sessions. . . . . the physical therapist and research team will constantly monitor the research participant’s skin and muscle for signs of muscle strain, joint sprain, and skin irritation, as signs of skin redness, swelling of joints, or spasticity can be indicators of injury in individuals with impaired sensation. If any signs of risks or discomfort are noted, the experiment or training session will be immediately discontinued and the site physician will be informed.” [emphasis added] Subject 203_BFZ attempted to get the joint X-rayed but was unable to on two occasions. She was finally X-rayed about one year later and diagnosed with a fractured tibia. However, the study interventions continued in the absence of X-rays and in the presence of swelling and instability until October 22, 2014, risking further damage to the joint and was a deviation from the IRB-approved protocol.
We have determined that the investigator did not conduct the study as described in the IRB-approved protocol. This is in violation of the regulatory requirements at 45 CFR 46.103(b)(4)(iii) which require that the IRB review and approve all proposed changes in a research activity prior to initiation of such changes during the period for which IRB approval has already been given, except when necessary to eliminate apparent immediate hazards to the subjects.
Corrective Action: UL is implementing the following corrective actions to address this determination: All study personnel listed on the site delegation log will be given a copy of each study protocol in which they are involved and will be required to attest that they have read and understand their responsibilities. Dr. Harkema has also hired a research nurse who will be performing weekly checks of both the physical therapy records and clinic notes. The physician listed on this study will be notified of any issues related to abnormal lab results or medical events and will document the findings and determinations. In addition, the principal investigator, physician, research nurse, and regulatory coordinator have received additional training related to protocol deviations, adverse events and unanticipated problems.
4) We note that your audits also found numerous additional findings of noncompliance in these research studies and have required the researchers to take corrective actions. Many instances of noncompliance identified in your audit were not violations of the HHS regulations for the protection of research subjects, and we do not describe them here. Additional violations identified in your audit that do pertain to the HHS regulations involved investigators not following the IRB approved protocol as required by 45 CFR 46.103(b)(4)(iii). The following are examples of the more substantive issues of noncompliance with the HHS regulations at 45 CFR 46.103(b)(4)(iii) that your audit revealed and the corrective actions you proposed:
a. The EPI-ORIGINAL protocol has a data safety monitoring plan outlined but there is no documentation that this has been followed to date. Corrective action: The principal investigator is in the process of executing a DSMB committee for all epidural stimulation related studies. The researchers have identified the members and have finalized the charter.
b. Lack of protocol-required source documents for the CV-EPI study leads to inability to verify if the protocol has been followed and data captured is valid, particularly related to inclusion and exclusion criteria. Corrective action: The principal investigator is revising the research procedures to implement the completion of an Eligibility Checklist which is completed and placed in the research medical record along with supporting documentation. These will be reviewed by the research manager, then further reviewed and signed off by the principal investigator to further document eligibility.
With respect to all of these determinations, these corrective actions are adequate to address the noncompliance.
5) We determine that there is no regulatory violation regarding the following allegations made by the complainant:
a. The NMES protocol specifically requires that participants not be on Baclofen, a centrally acting GABA agonist, but one subject took this medication for much of the study period, which the complainant alleged was contrary to HHS regulations at 45 CFR 46.103(b)(4)(iii). Dr. Harkema reportedly stated during meetings that she’ll use at least part of this subject’s data, though she doesn’t know during exactly which period(s) this subject was taking Baclofen nor the durable effect(s) of his having done so. We note your statement that the research participant informed Dr. Harkema and the research nurse that he was prescribed 10 mg of Baclofen when he was in the hospital for a viral infection but he did not take it during the study.
b. A subject involved in the EPI ORIGINAL protocol reported a right ankle ulcer. During training, the ulcer worsened and was eventually described as a “stage III” ulcer. It required surgical debridement at the Norton Wound Care Center and the subject was later seen on an urgent basis for cellulitis at the ulcer site. Another subject involved in the CV EPI protocol suffered bilateral foot ulcers during training, which were stage III on the right foot and stage II on the left foot. He was referred for evaluation and treatment at the Norton Wound Care Center, and his right-sided ulcer required surgical debridement. The complainant alleged that these events were not reported to the IRB, contrary to HHS regulations at 45 CFR 46.103(b)(5). We note that the consent form states: “Locomotor Training Risks . . . Skin irritation from the harness or hand placements of the trainers helping your legs (likely) . . . . Skin irritation from the sensors and/or wires (likely) . . .” Therefore, ulcers were not unanticipated problems that needed to be reported to the IRB. In addition, we note that the second research participant was not undergoing any training during the time period that he had the ulcers.
c. A subject was admitted to the hospital while participating in the NMES protocol and as a result nearly required an invasive procedure called a blood patch. We note that research participant 149_BWJ was admitted to Jewish Hospital on January 22, 2015 while also enrolled in the NMES study (Appendix #6). The complainant alleged that these events were not reported to the IRB, contrary to HHS regulations at 45 CFR 46.103(b)(5). The reason for consultation was a possible infection and the chief complaint of fever and chills. From our review, this event does not appear to be related to the study interventions.
d. A subject in the EPI ORIGINAL protocol suffered a broken hip during a training session on January 16, 2015 and that event was not reported to the IRB until January 30, 2015. The complainant alleged that this event was not promptly reported to the IRB, contrary to HHS regulations at 45 CFR 46.103(b)(5). This subject underwent multiple surgeries as a result of this fracture, and developed an infected hematoma with bacteremia as a result of one of them. As of August 15, the subject was still undergoing corrective surgical procedures, not all of which have been reported to the IRB. We note that fractures are anticipated and this is stated in the consent form. Thus, this event did not have to be reported as an unanticipated problem.
At this time, there should be no need for further involvement by our office in this matter. Please notify us if you identify new information which might alter this determination.
We appreciate the continued commitment of your institution to the protection of human research subjects. Please do not hesitate to contact me should you have any questions.
Kristina C. Borror, Ph.D.
Director, Division of Compliance Oversight
101 Wootton Parkway, Suite 200
Rockville, Maryland 20852
cc: Ms. Rebecca H. Higgins, Director, Human Subjects Protection Program, University of Louisville
Dr. Laura Clark, IRB Chairperson, University of Louisville
Dr. Peter Quesada, IRB Chairperson, University of Louisville
Dr. Susan Harkema, University of Louisville
Dr. Joanne Less, FDA
Ms. Bridget Foltz, FDA
Mr. John Tschida, Director, National Institute on Disability, Independent Living and Rehabilitation Research
Mr. Jeffrey Rodamar, Protection of Human Subjects Coordinator, U.S. Department of Education
Dr. Roderic I. Pettigrew, Director, National Institute of Biomedical Imaging and Bioengineering
Dr. Sherry Mills Rodgers, Director Office of Extramural Programs, National Institutes of Health (NIH)
Ms. Michelle Bulls, Director of OPERA, Office of Extramural Research, NIH Ms. Sarah Carr, Office of Science Policy, National Institutes of Health