February 7, 2003
Dr.
Leslie Ball
Office
for Human Research Protections
Department
of Health and Human Services
The
Tower Bldg., 1101 Wootton Pkwy., Suite 200
Rockville,
MD 20852
Subject:����������� UNC-Chapel Hill Request for HHS
Review under 45 CFR 46.407
����������� ���
����������� Re:� P50 HL 60280 � Dr. R. Boucher (Grant PI) �
Dr. T. Noah (Project PI)
Dear
Dr. Ball:
Please
accept this in response to Clifford Scharke�s letter to one of us (DKN), dated
October 24, 2002.� It is our
understanding that you have assumed responsibility for coordinating this
review.��� We appreciate the substantive
nature of questions raised, and have consulted further with IRB members and the
investigator, in order to answer them completely. We are providing the
additional information for each point raised by OHRP:
You
may already have some of these items from interim correspondence, but we are
enclosing a copy of the award application and the minutes of the IRB meetings
in which the deliberations regarding this study took place.� [Appendix IA and IB].� We are assuming that you still have all
materials previously submitted, so we are not re-enclosing all here.
The
scientific value of this study has been reviewed by the IRB and with Dr. Terry
Noah, the Principal Investigator.�
Background information supporting our analyses and conclusions is
summarized in the following paragraphs:
Prognosis and Proposed
Disease Model
This
study addresses the underlying pathogenesis of cystic fibrosis (CF).� This disease is fatal in 85 to 90 % of
individuals by the 4th decade of life without a lung
transplant.� There are proposed
mechanisms by which cystic fibrosis is thought to exert its effects on
pulmonary physiology.� In the model
supported by medical scientists at the University of North Carolina, there is
thought to be excess isotonic volume transport in the distal airways
with subsequent development of thickened mucus resulting in mucus plugging.� This mucus plugging predisposes the lungs to infection.�
This lack of water also impairs mucociliary clearance by decreasing mechanical advantage.� This decrease in mucociliary clearance leads to stasis of inhaled microbes. Pseudomonas has a special tendency to form biofilms.� Bacterial biofilm formation may occur early in life (weeks to months after birth).� Resulting inflammation is persistent and neutrophil-dominated, leading to progressive bronchiectasis and eventual death from respiratory failure.
The predicted effect of the above would be 1) accumulation of mucus (mucins plus other compounds) in distal airways, even prior to onset of infection/inflammation; and 2) a qualitative change in mucus density.� Because chronic infection becomes established so early in life in CF patients, this initiating event (accumulation of mucus) can likely be detected only in the first weeks to months after birth.� After that time, infection and inflammation (with secondary effects on mucus hypersecretion) are likely to obscure the original defect.
Existing Models in Cystic
Fibrosis
Pathology studies in older CF patients show increased expression of MUC5AC (one of the major mucins in airway) compared to normals1, though there has also been reported decreased MUC5AC expression in nasal epithelium from CF patients2.� [Appendix II � References].� Cell culture and mouse models (CFTR knockouts) for CF airway disease do show some evidence of reduced mucociliary clearance3.� However, these models have been disappointing for several reasons, as described by A.S. Verkman4� in a recent editorial:
�The testing of these hypotheses has presented a
formidable challenge because of difficulties in establishing suitable model
systems�cell culture models have been highly variable from laboratory to
laboratory; they cannot recapitulate the complex in vivo airway anatomy,
hormonal regulation, and cellular heterogeneity�.lung disease in mouse models
of cystic fibrosis is quite subtle.�
Also, there are a number of potentially important human versus mouse
species differences in airway physiology; airway submucosal glands are
essentially absent from mouse airways below the larynx, and the mouse airway
epithelium appears to express an alternative C1 channel that may substitute for
functionally defective CFTR.�
Measurements in intact normal versus cystic fibrosis human airways are
probably the most appropriate for studying ASL properties, recognizing the
caveat that human airway anatomy and function are altered in response to
recurrent infection and inflammation.�
Based
on the above, the IRB agreed with Dr. Noah that human studies remain of unique
value in CF research.�� Numerous infant
bronchoalveolar lavage fluid (BALF) studies have been performed and published
related to onset of infection and inflammation in CF, many from the PI�s
laboratory5-7, some for research indications only (at other
institutions), and
none
with adverse events reported.� To date,
no studies addressing mucus/mucin in early CF have been reported, nor do our
local investigators (among the leaders in the country in
this
area) know of any such studies in progress.�
This is because the methods for measurement of these factors have only
recently become available � and only at a handful of research centers.�
The proposed study thus
seeks to measure certain mucus and mucin characteristics in young CF patients,
while potentially providing the direct benefit of early identification of
bacterial infections for early initiation of specific treatment with
antibiotics.� Parameters measurable with
current technology include MUC5AC and MUC5B quantity (by ELISA) and
histological assessment of mucus plugs to determine their biochemical, cellular
and microbial contents.�
The
primary significance of the proposed study would be increased understanding
of� the pathogenesis of CF lung disease
at its earliest stages.� This could have
an impact on treatment, perhaps even for the subjects who would be involved in
the immediate study.� Several strategies
for improvement of mucus abnormalities in CF are actively being investigated at
present in older patients, including use of hypertonic saline8 ,
Nacystelyn9, and mannitol10.� These agents are designed to improve lysis and hydration of
mucus.� It is possible that promising
agents would move quickly to clinical trials in infants and children with CF,
if evidence favored an initial accumulation of mucus as the starting point for
CF airway disease.
Subjects
that would be included in this study are children from age 0-2 years.� All these children will have the diagnosis
of cystic fibrosis. However, the children will not have received any inhaled
steroids or oral drugs to reduce inflammation prior to bronchoscopy.� Children who have signs of active infection
or respiratory symptoms will be excluded.
This
study was reviewed by the convened IRB on 3 occasions:� 12/17/01, 02/04/02, and 04/15/02.� We are enclosing copies of minutes from all
three dates� [Appendix IB � Missing
Items].�� We call your attention to the
memo dated 02/05/02, in which we discuss the rationale for IRB disapproval and
our recommendation that this study be referred to OHRP for review by the panel
of experts convened under 45 CFR 46.407.
In
our memo dated 02/05/02, we note that bronchoscopy in an asymptomatic infant is not indicated.� Children at this age are unable to expectorate or provide sputum
samples.� Regardless of the pathogenesis
of this disease, these mechanisms ultimately lead to an increased incidence of
colonization and ultimately infection within the lung.�
When we reviewed this study on 02/04/02, with the information provided, we had concluded that bronchoscopy in asymptomatic infants was not indicated.��� Because we felt that bronchoscopy is a procedure which represents greater than minimal risk in this population, and because the prospect of direct benefit was uncertain, the IRB felt that
categories 46.404� and 46.405 were not applicable.� Similarly, because the IRB felt that the risk of bronchoscopy in an asymptomatic infant was of sufficient magnitude to be beyond
that described as a minor increase over minimal risk, we found that we could not approve this study under category 46.406.� These findings were reaffirmed in a third consideration by the IRB in April, 2002.
The IRB did conclude, however, that because of the scientific basis for this study and because there was a strong possibility of furthering the understanding of pulmonary disease in cystic fibrosis, the study had sufficient merit to be considered under Category 46.407.�
If a subsequent study were to show that early intervention with antibiotics in asymptomatic infants with colonization by pseudomonas delayed or prevented the progression of the pulmonary pathology in infants with cystic fibrosis, this would represent a shift in the thinking about the management of this disease.�
At
the time we made our initial determination in December 2001, and again in
February and April, 2002, we felt that the risk of the bronchoscopic procedure
would constitute greater than minimal risk, which was not offset by potential
benefit in the asymptomatic population.�
We did query Dr. Noah about the alternative of enrolling children who
already had lung disease (i.e. were symptomatic).� He noted that �the published rates for infection for
clinically-indicated procedures compared to research procedures are quite
similar.� In essence, clinical symptoms
are poor predictors of chronic bacterial airways infection in cystic
fibrosis.� Chronic infection, not acute
infection, is the main determinant of progressive lung disease and death�.� (See Noah response memo from March 2002).
Thus, limiting enrollment to symptomatic children would miss many who were already
infected.
Dr.
Noah�s response to the IRB memo dated 02/05/02 also addressed the question of
early treatment of pseudomonas or
other pathogens.� Recent information has
been presented at a meeting of investigators of cystic fibrosis� in Europe which looked at early intervention
with antibiotics.� The preliminary
results have not shown a problem with resistance to infection developing
early.� This data from the Copenhagen
Cystic Fibrosis Center also suggests that the outcomes are certainly not worse
and may be dramatically improved.� (T.
Noah, personal communication).
The
frequency of adverse events with bronchoscopy remain the same with repeated
bronchoscopies even if they are performed within a relatively short period of
time.� There
is
no cumulative effect.� Data from the
literature on pediatric flexible bronchoscopy is referenced in the table below,
taken from an article by Barbato11.�
|
|
< 5% of procedures |
5-10% of procedures |
>10% of procedures |
|
During flex. Bronchoscopy |
|
|
|
|
Bleeding |
33 |
2 |
- |
|
Bronchospasm |
34 |
2 |
2* |
|
Laryngospasm |
35 |
- |
1* |
|
Drug reactions |
32 |
2 |
1* |
|
After flex. Bronchoscopy |
|
|
|
|
Bronchospasm |
34 |
2 |
- |
|
Cough |
16 |
19 |
2 |
|
Fever |
30 |
5 |
3 |
|
Laryngospasm |
33 |
1 |
- |
Numbers are centers out of 51 surveyed.�� * Reported by centers performing <100 procedures per year.
The IRB considered the amount of compensation to subjects in the course of three separate reviews.� The IRB determined that the inducements offered in the study to both children and their parents are reasonable because of a 3-day stay in the hospital, as well as the travel and time loss for work.� It was felt that the inducement offered directly for the children sends a signal to the parents that this money should be used for the child�s benefit.� We would, however, be willing to reconsider this issue should the expert panel have other opinions.
OHRP questioned the lack of compensation for research-related injury.� While the IRB does not necessarily endorse this lack of coverage, the standard consent form language reflects current UNC policy.� As OHRP is aware, the regulations [45 CFR 46.116(a)(6)] only require that subjects be provided an explanation as to whether any compensation is available, and not that such compensation must be provided.� Regrettably, the IRB is not in
a position to alter institutional and/or state policy, on an ad hoc basis.� We have, however, secured a commitment from the PI and his Division to waive any professional fees stemming from medical care required to treat complications.� The parental permission form has been modified to provide further information for participating families.�� The form has�
also been modified to provide a space for each parent�s signature, in accordance with 45 CFR 46.408(b).�� [Appendix VI � Modified Parental Permission Form].�
Review by other
groups
This group, with considerable expertise in the area under study, concluded that the proposed study was implicitly approvable under 45 CFR 46.405 or 46.407.� While this came after and did not alter the determinations of our local IRB, it is reflective of the subjective nature of this process, and of the blurry boundary at which the risk-benefit ratio for this particular study lies.� In any case, the DSMB likewise encouraged Dr. Noah to pursue approval and conduct this study.
This study was originally included in the grant application that we have provided to you.� The proposed study modifies the population from children undergoing surgery for complications related to meconium ileus to the children described herein.� Despite this modification, the principal investigator for the grant, Dr. Richard Boucher, has agreed that this� project will be funded through the SCOR grant to UNC, as a substitute for the meconium ileus study.� Thus, the proposed study remains under NIH funding.
We appreciate your assistance and look forward to your further review. Please don�t hesitate to contact us if further questions arise.�
Sincerely,
Enclosures:
IB:�������������� Missing Items:�
Minutes of meetings of December 17, 2001, February 4, 2002 and April 15,
2002
Appendix� IV:�� ������������������� Complication
Data at UNC
Appendix
V:���� ����������� Modified
Parental Permission Form