on Abigail Alliance Petition below
Update: The FDA has
made substantial changes to the Expanded Access program
following consideration of comments from stakeholders.
See Access to Investigational
Drugs Outside of a
Clinical Trial (Expanded Access)
Regarding the Updated Expanded Access
It remains challenging for cancer survivors
in medical crisis to make use of the Expanded Access program, with drug
sponsors and demands on physician time appearing to be the primary
obstacles. (Neither the sponsors or physicians could be
forced to participate - noting also the challenge for physicians
to select an appropriate study drug (a single agent) in an advanced clinical
circumstance - that is arguably more likely to do more harm than
However, from my perspective, drug sponsors should consider setting up Expanded
Access trials for investigational drugs that they feel have promise and
genuine potential to address unmet needs for the
(1) Demand for a study drug would be a strong signal that
the new drug has the potential to fill an unmet need (a way to
demonstrate confidence in the investigational product),
(2) The data from the trial could be used to
support marketing approval - helping to validate safety and efficacy
the primary study, without impacting accrual.
(3) Because of new regulations, the sponsors can charge for providing the study drug
under expanded access to mitigate costs.
Karl Schwartz (January 2011)
Submitted Comment on Petition by Abigail Alliance
- September 4, 2003
Initial Approval Program to Expedite the Availability of Lifesaving
– Citizen Petition of the Abigail Alliance and the Washington Legal
To comment or endorse this letter
to send an email.
Mark McClellan, MD, PhD
Food and Drug Administration
Dear Dr. McClellan,
I am the president of Patients Against Lymphoma,
and have served as a patient consultant to the FDA.
My comments, however, represent only my personal perspectives,
which I believe to be relevant because of my training and seven-year
support of patients living with lymphatic cancers and their
I wish first to express my deep sympathy with the
needs of patients that have led to the Petition – I recognize as
legitimate the desire for patients to have autonomy and to have access
to investigational treatments. Facing death, a patient will
want to try an unproven path over a path proven to fail. In a
similar vein, it can be argued that it's illogical to try first what
is both toxic and ineffective before trying some
However, I have concerns with the Petition. I
will start with the proposal to allow for Tier 1 approvals based on
preliminary Phase I or II study outcomes.
Initially, the idea might sound good to patients, but in
practice it is likely to have the following important and undesirable
There will be increased risks to patients using Tier 1
Statistically, most drugs that complete Phase I will
be judged not suitable for approval.
Adding to the risks for patients would be that treating
physicians, instead of trained investigators, would monitor patients
receiving new and poorly-characterized drugs in local centers, perhaps
without adequate resources and time.
It will be more difficult to assess the merits and risks
of tier 1 approved drugs in these less controlled settings.
This could slow the pace of approval and thereby inadvertently
harm the larger patient community who must wait for full approval
before having access.
It proposes a commercial solution while making only
modest recommendations for enhancing the compassionate use [Expanded
which we address below.
It does not adequately
speak to the potential downside of reducing standards of evidence,
which can undermine the public confidence in marketed drugs and create
confusion about which drug is best and how it is best administered.
For example, if three [or forty] new therapies gain Tier 1 approval for a
condition, which of these is best, safest, or most dangerous?
It does not adequately
address how the sponsor can collect adequate and reliable data in
post-marketing studies in order to establish safety and efficacy after
provisional approval. [Clinical
practice cannot contribute to safety and efficacy data reliably.]
It can have negative effects on the completion of formal
clinical trials by distracting the sponsor from the primary goal of
winning full approval, and by confusing patients as to why they should
enroll in a study for a treatment that’s already approved and can be
purchased by some patients.
We can expect that insurance companies will not
reimburse patients for provisionally approved treatments, which is
required in order for sponsors to recoup investments and become
The proposed solution would exclude the majority of
patients, who will be unable to pay for Tier 1 drugs out of pocket.
Even if the petition changed FDA policy, drug sponsors
may not participate or may be less inclined to develop new drugs
because of the following concerns.
The program would be complex and distract from the more
profitable and therefore primary objective of winning full approval
– required for insurance reimbursement.
Concerns about litigation resulting from adverse events
that are more likely to result when going too fast with a new drug
(about which less is known in less controlled settings).
Adverse events in less controlled settings could be
considered as increasing the risk of not winning full FDA approval.
Producing drugs on demand in small orders can be
much more expensive. It would be difficult to predict manufacturing
requirements in practice, and difficult for FDA to confirm good
manufacturing practices. This could further delay or compromise
winning full approval.
Compassionate Use [Expanded Access] Reform - Access is Not Working
Update: Since this
commentary was submitted to FDA, the agency has made substantial
changes to the Expanded Access program following consideration of
comments from stakeholders.
See Access to Investigational
Drugs Outside of a Clinical Trial (Expanded Access)
proposes a commercial solution while making only modest
recommendations for enhancing the compassionate use program.
The compassionate use system is not accessible
and when implemented often amounts to too little too late. It requires
the cooperation and consent of the drug sponsor, the
treating physician, the FDA, and an IRB.
The majority of patients are not aware that the program exists,
and treating physicians are unlikely to inform patients or make
compassionate use recommendations [ or to have information sufficient
to make the recommendation].
Getting all the parties to cooperate takes considerable time,
energy, persistence and luck.
In advance of approval of a CU protocol many
important questions can be raised. Has the requesting patient and/or
physician done their homework? Is
it the best treatment option? Is
someone failing to recognize that it’s time to let nature take its
How sensible the decision is depends on the type
of cancer, the previous treatments, the health of the patient, the
mechanism and safety profile of the investigational agent, and whether
the potential to help is reasonably established by some preliminary
data - and perhaps if the therapeutic target is relevant to
I think it’s self-evident that a functioning
compassionate use system is needed for cancer patients because all too
often standard therapies are not effective, and all too often cancer
patients are not eligible for clinical trials.
Here are some well-known reasons patients
cannot enter clinical trials they desire:
The appropriate study might not be offered locally and
travel can be a hardship or impossible.
Target enrollment numbers in many phase I and II
studies are small.
Study criteria exclude many from participation.
Studies are typically designed to treat
patients in narrowly defined diagnostic and treatment
settings, based on calculations on the fastest way to
win approval. Large numbers of patients with similar
diseases, who could reasonably try the protocol, are often
About eligibility criteria for compassionate
(in addition to being ineligible for a clinical trial)
“Desperately ill” and having “immediately
as requirements for eligibility.
These are inappropriate criteria for access to CU
status implies advanced disease and a circumstance where no
therapy no matter how advanced is likely to benefit the patient. That
said, with the improving specificity and better toxicity profiles
of many emerging treatments it’s reasonable to expect that it
will be appropriate to administer CU protocols, increasingly, even
to patients with rapidly declining health and end-stage disease.
criteria for eligibility should be the following for patients who
have life-threatening cancers and are not eligible for clinical
standard therapies are toxic and have demonstrated little or no
ability to benefit patients for a given cancer.
the patient is refractory to standard treatments.
an investigational agent is established to be significantly less
toxic than standard treatments and has a potential advantage
that does not preclude the latter use of toxic standard
NOTE: This last criterion could be most appropriate and best
applied when there is no need for an immediate treatment
response – the patient is not desperately ill – such as when
in watchful waiting status.
A Proposal for Optimizing the Compassionate
(for cancer patients in need who are NOT eligible for clinical
The proposed changes to compassionate use (CU) in
the Petition do not go far enough, in my view, to address the
inability of the system to deliver appropriate investigational
therapies in a timely manner.
For example, it’s widely acknowledged that
access to CU protocols is mainly restricted by the drug sponsor’s
declining to make the drug available, and not by existing FDA
guidelines. This key problem was not addressed by the petition.
for optimizing the CU program:
Set up an NCI-directed panel to create CU trial
protocols in anticipation of needs that utilize promising new
agents, or new uses of existing agents, that appear promising for
patients in a variety of treatment settings. The panel should include
experts in the field, in consultation with patient advocates and
treating oncologists, with the primary goal of designing protocols
that address unmet needs.
NOTE: The selection of an
investigational agent for a CU protocol will be seen as an indicator
of the potential of the therapy to benefit patients, and therefore
sponsors who have confidence in their agents will be likely to
welcome the program.
The program should be conducted by the NCI and funded
by both public and commercial monies.
The program will require that drug sponsors provide
investigational and approved drugs to the NCI for CU studies, as part
of deal for the promise of patent protection for drugs that win
Would an expanded
CU system harm accrual in clinical trials?
By definition, a CU protocol is open to patients who are
not eligible for other studies. Unlike the present system, CU
protocols will be better powered than individual INDs, and they will
be structured like formal studies to provide credible data.
Administrators can also refer patients to formal studies for
the agent when openings exist.
be the benefits of providing ready-to-use CU protocols?
These protocols may help some patients who have
failed standard treatments, and they will provide meaningful
rationally-selected alternatives to the often-heard refrain that
“There is nothing more that we can do for you. You should get your
affairs in order.”
They will investigate promising protocols in settings
that are not currently explored in our commercially driven system;
for example, frontline biological approaches for indolent
They can be rationally designed and selected
by experts to treat many patients, instead of being selected by treating
physicians for individual patients in crisis.
They can be administered on a timely basis – unlike
single-patient INDs that take months to apply for and will
require the cooperation and consent of the drug sponsor, the
treating physician, the FDA, and an IRB.
They will be available to more patients, not just the
patients who have the ability, resources and information necessary to
apply for single-patient INDs.
Safeguards Against Misuse
We must recognize that giving patients the freedom
to try experimental therapies when trials are not available is
not necessarily in their best interest. Inherently, for new drugs,
less is known and significant dangers can exist. The outcomes
could conceivably be worse than had nothing been done. New is not
always better. These attempts may increase mortality, pain and
suffering, or both in some cases. However, this is true for any
clinical trial for cancer and should not be used as a rationale
not to provide rationally selected investigational treatments to patients ineligible
Existing safeguards against misuse need only to be
applied: It's the duty of the treating physician, the
principle investigators, and in some cases the patient's
guardians, to thoroughly inform the individual about the potential
risks and benefits of a proposed treatment, and to fully describe and
compare it to any viable alternatives. These safeguards should apply
to the initiation of any treatment – be it investigational or
standard, be it in a formal study or for compassionate use.
We also need to help patients and caregivers to
recognize that sometimes – when the cancer is advanced and health
has been compromised by treatment and disease – additional efforts will
do more harm than good, and allow these individuals to die
with dignity and help instead to minimize pain and suffering.
In conclusion, these two important goals must not
be considered mutually exclusive:
the best possible evidence-based standards to ensure that patients
have reliable information about the risks and benefits of marketed
treatments for given conditions, and increasing incentives for
companies to develop new therapies to meet our needs.
cancer patients with every reasonable chance to improve their
survival and quality of life.
A planned and optimized CU program will be
able to provide meaningful safety and efficacy data. It will be
much easier to make CU protocols available to patients in a timely
manner. These changes should not influence accrual in similar
formal clinical trials. The current CU program helps very few
and teaches us less. We think that the well-intentioned Petition breaks more than
it fixes. We think the correct remedy is not to weaken the
evidence-based evaluation system, but to ensure that additional CU or
expanded access protocols are developed and are offered when early
evidence suggests that an investigational protocol has the potential
to help cancer patients in need.
President, Patients Against Lymphoma