Genentech, Inc. (“Genentech”) submits these
comments in response to the announcement by the
Centers for Medicare & Medicaid Services (“CMS”)
that it plans to reconsider its national coverage
decision on the Clinical Trial Policy (“NCD”).

Genentech is a leading biotechnology company
focused on discovering, developing,
manufacturing, and commercializing
biotherapeutics that address serious unmet
medical needs. Genentech has discovered and
introduced over a dozen significant therapies for
serious and life-threatening diseases affecting
the lives of Medicare patients, including cancer,
heart disease, pulmonary disease, rheumatoid
arthritis, and age-related macular degeneration.
The conduct of clinical trials is critical to
Genentech’s exploration of new therapies that may
address unmet medical needs.

Genentech supports the decision of CMS to revisit
the NCD in order to clarify the circumstances
under which Medicare will pay for costs
associated with clinical trials. The NCD was
intended to encourage Medicare beneficiaries to
participate in clinical trials by eliminating the
uncertainty about whether Medicare would cover
routine costs associated with clinical trials.
However, the NCD itself has led to considerable
confusion in the research community and among
beneficiaries about the scope of Medicare
coverage. Because the legal and financial risks
of improperly billing Medicare for trial-related
costs are so severe, research sites and sponsors
cannot rely on the NCD. Consequently, the NCD
has not been as effective as it should be in
encouraging Medicare beneficiaries to participate
in clinical trials.

In amending the NCD, CMS should:

• Clarify that Medicare will pay for study-
related costs for Medicare beneficiaries,
regardless of whether the sponsor of the trial is
paying for the study-related costs of other study
participants whose costs are not covered by
insurance.
• Ensure that IND-exempt studies continue
to be qualified trials under the NCD.
• Clearly state that all clinical trials in
Phase I- IV are qualified trials under the NCD,
except for Phase I trials in healthy patients.

Changes to the NCD should seek to expand the
opportunities for Medicare beneficiaries to
participate in clinical trials by reducing the
barriers to participation rather than by enacting
burdensome and unworkable new requirements on
clinical trial sponsors and sites. Therefore,
CMS should not impose requirements that studies
include a representative sample of Medicare
beneficiaries.

I. Comments on Issues Identified in the
Tracking Sheet

The issues identified by CMS in the NCA Tracking
Sheet for Clinical Trial Policy are important
areas of concern and we present our comments and
recommendations on each of these issues below.

1. Clarify payment criteria for clinical
costs in research studies other than clinical
trials.

Comments: Genentech requests that CMS clarify
this issue. Genentech offers no comments on this
issue at this time because it is unclear what CMS
means by “clinical costs in research studies
other than clinical trials.”

2. Devise a strategy to ensure that Medicare
covered clinical studies are enrolled in the
National Institute of Health (NIH) clinical
trials registry website.

Comments: Genentech supports efforts to ensure
that information about clinical trials is made
available to physicians and potential clinical
trial subjects. Consistent with that position,
Genentech posts protocol information regarding
all of its Phase II – IV trials on the
www.clinicaltrials.gov web site. However, we
believe it is important to balance the need for
disclosure with appropriate regard for the
protection of confidential and proprietary
information. Therefore, CMS should not require
sponsors to disclose information relating to
Phase I trials, which represent early inquiries
into the safety, pharmacokinetics, and activity
of investigational therapies. Requiring sponsors
to disclose these studies would force them to
reveal confidential information about their
investigational products to competitors at an
early point in the development of the product.

3. Develop criteria to assure that any
Medicare covered research study includes a
representative sample of Medicare beneficiaries,
by demographic and clinical characteristics.

Comments: Genentech welcomes initiatives by CMS
that would encourage Medicare beneficiaries to
participate in clinical trials, including through
educating beneficiaries about clinical trials and
removing barriers to their participation in
clinical trials. However, Genentech is troubled
by the implication that CMS might attempt to
implement a policy that would require a clinical
trial to include a representative number of
Medicare beneficiaries in order to qualify for
coverage under the NCD. The U.S. Food and Drug
Administration (“FDA”) – and not CMS – is the
federal agency principally responsible for
regulating the conduct of clinical trials and CMS
should not be creating policy related to clinical
trial design and conduct. Moreover, such an
approach raises a number of significant concerns,
including the introduction of bias into trials,
ethical considerations, and practical concerns.
CMS should not include any requirement that a
clinical study include a representative number of
Medicare beneficiaries in order to qualify for
coverage under the NCD.

a. Scientific Integrity: A significant
concern in biomedical research is the elimination
of bias in research. A key aspect of addressing
bias is to ensure that patients that enroll in a
trial are typical of all patients who are
eligible for the trial. We recognize that a
variety of social and economic factors may raise
barriers to participation in clinical trials and
we welcome efforts that would remove barriers
that discourage enrollment in clinical trials.
However, CMS should refrain from creating patient
selection criteria that introduce a non-
scientific basis for patient selection, which
could itself become a source of bias.

b. Human Subject Protection: The Belmont
Report, developed in 1978 by the National
Commission for the Protection of Human Subjects
of Biomedical and Behavioral Research, sets forth
the basic ethical principles for the conduct of
human subject research, including the principle
of justice. According to this principle, the
selection of research subjects must be the result
of fair selection procedures and result in fair
selection outcomes. Implementing any requirement
under which status as a Medicare beneficiary
would be considered in determining eligibility
for a clinical trial violates the principle of
justice.

In addition, tying Medicare coverage of clinical
trial-related costs to enrollment of a certain
number of Medicare beneficiaries would create a
significant economic incentive to enroll a
sufficient number of such patients. This would
place undue pressure on researchers to convince
Medicare beneficiaries to enroll in clinical
studies and could lead to coercive behavior that
would undermine the informed consent process.

c. Practical Considerations: There are many
practical difficulties with tying Medicare
coverage to the enrollment of a representative
sample of Medicare patients. If Medicare
reimbursement for clinical trial costs were to
depend on the enrollment of a certain number of
Medicare beneficiaries, researchers and Medicare
beneficiaries would not know whether Medicare
would cover clinical trial costs until enrollment
had been completed. This would cause enormous
confusion and further discourage participation in
trials by Medicare beneficiaries, because they
would not know whether a given trial was covered
under the NCD when they were deciding to enroll
in the study. In addition, study sites would not
know whether trial-related items or services
could properly be billed to Medicare until
enrollment was completed. In the case of multi-
center studies, this difficulty would be further
compounded, because any one site would not know
about patient enrollment at other sites in the
study.

CMS must avoid making changes to the NCD that
create additional confusion about Medicare
coverage for clinical studies. CMS should not
condition coverage for clinical trial costs, or
for the drug once it is approved, on the
enrollment of Medicare beneficiaries, because
that approach is unworkable and would cause
further confusion among researchers and Medicare
beneficiaries. Instead, CMS should encourage
enrollment of Medicare beneficiaries by
clarifying the terms of the NCD and removing
barriers to participation in clinical trials.

4. Clarify the definitions of routine
clinical care costs and investigational costs in
clinical research studies including clinical
trials.

Comments: CMS should restate the NCD so that it
is clear that Medicare will pay for study-related
costs for Medicare beneficiaries, regardless of
whether the sponsor of the trial is paying, or
agrees to pay, for the study-related costs of
other study participants whose costs are not
covered by insurance.

One of the principal sources of confusion
regarding the NCD is the issue of when Medicare
will refuse to pay for clinical trial costs
because a sponsor has agreed to protect study
participants from having to pay for such costs
out of their own pockets. This confusion has two
sources: the NCD itself and the Medicare
Secondary Payer (MSP) principle.

In developing the NCD, CMS agreed to cover
routine costs associated with clinical trials,
which are not covered by many other insurers.
However, the NCD states that Medicare will not
pay for clinical trial costs if they are
for “items or services customarily provided by
the research sponsors free of charge for any
enrollee in the trial.” This creates a dilemma
for research sponsors. If a research sponsor
agrees to cover clinical trial costs for patients
not covered by Medicare, there is a risk that
Medicare would refuse to cover those costs for
Medicare beneficiaries. This puts the research
sponsor in the position of choosing whether to
cover all research-related costs for all trial
participants, or forcing study participants who
are not Medicare beneficiaries to bear the costs
of participating in the study.

The MSP rules add to the confusion. Under the
MSP rule, Medicare may not pay for any item or
service that has been made is reasonably expected
to be made under a “primary plan.” A “primary
plan” is defined to include (1) a worker’s
compensation law or plan, (2) no-fault insurance,
(3) an automobile or liability insurance policy
or plan (including a self-insured plan), and (4)
a group health plan or large group health plan.
CMS has interpreted this rule to prohibit
coverage for study-related injuries if a sponsor
has agreed to pay for such costs. This
interpretation rests on a research sponsor being
considered a “primary plan”, a view that has no
statutory support. Further, this interpretation
forces a sponsor to choose between bearing all
costs for trial-related injuries (an outcome
inconsistent with the purpose of the NCD) or
bearing no such costs (forcing non-Medicare
beneficiaries to pay for trial-related costs out
of their own pocket).

The existing NCD creates confusion among sponsors
and research sites. It is important that CMS
address these issues and revise the NCD to
clarify that Medicare will cover trial-related
costs regardless of whether the sponsor is paying
for costs incurred by non-Medicare
beneficiaries. Any other interpretation and
application of the NCD and MSP rules would be
inconsistent with the underlying policy goals of
the NCD.

5. Remove the self-certification process
that was never implemented.

Comments: CMS should continue to consider all
IND-exempt trials as automatically qualified to
receive Medicare coverage under the NCD if it
removes the self-certification process.

Many important studies exploring new uses for
marketed products are conducted by investigators
who are not required to conduct such trials under
an IND. This is particularly true in the area of
oncology. The U.S. Food and Drug Administration
(“FDA”) launched its Reinventing the Regulations
of Cancer Drugs initiative in 1996 with the goal
of accelerating the approval of cancer therapies
and encouraging new uses of marketed products in
cancer treatment. As part of that initiative,
FDA explained that many sponsors were submitting
INDs for trials of marketed products in off-label
indications when no such INDs were required, and
stated that the agency would no longer accept
INDs that were exempt under the applicable
regulations. In 2004, FDA issued a guidance
document to clarify the circumstances when no IND
is required for studies of marketed products in
cancer research. FDA noted in that guidance
document that oncologists commonly modify labeled
dosing recommendations as part of their clinical
practice. FDA further stated that off-label use
of cancer therapies in the context of clinical
trials does not appear to significantly increase
risk to patients, and that an investigator could
conclude that such a study could be conducted
without an IND.

In oncology, IND-exempt trials are part of an
established federal regulatory mechanism designed
to expedite the approval of cancer therapies and
encourage new uses of marketed products in cancer
treatment. However, even outside the oncology
setting, IND-exempt trials play an important role
in exploring new uses for marketed products. The
current NCD covers the routine costs of a
clinical trial conducted under an IND, even when
the study is purely investigational and involves
drugs or biologicals that have not been approved
by the FDA. It is therefore incongruous that CMS
would not extend the same degree of coverage to
IND-exempt trials which involve lawfully marketed
drug or biological products.

Failure to automatically grant IND-exempt
trials “deemed” status would undermine the goals
of the Medicare clinical trials NCD. IND-exempt
trials represent an important category of
clinical trials. In the area of oncology,
thousands of cancer patients are enrolled in
studies that are not required to be conducted
under an IND. If deemed status is not extended
to IND-exempt investigations, Medicare
beneficiaries will be discouraged from
participating in the many clinical trials
exploring investigational cancer treatment
options using drugs that are already available.
CMS must ensure that the NCD continues to provide
coverage for clinical studies that are exempt
from the IND regulations.

6. Clarify the scientific and technical
roles of Federal agencies in overseeing IND
Exempt trials.

Comments: CMS has no statutory authority to
clarify the roles of other Federal agencies in
overseeing clinical trials. In any event, no
such clarification is required, because the
conduct of those trials is regulated by FDA and
subject to oversight by Institutional Review
Boards (IRBs).

A clinical study that is exempt from the
requirements of an IND is not exempt from
regulatory oversight. In addition to meeting the
specific regulatory requirements to be considered
exempt from the IND regulations, such studies
must be conducted in compliance with IRB and
informed consent regulations (set forth at 21
C.F.R. Parts 50 and 56). By definition, IND-
exempt trials are generally not intended to
support FDA approval of a new indication or a
significant change in product labeling.
Nonetheless, these IND-exempt trials commonly
meet rigorous standards for trial design, data
collection and reporting of results in order to
pass IRB scrutiny, attract research funding, and
qualify for presentation and publication.

In revising the NCD, CMS should focus on
providing clarification and eliminating the
obstacles to greater participation in clinical
trials by Medicare beneficiaries, and not assume
the obligations of other agencies in the
regulation of clinical trials.

7. Determine if coverage of routine clinical
care costs is warranted for studies beyond those
covered by the current policy.

Comments: Genentech requests that CMS clarify
this issue. Genentech offers no comments on this
issue at this time because it is unclear what CMS
means by “studies beyond those covered by the
current policy.”

8. Clarify how items/services that do not
meet the requirements of 1862(a)(1)(A) but are of
potential benefit can be covered in clinical
research studies as an outcome of the National
Coverage Determination process.

Comments: The above issue relates to the
Coverage with Evidence Development (CED) Guidance
that was issued by CMS on July 12, 2006.
Genentech will be separately submitting comments
on the CED Guidance and we request that CMS take
our CED comments into consideration as they
relate to the NCD.

9. Clarify whether and under what conditions
an item/service non-covered nationally may be
covered in the context of clinical research to
elucidate the impact of the item or service on
health outcomes in Medicare beneficiaries.

Comments: The above issue relates to the
Coverage with Evidence Development (CED) Guidance
that was issued by CMS on July 12, 2006.
Genentech will be separately submitting comments
on the CED Guidance and we request that CMS take
our CED comments into consideration as they
relate to the NCD.

10. Discuss Medicare policy for payment of
humanitarian use device (HUD) costs.

Comments: Genentech offers no comments on this
issue at this time.


II. Coverage of Phase I and Phase II Studies

In addition to the specific issues CMS identified
in the tracking sheet, we think it is critical
that CMS address in its reconsideration of the
NCD coverage for Phase I and Phase II studies.

One aspect of the NCD that has caused confusion
and led to inconsistent coverage determinations
by local Carrier Medical Directors is the
requirement that clinical trials have therapeutic
intent. The NCD states that “any clinical trial
receiving Medicare coverage of routine costs
must . . . not be designed exclusively to test
toxicity or disease pathology. It must have
therapeutic intent.” It further states that, to
qualify for coverage, the “principal purpose” of
the trial must be “to test whether the
intervention potentially improves the
participants’ health outcomes.” There have been
conflicting interpretations regarding the meaning
of these provisions. Coverage for Phase I
studies is frequently denied and, under some
restrictive interpretations of the NCD, coverage
would be limited mainly to Phase III studies.
These narrower restrictions limit access of
Medicare beneficiaries to promising new
investigational drugs in early stages of their
development.

In revising the NCD, CMS should clarify that all
Phase II studies as well as all Phase I studies,
except for those conducted in healthy patients,
are eligible for coverage under the NCD.
Although Phase I trials are frequently conducted
in healthy volunteers to test dose ranges and
pharmacokinetics, clinical trials in oncology are
typically conducted in sick patients where
evidence of response is frequently collected in
addition to safety and pharmacokinetic data.
Patients often enroll in Phase I trials when
there is no standard of care treatment to combat
their disease or the standard treatments have
failed. Phase II studies are studies conducted
to make preliminary determinations of efficacy
and to collect additional data regarding safety.
These trials can represent the last hope of
therapeutic benefit for ill patients without
other treatment options, and can help determine
whether the development of a drug candidate
should go forward. According to the American
Society of Clinical Oncology, in recommending
participation in a clinical trial “[i]t should
not be relevant whether the scientific goals of
the study are to determine toxicity and
pharmacokinetics (phase I) or to ascertain the
response rate (phase II) because the patient is
receiving an appropriate treatment for his or her
disease . . . . [in a phase I trial] there is a
strong preclinical rationale for bringing the
drug into the clinic with the expectation of
positive clinical outcomes for some
patients.”

Denying coverage for Phase I and Phase II
clinical trials would be contrary to the intent
and spirit of the NCD. Given their importance to
patients and to the drug development process,
Medicare should clarify that costs for all Phase
II trials and all Phase I trials, except for
those conducted in healthy patients, are covered
under the NCD.


III. Conclusion

Genentech appreciates the opportunity to comment
on the reconsideration of the NCD. We urge CMS
to clarify those aspects of the NCD that have
prevented its more widespread application to the
many clinical trials that are exploring new
therapies for serious conditions affecting
Medicare beneficiaries.

Please contact either Heidi Wagner at (202) 296-
7272 or Michael Listgarten at (650) 467-6827
should you have any questions.

Respectfully submitted,


Walter Moore
Vice President, Government Affairs