1. INNOVATION AND CHOICE
REPORT ON THE PROMISE OF AND IMPEDIMENTS TO
DRUG COCKTAILS
Executive Summary
Innovation and Choice
(202) 556-0614

2. EXECUTIVE
SUMMARY
Executive Summary • Drug Cocktails
WE PRESENT THE NARRATIVE of this report and the recommendations which flow
from it to the United States Congress, the President of the United States, applicable
agencies, and the American public for their consideration. This report is the product of
numerous discussions with statisticians, economists, researchers, physicians, patients,
regulatory officials and the general public.
In our country’s effort to improve the quality of healthcare, most especially in
defeating dynamic and debilitating conditions like cancer, ALS, and Alzheimer’s, it is
imperative that we recognize the benefits of combination therapies (i.e., drug cocktails)
and create statutory incentives to stimulate the development of such combinational
products.
These cocktails are likely to result in lower side effects and greater effectiveness
when compared to conventional therapies, as they can attack multiple different sites of
the same cellular-signaling cascade and multiple different molecular problems unique to
diseased/infected cells, at one time, leading to lower dosages needed for each molecular
mechanism to get the same or greater of an effect on the aggregate (survival or disease
response). With this in mind, it is paramount to medical progress that we stimulate the
amount of drug cocktails investigated.
BACKGROUND
What are Combination Products
A combinational product, as defined by 21 CFR 3.2(e), refers to the combination
of two or more separate components—be it drugs, devices, or biological products (a
substance derived from a living organism)—for the purposes of obtaining a primary
endpoint. The specific definition for a combination product is as follows:
(1) A product comprised of two or more regulated components that are
physically, chemically, or otherwise combined or mixed to produce a
single entity;
(2) Two or more separate products packaged together in a single package or
as a unit and comprised of drug and device products, device and biological
products, or biological and drug products, and
(3) A drug, device, or biological product packaged separately that according
to its investigational plan or proposed labeling is intended for use only
with an approved individually specified drug, device, or biological product
where both are required to achieve the intended use, indication, or effect
and where upon approval of the proposed product the labeling of the
approved product would need to be changed, e.g., to reflect a change in
intended use, dosage form, strength, route of administration, or significant
change in dose; or
2

3. (4) Any investigational drug, device, or biological product packaged
separately that according to its proposed labeling is for use only with
another individually specified investigational drug, device, or biological
product where both are required to achieve the intended use, indication, or
effect.
Once a combination product is developed, sponsors submit a Request for
Designation where the Food and Drug Administration (“FDA”) then decides which
department has oversight over that therapy. This decision of department designation is
based off the primary mode of action (PMOA) of the therapy, which is determined by an
algorithm if one is not able to determine, arbitrarily, which component is the primary
agent of action of the therapy.
BENEFITS FROM COMBINATIONAL PRODUCTS
Combination therapies (i.e., drug cocktails):
1. Reduce the probability a condition, infection, or disease develops resistance to the
therapy administered, as multiple active modes of action target a particular disease or
infection at the same time. Thus, even if a disease/infection becomes resistant to one
mechanism of action, the other agent still combats the disease or infection.
2. Often result in more effectively combating an ailment when compared with
administering components individually with the same dose. With many diseases and
conditions the result of multiple molecular dysfunctions and infections having
multiple routes to attack, combination therapies often result in an increased
effectiveness against such ailments. This is because combination therapies often
attack a single disease at multiple molecular points at a time.
Secondly, there is the concept of drug synergy. Drug synergy refers to when
the benefit of a combination therapy is greater than the sum of the effects of each
therapy when administered individually. This occurs when the components which
make up a drug cocktail are partially dependent on each other’s action.
3. Often decrease the amount of side effects incurred, when compared with each therapy
given individually at their maximum tolerable dose (“MTD”), by administering the
combination of multiple independent components at lower doses than would
otherwise be administered. However, there is the risk components will not be
independent and will result in greater than expected side effects. This risk often
deters therapy sponsors from investigating drug cocktails, as one severe side effect
from the combination of two or more components can result in the discontinuation of
an experimental trial.
WHEN AND HOW COMBINATION PRODUCTS ARE DEVELOPED IS
IMPORTANT
1. Combining already approved therapies through off-label prescription /
physician created cocktails
Currently, the FDA provides no oversight over off-label prescription. Once a
therapy is approved for use, it can be prescribed for any indication and at any dosage.
3

4. This freedom allows physicians to concoct their own cocktails of therapies, which is
ever-present in the field of cancer. It is estimated that 60% of oncologists prescribe
therapies on an off-label basis. The reason for this is because cancer is a disease of
multiple molecular dysfunctions at the root of cell-growth. Recognizing multiple means
of combating this disease, physicians often prescribe multiple drugs for cancer patients at
a time.
The downside to this off-label prescription is these cocktail therapies were not
tested in concert for safety and efficacy. This means the dosage and cocktail prescribed
may easily result in deleterious side effects. As well, physicians are not trained to make
their own dosage recommendations, thus discovering the right dose for each individual
component and patient is a challenge in itself.
2. Development of a drug cocktail by combining already approved therapies
with investigational or other already approved treatments
Many sponsors know certain combinations may be best to treat a certain illness,
yet they will still investigate components separately first to gain marketing exclusivity for
each individual component. This leads to a prolonged time for society to realize the full
benefits of combination therapies, which in the case of seriously compromising illnesses
means countless lives lost.
Examples follow:
Atripla refers to a drug cocktail which combined and tested three already
approved therapies for HIV. This therapy combination of antiretroviral components
severely impedes the progression of HIV by attacking it at multiple points in its life-
cycle.
Many chemotherapy combinations for cancer are also developed by performing
clinical trials with an already approved chemotherapeutic agent along with an
investigational new agent.
3. Many components of cocktails are investigated individually first by therapy
sponsors, despite obvious preclinical data and benefits supporting greater
effectiveness and specificity when such components are administered in
concert.
There are multiple components in combination therapies that must each be
isolated and proven to contribute to the overall benefit of the therapy. Safety is an
additional concern because, as these therapies are more potent, they are also more likely
to have adverse side effects. This is why the dosage of cocktails is often 10% less than
the dosage would be if each therapy was approved for use individually. The key benefit
of cocktails is the ability to specifically target a disease, infection, or condition, with
many small targets each hitting that one disease/infection site. In turn, this realizes a
grand, gestalt effect on defeating many diseases and infections.
However, with this specificity also comes with the less desired goal of a smaller
sales market for pharmaceuticals and therapy sponsors to extract profits; thereby, creating
a disincentive to produce such a cocktail unless the sponsor knows it will still reach a big
enough population. This does not appear to be the case in most instances, as sponsors
continually show they would rather test a therapy separately, have it be approved, and
4

5. then work a cocktail, than test the cocktail originally. This is evident in the case of
FUS1/p53 dual gene therapy and MD Anderson’s choice to test each therapy individually
rather than in concert despite obvious synergistic effects in mice.
It has been well documented that a therapy sponsor’s goal is to extract the most
profits possible, while providing small changes to already approved therapies to gain
additional patent protections. Over three-quarters of all drugs approved by the FDA are
new combinations or formulations of therapies rather than new chemical entities. This
means that after a therapy is developed, sponsors try to add to it rather than discovering
novel treatments.
In an effort to secure market exclusivity, sponsors will not develop cocktails until
they have the patents on each component or such components are already approved. The
reason is because the investigation of two unproven and untested components is viewed
as too risky with regards to unforeseen severe side effects, and sponsors would rather
secure each individual patent separately before investigating a cocktail to realize broader
market bases for each.
If a therapy is already developed and approved, a sponsor already has the
marketing exclusivity for that market; thus, additional cocktails are tested after these
pharmaceuticals have a lock on this market. The reason a sponsor would rather have this
than a combination therapy approved is because a single active agent is potentially
beneficial for multiple ailments rather than just a few. The more specific the therapy (the
more active components), the less is a sponsor’s market base.
4. Cases where all components of a cocktail are investigational
No examples of drug cocktails were found in which all components were new
chemical entities. The reason for this is likely because sponsors would rather gain
marketing exclusivity for each individual component to realize a greater market-base,
before undertaking the more risky cocktail, which has a more slim market.
EXAMPLES OF THE BENEFITS OF COMBINATION PRODUCTS
HIV
Drug cocktails are most commonly known in the field of HIV, with many patients
afflicted taking a treatment cocktail of three to four antiretroviral therapies. This unique
combination of active components, working in concert, has the effect of drastically
reducing HIV life-cycle by impinging on different points in this life-cycle. These
cocktails can be taken as four separate pills which must be monitored distinctly, or
together in the form of one pill and one dosage. As the human body does not have
beneficial viruses, these have very few side effects.
Alzheimer’s
Alzheimer’s is thought to be caused by the accumulation of amyloid-beta plaque
in brain cells. However, studies show this toxic protein is reduced when either of
enzymes beta-secretase or gamma-secretase are inhibited. It has also been discovered
that overly inhibiting these enzymes produce dangerous side effects – inhibiting too much
gamma-secretase leads to developmental defects, skin tumors, and a shortened life span;
5

6. while inhibiting too much beta-secretase impairs nerve function and causes
schizophrenia-like symptoms in mice.
In a recent study, published in Science Translational Medicine, aged Alzheimer’s-
like mice were given a cocktail of inhibitors that moderately inhibited both beta-secretase
and gamma-secretase. This reduced the production of amyloid-beta plaque, without
adverse side effects.
Cancer
Effectiveness
It is said the reason there is no cure for cancer is because cancer is not one
disease, but multiple diseases. What physicians and researchers mean by this is the fact
that cancer is simply the uncontrolled overgrowth of any cell in your body, but the cause
of this uncontrolled overgrowth is rooted in multiple different molecular dysfunctions in
the same cell. Moreover, when a cancer cell divides, those newly formed cells have
different mutations than the previous. This creates an environment where there are
multiple different types of cancer in the same very patient. To effectively combat this
ailment, it is hypothesized by many physicians and researchers that drug cocktails and
vaccines are adept to this challenge via their special ability to target specific molecules
within a cell.
Targeting potent therapies to avoid toxic side effects
Combining plasmodium (bacterial) proteins specific for liver cells with
chemotherapeutic agent doxorubicin, researchers at the University of California, Irvine
are able to specifically target liver cells with this combination therapy, effectively
minimizing side effects felt by the heart and other organs that were previously affected.
Therapies that work on their molecular target, and are safe, but not approved by
the FDA
Often times when individual therapies are not beneficial in combating illnesses,
infections, and diseases, combination products will realize desired effects. Hence, there
is the importance for therapies which effectively accomplish their molecular object (i.e.,
inhibit molecule A), and are safe, to be approved for licensing for development as
potential combination therapies.
Many therapies do not meet the FDA’s statistical cutoff for efficacy on primary
endpoints, but still provide valuable benefits to patients with little to no side effects. The
therapies I speak of are those that provide a three month benefit to cancer patients instead
of four. While it is not clear if these therapies provide benefit to patients better than
chance would (which is the purpose of statistical testing), many of them do perform their
function (inhibit telomerase or a specific kinase) at a statistically significant level. This
means that they do what they’re supposed to at a significant level, but the level of benefit
to patients’ survival rate or disease progression isn’t as grand as what is necessary to get a
therapy approved. It is proposed that these therapies should be allowed for marketing or
licensing under a separate approval type. The reason for this is because these therapies
are ideal for adjuvant therapies or cocktails (therapies that do not yield adverse effects,
but still perform their molecular purpose to a statistically significant level).
6

7. RECOMMENDATIONS
A. Break the FDA Approval Codes into three different approval types to ensure
hundreds of millions of dollars are not wasted developing therapies that are
potentially beneficial for combination therapies (but which do NOT produce
statistically significant effectiveness alone):
i. Approval type A-1: Marketing approval to the public of a therapy, with the
indication of a particular molecular or genetic characteristic and disease type, is
granted if a therapy yields statistically significant results for that disease (in
primary endpoints like survival rate or disease progression), shows the expected
molecular response in patients, and shows safety.
A brief example of how this would work is: one would take a stratified
random sample of people with a given disease, stratify them into genetic /
molecular characteristics; stratify those groups into disease stage;
randomly allocate these groups into treatment and control groups, and test
for results.
ii. Approval type A-2 (similar to Accelerated Approval based on surrogate
endpoints): Marketing approval of a therapy to licensed physicians and physician
groups, with the indication of a molecular or genetic characteristic and a
particular disease, is granted if a therapy yields a statistically significant
molecular response (also known as, “biological effect”) (i.e., Tarceva shows
inhibition of a cell’s EGFR tyrosine kinase) and shows safety, but does not yield
statistically significant results with regards to primary endpoints like disease
progression / survival rate. A pharmaceutical does not apply for this but can
receive approval under type A-2 if it meets what is described above after applying
for approval type A-1 or through approval type B (described below). *The
therapy does not have to result in any disease response in order to obtain
“physician marketing approval.”*
SUMMARY: Approval type A2 is granted for a particular molecular or
genetic characteristic and disease type, as the therapy yields positive
results with respect to its mechanism of action working, but not with
regards to progression / disease response. These therapies are ideal
candidates for drug cocktails and adjuvant therapies as they do not work
by themselves but may work well if you combine them with other
therapies. A rigorous demand for safety is maintained of course.
iii. Approval type B (the current method of approval) without stratification of
molecular or genetic characteristics. This can be a supplemental application, or
granted without the need for further statistical testing, if a sponsor’s study for
Approval Type A1 produces certain results. This approval type is for indications
with patients with unknown molecular / genetic characteristics, with its purpose
being if the patient needs a therapy but cannot afford to pay for specific testing or
it is not covered by their insurance (or if a sponsor cannot afford to test via
stratification of molecular/genetic traits). This approval type allows patients that
do not know if they have a specific molecular trait to take a therapy and insurers
7

8. should be mandated to cover at least a small portion of it (if it is granted under
Type B).
SUMMARY: This application allows sponsors to market their therapy
and patients to obtain a therapy, with unknown molecular traits but a
known disease type. This requires sponsors to run a study and perform
randomized statistical tests to measure the effectiveness of a therapy in
people with a particular disease type (that may or may not exhibit a
molecular or genetic characteristic), or pool data from a previous study
(i.e., those performed in A1 trials).
B. Offer unique incentives for the investigation of combinational therapies whose
components are not previously approved. (The notion behind this is to mitigate
the costs pharmaceuticals face with less of a market base with such therapies.)
C. Create a symposium where the FDA, pharmaceuticals, and researchers all come
together to brainstorm potential drug combinations that can be made into a single
therapy. Some mechanism to facilitate this interaction is gravely needed. We
currently have the American Association for Cancer Research, the American
Society of Clinical Oncology, and other putting on symposiums. However, we
need the focus of this symposium to be brainstorming the creation of drug
cocktails that can more effectively combat illnesses rather than sharing
information and giving presentations on the latest research or clinical results.
CONCLUSION
By combining small modes of action to target an ailment at the same time, it has
been proven that side effects can be minimized and effectiveness maximized. As seen,
there are many ways to develop such combinational therapies – physicians can develop a
cocktail from already approved therapies, sponsors can investigate novel cocktails from
already approved therapies, sponsors can investigate novel cocktails by combining
investigational therapies with already approved therapies, or sponsors can investigate
combination therapies consistent of completely new active components. We have also
seen that it is crucial to the benefit of society at large, if a firm knows a cocktail to be
more beneficial than treatment by individual components, that firms investigate that
cocktail first and foremost. Society cannot have companies developing therapies
individually despite knowing there are better benefits to be developed, just so that
company can have patent protections for a larger market.
Given this, and the notion individual therapies in investigation which do not yield
statistically significant marks with regards to primary endpoints like survival and disease
regression, but are safe and effective in their target action, it is imperative we rewrite the
approval codes to be consistent with such knowledge. And, while we are naturally averse
to change, it is for the benefits of society at large that we must create incentives and
regulatory changes to realize the benefits of combination products.
We look forward to a comprehensive discussion on the merits of what we have
recommended, and we will participate vigorously in this discussion.
8