GS: AMGN(OP/N): Meetings with CEO confirmed
long-term positive view
52-Week Range US$72-54
YTD Price Change -11.47%
Market Cap US$73.3bn
EPS Growth Estimate 23%
Fiscal Year (ending in Dec)
2003 2004E 2005E
US$1.90 US$2.39 US$2.84
— 22.9X 19.3X
— 1.1X 1.0X
We met w/ AMGN's's CEO who was positive about the state of AMGN's business. He
discussed the following: (1) decision to grow the business organically (2) analysis of
Medicare reimbursement in 2005 (3) U.S. entry of generics or CERA (Roche) unlikely &
(4) growing pipeline with start of Phase III trial of AMG162 in osteoporosis in 2004.
AMGN shares have underperformed partly due to concerns about reimbursement cuts &
competition. While the exact impact is difficult to predict, we believe the potential
disruptions should be manageable. Generic threat is unlikely until next decade in U.S. &
modest in Europe. We maintain our Outperform rating based on solid growth, an
increasingly visible pipeline and attractive valuation. AMGN's PE is 23x our 2004 EPS
and PEG 1.0 vs 40x and 1.9 PEG for biotech, 19x and 1.8 PEG for drugs, & 26x and 1.9
PEG for LLY. Risks include slower sales, reimbursement, patent disputes, and
development failures. Our coverage view remains Neutral.
We traveled with Amgen's CEO, Mr. Kevin Sharer, in Europe last Friday. Mr. Sharer was
positive about the operations of Amgen and the growing pipeline. Amgen shares have
been under pressure due to concerns such as reimbursement cuts, competitive threats
from generics and Roche's CERA, pipeline gap in the near term and the potential for a major
dilutive acquisition. Mr. Sharer's discussions covered these issues.
While the reimbursement changes to be implemented by Medicare in 2005 are complex and the
effect on usage, especially temporary disruptions, is difficult to predict, our scenario analysis
suggests that the impact on earnings, if any, should be manageable.
We continue to view the entry of generics to Amgen's products to be unlikely in this decade in the
U.S. and the impact on profits modest in Europe. Roche's CERA will probably not reach the
market in Europe until 2007/08 and should face major patent hurdles from Amgen in the United
States. The efficacy of CERA still needs to be demonstrated in Phase III trials which just began.
The long half life of CERA may make titration by physicians difficult and lead to more clotting
problems.
We do not expect any major product launch from Amgen in 2005/06. However, we estimate that
Amgen's product sales should grow 21% in 2004 and 14% in 2005, leading to EPS growth of 26%
and 19%, respectively. The higher growth rate for EPS versus sales is due to economies of scale,
improved efficiencies and lower tax rate.
To estimate the sales from new product/indications required to sustain 10%- 20% sales growth or
13%-23% EPS growth in 2006-2010, we have performed a scenario analysis using the 2005 sales
as baseline. Total sales should approximate $10.9 billion in 2005. Assuming the CAGR in sales for
products launched before 2004 will be 7% in 2006-2010, the implied sales from new
products/indication required to provide total sales growth of 10%, 15%, and 20% would be $2.3
billion, $6.7 billion, and $11.9 billion, respectively. We believe the hurdle to achieve 10%-15%
sales growth or 13-18% EPS growth is not unusually challenging considering the current pipeline
which includes:
* Palifermin to be filed with the FDA for mucositis this year, * ABX-EGF in Phase III trials for
colorectal cancer, * AMG162 to enter Phase III trials for osteoporosis in 2004, * AMG714,
AMG531 and GDNF to complete Phase II trials for rheumatoid arthritis, idiopathic
thrombocytopenic purpura (low platelets), and Parkinson's disease respectively, in 2004 * Ongoing
Phase IV trials on new indications for Aranesp, Neupogen/Neulasta, and Sensipar.
Amgen shares are trading at 23x our 2004 EPS or PEG 1.0 versus biotech average of 40x and 1.9
PEG, drug average of 19x and 1.8 PEG and Eli Lilly's PE of 26x and 1.9 PEG. On 2005 EPS,
Amgen's PE is at 19x and PEG of 0.8 versus biotech group average of 31x and 1.5 PEG, drug
average PE of 17x and PEG of 1.6, and Eli Lilly's PE of 22x and PEG of 1.6.
1. FOCUS ON ORGANIC GROWTH
Mr. Sharer discussed his strategy to grow Amgen by focusing on current products and developing
innovative drugs to treat grave illnesses. He does not favor diversification away from human
therapeutics nor major acquisitions. However, the company will continue to in-license products
and consider small acquisitions.
2. AWP REFORM LIKELY MANAGEABLE BUT TEMPORARY DISRUPTIONS CANNOT
BE RULED OUT
The Medicare Modernization Act (MMA) passed in late 2003 cut the Medicare reimbursement to
physicians on drugs administered in doctors' offices. Prior to 2004, reimbursement to physicians on
drugs administered in physician offices, including Aranesp, Procrit, Neupogen, and Neulasta, was
based on 95% of the average wholesaler price (AWP). The Congress and the Center of Medicare
and Medicaid Services (CMS) viewed the reimbursement to be excessive because the average
selling price (ASP) at which certain drugs are sold to physicians can be below 95% of AWP. In
2004, drugs are reimbursed at AWP minus 15% (85% AWP) with certain exceptions (Rituxan is
reimbursed at 81% AWP). In 2005, reimbursement will be at ASP plus 6%. In 2006, physicians
have a choice of being reimbursed under ASP or obtain drugs from regional contractors which will
contract with CMS under competitive bidding.
CMS defines ASP as the manufacturer's sales to all purchasers in the United States (excluding units
sold to government entities or at nominal cost) for each National Drug Code (NDC) for a quarter
divided by the total number of applicable units of that NDC sold by the manufacturer in that
quarter. For the purposes of the ASP calculation, the 'unit' is the product represented by the NDC
of the Act. Sales to government entities will be excluded from the ASP calculation but most rebates
and discounts will be included in ASP.
ASP data will be due from manufacturers on the 30th of every month following the end of the
calendar quarter. Data from the second or third quarter of 2004 will be used to determine payment
starting January 1, 2005. Reimbursement will be updated on a quarterly basis thereafter.
Under the ASP system, drugs with the larger discount will be more vulnerable to shift in user mix.
For example, if a product with a list price of $100 is sold to large volume users at 30% discount
and to low volume users at no discount, then the ASP would be between $70-100. With 2005
reimbursement at ASP + 6%, the large volume users can generate a profit even though the level of
profit will vary depending on the purchase price paid by the users. However, a small volume user
receiving little or no discount may incur a loss, leading to fewer prescriptions or referral of the
patients to other clinics or hospitals.
In the case of Aranesp and Procrit, the prices paid by oncology clinics are generally the lowest,
followed by hospitals and then by retail pharmacies and long-term care facilities. Currently,
Aranesp is the market leader in oncology clinics whereas Procrit dominates in the other channels.
Therefore, the ASP for Aranesp, as calculated based on the weighted average of all channels, will
likely be lower than for Procrit, leading to a lower level of reimbursement when the ASP system
becomes effective in January 2005. We expect Amgen to aggressively gain share in channels
outside of oncology clinics to raise the effective ASP which will be adjusted quarterly. Amgen
might also use other tactics to gain market share in oncology clinics without increasing discounts.
In the next few years, we continue to expect the anemia market in cancer to grow at 5-10%. The
market share of Aranesp should increase to above 50% from the current mid 30% based on more
convenient dosing. However, temporary disruption in usage, especially in late 2004/05 as
physicians evaluate the impact of ASP, cannot be ruled out. We have performed a scenario
analysis by assuming 5%-15% of cancer drugs may be "over-used" by some physicians who are
driven by financial incentives. Therefore, reduction in reimbursement may lead to a comparable
"shrinkage" in the market. We estimate that the reduction could reduce 2005 EPS by 1-3%, which
should be manageable.
Geographic expansion should also drive the profit contribution from Aranesp. Amgen's joint
venture with Kirin, Kirin Amgen, should file a marketing application on Aranesp in Japan by
yearend. Approval should be granted in 2006. In the long term, new indications, such as congestive
heart failure (Phase II ongoing) might also boost sales of Aranesp.
For Neupogen/Neulasta, competition in the U.S. is limited to Schering AG's Leukine which has
about 10% market share despite a lower price. The shift to ASP should have a neutral to slightly
positive impact on Neupogen/Neulasta. Japanese approval of Neulasta might be granted in 2007.
3. CHANGE IN EPOGEN REIMBURSEMENT SHOULD HAVE MODEST IMPACT
For dialysis providers, the Medicare reimbursement of Epogen will be based on ASP in 2005 instead of 95% AWP. The foregone profit from the spread between the
cost and the previous reimbursement level of Epogen will be bundled into the composite rate for
dialysis services. The change is designed to remove direct financial incentive from administering
Epogen without significantly affecting the overall profit of dialysis providers. Based on the report
by the Inspector General in May 2004, the average discount of Epogen is about 12% to large
volume users and 5% to small volume users. CMS will use this information to calculate the
difference, or spread, between the Medicare reimbursement for Epogen and the average cost to
facilities to acquire those products. This information will be used in an upcoming regulation that
will create the new case-mix adjusted payment system for 2005. The ESRD payment regulation is
due to be published in proposed form between May and July of this year, but no later than August
1. Due to the relatively low level of discount and our belief that most physicians would still target
hematocrits close to the current level of around 36%, the long-term impact on usage should be
modest. Our scenario analysis indicates that if the reimbursement change will lead to 5% decline
in Epogen sales in 2005, Amgen's EPS would be lowered by 1%-2%.
4. GENERIC EPOGEN AND CERA UNLIKELY BEFORE 2006/07 IN EUROPE
The patent on Epogen will expire in Europe in late 2004 and the guidelines on approving generic
biologics should be issued in the European Union in 2004. However, the requirement for clinical
trials should be extensive. Therefore, generic Epogen will unlikely reach the market in Europe
before 2006/07. In the U.S., the patents on Epogen will not expire until 2013. Even though the
FDA is expected to issue guidelines on generic biologics in 2004, Congressional approval will be
required before the guidelines can be formally adopted. We do not expect Congressional approval
before 2005.
Roche's CERA, a long acting EPO, might be launched in Europe in 2007/08, assuming it is safe
and efficacious. Phase III trials on anemia in dialysis just started in H1/04 and are expected to
begin for chemotherapy related anemia later this year. In the U.S., we believe the patent hurdle for
CERA is high unless Amgen's patent protection is weakened by the legal disputes against
Transkaryotic Therapies (TKT). We expect a decision on the TKT trial in 2004. Regardless of the
verdict, we expect further appeals. However, the long half life of CERA may make by physicians
titration difficult and could lead to more side effects, such as clotting.
Aranesp sales in Europe represent about 10% of Amgen's total sales and probably less than 10% of
profit. Therefore, EPS impact from generic EPP and CERA should be modest.
5. PHASE III TRIAL ON AMG 162 ON OSTEOPOROSIS WILL START IN 2004
AMG 162 is a humanized antibody to RANK ligand which plays a critical role in bone resorption.
In Phase I/II trials, one subcutaneous injection of high dose AMG 162 seemed to be effective in
reducing bone turnover and increasing the spine and hip bone mineral density in postmenopausal
women with osteoporosis for 6 months. Data at 12 months should be available in H2/04. The
efficacy of AMG162 appeared to be better than that of the oral biphosphonate, Fosamax (Merck).
Unlike Fosamax, AMG 162 might be synergistic with parathyroid hormone (PTH), including
Forteo (Eli Lilly) and Preos (NPS). We believe the sales potential could exceed $1 billion. Mr.
Sharer indicated that Amgen just finalized the plans for large Phase III trials involving several
thousand women with osteoporosis.
6. PALIFERMIN
Amgen is on track to file an FDA application on Palifermin for mucositis in blood cancer patients
receiving bone marrow transplantation (BMT) this year. We estimated initial sales potential to be
$200MM. However, the potential could be increased to $1 billion if Palifermin is effective in
treating mucositis in patients with solid tumors.
I, Maykin Ho, PhD, hereby certify th |
