GS: RNVS (IL/N): Additional data on Cerovive
showed improved functional outcome & are
within expectation.
52-Week Range US$19-7
YTD Price Change 18.92%
Market Cap US$419.4mn
EPS Growth Estimate NA
Fiscal Year (ending in Dec)
2004 2005E 2006E
US$-1.50 US$-1.77 US$-1.75
Additional data on the first PIII study (SAINT-1) of Cerovive on stroke presented on 5/28
were generally within expectation. Cerovive improved functional outcome significantly
but had minimal impact on neurological outcome. The functional improvement seems to
be more modest versus that seen with Activase (the standard therapy) but consistent with
the different mechanisms of action. However, use of Activase with Cerovive did not
increase bleeding, which is a positive, in our view. RNVS shares have appreciated about
150% since the release of the SAINT-1 data and up 19% YTD. We believe there could be
further upside to the shares if the SAINT-2 data is positive in late 2005/early 2006.
However, the shares may come under pressure in the near term because the magnitude of
clinical benefit based on SAINT-1 might be difficult to interpret. We maintain our In-Line
rating and Neutral coverage view
If the 2nd PIII study (SAINT-2) duplicates the results of SAINT-1, the data should be
sufficient to support an FDA application. We continue to believe that the sales potential of
Cerovive exceeds $1B, assuming approval & launch in 2007. Risks are development
failures, cost containment, reliance on corporate partners, limited commercial experience
& need for additional capital.
1. ADDITIONAL DATA ON CEROVIVE GENERALLY WITHIN EXPECATION
At the European Stroke Conference (ESC) on May 28 in Bologna, Italy, the chief investigator of
Cerovive presented further data from the first Phase 3 trial (SAINT-1). The primary endpoint is the
change at 90 days in the Modified Rankin Scale (MRS) which is a measure of functional outcome,
while the secondary/co-primary endpoint is the National Institute of Health Stroke Scale (NIHSS)
which evaluates neurological outcome. The presentation at the ESC was generally within
expectations. As previously announced, Cerovive met the primary endpoint by reducing disability of
ischemic stroke, based on the MRS (p=0.038), but showed minimal benefit on the NIHSS
(p=0.864). Cerovive did not increase bleeding in patients treated with Activase (tPA, Genentech),
which we view as a positive. The improvement on MRS appears modest, with 3.6% more patients
scoring at MRS ?3 and 2.4% more patients scoring at MRS ?1. MRS 3 indicates moderate disability
with the ability to walk without assistance, MRS 1 indicates no significant disability and MRS 0
indicates no symptoms. However, we note that the data should be viewed in the context of the high
failure rate in development of stroke therapeutics and the major unmet medical need for stroke.
About 10% of the patients randomized for treatment in SAINT-1 had less than 75% of the intended
infusion of 71 hours and/or protocol violations. A per protocol analysis excluding these patients
showed slightly better data, with 4.5% and 3.1% of patients scoring at MRS ?3 and ?1, respectively
(p=0.028). Please see Exhibit 1 for details.
In addition to the MRS and NIHSS, analysis was also conducted on a third scale, the Barthel Index.
If responders are defined as those having ?95 on Barthel, then the odds ratio was 1.16 in favor of
Cerovive (p=0.182). The odds ratio for MRS was about 1.2. For NIHSS, the posthoc analysis on
responders (NIHSS ?1) was about 1.1. Subgroup analysis on concomitant use of tPA, time of
therapy from symptom onset, severity of stroke, age, diabetes and glucose did not show any major
drivers of the MRS improvement. The patients in SAINT-1 were relatively well balanced between
the placebo and Cerovive groups. There was a slightly higher percentage of patients with prior
stroke but lower percentage of previous transient ischemic attacks (TIA) in the placebo group. Both
conditions are considered high risk factors for stroke. All patients had their strokes confirmed by
computed tomography (CT) scans and were conscious. The inclusion criteria is NIHSS ?6 including
limb weakness and onset of time to therapy of <6 hours, with forced allocation to achieve a mean of
4 hours.
Our extrapolation of the data on the number of patients with a shift toward improvement in the
MRS scores suggests a response rate of about 6%. In the Phase 3 study of Activase, the standard
therapy of ischemic strokes, conducted in the U.S., there were 5% and 13% more patients with MRS
?3 and ?1, respectively. The more modest impact of a neuroprotectant, such as Cerovive, versus that
of a thrombolytic (tPA), is consistent with the mechanism of action, as the former reduces tissue
damage whereas the latter restores blood flow. The inclusion criteria for the Activase study are
more stringent than that for Cerovive due to safety concerns, especially bleeding. Therefore, the
number of patients who can potentially benefit from Cerovive therapy may be higher than that for
Activase.
2. REGULATORY FILING CONTINGENT UPON RESULTS FROM SAINT-2 AND CHANT
BY EARLY 2006
Renovis management indicated that the FDA considers MRS as the sole primary endpoint and the
NIHSS as a secondary endpoint. Therefore, if the data from SAINT-2, the second Phase 3 study
with identical design as SAINT-1, are positive, there should be sufficient data to support an FDA
application. Results from SAINT-2 are expected in late 2005/early 2006. In Europe, the regulatory
agencies recommended that efficacy be supported by 2 stroke scales based on guidelines issued
about 4 years ago. Therefore, some negotiations will likely be required before European filing.
However, for most applications on drugs with high unmet need, such as Cerovive, the FDA and
other regulatory agencies will likely consider the totality of the data on efficacy, safety and clinical
relevance.
AstraZeneca will probably wait for the results of a Phase II/III trial (CHANT) before filing
marketing applications in the U.S. and Europe in 2006. The CHANT study was designed to evaluate
the safety of Cerovive in about 600 patients with acute hemorrhagic stroke. Enrollment started in
8/04 and should complete in H2/05, leading to results in late 2005/early 2006.
On May 4, 2005, management indicated that an independent data and safety monitoring board
(DSMB) conducted a planned safety review on the first 200 patients. The DSMB recommended that
the study proceed according to plan, which implies no major safety concerns. We believe the
interim analysis is encouraging. If the final data on 600 patients support safety of Cerovive in
CHANT, then CT scans may not have to be performed before Cerovive therapy to rule out
hemorrhagic strokes, thus reducing the time to therapy. Reduction in time to therapy is correlated
with better outcome. Thrombolytics, such as tPA, are used in less than 5% of eligible stroke patients
partly because therapy has to be within 3 hours of symptom onset and hemorrhagic strokes need to
be ruled out by CT scanning.
3. SALES POTENTIAL OF CEROVIVE APPROXIMATES $1.5B
There are about 700,000 stroke victims in the U.S. and 800,000 in Europe per year. Current therapy
is inadequate. Assuming $3,000 per course of therapy and peak penetration of about 30% for
patients presenting within 6 hours, the sales potential of Cerovive could approach $1.5B in the U.S.
and Europe assuming launch in 2007. Approximately 85% of strokes are ischemic strokes, caused
by blockage of the vessels supplying blood to brain tissues. The remaining 15%, known as
hemorrhagic strokes, are caused by rupture of blood vessels to the brain. The price of Cerovive and
penetration will depend largely on the risk benefit profile.
4. VALUATION ANALYSIS
Assuming FDA filing in 2006 and launch in 2007, sales of $1B in 2009, 15% net royalty to
Renovis, and a discount rate of 20%, the implied present value of Cerovive to Renovis at 5X sales
approximates $362MM in 2005 or $15/share. Together with $3 in cash and investor optimism about
indications beyond stroke, the shares have traded above $15 lately. There could be further upside if
SAINT-2 and CHANT are positive. The implied present value using different levels of sales in 2009
and discount rates of 15% and 25% are summarized in Table 1:
Table 1: Sensitivity analysis on implied present value ($/share) of Renovis
Discount rate 15% 20% 25% 2009 $500 $8.8 $7.4 $6.3 Sales ($MM) $1,000 $17.5 $14.8 $12.6
$1,500 $26.3 $22.2 $18.8 $2,000 $35.1 $29.6 $25.1
Source: Company data and GS research estimates
5. POTENTIAL IMPACT OF SAINT-1 ON SAINT-2
AstraZeneca, Renovis' development and marketing partner for Cerovive, started two Phase III trials
of Cerovive in acute ischemic stroke in May 2003. Each involves over 1,700 patients in more than
200 centers. The first study, SAINT-1, was conducted in about 200 centers in Europe, Asia,
Australia and S. Africa, whereas the second study, SAINT-2, is enrolling patients in the Americas.
Patients will receive a one hour loading dose of Cerovive intravenously, followed by 71 hours of
hourly infusion. The target plasma concentration of Cerovive is over 260 micromolar.
The design of the SAINT-2 study is the same as that of SAINT-1. Positive results in SAINT-1
should increase investor confidence that SAINT-2 would show favorable data, assuming the
baseline characteristics and handling of the patients are similar. However, we not
e that medical practice and patients can vary among countries.
The SAINT-1 data may lead to faster enrollment of patients in SAINT-2 so that data from the
second study might be available earlier than the target time (late 2005/early 2006). Management
may also decide to expand the patient number so as to increase the power of the study (a higher
likelihood of meeting the primary endpoint).
I, Maykin Ho, PhD, hereby certify |
