Piper: RNVS :Robust Cerovive Data Support Bullish Outlook
May 31, 2005
(RNVS - $17.10)
Outperform Volatility: Medium
Robust Cerovive Data Support Bullish Outlook
Mark E. Karvosky, Sr. Research Analyst Reason for Report:
212 284-9324, mark.e.karvosky@pjc.com
Swayampakula Ramakanth, Ph.D.,
Research Analyst
212 284-9326,
Swayampakula.X.Ramakanth@pjc.com
KEY POINTS:
* Over the weekend, better-than-expected Cerovive data were presented that
indicate Cerovive has robust and well-distributed efficacy profile and a
clean safety profile. In addition, a posthoc analysis showed that combining
Cerovive with tPA significantly decreased the incidence of Intracerebral
Hemorrhage (ICH).
* Robust Efficacy. Based on the data presented over the weekend we calculate
that between 8.1-12.7% of patients responded to Cerovive (we est. 9-10%
treatment effect). Said another way, between 69-108, out of 850 patients,
responded to Cerovive. Clinicians at the conference indicated that a 3-5%
treatment effect would have been very clinically meaningful. Additionally,
Cerovive's effect was also well-distributed, indicated by data showing that
efficacy was not driven by any particular subgroup, including treatment
time, age, stroke severity or concurrent tPA usage. Secondary endpoints,
NIHSS (p=0.864), Barthel (p=0.182).
* Clean Safety " Improving tPA Safety Profile? As expected, administration of
Cerovive did not result in any increase in adverse events. Additionally a
surprise positive was seen in a posthoc analysis of the data that indicated
Cerovive reduced the number of patients (p=0.005) experiencing an ICH after
treatment with tPA. ICH occurs when a blood vessel in the brain bursts,
causing a patient to transform from an ischemic stroke (blockage) to a
hemorrhagic stroke (bleeding). Conversion to ICH is the major drawback of
tPA, therefore the assertion that Cerovive may reduce the incidence in ICH
is a big deal.
* tPA and Cerovive " Very Different. We caution investors against comparing
Cerovive's effect to tPA's effect in the NINDS trial (basis for FDA
approval) when evaluating Cerovive. First, Cerovive is extremely safe and if
approved would likely be able to be given to all stroke victims, while tPA
has very serious side-effects and can only be administered to patients
within 3 hours of stroke onset (used in 1-2% of patients). Second, due to
tPA's side effects, the tPA trial was designed very differently than the
Cerovive trials. For tPA the FDA was interested in how many patients were
fully recovered (0/1 on mRankin) at 3 months. However, based on Cerovive's
safety profile, the FDA was interested in an analysis of Cerovive's overall
treatment effect across the mRankin scale which we outline below.
* Conclusion. The data presented over the weekend reflect a more robust effect
then we had expected. We believe these data support our notion that as long
as the Saint II data are positive Cerovive is an approvable drug with a
strong safety and efficacy profile that would likely make it the standard of
care in stroke. Additionally early data indicate that Cerovive could improve
tPA's safety profile.
Price: $17.10
52 Week High: $19.25
52 Week Low: $6.34
Price Tgt -- $32.00
Shares Out (mil): 24.5
Market Cap. (mil): $418.9
Avg Daily Vol (000): 988
Book Value/Share: $3.03
Cash Per Share: $3.13
Debt to Total Capital: 0%
Est LT EPS Growth: NM
P/E to LT EPS Growth (FY06): NA
Est Next Rep Date: 07/30/2005
Fiscal Year End: Dec
INVESTMENT RECOMMENDATION:
Our $32 price target is derived by applying a 7x multiple to our 2008 revenue
estimate of $192M, discounted 20%.
RISKS TO ACHIEVEMENT OF TARGET PRICE:
Risks include clinical trial failures or Reg delays (Cerovive and REN1654).
COMPANY DESCRIPTION:
Renovis develops treatments for neurological diseases.
Understanding The Effect. Understanding Cerovive's treatment effect is
somewhat complex because you need to consider all potential improvements
within mRankin, not just the patients that improved to 0 or 1. Additionally,
there is no baseline mRankin score, so in order to estimate the impact that
Cerovive had on patients you must look at the difference between the placebo
and Cerovive arms. Below we walk through the method by which we calculated the
low end of the treatment effect, 8.1%, and the high end of the treatment
effect, 12.7%.
How We Came To A Baseline Effect Of 8.1%. For our first example we will
consider the minimum treatment effect of Cerovive. If we look at the data from
the trial in the table below we see that there was a 3.7% decrease in the
number of patients in the 4 category; additionally we see that there was a
4.4% increase in the number of patients in the 0 category. During the investor
Q/A session AstraZeneca stated that it was safe to assume that based on the
mechanism of Cerovive, that the 3.7% and 4.4% represented different patients.
In other words, it was unlikely that the 3.7% of patients that left the 4
Category had improved all the way to the 0 Category. Based on this assumption
Astra said they believed that at a minimum, 8.1% of patients in the Cerovive
arm experienced a treatment effect of at least 1 point on the mRankin scale.
That said, we must point out that this calculation only considers the patients
with a 4 or 0 and overlooks the likelihood that patients did move within the
1, 2 and 3 categories.
% of Patients In Each mRankin Category
mRankin Score 0 1 2 3 4 5/Death
Placebo 11.0% 20.0% 11.7% 12.7% 20.6% 24.0%
Cerovive 15.4% 18.0% 11.4% 14.2% 16.9% 24.0%
% Dec (Absolute) 4.4% -2.0% -0.3% 1.5% -3.7%
Source: RNVS/AZN Presentation and PJC Company Estimates
How We Came To A Maximum Effect Of 12.7%. Our second example will look at how
we calculated the maximum treatment effect that could have occurred in the
Cerovive treated arm. First, we started with the assumption that any patient
that responded to Cerovive experienced only a one point improvement in mRankin
" this is the most aggressive assumption you can make in terms of calculating
the number of patients that responded to Cerovive. We then started on the
right-hand side of the table below and calculated how many patients shifted in
each mRankin Category.
mRankin 5/Death. If you start on the right-hand side you will see that there
was no change in the number of patients that were dead or scored a 5 on
mRankin. No patients moved into or out of this category.
# of Patients In Each mRankin Category
|mRankin Score |0 |1 |2 |3 |4 |5/Death|
|Placebo |93 |170 |99 |108|175 |204 |
|Cerovive |131|153 |97 |121|144 |204 |
|Increase (Decrease)|38 |(17)|(2)|13 |(31)|0 |
|Patient Responders |38 |20 |18 |31 |0 | |
Source: RNVS/AZN Presentation and PJC Company Estimates
mRankin 4. Next we will consider patients scoring a 4 on the mRankin scale. If
you look at the table below you will see that 31 fewer patients had an mRankin
of 4 than in the placebo arm. Therefore we assume that those 31 patients had a
one point improvement and became 3's on the mRankin Scale.
# of Patients In Each mRankin Category
|mRankin Score |0 |1 |2 |3 |4 |5/Death|
|Placebo |93 |170 |99 |108|175 |204 |
|Cerovive |131|153 |97 |121|144 |204 |
|Increase (Decrease)|38 |(17)|(2)|13 |(31)|0 |
|Patient Responders |38 |20 |18 |31 |0 | |
Source: RNVS/AZN Presentation and PJC Company Estimates
mRankin 3. Again looking at the table below you will see that 13 more patients
receiving Cerovive scored a 3 on the mRankin scale. But remember that 31
patients moved from 4 to 3 and those patients only improved by 1 point,
therefore we assume that 18 patients that would have scored a 3 on mRankin
were improved to a 2.
# of Patients In Each mRankin Category
|mRankin Score |0 |1 |2 |3 |4 |5/Death|
|Placebo |93 |170 |99 |108|175 |204 |
|Cerovive |131|153 |97 |121|144 |204 |
|Increase (Decrease)|38 |(17)|(2)|13 |(31)|0 |
|Patient Responders |38 |20 |18 |31 |0 | |
Source: RNVS/AZN Presentation and PJC Company Estimates
mRankin 2. The number of patients scoring a 2 on mRankin decreased by 2,
however remember that we had 18 patients in the 3 category that were improved
to 2, therefore if the category shrunk overall by 2 patients, we can assume
that 20 patients improved from a 2 to a 1 on the mRankin.
# of Patients In Each mRankin Category
|mRankin Score |0 |1 |2 |3 |4 |5/Death|
|Placebo |93 |170 |99 |108|175 |204 |
|Cerovive |131|153 |97 |121|144 |204 |
|Increase (Decrease)|38 |(17)|(2)|13 |(31)|0 |
|Patient Responders |38 |20 |18 |31 |0 | |
Source: RNVS/AZN Presentation and PJC Company Estimates
mRankin 1. The number of patients scoring a 1 on the mRankin decreased by 17,
however recall that 20 patients improved from 2 to 1, therefore in order for
the category to decrease by 17 patients overall we assume 38 patients (numbers
slightly off due to additive effect of rounding) improved from a 1 to a 0.
# of Patients In Each mRankin Category
|mRankin Score |0 |1 |2 |3 |4 |5/Death|
|Placebo |93 |170 |99 |108|175 |204 |
|Cerovive |131|153 |97 |121|144 |204 |
|Increase (Decrease)|38 |(17)|(2)|13 |(31)|0 |
|Patient Responders |38 |20 |18 |31 |0 | |
Source: RNVS/AZN Presentation and PJC Company Estimates
mRankin 0. We can see in the table that the mRankin 0 category grew by 38
patients, which means that 38 more patients experienced no symptoms of stroke
at 3-months when they got Cerovive, compared to placebo.
# of Patients In Each mRankin Category
|mRankin Score |0 |1 |2 |3 |4 |5/Death|
|Placebo |93 |170 |99 |108|175 |204 |
|Cerovive |131|153 |97 |121|144 |204 |
|Increase (Decrease)|38 |(17)|(2)|13 |(31)|0 |
|Patient Responders |38 |20 |18 |31 |0 | |
Source: RNVS/AZN Presentation and PJC Company Estimates
Calculating Treatment Effect. If you now go back and sum all of the patients
that had a 1 point move on the mRankin Scale, with rounding you get 108
patients experienced a 1-point improvement on the mRankin Scale. Therefore
assuming Cerovive can only improve patients by 1 point on the mRankin, 108
(again rounding skews the numbers slightly) out of 850 or 12.7% of treated
patients experienced a 1-point improvement.
# of Patients In Each mRankin Category
|mRankin Score |0 |1 |2 |3 |4 |5/Death|
|Placebo |93 |170 |99 |108|175 |204 |
|Cerovive |131|153 |97 |121|144 |204 |
|Increase (Decrease)|38 |(17)|(2)|13 |(31)|0 |
|Patient Responders |38 |20 |18 |31 |0 | |
Source: RNVS/AZN Presentation and PJC Company Estimates
Real Effect. We believe that these two scenarios describe the lowest and
highest range of patients that could have experienced improvement from
Cerovive. In reality it is unlikely that Cerovive treatment only improves
patients by 1 point on the mRankin and alternatively it is just as unlikely
that the only effect was seen in patients from the mRankin 4 and mRankin 0
categories. We therefore estimate conservatively that Cerovive likely improved
outcome in 9-10% of patients.
Understanding The Impact " mRankin. In order to best understand the impact of
moving one point on the mRankin scale we have included a table that describes
each mRankin category. While some might question the clinical relevance of a
one point move on the mRankin scale we will point out that moving from 4 to 3
implies that a patient essentially care for themselves (walk without
assistance and attended to their own bodily needs). So we view a 1 point move
as very clinically meaningful.
Modified Rankin Scale
|Score|Description |
|0 |No symptoms at all |
| | |
|1 |No significant disability despite symptoms; able to carry out all |
| |usual duties and activities |
|2 |Slight disability; unable to carry out all previous activities, but|
| |able to look after own affairs without assistance |
|3 |Moderate disability; requiring some help, but able to walk without |
| |assistance |
| | |
|4 |Moderately severe disability; unable to walk without assistance and|
| |unable to attend to own bodily needs without assistance |
|5 |Severe disability; bedridden, incontinent and requiring constant |
| |nursing care and attention |
| | |
|6 |Dead |
Source: www.strokecenter.org
Important Research Disclosures
------------------------------------------------------------------------------
Analyst Certification - Mark E. Karvosky, Sr. Research Analyst
The views expressed in this report, including the Key Points and Risk sections
in particular, accurate |
