CS: Sell Off Overdone - Reiterate Outperform
Conclusion: With the stock down ~40% on the day from already depressed levels,
we think investors over reacted to the R788 data. That is not to say we do not think
R788 has issues, but we think most have already been disclosed and the new data
do not dramatically change our view. Some concern was raised by differences in
efficacy in Mexico versus the US, but adjusting for the higher placebo rate in Mexico,
the benefit in ACR scores was similar across geographies. In terms of toxicity, R788
had a dose dependent increase in blood pressure (BP), which should not have been
a surprise as the company has talked about the effect on BP since the data were first
presented. Still, they presented the magnitude for the first time and there is no
question that this is one of the risks of the program. That said, we remind investors
that this is a drug to treat RA, which has a higher risk tolerance compared to other
indications and look at Actemra, which showed a meaningful increase in lipids that
was not a cause for significant concern at a recent FDA panel meeting. We believe
the other toxicities, such as the LFT abnormalities and neutropenia were also fully
disclosed for months and should not have sparked any new concern.
What’s New? Rigel presented its Phase IIa data at the annual American College of
Rheumatology meeting. While efficacy had already been reported, there were some
new information on the break out between patients enrolled in Mexico and the US, as
well as additional toxicity data (See Exhibit 1 & 2).
Implication: We were surprised to the stock reaction to today’s data. That said,
based on the compounds profile, we believe we have been appropriately conservative
in our approach to R788’s probability of success (~45%) and think today’s reaction
was over blown. That said, in today’s market we recognize a return to earlier levels
will likely take time. We look towards a partnership announcement in 2009 as the
next important catalyst and PIIb data later in 2009 and remain Outperform rated.
Rating OUTPERFORM* [V]
Price (27 Oct 08, US$) 8.84
Target price (US$) 37.00¹
52-week price range 28.19 - 6.67
Market cap. (US$ m) 526.11
Enterprise value (US$ m) 402.74
*Stock ratings are relative to the relevant country index.
¹Target price is for 12 months.
Research Analysts
Michael Aberman, M.D.
Y. Katherine Xu, Ph.D.
Ying Huang, Ph.D.
ying.huang.2@credit-suisse.com
Nikhil Khetarpal
27 October 2008
Rigel Pharmaceuticals Inc. (RIGL) 2
Key Takeaways From ACR
Higher Placebo Response Rate Confounds Absolute ACR Response Rates – But
Magnitude Of Benefit Over Placebo The Same Across Geographies
One of the first areas of concerns that was raised from today’s data presentation was the
fact that patients enrolled in Mexico had a higher placebo response rate than in the US. A
high placebo response could also lead to higher treatment response. This is a particular
concern since the ratio of Mexican patients to US patients was higher in the higher dosing
groups, which could skew the data in favor of R788. However, one way to remove this
potential bias is to analyze the data separately for each country. This allows one to see if
the magnitude of the benefit is similar over placebo as opposed to only being driven by
geographic differences. As can be seen in Exhibit 1, the magnitude of benefit over
placebo in ACR20, ACR50, and ACR 70 scores is equivalent regardless of the country
where the patient is treated. While it is true that one can see the “absolute” ACR 20 score
is lower in the US than Mexico, the important measure is the benefit over placebo. This
benefit of ~20-26% is comparable to what has been seen with TNF inhibitors at 3-months
and continues to convince us that R788 is as efficacious as those injectable agents.
Exhibit 1: Efficacy of R788 by Country
Total Placebo 100 MG 150 MG Combined 100/150 MG
N 47 46 49 47 96
ACR20 18 38% 15 33% 32 65% 34 72% 66 69%
ACR50 9 19% 8 17% 24 49% 27 57% 51 53%
ACR70 2 4% 1 2% 16 33% 19 40% 35 36%
US Placebo 100 MG 150 MG Combined 100/150 MG
N 25 46 21 5 26
ACR20 6 24% 15 33% 11 52% 2 40% 13 50%
ACR50 1 4% 8 17% 6 29% 2 40% 8 31%
ACR70 0 0% 1 2% 3 14% 2 40% 5 19%
Mexico Placebo 100 MG 150 MG Combined 100/150 MG
N 22 0 28 42 70
ACR20 12 55% 0 21 75% 32 76% 53 76%
ACR50 8 36% 0 18 64% 25 60% 43 61%
ACR70 2 9% 0 13 46% 17 40% 30 43%
50 MG
50 MG
50 MG
Source: Company data, Credit Suisse estimates
Blood Pressure Elevation an Issue, But Not A Game Stopper
The next most important topic of debate at ACR was the issue of R788’s blood pressure
elevation. Specifically, the mean increase in blood pressure elevation over baseline was 4
mmHg in the 100mg PO BID group, which is the highest dose moving forward in the
Phase IIb trials enrolling. With the FDA’s increased scrutiny over cardiac toxicity and the
well known association of elevated blood pressure with cardiac events, this toxicity is not a
non-issue. That said, the company had previously disclosed elevated BP as a toxicity with
R788 and this should not be a surprise to investors. We also note that elevated BP is
relatively easy to treat and those patients who actually had hypertension as an adverse
event were able to be managed by either dose reductions, discontinuation, and/or the
addition of anti-hypertensive medications.
Perhaps more important, we understand that at least one investigator not involved in the
R788 program suggested that this level of elevated BP would be a show stopper clinically.
We do not think that is the case and point to other drugs that treat RA, such as the anti-
TNF inhibitors, that have a laundry list of toxicities (including lymphoma), which reflects the
medical benefit afforded in this indication. We also look at Actemra, a drug that has
considerable excitement within the rheumatology community yet causes significant
elevations in lipids. Like blood pressure, elevated lipids correlate with an increase in
cardiac events. And, while this issue was discussed at a recent advisory committee
meeting for Actemra, it was not a contentious issue for the FDA.
We are very comforted by the fact that despite the blood pressure elevation is included in
the investigator’s brochure used in the ongoing Phase IIb trials and has been highlighted
(according to the company) at several investigator meetings, the trial is enrolling ahead of
expectations. This suggests to us that clinicians who treat patients with RA are not overly
concerned with this level of BP elevation. Again, we are not saying that the elevated BP
will not be an ultimate concern for the FDA, but we think this is adequately reflected in our
probability of success. We also believe the company is taking an appropriately aggressive
approach to this issue, trying to ensure that patients in its clinical trials have properly
managed BP to reduce this signal in future trials.
Lastly, one area that has raised concern is the magnitude of the BP rise in those patients
who had hypertension as an adverse event. While the complete patient level data were
not disclosed, the company disclosed that the BP went up as much as 20-30 mmHg. This
magnitude is concerning in that it could precipitate significant morbidity acutely, such as a
cardiac event. While the rate of hypertension as an adverse event was relatively rare, this
is probably the biggest risk to the program and one that bears watching in the Phase IIb
program.
Liver Function Abnormalities Not Much Different From Methotrexate Alone
Another concern for R788 has been its effect on liver enzymes. Again, this is an issue that
has been fully disclosed by Rigel ahead of this meeting. Still, it is worth noting that the
rate of LFT abnormalities considered clinically relevant (e.g. >2X ULN) were similar in the
100mg PO BID group (the dose moving forward) and the placebo arm in this trial (see
Exhibit 2). That said, there are clearly some additive liver abnormalities as evidenced by
the increase in patients who had LFT levels of more than 1.2X ULN. We think such
increases (that are less than 2XULN) are less likely to cause a concern either with the
FDA or the clinical community. In the open label extension data, also presented at this
meeting, there were an additional five patients who had a single isolated elevation of 3X
ULN, two who had this confirmed at a later date. Based on data that were presented at
this same meeting, this level of LFT abnormality is likely no different that what could be
expected with methotrexate alone, although with no control arm, it is hard to prove this
hypothesis.
Exhibit 2: Liver Function Profile of R788
Placebo 50 mg bid 100 mg bid 150 mg bid
Neutrophils <1500 0 1 (2%) 5 (14%) 14 (30%)
ALT > 1.2 XULN 4 (8%) 5 (11%) 12 (25%) 22 (48%)
ALT > 2XULN X2 1 (2%) 0 0 4 (8%)
ALT > 3XULN 2 (4%) 0 0 3 (6%)
AST > 3XULN or > 2XULN X2 1 (2%) 0 1 (2%) 1 (2%)
Source: Company data, Credit Suisse estimates
Neutropenia Also Not New
Like the liver toxicity, there was nothing new about the neutropenia described in the data
presentation at the ACR meeting. Neutropenia is clearly a dose dependent toxicity with
R788 and occurred in 14% of patients in the 100mg BID arm. We think that
rheumatologists will be very comfortable monitoring this toxicity and are comforted by the
fact that neutrophils return to normal after discontinuation of therapy (and/or dose
27 October 2008
Rigel Pharmaceuticals Inc. (RIGL) 4
reductions). In the open label extension, only 3 patients (out of 78) had confirmed (>1
episode) of neutropenia and only 1 discontinued therapy for neutropenia.
Bottom Line – New Data Do Not Meaningfully Change Our View
We are not arguing that some of the new disclosures about the side effect profile do not
increase our level of concern over the long term prospects of R788. However, our view
has always been that R788 was a relatively early in development and had a toxicity profile
that could ultimately prevent approval in RA. In fact, our base case has only a 45%
probability of R788 coming to market. We do not think today’s data should meaningfully
change this probability. We will continue to get feedback from clinicians to see if we need
to change our view. |
