Citi: Pmab to Become EGFR MAb of Choice; Raising ABGX Target Price to $17 November 3, 2005 SUMMARY * We provide more info on Ph III Pmab data. The hazard ratio was 0.54 indicating a 46% reduction in tumor progression rate. This was derived from the Kaplan-Meier curves for PFS for the two arms, where the curves separated at 8 wks (time of 1st scan) & remained separated. The 95% confidence intervals did not overlap suggesting the Pmab benefit remained robust.
* With these data, we believe Pmab is likely to become the EGFR Mab of choice given its better safety profile (lower infusion rxns, no premedication) and greater flexibility (every 2 wks) vs. Erbitux (every wk, premed required).
* We have raised our Pmab U.S. sales ests to $225MM from $188MM in FY07, to $334MM from $295MM for FY08, & to $592MM from $510MM for FY09 driving our ABGX EPS ests to ($0.43) from ($0.55) for FY07, to $0.00 from ($0.12) for FY08 & to $1.15 from $0.89 for FY09. Our ABGX target price is revised to $17 per share from $13 per share.
SUMMARY VALUATION AND RECOMMENDATION DATA
OPINION
This morning, partners Amgen and Abgenix held a conference call to discuss the positive top-line results of the pivotal Phase III International trial (Study 408), which compared panitumumab (Pmab) in combination with best supportive care versus best supportive care (BSC) in patients with advanced (third-line) colorectal cancer (CRC). The 463-patient study achieved a statistically significant benefit in the primary endpoint of progression-free survival (PFS). Specifically, the companies indicated that the hazard ratio was 0.54 in favor of Pmab indicating that treatment with Pmab provided a 46% decrease in tumor progression rate with a p value of 0.000000001. These data were derived from a Kaplan-Meier analysis of the primary endpoint of PFS, where the curves showed a distinct separation beginning at eight weeks, the time of the first radiographic scan to assess tumor progression, and remained separated. The companies indicated that the 95% confidence intervals (i.e., error bars) for the two curves never overlapped implying that the data was exceptionally robust.
Given the advanced stage of disease for these patients, Amgen and Abgenix highlighted how almost all patients in the best supportive care arm had progressed by the first 8-week scan, and therefore, suggested that an analysis of the rate of tumor progression versus median PFS may be more useful in assessing treatment effect. Patients treated with Pmab plus BSC who achieved a partial response or had stable disease accounted for a significant amount of the differential in rate of tumor progression. According to the companies, these patients represented about 40% of patients in this treatment arm. To further underscore the strength of the data, the companies indicated that various sensitivity analyses, including the region of the tumor (colon versus rectal) and the number of prior chemotherapy regimens, confirmed the robust benefit of the Pmab plus BSC arm compared to best supportive care alone. The companies also indicated that this benefit was also independent of whether the patient was assessed with a scan at the pre-specified time points (i.e., initially at 8 weeks then every 4 weeks for the next two scans and then every 8 weeks) or at more frequent intervals for scanning in either treatment arm. As a reminder, physicians were allowed to conduct a radiographic scan at any time he/she believed the patient may be experiencing disease progression (i.e., potential tumor progression). In our opinion, the consistency of the data showing a benefit with Pmab plus BSC under a variety of sensitivity analyses highlights the robust efficacy of this compound. In addition, the companies indicated that response to Pmab was not differentiated by level of EGFR expression. We note that the product label for Erbitux indicates that patients need to be assessed for EGFR expression levels prior to receiving treatment as the study for Erbitux had shown a differential response based on levels of EGFR expression.
The secondary endpoint in study 408 of objective response rate was also met. While the specific response rate data as well as the median PFS were not disclosed, Amgen and Abgenix indicated that the response rate observed was consistent with results from prior studies including: (1) the two ongoing U.S. studies, one in third-line colorectal cancer (study 167) and another in patients with low or negative EGFR expression as a third-line treatment for CRC (study 250) where the response rate was approximately 10%; and (2) the previous Phase II monotherapy study of Pmab where the response rate was approximately 9%. In addition, another secondary endpoint in study 408 of overall survival will be evaluated at the 12-month time point. The overall survival data are expected to become available in the second quarter of 2006. The companies indicated that the median follow-up to date is approximately 20 weeks. Despite the fact that the review of the overall survival data is ongoing, the companies indicated the undisclosed median progression free survival (PFS) data is likely to remain unchanged as there have been 401 progression events already among the 463 patients enrolled. We believe the detailed data, including primary and secondary endpoints as well as detailed safety data, will be presented at the 2006 American Society of Clinical Oncology (ASCO) meeting next June.
We had expected there was a high likelihood of positive top-line results from this Phase III Pmab study given data from prior clinical studies and from clinical studies of Bristol-Myers Squibb/ImClone's Erbitux in similar settings. As a reminder, at this year's ASCO meeting, Amgen and Abgenix released data from a Phase II study indicating an overall response rate of 9%, a median progression-free survival time of 13.6 weeks, and a median overall survival time of 37.6 weeks (median follow up of 30.9 weeks) when Pmab was administered as a single agent therapy in patients with positive EGFR tumor expression who failed standard chemotherapy (i.e., a fluoropyrimidine +/- leucovorin and either irinotecan or oxaliplatin or both). A summary of these results are provided below:
Phase II Panitumumab Monotherapy Study - Objective Response Data
Cohort A (n=105) Cohort B (n=44) Total (n=148) Overall response 7 patients (7%) 6 patients (14%) 13 patients (9%) - Partial response 7 patients (7%) 6 patients (14%) 13 patients (9%) - Complete response 0 patients (0%) 0 patients (14%) 0 patients (0%) Stable disease 36 patients (34%) 7 patients (6%) 43 patients (29%) Progressive disease 35 patients (33%) 24 patients (56%) 59 patients (40%) Cohort A: 2+ or 3+ EGFr overexpression by IHC in =10% of tumors
Cohort B: 2+ or 3+ EGFr overexpression by IHC in <10% of tumors, but 1+ or 2+ or 3+ in =10% of tumor cells
Source: 2004 ASCO Meeting, Presentation, Hecht et al. and Nov. 2005 Chemotherapy Foundation Meeting
Phase II Panitumumab Monotherapy Study -- Survival Data
Kaplan-Meier data, in Cohort A (n=100) Cohort B (n=41) Total (n=141) weeks Median PFS, 95% CI 16.0 (9.6, 16.4) 8.3 (8.0, 16.3) 13.6 (8.3, 16.1) Median Overall 39.9 (23.3, 44.7) 37.5 (24.1, 47.7) 37.6 (25.6, 42.4) Survival, 95% CI Cohort A: 2+ or 3+ EGFr overexpression by IHC in =10% of tumors
Cohort B: 2+ or 3+ EGFr overexpression by IHC in <10% of tumors, but 1+ or 2+ or 3+ in =10% of tumor cells
Source: 2004 ASCO Meeting, Presentation, Hecht et al. and Nov. 2005 Chemotherapy Foundation Meeting
Pivotal Phase III Study Design (Study 408): The randomized, controlled trial enrolled 463 patients in Europe, Australia, and Canada who had failed more than two chemotherapy regimens, including fluorouracil (5-FU), irinotecan, AND oxaliplatin. Some patients had also received prior treatment of Avastin but not within three months of being enrolled in this study. The study was designed to evaluate the effectiveness of best supportive care plus panitumumab (n=231) as a stand-alone therapy administered every other week versus best supportive care in patients (n=232) with spreading colorectal cancer who have failed standard chemotherapy. Patients were randomized to receive 6 mg/kg IV every other week of panitumumab with no pre-medication necessary. The first scan was scheduled at 8 weeks and then at every 4-week interval for the next two scans, and then every 8 weeks thereafter. After 48 weeks, scans are scheduled for every three months. Physicians could also elect to scan patients if physicians felt a patient was showing signs of disease progression. The local physician reviewed each of the scans, but the final data analysis was conducted by an independent central review committee. The results from this independent committee were disclosed today and are expected to be presented at ASCO next year. As noted, the study was powered to show a 33% decrease in tumor progression rate. Secondary endpoints included response rate, overall survival, and safety.
Safety Data. In terms of safety, the companies indicated that panitumumab was well tolerated. The most common side effect was acneiform rash, a known reaction of this class of compounds. Other less commonly observed side effects were fatigue, nausea, and mild diarrhea. In addition, the companies indicated that commonly observed side effects with this class of EGFR inhibitors, including conjunctivitis (approximately 17% incidence rate), folliculitis, nail infections, and others, were seen in this study, although none of these occurred with high frequency at severe levels (i.e., Grade 3 or 4). As a reminder, data released yesterday from a Phase II 24-patient study of panitumumab in combination with chemotherapy in first-line colorectal cancer at the Paris ECCO (European Cancer Conference) meeting indicated that conjunctivitis occurred at a rate of 17% (none at Grade 3 or 4), which raised investor concerns. We note that the Erbitux BOND study showed a similar rate of conjunctivitis (14%, any grade) in patients treated with Erbitux and irinotecan, and as indicated, we believe this is a relatively common yet mild side effect observed with this class of compounds.
The dropout rate in this study was relatively low. Overall, only 13 patients (6%) in the panitumumab plus best supportive care arm discontinued therapy due to adverse events, which was two times higher than the best supportive care arm alone. Importantly, Amgen and Abgenix also emphasized that there were only rare, if any, severe infusion reactions and symptoms associated with this were rare. As a result, administration of panitumumab did not require pre- medication. This is in direct contrast to competitor ImClone/Bristol-Myers Squibb's Erbitux, which has a black box warning on its label related to the potential for severe infusion and hypersensitivity reactions. Given the potential for severe infusion reactions, particularly in the first infusion, pre-medication with administration of Erbitux is typically required.
Regulatory Strategy. Based on these results and fast-track designation for panitumumab, Amgen plans to initiate a rolling regulatory submission later this year, which is expected to be complete by Q1 2006. The companies plan to request a priority review (i.e., 6-month) given the strength of the data and the need for improved treatments for this advanced disease. Whether the BLA is also considered for accelerated approval, which would require further studies to assess the impact on survival in the future, is likely to be dependent on the outcome of the overall survival data analysis from this Phase III study.
Revising Our Abgenix Financial Model. Given these positive Phase III panitumumab results, we have revised our panitumumab U.S. sales estimates in our financial model. As a reminder, Abgenix has a 50/50 profit-sharing arrangement with Amgen on panitumumab. We have revised our U.S. Pmab sales estimates to $225 million from $188 million for fiscal 2007, to $334 million from $295 million for fiscal 2008, and to $592 million from $510 million for fiscal 2009. Thus, we have adjusted Abgenix's profit share of the panitumumab revenues to $70 million from $58 million for fiscal 2007, to $111 million from $100 million for fiscal 2008 and to $195 million from $170 million for fiscal 2009. Consequently, our EPS estimates are adjusted to ($0.43) from ($0.55) for fiscal 2007, $0.00 from ($0.12) for fiscal 2008 and $1.15 from $0.89 for fiscal 2009.
Our Abgenix financial model continues to suggest the company will achieve profitability in fiscal 2009 with our revised EPS estimate of $1.15 (previously $0.89) per share. We have lowered our discount rate to 25%-30% (previously 30%-35%), as we believe Panitumumab is a potential approvable product with less regulatory risk now given the positive Phase III top-line results released today. Our target price is revised to $17 per share from $13 per share based on applying a P/E multiple of 30-35x and a discount rate of 25%-30% to our 2009 earnings of $1.15. We believe this multiple is appropriate when one examines historically what investors have been willing to apply on biotechnology companies with a lead compound in advanced stages of development and significant growth prospects. Such companies typically trade at a multiple in the range of 23x-39x, by our estimates, with a mean of approximately low 30s. Given the positive pivotal data from Abgenix's lead product candidate (i.e., panitumumab), we believe that Abgenix should trade at a multiple near the high- end of this comparable range. We note that Abgenix is partnered with a major biotechnology company, Amgen, for this program.
Impact to Amgen Financial Model. We have yet to include any panitumumab- related revenue (from colorectal cancer or other indications) in our Amgen model, but we estimate these revenues will have a minimal impact on our fiscal 2006 EPS estimate of $3.66 for Amgen. For fiscal 2007, we estimate the potential impact will still be modest of approximately $0.03-$0.05 to our EPS estimate of $4.22 for Amgen.
ANALYSIS OF COMPETITIVE LANDSCAPE -- PANITUMUMAB VS. ERBITUX
We expect many investors may try to draw comparisons between these results and those for Bristol-Myers Squibb/ImClone's Erbitux. However, there are distinct differences between the studies for Panitumumab and Erbitux in terms of patient population and endpoints. As a reminder, Erbitux is approved for use once weekly in combination with irinotecan or as a monotherapy based on data from three trials conducted in patients with EGFR-expressing, metastatic colorectal cancer. Patients included in the Erbitux studies had progressive disease after receiving an irinotecan-containing regimen. As noted previously, patients in the Phase III panitumumab study had failed more than two chemotherapy regimens including fluorouracil, irinotecan, and oxaliplatin. Therefore, patients enrolled in the panitumumab study represented more heavily pre-treated patients. Specifically, all patients (n=463) enrolled in the panitumumab study had failed more than two standard chemotherapy regimens as compared to only about 44% (44/111) of patients in the Erbitux BOND study (see Table 1 below).
We believe the endpoint of objective response rates (ORR) may serve as the most suitable measure for comparison given the limited data released at this time. We provide data from the largest randomized, controlled study for Erbitux called the BOND study in Table 1. In this study, 329 patients were randomized to receive either Erbitux plus irinotecan (218 patients) or Erbitux monotherapy (111 patients). Of the 329 patients, approximately two-thirds of the patients had previously failed oxaliplatin treatment, 58% had colon cancer, and 40% had renal cancer. The Erbitux dosing regimen for both arms of the trial was as follows: 400 mg/mm for the initial dose; 250 mg/mm weekly until either disease progression or some toxicity was reached. The most directly comparable results to Panitumumab monotherapy studies are highlighted.
Table 1: Erbitux BOND study - Objective Response Rate (ORR) Per Independent Review
Source: Erbitux label.
The BOND study for Erbitux also measured time to progression (TTP) as a secondary endpoint, which may be considered the closest measurement to the primary endpoint of progression-free survival (PFS) in the Phase III panitumumab study. This may be especially relevant since Erbitux's TTP definition included the event of death, which is also a necessary criterion for PFS. Specifically, in the Erbitux BOND study, TTP was defined as the period from the date of randomization to the first observation of disease progression or to death from any cause within 60 days after randomization or the most recent tumor assessment (Source: New England Journal of Medicine article on the Erbitux BOND study published on July 22, 2004). Depending on the study design, TTP as an endpoint can includes any disease progression or death whereas PFS always includes death. While the definition for TTP in the Erbitux BOND study includes death as an event, we note that both these endpoints are relatively subjective and may be dependent on when MRI scans are conducted to monitor disease progression. Tumor response for the Erbitux BOND study was evaluated every 6 weeks for the first 24 weeks and then every 3 months thereafter. The frequency of scans in the pivotal Phase III Panitumumab study was an initial scan at 8 weeks followed by a scan every 4 weeks following a cycle of treatment. Although, according to Abgenix, there were provisions in the Phase III study to allow the physician to conduct a scan if progression was suspected based on a clinical assessment. With this as a cautionary background, the most directly comparable results from the Erbitux BOND study to the Panitumumab monotherapy study are highlighted in Table 2 below.
Table 2: Time to Progression Per Independent Review
Source: Erbitux label.
We believe it is important to note in the discussion of endpoints that Erbitux was approved under accelerated approval, and still has an unfulfilled postmarketing commitment to the FDA to demonstrate a survival benefit. To meet its obligation to the FDA, Bristol-Myers Squibb/ImClone are performing a Phase III survival trial (BMS-006) in the second-line setting. This study, which is being conducted with Merck KgaA, opened for patient enrollment in May 2003. According to ImClone, the BMS-006 trial tests the addition of Erbitux to irinotecan in patients who have failed oxaliplatin, 5-fluorouracil, and leucovorin in the first line of therapy. A second study, BMS-014, was discontinued and removed from the list of post-approval commitments with the FDA's approval because of decreased feasibility. Data from the BMS-006 study is expected by late 2006. We note that the FDA's Oncologic Drugs Advisory Committee is planning to review unfulfilled commitments for Erbitux and other oncology products at an upcoming meeting.
Panitumumab's Differential Safety and Dosing Benefit. As mentioned previously, we believe the objective response clinical data suggests Panitumumab is at least equivalent in efficacy to ImClone/Bristol-Myers Squibb's Erbitux. Thus, we believe panitumumab has the opportunity to become the EGFR monoclonal antibody of choice based on its better safety profile (lower infusion reactions, no pre-medication) and greater flexibility (every 2 weeks) vs. Erbitux (every week, pre-medication required). Specifically, Panitumumab has shown lower infusion reactions with no need for pre-medication (better safety profile) and greater administration flexibility (every two weeks). Further, we note that data from a Phase I study presented at the 2005 ASCO meeting reveal that Panitumumab has anti-tumor activity in a range of flexible dosing regimens, including every week, every other week, and every three weeks.
We note Erbitux's label was updated on September 1, 2005 to primarily reflect the recommendation for a one-hour observation period following Erbitux infusions. The updated label indicates that longer observation periods may be necessary for those who experience infusion reactions. This information was distributed by a subsequent Dear Healthcare Provider letter. We remind investors that since the product's launch, Erbitux's label has included a black box warning related to infusion reactions, which occur in approximately 3% of patients and are characterized by rapid onset of airway obstruction, urticaria, and hypotension.
Planning for Panitumumab Expansion. While today's top-line results in advanced colorectal cancer are significant, we believe Amgen and Abgenix are aiming to carve out larger market opportunities for Panitumumab by moving it into the earlier setting in colorectal cancer in combination with other targeted therapies and traditional chemotherapeutic regimens. To support this goal, the companies announced earlier this year the initiation of the PACCE study (Panitumumab Advanced Colorectal Cancer Evaluation). As a reminder, this trial is a randomized, multi-center, open-label study that is targeted to enroll approximately 1,000 patients to evaluate the potential benefits of adding Panitumumab administered every other week to Avastin and either oxaliplatin or irinotecan-based chemotherapy for the first-line treatment of metastatic colorectal cancer. The primary endpoint for this study is progression free survival and secondary endpoints include response rate, overall survival, and safety. We expect this interim analysis will be available for an internal review in Q4 2005 with the data expected to be presented in 2006 at an appropriate medical meeting. Since this study just began enrolling at the end of April 2005, the interim analysis release may reflect a surrogate endpoint, such as response rate. We note that the pivotal study for Avastin plus standard chemotherapy in the first line metastatic setting, the median overall survival for patients was nearly 11 months. On today's call, Amgen's Senior Vice President of R&D, Roger Perlmutter, MD, PhD, conveyed his belief that panitumumab is best used in an earlier CRC setting rather than in the salvage setting as was investigated in study 408. We note that success in this setting would greatly expand the opportunity for panitumumab.
VALUATION AND RISKS -- COMPANIES DISCUSSED
Amgen Inc. (AMGN--$77.51; 1M)
Valuation
Our target price is $100. We believe Amgen should trade on par with the large- cap biotech companies given the robust earnings growth and advancing pipeline. The large-cap biotech companies currently trade at a group average PEG (P/E to growth) ratio of approximately 1.4x (range: 1.2-x1.9x) and an average P/E in the high-20s (range: 19x-53x). Our target price of $100 per share is derived by applying an approximate PEG ratio of 1.4x and P/E multiple in the high-20s to our 2006 EPS estimate of $3.66.
We have also applied a second valuation approach utilizing an analysis of residual earnings. In other words, the value of the equity is assessed by adding the present value of the residual earnings plus the book value of the company. Residual earnings is defined as the earnings stream remaining after accounting for the required return on investment, and is determined by subtracting the multiple of the required return on investment times the book value from projected earnings per year. Using this analysis for Amgen, we calculated a book value per share of $14.92 and assumed a time horizon of three years. Our estimate for the discount rate is 9.0%, which assumes a beta of 1.00 for AMGN, a market risk premium of 5.0% and a risk free rate of 4.0%. In addition, we assumed a constant growth rate of 7.5% for the biotech industry. Using this approach, we also derive a target price of approximately $100-$105 per share, based on earnings estimates.
Risks
We rate Amgen Medium Risk due to the moderate predictability of the company's financial results, strong cash flow, low debt-to-earnings ratio (15%), and moderate stock price volatility compared to other large-cap biotechnology companies. Key risks we see to Amgen achieving our valuation target include the following: any delay in clinical development or regulatory approval of developing products could have a material impact on the company's earnings. From time to time, the Centers for Medicare/Medicaid Services (CMS) proposes changes in the reimbursement for drugs, such as Epogen, Aranesp, Neupogen and Neulasta, and other products which could limit the pricing and use of these drugs. Additionally, if sales for key products such as Aranesp, Neulasta, Epogen, Neupogen or Enbrel fail to grow, shares would likely be impacted.
If the impact on the company from any of these factors proves to be greater than we anticipate, the stock will likely have difficulty achieving our target price.
Abgenix, Inc. (ABGX--$12.90; 2S)
Valuation
In order to value the company, we have utilized a discounted earnings analysis and applied a discount rate and P/E multiple, both of which we believe are appropriate for a high-growth biotechnology company with a Phase III program, to a projected EPS estimate for the company. Our financial model suggests the company will achieve profitability in fiscal 2009 with our revised EPS estimate of $1.15 (previously $0.89) per share. We have lowered our discount rate to 25%-30% (previously 30%-35%), as we believe Panitumumab is a potential approvable product with less regulatory risk now given the positive Phase III top-line results released today. Our target price is revised to $17 per share from $13 per share based on applying a P/E multiple of 30-35x and a discount rate of 25%-30% to our 2009 earnings of $1.15. We believe this multiple is appropriate when one examines historically what investors have been willing to apply on biotechnology companies with a lead compound in advanced stages of development with significant growth prospects. Such companies typically trade at a multiple in the range of 23x-39x, by our estimates, with a mean of approximately low 30s. Given the positive pivotal data from Abgenix's lead product candidate (i.e., panitumumab), we believe that Abgenix should trade at a multiple near the high-end of this comparable range. We note that Abgenix is partnered with a major biotechnology company, Amgen, for this program.
We have also assessed Abgenix's valuation using a discounted cash flow (DCF) analysis. Our DCF analysis assumes a time horizon of 10 years, with 2015 serving as our year for assessing a terminal value. Our estimate for a cost of equity is revised to approximately 9% from 10%. Our cost of equity assumes a beta of 1.15 (previously 1.10), a market risk premium of 3.8% (previously 5.0%), and a risk free rate of approximately 4.21% (previously 4.50%). We have also applied a terminal growth rate of 6.0%. Using this approach, we arrive at a target price of approximately $17 per share.
Risks
We believe a Speculative Risk rating is warranted for Abgenix given the company's dependence on panitumumab (ABX-EGF) and the high volatility of its shares. We outline the Risks we see to Abgenix achieving our valuation target below.
Like all biotechnology companies developing proprietary products, Abgenix is subject to clinical development setbacks, which could delay or hamper profitability. Currently there is an acute shortage of manufacturing capacity in the monoclonal antibody area and many companies, including Abgenix, are building new commercial-scale facilities to address this issue. Furthermore, any patent issues in the EGFr antagonist field will likely have a negative effect on the shares of Abgenix. Abgenix is also highly dependent on its corporate partners, Amgen and AstraZeneca for funding and commercialization support. Any termination of these relationships would be materially detrimental to Abgenix.
We also note, however, that if Abgenix's lead product, panitumumab, demonstrates positive results in other studies that this could represent an opportunity for the stock's price appreciation.
Investment Thesis
We rate the shares of Abgenix (ABGX) Hold/Speculative risk (2S) with a target price of $17 per share (previously $13). Continued progress of the company's late-stage clinical pipeline remains a key focus. In this regard, positive news on the development of panitumumab (ABX-EGF), a fully human monoclonal antibody targeted to the EGF receptor for cancer, will be critical for significant appreciation of the stock. The company in conjunction with partner, Amgen, is conducting a number of clinical studies of panitumumab in three cancer indications: colorectal cancer, renal cell carcinoma, and non- small cell lung cancer. The pivotal Phase III program for panitumumab is focused on the development of panitumumab as a monotherapy in refractory colorectal cancer patients. The company and its partner, Amgen, continue to expect to initiate a BLA filing by year-end and complete this filing by Q1 2006. We continue to believe the market for EGFR inhibitors in cancer is sizable enough to support multiple players.
ANALYST CERTIFICATION APPENDIX A-1
I, Elise Wang, research analyst ... |