This is from a company that has not been too positive on Onyx.
Bernstein "CELG: Pom Ph III, MM015 Support Near-Continuous IMID in MM; Approval, Prem Price Likely, Kyprolis Will Slow. $96 target"
CELG: Pom Ph III, MM015 Support Near-Continuous IMID in MM; Approval, Prem Price Likely, Kyprolis Will Slow
Celgene – Target $96 – Overweight
PYCC – Market Weight
ONXX – Market Weight
Celgene's eagerly awaited Pomalidomide (POM) phase III trial was presented as the final late-breaking abstract in the final session of the final day of this year's American Society of Hematology meeting in Atlanta yesterday. Attendees were rewarded with an impressive improvement in the progression free survival and overall survival from the addition of POM to standard low dose dexamethasone; these results are likely to lead to rapid approval and adoption, at least in this population of refractory myeloma patients; short term it will also cause some softness in the otherwise robust early adoption of Onyx's Kyprolis for the same indication. We will be conducting a conference call for investors with two leading Myeloma investors to discuss these and other results presented at ASH today, Wednesday December 12, at 11am.
· So What Exactly was the Efficacy? Impressive Confirmation of Phase II. MM003 is a 455 patient phase III trial in refractory heavily pre treated multiple myeloma patients. In the active (POM + Low dose dex) arm the Progression Free Survival (PFS) was 3.6 months compared to 1.8 months in the Dex only arm; the overall survival (OS) was not reached but is likely to exceed 11 months compared to the control arm overall survival of 7.8 months (Exhibit 1, Exhibit 2). The reduction in the risk of progression was 52% (HR = 0.48) and in risk of death was 47% (HR = 0.53 ) and the p value for both PFS and OS benefit were highly significant. Even though these numbers all seem relatively small, the magnitude of the benefit and the size of the signal are likely to be impressive to physicians, particularly since some 30% of patients in the control arm subsequently crossed over to POM after they failed, thus reducing the apparent survival difference between the two arms.
· Noteworthy (and reassuring) that Efficacy was Maintained in Revlimid Refractory Patients. To Celgene's credit, this trial had strict criteria for entry requiring significant prior exposure to both the standard early line drugs in MM, Revlimid and Velcade, In fact ~72% of patients in the trial met a strict definition for Revlimid and Velcade refractory (ie failed on or immediately after treatment) (Exhibit 2). Whether patients were refractory or just relapsed, and whether Velcade refractory, or Revlimid refractory or refractory to both, they all showed highly statistically significant improvements in PFS and OS from POM. This will certainly reassure investigators, physicians, patients and investors that POM can be given to refractory patients regardless of their prior treatment history. It also sets the product up well for combination use with virtually any other agent physicians would like to offer to their refractory patients (information about those combinations started to emerge at this meeting and is likely to feature in future myeloma sessions and conferences).
· How About the Safety? Another Clean iMID. Generally iMID's have been relatively well tolerated (as far as cancer treatments go and notwithstanding their teratogenicity risks). Each of the three generations of compounds have had their idiosyncratic risks, and they have become progressively less troubling as the potency has increased and dose decreased. It is hard to declare yet that POM is the safest iMID yet, given the highly pre-treated patients that it has been used in so far, but it seems reasonable to declare it as safe or safer than Revlimid. The rate of neutropenia was increased in the combination arm to 42% of patients, compared to 15% in the HiDex arm, and 7% of patients had Gr ¾ neutropenia compared to 0 in the HiDex arm. Surprisingly, however, there was no increase in low platelets or anemia in the combination arm, which is a potential advantage over Revlimid. The rates of infections were similar between the two arms, as were the rates of neuropathy, thromboembolism and discontinuation due to AE's. There were no SPM's reported in the trial.
· How Long Were Patients Treated? 7-8 Months Seems Right. One of the critical questions about any new treatment in patients with late stage disease is the likely duration of therapy. We have previously suggested that the initial duration of therapy for new medicines in this setting in the real world could be short; this is likely to also be the case initially for POM . In the phase III trial patients were treated until progression, and the median duration of response was 8.5 months; this suggests that with the PFS of around 3.6 months, there were some patients who did not respond at all, or failed within a month or two, and then responders who benefited from the drug for 8-9 months. This is comparable to the duration of response in the Kyprolis phase II trial of 7.8 months. In our forecast for POM we modeled initial duration of therapy of 7 months, increasing as the drug is used more in combinations.
· How Does this Compare to Kyprolis and Other Alternatives? Very Similar. It is tempting to compare these results to both the previously disclosed and published pom phase II open label results (RR 34%., PFS 4.6 months, OS 16.5 months) and the equivalent results for Kyprolis (23%., duration of response 7.8 months, OS 15.4 months) in similar patients (Exhibits 3, 4). This comparison is a little unfair since there is really no comparable phase III trial result for Kyprolis at this stage. But the response rate and PFS in this trial compare favorably with both the prior phase II open label trial for pomalidomide and the Kyprolis trial; the OS in the two phase II trials is now relatively mature, and it is likely that patients in this phase III trial could approach the 15-16 month OS when the active arm OS results are mature. It is important to note that the prior phase II trial comparing pom monotherapy with pom plus dex showed that pom monotherapy had only modest responses; the Kyprolis study responses did not include a dexamethasone "boost"; in real life most physicians are now using Kyprolis with dex or prednisone and responses should be significantly better than seen in the open label trial.
· So When Should Patients Get iMIDs and for How Long? In addition to these heralded phase III results for POM, investigators also presented updated results and analyses from the MM015 trial of front line Revlimid and Revlimid maintenance. In one of the oral presentations, it was surprising to see patients who had failed front line therapy in the MM015 active arms, then be re-treated with Revlimid, and still show impressive responses. Specifically, of the 81 patients in the MPR-R of MM015 who failed, and then had further treatment, the Median TTP for that second line regimen was 18 months if it contained Revlimid, 14 months if it contained Velcade and 6 months if it contained neither agent. The duration from 2nd line to 3rd line therapy was comparable for patients coming off the MPR-R arm, the MPR arm and the MP arm. In all three arms, the duration of apparent response to second line treatment was significantly longer if the second line regimen contained Revlimid.
· What Comes Next? Go, Go, Go The PDUFA date for POM is February 10, and the product's key data and results are now all in the public domain. The FDA have not announced an advisory committee for the product, and given the data, the precedent set by carfilzomib and the well defined mechanism of action, there does not seem to be any obstacle to approval. Celgene have suggested that they will be prepared to launch the product immediately, and they certainly communicated their urgency about the launch during conversations at ASH.
· Arguments about why POM has risks largely spurious. We have heard a number of flimsy arguments about why the POM approval may have risks. The first is that the remaining "unmet medical need" in refractory myeloma may no longer exist, in the FDA's eyes, after their approval of Kyprolis and therefore the approval could be denied, delayed or restricted. We believe the unmet need still exists, and even though technically POM's efficacy has not been established in Kyprolis-failures, we do not expect the FDA to be deterred by this limitation. At the time POM was studied, and submitted, the unmet need existed, and they established their efficacy in a similar high unmet need population as Kyprolis. A second variation of this argument is that the approval of POM might be more restricted than the approval for Kyprolis, limited, for example to dual refractory patients, rather than just patients previously exposed to Velcade and Revlimid (truly refractory patients are smaller in number and likely to have shorter treatment duration). We believe such a restriction is unlikely, given the breadth of patients studied in the POM phase II trial (and the responses seen in the phase III). Lastly, investigators and clinicians at ASH indicated to us that any such labeling restriction that might emerge from the FDA discussions would have little impact on their use. Clearly the drug will be used widely in relapsed and refractory patients, and it would simply not be feasible for payers or other authorities to require documentation of "refractoriness" over intolerance or relapse.
· POM MM003 trial unlikely to be in the label. One question that has surfaced is whether these impressive MM003 results will be in the initial label for POM. We believe this is very unlikely. Adding this trial to the label would be a significant regulatory amendment for Celgene and the FDA, and one that would certainly delay the approval by 3 or more months given timing. We don't see any sign that Celgene wants to trigger such a delay in the initial approval, and expect this data to be formally submitted to the FDA in 2013. It will certainly be part of the initial submission to the EMEA, and given what we are hearing about timing for the Kyprolis ASPIRE trial (possibly running to late 2013 or early 2014), Celgene are likely to reach the market in Europe a year or more ahead of Onyx.
· What should we expect for pricing? How high is enough? It is well known by now that pomalidomide currently has shorter patent life than Revlimid (although depending on approval date in Europe market exclusivity there could be comparable); Celgene have also focused their development efforts for POM on refractory and relapsed myeloma, and made it clear that POM is a salvage drug only, rather than a replacement for Revlimid. Given those conditions it seems inevitable that POM will be priced at a considerable premium to Revlimid, perhaps as much as 50%. In our current forecasts we assume that Revlimid is priced in the range of $10,500 per patient per month (~30% premium to Revlimid), which is a conservative interpretation of this positioning. Most investors will have similar expectations but it is certainly possible that Celgene take a more aggressive approach; we see limited variability in demand up to around $12-14,000 per patient per month.
· So how will POM launch? We expect Celgene to add significant new promotional resources to support the launch of POM in the US, and would expect those expenses to be included in the company's full year financial guidance in January. Given the independent revenue opportunities for Revlimid and POM it makes sense to us, at least in the initial launch year or two, to have dedicated field support for POM. That support is likely to be an incremental 40-50 salespeople in the US, adding an additional $20-30mm to expenses. With this support, POM should launch solidly, although with a more gradual pattern of adoption than Kyprolis, given the erosion of the accumulated refractory patient pool by the preceding launch. There are likely to be some patients failing Kyprolis by the time POM launches, or who prefer an oral medication, who will discontinue Kyprolis and switch to POM. After that small initial bolus, POM and Kyprolis are likely to initially compete for new starts among dual (iMID and Proteasome inhibitor) experienced patients; over time, as in the case of Revlimid and Velcade, the two drugs will start to be used in combination rather than as substitutes, but that practice is likely to take several years to become widespread.
· We forecast a modest start for POM in 2013, but rapid growth through 2020. In our model we now have POM at a 100% POS, and assume that it launches slowly initially (waiting for reimbursement and competing with Kyprolis) generating revenue of $85mm in 2013. We do forecast rapid growth to $500mm in US sales by 2015, and then to $1.15bn globally by 2016. The recent consensus estimates are uncertain, but appear to be higher than us for 2013, and then lower longer term; we expect them to change in the coming months after this disclosure and following the approval and launch. In our model some 1500 patients are treated with POM in 2013, which more than doubles in 2013 and reaches more than 8,000 patients (in the US) and a 7.5 month duration by 2016.
· What is the outlook in Europe? This study was reportedly conducted exactly in line with the demands of regulators in Europe, and certainly meets the standard of a comparison to an existing standard of care. It is very difficult to see this application failing in Europe, although Celgene have certainly stolen defeat from even more obvious regulatory opportunities in the past. Under normal circumstances POM would be expected to be almost equally successful in markets outside the US, compared to the US, but it still faces the roadblock of Revlimid's failure to move "upstream" in those geographies. For this reason we forecast OUS revenue significantly below US sales through 2018.
· What does this means for Kyprolis? We think the simple fact of competition will initially hurt the Kyprolis launch trajectory, and do not share the belief that some investors and investigators have outlined that Kyprolis will simply just continue to grow through the launch of POM. This does not mean we are disinterested in Onyx or that Kyprolis sales will go down; instead reported results in Q1 and Q2 are likely to reflect slowing growth in the face of discontinuations by refractory failing patients after short duration of treatment on Kyprolis, switching to a trial of POM, as well as sharing of new patient starts (which drive sales for a relatively short duration therapy) with POM. Over time we do forecast that the two drugs will ultimately be used together in many patients, and that Kyprolis will eventually migrate to earlier lines of treatment, but near term these two phenomena are likely to be relatively uncommon. For this reason we see sales plateauing for Onyx in Q1 and Q2, at least until their market share of new starts with POM stabilizes.
· What else lies ahead for Celgene? While Celgene's stock has started to act more positively in recent weeks, we still believe the stock has more upside than other large cap biotech stocks. In the coming months the stock is likely to react positively to a timely or early approval of POM and aggressive pricing. The company seems wedded to pre-announcing Q4 financial results and providing guidance in early January, and this could be a near term downside risk – increased spending, and/or cautious initial revenue guidance, could alarm investors temporarily. Beyond the near year, Q4 results and the POM launch, investors will start to speculate with renewed energy about the risks associated with the upcoming MM020 trial analysis for Revlimid. We recently published our analysis of this event, and our view of its high probability of success, and that judgment was reinforced in many conversations with investigators and KOL's this weekend. We continue to rate Celgene's stock Outperform, with a target price of $96.
We rate Celgene Outperform with a target price of $96. Celgene has taken an old and discarded drug (thalidomide) and invented a whole new class of compounds and a novel approach to cancer treatment based on research into thalidomide’s original activity. Although it has taken many years for that strategy to come to fruition, it now appears that Celgene will capture significant incremental use of its first novel compound in this family, Revlimid, in its two existing indications, myeloma and myelodysplastic syndrome, as well as in other diseases such as NHL and CLL and potentially even in solid tumors.
Over the next three years, we expect Celgene to deliver ~22% EPS CAGR. The acquisition of Abraxis boosts the company's long term growth outlook, consolidates its position in oncology and leverages the company's established global development and commercialization infrastructure. The returns on this acquisition will depend on successful development in lung cancer at least, and potentially pancreatic and other indications too, as well as international expansion.
We rate Onyx Pharmaceuticals Market-perform with a target price of $82. With the stock's recent strong performance, it now largely reflects the full value of carfilzomib, at least in its immediate refractory myeloma setting. Onyx transitions now from a royalty and development company to a revenue and commercial one, and for some such companies there are hiccoughs in this transition. Nevertheless we believe carfilzomib has revenue potential in the $1-1.5bn range, and this, along with the company's secure royalty/revenue sharing profit stream from Nexavar/regorafenib, should support a valuation in the $5.5-6 bb range.
Expenses are high at ONXX, and we expect further increases in R&D to fund phase III and phase IV trials of carfilzomib and in SG&A to fund the launch expenses for Kyprolis. Longer term the company will also need to absorb full manufacturing costs for carfilzomib, as well as substantial milestone payments to the original investors in Proteolix. We currently forecast profitability in 2014, and then rapid earnings escalation in later years based on substantial operating leverage and recurring collaborator revenue covering ongoing expenses. We now see ~$9 in earnings being achievable by 2017, with $11-14/share achievable in later years. As the company transitions from development to commercialization, we see an acquisition being increasingly likely, although this is not part of our valuation or launch thesis.
0users liked this posts0users disliked this posts
Sentiment: Strong Buy