Don’t blame Soriot, he’s just doing his job
The market might have shrugged off Pascal Soriot’s appraisal of AstraZeneca’s long-term prospects as a pipe dream, but it should not lose sight of the chief executive’s most important goal: to extract as high a price from Pfizer as possible.
For his peak sales assumptions of many of Astra’s key assets to be way above sellside consensus, as EvaluatePharma data bear out, is to be expected, though a couple of key assets are being downplayed (see table below). The main point of separation from analysts is in Astra’s highlighting of non-risk-adjusted revenues.
Then again, Mr Soriot is just doing his job in trying to present Astra as a company that might one day be worth a lot, but one that today Pfizer is trying to pick up on the cheap. Using sales without risk adjustment is merely an attempt to illustrate this.
On the upside, the biggest outliers in Astra’s presentation yesterday are the anti-PD-L1 oncology project MEDI4736, with a massive $6bn in internal peak sales forecast, the COPD combination PT003 ($4bn) and the anti-Alzheimer’s BACE1 inhibitor AZD3293 ($5bn).
| Non-risk-adjusted peak sales ($m) | ||||
| Project | Pharma class | Status | EvaluatePharma consensus | AstraZeneca forecast |
| MEDI4736 | Anti-PD-L1 MAb | Phase II | 2,000 | 6,500 |
| AZD9291 | EGFR inhibitor | Phase II | 1,500 | 3,000 |
| Tremelimumab | Anti-CTLA4 MAb | Phase II | 1,500 | Not disclosed |
| MEDI-6469 | Anti-OX40 MAb | Phase I | 1,500 | Not disclosed |
| Olaparib | PARP inhibitor | Filed | 1,092 | 2,000 |
| Lesinurad | URAT1 inhibitor | Phase III | 1,064 | Not disclosed |
| Brodalumab | Anti-IL-17 MAb | Phase III | 1,009 | Not disclosed |
| Benralizumab | Anti-IL-5 MAb | Phase III | 1,000 | 2,000 |
| Cediranib (Recentin) | VEGFr TK inhibitor | Phase III | 1,000 | Not disclosed |
| MEDI-565 | Anti-CEA BiTE MAb | Phase I | 1,000 | Not disclosed |
| Mavrilimumab | Anti-GM-CSFR MAb | Phase II | 917 | Not disclosed |
| PT003 (formoterol + glycopyrrolate) | LABA & LAMA | Phase III | 742 | 4,000 |
| Roxadustat | HIF-PH inhibitor | Phase III | 700 | Not disclosed |
| Selumetinib | MEK inhibitor | Phase III | 675 | Not disclosed |
| MEDI-545 (sifalimumab) | Anti-interferon alpha Mab | Phase II | 625 | 500 |
| MEDI-546 (anifrolumab) | Anti-interferon alpha Mab | Phase II | 475 | 500 |
| AZD3293 | BACE1 inhibitor | Phase I | None | 5,000 |
| Saxagliptin + dapagliflozin | DPP-IV inh & SGLT2 inh | Phase III | None | 3,000 |
| Total AstraZeneca R&D portfolio | 38,383 | 63,000 | ||
In fairness to Astra, the group did say that internally it has applied only a 9% success probability to AZD3293; nontheless the fact is that its unadjusted $5bn number is now out there. Moreover, it admits that peak unadjusted sellside consensus is only $0.5-3bn.
Meanwhile, PT003, the star asset of the group’s acquisition of Pearl Therapeutics, is perhaps not fully factored in by the sellside given limited information about Pearl, while the saxagliptin-dapagliflozin combination is not in most sellside models, which Astra admits.
On the downside, it might have come as a surprise to many that Astra did not even mention its own expectations for tremelimumab – reckoned by the sellside to be its third-most valuable R&D asset – or MEDI-6469. It is likely that some of these are captured in the huge number for MEDI4736, which includes combinations.
Looking at the group’s forecasts for established brands reveals another surprise – the expectations for Brilinta sales to reach $3.5bn by 2023. Such a leap from sellside consensus of just $1.6bn in 2020 stretches credibility. The huge disparity in diabetes franchise expectations likely results from as-yet unmodified sellside models following Astra’s buyout of the Bristol-Myers Squibb venture.
There is also much variability within the sellside; EvaluatePharma consensus for 2020 sales of $25.2bn, for instance, is the mean of between $15.0bn and $38.8bn – numbers that reflect only projects that appear in analyst models. Overall, however, Astra’s estimate of risk-unadjusted peak pipeline sales at $63bn is wildly at odds with EvaluatePharma’s sellside consensus of $38.4bn.
| Risk-adjusted sales ($m) | ||
| 2020e (EvaluatePharma) | 2023e (AstraZeneca) | |
| Brilinta | 1,556 | 3,500 |
| Total diabetes | 3,921 | 8,000 |
| Respiratory, incl R&D | 8,015 | 8,000 |
| Total company | 25,229 | 45,000 |
Moreover, the group’s internal expectations are somewhat opaque – Mr Soriot has not, for instance, specified what type of risk is or is not being accounted for. Analysts at Bryan Garnier nailed the point when they asked today: “Is it reasonable to expect AstraZeneca to take 20% share of the immune-oncology [market] in 2023?”
The other unknown in all this is the stance of Astra’s shareholders should Pfizer take the bid hostile – an eventuality that at present is purely theoretical. Mr Soriot’s attempts to play up his company’s long-term prospects will only go so far before they reach a level that a majority of Astra’s investors would judge to be unrealistic.
After all, during the biopharma bull market Astra stock largely treaded water, buoyed by the group’s dividend policy, but right now, thanks to the Pfizer approach, the shares are near an all-time high. At some point long-suffering investors will realise that they might never see such an opportunity again, and will want to take the money and run.
They must be particularly wary of Mr Soriot overplaying his hand. For his part, the chief executive will need no reminder that his primary fiduciary duty is to generate the greatest possible return for Astra’s investors.
So far he has done precisely that.
To contact the writers of this story email Jacob Plieth or Amy Brown in London at [email protected] or follow @JacobEPVantage or @AmyEPVantage on Twitter
