The purpose of this article is to discuss a method of forecasting profitability in the biopharmaceutical industry where profits are rare. To identify the stocks with earnings-driven price appreciation, I screened for biotechnology stocks not based in China with consistent 3- and 5-year growth in earnings and with positive profit margins for the trailing twelve months. Of 273 companies, only seven biopharmaceuticals met those criteria.
These seven profitable biopharmaceuticals are ranked below in order of annualized 5-year earnings growth from highest to lowest:
Also meeting the screening criteria, but not included because they do not develop drugs, were PDL Biopharma, Tekmira, and Myriad Genetics.
The next step in the process was to attempt to identify characteristics that these successful companies had five years ago when they were less profitable. I reviewed Form 10-Ks that covered the year 2008 for each company. I did not read footnotes or look in obscure locations for dirty secrets since this was about success, not failure. I read the business overview to get a sense for the focus and capabilities of each company, what drugs they had on the market and what drugs were in the pipeline in 2008. I focused on top and bottom lines of financial statements, on cash in relation to operating expenses, on R&D and PP&E in relation to revenues and total assets, and I attempted to get a sense of the management's outlook.
Key findings related to 5-year earnings growth.
Market presence & source of revenue - all seven of these companies had approved products on the market and were generating revenues at least in part from product sales.
UTHR in 2008 had one non-biological drug in two formulations: intravenous and subcutaneous treprostinil for pulmonary hypertension, and a monitoring device, both of which were approved and commercial. Sales of Remodulin® treprostinil worldwide in 2008 were $297M or 96% of total revenues based in part on proof of improved survival in patients with the potentially lethal disease, pulmonary hypertension. UTHR bought most of its patent portfolio for treating pulmonary hypertension from Glaxo, Pfizer and Eli Lilly.
CELG in 2008 had five non-biological drugs on the market: REVLIMID® lenalidomide, an immunomodulatory drug used for myeloma, THALOMID® thalidomide also for myeloma (patents expiring 2018-2020), VIDAZA® azacitidine, an orphan drug when used for acute myeloid leukemia and also used for myelodysplastic syndromes, ALKERAN® melphalan for myeloma and ovarian carcinoma, and FOCALIN® methylphenidate for ADHD. Drug sales in 2008 accounted for 95% of $2.3B in revenue. CELG had 175 patents issued as of 2008.
REGN received in February 2008 FDA approval to market its first commercial product, ARCALYST rilonacept, a monoclonal antibody and interleukin-1β trap that became the first approved therapy for CAPS, a group of rare, hereditary inflammatory conditions. ARCALYST sales accounted for $6M or 3% of total revenue in 2008.
BIIB in 2008 already had approval to market AVONEX® interferon beta-1a for multiple sclerosis, RITUXAN® rituximab for lymphoma and rheumatoid arthritis, TYSABRI® natalizumab for multiple sclerosis and Crohn's disease, and FUMADERM® dimethylfumarate for psoriasis. Product sales accounted for 62% of $4B in revenues for 2008. BIIB was also receiving revenue from royalties on Angiomax® bivalirudin for anticoagulation.
ACOR in 2008 derived all of its $48M in revenue from sales of Zanaflex® tizanidine, a non-biological drug used to reduce muscle spasticity.
AMGN in 2008 derived 98% of $15B in revenue from sales of Aranesp® darbepoetin alfa, EPOGEN® epoetin alfa, Neulasta® pegfilgrastim, EUPOGEN® filgrastim all for the stimulation of cells in the bone marrow, and with patents expiring 2012-2015; and Enbrel® etanercept for the treatment of inflammatory disorders such as rheumatoid arthritis. This patent expired in 2012.
GILD had by 2008 won US and other countries' drug agencies approvals in each of 11 consecutive years: Truvada® (emtricitabine and tenofovir), Viread® tenofovir, Emtriva® emtricitabine, Atripla® (combination efavirenz/emtricitabine/tenofovir) all for HIV infection; Hepsera® adefovir for chronic hepatitis B; AmBisome® amphotericin B liposome; Letairis® ambrisentan and Flolan® epoprostenol for pulmonary arterial hypertension; and Vistide® cidofovir for cytomegalovirus retinitis. Product sales accounted for 95% of $5.3B in revenue for 2008. GILD was also receiving royalties from Roche for Tamiflu®. US patent were to expire for one in 2010, one in 2014, one in 2015, two in 2016, two in 2017, and three in 2021.
Five-year market cap appreciation, as shown below, was not as good for companies generating most revenues from sales (UTHR, CELG, ACOR, AMGN, and GILD) as it was for those that were generating most revenues from other sources like royalties: REGN and BIIB.
Pipeline - all seven biopharmaceuticals had multiple drug pipelines in 2008:
UTHR, which had the highest 5-year earnings growth at 67% per annum, had only three drugs in clinical trials, one of which had already been approved in a different formulation. UTHR had proposals but no patients enrolled in advanced clinical trials of any biologicals in 2008.
CELG in 2008 had one drug for one cancer in Phase 3, three drugs for five indications in Phase 2, and two drugs for two indications in Phase 1. CELG's expertise is in small molecules, not biologicals.
REGN had nine Phase 3 trials on rilonacept and aflibercept. REGN partnered with Sanofi for use of aflibercept for cancer and with Bayer for its use in patients with age-related macular degeneration and diabetic macular edema, going up against Genentech's Lucentis®. Three drugs were preclinical: REGN88 against the interleukin-6 receptor, REGN421 an anti-angiogenesis antibody, and REGN475 against nerve growth factor for the treatment of pain.
BIIB in 2008 had 22 pipeline products in Phase 2 trials or beyond, almost all of which were biologicals.
ACOR had by far the weakest pipeline with only three biologicals, all in the preclinical phase of development.
AMGN's pipeline in 2008 was the deepest with five drugs for ten indications in Phase 3, thirteen for twelve indications in Phase 2, and thirteen for thirteen indications in Phase 1. Most of these are biologicals.
GILD had one new drug application for inhaled aztreonam for treatment of bacterial infection in patients with cystic fibrosis; Phase 3 drugs included one for HIV, one for hypertension, one for pulmonary fibrosis; Phase 2 drugs included one for HIV, one for bacterial infections, one for liver cirrhosis; there were three drugs in Phase 1.
Looking at market cap appreciation subsequent to 2008, I did not see an obvious correlation with depth of the pipeline, AMGN having the lowest 5-year market cap growth while having a very deep pipeline.
Financial strength - except for AMGN, each company was enjoying double and triple-digit growth in revenue. REGN and ACOR reported net losses for the year; UTHR and CELG also reported net losses due to acquisition costs in 2008. GILD and AMGN reported net profits for 2008 as did BIIB for the 5th straight year. AMGN had by far the most cash ($9.5B) followed by CELG ($2.2B), GILD ($1.8B), BIIB ($1.3B), and REGN ($0.5B); thus there was actually a weak inverse correlation (R-squared = 0.19) between the cash position and subsequent 5-year EPS growth. However, BIIB, AMGN, GILD were all feeling confident enough of their financial outlook to reduce their share counts. Following its IPO in 2006, ACOR nearly doubled its share count by late 2008. CELG and REGN share count increased about 2% in 2008 versus the prior year. UTHR share count was stable.
It appears to me that market cap appreciation, as discussed below, is driven more by revenue growth than by any other financial metric in the Biotech Industry. Also, to reiterate, revenue derived from fees and royalties is a better driver of stock performance than revenues from product sales.
Research & Development - the percent of revenue spent by UTHR on R&D was 85%; however, 62% of the R&D was due to $150M in upfront fees paid to Lilly in December 2008 for licensing, manufacturing and supply agreements.
CELG reported organic R&D of $931M, which was 43% of revenue, but $303M of that was royalty payment to Pfizer related to unapproved forms of VIDAZA.
REGN spending of $278M on R&D was 116% of total revenue, and is divided into six categories and projects on pages 40-41 of Form 10-K.
BIIB spent over $1B on R&D in 2008 which was 26% of revenue.
ACOR provided line items for its $36M of 2008 R&D expenses which was 77% of revenues. $8M was for regulatory affairs, $9M for outsourced R&D, $0.25M was for issuance of stock options to non-employees.
AMGN spent 15% of revenue on R&D.
GILD spent 14% of revenue on R&D.
REGN is the standout in this category. However, because I observed so much variability in how R&D was broken down, and the different types of expenses (even some marketing perhaps) that were included, I am personally reluctant to attach much significance to R&D as a simple line item. Doctoral headcount, though not always reported in Form 10-Ks, might be a better predictor than R&D budget.
Manufacturing capability - UTHR had the 2nd smallest headcount of the 7 profitable companies I reviewed with 360 employees. It owned offices and facilities for the synthesis of treprostinil and monoclonal antibodies. Property, plant and equipment was valued at $221M (25% of total assets) and was up more than triple over the year due to the construction of a manufacturing facility for oral treprostinil in Research Triangle. However, as of end 2008, Baxter was formulating the active ingredient used in the manufacturing of Remodulin. Eli Lilly was paid to manufacture tadalafil.
CELG had a headcount of 2,441 full-time employees over half of which were engaged primarily in R&D. Four acquisitions gave CELG its own manufacturing capabilities: $249M or 6% of its total assets tied up in PP&E.
REGN was already in 2008 an asset-rich company with manufacturing capability, and 919 total full-time employees including 85 doctoral level scientists. Net PP&E was $88M or 13% of total assets. Spending on PP&E was $34.9M in 2008 at which time there were plans to spend $50-60M more in 2009.
BIIB in 2008 had half a million square feet of space including a biologics manufacturing facility, a large scale manufacturing plant, a large-scale purification facility, and a laboratory office building. Net PP&E was $1.59B which was 19% of total assets. BIIB had the largest headcount at 4,700.
ACOR in 2008 carried only $2.3M of PP&E on its balance sheet which represented less than 1% of its total assets. ACOR did not have the capability to manufacture its own drugs. In its Form 10-K, ACOR boasted it had a 61-person professional sales force and did not give a headcount of doctoral scientists.
AMGN in 2008 had ample capability to do its own manufacturing with $5.88B of PP&E which was 16% of total assets. I could not find the number of employees in the 2008 Form 10-K.
GILD in 2008 had 3,441 full-time employees, and had acquired pharmaceutical companies NeXstar in 1999, Triangle in 2003, and Corus in 2006; Raylo Chemicals and Myogen in 2006. PP&E of $529M accounted for 8% of total assets on its balance sheet. As of 2008, GILD had also licensed an India-based pharmaceutical to manufacture and distribute generic versions in the developing world, and had expanded its HIV Drug Access Program to 97 countries to cover 70% of the world's affected population.
Every one of these companies except ACOR had its own manufacturing capabilities. However, PP&E as a percentage of total assets was not a powerful predictor of subsequent 5-year market cap appreciation as illustrated by AMGN which had the lowest 5-year market cap appreciation despite the large value of its hard assets.
Effect of earnings growth on market cap appreciation. Between late 2008 and late 2013, there was a very weak correlation between EPS growth and growth in market cap or total return: R-squared = 0.04, slope = +0.23. The list of biopharmaceutical EPS growers are again ranked below in descending order of earnings growth, but percentages in the right column are 5-year total growth in market cap:
The 5-year total market cap growth of 1,870% (72% annualized return) for REGN was disproportionately high compared to its annualized EPS growth of 34%. BIIB market cap and annualized return (42%) also outperformed its EPS growth (24%), while ACOR underperformed relative to its EPS growth as did the two biggest EPS growers, CELG, and UTHR.
REGN as top market performer, crushing its peers since 2008, is worthy of comment. REGN was in 2008 an asset-rich company with manufacturing capability and 85 doctoral level scientists. This company, though it had only launched one drug, had a deep pipeline and the brainpower, the technical know-how and tools for genetically engineering its own hard-to-duplicate, large molecule, monoclonal antibodies. REGN doubled revenues from 2006 to 2007 then again from 2007 to 2008, but mostly from royalties and contracts rather than product sales. In contrast, UTHR and CELG grew revenues from sales, yet the market put a lower premium on those revenues compared to REGN's. I think that REGN's capabilities were viewed as a huge barrier to entry, and investors placed a high premium on that.
AMGN as worst market performer is also worthy of comment. This was also an asset-rich company with 16% of assets in PP&E. I cannot find a headcount of doctoral scientists in its 2008 Form 10-K, but AMGN had the largest - with the possible exception of BIIB - pipeline of new drugs. AMGN had 5 large-molecule drugs already on the market. However, revenue growth in 2008 was the lowest of the seven companies. Management's outlook was very downbeat in its 2008 annual report, the most pessimistic of the cohort. This would have put me off. I am not sure what kept investors away between late 2008 and late '12; but since then the threat of generics, the so-called biosimilars has emerged and has been discussed in a June 6, 2013 Seeking Alpha article; and on September 5, 2012 in another.
Implications for investing now. In other sectors and industries, stock prices are driven by earnings. The high stock prices in relation to earnings for biopharmaceuticals (aggregate P:E TTM = 427) tell me that biotech is an altogether different world of investing where price is determined by revenue growth, perceived technological moat, and perceived future success of drugs in the pipeline, and not by profits from drugs already on the market. Here are my recommendations for biopharmaceutical investing based on this exercise:
- Invest in companies that have at least one drug already on the market. I personally want drugs that change clinical outcomes and not just symptoms.
- Look for growing revenues and for companies with their own manufacturing capabilities. Stocks of asset heavy companies tended to outperform; AMGN was an exception.
- Pipeline matters, but since everyone seems obsessed with that, market cap gets pushed to high double-digit multiples of revenue in companies with deep pipelines. On the other hand, if the pipeline looks very weak as it did for ACOR in 2008, I would stay away. Advanced clinical trials (Phase 3) count more than less advanced trials. I do not consider preclinical trials investable if a company does not have a track record of getting the drugs to Phase 3.
- R&D budget also matters: the paragraph above shows that the percentage of revenue spent on R&D was directly related to subsequent 5-year earnings growth among these seven; however, the median value of about 22% was low compared to the current median of about 80% for the industry. My impression from reading Form 10-Ks is that the importance of R&D expenditures in stock selection can be overrated if you don't look at the details.
- Although not the primary driver of market cap appreciation, I prefer to invest in companies that have the financial ability to take care of their shareholders by reducing or at least keeping stable their share count. Only ACOR failed to meet that criterion in 2008 and has steadily increased share count since then. REGN has increased share count by 24% in the past 5 years, but the CEO owns over half the common shares if restricted shares are included in the count. AMGN has reduced its share count by 26% since 2008 and even pays a small dividend.
Now we should apply what we have learned about what these companies had going for them in 2008 to find the companies that are now poised for similar growth. I have screened and ranked biotech companies for 1-, 3-, and 5-year revenue growth. The top-10 ranked biopharmaceuticals with at least one product on the market included Incyte Corp. (INCY), Halozyme Therapeutics (HALO), REGN, and Seattle Genetics (SGEN). I recommend looking at the pipelines, at a detailed description of the R&D, collaborators, and manufacturing capabilities of these companies or others farther down on your stock screener. Look for webcasts of presentations at Healthcare conferences - this might get technical, but even an investor without a science or medical background can get a feel for the depth of the science going into a company's pipeline.