On Predictions and Probabilities: The Year Ahead
Pharma’s Cutting Edge Vol. 3 Number 1 - January 2005
On Predictions and Probabilities: The Year Ahead
Regular readers will know that I am not generally drawn towards prediction-making. There seem so many unpredictable variables, and no matter how well-reasoned, most predictions end up sounding like amateurish twaddle. I much prefer presenting detailed facts and expert opinions, estimating outcome probabilities and delineating major risk factors. This is generally how therapeutic program and company valuations are derived (i.e. discounted future value method).
We first gather facts, including stage of programs, regulatory guidelines and precedents, intellectual property protections, etc. We then add to those facts expert opinions that help put the facts into context (e.g., the quality of a drug program with reference to regulatory and marketing needs). Armed with these facts and opinions, we create parameter assumptions and deduce probabilities of success (by drug, by stage of development, by time period, etc.). These assumptions and probabilities are then used to adjust our expectations of future value. After applying discount factors, future values translate into present values. For readers who don’t regularly perform program or company valuations, it probably sounds easy, but it’s difficult to do well, and it’s prone to tremendous uncertainty under the best of circumstances.
For an example of what the end result looks like, check out a small-company valuation, we’ve posted on the crownstonegroup.com website. Play around with the assumptions, and you will get a sense of the variability inherent in such valuations.Which brings me back to making predictions. Predictions are also the product of deductions and inferences based on fact and expert opinions. So, why does a valuation exercise employing many assumptions and success probabilities strike us as an acceptable means of comparing investment opportunities, while predictions seem distasteful…more guilty pleasure than valid planning tool?
Personally, I think it all boils down to semantics; the word prediction connotes an ill-reasoned excuse to pontificate. Predictions conjure up the visceral responses we have to pretense and carelessness. But probabilities and risk factors usually connote thoughtful, well-reasoned deductions and inferences.
It’s like the difference between the industry executive who makes a decision after a dozen committee meetings using his expertly honed intuition, and the industry executive who makes a decision after sorting and digesting pertinent facts and opinions from a few trusted sources using her expertly honed intuition. The former strikes observers as a careful decision-maker, and the latter looks like a renegade.So, what are my probable outcomes and risk factors—my predictions—for the coming year? In no particular order, here are the biggies:
1. CMS and FDA band together to create a program for new drug and device evaluations that employs cost and relative value considerations for the first time in U.S. regulatory history. I doubt cost and value will ever be important deciding factors for marketing approval, but they will be scrutinized as never before. Any government scheme will have a bearing not only on the many government-run and -coordinated health plans but also on the entire private sector as well. My only reluctance in making this call is the timing; I have no doubt about the inevitability of a new drug evaluation system that includes features similar to those of UK’s NICE and Australia’s PBAC. But will it come in 2005? I think it will be gradually phased in beginning late this year.
2. FDA reconfigures its organization to provide more autonomy for the drug safety division. The newly autonomous division, in conjunction with CMS, proposes a standing post-marketing surveillance system that relies on prospective surveillance of Medicare drug beneficiaries to—again for the first time in U.S. history—track incidence rates and case mortality rates for all adverse events reported by seniors enrolled in Medicare Part D. Similar surveillance programs, coordinated by FDA will be initiated within the private sector, creating the world’s most sophisticated post-marketing surveillance system by 2007.
3. Industry-wide liability protection for drug and device makers will not be enacted into law in the U.S. this year. However, lawmakers are likely to expand current incentives for industry to develop ways of countering biological and chemical threats and improving the supply of prophylactic vaccines. Particularly for the latter, these incentives will include some liability shields.
4. Follow-on protein therapeutics. I think this will be the compromised name for what the generics industry would prefer to be called “biogenerics” and what innovators would like to call “not in my lifetime.” The first U.S. approval will happen in 2005, although it will be for a follow-protein whose innovator cousin was approved by CDER under regulation 505(b) governing approval of drugs via the NDA route. This is because the 505(b)(2) regulation in existence allows for such approval. The only thing holding back the first 505(b)(2) follow-on protein approval—likely to be Sandoz’ growth hormone—is FDA publication of scientific guidance specific for proteins and evidence that the legal challenges to 505(b)(2) by Pfizer et al will be futile. I anticipate both of these hurdles to be cleared by year’s end. But approval does not equal launch, and I anticipate the first launch to be delayed—because of litigation—until late 2006 or 2007. Conversely, approval of the first follow-on protein outside of 505(b)(2) won’t happen until late 2006 at the earliest.
5. Investing in the healthcare sector will again be a mixed bag for most of 2005, with most drug makers and emerging drug companies underperforming their reference market-cap indices. However, device makers and diversified manufacturers, like J&J, should continue to out-perform as investors take note of the increasing willingness of U.S. private and public sector third-party payers to cover an increasing variety of “quality-of-life” products. The situation for non-diversified companies will improve towards year-end, however, as the environment for drug reimbursement in 2006 clarifies, and lowered expectations for sector growth become widely adopted. This turnaround should mark the beginning of a fairly robust period of market value appreciation for much of the drug sector, as it climbs closer towards its long-term valuation averages. Emerging companies should also pick up towards year-end.
6. There will continue to be a reasonably healthy market for healthcare IPOs in 2005, although it is likely to cool off a bit from 2004. As in 2004, companies will find themselves relying on “fast follow-on” secondary equity and debt offerings to fully fund their clinical programs and launches, creating turbulence for investors with short-term horizons. On the bright side, 2004’s IPO class had an overall positive return (albeit with very high volatility), and I anticipate the same in 2005. The most successful IPO for 2005? I’ll go out on a limb and guess that Synta will get the $115 M they’re looking for to be the year’s most successful offering.
7. Industry consolidation will continue. We will see more diversification of businesses within major pharma, as some of the less successful non-diversified players seek to emulate J&J’s long-term success. This diversification will include pharma acquisitions of device and diagnostics manufacturers in particular. I also anticipate that at least one major pharma will swallow a big-name biotech; likely candidates to be consumed include Chiron and Millennium.
