Message Font: Serif | Sans-Serif

No. of Recommendations: 8
New Formula Presented

In response to the previous suggestions from this thread on the Biotechnology Board I have come up with a new formula. Also, I have compiled the comments and suggestions with rebuttals at the end (by myself and others). Credit for changes and suggestions appear in []. Sorry if I missed anyone. See the next post for the actual numbers.

New formula for the calculation of ClinIndex (CI):

CI = Phase I Candidates * .20 + Phase II Candidates * .286 + Phase III Candidates * .606

Candidates were only counted once even if under testing in multiple trials for multiple candidates. I believe that this concurs with the statistics compiled by Digger 65 and biostatprof. See the following link:

http://boards.fool.com/Message.asp?id=1020007000752017&sort=id

As a second part of the formula I have calculated a value screen that ranks the candidates normalized by market capitalization. This shows how much in market cap one is paying for each drug that is statistically likely to survive clinical testing. I call it the price to clinical index ratio (P/CI). Its similar to the P/E.

P/CI = market capitalization / ClinIndex

Restatement of the Purpose of the Formula

The purpose was to come up with a single number conveys the strength of the company's drug candidate pipeline. This was done by statistically estimating the number of current drug candidates will successfully reach the market.

Keep the following guidelines in mind when using the ClinIndex.

Use the formula as a screen only and follow-up with due diligence. Its purpose is not to make the buy/sell decision for you. It is not like a mechanical investing method such as the Foolish Four or PEG5 which is the beginning and end of the inquiry. It is more properly used to identify companies that stand out from the crowd for further investigation. It is a labor saving device.

It can also be used as a warning sign that a company you wanted to invest in may have a weakness. If the company has a low ClinIndex then the follow up diligence had better reveal that the few candidates are potential blockbusters.

It indicates only the strength of the pipeline and not the strength of the company's intangibles, current drugs already on the market, marketing savvy, other revenue streams, etc.

Modify the numbers to concur with possibility of making it to market. [biostatprof, Digger65]

Done, see above

Eliminate multiple indications for the same candidate. [biostatprof]

Done, see above.

Some drugs may be blockbusters, multiply by expected revenue or otherwise factor in potential market size. [biostatprof, YoHo00, SamIAm99]

I did not do this for the following reasons:

1. Simplicity and ease of understanding. If the calculations become too complex then we lose track of the rationale behind it and the weaknesses. This has been cited as a consideration in the Foolish Four approach. [aps]

2. Uncertainty of market projections, i.e., it gets too speculative. [GPFE]

3. Massive amounts of work for a simple index that is not going to be used as the sole purchasing criteria, i.e., it is not the same as a PEG 5 or other mechanical screen. [ElricSeven]

4. This information can be verified in follow-up due diligence. [aps3]

Include the probability of each company's execution based on former successes. [GPFE]

This has not been done. Mostly rebutted by the above arguments of workload and simplicity. One additional argument is that the population of previous drugs for these companies tends to be very small. Insufficient size of the population leads to erroneous statistics. In lay terms, they chance that they “got lucky” on 2 of 2 candidates masks their true forward looking chance of success. Maybe in ten years we will be able to do this. [ElricSeven]

Genomics companies have faster execution of clinical trials and a greater chance of success in their new paradigm. [tomheadrick]

Damn, I hope so. However, all the companies that we are comparing are genomics companies, so that factor washes out. Unless, perhaps, one company is better than the other at genomics. That is best covered by a follow up with due diligence, where your “gut” instinct plays a greater role. [ElricSeven]

There is no evidence that being a genomics company has an effect on the speed of trials. [FushiTurazu]

Run some type of backtesting. [Digger65]

Backtesting only works on “mechanical methods” like the Foolish Four or PEG 5 that are used as the sole criteria for making a stock purchase. [ElricSeven]

We know that candidates in clinical trials MUST be a component of a pharmaceutical company's success. If the company never has a candidate in clinical trials it will fail. If it has many, then it has a much better chance of success. No one argued with this and its quite intuitive. Therefore, a lower ClinIndex is a negative factor and a high one is a positive factor. Don't use it as the only factor and you should be fine. [ElricSeven]

Finally, some pithy words from biostatprof:

“Quick screens are just as important as longwinded due diligence.”

“[You] can't read every SEC report on every company.”