Say you’re a healthcare entrepreneur. And say you meet a brilliant immunologist at the American Society of Clinical Oncology conference, and she has very exciting data demonstrating full tumor regression in a genetically engineered mouse model of human breast cancer. It’s clear that she has discovered a critical step in cancer development that could be a great new drug target.

Time to license the intellectual property, right? Get this scientist onto the board of your new start-up, then turn her research into a treatment and, of course, profits.

Actually, that’s probably not such a great idea.

Here’s why it’s not the right time to pull the trigger - and why it may never be:

1.     Most early science will not lead to drug candidates that pass feasibility and toxicology requirements to qualify for human trials.  Human biology is complicated, and no one can predict what will or won’t work.

2.     Once you raise money and recruit co-founders, you’re in the business of convincing them that you will be successful. No matter what. In so doing, you incentivize scientists to design experiments that will support your business plan. This corrupts the conduct of the most important testing and encourages your scientists to generate ‘good enough’ data  - good enough to keep the funding rolling in.

3.     Capital that could be going directly into research will instead be spent on the all-too-familiar costs of doing business: salaries, attorneys, accountants, boards of directors, etc.

4.     As all these parties become more invested in the success of certain experiments and approaches, the less willing they become to admit it when things-;even small things--inevitably go wrong. Dismantling a team and a company is difficult, and it isn’t any fun.

5.     Even the best outcome - generation of a compound that achieves Investigational New Drug status- carries a burdensome economic sidecar. To generate attractive returns for your earliest-stage investors, you have to begin recuperating not only your company’s scientific investment, but also the general and administrative costs, which have probably eaten up one-third to one-half of the total capital invested. And the road to a marketable drug has just begun.

6.     In the pursuit of ‘good enough,’ it’s likely that your team has passed over other approaches that may have led to better drugs. Now that capital is gone.

Can you tell that I’ve seen this movie a few times? Here’s when you might consider investing:

1.     When the science has become a platform--a fundamental new approach to generating many new drugs. Now you’re jumping into something that’s more akin to an engineering start-up.

2.     When a bonafide drug molecule (better, several) has been advanced to the human trials stage. This is still a high-risk investment, but at this point your ultimate buyers - the pharmaceutical companies - will be lining up to partner with venture capitalists and entrepreneurs to co-invest. That greatly raises your chances of profitable success.

In medicine, forming a new company around early science can easily produce a  misalignment of incentives and waste precious capital. Early-stage biomedical assets should be managed and tested as a portfolio of ideas with no bets or promises placed on any single one. This will allow the unbiased pursuit of the projects that survive rigorous vetting and testing, all the way through to drug molecules that are ready to be used in the treatment of actual disease.