Which questions should we be asking in preclinical testing?

Immuneed Chief Operating Officer share first-hand experience…

…from taking a drug candidate from preclinical testing to first in human trials – without testing on animals. “It was clear from the outset that safety testing on mice would have no relevance for our drug’s potential to succeed in humans. Nonetheless, many said it would never be approved for first in human trials..

How can investing more time and thought before preclinical testing make your drug’s evaluation more efficient?

The FDA has issued new draft guidelines for safety in evaluating the immunotoxicity of drugs and biologics. Among other things, the FDA now highlights the importance of obtaining pharmacological data using human cells before submitting immunostimulatory candidates. This emphasis on human-based preclinical methods stems clearly from the TGN 1412 incident,[1] especially as it’s mentioned in conjunction with cytokine release syndrome. The shift is important, if only a partial step. The guidance is still general, and it overlooks some systemic risk factors. What about platelets? What about the complement system? Those messy little human reactions make a big difference. The fact that they aren’t noted raises some larger questions: what are guidelines for, and which questions should we be asking in preclinical testing?

Guidelines, not checklists

As the name suggests, guidelines are guidance, not checklists. It isn’t the authorities who write them that carry the responsibility. It’s the sponsors. So instead of asking which tests will meet the guidelines, one should ask, “Which tests are relevant for my drug candidate?” It’s not as obvious as it sounds, which has always surprised me – even when I was doing my first research as part of the development of an adenovirus drug. Published research, which had been done by others with similar drug candidates, made clear that our drug type would produce fatal liver toxicity in mice, because it wouldn’t bind to mouse blood cells. However,  it would bind to human red blood cells, thus producing a totally different distribution. In other words, it was clear from the outset that safety testing on mice would have no relevance for our drug’s potential to succeed in humans. Nonetheless, many said it would never be approved for first in human (FIH) trials if we didn’t test on mice because animal testing was ‘expected’. (No, we didn’t do safety testing on mice. Yes, our drug was approved for FIH without it.)

Relevance should determine the method

The issue isn’t whether animal models are good or bad, but whether individual assays are relevant for your drug. Are you developing your drug to cross things off on paper, or to achieve a therapeutic effect in the human body? If it’s the latter, you need assays that identify possible effects and issues, so that you truly know your drug’s mode of action and safety profile. It’s neither about animal vs. human nor about simple vs. complex. In the infamous TGN 1412 example, promising preclinical testing had been done on macaques. While primates are extremely similar to humans, the expression of the surface receptor differed between species. This particular molecule was down-regulated by the macaque cells, but not by human cells. Yet signs of activation had shown up in studies on rodents, which can be deemed far simpler and less related to humans [2]. Although they were relevant, the rodents were ignored in favor of the ‘obvious’ primate analogy. Given the costs of primate testing, using a human-based assay from the beginning would have yielded a more relevant answer for the money. But the rodent study held the key already – because the bottom line is relevance, not complexity.

Where does the most value lie?

The TGN 1412 example is telling, because many sponsors don’t bother confirming a negative. If an issue appears in one test they’ll often try another, but seldom do they confirm a problem-free result. Maybe this is due to expense, but what is the cost for extra preclinical testing in comparison to clinical trial costs? Knowing for certain that your drug works as you think it does, and that it doesn’t risk causing people harm, should have a value far greater than the preclinical cost.

Fast may not mean first

We all want a success story, just as we all want to move quickly and make money from our innovation. But taking time to ask the right questions at the preclinical stage can actually speed up the whole process. Even if your competitors have a head start, you can win the race if you have a better understanding of how your drug works – and a well-prepared mitigation plan for the clinical development. When competitors get results they don’t understand, where the control performs better than the drug itself, you will have real information to act on. Moreover, you’ll be able to defend your results to your financial backers.

The questions that matter

So, if asking which tests are expected is the wrong question, what are the right questions? Here’s my shortlist:

1. What information do I need about my drug, and which testing methods can reasonably be expected to provide it?
2. How can I diversify my testing, using the right mix of models and assays to provide a comprehensive and reliable picture?
3. Are the negative results I see true negatives, or could I have missed something in my choice of assay?

With those three questions in mind, you can go further and faster in both preclinical work and the clinical work that follows. We at Immuneed are happy to help you move forward. We can assist you not only with testing, but also in determining if our platform and assays are right for your drug candidate. With any service provider, that should always be the first step. Simply get in touch.

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2964774/

[2] https://pubmed.ncbi.nlm.nih.gov/27191544/