As has been made all too clear by the Covid-19 pandemic, the ability to rapidly develop drugs and vaccines is crucial to saving lives. In a recent talk, Dr. Toshihiko Nishimura detailed his research that promises to shrink the duration of the trickiest part of the development process: testing in humans.

Dr. Nishimura is Director of the Stanford University Lab for Drug, Device Development and Regulatory Science (SLDDDRS). Speaking at Upgrade 2020, the NTT Research Summit, he explained how his team is working to repair a shortcoming in testing traditionally performed on mice by giving the mice more human-like qualities.

“In drug development, we have three major pains. One, exceedingly long duration, such as 20 years, huge budget, and very low success rate,” Dr. Nishimura said. Generally, the drug development process follows four stages: discovery, preclinical, clinical trials and regulatory. Of the four, the clinical trials is typically the longest, around 10 years on average, historically.

Some 90% of clinically tested drug compounds ultimately fail, he said. The U.S. Food and Drug Administration confirmed that one of the reasons for the low success rate is there’s a large gap between preclinical proof of concepts (POCs) and clinical POCs.

Dr. Nishimura’s team set out to test their hypothesis that the reason for the gap is data from the mice used in pre-clinical stage misrepresents the results of the drug under test. The reason for that is because mice are missing certain genes and enzymes that humans have, most notably affecting the immune system and liver metabolism.

The solution is to “humanize” the mice by introducing the missing genes or enzymes. This, in effect, is the same procedure used in the development of the polio vaccine, led by Dr. Tatsuji Nomura, who Dr. Nishimura said was his mentor. Dr. Nomura pioneered development of validation system using the Tg-PVR mouse, which carried the human poliovirus receptor (PVR) gene. That enabled oral polio vaccines to be tested on mice rather than primates, thus adhering to the Russel and Burch 3R principle.

Over the years, some 100 variants of humanized mice, known as an NOG mice, have been developed. Dr. Nishimura was able to reconstitute human red and white blood cells as well as platelets in the mice to create different variants of mice with humanized liver and immune system characteristics.

“We have Asian liver humanized mice, African-American humanized liver mice, Caucasian humanized liver mice,” he said. He’s also created mice to mimic diseased livers, including congenital liver disease and Alagille syndrome, as well as cancer.

The objective is to use the humanized mice at any stage of drug and vaccine development. But the first was to get at that discrepancy between results in preclinical and clinical trials. In 1993, for example, tests of Fialuridine (FIAU), which was being investigated as a therapy for hepatitis B, resulted in unexpected toxicity in humans, leading to the death of 5 out of 15 patients.

“The traditional mice model did not predict this severe side outcome while the humanized mouse liver model demonstrated the test within a few days,” Dr. Nishimura said.

In thinking about Covid-19 vaccine development, he noted the clinical phase was slowest because it required around 30,000 subjects for each potential vaccine. Humanized mice have the potential to shorten that phase by helping researchers pinpoint stumbling blocks earlier in the process.

For the full transcript of Toshihiko Nishimura’s presentation, click here.

Watch Toshihiko Nishimura’s full presentation below.