Normally when you read a scientific paper they give you a confidence interval for the effect they investigated in their abstract. This Cell paper doesn't. If they could actually tell you how much years of life the mouse gained through the therapy they would put that into the abstract and force you to read the whole paper to get the number.

Are they really moving to humans trials based on nothing but 1 or 2 mouse studies based on biomarkers?

The biomarkers seems interesting. It's useful to know whether humans react the same way towards them. This is an acedemic group making an experiment in humans and not a big pharma company testing a prospective drug.

Given the drug to 5 humans and seeing whether their biomarkers react the same way the one of the mouse isn't as expensive as the later trial phases.

From a pharmaceutical perspective the drug might be useful even if it doesn't help to fight aging in the general population. There are probably some illnesses where it's very important to upregulate mitrochondria.

Other things that might you doubt that the drug is useful for fighting human aging is that it works similarly to calorie restriction. Calorie restriction does increase mouse lifespan but last year we learned that it didn't do well with the apes.

Given the cost of $50,000 a day for a human that the guardian article cites, a drug company would have to first focus on actually investigating related compounds and search for a compound that they can produce at a reasonable price before rushing to get a drug through the approval process.