Scientists Caution that Phasing Out Animal Research Prematurely Could Risk Women’s Health

When Irina Kovlyagina started her graduate studies at the University Medical Centre Mainz in 2017, she was one of the first people in the institute to use female animals for her experiments. “I was like, ‘Why [do] we take only males? It costs a fortune to produce this [animal colony]. Why don't we also take females?’” recalled Kovlyagina, now a postdoctoral researcher at the same institute focusing on translational psychiatry and behavioral neuroscience.

Historically, scientists largely used male animals for basic and preclinical biomedical research, a bias that plagued the field of neuroscience the most.^1,2 The National Institutes of Health’s (NIH) implementation of the “Sex as a Biological Variable” (SABV) policy in 2016—which mandated that sex be factored into biomedical research—sought to change this.

Eventually, while researchers increasingly studied both sexes, only a minority of the research used a balanced ratio of male and female models.³ So, disease mechanisms, drug actions, and stress responses still largely reflect male physiology. In vivo models, which can capture sex-specific biology, could help increase female representation.⁴

According to Kovlyagina, discontinuing animal research could widen the gap toward understanding biological pathways critical to women’s health. This motivated Kovlyagina to team up with Ivana Jaric, a neuroendocrinologist at the University of Zurich studying female-specific susceptibility to psychiatric disorders and developing guidelines to make sex-inclusive rodent studies more translationally relevant. Together, Kovlyagina and Jaric set out to explore how moving away from animal research would affect this bias.

In a commentary published recently in Nature Neuroscience, Kovlyagina and Jaric argued that prematurely phasing out animal testing would lock in biomedicine’s male-default bias.⁵ According to them, the scientific field runs the risk of preserving sex inequities by discontinuing in vivo work before researchers have used it to correct sex bias in biomedicine, which would have dire consequences for women’s health.

“I agree with [the] sentiment that we need to watch out for what we're going to be missing if we…phase out animal research too rapidly,” said Tracy Young-Pearse, a neurology researcher at Harvard University who was not associated with the commentary. Young-Pearse uses both in vivo and in vitro approaches to study the impact of sex on cellular mechanisms underlying neuropsychiatric diseases. “There's been decades and decades of male mouse dominated studies, and we're just starting to catch up. And then, [prematurely phasing out animal work] would hinder that progress,” she explained.

Monthly Cycles and the Historical Seeds of the Male-Default Bias

In the 1900s, scientists turned to male animals for their experiments, reserving female animals for breeding purposes. They justified experimenting with male animals by saying male data was less variable due to the absence of monthly cycles.⁶ According to one report, 80 percent of the publications that specified sex studied only male animals.⁷

With increasing awareness of the bias, researchers eventually started studying female animals and found no evidence of increased variability compared to male animals.⁶ The NIH’s enforcement of the SABV research policy further set the ball rolling toward equitable representation.

In part due to this, the proportion of publications studying both sexes shot up from 38 percent in 2009 to 68 percent in 2019. However, only about 17 percent of the papers surveyed used an equal number of male and female animals, revealing that the male-default bias persisted.³

According to Jaric, this bias eventually leaked down to new approach methodologies (NAMs), which have been proposed as alternatives to in vivo testing. For example, she said, while AI-based models are critical to the biomedical research pipeline, their outputs are only as good as the data they are trained on, and that data remains highly male-biased, at least in neuroscience. “If we train the network on already male-biased data, we are risking again producing results which are not relevant [to female physiology],” she said.

Organoids and stem cell-based in vitro systems derived from human tissue could be tailored to female biology.⁸ However, cellular sex in such experiments remains under-investigated, indicating that the sex bias that plagued animal research persists in several in vitro studies too.⁹

The Limits of Studying Complex Biology in a Dish

There is an additional problem with studying some sex-specific conditions using in vitro systems. “What we can model [in vitro] is brain cells that have two X chromosomes [and] brain cells that have an X and a Y chromosome,” said Young-Pearse. While these are useful to study the biology of sex-based genetic differences, she noted that investigating the influence of factors like hormones in brain cell biology remains challenging.

“We can do things like add gonadal hormones at concentrations that are relevant to the brain, but that's not the same as all the downstream effects on peripheral tissues that could then culminate in effects on the brain,” Young-Pearse explained.

Jaric agreed, “There is no simple model which can just give us a very specific connection between the brain and different organs and hormones and immune profile.” She added that organoids are extremely valuable and can meaningfully reduce the number of animals needed, but they cannot replace animals yet.

Young-Pearse concurred, noting that her team uses this approach in her lab. While a lot of their work uses stem cell-based models to understand neurodegenerative or neurodevelopmental disorders, they also work on mice. “We kept animal research; we've tried to minimize it only to those things that we cannot address in a dish,” she said.

One of these, according to Kovlyagina, is behavioral biology, such as observing anxiety in mouse models. She added that human-based in vitro models do not necessarily solve the ethics of using animals in research. “It's also important to remember that organoids are not animal free. They're using a lot of animal products,” she said. Indeed, while scientists are exploring alternatives, most approaches to generate organoids rely on using animal-derived extracellular matrices.¹⁰

However, animal models face criticism in not capturing human biology. Only five percent of animal-tested therapies eventually obtain regulatory approval for use in people.¹¹ Given this, is it still important to study sex differences in vivo?

Kovlyagina and Jaric noted that this statistic is more likely a reflection on the complexity of the regulatory process and potential study design flaws. They added that drugs could potentially fail for many reasons beyond the use of animal models. “There's a lot of excellent science that happens with every model system, and there's a lot [of] non-excellent science that happens in every model,” agreed Young-Pearse. “What is the percentage for other non-animal studies that leads to useful therapeutics?” said asked.

The Way Toward Sex-Inclusive Research

Despite noting the importance of continuing studies in animals, Kovlyagina and Jaric said this does not mean keeping animal research as is will solve the sex bias problem. First and foremost, scientists must include SABV as a gold standard, not an exception, by using animals of both sexes for their experiments. This need not translate to increasing the number of animals, “Because these [female] animals are born anyway,” said Kovlyagina.

Both Kovlyagina and Jaric emphasized the importance of complementing in vivo research with NAMs such as in vitro and in silico platforms to give the full picture of human health. “In biology the more models you have, the better understanding you have,” said Kovlyagina. Jaric noted that those using NAMs must also be cognizant of using sex-inclusive research design to use these systems to their full potential. “We cannot understand the whole if we investigate half,” she said.

Young-Pearse agreed, “How science has always thrived is by having these complementary systems come together to really address the biggest problems of our time.” She believes that encouraging more collaborations could combine the expertise of scientists working on different model systems.

She also noted the importance of having conversations about increasing equity in sex research. “To do that in a thoughtful way requires open dialogue, open discourse on the subject. And I think that's what we're doing here,” she said.

Kovlyagina agreed, adding that she hopes that this commentary helps bring sex inequities in research into the limelight and draws attention to the consequences of a premature animal research phase-out to women’s health. “This commentary was to raise awareness about that blind spot,” agreed Jaric. “If you phase out animal experiments, it will have direct consequences to biomedicine, to drug development, and then there is no way back.”