“A substantial health gap between men’s and women’s health remains”

Reliable, long-term data on women’s health are still lacking. This so-called Gender Health Gap points to a medical system that, for decades, has oriented itself more toward the male than the female body. As a result, our understanding of female biology has remained incomplete, with tangible consequences for diagnosis and treatment.

This is where Ângela Gonçalves’ work begins. The DKFZ professor has spent years studying the fundamental biological processes of menstrual cycles and menopause. She combines laboratory experiments with computational analyses to understand how early changes in tissue can lead to cancer. Her research was recognized at the Falling Walls Conference, earning her the title of 'Science Breakthrough of the Year'.

Ms Gonçalves, although it has long been recognized that women respond differently to diseases and therapies, their biology has been poorly studied. Why?

This recognition has been partial, inconsistent and has not translated into systematic change. A substantial health gap between men’s and women’s health remains. In women, research often focuses on high-mortality diseases, but the huge burden of disability-inducing conditions such as menopause-related symptoms, PMS, migraine or endometriosis is still overlooked. Reasons include a lack of sex-disaggregated data, social biases that trivialise symptoms, and the underrepresentation of women in research leadership. In addition, many of these conditions do not fit neatly into medical specialties, making funding more difficult. Together, these factors have created a cycle in which female-specific conditions have been overlooked and underfunded.

Your data show that the menstrual cycle can leave behind fibrosis, inflammation and mutations over many years. Why were these possible consequences overlooked?

The link between lifetime number of menstrual cycles and cancer risk has been recognized for decades. But the molecular mechanisms remain unclear, which limits the development of meaningful prevention strategies. A particular challenge has been that most tissue available to researchers comes from patients who are already unwell. As a result, we lack “normal” molecular snapshots of healthy women as they age. Common processes such as cyclical tissue injury and repair have therefore not been studied in sufficient depth.

Your work brings together laboratory research, modern genomics – that is, the comprehensive analysis of genetic material – and artificial intelligence. What becomes possible through this combination?

Bringing laboratory research together with modern genomics and AI makes it possible to see the full complexity of how thousands of molecular processes interact across cells and tissues. This reveals patterns of inflammation, mutation and stress that traditional single-gene approaches cannot capture. Because genomics produces enormous datasets, AI is essential for integrating them, uncovering hidden correlations and detecting early deviations from healthy biology. Through this combination, we can identify the earliest molecular signs of disease and understand how and why these changes develop.

How does work at the lab bench complement computational analysis?

Our work follows a cycle: in the lab we generate large datasets, which we analyse to identify patterns and derive hypotheses. We then return to the lab to test whether these hypotheses hold up experimentally. The results feed back into the next round of analysis, refining our models and leading to more precise questions. Experiment and data science continuously inform one another.

What does single-cell research reveal that remains invisible in bulk tissue?

Single-cell genomics allows us to examine each cell type separately. Bulk data blend all cell types and can obscure important processes or generate contradictory results. Single-cell analysis shows that different cell types within the same tissue can react in opposite ways: an “increazed” signal may come from a small population of cells, while others show the opposite response. This allows us to identify which cell types truly drive pathological change.

Where does AI make the biggest difference in your work?

When dealing with extremely large datasets from single-cell, spatial and imaging technologies, classical analyses quickly reach their limits. AI helps find patterns in these data, identify gaps and make experiments comparable. In imaging, it can automate image recognition and reveal rare cell states and subtle alterations that humans often overlook.

You also analyse menstrual samples. What can they reveal about women’s health?

Menstrual samples provide a unique, non-invasive window into the endometrium, the tissue that lines the uterus. Because healthy women are not routinely biopsied, most previous data come from clinical patients. Menstrual fluid instead contains direct information about normal, healthy tissue—something blood tests and imaging cannot provide.

Which technology will best advance women’s health in the coming decades?

Wearables. Unlike traditional clinical measurements, which offer only brief snapshots, wearables provide continuous physiological data in a non-invasive and relatively inexpensive way. Combined with molecular measurements, they allow a far more accurate and personalized picture of health, risk trajectories and early warning signs.

Was it difficult to obtain funding for women’s health research?

Not particularly. Most of the men and women I spoke to recognized immediately that these questions matter. Yet the field as a whole is underfunded, because too few people actively advocate for it. To make lasting progress, we need more researchers, funders and institutions that treat women’s health as a core scientific and medical challenge - not a niche topic.

What questions do you want to address in your next research projects?

A central question is how early precancerous lesions arise in the fallopian tube, that is small abnormal changes in tissue that can later lead to ovarian cancer. We still do not fully understand which early changes matter or why some women develop these lesions while others do not. I aim to map the earliest molecular events, across age, hormonal states and genetic backgrounds, and identify what triggers them. The immune system and inflammation likely play a major role. The goal is to detect risk so early that no lesion forms at all, and to find strategies that prevent cancer from taking hold.