K. Albin Johansson Cancer Research Fellow (4/2020-3/2023) Leena Latonen is a cell biologist (M.Sc.) with a PhD and Title of Docent in cancer biology. In addition, she has a special training in disease modelling. Leena studied cell biology in the University of Jyväskylä but did her PhD in the University of Helsinki. Research has taken her abroad, to the Karolinska Institute in Sweden and to the University of California in San Diego. Most of the Finnish universities are also familiar to Leena since she used to work in the University of Tampere and is now affiliated to the University of Eastern Finland.
The world too small for the human eye to detect
Leena has always been interested in nature and living organisms. In high school, her biology teacher showed a book containing scanning electron microscopy pictures of different biological material, such as flowers and insects. That got her hooked with the world too small for the human eye to detect. Leena can still remember, how impressive the needle of a mosquito looked like with the saw-pattern. Actually, Leena is still on the path to figure out more of what we can’t see and don’t instantly understand.
Leena’s main interest is molecular mechanisms of cancer. She wants to understand what makes cancer cells tick and how they achieve the miniature form of eternal life with their ability to overcome all set boundaries and find ways around human attempts to kill them. In Leena’s opinion, with the amazing potential for evolution, cancer is both fascinating and scary: a threat from within.
Method development, novel biomarkers and patient samples
Latonen Lab - Cancer stress biology, Leena’s research group, is working on identifying previously unknown mechanisms that enable cancer development and resistance to drugs. The main aim is to develop more efficient treatments for the patients. To achieve the goal, the Group searches for cues in patient tumors of what has gone wrong in the molecular level, identify the functionally relevant mistakes with help of cultured cells, and study how they affect cancer drug responses and how tumors grow in tissues.
Since cancer research is in general lacking handy and quantitative tools to efficiently study cancer both in research labs and in the clinics, Latonen Lab also develops better methods for this. They work a lot around quantitative histopathology and use deep learning to help in cancer cell analysis. The Group also searches for better markers to identify which patients will get a lethal form of the disease and need radical treatment. This is especially important in prostate cancer, where there is overdiagnosis and many patients may gain more harm than good of the knowledge of having a cancer. On the other hand, those patients that really need radical treatment need to be identified earlier to be able to help them before it is too late.
Virtual H&E staining and AI-based models
Leena and her colleagues published recently an article called Unstained Tissue Imaging and Virtual Hematoxylin and Eosin Staining of Histologic Whole Slide Images that showed how to replace chemical tissue stainings with a virtual staining. The chemical staining used in histopathology is a cornerstone method in cancer research to study tumor tissue, as well as in the clinics to grade the tumor and give the patient a prognosis. The study showed that it is possible to omit the use of the staining chemicals, image the tissue sections unstained, and build a deep learning model to do the staining virtually based on the unstained tissue. The published work so far is preclinical and there is still work to do to develop the laboratory methods and the algorithms to be more accurate. In addition, testing their performance with different cancer types in crucial as well as validating the algorithms for clinical use.
However, Leena and her colleagues already have promising results from patients’ breast cancer tissue samples, and they believe this AI-based approach will transform histopathological analysis. Currently, the computing is non-trivial and costly, but it is predicted that soon the method will be more economic and more sustainable than the one that is currently used. With these new tools, the patients will in the future get their diagnoses and prognoses faster. In addition, these tools might be involved in decision making during cancer surgery, especially when tissue-sparing techniques are considered. The authors believe that virtual tissue staining will also lower expenses, eventually helping to ensure sufficient resources with the increasing burden of cancer costs.
Neurodiversity-related psychology, family, cottage and culture
Apart from work, Leena spends some of her spare time in science-related activities. She is involved in a foundation and is interested in neurodiversity-related psychology. Various activities with her teenaged daughters, the family pets, garden, and at the family cottage counterbalance to work. Leena also visits the theatre with her mother and does some handicrafts every year: at least a pair or two of woolen socks to Santa’s gift bag.