Professor Ewa Sitnicka Quinn, from Lund University’s Faculty of Medicine, discusses her work on NK cell development and function, as well as clinical applications of NK cells.
In a broad sense, Professor Ewa Sitnicka Quinn’s group at Lund University’s Faculty of Medicine investigates the cellular stages and regulatory pathways governing the development of lymphoid cells in humans and in mice. The knowledge they hope to gain has the potential to lead to a better understanding of hematopoietic disorders such as immune-deficiencies, myeloproliferative disease and leukemia.
Here, Sitnicka Quinn discusses the contribution of natural killer (NK) cells to the immunosurveillance of human cancer, defects in NK cell generation and cytotoxicity, and the benefits of a ‘universal’ off the shelf cell therapy, as well as other areas.
The contribution of natural killer (NK) cells to the immunosurveillance of human cancer remains debatable – where, then, would you like to see future research priorities lie?
From a historic point of view, the concept that the immune system controls and prevents cancer was first introduced by Paul Ehrich in 1909, and that theory was further developed in the 1950s. Since then, it has been generally accepted that defects in the immune system function (immune deficiency) lead to a higher incidence of cancer.
NK cells were first identified in the 1970s, based on their very unique natural ability to kill tumour cells without pre-stimulation, which differentiates them from T cells. In addition, NK cells also produce cytokines which can activate other cells in the immune systems, such as T cells and B cells, and so they are often presented as cells that can make a bridge between innate and adaptive immunity.
A further very important property that has been discovered more recently is that NK cells can mediate memory type of responses. That is, memory-like NK cells ‘remember’ a previous activation, display more potent response, and have longer life span. Typically, NK cells have been estimated to live around four-to-five months, while memory-like NK cells can persist for up to two years.
It is perhaps also important to explain, from a biological perspective, why NK cells are interesting targets for cellular immune therapy (alone or together with T cells).
NK cells, in contrast to T cells or NKT cells, do not express T cell receptors; their activation is regulated by the balance of the signals from inhibitory and activating receptors. NK cells recognise target cells (such as malignant cells, or cells infected with a virus), in the context of self-MHC class I molecules that bind to inhibitory KIR receptors present on the NK cells. If this receptor is bound, then the activity of NK cell is inhibited, however, when MHC class I molecules are lost or expressed at a very low level, the inhibition is broken and NK cell activity is triggered. Activating receptors include NKG2D receptor and NCR killing receptors, as well as CD16 that mediates antibody dependent cytotoxicity (ADCC).
Another important property, as previously mentioned, is that NK cells do not require previous stimulation; they are the first line of the body’s natural defence, and they are part of the so called ‘innate immune system’, which means that they respond and eliminate the pathogen very quickly.
Since their discovery, NK cells have been extensively studied with the respect to their anti-tumour properties, and there is already evidence that NK cell-based immunotherapies show promising results, while other more general studies on this area have predicted that within the next 10 years immunotherapy will constitute about 60% of all cancer treatments.
Coming back to the evidence of NK cells being engaged in immuno-surveillance, an example here is the large cohort study undertaken in Japan over the course of 11 years. This involved the study of 3,500 people in respect to NK cell activity and, based on this study, it has been concluded that high NK cell activity coincides with low cancer incidence, whereas in cases with low cell activity, increased cancer incidents were found. This study also found more than 150 people within the cohort with a previously undiagnosed cancer. Therefore, via a statistical analysis of the population, this demonstrates that NK cells are directly involved in host defence mechanisms preventing cancer development. This is something that is also evident from experimental mouse studies, demonstrating that NK cells were mediating rejection responses and killing of transplanted tumour cells.
Furthermore, in the clinic, in allogeneic hematopoietic stem cell transplantation, it has also been shown that allo-reactive mismatched NK cells are very efficient in providing the graft-versus-leukaemia (GVL) effect in the treatment of patients with acute myeloid leukaemia (AML). Here, donor NK cells are designed to be mismatched with recipients’ cells (based on KIR receptors on NK cells and MHC class I), meaning that the transplanted NK cells would be activated to kill leukaemic cells present in the patient.
Alongside this, other efforts have shown that NK cells taken from leukaemia patients, being activated ex vivo and re-transplanted back into the same patient can kill blast cells without affecting the healthy cells. Then, in the case of solid tumours, NK cells migrate to the tissue surrounding the tumours and, typically, they are found in the areas where the MHC class I expression is low as they recognise this as a potential target.
There is also evidence that mutations that impact NK cell or T cell functionality are associated with a higher risk of cancers. For example, patients with mutations within a gene encoding perforin, a protein involved in the killing of target cells by NK cells, and cytotoxic T cells, often develop lymphoma and leukaemia. Also, mutations in transcription factor GATA2 lead to defects in several immune cells, including NK cells, and patients then frequently develop myelodysplastic syndrome (MDS) which progresses to leukaemia.
There is evidence that individual cancers may vary in their susceptibility to NK cells: some are killed by NK cells, and some may, indeed, escape.
In studies on NK cells in solid tumours it has been found that the tumour environment can induce NK cell suppression. Furthermore, these studies have also shown that tumour cells have many ways to evade NK cell activity. For instance, they could release factors like prostaglandins or transforming growth factor-beta or activate platelets or components such as collagen or salicylic acid, all of which supress NK cell activity.
Similarly, in hematopoietic malignancies such as AML and MDS there is a defect in NK cells and, in many cases, this defect can be used as a prognostic factor – a more severe stage of the disease is associated with more serve NK impairment; there are less NK cells, and they are less active (the expression of the killing receptors is reduced).
The interaction is thus akin to a tug of war; on one side, cancer cells can escape surveillance by making NK cells weaker or not active, and on the other, when NK cells are activated tumour cells can be eliminated. As such, we are now exploring how to activate NK cells, and it will be very important for future research to better understand how the disease environment impacts NK cell development, both in solid and haematological tumours, and, moreover, to what extent NK cell defects contribute to the disease onset and progression.
You have conducted a review of the associations between inherited human diseases and NK cell development as well as function, with a particular focus on defects in NK cell exocytosis and cytotoxicity – what did you find?
In one of the experimental disease models in the lab, we are investigating NK cell development in myeloproliferative diseases. This involves a transgenic mouse model which expresses human mutated gene FLT3-ITD, the gene which represents a most frequently mutated gene found in human leukaemia. From this study, we found that in the mice which carry this mutation, the number of NK cell progenitors and the number of mature NK cells is significantly reduced, and, moreover, that the function of the NK cells in terms of producing cytokines and killing tumour cells similarly decreased.
In this study, we have been also trying to understand whether what we are seeing is an intrinsic or an extrinsic effect, and we found that it is in fact intrinsic to the cells, because when we transplanted bone marrow progenitors from FLT3-ITD donor mice into the healthy recipient mice, the generated NK cells behaved the same as the donor cells. In these experiments, in addition to the mutated cells, we also transplanted normal healthy bone marrow cells, and interestingly, after some time, we saw that the production of NK cells from these healthy bone marrow cells was also affected. This suggests that in addition to intrinsic defects in hematopoietic cells carrying FLT3-ITD, the disease environment does indeed impact NK cell generation.
In an on-going project, we have now begun to purify NK cells from the healthy control mice and their mutated counterparts and to perform genetic screening to compare the gene profiles from these cells in order to see whether any regulatory pathways are affected or if there are any changes in expression of genes which are encoding receptors involved in cellular killing; we want to understand what the potential mechanisms behind this NK cell impairment are and, if we find the regulatory pathway or missing receptor, we want to reintroduce it to develop a therapeutic approach which will allow to restore NK cells and correct their function.
We also use other mouse models to try to understand how, for example, the loss of the GATA2 transcription factor affects NK cells and how this leads to NK cell deficiency.
One of the approaches we employ here is to use mice that are deficient in GATA2, while with our colleagues in the clinic we are also generating induced pluripotent stem cells (iPSC) from the patients carrying GATA2 mutations. Using these iPSC cells, we will investigate how NK cell development is impaired, including how and why and what kind of mechanism is behind it. In this way, we hope to develop an ex vivo model using human cells to study this disease.
Most of the work in my lab focuses on basic research, but we have also begun a collaboration with the Division of Clinical Genetics in order to gain a more in-depth understanding of the impact of myeloproliferative disease and AML on the development and function of NK cells. Here, we will conduct studies using patient materials, and then we will combine the data from the analysis of NK cell differentiation and function with the detailed genetic analysis obtained from these patients.
Multiple on-going clinical trials are focusing on applying NK cells for the treatment of different type of tumours. How do you think these applications will further develop? What
would the benefits of a ‘universal’ off the shelf cell therapy product be?
I would like to stress here again that not being a clinician, my answers are based on current literature.
In the last few years, there were around 300 registered clinical trials for different applications of NK cells; this shows that this is an area that is studied quite intensively. In the clinic, the current emphasis is being placed on boosting NK cell activity by using different means, including cytokines such as interleukin-15 and other factors which are known to be important for NK cells and which could increase their numbers and activity. Another approach is to use treatment with antibodies to enhance antibody-dependent cell-mediated cytotoxicity (ADCC).
James P Allison and Tasuku Honjo, last year’s Nobel Prize laureates, discovered that CDLA-4 and PD1 proteins function as a T cell brake and a specific inhibition of this checkpoint activates T cells; this finding is now a basis for the ‘immune checkpoint’ T cell-based therapy for cancer patients. A similar approach can be used to inhibit NK cell specific checkpoint by specifically targeting with antibodies two inhibitory receptors NKG2A and KIR to activate NK cells.
There are different sources of NK cells for adoptive transfer (the adoptive transfer of NK cells would improve the persistency of their activation and numbers). These include umbilical cord blood, bone marrow, peripheral blood, embryonic stem cells, or induced pluripotent stem cells (iPSC).
I think that to fully benefit from the therapeutic NK cell properties there is a need to better understand their development and regulation, as these processes are not yet fully characterised, despite an appreciation of NK cell potential. The field will benefit significantly from extending this knowledge.
When it comes to NK cell therapies, I believe that they will come to be applied in the treatment of different types of cancer, in both allogeneic and autologous settings, and that therapeutic strategies could be combined. For example, in autologous settings, cytokines could be used to increase the number of NK cells, while antibodies blocking KIR and NKG2A inhibitory receptors can be used in combination to boost NK cell activity.
Here, the therapy would be specifically designed for individual patients or for individual types of cancer. One of the challenges of using a therapy that utilises allogeneic NK cells is to obtain large enough numbers of cells for the treatment and also ‘produce’ them in a uniformed way so that they are fully mature and functional.
A potential way of overcoming this problem is to use NK cells that are generated from human embryonic stem cells (ES) or induced pluripotent stem cells (iPSC) that can be now routinely generated from a variety of sources such as skin biopsy or peripheral blood. It has been shown that both ES and iPSC cells can efficiently produce NK cells with functional properties comparable to the natural counterparts. However, generation of NK cells from iPSC would allow by introducing different properties and modulating their phenotype and function to design a ‘synthetic killer cell’ that expresses specific receptors, has a high specificity and high effector function, while at the same time is resistant to tumour evasion and efficient in homing and persistence.
One would imagine that such a cellular therapy product could be stored in a cell bank and provided on request. Theoretically at least, this can be done. However, its realisation will require a better understanding of NK cells and their regulation, and one should also take into consideration the associated costs – what would the cost be to individually design NK cell products or NK cell therapies for individual patients, for instance? Or is it better to have a universal product that can be used to treat everyone and is ready to go?
Moving forwards, what will be the main focus of your own research?
I plan to focus on the pathways which are important for NK cell development. My group identified a novel human progenitor that is restricted to NK lineage and which only produces NK cells. We want to establish a map of NK cell development (define sequential differentiation steps), which we could then use to characterise factors that are critical in generation of NK cells, progressing from progenitors to mature NK cells. We will also use this human progenitor as a model.
In addition, we are applying the disease models to search for novel regulatory pathways that have not been previously identified.
As already mentioned, we have a leukaemia model that allows us to follow how the progression of the disease affects NK cell development, and comparing gene expression profiles between healthy NK cells and NK cells from different disease stages may help us to identify mechanisms responsible for NK cell defects.
Finally, in collaboration with one of our colleagues at Oslo University, we want to use the NK cell progenitor that we have identified as a target cell population to expand NK cells, instead of iPSC cells that can give rise to many other different cell types in addition to NK cells.
Ewa Sitnicka Quinn
Division of Molecular Hematology
+46 46 2224676