Case Study 1: Understanding the biology of early disease in an array of human malignancies

Summary

• We have developed new mouse models that recapitulate the onset of human malignancies including colorectal cancer , hepatocellular carcinoma and malignant mesothelioma

• We have identified key biological processes driving early tumour development and investigated the mechanisms through which they achieve this

• We have uncovered critical environmental factors that influence disease progression, biomarkers to improve early diagnosis and potential therapeutic approaches for targeting early disease

Development of models of early disease. A. In CRC, (1) acquisition of mutant APC leads to activation and secretion of Notum, (2) which leads to expansion of mutant stem cells and their migration to neighbouring crypts. (3) Notum inhibitors or fasting reduce the competitive advantage APC mutant cells. B. (4) Using mouse models of HCC, we have identified a novel circulating metabolite produced by the β-catenin target glutamine synthetase (5) which correlates with HCC burden and could act as a biomarker of early disease. C. In malignant mesothelioma, exposure to asbestos leads to macrophage recruitment and an unresolved inflammatory response (6). Inability to clear asbestos results in acquisition of mutations in genes and subsequent malignant disease (7). Generation of a mouse model that carries mutations commonly found in human in combination with intrapleural delivery of asbestos results in disease that recapitulates human malignancy (8).

Rapid diagnosis of cancer at an early stage is vital for improving response and survival rates. However, understanding of early disease is often missing, as are reliable and non-invasive techniques for diagnosis. Since 2017, the CRUK Beatson Institute has invested in the development of models of early disease to address this knowledge gap. These models are now delivering important new insights in the biology of early disease and identifying mechanisms of detection and treatment at early stages.

The use of genetically modified mice has allowed us to study the impact of specific mutations throughout the course of cancer development. Importantly, using model systems in which we introduce mutations equivalent to those found in human disease, we can investigate the consequences and effects on the whole organism at very early stages. This approach has furnished us with new insights into the processes that cause cancer, identified key markers for the early detection of this, and is enabling development of novel early intervention/prevention strategies. Additionally, these models have been aligned to patient populations, which will ultimately allow clinical translation of these findings.

During the last five years, we have gained several insights into fundamental processes that underpin development of colorectal cancer (CRC) following loss of the tumour suppressor gene Adenomatous polyposis coli (APC), which is a very common founder mutation in human CRC (~80% of cases). These include understanding key mechanisms by which APC-deficient stem cells out-compete their wild-type neighbours, and how mutant crypts can expand laterally into neighbouring crypt populations and, crucially, how this process can be targeted, for example, through inhibition of the palmitoyltransferase, Notum (1-4). Furthermore, environmental factors – such as fasting – can increase stem cell number and thus reduce the likelihood of APC-deficient cells out-competing their neighbours (5) (N.B. the impact of dietary and metabolic intervention is discussed in case study 2). By working with the CRUK Clinical Trials Unit in Glasgow, we have drawn up secondary prevention protocols to allow us to translate these findings in the next five years.

The power of preclinical models rests upon their ability to faithfully recapitulate human disease. To this end, we have made great strides in developing new, clinically relevant models that align to specific subsets of early human disease. Our work modelling right-sided CRC through the introduction of BRAF mutations resulted in the identification of two distinct classes of lesions – a WNT-high, Lgr5-high, canonical stem cell-driven fate, and a mutant BRAF-associated fate characterised by regenerative, foetal-associated stem populations, and lacking canonical stem markers (6). This animal-based approach exactly aligned with a study of early human lesions via single-cell approaches (7). Moreover, we have used signatures from early stage-disseminating CRC to design new mouse models that recapitulate these virulent cancers, and to identify key mechanistic pathways that underpin early progression of disease (8).

From 2017 onwards, we selected and supported the development of models in two key areas: hepatocellular carcinoma (HCC) and malignant mesothelioma. We felt that this was particularly important due to the lack of physiologically relevant models of these diseases of unmet need, and their impacts upon high-risk local populations within Scotland and beyond. The Glasgow-based CRUK PREDICT-Meso Accelerator (led by Kevin Blyth) has, as one key component, a study that follows patients with benign and premalignant mesothelial disease, generating a unique resource for the discovery of early events in human disease development. In parallel, the Murphy group has developed new mouse models of mesothelioma that closely recapitulate human disease, which we are already using for intervention studies (9). Murphy and his collaborators in Glasgow and Cambridge have recently been awarded two CRUK Programme grants, one in early detection (IAMMED-Meso) and the other (REMIT) to expand this work further. In IAMMED-Meso, deep multi-omics and advanced histopathology techniques are being applied both to our temporally tractable mouse models and to longitudinal patient sample cohorts to define molecular and cellular features predictive of malignant progression. REMIT examines the role of innate and adaptive immune populations in disease aetiology and seeks to identify specific immunological processes that control premalignant neoplastic lesions and the mechanisms of immune evasion that enable their escape.

In HCC, the Bird group have now generated over 35 new models of HCC that we can accurately position in relation to human disease and use to follow disease longitudinally. Using this suite of models, we have already discovered novel circulating metabolites for β-catenin mutated HCC, a disease variant with particularly poor outlook (10) and received a catalyst grant for biomarker detection in both models and patient samples. Furthermore, we are integrated into the CRUK HUNTER HCC Accelerator. In addition, Tom Bird has recently received funding from Perspectum to run an early detection trial in high-risk liver patients that will compare MRI imaging with standard surveillance ultrasound. These same patients will be monitored with longitudinal blood and urine samples, permitting window-of-opportunity drug studies prior to any cancer resection surgery. This provides a uniquely valuable opportunity to translate our early detection/prevention work from our models to patients in the next five years.

References

1.  Huels DJ et al. Wnt ligands influence tumour initiation by controlling the number of intestinal stem cells. Nat Commun. 2018; 9: 1132
2.  Johansson J et al. RAL GTPases drive intestinal stem cell function and regeneration through internalization of WNT signalosomes. Cell Stem Cell. 2019; 24: 592-607
3. Flanagan DJ et al. NOTUM from Apc-mutant cells biases clonal competition to initiate cancer. Nature. 2021; 594: 430-5
4. Lähde M et al. Expression of R-Spondin 1 in ApcMin/+ mice suppresses growth of intestinal adenomas by altering Wnt and transforming Growth Factor Beta Signaling. Gastroenterology. 2021;160: 245-59
5. Bruens L et al. Calorie restriction increases the number of competing stem cells and decreases mutation retention in the intestine. Cell Rep. 2020; 32: 107937
6. Leach JDG et al. Oncogenic BRAF, unrestrained by TGFbeta-receptor signalling, drives right-sided colonic tumorigenesis. Nat Commun. 2021; 12: 3464
7. Chen B et al. Differential pre-malignant programs and microenvironment chart distinct paths to malignancy in human colorectal polyps. Cell. 2021; 184: 6262-80
8. Flanagan DJ et al. Epithelial TGFβ engages growth-factor signalling to circumvent apoptosis and drive intestinal tumourigenesis with aggressive features. Nat Commun. 2022; 13: 7551
9. Grosso S et al. The pathogenesis of mesothelioma is driven by a dysregulated translatome. Nat Commun. 2021; 12: 4920
10. Villar VH et al. Hepatic glutamine synthetase controls N5-methylglutamine in homeostasis and cancer. Nat Chem Biol (2022), online ahead of print