Mapping longitudinal squamous cell lung cancer pathogenesis in pursuit of a preventative therapy

Lung cancer kills more men and women than any other cancer. People with lung cancer have a 5% chance of living ten years. Around 75% of lung cancer is attributable to smoking but while rates of smoking are decreasing in the West they are spiralling in the Far East, India and Africa and the global burden will likely keep escalating higher than the current 2,000,000 cases per year. Despite the new treatments for advanced lung cancer with targeted genetic therapies, lung cancer survival did not change between 2004 and 2014.

Lung cancers develop from the build-up of genetic damage in the individual cells of our lungs. This can cause some cells to begin growing faster than their neighbours and become what we call a dominant 'clone' of cells. The cells may change their appearance becoming detectable to doctors on the surface of the airways as what we call 'pre-cancerous lesions'. Importantly these cells must also adapt and hide from our immune system which should identify cells harbouring damaged genes and remove them; this is called 'immune evasion'. These lesions have the ability over time to progress to a full-blown cancer.

We have discovered that pre-cancerous lesions in the airways are not always destined to become cancer. This observation brings with it several possibilities. Can we determine which precancerous areas are going to progress to cancer so we can deliver treatments earlier? By understanding why some lesions progress and some regress can we develop new therapies preventing lesion progression to cancer? Finally, does this knowledge enable us to potentially prevent airway cancers altogether?

Over the last 15 years we have built internationally unique cohorts of patients with pre-cancerous airway lesions and followed the lesions and patients' clinical outcomes over time. We use biopsies to map the genetic and immune landscapes of the lesions. This has allowed us to compare lesions that progress to cancer to those that harmlessly regress. We have discovered a large number of genetic disturbances and can use these to accurately predict lesion progression. More recently we have also discovered a remarkable regenerative capacity of the lung: that in the right environment, highly damaged but microscopically 'normal' airway cells are replaced with cells undamaged by tobacco.

However, we have no knowledge how these genetic and immune observations change over time. Only with this knowledge can we understand the key mechanisms involved in cancer development and then mimic these abnormalities in experimental systems. This would enable us to define which changes are the most important and which we could potentially target to stop lung cancer forming. This is the aim of this programme of work and we have five interlinking workstreams:

1. Map how the genetic and cellular make-up of pre-cancerous lesions change over time
2. Document the key interactions our immune system makes with pre-cancerous cells, potentially enabling us to develop therapies that would encourage our own immune system to eliminate these early lesions
3. Using biopsies from patients already undergoing immune therapies for lung cancers we will use extra biopsies to examine the effects of an immune treatment on both pre-cancerous lesions and the normal epithelium - demonstrating this type of intervention is both possible and potentially effective
4. We will examine what appears to be the key, but ill-understood role of overexpression (amplification) of a small part of chromosome 3 in pre-cancerous lesions. We have found all of the pre-cancerous lesions that progress in our cohort have the same repetition of a small part of chromosome 3; this region has several interesting genes on it which we believe are the key to a normal cell's ability to progress to a cancerous destiny
5. We will use our knowledge to examine potential preventative therapies in a mouse model of early lung cancer

University College London Hospital started collection of the first cohort of pre-cancerous airway lesions in 1998. We described the clinical equipoise of these lesions progressing or regressing and documented the key transcriptional, genetic and epigenetic differences in these lesions. We then described the immune microenvironment of these lesions and indicators of regression; highlighting the potential to influence the microenvironment towards lesion elimination. Most recently we have documented the genetic damage in normal epithelium from tobacco smoke, but importantly highlighted the previous unknown regenerative potential of a low number of unmutated cells when tobacco smoke is withdrawn. We have used bulk transcription, DNA sequencing and epigenetics array studies combined with protein studies including multicoloured immunofluorescence. However, to understand the pathobiology of cancer development in more detail we need to examine the clonal dynamics of precancerous lesion growth, charting genetic and epigenetic abnormalities and immunophenotypes, including antigen-T cell interactions over time and space within the lesions and their surrounding tissues. To define mechanisms that can be targeted therapeutically we will: 1. Define the molecular and cellular changes in cancer development over time, 2. Understand the immune escape/regulatory axes between T cells and other immune/stromal populations, 3. Demonstrate in humans that airway clones can be influenced by therapy, 4. Delineate the role of the 3q amplicon which is necessary for lesion progression and 5. Clarify the cellular origin of SqCC in a murine model and trial our best prevention strategies. The work is enabled by the approved funding for the clinical cohorts and tissue collections and the excellent facilities at UCL. The findings of this programme are likely applicable to a broad range of other cancer types.