Defining the microenvironmental contexture of the biopsy-naive, MRI-characterized prostate.

Prostate cancer is commonly diagnosed through a blood test and a needle biopsy through the rectum. Although this random sampling of the prostate picks up some cancers, many can be missed. In addition, biopsies through the rectum carry a significant risk of infection and bleeding.

In contrast, MRI is a promising way to image the prostate: tumours that are likely to require treatment can be seen and men without visible tumours can be reassured. MRI also targets the urologist's biopsy needle through the skin to suspicious prostate areas with accuracy, efficiency and reduces the risk of any side effects.

It is estimated that in men who are being tested for prostate cancer, a suspicious area on MRI that looks like it might be a tumour is not actually cancer in up to 50% of cases. These so-called false-positive MRIs lead to men suffering the physical and psychological consequences of a biopsy without any actual benefit.

We do not know the how these false-positive MRIs come about, but there is evidence that suggests that cells of the immune system and the cells surrounding prostate glands (called the "stroma") play a role. These inflammatory and stromal cells influence the way tissue grows and appears and can contribute towards a positive MRI without cancer being present.

Many researchers also believe that certain types of inflammation contribute to the transformation of prostate cells into cancerous cells, as well as encouraging the growth of established tumours. Truly understanding these biological processes are necessary for developing new ways of preventing or treating early prostate cancer without surgery.

Unfortunately, few resources are being allocated to define the links between inflammation, stromal cells, prostate cancer and MRI imaging.

In a one-of-a-kind clinical trial called PROMIS organised by UCL investigators, men with suspected cancer had an MRI followed by extensive biopsies of their whole prostate. For this project, we identified PROMIS participants who have suspicious areas on MRI but no cancer on biopsy. We were surprised to see that most of these men had inflammation and stroma in their tissue, sometimes associated with cellular changes that are seen before cancer.

My goal for this fellowship is to characterize inflammation and stroma in the biopsy samples of PROMIS participants, along with any pre-cancerous changes. This will be done through immunohistochemistry, a powerful technique where antibodies that recognise different immune cells are used to visualise individual cells in the prostate samples. I will also analyse the MRI images of men on PROMIS using specialized software and look for imaging features that can differentiate cancer from inflammation or other conditions. Finally, I will combine tissue findings with MRI images and understand how the two are associated.

The proposed research will allow us to tell whether a suspicious area on an MRI is unlikely to be prostate cancer. This could drastically reduce the number of unnecessary biopsies and provide reassurance to men who are being tested for prostate cancer for the first time. We will also understand how inflammation and an altered prostate environment might promote cancer development in the first place. This will allow us to develop strategies that prevent cancer in its very early stages when it is most easily treated and cured.

Aims:
1. Define the microenvironmental contexture of non-cancerous tissue within the naïve prostate.
2. Establish the scientific prerequisites for the deconvolution of positive MRIs in biopsy-naïve men without clinically significant tumours.

Objectives:
1. Define the constitution of non-cancerous biopsy tissue derived from PSA-screened, biopsy-naïve prostates of PROMIS participants with and without clinically significant tumours.
2. Decompose the microenvironmental contexture of the same tissue by specifically identifying immune cell subsets, cytokines, stromal components and cancer precursors.
3. Extract quantitative features from PROMIS MRIs that effectively discriminate non-cancer from cancer.
4. Create stromal immunohistochemical and transcriptional maps that can be aligned to MRI regions of interest and investigate their relationship with MRI visibility.

Methodology:
I will annotate inflammation, stroma and cancer precursors on digital H&E images of PROMIS biopsy cores. I will also perform immunohistochemistry and in situ hybridization in the same tissue to identify specific immune cell subsets (including T and B lymphocytes, NK cells and macrophages), Th1/Th2 cytokines, stromal proteins and HGPIN or ASAP. Stromal transcriptomic signatures will be obtained through tissue dissection, RNA extraction and sequencing and, finally, quantitative feature analysis of MRI areas and their alignment with tissue findings will be performed using specialized software in collaboration with imaging experts.

Scientific and medical opportunities:
The ability to identify pro-tumourigenic microenvironments in patients with and without significant cancer will reform existing diagnostic pathways and expedite novel research on early disease prevention or modulation. The extraction of MRI features that discriminate inflammation from cancer will also spare some men from unwarranted biopsies.

This research project will benefit:

1. Patients with suspected cancer who are candidates for biopsy. In a representative subgroup of biopsy-naïve PROMIS participants without clinically significant tumours, over 60% had an indeterminate or clearly visible lesion on MRI. Such MRI phenotypes are not well defined and these patients would receive a targeted biopsy in the clinical context. As risk assessment with MRI and MRI-targeted biopsies are becoming the standard for men with suspected cancer (Kasivisvanathan et al, 2018), MRI features that non-invasively discriminate cancer from non-cancer are urgently needed. The work proposed for this fellowship will derive such features from MRIs of well-interrogated, biopsy-naïve prostates and could help men avoid unwarranted biopsies in the future.

2. Patients with suspected cancer who had a biopsy. PSA testing has resulted in a diagnostic migration towards smaller, low-grade disease that is often actively surveyed (Welch and Albertsen, 2009). Disease-modulating strategies controlling clinically insignificant disease and preventing progression are becoming increasingly necessary, but non-existent. This is primarily because of a gap in our understanding of the prostatic microenvironmental contexture in men with suspected cancer. The proposed fellowship will establish scientific prerequisites for bridging that gap and designing disease modulating strategies.
In addition, inflammation is an enabling characteristic of malignancy and a common finding on biopsy, but the implications of its presence are poorly understood. Defining a pro-tumourigenic profile in the prostate could have prognostic utility for men who had a biopsy and stimulate further research on prostate cancer prevention.
Collateral psychological benefits of (1) and (2) will be substantial for patients and promote quality of life.

3. Policy makers and charities. Prostate Cancer UK clearly states there is currently no effective way to prevent prostate cancer (prostatecanceruk.org/prostate-information/are-you-at-risk/can-i-reduce-my-risk). The proposed work is, therefore, necessary. Existing NICE recommendations on prostate MRI interpretation and diagnostic standards could also be substantially upgraded.

4. Host institution. Within UCL and UCLH, knowledge of prostate cancer biology will be expanded and more research ideas will be generated, allowing applications for further research funding. This would improve the group's global standing and expand collaborations with other leading institutions.

5. Researchers and academic community. Characterizing prostate tissue derived from this unique cohort will provide a deeper mechanistic understanding of the disease process and new ways of analyzing microenvironmental determinants in men undergoing a biopsy. Protocols and results generated during my fellowship can be disseminated according to the Communications Plan.

6. Industry. Deriving MRI features that discriminate cancerous from non-cancerous areas will be of great interest to the imaging industry. This could lead to further funding and new commercial applications. Defining immune and stromal perturbations that are amenable to intervention will also be of great interest to the biotechnology and pharmaceutical industries. UCL is perfectly positioned to take advantage of such commercial possibilities through its UCL Business Branch.

7. Economy. Effective risk stratification, prevention of unnecessary biopsies and the adoption of preventive strategies will have a positive economic impact on the national health service and the UK economy as a whole. In addition, the industrial impact outlined in (6) will further increase the global financial competitiveness of the UK.

Although all these aims are ambitious, they could be achievable 5-10 years after the completion of my fellowship.