Traditionally, most cases of localised prostate cancer were thought to be solely composed of adenocarcinoma, with epithelial cancer cells in direct contact with the surrounding stromal microenvironment. However, there is increasing recognition that other histological patterns can be present in the same tumour. This includes intraductal carcinoma of the prostate (IDC-P), where tumour cells grow within normal ducts and acini instead of the stroma. Although IDC-P is associated with poor prognosis, it has long been overlooked and not routinely reported by pathologists. Therefore, we sought to define the prevalence and biological relevance of IDC-P in high risk prostate cancer.
To determine the prevalence of IDC-P, we conducted a systematic review according to PRISMA guidelines. This analysis of >7000 patient specimens revealed that the prevalence of IDC-P rises from 2.1% in low-risk cohorts to 56.0% in cohorts that developed metastatic or recurrent disease. Since these data showed that IDC-P is a common feature of high risk prostate cancer, we studied the biological features of IDC-P using primary patient-derived xenografts (PDXs). IDC-P was successfully grown in PDXs from men with high risk and/or familial prostate cancer for up to 800 days, demonstrating the persistence of this grown pattern of prostate cancer in vivo. In classical castration-hormone regeneration experiments, we showed that IDC-P contains a subset of “castrate tolerant” cells that withstand androgen deprivation and can subsequently repopulate the tumour, confirming their tumourigenicity. Finally, we used whole genome sequencing to compare the genomic features of matched IDC-P and adenocarcinoma that was microdissected from patient specimens. These data showed that the two pathologies arise from a common tumour clone and then diverge with cancer evolution.
In summary, greater recognition and reporting of IDC-P is warranted. New experimental models that mimic the unique microenvironment of this distinctive form of prostate cancer are required.