Identification of potential driver mutations in rare prostate cancer subtype

From the Haffner, Lee, and Ha Labs, Prostate Cancer and Biostatistics and Computational Biology Programs, Cancer Consortium.

Organs are built from layers of tissues composed of specialized cell types. The prostate, for instance, is lined with two layers: a luminal cell layer and a basal cell layer. Most prostate cancers arise from mutations in the luminal, or secretory, cells. However, on very rare occasions, doctors find patients with tumors apparently caused by mutations in the basal cells. These prostatic Basal Cell Carcinomas (BCCs) are so rare that only a couple dozen cases have been described in the literature; nonetheless, BCC has a high incidence of metastasis (10% of patients) and disease recurrence (>40% of patients) after initial therapy. BCC-derived metastases also tend to land in the liver, lungs, and bowels, as compared to acinar adenocarcinoma (AAC) metastases that are more likely to land in the bone. Beyond this, very little is known about the genomic features of BCC. Dr. Jin-Yih (Nick) Low, of the Haffner Lab in the Human Biology Division, set out to change that, and published his findings recently in the American Journal of Pathology.

An image from a microscopic slide of prostatic basal cell carcinoma, sampled from a patient's tumor, that has been stained with hematoxylin and eosin.
Hematoxylin and eosin-stained micrograph of a thin section of prostatic basal cell carcinoma (BCC), an extremely rare tumor type that Dr. Low of the Haffner Lab sought to characterize. Figure adapted from publication.

As a rare cancer, prostatic basal cell carcinoma is unsurprisingly “understudied, and we would like to understand the genetic landscape” that drives this malignancy, Dr. Low told me. Despite clinical samples being “hard to get […] we identified several potential driver genes that may play a role.” Because samples of prostatic BCC are so hard to come by, the authors had to rely on archival, formalin-fixed, paraffin-embedded samples provided by Dr. Jonathan I Epstein at the Johns Hopkins University School of Medicine. The authors used microscopy to identify cancerous areas in the BCC tissue samples, then dissected these regions out to extract DNA for whole-genome sequencing – the first time such technology has been applied to this tumor type. They found some promising similarities between prostatic BCC and basal cell carcinomas that occur in other organs. “Our sequencing result reveals a mutation in [the] CYLD gene which is also mutated in (non-prostatic) basal cell carcinoma,” Dr. Low said. Even more interestingly, “histologically these two tumors -- prostatic and regular basal cell carcinoma -- actually look identical under the microscope (different cancers, different anatomic site, identical histology!)”. When the authors knocked down the CYLD gene with shRNAs in vitro, the cells started to proliferate significantly faster – suggesting this mutation might be important for BCC progression. In addition to the mutations in CYLD, the authors found recurrent loss of chromosome 16 and putative driver gene alterations in KIT, DENND3, PTPRU, and MGA, that they would like to investigate further in future studies.

“We hope to work with more clinical samples to further support our current findings, and also to further understand the biology of this rare cancer,” Dr. Low confided. Additional samples could lead to further mechanistic insights into how the putative driver mutations identified in this work lead to prostatic BCC progression, and what cures might be more effective than current standard-of-care. Dr. Low adds finally that “the Cancer Consortium provided a network that allowed us to collaborate with computer biologists to make this work possible.”

Low JY, Ko M, Hanratty B, Patel RA, Bhamidipati A, Heaphy CM, Sayar E, Lee JK, Li S, De Marzo AM, Nelson WG, Gupta A, Yegnasubramanian S, Ha G, Epstein JI, Haffner MC. 2022. Genomic characterization of prostatic basal cell carcinoma. American Journal of Pathology. S0002-9440(22)00320-0. doi: 10.1016/j.ajpath.2022.09.010.

This work was funded by the National Institutes of Health, the US Department of Defense Prostate Research Program, the Doris Duke Charitable Foundation, the Prostate Cancer Foundation, the Safeway Foundation, and the Richard M. Lucas Foundation.

Cancer Consortium members Drs. Michael Haffner and John K Lee (Prostate Cancer Program) and Dr. Gavin Ha (Biostatistics and Computational Biology Program) contributed to this work.