Umeå University's logo

We previously identified three molecular subtypes of prostate cancer (PC) bone metastases, MetA-C, with MetB linked to poor prognosis after androgen deprivation therapy (ADT). This study analyzed epithelial and stromal markers using immunohistochemistry, focusing on their relationship to MetA-C subtypes, spatial heterogeneities, and clinical outcomes after ADT. High tumor proliferation and low PSA expression were associated with MetB and poor outcomes after ADT. Most metastases contained tumor epithelial subclones with different morphologies. In the metastasis stroma, blood vessels and fibroblast-like cells expressed smooth muscle actin (SMA), platelet-derived growth factor β, stroma-derived factor 1 (SDF1), periostin (POSTN), and decorin (DCN). Compared to each other, MetB metastases had higher SMA and ERG + endothelial cell densities, while MetA cases showed higher SDF1 and DCN levels. Accordingly, high POSTN and ERG + densities were associated with poor outcomes after ADT, whereas high DCN indicated favorable prognosis. Low levels of AR-positive stromal cells were linked to poor outcomes. Macrophage and T-lymphocyte densities showed no significant associations with metastases subtypes or outcome. Two stroma subtypes were identified: subtype 1 with higher bone content, lower vessel density, MetA-enrichment and better prognosis compared to subtype 2 that exhibited higher tumor proliferation and lower PSA expression. Most metastases contained regions of both stroma subtypes.

Background: Prostate cancer spinal bone metastases can have a radiographic profile that mimics multiple myeloma.

Objective: To analyse the presence and prognostic value of myeloma-like prostate cancer bone metastases and its relation to known clinical, molecular, and morphological prognostic markers.

Design, setting, and participants: A cohort of 110 patients with prostate cancer who underwent surgery for metastatic spinal cord compression (MSCC) was analysed. Spinal bone metastases were classified as myeloma like (n = 20) or non–myeloma like (n = 90) based on magnetic resonance imaging prior to surgery. An immunohistochemical analysis of metastasis samples was performed to assess tumour cell proliferation (percentage of Ki67-positive cells) and the expression levels of prostate-specific antigen (PSA) and androgen receptor (AR). The metastasis subtypes MetA, MetB, and MetC were determined from transcriptomic profiling.

Outcome measurements and statistical analysis: Survival curves were compared with the log-rank test. Univariate and multivariate Cox proportional hazard models were used to assess the effects of prognostic variables. Groups were compared using the Mann-Whitney U test for continuous variables and the chi-square test for categorical variables.

Results and limitations: Patients with the myeloma-like metastatic pattern had median survival after surgery for MSCC of 1.7 (range 0.1–33) mo, while the median survival period of those with the non–myeloma-like pattern was 13 (range 0–140) mo (p < 0.001). The myeloma-like appearance had an independent prognostic value for the risk of death after MSCC surgery (adjusted hazard ratio 2.4, p = 0.012). Postoperative neurological function was significantly reduced in the myeloma-like group. No association was found between the myeloma-like pattern and morphological markers of known relevance for this patient group: the transcriptomic subtypes MetA, MetB, and MetC; tumour cell proliferation; and AR and PSA expression.

Conclusions: A myeloma-like metastatic pattern identifies an important subtype of metastatic prostate cancer associated with poor survival and neurological outcomes after surgery for MSCC.

Patient summary: This study describes a novel radiographic pattern of prostate cancer bone metastases and its relation to poor patient prognosis.

MicroRNAs (miRNAs) are aberrantly expressed in prostate cancer (PC), but comprehensive knowledge about their levels and function in metastatic PC is lacking. Here, we explored the differential expression of miRNA profiles during PC progression to bone metastasis, and further focused on the downregulation of miRNA-23c and -4328 and their impact on PC growth in experimental models. Using microarray screening, the levels of 1510 miRNAs were compared between bone metastases (n = 14), localized PC (n = 7) and benign prostate tissue (n = 7). Differentially expressed miRNAs (n = 4 increased and n = 75 decreased, p < 0.05) were identified, of which miRNA-1, -23c, -143-3p, -143-5p, -145-3p, -205-5p, -221-3p, -222-3p and -4328 showed consistent downregulation during disease progression (benign > localized PC > bone metastases). The downregulation of miRNA-23c and -4328 was confirmed by reverse transcription and quantitative polymerase chain reaction analysis of 67 metastasis, 12 localized PC and 12 benign prostate tissue samples. The stable overexpression of miRNA-23c and -4328 in the 22Rv1 and PC-3 cell lines resulted in reduced PC cell growth in vitro, and in the secretion of high levels of miRNA-23c (but not -4328) in extracellular vesicles. However, no tumor suppressive effects were observed from miRNA-23c overexpression in PC-3 cells subcutaneously grown in mice. In conclusion, bone metastases display a profound reduction of miRNA levels compared to localized PC and benign disease. The downregulation of those miRNAs, including miRNA-23c and -4328, may lead to a loss of tumor suppressive effects and provide biomarker and therapeutic possibilities that deserve to be further explored.

To improve treatment of metastatic prostate cancer, the biology of metastases needs to be understood. We recently described three subtypes of prostate cancer bone metastases (MetA-C), based on differential gene expression. The aim of this study was to verify the clinical relevance of these subtypes, and to explore their biology and relations to genetic drivers. Freshly-frozen metastasis samples were obtained as hormone-naive (n=17), short-term castrated (n=21) or castration resistant (n=65) from a total of 67 patients. Previously published sequencing data from 573 metastasis samples was also analyzed. Through transcriptome profiling and sample classification based on a set of predefined MetA-C-differentiating genes, we found that most metastases were heterogeneous for the MetA-C subtypes. Overall, MetA was the most common subtype, while MetB was significantly enriched in castration-resistant samples and in liver metastases, and consistently associated with poor prognosis. By gene set enrichment analysis, the phenotype of MetA was described by high androgen response, protein secretion and adipogenesis, MetB by high cell cycle activity and DNA repair, and MetC by epithelial-to-mesenchymal transition and inflammation. The MetB subtype demonstrated single-nucleotide variants of RB transcriptional corepressor 1 (RB1) and loss of 21 genes at chromosome 13, including RB1, but provided independent prognostic value to those genetic aberrations. In conclusion, a distinct set of gene transcripts can be used to classify prostate cancer metastases into the subtypes MetA-C. The MetA-C subtypes show diverse biology, organ tropism and prognosis. The MetA-C classification may be used independently, or in combination with genetic markers, primarily to identify MetB patients in need of complementary therapy to conventional androgen-receptor-targeting treatments.

Prostate cancer (PC) bone metastases can be divided into transcriptomic subtypes, by us termed MetA-C. The MetB subtype, constituting about 20% of the cases, is characterized by high cell cycle activity, low androgen receptor (AR) activity, and a limited response to standard androgen deprivation therapy (ADT). Complementary treatments should preferably be introduced early on if the risk of developing metastases of the MetB subtype is predicted to behigh. In this study, we therefore examined if the bone metastatic subtype and patient outcome after ADT could be predicted by immunohistochemical analysis of epithelial and stromal cell markers in primary tumor biopsies obtained at diagnosis (n = 98). In this advanced patient group, primary tumor International Society of Urological Pathology (ISUP) grade was not associated with outcome or metastasis subtype. In contrast, high tumor cell Ki67 labeling (proliferation) in combination with low tumor cell immunoreactivity for PSA, and a low fraction of AR positive stroma cells in the primary tumors were prognostic for poor survival after ADT. Accordingly, the same tissue markers were associated with developing metastases enriched for the aggressive MetB subtype. The development of the contrasting MetA subtype, showing the best response to ADT, could be predicted by the opposite staining pattern. We conclude that outcome after ADT and metastasis subtype can, at least to some extent, be predicted by analysis of primary tumor characteristics, such as tumor cell proliferation and PSA expression, and AR expression in stromal cells.

Low-affinity immunoglobulin gamma Fc region receptor III-A (FcγRIIIa) is a cell surface protein that belongs to a family of Fc receptors that facilitate the protective function of the immune system against pathogens. However, the role of FcγRIIIa in prostate cancer (PCa) progression remained unknown. In this study, we found that FcγRIIIa expression was present in PCa cells and its level was significantly higher in metastatic lesions than in primary tumors from the PCa cohort (P = 0.006). PCa patients with an elevated level of FcγRIIIa expression had poorer biochemical recurrence (BCR)-free survival compared with those with lower FcγRIIIa expression, suggesting that FcγRIIIa is of clinical importance in PCa. We demonstrated that overexpression of FcγRIIIa increased the proliferative ability of PCa cell line C4-2 cells, which was accompanied by the upregulation of androgen receptor (AR) and phosphatidylinositol-4-phosphate 5-kinase alpha (PIP5Kα), which are the key players in controlling PCa progression. Conversely, targeted inhibition of FcγRIIIa via siRNA-mediated knockdown or using its inhibitory antibody suppressed growth of xenograft PC-3 and PC-3M prostate tumors and reduced distant metastasis in xenograft mouse models. We further showed that elevated expression of AR enhanced FcγRIIIa expression, whereas inhibition of AR activity using enzalutamide led to a significant downregulation of FcγRIIIa protein expression. Similarly, inhibition of PIP5K1α decreased FcγRIIIa expression in PCa cells. FcγRIIIa physically interacted with PIP5K1α and AR via formation of protein-protein complexes, suggesting that FcγRIIIa is functionally associated with AR and PIP5K1α in PCa cells. Our study identified FcγRIIIa as an important factor in promoting PCa growth and invasion. Further, the elevated activation of FcγRIII and AR and PIP5K1α pathways may cooperatively promote PCa growth and invasion. Thus, FcγRIIIa may serve as a potential new target for improved treatment of metastatic and castration-resistant PCa.

PIP5K1α has emerged as a promising drug target for the treatment of castration-resistant prostate cancer (CRPC), as it acts upstream of the PI3K/AKT signaling pathway to promote prostate cancer (PCa) growth, survival and invasion. However, little is known of the molecular actions of PIP5K1α in this process. Here, we show that siRNA-mediated knockdown of PIP5K1α and blockade of PIP5K1α action using its small molecule inhibitor ISA-2011B suppress growth and invasion of CRPC cells. We demonstrate that targeted deletion of the N-terminal domain of PIP5K1α in CRPC cells results in reduced growth and migratory ability of cancer cells. Further, the xenograft tumors lacking the N-terminal domain of PIP5K1α exhibited reduced tumor growth and aggressiveness in xenograft mice as compared to that of controls. The N-terminal domain of PIP5K1α is required for regulation of mRNA expression and protein stability of PIP5K1α. This suggests that the expression and oncogenic activity of PIP5K1α are in part dependent on its N-terminal domain. We further show that PIP5K1α acts as an upstream regulator of the androgen receptor (AR) and AR target genes including CDK1 and MMP9 that are key factors promoting growth, survival and invasion of PCa cells. ISA-2011B exhibited a significant inhibitory effect on AR target genes including CDK1 and MMP9 in CRPC cells with wild-type PIP5K1α and in CRPC cells lacking the N-terminal domain of PIP5K1α. These results indicate that the growth of PIP5K1α-dependent tumors is in part dependent on the integrity of the N-terminal sequence of this kinase. Our study identifies a novel functional mechanism involving PIP5K1α, confirming that PIP5K1α is an intriguing target for cancer treatment, especially for treatment of CRPC.

Selective ERα modulator, tamoxifen, is well tolerated in a heavily pretreated castration‐resistant prostate cancer (PCa) patient cohort. However, its targeted gene network and whether expression of intratumor ERα due to androgen‐deprivation therapy (ADT) may play a role in PCa progression is unknown. In this study, we examined the inhibitory effect of tamoxifen on castration‐resistant PCa in vitro and in vivo. We found that tamoxifen is a potent compound that induced a high degree of apoptosis and significantly suppressed growth of xenograft tumors in mice, at a degree comparable to ISA‐2011B, an inhibitor of PIP5K1α that acts upstream of PI3K/AKT survival signaling pathway. Moreover, depletion of tumor‐associated macrophages using clodronate in combination with tamoxifen increased inhibitory effect of tamoxifen on aggressive prostate tumors. We showed that both tamoxifen and ISA‐2011B exert their on‐target effects on prostate cancer cells by targeting cyclin D1 and PIP5K1α/AKT network and the interlinked estrogen signaling. Combination treatment using tamoxifen together with ISA‐2011B resulted in tumor regression and had superior inhibitory effect compared with that of tamoxifen or ISA‐2011B alone. We have identified sets of genes that are specifically targeted by tamoxifen, ISA‐2011B or combination of both agents by RNA‐seq. We discovered that alterations in unique gene signatures, in particular estrogen‐related marker genes are associated with poor patient disease‐free survival. We further showed that ERα interacted with PIP5K1α through formation of protein complexes in the nucleus, suggesting a functional link. Our finding is the first to suggest a new therapeutic potential to inhibit or utilize the mechanisms related to ERα, PIP5K1α/AKT network and MMP9/VEGF signaling axis, providing a strategy to treat castration‐resistant ER‐positive subtype of prostate cancer tumors with metastatic potential.

Cancer cells facilitate growth and metastasis by using multiple signals from the cancer-associated microenvironment. However, it remains poorly understood whether prostate cancer (PCa) cells may recruit and utilize bone marrow cells for their growth and survival. Furthermore, the regulatory mechanisms underlying interactions between PCa cells and bone marrow cells are obscure. In this study, we isolated bone marrow cells that mainly constituted populations that were positive for CD11b and Gr1 antigens from xenograft PC-3 tumor tissues from athymic nu/nu mice. We found that the tumor-infiltrated cells alone were unable to form tumor spheroids, even with increased amounts and time. By contrast, the tumor-infiltrated cells together with PCa cells formed large numbers of tumor spheroids compared with PCa cells alone. We further utilized xenograft athymic nu/nu mice bearing bone metastatic lesions. We demonstrated that PCa cells were unable to survive and give rise to colony-forming units (CFUs) in media that were used for hematopoietic cell colony-formation unit (CFU) assays. By contrast, PC-3M cells survived when bone marrow cells were present and gave rise to CFUs. Our results showed that PCa cells required bone marrow cells to support their growth and survival and establish bone metastasis in the host environment. We showed that PCa cells that were treated with either siRNA for PIP5K1α or its specific inhibitor, ISA-2011B, were unable to survive and produce tumor spheroids, together with bone marrow cells. Given that the elevated expression of PIP5K1α was specific for PCa cells and was associated with the induced expression of VEGF receptor 2 in PCa cells, our findings suggest that cancer cells may utilize PIP5K1α-mediated receptor signaling to recruit growth factors and ligands from the bone marrow-derived cells. Taken together, our study suggests a new mechanism that enables PCa cells to gain proliferative and invasive advantages within their associated host microenvironment. Therapeutic interventions using PIP5K1α inhibitors may not only inhibit tumor invasion and metastasis but also enhance the host immune system.

Prostate cancer (PCa) is one of the most common cancer types and the fifth cancer-related cause of death among Western world men.  The sex steroid hormone, androgen and androgen receptor (AR) play important roles in PCa progression. Herewith, androgen deprivation therapy (ADT) is used as a regimen for PCa, but inevitably leads to development of castration-resistant PCa (CRPC) and distant metastasis. No effective treatment for metastatic PCa currently exists. Furthermore, it remains poorly understood whether and how the steroid hormone signaling in cooperation with multiple pathways that control proliferation, survival and invasion of cancer cells may contribute to metastatic dissemination and growth.

The aims of my PhD thesis focused on: (i) studying the clinical importance of estrogen- and androgen-related signaling pathways in promoting homing and metastatic growth of PCa cells in bone, (ii) gaining deeper understanding of the underlying mechanisms that facilitate PCa metastasis and treatment resistance, with focus on phosphatidylinositol-4-phosphate 5-kinase type-1 alpha (PIP5K1α), estrogen- and androgen receptor signaling, (iii) testing and characterizing the therapeutic potential of PIP5K1α inhibitor in combination with anti-estrogen or anti-androgen agents to improve treatment and overcome treatment resistance in CRPC.

In my thesis work we have shown that key biomarker genes exhibited unique expression profiles and signatures in PCa subtypes within large patient cohorts. Alterations in androgen- and estrogen-related biomarkers and PIP5K1α/Akt pathways were associated with poor patient outcome. We further discovered that CRPC cells and cancer stem-like cells utilized estrogen-associated factors including aromatase and estrogen receptor alpha (ERα), as well as cyclin A1, a key cell cycle regulator, to gain proliferative advantage, and to survive and metastasize to distant organs.

We found that the interaction between PIP5K1α and AR splice variant AR-V7 contributed to enzalutamide resistance. In series of in vivo treatment experiments using tumor xenograft mice, we demonstrated that ISA-2011B alone or in combination with enzalutamide had great therapeutic potential to suppress growth of tumors that had elevated levels of PI3K/Akt and AR-V7, and that were resistant to enzalutamide monotherapy.

We further showed that combination treatment using tamoxifen together with ISA-2011B selectively blocked elevated ERα/cyclin D1 and PIP5K1α/Akt, leading to tumor regression and had superior inhibitory effect over monotherapy in xenograft mice.

My studies therefore suggest that steroid hormone receptors, PIP5K1α signaling cascade and multiple cellular pathways cooperatively promote PCa progression. Taken together, the reported findings are the first to suggest a new therapeutic potential to inhibit or utilize the mechanisms related to ERα and PIP5K1α/Akt network, and provide a new therapeutic strategy to treat castration-resistant ER-positive subtype of tumors with metastatic potential.