EURURO Vol. 72 No. 4

E104 656

Letter to the Editor
Reply to Isabel Heidegger, Renate Pichler, and Andreas
Pircher’s Letter to the Editor re: Erik Bovinder Ylitalo,
Elin Thysell, Emma Jernberg, et al. Subgroups of
Castration-resistant Prostate Cancer Bone Metastases
Defined Through an Inverse Relationship Between
Androgen Receptor Activity and Immune Response.
Eur Urol 2017;71:776–87
We appreciate the letter from Heidegger et al highlighting
the potential importance of our current work suggesting
two molecular subtypes of prostate cancer (PC) bone
metastases according to high and low androgen receptor
(AR) activity in tumor cells . Metastases with high AR
activity had low expression of genes involved in MHC class I
antigen presentation and less immune cell infiltration than
metastases with lowAR activity. As Heidegger et al correctly
point out, pathway analysis of gene expression signatures
further indicated metabolic differences between the two
subtypes, with greater cholesterol biosynthesis suggested
in metastases with high AR activity. The functional
importance of high cholesterol biosynthesis was not
evaluated in our study , but certainly deserves further
attention. Fortunately, the recent paper by Alfaqui and co-
workers shed some light on this by providing evidence of
CYP27A1 downregulation during PC progression, as well as
functional evidence that lower CYP27A1 levels lead to lower
27-hydroxycholesterol (27HC) levels and thereby dimin-
ished feedback regulation of cholesterol homeostasis. By
combining transcriptomics data from two of our previous
studies , we can verify lower CYP27A1 mRNA levels in
our bone [1_TD$DIFF]metastasis cohort compared to radical prostatec-
tomy samples. The CYP27A1 mRNA levels were significantly
lower in malignant (n = 13) compared to nonmalignant
(n = 12) prostate tissue (2.6-fold; p = 0.034), and were
further lower in untreated (1.6-fold; p = 0.018; n = 12)
and castration-resistant (1.4-fold; p = 0.017; n = 56) bone
metastases compared to primary PC. Interestingly, and as
anticipated by Heidegger et al, the bone metastasis subtype
we identified as AR-driven showed clearly lower
CYP27A1 expression than the non–AR-driven subtype
(2.6-fold; p = 0.005). Notably, the CYP27A1 mRNA levels
in non–AR-driven PC bone metastases were comparable to
levels in bone metastases from patients with different
cancers. This is in line with results from a previous study, in
which we used gas chromatography and mass spectrometry
[2_TD$DIFF]analysis to study the metabolome of bone metastases from
the same patient cohort, and observed high cholesterol
levels in PC bone metastases but not in bone metastases of
different origin . Accordingly, AR-driven bone metastases
expressed higher levels of HMGCR (coding for the rate-
limiting enzyme in cholesterol biosynthesis; p = 0.003) as
well as SCARB1 and LDLR (receptors for high- and low-
density lipoprotein cholesterol, respectively; p = 0.033 and
p = 0.079), and lower levels of ABCA1 (cholesterol efflux
pump; p = 0.003) when compared to non–AR-driven cases.
No differences in expression were observed for sulfotrans-
ferases between AR-driven and non–AR-driven bone
metastases, as might have been anticipated . It is
noteworthy that AR-driven bone metastases expressed
significantly higher levels of GNMT than non–AR-driven PC
metastases and metastases of different origin (data not
shown), although no differences in sarcosine levels could be
detected
Taken together, our data strengthen the hypothesis
highlighted by Heidegger et al regarding the contribution of
dysregulated cholesterol homeostasis to PC pathogenesis,
and indicate the probability that patients with castration-
resistant PC and AR-driven [3_TD$DIFF]metastases would benefit from
treatments inhibiting cholesterol synthesis and influx.
Conflicts of interest: The authors have nothing to disclose.
References
[1]
[2]
[3]
.
[4]
available at
journal homepage:
DOIs of original articles: .
0302-2838/# 2017 European Association of Urology. Published by Elsevier B.V. All rights reserved.