Fankhauser and colleagues recently report on pathways of androgen metabolism and dihydrotestosterone (DHT) synthesis in prostate cancer (1). Two possibilities for the fate of the adrenal steroid androstenedione in peripheral tissues are conversion to testosterone or 5α-reduction to 5α-androstanedione (5α-dione), the latter of which we previously reported predominates in multiple cell line models and clinical metastatic castration-resistant prostate cancer tissues, promotes tumor progression (2), and is probably driven by the upregulation of 5α-reductase isoenzyme-1, which prefers androstenedione over testosterone as its substrate (3).

In the Statement of Translational Relevance, Fankhauser and colleagues state they have “performed a comprehensive screen of the steroidogenic potential” and that they “found no evidence of significant contribution from the previously described ‘backdoor’ or ‘5α-dione’ pathway to androgen synthesis.” Unfortunately, their experimental design cannot determine the order of these metabolic steps, and, thus, their data cannot support their conclusions. Their bold conclusions were drawn from data with experiments that incubated prostatic tissues with steroid precursors with a single technical replicate per patient, for a single time point of 96 hours. First, a time course is necessary to determine reaction sequence. Second, 5α-dione was never measured. Third, their conclusions fail to account for the possibility that transient intermediates are metabolized without accumulation. Ninety-six hours incubation in their experiments is a late time point that may miss the appearance of intermediate metabolites and at which time the majority of 5α-reduced androgens (including DHT) are often found conjugated by glucuronidation and thereby not extracted in the organic phase, unless there is treatment with glucuronidase, which was not reported in this study. Contrary to their conclusions, the data in Fig. 2 show that AD is primarily metabolized to the 5α-reduced metabolite androsterone, which occurs through the intermediacy of 5α-dione. Because the 5α-reduced androgens are reversibly interconverted in cells, it is likely that DHT is formed but is a transient product.

Furthermore, we were most concerned that “no evidence” of conversion to 5α-dione in their “comprehensive screen” was concluded despite the fact that they did not even look for 5α-dione. This is the equivalent of stating with certainty that there is nothing behind door #2 without ever opening the door. These authors state that they did not “detect any significant production of DHT…directly from androstenedione by the ‘5α-dione’ pathway (Supplementary Fig. S1)….” However, the referenced Supplementary Figure shows no data from androstenedione incubations and again did not assess for 5α-dione.

See the Response, p. 4972

R.J. Auchus reports receiving other commercial research support from Janssen Pharmaceuticals and is a consultant/advisory board member for BioMarin Pharmaceuticals, Tokai Pharmaceuticals, and Viamet Pharmaceuticals. No potential conflicts of interest were disclosed by the other author.

1.
Fankhauser
M
,
Tan
Y
,
Macintyre
G
,
Haviv
I
,
Hong
MK
,
Nguyen
T
, et al
Canonical androstenedione reduction is the predominant source of signalling androgens in hormone refractory prostate cancer
.
Clin Cancer Res
2014 Apr 25
.
[Epub ahead of print]
.
2.
Chang
KH
,
Li
R
,
Papari-Zareei
M
,
Watumull
L
,
Zhao
YD
,
Auchus
RJ
, et al
Dihydrotestosterone synthesis bypasses testosterone to drive castration-resistant prostate cancer
.
Proc Natl Acad Sci U S A
2011
;
108
:
13728
33
.
3.
Thigpen
AE
,
Cala
KM
,
Russell
DW
. 
Characterization of Chinese hamster ovary cell lines expressing human steroid 5 alpha-reductase isozymes
.
J Biol Chem
1993
;
268
:
17404
12
.