cancers

Review PET Imaging Agents (FES, FFNP, and FDHT) for Estrogen, Androgen, and Progesterone Receptors to Improve Management of Breast and Prostate Cancers by Functional Imaging

John A. Katzenellenbogen

Department of Chemistry and Cancer Center at Illinois, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA; [email protected], Tel.: +1-217-333-6310; Fax: +1-217-333-7325



Received: 1 June 2020; Accepted: 17 July 2020; Published: 23 July 2020


Abstract: Many breast and prostate cancers are driven by the action of steroid hormones on their cognate receptors in primary tumors and in metastases, and endocrine therapies that inhibit hormone production or block the action of these receptors provide clinical benefit to many but not all of these cancer patients. Because it is difficult to predict which individuals will be helped by endocrine therapies and which will not, positron emission tomography (PET) imaging of estrogen receptor (ER) and progesterone receptor (PgR) in breast cancer, and androgen receptor (AR) in prostate cancer can provide useful, often functional, information on the likelihood of endocrine therapy response in individual patients. This review covers our development of three PET imaging agents, 16α-[18F]fluoroestradiol (FES) for ER, 21-[18F]fluoro-furanyl-nor-progesterone (FFNP) for PgR, and 16β-[18F]fluoro-5α-dihydrotestosterone (FDHT) for AR, and the evolution of their clinical use. For these agents, the pathway from concept through development tracks with an emerging understanding of critical performance criteria that is needed for successful PET imaging of these low-abundance receptor targets. Progress in the ongoing evaluation of what they can add to the clinical management of breast and prostate cancers reflects our increased understanding of these diseases and of optimal strategies for predicting the success of clinical endocrine therapies.

Keywords: FES; FFNP; FDHT; receptor-targeting; PET imaging; radiopharmaceuticals; endocrine therapy; breast cancer; prostate cancer; hormone-challenge test

1. Introduction Steroid receptors are transcription factors that play major roles during development, and in both adult men and women, they control reproduction and regulate many other organ systems; they are also important regulators of the growth and spread of various hormone-dependent cancers. This review will cover principally the development of agents to image the estrogen receptor (ER) and the progesterone receptor (PgR) in breast cancer and the androgen receptor (AR) in prostate cancer. Imaging agents for these targets have been labeled with different radioisotopes (various radiohalogens, carbon-11, and radiometals) and have been developed for use with different tomographic imaging devices. Here, I will cover the development of three positron emission tomography (PET) imaging agents labeled with fluorine-18 that have reached the clinic and continue to be of great utility and actively investigated: 16α-[18F]fluoroestradiol (FES) and 21-[18F]fluoro-furanyl-nor-progesterone (FFNP) for breast cancer and 16β-[18F]fluoro-5α-dihydrotestosterone (FDHT) for prostate cancer (Figure1).

Cancers 2020, 12, 2020; doi:10.3390/cancers12082020 www.mdpi.com/journal/cancers Cancers 2020, 12, 2020 2 of 32 CancersCancers 2020 2020, 12, 12, ,x x 22 of of 32 32

EstrogenEstrogen Receptor Receptor Breast Cancer Breast Cancer FESFES

ProgesteroneProgesterone Receptor Receptor BreastBreast Cancer Cancer FFNPFFNP

AndrogenAndrogen Receptor Receptor ProstateProstate Cancer Cancer FDHTFDHT

FigureFigure 1. 1. Structures Structures of of 16 16 αα-[-[1818F]fluoroestradiolF]fluoroestradiolF]fluoroestradiol (FES), (FES), (FES), 21-[ 21-[ 21-[181818F]fluoro-furanyl-nor-progesteroneF]fluoro-furanyl-nor-progesteroneF]fluoro-furanyl-nor-progesterone (FFNP) (FFNP) andand 16 16ββ-[-[1818F]fluoro-5F]fluoro-5F]fluoro-5αα-dihydrotestosteroneα-dihydrotestosterone-dihydrotestosterone (FDHT), (FDHT), (FDHT), with with with thei thei theirr rhormone hormone hormone receptor receptor receptor targets targets and and use use in in cancercancer noted noted at at left. left.

TheThe initial initial synthesis, synthesis, radiolabeling, radiolabeling, and and in in vivo vivo studies studies in in animal animal models models and and humans humans for for these these threethree PET PET probes probesprobes were werewere initiated initiatedinitiated in inin the thethe 1970s 1970s1970s and andand 1980s, 1980s,1980s, and andand culminated culminatedculminated with withwith first-in-man/woman first-in-manfirst-in-man/woman/woman reportsreports for for FES FES in in 1984 1984 [1,2], [[1,2],1,2], FDHT FDHTFDHT in inin 2005 20052005 [3], [[3],3], and andand FFNP FFNPFFNP in inin 2012 20122012 [4]. [[4].4]. For For several several prior prior years, years, early early investigationsinvestigations leadingleading to toto the thethe development developmentdevelopment of these ofof thesthes agentsee agentsagents animated animatedanimated the frontier thethe frontierfrontier of the rapidly ofof thethe evolving rapidlyrapidly evolvingfieldevolving of receptor-binding field field of of receptor receptor radiotracers.-binding-binding radiotracers. radiotracers. The results The The from results results our researchfrom from our our and research research that of and others,and that that and of of others, theothers, insights and and thewethe insights allinsights gained we we from all all gained them,gained werefrom fromoften them, them, taking were were theoften often lead taking taking in defining the the lead lead the in in technical defining defining advances the the technical technical and conceptualadvances advances andunderstandingand conceptual conceptual understanding thatunderstanding are now common that that are are knowledgenow now commo commo innn theknowledge knowledge design and in in the developmentthe design design and and of development development radiotracers forof of radiotracerslimitedradiotracers capacity for for limited limited high acapacity fficapacitynity receptor hi highgh affinity affinity targets. receptor receptor The technicaltargets. targets. The The advances technical technical were advances advances in isotope were were production; in in isotope isotope production;chemicalproduction; reactivity chemical chemical of reactivity reactivity the radionuclide; of of the the radionuclide; radionuclide; and efficiency, and and ease, efficiency, efficiency, and speed ease, ease, of and and radiolabeling speed speed of of radiolabeling radiolabeling and product andpurification;and product product thepurification;purification; conceptual the the understandings conceptual conceptual understand understand were of performanceingsings were were of criteriaof performance performance critical for criteria criteria receptor-targeted critical critical for for receptor-targetedradiotracers,receptor-targeted i.e., radiotracers, targetradiotracers, binding i.e., i.e., affi target nity;target selectivitybindin bindingg affinity; affinity; and capacity; selectivity selectivity imaging and and probecapacity;capacity; molar imaging imaging activity; probe probe and molardistribution,molar activity; activity; metabolism, and and distribution, distribution, and clearancemetabolism, metabolism, behavior. and and clea clea Attentionrancerance behavior. behavior. to all Attentionof Attention these factors to to all all of wasof these these needed factors factors to wasproducewas needed needed PET to to probes produce produce that PET PET gave probes probes informative that that gave gavein vivo informative informativeimages, becausein in vivo vivo images, onlyimages, low because because levels of only only these low low high levels levels affinity of of thesereceptorsthese high high areaffinity affinity present receptors receptors in human are are present breastpresent andin in human human prostate breast breast tumors. and and Inprostate prostate the three tumors. tumors. sections In In the the that three three follow, sections sections these thatfacetsthat follow, follow, of technical these these facets facets and conceptualof of technical technical advancesand and conceptual conceptual are wrapped advances advances into are are historical wrapped wrapped accounts into into historical historical of the developmentaccounts accounts of of the the of developmentFES,development FDHT, and ofof FFNP,FES,FES, FDHT, includingFDHT, andand informative FFNP,FFNP, includingincluding missteps informativeinformative that occurred misstepsmissteps along the thatthat development occurredoccurred along pathways.along thethe developmentEachdevelopment of these pathways. threepathways. sections Each Each ends of of these these with three athree description sections sections ofen ends edsff ortswith with to a a improvedescription description these of of threeefforts efforts clinically to to improve improve used these these PET threeimagingthree clinically clinically agents, used used including PET PET imaging imaging some agents, suggestionsagents, including including for future some some studiessuggestions suggestions that for mightfor future future be worthwhile.studies studies that that might Themight final be be worthwhile.sectionworthwhile. provides The The final final a summary section section provides ofprovides clinical a a summary investigationssummary of of clinical clinical involving investigations investigations these three involving involving agents, these withthese anthree three indication agents, agents, withthatwith an manyan indication indication new studies that that many many are ongoingnew new studies studies and are othersare ongoin ongoin stillgg emerging.and and others others Whilestill still emerging emerging discussion,. .While While here, discussion, discussion, centers on here, here, the centersusecenters of theseon on the the PET use use probesof of these these in PET breastPET probes probes and prostatein in breast breast cancers, and and prostate prostate in principle, cancers, cancers, they in in princi mightprinciple,ple, also they they be usefulmight might inalsoalso other be be usefultypesuseful in ofin other cancersother types types that of of are cancers cancers regulated that that are are by regu hormoneregulatedlated by actionby hormone hormone through action action ER, through through PgR, or ER, AR.ER, PgR, PgR, or or AR. AR. ThroughoutThroughout the the the sections sections sections on on on probe probe probe development, development, development, we we we will will will make make make frequent frequent frequent use use use of of several ofseveral several terms terms terms to to characterizetocharacterize characterize thethe the bindingbinding binding behaviorbehavior behavior ofof of receptor-targetedreceptor-targeted receptor-targeted radiotracers;radiotracers; radiotracers; hence,hence, hence, thesethese these deservedeserve some somesome explanationexplanation (Figure (Figure (Figure 2). 2).2).

RelativeRelative Binding Binding Affinity Affinity (RBA) (RBA) BindingBinding Selectivity Selectivity Index Index (BSI) (BSI) = = Non-SpecificNon-Specific Binding Binding (NSB) (NSB)

ValuesValues are are indexed indexed to to estradiol, estradiol, the the standard: standard: RBARBA = = 100, 100, NBS NBS = = 1, 1, so so BSI BSI = = 100 100

Figure 2. Definitions of relative binding affinity (RBA), nonspecific binding (NSB), and binding FigureFigure 2.2. DefinitionsDefinitions ofof relativerelative bindingbinding affinityaffinity (R(RBA),BA), nonspecificnonspecific bindinbindingg (NSB),(NSB), andand bindingbinding selectivity index (BSI). selectivityselectivity index index (BSI). (BSI).

• RBA and Ki:Affinity for the intended target is expressed relative to a standard ligand, typically •• RBA and Ki i: Affinity for the intended target is expressed relative to a standard ligand, typically RBAa tritiated and K version: Affinity of for a knownthe intended high-a targetffinity is ligand expressed used relative as a tracer to a standard in convenient ligand, radiometric typically a tritiated version of a known high-affinity ligand used as a tracer in convenient radiometric acompetitive tritiated version binding of assays,a known such high-affinity as [3H]estradiol ligand for used ER[ 5as]. a Thus, tracer for in ER, convenient the standard, radiometric estradiol competitivecompetitive binding binding assays, assays, such such as as [ 3[H]estradiol3H]estradiol for for ER ER [5]. [5]. Thus, Thus, for for ER, ER, the the standard, standard, estradiol estradiol (which(which has has a a K Kd dvalue value of of 0.2 0.2 nM), nM), is is assigned assigned a a relative relative binding binding affinity affinity ( RBA(RBA) )value value of of 100, 100, with with lowerlower affinity affinity candidate candidate tracers tracers having having RBA RBA valu valueses below below 100, 100, and and higher higher affinity affinity ones ones above above Cancers 2020, 12, 2020 3 of 32

(which has a Kd value of 0.2 nM), is assigned a relative binding affinity (RBA) value of 100, with lower affinity candidate tracers having RBA values below 100, and higher affinity ones above 100. If of interest, K values can be calculated from RBA values: K (in nM) = (0.2 100)/RBA. i i × The standard compound for PgR is R5020, with a Kd of 0.4 nM [6,7], and for AR is methyltrienolone (R1881), with a Kd of 0.2 nM [6]. NSB and BSI: Nonspecific binding (NSB) is high-capacity, low-affinity, off-target binding, • to proteins such as serum albumin and possibly lipid phases, and it is also measured relative to that of the standard ligand, but in this case, the standard is assigned an NSB value of 1. As a measure of target binding selectivity, we defined the ratio of specific to nonspecific binding as the binding selectivity index, BSI: BSI = RBA/NSB [8,9]. Thus, the standard ligand has a BSI of 100 (i.e., 100/1), and candidate tracers with higher selectivity have BSI values above 100, whereas those with lower selectivity have BSI values below 100. Notably, unlike RBA, BSI is sensitive to both target binding affinity and nonspecific binding.

NSB can be measured experimentally by various, though somewhat tedious, methods such as equilibrium dialysis with serum or serum albumin [9]. We and others have found that NSB is proportional to compound lipophilicity, which is typically expressed as an octanol–water partition coefficient, LogPo/w, and can be measured more conveniently by octanol–water shake flash methods [7] or by reversed-phase HPLC after suitable calibration [10]. Even more conveniently and of equal accuracy, NSB values can be based on calculated LogPo/w values [9].

2. FES, a PET Imaging Agent for Estrogen Receptors in Breast Cancers

2.1. A Prelude to In Vivo Imaging of ER: Finding a Ligand (Steroidal vs. Nonsteroidal) and a Radionuclide (Iodine, Bromine, or Fluorine) Enabling Effective and Selective Uptake by ER-Target Tissues Our earliest interest in preparing gamma-emitting estrogens was not at all for the purpose of in vivo imaging, but rather to study the kinetics of estrogen uptake in ER-positive tissues, followed by the exchange of estrogens once bound by the receptor. We planned to do this by preparing iodine-125 and iodine-131 labeled hexestrol, a nonsteroidal ligand for ER. Hexestrol has very high affinity for ER (RBA = 300; Kd = 0.07 nM), and while progressive iodine substitution reduced binding affinity, monoiodohexestrol having a single iodine adjacent to one phenol still bound quite well to ER (RBA 14; Kd 1.4 nM) [11,12] (Figure3). It was fortuitous that early in 1974, when I presented early phases of this work at a meeting of the American Chemical Society in Los Angeles, I learned of the possibility of using such compounds for in vivo imaging of ER-positive breast cancer by gamma scintigraphy. Shortly thereafter, we performed our first biodistribution study of [125I]iodohexestrol in immature female rats; immature animals were used because their endogenous levels of estradiol are low and would not compete with probe binding to ER. In addition, it was well known that high molar activity [3H]estradiol was taken up avidly and selectivity by the uterus [13,14]. The results with [125I]iodohexestrol, however, were disappointing, but instructive: unlike [3H]estradiol, uterine uptake of [125I]iodohexestrol was only modest and was not blocked by unlabeled estradiol [12] (Figure3). Cancers 2020, 12, x 3 of 32

100. If of interest, Ki values can be calculated from RBA values: Ki (in nM) = (0.2 × 100)/RBA. The standard compound for PgR is R5020, with a Kd of 0.4 nM [6,7], and for AR is methyltrienolone (R1881), with a Kd of 0.2 nM [6]. • NSB and BSI: Nonspecific binding (NSB) is high-capacity, low-affinity, off-target binding, to proteins such as serum albumin and possibly lipid phases, and it is also measured relative to that of the standard ligand, but in this case, the standard is assigned an NSB value of 1. As a measure of target binding selectivity, we defined the ratio of specific to nonspecific binding as the binding selectivity index, BSI: BSI = RBA/NSB [8,9]. Thus, the standard ligand has a BSI of 100 (i.e., 100/1), and candidate tracers with higher selectivity have BSI values above 100, whereas those with lower selectivity have BSI values below 100. Notably, unlike RBA, BSI is sensitive to both target binding affinity and nonspecific binding. NSB can be measured experimentally by various, though somewhat tedious, methods such as equilibrium dialysis with serum or serum albumin [9]. We and others have found that NSB is proportional to compound lipophilicity, which is typically expressed as an octanol–water partition coefficient, LogPo/w, and can be measured more conveniently by octanol–water shake flash methods [7] or by reversed-phase HPLC after suitable calibration [10]. Even more conveniently and of equal accuracy, NSB values can be based on calculated LogPo/w values [9].

2. FES, a PET Imaging Agent for Estrogen Receptors in Breast Cancers

2.1. A Prelude to In Vivo Imaging of ER: Finding a Ligand (Steroidal vs. Nonsteroidal) and a Radionuclide (Iodine, Bromine, or Fluorine) Enabling Effective and Selective Uptake by ER-Target Tissues Our earliest interest in preparing gamma-emitting estrogens was not at all for the purpose of in vivo imaging, but rather to study the kinetics of estrogen uptake in ER-positive tissues, followed by the exchange of estrogens once bound by the receptor. We planned to do this by preparing iodine- 125 and iodine-131 labeled hexestrol, a nonsteroidal ligand for ER. Hexestrol has very high affinity for ER (RBA = 300; Kd = 0.07 nM), and while progressive iodine substitution reduced binding affinity, monoiodohexestrol having a single iodine adjacent to one phenol still bound quite well to ER (RBA 14; Kd 1.4 nM) [11,12] (Figure 3). It was fortuitous that early in 1974, when I presented early phases of this work at a meeting of the American Chemical Society in Los Angeles, I learned of the possibility of using such compounds for in vivo imaging of ER-positive breast cancer by gamma scintigraphy. Shortly thereafter, we performed our first biodistribution study of [125I]iodohexestrol in immature female rats; immature animals were used because their endogenous levels of estradiol are low and would not compete with probe binding to ER. In addition, it was well known that high molar activity [3H]estradiol was taken up avidly and selectivity by the uterus [13,14]. The results with 125 3 [CancersI]iodohexestrol,2020, 12, 2020 however, were disappointing, but instructive: unlike [ H]estradiol, uterine uptake4 of 32 of [125I]iodohexestrol was only modest and was not blocked by unlabeled estradiol [12] (Figure 3).

Figure 3. 3. EstrogenEstrogen receptor receptor ligands. ligands.(A) Estradiol, (A the) Estradiol,endogenous the estrogenic endogenous hormone, estrogenic (B) o- Iodohexestrol,hormone, (B ) (Co)-Iodohexestrol, bromoestradiols, (C )16 bromoestradiols,α-[77Br]bromoestradiol 16α -[(BE2)77Br]bromoestradiol and 16α-[77Br]bromo-11 (BE2) andβ- methoxyestradiol16α-[77Br]bromo-11 (BME2).β-methoxyestradiol Under each compound, (BME2). Underthe numbers each compound, indicate the the binding numbers selectivity indicate index the (BSI),binding which selectivity is the indexratio (BSI),of the which relative is thebinding ratio ofaffinity the relative (RBA) binding to the anonspeffinity (RBA)cific binding to the nonspecific (NSB); all threebinding values (NSB); are allscaled three relative values to are estradiol scaled relativeas the standard, to estradiol set asat 100 the. standard,The number set of at asterisks 100. The indicate number XX theof asterisks degree of indicate uptake theefficiency degree and of uptake selectivity efficiency in the anduterus selectivity of immature in the rats. uterus XX indicates of immature no estrogen- rats. specificindicates uterine no estrogen-specific uptake. uterine uptake.

Initially puzzled by these results, we took a deeper look at the binding characteristics of [125I]iodohexestrol to consider not only ER binding affinity but also nonspecific binding, NSB, i.e., binding to abundant low-affinity components such as serum albumin and lipid phases. In doing this, we articulated the concept of the binding selectivity index, BSI, which accounts of both the desired high affinity target binding and the undesired nonspecific, off-target binding (see Figures2 and3)[ 9]. Estradiol by definition had a BSI value of 100 (RBA of 100/NSB of 1), but by this metric, iodohexestrol, with both a lower RBA of 14 and a much higher NSB of 10, had it an embarrassingly low BSI of 1.4 [12]. Clearly, the very low BSI value for [125I]iodohexestrol provided some explanation for its poor biodistribution. Later, however, we found another reason: the structure of iodohexestrol sufficiently resembles the terminus of thyroid hormone that it is tightly bound by transthyrethrin, a thyroxine-binding prealbumin that is an abundant in serum protein. This off-target interaction overshadows any hint of the uptake of iodohexestrol by the high affinity but limited capacity of ER in the uterus. The off-target binding of iodohexestrol to a serum protein also foreshadowed marked effects we later found with another serum binding protein, sex-hormone binding globulin (SHBG), a protein thought to facilitate the in vivo distribution of certain steroids. As will be noted later, binding to SHBG can have marked effects, both good and bad, on the biodistribution of ER and AR-targeted PET radiotracers that merits further investigation [15–18].) Fortuitously, later in 1974, I met and joined forces with Michael Welch a Professor of Radiological Sciences at Washington University Medical School in St. Louis, about a 3-h drive from the University of Illinois at Urbana-Champaign; this began a fruitful collaboration in radiopharmaceutical development that extended for nearly 38 years. At the time, other researchers interested in imaging ER in breast cancers were seeing favorable biodistribution using steroidal estrogens radioiodinated at different positions [19–25]. At the time, Dr. Welch was receiving bromine-77 shipments from Los Alamos National Laboratory for other purposes (protein labeling); so, we began to prepare bromoestrogens. In general, 16α-bromoestradiol (BE2), readily prepared by electrophilic bromination of a 17-enol acetate precursor, bound to ER with greater affinity than estradiol (Figure3)[ 26], and BE2 and other analogs could be radiobrominated efficiently on a tracer scale [27], giving products with very high molar activities [28,29]. Biodistribution studies confirmed that these analogs were taken up efficiently and selectively by the uterus of immature female rats, now nicely mimicking the selective uptake of [3H]estradiol [27]. In an adult rat, BE2 was also taken up selectively by ERα-positive DMBA (dimethylbenzanthracene)-induced mammary tumors [27]. The lesson we learned from iodohexestrol that both high receptor affinity and low nonspecific binding were important was confirmed by the behavior of these radiobrominated estrogens, because their BSI values (shown in Figure3) were more useful predictors of their in vivo uptake efficiency and selectivity than their binding affinity (RBA) values alone, particularly in predicting the retention of a tracer by the uterus [30]. Compared to the parent bromoestrogen, 16α-[77Br]bromoestradiol (BE2), with Cancers 2020, 12, 2020 5 of 32 a BSI of 84 (130/1.5) was similar to estradiol; the more polar 16α-[77Br]bromo-11β-methoxyestradiol (BME2) actually bound to ER 4-fold less well (RBA 39) but had a 10-fold lower NSB (0.1), giving it a much higher BSI value of 390 (39/0.1). Indeed, the uptake of BME2 by the uterus was greater than that of BE2. Its target tissue retention, in particular, was much longer (t1/2 of 6 h for BE2 vs. 24 h for BME2) [28], clearly showing evidence for the pronounced target-tissue retention that can be displayed by compounds having high BSI values [30,31]. These radiobrominated estrogens were simple to prepare, had excellent binding characteristics, and gave very efficient and ER-selective biodistributions. Our attempts to use them for in vivo imaging with 2D gamma cameras, however, was confounded by the high emission energy of bromine-77. Some accumulation of [77Br]BE2 activity above a relatively high background could be noted in DMBA-induced mammary tumors in rats [26,27], but in humans, the uptake of [77Br]BE2 in ER-positive breast tumors was obscured by extensive scattering from other sites [32]. This prompted us to switch isotopes to fluorine-18 to benefit from its well-recognized advantages of an element that is small and forms very strong bonds to carbon. While its high electronegativity gives the C-F bond a large dipole moment, the potent estrogen, estriol, also has a polar OH group at the 16α position and binds well to ER. Our switch to a positron-emitting radionuclide fit well with the rapid development of PET imaging technology at Washington University, but at this early time, there were challenges both in the cyclotron production of fluorine-18 and its use for radiolabeling. Fluorine-18 was being produced in gas targets (20Ne(d,α)18F) and was used to prepare 2-[18F]fluoro-2-deoxyglucose by an electrophilic reaction using 18 [ F]F2.Efficient extraction of activity from the gas target required carrier F2, which reduced molar activity far below what is useful for receptor imaging. For nucleophilic radiofluorination, which we planned to use, efficient extraction of the activity from gas targets without dilution (e.g., as H [18F]F) proved to be very cumbersome [33]. Soon thereafter, fluorine-18 production in oxygen-18 water targets (18O(p,n)18F) gave large quantities of [18F]fluoride ion that was easily recovered from the target as an aqueous solution. Rapid drying by azeotropic distillation with acetonitrile in the presence of lipophilic cations or cation chelators produced [18F]fluorine in an organic soluble, chemically reactive form [1,2,34]. A final technical hurdle required us to trace the source of fluorine-19 that was inadvertently and markedly reducing the specific activity of our fluorine-18 radiolabeled products [2]. The source was Teflon tubing used for activity transfers from the target and during synthesis that released fluoride-19 ion when exposed to hard radiation. Replacement with polyethylene tubing provided us with the very high molar activities required for in vivo imaging of low abundance targets such as ER [2]. A more extensive study of fluoride ion release from Teflon lines during radiochemical syntheses has recently been published [35].

2.2. An Exploration of Various [18F]Estrogens and the Birth of FES We prepared a number of fluorine-labeled steroidal and nonsteroidal estrogens [1,36–38], but soon focused on the 16 epimers of fluoroestradiol, which were prepared by [18F]fluoride ion displacement of the corresponding epimeric 16 trifloxy estrones, which proceeded with inversion of stereochemistry; the second triflate protected the 3-phenol (Figure4). Subsequent reduction of the 17-ketone was done with LiAlH , first at low temperature ( 78 C) to give the 17β-epimeric alcohol with good selectivity 4 − ◦ and then warming to RT to cleave the 3-O-triflate [1]. The 16α epimer of the product bound twice as well to ER than did the 16β isomer, giving BSI values of 114 (80/0.7) for 16α and 43 (30/0.7) for 16β [39]. Thus, 16α-[18F]fluoroestradiol, which we named, FES, became our tracer of interest. We could routinely obtain FES with molar activities well in excess of 1000 Ci/mmol (37 GBq/µmol) [2], which were sufficient for in vivo PET imaging [40]. Following our original publication of the synthesis of FES [2], alternative routes have been reported, some of which avoid stereochemical issues at C-17 [41–43] and are more adaptable to automated synthesis systems [44–46], which were not available when we first developed our synthesis. Cancers 2020, 12, x 5 of 32

Our switch to a positron-emitting radionuclide fit well with the rapid development of PET imaging technology at Washington University, but at this early time, there were challenges both in the cyclotron production of fluorine-18 and its use for radiolabeling. Fluorine-18 was being produced in gas targets (20Ne(d,α)18F) and was used to prepare 2-[18F]fluoro-2-deoxyglucose by an electrophilic reaction using [18F]F2. Efficient extraction of activity from the gas target required carrier F2, which reduced molar activity far below what is useful for receptor imaging. For nucleophilic radiofluorination, which we planned to use, efficient extraction of the activity from gas targets without dilution (e.g., as H [18F]F) proved to be very cumbersome [33]. Soon thereafter, fluorine-18 production in oxygen-18 water targets (18O(p,n)18F) gave large quantities of [18F]fluoride ion that was easily recovered from the target as an aqueous solution. Rapid drying by azeotropic distillation with acetonitrile in the presence of lipophilic cations or cation chelators produced [18F]fluorine in an organic soluble, chemically reactive form [1,2,34]. A final technical hurdle required us to trace the source of fluorine-19 that was inadvertently and markedly reducing the specific activity of our fluorine-18 radiolabeled products [2]. The source was Teflon tubing used for activity transfers from the target and during synthesis that released fluoride-19 ion when exposed to hard radiation. Replacement with polyethylene tubing provided us with the very high molar activities required for in vivo imaging of low abundance targets such as ER [2]. A more extensive study of fluoride ion release from Teflon lines during radiochemical syntheses has recently been published [35].

2.2. An Exploration of various [18F]Estrogens and the Birth of FES We prepared a number of fluorine-labeled steroidal and nonsteroidal estrogens [1,36–38], but soon focused on the 16 epimers of fluoroestradiol, which were prepared by [18F]fluoride ion displacement of the corresponding epimeric 16 trifloxy estrones, which proceeded with inversion of stereochemistry; the second triflate protected the 3-phenol (Figure 4). Subsequent reduction of the 17- ketone was done with LiAlH4, first at low temperature (−78 °C) to give the 17β-epimeric alcohol with good selectivity and then warming to RT to cleave the 3-O-triflate [1]. The 16α epimer of the product bound twice as well to ER than did the 16β isomer, giving BSI values of 114 (80/0.7) for 16α and 43 (30/0.7) for 16β [39]. Thus, 16α-[18F]fluoroestradiol, which we named, FES, became our tracer of interest. We could routinely obtain FES with molar activities well in excess of 1000 Ci/mmol (37 GBq/μmol) [2], which were sufficient for in vivo PET imaging [40]. Following our original publication of the synthesis of FES [2], alternative routes have been reported, some of which avoid stereochemical

Cancersissues2020 at C-17, 12, 2020 [41–43] and are more adaptable to automated synthesis systems [44–46], which6 were of 32 not available when we first developed our synthesis.

. 18 α FigureFigure 4.4. SynthesisSynthesis ofof thethe twotwo 16-[16-[18F]fluoroestradiolsF]fluoroestradiols and their binding values.values. FES (the 16α-epimer, bottombottom right)right) waswas carriedcarried forwardforward forfor positronpositron emissionemission tomographytomography (PET) (PET) imaging imaging studies. studies. In biodistribution studies, FES showed high uptake in the uterus and ovary of immature female In biodistribution studies, FES showed high uptake in the uterus and ovary of immature female rats that was nearly completely reversed by a blocking dose of unlabeled E2 (Figure5). Uptake in rats that was nearly completely reversed by a blocking dose of unlabeled E2 (Figure 5). Uptake in nontarget organs varied, but was unchanged by the blocking dose of E2. Defluorination was minimal, Cancersnontarget 2020 ,organs 12, x varied, but was unchanged by the blocking dose of E2. Defluorination was minimal,6 of 32 noted by the absence of uptake in bone. Other F-18 labeled estrogens we investigated around the noted by the absence of uptake in bone. Other F-18 labeled estrogens we investigated around the samesame time showed showed no no particular particular advantage advantage over over FES FES [2]; [2 so,]; so,FES, FES, on which on which we published we published first firstin 1984 in 1984[2], became [2], became the focus the focus of our of our clinical clinical development development efforts, efforts, all all done done in in collaboration collaboration with with Washington UniversityUniversity colleagues.colleagues.

FigureFigure 5.5.Tissue Tissue biodistribution biodistribution of FESof FES at 1 at h in1 immatureh in immature rats alone rats oralone blocked or blocked with unlabeled with unlabeled estradiol. (Preparedestradiol. (Prepared from data infrom reference data in [ 2reference].) [2].)

ItIt isis worthworth noting noting that that in in subsequent subsequent years, years, we we did did undertake undertake a substantial a substantial effort effort to improve to improve upon FESupon in FES terms in ofterms its binding of its selectivitybinding selectivity and its target and uptakeits target effi ciencyuptake and efficiency selectivity and by selectivity introducing by variousintroducing substituents various atsubstituents the 11β and at 17 theα positions. 11β and 17 Theseα positions. studies areThese described studiesin are detail described elsewhere in detail [47], butelsewhere in essence, [47], althoughbut in essence, we were although able towe obtain were able compounds to obtain that compounds had better that RBA had and better BSI RBA binding and valuesBSI binding and were values substantially and were to markedly substantially better to than markedly FES in terms better of theirthan biodistribution FES in terms in of rodents, their theybiodistribution were not better in rodents, in humans they [ 18were]. This not unexpected better in humans species [18]. difference This unexpected appears to involve species didifferencefferences inappears the binding to involve characteristics differences of serumin the steroidbinding binding characteristics proteins of in rodentsserum steroid (α-fetoprotein) binding vs.proteins humans in (sexrodents hormone (α-fetoprotein) binding globulin). vs. humans It (sex is an hormone issue that binding is still notglobulin). well understood, It is an issue and that also is still appears not well to aunderstood,ffect the behavior and also of the appears PET imaging to affect agents the behavior for the androgen of the PET receptor imaging (see Sectionagents 3for). Morethe androgen recently, versionsreceptor of(see FES Section in which 3). More strategically recently, placed versions substituents, of FES in includingwhich strategically unlabeled placed fluorine, substituents, appear to enhanceincluding binding unlabeled characteristics fluorine, appear and reduce to enhance metabolic binding clearance, characteristics are reported and to provide reduce somewhatmetabolic betterclearance, images are thanreported those to obtained provide withsomewhat FES [48 better]. images than those obtained with FES [48]. In addition to the preclinical toxicology studies required at the time for the use of high molar activity radiotracers in humans, we examined the behavior of FES in rodent models of breast cancer. In DMBA-induced mammary tumors in adult female rats, we examined the time course of FES tumor uptake and metabolism and looked for correlations between metrics of blood volume and blood flow in tumors with ER levels in tumors and in the uterus; these relationships proved to be complex and difficult to interpret [49]. On the other hand, a dose–response study of FES in immature female rats was very illuminating. The uptake of FES in ER-rich tissues (uterus and ovary) became saturated only at the higher mass dose levels, whereas tissues with lower ER levels, such as muscle and esophagus, showed low but clearly measurable levels of specific uptake of FES activity that was saturated at much lower mass dose levels [50]. To us, this indicated that while the uptake of FES in very ER-rich target organs might be “flow limited” and not proportional to high ER levels [31], the specific uptake by the lower levels of ER found in breast tumors were likely to be proportional to receptor levels [40]. In 1988, we published the first PET images of ER-positive breast tumors in women [51], and one image, presented in advance of our publication at the 1987 national meeting of the Society of Nuclear Medicine, was highlighted by Professor Henry Wagner as “Image of the Year” (Figure 6) [52]. Because it was the first PET image of a receptor in cancer, it represented a milestone in the development of receptor-targeted radiotracers for PET imaging. A discussion of the clinical utility of FES-PET imaging ER in breast cancer, as well as that for FDHT-PET and FFNP-PET, is given in Section 5. Cancers 2020, 12, 2020 7 of 32

In addition to the preclinical toxicology studies required at the time for the use of high molar activity radiotracers in humans, we examined the behavior of FES in rodent models of breast cancer. In DMBA-induced mammary tumors in adult female rats, we examined the time course of FES tumor uptake and metabolism and looked for correlations between metrics of blood volume and blood flow in tumors with ER levels in tumors and in the uterus; these relationships proved to be complex and difficult to interpret [49]. On the other hand, a dose–response study of FES in immature female rats was very illuminating. The uptake of FES in ER-rich tissues (uterus and ovary) became saturated only at the higher mass dose levels, whereas tissues with lower ER levels, such as muscle and esophagus, showed low but clearly measurable levels of specific uptake of FES activity that was saturated at much lower mass dose levels [50]. To us, this indicated that while the uptake of FES in very ER-rich target organs might be “flow limited” and not proportional to high ER levels [31], the specific uptake by the lower levels of ER found in breast tumors were likely to be proportional to receptor levels [40]. In 1988, we published the first PET images of ER-positive breast tumors in women [51], and one image, presented in advance of our publication at the 1987 national meeting of the Society of Nuclear Medicine, was highlighted by Professor Henry Wagner as “Image of the Year” (Figure6)[ 52]. Because it was the first PET image of a receptor in cancer, it represented a milestone in the development of receptor-targeted radiotracers for PET imaging. A discussion of the clinical utility of FES-PET imaging CancersER in 2020 breast, 12, cancer, x as well as that for FDHT-PET and FFNP-PET, is given in Section5. 7 of 32

Figure 6. FES-PET of a breast tumor in a woman with breast cancer. Image of the Year at the Figure 6. FES-PET of a breast tumor in a woman with breast cancer. Image of the Year at the 1987 1987 National Meeting of the Society of Nuclear Medicine. (Image provided by Drs. Michael Welch National Meeting of the Society of Nuclear Medicine. (Image provided by Drs. Michael Welch and and Barry Siegel, Washington University Medical School; discussed (but not shown) in reference [52].) Barry Siegel, Washington University Medical School; discussed (but not shown) in reference [52].) The focus of this review on PET imaging of steroid receptors in cancer, using radiofluorinated probes,The does focus not of providethis review an adequate on PET imaging opportunity of steroid to cover receptors other ER in probecancer, development using radiofluorinated and clinical probes,imaging does work not using provide radioiodinated an adequate steroids. opportunity Some ofto thesecover have other very ER goodprobe ER development binding affi nity,and showedclinical imagingpromise inwork biodistribution using radioiodinated studies in rodents steroids. [20 Some] and proceededof these have into interestingvery good clinicalER binding studies affinity, [53,54]. showedDetails canpromise be found in biodistribution in the references studies cited in and rode [55nts]. [20] The biodistributionand proceeded ofinto some interesting radioiodinated clinical studiesnonsteroidal [53,54]. estrogens Details alsocan looksbe found very in promising the references [56], and cited some and work [55]. has The been biodistribution done to advance of some these radioiodinatedand related agents nonsteroidal for both estrogens imaging and also targetedlooks veryin vivopromisingradiotherapy [56], and in some preclinical work has breast been cancer done tomodels advance [57 ,these58]. Note and isrelated made agents of other for recent both reviewsimaging covering and targeted the development in vivo radiotherapy of PET imaging in preclinical agents breastfor ER cancer [47,59]. models [57,58]. Note is made of other recent reviews covering the development of PET imaging agents for ER [47,59]. 3. FFNP, a PET Imaging Agent for Progesterone Receptors in Breast Cancers 3. FFNP, a PET Imaging Agent for Progesterone Receptors in Breast Cancers 3.1. Bumps on the Road to the Discovery of a PET Radiotracer for PgR in Breast Cancer 3.1. Bumps on the Road to the Discovery of a PET Radiotracer for PgR in Breast Cancer Early reports of PET imaging agents for the progesterone receptor (PgR), published in the 1970s and 1980s,Early reports described of PET the synthesisimaging agents of 21-[ for18F]fluoroprogesterone, the progesterone receptor a direct (PgR), analog published of the endogenousin the 1970s andprogestational 1980s, described hormone, the progesterone, synthesis of 21-[ as well18F]fluoroprogesterone, as some analogs of this a compounddirect analog and of ones the substitutedendogenous at progestationaldifferent sites; however,hormone, no progeste PgR bindingrone, was as reported, well as andsome biodistribution analogs of andthis metabolism compound studies and ones were substitutedlimited and generallyat different disappointing sites; however, [60–62 no]. High-aPgR bindingffinity radioiodinated was reported, ligands and biodistribution for PgR, reported and at metabolismthe same time, studies appeared were more limited promising, and generally especially disappointing (Z)-17β-hydroxy-17 [60–62]. αHigh-affinity-(2-iodovinyl)-4-estren-3-one radioiodinated ligands for PgR, reported at the same time, appeared more promising, especially (Z)-17β-hydroxy- 17α-(2-iodovinyl)-4-estren-3-one and the bromo analog [63–67], and these showed some evidence of PgR-specific uptake in the uterus of estrogen-primed immature rats [24,68]. We began our development of PgR PET probes knowing that progesterone itself is not a very high- affinity ligand for PgR (Kd = 8 nM) [6,69]. Hence, we focused on two much higher affinity ligands that had been developed by pharmaceutical companies, the Roussel compound R5020 (Kd = 1 nM) [70] and the Organon compound ORG2058 (Kd = 0.5 nM) [71] (Figure 7). As will be noted below, to some degree, structural changes that boost the binding affinity of androgens for AR also worked for the progestins: ORG2058 is a 19-nor analog of progesterone with an additional 16α-ethyl group, and R5020 has an extended A,B-ring dienone with flanking methyl groups at the 17α and 21 positions [6,69].

Endogenous Progestin Synthetic Progestins

CH3 OH O O

CH3 H CH3

O O

Progesterone (Kd = 8 nM) R5020 (Kd = 1 nM) ORG2058 (Kd = 0.5 nM) (RBA = 13; Poor BioD) (RBA = 100; Good BioD) (RBA = 200; Better BioD)

Figure 7. Natural and synthetic ligands for the progesterone receptor (PgR). Binding affinities and biodistribution (BioD) properties. Cancers 2020, 12, x 7 of 32

Figure 6. FES-PET of a breast tumor in a woman with breast cancer. Image of the Year at the 1987 National Meeting of the Society of Nuclear Medicine. (Image provided by Drs. Michael Welch and Barry Siegel, Washington University Medical School; discussed (but not shown) in reference [52].)

The focus of this review on PET imaging of steroid receptors in cancer, using radiofluorinated probes, does not provide an adequate opportunity to cover other ER probe development and clinical imaging work using radioiodinated steroids. Some of these have very good ER binding affinity, showed promise in biodistribution studies in rodents [20] and proceeded into interesting clinical studies [53,54]. Details can be found in the references cited and [55]. The biodistribution of some radioiodinated nonsteroidal estrogens also looks very promising [56], and some work has been done to advance these and related agents for both imaging and targeted in vivo radiotherapy in preclinical breast cancer models [57,58]. Note is made of other recent reviews covering the development of PET imaging agents for ER [47,59].

3. FFNP, a PET Imaging Agent for Progesterone Receptors in Breast Cancers

3.1. Bumps on the Road to the Discovery of a PET Radiotracer for PgR in Breast Cancer Early reports of PET imaging agents for the progesterone receptor (PgR), published in the 1970s and 1980s, described the synthesis of 21-[18F]fluoroprogesterone, a direct analog of the endogenous Cancersprogestational2020, 12, 2020 hormone, progesterone, as well as some analogs of this compound and 8ones of 32 substituted at different sites; however, no PgR binding was reported, and biodistribution and metabolism studies were limited and generally disappointing [60–62]. High-affinity radioiodinated ligandsand the for bromo PgR, analog reported [63– at67 ],the and same these time, showed appeared some more evidence promising, of PgR-specific especially uptake (Z)-17 inβ the-hydroxy- uterus 17ofα estrogen-primed-(2-iodovinyl)-4-estren-3-one immature rats and [24 ,the68]. bromo analog [63–67], and these showed some evidence of PgR-specificWe began uptake our developmentin the uterus of es PgRtrogen-primed PET probes immature knowing thatrats progesterone[24,68]. itself is not a very high-aWeffi begannity ligand our development for PgR (K dof= PgR8 nM) PET [probes6,69]. Hence,knowing we that focused progesterone on two itself much is not higher a very affi high-nity ligands that had been developed by pharmaceutical companies, the Roussel compound R5020 (K affinity ligand for PgR (Kd = 8 nM) [6,69]. Hence, we focused on two much higher affinity ligands thatd = 1 nM) [70] and the Organon compound ORG2058 (K = 0.5 nM) [71] (Figure7). As will be noted had been developed by pharmaceutical companies, the Rousseld compound R5020 (Kd = 1 nM) [70] and below, to some degree, structural changes that boost the binding affinity of androgens for AR also the Organon compound ORG2058 (Kd = 0.5 nM) [71] (Figure 7). As will be noted below, to some degree, workedstructural for changes the progestins: that boost ORG2058 the binding is a affinity 19-nor analogof androgens of progesterone for AR also with worked an additional for the progestins: 16α-ethyl ORG2058group, and is R5020a 19-nor has analog an extended of progesterone A,B-ring with dienone an additional with flanking 16α-ethyl methyl group, groups and at R5020 the 17 hasα and an extended21 positions A,B-ring [6,69]. dienone with flanking methyl groups at the 17α and 21 positions [6,69].

Endogenous Progestin Synthetic Progestins

CH3 OH O O

CH3 H CH3

O O

Progesterone (Kd = 8 nM) R5020 (Kd = 1 nM) ORG2058 (Kd = 0.5 nM) (RBA = 13; Poor BioD) (RBA = 100; Good BioD) (RBA = 200; Better BioD)

Figure 7. Natural and synthetic ligands fo forr the progesterone receptor (PgR).(PgR). Binding aaffinitiesffinities and biodistribution (BioD) properties. Cancers 2020, 12, x 8 of 32 Because these high-affinity PgR ligands were widely used as radiotracers in competitive binding assays,Because they werethese available high-affinity in tritium-labeled PgR ligands were form. widely R5020, used the as better-known radiotracers in and competitive generally binding favored assays,tracer, wasthey assigned were available a relative in tritium-labeled binding affinity form. (RBA) R5020, of 100, the which better-known gave ORG2058 and generally a higher favored RBA of tracer,200. In was biodistribution assigned a relative studies, binding both tritium-labeled affinity (RBA) progestins of 100, which showed gave high, ORG2058 specific a (i.e.,higher blockable) RBA of 200.uptake In biodistribution in the uterus of studies, immature both rats tritium-labeled that had been progestins primed with showed estradiol high, tospecific increase (i.e., the blockable) levels of uptakeuterine in PgR the [72 uterus]. Notably, of immature due to itsrats low that binding had been affinity, primed tritium-labeled with estradiol progesterone to increase showed the levels poor of uterinePgR-selective PgR [72]. uptake Notably, by the due uterus to its [ 72low]. binding affinity, tritium-labeled progesterone showed poor PgR-selectiveThe 21-hydroxy uptake groupby the inuterus ORG2058 [72]. was an obvious site for fluorine substitution and had already beenThe explored 21-hydroxy by researchers group in atORG2058 Organon was [73 an]. Fluorine-18obvious site labelingfor fluorine by fluoridesubstitution ion displacementand had already of beenthe corresponding explored by researchers triflate gave at 21-fluoro-16Organon [73].α-ethyl-nor-progesterone Fluorine-18 labeling by (FENP)fluoride (Figure ion displacement8), which bound of the correspondingto PgR 3-times triflate better gave than 21-fluoro-16 ORG2058 andα-ethyl-nor-progesterone 60-times better than (FENP) progesterone, (Figure giving8), which it abound remarkably to PgR 3-timeshigh binding better than affinity ORG2058 (RBA and 700; 60-times Ki = 0.14 better nM) than [74 progesterone,]. While there giving was it a no remarkably such obvious high binding site for affinityfluorine (RBA substitution 700; Ki = in 0.14 R5020, nM) two[74]. epimeric While there 21-hydroxy was no such metabolites obvious ofsite R5020 for fluorine were known substitution and still in R5020,had reasonably two epimeric good 21-hydroxy binding ametabolitesffinities [75 of]. R5020 Fluorine were for known hydroxyl and still substitution had reasonably at these good sites binding gave affinitiesepimeric [75]. products Fluorine having for morehydroxyl modest substitution binding at affi thesenities sites (RBA gave values epimeric of 11 andproducts 45) (Figure having8 )[more75]. Allmodest three binding of these affinities products (RBA showed values effi ofcient 11 and and 45) specific (Figure uptake 8) [75]. in All the three uterus of of these estrogen-primed products showed rats. efficientNot surprisingly, and specific however, uptake thein the uptake uterus of of the estrogen-primed very high affinity rats. FENP Not wassurprisingly, clearly the however, most effi thecient uptake and ofselective the very [74 high,75]. affinity FENP was clearly the most efficient and selective [74,75].

R5020 Analogs ORG2058 Analog

X, Y = H, F or F, H (RBA = 11, 45) FENP (RBA = 700)

Figure 8. 8. Fluorine-18Fluorine-18 labeled labeled analogs analogs of the of high the affi highnity affi PgRnity ligands, PgR ligands, R5020, and R5020, ORG2058. and ORG2058. Binding affinitiesBinding aareffinities relative are to relative R5020, to set R5020, at 100. set at 100.

Encouraged by these promising preclinical studies, we tried to use FENP to image ER and PgR- positive breast tumors in women and were most surprised to find that there was little tumor uptake at all [76]. At the same time, a group in the Netherlands, who were also investigating FENP and related compounds for PgR imaging, had the same experience [77,78]. When the Netherlands group examined FENP metabolism in humans, they found that a very active 20-hydroxysteroid hydrogenase rapidly converts circulating FFNP to the low affinity 20-hydroxy metabolite, presumed to be the 20α epimer (Figure 9) [79]. In humans, this active dehydrogenase appears to be in blood cells, but curiously, it is not present in rodents and thus, did not undermine the very favorable biodistribution of FENP in our preclinical models [77,80]. Clearly, this marked and unexpected species difference in steroid metabolism made FENP unsuitable for imaging breast cancer in humans [76]. In some ways, the finding of compounds that were promising in rodent models proved most unsuitable for imaging in humans was reminiscent of our experience with the behavior of certain modified FES [18] and FDHT analogs [17], whereas in those cases for a different reason: lack of sufficient binding affinity for SHBG.

FENP 20-Dihydro-FENP High Affinity Very Low Affinity

Figure 9. Rapid metabolic inactivation of FENP in humans (but not in rats), presumed due to a very active 20-hydroxysteroid dehydrogenase in human blood, but absent from rodents. Cancers 2020, 12, x 8 of 32

Because these high-affinity PgR ligands were widely used as radiotracers in competitive binding assays, they were available in tritium-labeled form. R5020, the better-known and generally favored tracer, was assigned a relative binding affinity (RBA) of 100, which gave ORG2058 a higher RBA of 200. In biodistribution studies, both tritium-labeled progestins showed high, specific (i.e., blockable) uptake in the uterus of immature rats that had been primed with estradiol to increase the levels of uterine PgR [72]. Notably, due to its low binding affinity, tritium-labeled progesterone showed poor PgR-selective uptake by the uterus [72]. The 21-hydroxy group in ORG2058 was an obvious site for fluorine substitution and had already been explored by researchers at Organon [73]. Fluorine-18 labeling by fluoride ion displacement of the corresponding triflate gave 21-fluoro-16α-ethyl-nor-progesterone (FENP) (Figure 8), which bound to PgR 3-times better than ORG2058 and 60-times better than progesterone, giving it a remarkably high binding affinity (RBA 700; Ki = 0.14 nM) [74]. While there was no such obvious site for fluorine substitution in R5020, two epimeric 21-hydroxy metabolites of R5020 were known and still had reasonably good binding affinities [75]. Fluorine for hydroxyl substitution at these sites gave epimeric products having more modest binding affinities (RBA values of 11 and 45) (Figure 8) [75]. All three of these products showed efficient and specific uptake in the uterus of estrogen-primed rats. Not surprisingly, however, the uptake of the very high affinity FENP was clearly the most efficient and selective [74,75].

R5020 Analogs ORG2058 Analog

X, Y = H, F or F, H (RBA = 11, 45) FENP (RBA = 700)

Figure 8. Fluorine-18 labeled analogs of the high affinity PgR ligands, R5020, and ORG2058. Binding Cancersaffinities2020, 12, 2020are relative to R5020, set at 100. 9 of 32

Encouraged by these promising preclinical studies, we tried to use FENP to image ER and PgR- positiveEncouraged breast tumors by these in women promising and were preclinical most su studies,rprised weto find tried that to there use FENP was little to image tumor ER uptake and PgR-positiveat all [76]. At breastthe same tumors time, in a women group in and the were Netherlands, most surprised who were to find also that investigating there was little FENP tumor and uptakerelated atcompounds all [76]. At for the PgR same imaging, time, a had group the in same the Netherlands,experience [77,78]. who were When also the investigating Netherlands FENPgroup andexamined related FENP compounds metabolism for PgR in imaging, humans, had they the samefound experience that a very [77,78 active]. When 20-hydroxysteroid the Netherlands grouphydrogenase examined rapidly FENP converts metabolism circulating in humans, FFNP to theythe low found affinity that 20-hydroxy a very active me 20-hydroxysteroidtabolite, presumed hydrogenaseto be the 20α rapidly epimer converts (Figure circulating9) [79]. In FFNPhumans, to the this low active affinity dehydrogenase 20-hydroxy metabolite, appears to presumed be in blood to becells, the but 20α curiously,epimer (Figure it is 9not)[ 79present]. In humans, in rodents this activeand thus, dehydrogenase did not undermine appears tothe be very in blood favorable cells, butbiodistribution curiously, it isof not FENP present in inour rodents preclinical and thus, models did not[77,80]. undermine Clearly, the this very marked favorable and biodistribution unexpected ofspecies FENP difference in our preclinical in steroid models metabolism [77,80 ].made Clearly, FENP this unsuitable marked and for imaging unexpected breast species cancer di ffinerence humans in steroid[76]. In metabolismsome ways, madethe finding FENP unsuitableof compounds for imagingthat were breast promising cancer in in rodent humans models [76]. In proved some ways, most theunsuitable finding offor compounds imaging in thathumans were was promising reminiscent in rodent of our models experience proved with most the unsuitable behavior for of imaging certain inmodified humans FES was [18] reminiscent and FDHT of our analogs experience [17], withwhereas the behavior in those of cases certain for modified a different FES reason: [18] and lack FDHT of analogssufficient [17 binding], whereas affinity in those for SHBG. cases for a different reason: lack of sufficient binding affinity for SHBG.

FENP 20-Dihydro-FENP High Affinity Very Low Affinity

FigureFigure 9.9. RapidRapid metabolicmetabolic inactivation of FENP in humans (but not in rats),rats), presumedpresumed due toto aa veryvery activeactive 20-hydroxysteroid20-hydroxysteroid dehydrogenasedehydrogenase inin humanhuman blood,blood, butbut absentabsentfrom fromrodents. rodents. 3.2. Defeating a Metabolic Inactivation in Human Blood Led to the Discovery of FFNP as an Effective PET Imaging Agent for PgR in Breast Cancer From a number of early medicinal chemistry studies, we were aware that PgR was very tolerant of large substituents on the alpha face of the D-ring of ligands, and we found that researchers at Squibb in the 1960s had developed potent progestins having acetals and ketals appended to 16α,17α-dihydroxy progestins (Figure 10)[81,82]. We thought that these bulky substituents, situated directly adjacent to the metabolically vulnerable 20-keto group of progestins, might sterically impede their inactivation by the 20-hydroxysteroid dehydrogenase. Therefore, we undertook a systemic investigation of fluoro-19-norprogestins having various 16α,17α-dioxolane substituents (initially to make affinity labels for PgR) [83]. In one series, the fluorine substitution was on the para position of dioxolanes from benzaldehyde or acetophenone, with radiofluorination being possible by nucleophilic aromatic substitution on the corresponding p-nitro compound prior to dioxolane formation [7]. Alternatively, fluorine substitution at C21 could be done as a last step on the 21-triflate prepared from preformed dioxolanes of unsubstituted benzaldehyde or acetophenone. Some of these compounds were quite lipophilic, and their biodistribution selectivity was not very high [7]. To reduce lipophilicity and nonspecific binding, we turned to dioxolanes prepared from more polar arene carbonyl compounds, notably furfural or acetylfuran [80]. With all of these dioxolanes, the arene substituent can either be endo (folded under the D-ring) or exo (extended outward from the D-ring); these epimers could be well separated by chromatography, and the endo stereoisomer had much higher affinity [7,80]. Cancers 2020, 12, x 9 of 32 Cancers 2020, 12, x 9 of 32 3.2. Defeating a Metabolic Inactivation in Human Blood Led to the Discovery of FFNP as an Effective PET 3.2.Imaging Defeating Agent a forMetabolic PgR in InactivatiBreast Canceron in Human Blood Led to the Discovery of FFNP as an Effective PET Imaging Agent for PgR in Breast Cancer From a number of early medicinal chemistry studies, we were aware that PgR was very tolerant of largeFrom substituents a number on of theearly alpha medicinal face of chemistrythe D-ring studieof ligands,s, we andwere we aware found that that PgR researchers was very at tolerant Squibb ofin thelarge 1960s substituents had developed on the potentalpha face progestins of the D-ring having of acetals ligands, and and ketals we appendedfound that toresearchers 16α,17α-dihydroxy at Squibb inprogestins the 1960s (Figure had developed 10) [81,82]. potent We progestins thought that having thes acetalse bulky and substituents, ketals appended situated to directly16α,17α -dihydroxyadjacent to progestinsthe metabolically (Figure vulnerable 10) [81,82]. 20-keto We thought group thatof proges thesetins, bulky might substituents, sterically impedesituated their directly inactivation adjacent byto the metabolically20-hydroxysteroid vulnerable dehydrogenase. 20-keto group Therefore, of proges wetins, undertook might sterically a systemic impede investigation their inactivation of fluoro-19- by thenorprogestins 20-hydroxysteroid having various dehydrogenase. 16α,17α-dioxolane Therefore, substituents we undertook (initially a systemic to make investigation affinity labels of fluoro-19- for PgR) norprogestins[83]. In one series, having the variousfluorine 16 substitutionα,17α-dioxolane was on substituents the para position (initially of todioxolanes make affinity from labels benzaldehyde for PgR) [83].or acetophenone, In one series, withthe fluorine radiofluorin substitutionation being was onpossible the para by positionnucleophilic of dioxolanes aromatic fromsubstitution benzaldehyde on the orcorresponding acetophenone, p-nitro with compound radiofluorin prioration to being dioxolane possible formatio by nucleophilicn [7]. Alternatively, aromatic fluorine substitution substitution on the correspondingat C21 could bep-nitro done compound as a last priorstep toon dioxolane the 21-triflate formatio preparedn [7]. Alternatively, from preformed fluorine dioxolanes substitution of atunsubstituted C21 could benzaldehydebe done as a orlast ac etophenone.step on the Some21-triflate of these prepared compounds from werepreformed quite lipophilic,dioxolanes and of unsubstitutedtheir biodistribution benzaldehyde selectivity or was acetophenone. not very high Some [7]. Toof thesereduce comp lipophilicityounds were and quitenonspecific lipophilic, binding, and theirwe turned biodistribution to dioxolanes selectivity prepared was fromnot very more high polar [7]. areneTo reduce carbonyl lipophilicity compounds, and nonspecific notably furfural binding, or weacetylfuran turned to[80]. dioxolanes With all of prepared these dioxolanes, from more the polar arene arene substituent carbonyl can compounds, either be endo notably (folded furfural under the or acetylfuran [80]. With all of these dioxolanes, the arene substituent can either be endo (folded under the CancersD-ring)2020 or, 12exo, 2020 (extended outward from the D-ring); these epimers could be well separated10 of by 32 D-ring)chromatography, or exo (extended and the endo outward stereoisom fromer the had D-ring much); higher these affinityepimers [7,80]. could be well separated by chromatography, and the endo stereoisomer had much higher affinity [7,80].

R = H (acetal); CH3 (ketal) R = H (acetal); CH (ketal) 3 Figure 10. Compound 19-norprogestins with 16α,17,17α-dioxolanes from benzaldehydebenzaldehyde/acetophenone/acetophenone (left)Figure andand 10. furfuralfurfural/furanyl Compound/furanyl 19-norprogestins methyl methyl ketone ketone (right).with (right). 16 Theα The,17 stericα steric-dioxolanes bulk bulk near near from the the Cben C= zaldehyde/acetophenoneO= O group group was was intended intended to block(left)to block and its its metabolic furfural/furanyl metabolic inactivation inactivation methyl by ketoneby 20-hydroxysteroid 20-hydroxysteroid (right). The dehydrogenasesteric dehydrogenase bulk near action.the action. C = O group was intended to block its metabolic inactivation by 20-hydroxysteroid dehydrogenase action. Based on on the the PgR PgR binding binding affinities affinities and andcalculated calculated NSB values NSB values of the fluorine-19 of the fluorine-19 substituted substituted 16α,17α- α α 16dioxolanes,,17Based-dioxolanes, on we the selected PgR we binding a selectednumber affinities for a number fluorine-18 and calculated for fluorine-18radiolabeling NSB values radiolabeling and of in the vivo fluorine-19 biodistribution and in vivo substitutedbiodistribution studies. 16α All,17 αof- studies.dioxolanes,them gave All clear, we of selected themblockable gave a number uptake clear, for in blockable thefluorine-18 uterus uptake of radiolabeling our in adult the estrogen-primed uterus and in of vivo our biodistribution adult rat preclinical estrogen-primed studies. model All [80]. rat of preclinicalthemTheir gavetarget clear, model tissue blockable [uptake80]. Their uptakeselectivity target in tissue the(uterus/muscle uterus uptake of our selectivity and adult uterus/blood estrogen-primed (uterus/muscle ratios) rat andreflected preclinical uterus both/blood model their ratios) [80].PgR reflectedTheirbinding target affinities both tissue their and uptake PgR their binding selectivity lipophilicities. affi (uterus/musclenities In and fact, their the and lipophilicities.correlation uterus/blood between In ratios) fact, the the reflectedtarget/nontarget correlation both their between uptake PgR thebindingratios target and affinities/ nontargetthe binding and uptake theirselectivity lipophilicities. ratios index and (BSI) the Inbinding forfact, thes thee selectivity compoundscorrelation index betweenwas (BSI)much the for better target/nontarget these than compounds with just uptake their was muchratiosRBA values; and better the uptake than binding with in fatselectivity just also their correlated RBAindex values; (BSI) nicely for uptake with thes theire in compounds fat NSB also values correlated was (Figure much nicely 11) better [80]. with than This their withhighlights NSB just values their the (FigureRBAvalue values; of 11 considering)[ 80uptake]. This in target-specific highlights fat also correlated the vs value. nonspecific nicely of considering with binding their NSB target-specific affinity, values embodied (Figure vs. nonspecific 11)in the [80]. BSI This value, binding highlights as a a ffiusefulnity, the embodiedvaluepredictor of considering of in in thevivo BSI uptake target-specific value, in astarget a usefulvs tissues. nonspecific predictor (see Sections binding of in 1 vivo andaffinity, 2.1)uptake [40].embodied in target in the tissues BSI value, (see Sectionsas a useful1 andpredictor 2.1)[ 40of ].in vivo uptake in target tissues (see Sections 1 and 2.1) [40].

Figure 11. Correlations between (A) binding selectivity index (BSI) and PgR-specific uterine uptake, and (B) nonspecific binding (NSB) and uptake into fat for five different PgR imaging agents. (Adapted and reprinted with permission from reference [80]. Copyright 1995, American Chemical Society.)

The results from these extensive biodistribution studies indicated that the more hydrophilic acetals from furfural had the most favorable combination of high PgR-mediated target tissue uptake and low nonspecific binding, the endo isomer being the best, as expected. For convenience, we named this compound FFNP (for 21-[18F]Fluoro-16α,17α-endo-Furanyl-19-Nor-Progesterone) (Figure 12A). In more recent preclinical imaging studies in mice, FFNP showed efficient and selective PgR-mediated uptake in a syngeneic mouse mammary cancer system [84,85]. Clinical studies with FFNP in breast cancer are discussed in Section 5.2. Clinical studies with FFNP will likely be facilitated by an expedited radiosynthetic method reported recently [86]. An interesting comparison can be made between the PgR binding affinities and biodistribution properties of FFNP and FENP (Figure 12B): FENP has nearly fourfold higher binding affinity, yet the uptake of FFNP by the uterus of estrogen-primed immature rats was as good as that for FENP in terms of %ID/g and nearly as good in terms of tissue to muscle ratio (T/M) and blocking by a high dose of unlabeled compound; of course, irrespective of its high PgR binding affinity, FENP failed rather dramatically for clinical PgR PET imaging in vivo, because of its rapid metabolic inactivation by the 20-hydroxysteroid dehydrogenase activity found in humans (Figure9). Cancers 2020, 12, x 10 of 32

Figure 11. Correlations between (A) binding selectivity index (BSI) and PgR-specific uterine uptake, and (B) nonspecific binding (NSB) and uptake into fat for five different PgR imaging agents. (Adapted and reprinted with permission from reference [80]. Copyright 1995, American Chemical Society.)

The results from these extensive biodistribution studies indicated that the more hydrophilic acetals from furfural had the most favorable combination of high PgR-mediated target tissue uptake and low nonspecific binding, the endo isomer being the best, as expected. For convenience, we named this compound FFNP (for 21-[18F]Fluoro-16α,17α-endo-Furanyl-19-Nor-Progesterone) (Figure 12A). In more recent preclinical imaging studies in mice, FFNP showed efficient and selective PgR- mediated uptake in a syngeneic mouse mammary cancer system [84,85]. Clinical studies with FFNP in breast cancer are discussed in Section 5.2. Clinical studies with FFNP will likely be facilitated by an expedited radiosynthetic method reported recently [86]. An interesting comparison can be made between the PgR binding affinities and biodistribution properties of FFNP and FENP (Figure 12B): FENP has nearly fourfold higher binding affinity, yet the uptake of FFNP by the uterus of estrogen- primed immature rats was as good as that for FENP in terms of %ID/g and nearly as good in terms of tissue to muscle ratio (T/M) and blocking by a high dose of unlabeled compound; of course, irrespective of its high PgR binding affinity, FENP failed rather dramatically for clinical PgR PET imagingCancers 2020 in, 12vivo,, 2020 because of its rapid metabolic inactivation by the 20-hydroxysteroid dehydrogenase11 of 32 activity found in humans (Figure 9).

18 A R-SO2-O F O H O H O 18 O n-Bu4 [ F]F H O O H O O

O O Triflate or Methanesulfonate FFNP (RBA = 190) Precursor

B FENP FFNP % ID/g [blocked] 6.4 [0.4] 6.8 [1.7] T/M [blocked] 16.0 [1.6] 11.6 [2.3] RBA 700 190

FigureFigure 12. ((AA)) Synthesis Synthesis of of FFNP, 21-[21-[1818F]fluoro-furanyl-nor-progesterone.F]fluoro-furanyl-nor-progesterone. (B (B) )Comparison Comparison of of the the bindingbinding and and biodistribution biodistribution properties properties of of FFNP FFNP vs. vs. FENP.

3.3.3.3. Alternative Alternative Approaches Approaches and and Efforts Efforts to to Improve Improve PET PET Imaging Imaging Agents Agents for for PgR WhileWhile FFNP hashas very very good good binding bindin agffi affinitynity for PgR,for PgR, it also it bindsalso binds well to well the glucocorticoidto the glucocorticoid receptor receptor(GR) [87 ].(GR) GR is[87]. a challenging GR is a challenging target for PET target imaging for PET [88]; imaging however, [88]; because however, there are because significant there levels are significantof GR in many levels breast of GR cancers in many [89 breast], we cancers had concerns [89], we that had GR concerns binding that by GR FFNP binding might by confound FFNP might PET confoundimaging of PET PgR imaging in breast of cancer. PgR in Assurancesbreast cancer. that As thissurances might that not bethis a problemmight not came be a from problem preclinical came fromstudies preclinical in which studies we found in thatwhich FFNP we found uptake that in syngeneic FFNP uptake murine in syngeneic mammary murine tumors mammary could be effi tumorsciently blocked by coadministration of a blocking dose of the specific PgR ligand, R5020, but not by the specific couldCancers be 2020 efficiently, 12, x blocked by coadministration of a blocking dose of the specific PgR ligand, R5020,11 of 32 butGR not ligand, by the dexamethasone specific GR ligand, [84]. Nevertheless, dexamethasone we [84]. were Nevertheless, curious whether we wewere might curious be ablewhether to avoid we mightderivativesany issues be able with have to oavoid ffbeen-target any investigated, GR issues binding with but by off-target developing showed GR lim abindingited PgR radiotracerpromise by developing for having imaging a PgR very PgR radiotracer low in a ffivivonity having[92,93]. for GR, veryHence,by basing low at affinity this them point, on for tanaproget, GR,FFNP by is basing the a most nonsteroidal them advanced on tanapr ligand PEToget, withimaging a nonsteroidal high agent PgR for but ligand PgR very in lowwith breast GR high cancer binding PgR butpatients, affi verynity low(Figurewhere GR it 13binding appearsA) [90 ].affinity to have (Figure good selectivity 13A) [90]. for PgR (see Section 5.2). We prepared a number of fluorine-substituted tanaproget analogs with high PgR and low GR binding affinities [91]; we preparedA one of them, fluoropropyl tanaproget, FPTP (PgR RBA 189 and GR RBA 0.9 [dexamethasone, standard, RBA 100]) (Figure 13B), in fluorine-18 labeled form [87]. FPTP showed excellent uptake in the uterus of estrogen-primed rats that was PgR-mediated and equivalent to that of FENP and FFNP, and its activityDexamethasone in bone was considerablyTanaproget less than that of these two [87]. Based on its structure, FPTP should be unaffectedGR = 100 by the 20-hydroxysteroidGR = low hydrogenase activity in PgR = 12 PgR = 150 humans that undermined the utility of FENP [79]. So, FPTP is an obvious candidate for clinical translation, but given that FFNP was already being advanced into clinical studies by our groups (NCT02455453) [4], we madeB no further efforts to develop FPTP. Other fluorine-18 tanaproget

FFNP FPTP GR = 365 GR = 0.9 PgR = 190 PgR = 189

Figure 13. ((AA)) Nonsteroidal Nonsteroidal PgR PgR ligand, ligand, tanaproget, tanaproget, compared compared to glucocorticoid receptor (GR) ligand dexamethasone. ( B)) Comparison Comparison of FFNP and FPTP RBA to GR vs. PgR.

SomeWe prepared iodine a numberand bromine-substituted of fluorine-substituted analogs tanaproget of 16 analogsα,17α-dihydroxy with high PgRnorprogesterone and low GR acetophenonebinding affinities dioxolanes [91]; we preparedwere prepared one of them,[94]. Th fluoropropyley have good tanaproget, PgR binding FPTP (PgRaffinity RBA and 189 show and efficientGR RBA and 0.9 selective [dexamethasone, PgR-mediated standard, uptake RBA in 100])vivo; with (Figure alternative 13B), in radionuclides fluorine-18 labeled (iodine-123 form [and87]. bromine-76/77),FPTP showed excellent they might uptake be useful in the for uterus radiotherape of estrogen-primedutic applications rats [94]. that Note was PgR-mediatedis made of another and recentequivalent review to thatcovering of FENP the anddevelopmen FFNP, andt of its PET activity imaging in bone agents was for considerably PgR [59]. less than that of these two [Within87]. Based the superfamily on its structure, of nuclear FPTP hormone should be recept unafforsected [95], by PgR the is 20-hydroxysteroid a member of the 3C hydrogenase subfamily, whichactivity includes in humans AR that (see undermined Section 4 below), the utility as ofwell FENP as the [79 ].glucocorticoid So, FPTP is an receptor obvious (GR) candidate and the for mineralocorticoidclinical translation, receptor but given (MR). that Although FFNP was less alreadythoroughly being investigated advanced than into the clinical three studies receptors, by ourER, PgR,groups and (NCT02455453) AR, discussed [ 4here], we in made detail, no GR further and MR efforts certainly to develop play FPTP.roles in Other various fluorine-18 cancers, tanaproget including breastderivatives [89,96,97] have and been possibly investigated, prostate but cancers. showed Hence, limited it is promiseworth noting for imaging that some PgR effortsin vivo have[92 been,93]. made to develop imaging agents for these nuclear receptors: for GR [88,98–101] and for MR [88]. Limited in vivo studies using as endogenous targets for GR and MR, on liver and kidney of adrenalectomized adult male rats, respectively, were not promising [88]. Unlike ER, PgR, and AR, GR and MR need to respond physiologically to rapidly changing levels of their cognate corticosteroids, cortisol and aldosterone, so they tend not to retain their receptor-bound ligands as is needed for image contrast to develop. Thus, they are likely to be very challenging targets for PET imaging and will require particularly high-performing radiotracers.

4. FDHT, a PET Imaging Agent for Androgen Receptors in Prostate Cancers

4.1. A Prelude to In Vivo Imaging of AR: Optimizing the Binding Characteristics and Pharmacokinetic Properties of Radiotracers Leading to FDHT Some early reports described radiobrominated and radioiodinated androgens for imaging AR in prostate cancer, but their AR binding was poor or not characterized, they were chemically or metabolically unstable, or they did not show clear evidence of AR-selective uptake in rodent prostate [102–105]. More recent work on radioiodinated androgens was more promising, and those with 7α- iodo or 17α-iodovinyl groups bound well to AR, although those lacking a C19 methyl group (4- estren-3-one core) begin to resemble synthetic progestins and risked having considerable affinity for PgR. In vivo biodistribution studies of uptake of these compounds in rat prostate were either limited or largely unpromising [24,64,66,68,106,107]. Because of our success with FES and the increasing availability of PET imaging, early on we focused our development of AR PET imaging agents on steroids labeled with fluorine-18. Prior to undertaking the preparation of candidate AR-targeting probes, we conducted a review of what was Cancers 2020, 12, 2020 12 of 32

Hence, at this point, FFNP is the most advanced PET imaging agent for PgR in breast cancer patients, where it appears to have good selectivity for PgR (see Section 5.2). Some iodine and bromine-substituted analogs of 16α,17α-dihydroxy norprogesterone acetophenone dioxolanes were prepared [94]. They have good PgR binding affinity and show efficient and selective PgR-mediated uptake in vivo; with alternative radionuclides (iodine-123 and bromine-76/77), they might be useful for radiotherapeutic applications [94]. Note is made of another recent review covering the development of PET imaging agents for PgR [59]. Within the superfamily of nuclear hormone receptors [95], PgR is a member of the 3C subfamily, which includes AR (see Section4 below), as well as the glucocorticoid receptor (GR) and the mineralocorticoid receptor (MR). Although less thoroughly investigated than the three receptors, ER, PgR, and AR, discussed here in detail, GR and MR certainly play roles in various cancers, including breast [89,96,97] and possibly prostate cancers. Hence, it is worth noting that some efforts have been made to develop imaging agents for these nuclear receptors: for GR [88,98–101] and for MR [88]. Limited in vivo studies using as endogenous targets for GR and MR, on liver and kidney of adrenalectomized adult male rats, respectively, were not promising [88]. Unlike ER, PgR, and AR, GR and MR need to respond physiologically to rapidly changing levels of their cognate corticosteroids, cortisol and aldosterone, so they tend not to retain their receptor-bound ligands as is needed for image contrast to develop. Thus, they are likely to be very challenging targets for PET imaging and will require particularly high-performing radiotracers.

4. FDHT, a PET Imaging Agent for Androgen Receptors in Prostate Cancers

4.1. A Prelude to In Vivo Imaging of AR: Optimizing the Binding Characteristics and Pharmacokinetic Properties of Radiotracers Leading to FDHT Some early reports described radiobrominated and radioiodinated androgens for imaging AR in prostate cancer, but their AR binding was poor or not characterized, they were chemically or metabolically unstable, or they did not show clear evidence of AR-selective uptake in rodent prostate [102–105]. More recent work on radioiodinated androgens was more promising, and those with 7α-iodo or 17α-iodovinyl groups bound well to AR, although those lacking a C19 methyl group (4-estren-3-one core) begin to resemble synthetic progestins and risked having considerable affinity for PgR. In vivo biodistribution studies of uptake of these compounds in rat prostate were either limited or largely unpromising [24,64,66,68,106,107]. Because of our success with FES and the increasing availability of PET imaging, early on we focused our development of AR PET imaging agents on steroids labeled with fluorine-18. Prior to undertaking the preparation of candidate AR-targeting probes, we conducted a review of what was known in the literature about structure-activity relationships (SAR) of androgens [6]. (At the same time, we reviewed the SAR of progestins in preparation for our later investigations of ligands for PET imaging of the progesterone receptor (PgR) (see Section 3.1).) We supplemented what we learned from the literature on the SAR of AR ligands with some focused syntheses of fluorine-substituted analogs, the results of which were published in a review article [69]. As was the case with PgR, we noted that the AR binding affinity of testosterone, the principal circulating androgen, was not exceptionally high (Figure 14A). Compared to the subnanomolar affinity of estradiol for ER (Kd = 0.2 nM), the affinity of testosterone for AR was 10-fold lower (Kd = 2 nM), although the natural metabolite, 5α-dihydrotestosterone (DHT), generated in the prostate by the action of 5α-reductase, binds 10-fold better (Kd = 0.2 nM). In addition, many androgens are known to undergo very rapid metabolism, clearing at rates substantially greater than that of estrogens [6,69,108]. Considering both of these issues, we included in our studies some synthetic androgens known to have both very high binding affinity for AR and to be relatively resistant to metabolic clearance, such as mibolerone and methyltrienolone (Figure 14B). Cancers 2020, 12, x 12 of 32 known in the literature about structure-activity relationships (SAR) of androgens [6]. (At the same time, we reviewed the SAR of progestins in preparation for our later investigations of ligands for PET imaging of the progesterone receptor (PgR) (see Section 3.1).) We supplemented what we learned from the literature on the SAR of AR ligands with some focused syntheses of fluorine-substituted analogs, the results of which were published in a review article [69]. As was the case with PgR, we noted that the AR binding affinity of testosterone, the principal circulating androgen, was not exceptionally high (Figure 14A). Compared to the subnanomolar affinity of estradiol for ER (Kd = 0.2 nM), the affinity of testosterone for AR was 10-fold lower (Kd = 2 nM), although the natural metabolite, 5α-dihydrotestosterone (DHT), generated in the prostate by the action of 5α-reductase, binds 10-fold better (Kd = 0.2 nM). In addition, many androgens are known to undergo very rapid metabolism, clearing at rates substantially greater than that of estrogens [6,69,108]. Considering both of these issues, we included in our studies some synthetic androgens

Cancersknown2020 to, 12have, 2020 both very high binding affinity for AR and to be relatively resistant to metabolic13 of 32 clearance, such as mibolerone and methyltrienolone (Figure 14B).

ABEndogenous Androgens Synthetic Androgens

Testosterone (K = 2 nM)DHT (K = 0.2 nM) Various (K ~ 0.2 nM) d d d Figure 14. ((A)) Principal Principal endogenous endogenous androgens and ( B) high-affinity high-affinity synthetic androgens.

We prepared aa number of androgens (Figure 1515,, top) based on testosterone and four high affinity affinity ligand series: series: 5α 5α-dihydrotesterone-dihydrotesterone (DHT), (DHT), 19-nortestosterone 19-nortestosterone,, mibolerone, mibolerone, and and methyltrienolone methyltrienolone (the (thelast, last,a compound a compound developed developed by Roussel, by Roussel, also named also named R1881). R1881). As with Asthe withestrog theens, estrogens, binding affinities binding aareffi nitiesgiven areon a given percent on scale a percent relative scale to relativethat of the to thatmost of commonly the most used commonly radiotracer used for radiotracer competitive for competitiveradiometric radiometricbinding assays, binding methyltrienolone, assays, methyltrienolone, which was whichassigned was an assigned RBA value an RBA of 100 value (Kd of= 0.2 100 nM) (Kd =[6].0.2 Both nM) reduction [6]. Both reductionof the 4,5-doub of thele 4,5-double bond in testosterone bond in testosterone (which occurs (which endogenously occurs endogenously by the byaction the actionof 5α-dihydroreductase of 5α-dihydroreductase in the in prostate the prostate)) or removal or removal of its of its19 19methyl methyl group group (as (as a asynthetic synthetic analog) analog) increased binding aaffinityffinity by 5–10-fold. Strategic me methylthyl substitution in the analogs, mibolerone and methyltrienolone, aaffordedfforded bothboth highhigh binding aaffinityffinity and was intended to provide relative resistance toCancers metabolism, 2020, 12, x as did otherother structuralstructural alterations inin methyltrienolonemethyltrienolone [[6,69].6,69]. 13 of 32 Below each of the five parental structures are a selection of fluorine-substituted analogs we investigated, showing their binding affinities relative to that of methyltrienolone (Figure 15, bottom). In most cases, introduction of the fluorine could be done conveniently late in the synthesis, by fluoride ion displacement of a triflate precursor [109,110], or in one case, a spirocyclic sulfate precursor [109,111]. The only one synthesized by a different route was 11β-fluoro-5α- dihydrotestosterone, in which fluorine was introduced by an unusual bromo-fluorination reaction on a Δ9,(11) precursor, followed by rapid removal of a 9α-bromine by radical reduction with tributyltin hydride under forcing conditions [112]. Protecting groups that were sometimes needed could then be removed [109,111]. The fluorine-substitution and any needed deprotection reactions proceeded rapidly and under mild conditions, and was later readily adapted to radiolabeling with [18F]fluoride ion [109,111,112]. In most cases, fluorine substitution reduced binding affinity a few fold, and sometimes to a greater extent; nevertheless, several of the fluorine-substituted analogs still had subnanomolar binding affinities (RBA >20), and together with their low NSB values, their BSI values were quite reasonable.

Figure 15. Five androgen receptor (AR) ligand groups ( top) with fluorine-substituted fluorine-substituted analogs analogs below. below. Numbers are AR bindingbinding aaffinitiesffinities relativerelative toto methyltrienolonemethyltrienolone (R1881).(R1881). The three circled structuresstructures werewere selectedselected forfor furtherfurther study.study.

BelowBefore eachstudying of the the five biodistribution parental structures of all of are these a selection analogs in of a fluorine-substituted preclinical model, we analogs carefully we investigated,analyzed the showinglevel of AR their in bindingthe prostate affinities of adult relative male to rats that and of methyltrienolonethe degree to which (Figure it was 15 ,occupied bottom). Inby mostthe normally cases, introduction high circulating of the fluorine levels couldof testosterone. be done conveniently We did this late by in thecomparing synthesis, the by effect fluoride of ionreducing displacement endogenous of a triflate testosterone precursor levels [109 by,110 castrati], or inon one vs. case, by feedback a spirocyclic inhibi sulfatetion precursorfrom pretreatment [109,111]. Thewith onlyhigh one doses synthesized of an estrogen by a di (diethylstilfferent routebesterol, was 11 DES)β-fluoro-5 on theα-dihydrotestosterone, biodistribution of [3 inH]R1881, which 9,(11) fluorine[3H]testosterone, was introduced [3H]19-nortestosterone, by an unusual bromo-fluorination [3H]DHT, and [3H]mibolerone. reaction on In a ∆ all cases,precursor, we found followed robust byand rapid AR-selective removal uptake of a 9α in-bromine the prostate, by radical and reductionalthough castration with tributyltin more hydridemarkedly under reduced forcing AR conditionsoccupancy [by112 endogenous]. Protecting androgen groups thats, high-dose were sometimes estrogen neededpretreatme couldnt was then nearly be removed as effective [109,111 and]. Theproved fluorine-substitution much more convenient and any needed[113]. While deprotection some of reactions these proceededAR ligands rapidly (notably, and underR1881 mildand 18 conditions,mibolerone) and also was have later some readily affinity adapted for PgR, to radiolabeling the prostate with uptake [ F]fluoride of these ionfive [ 109tritiated,111,112 test]. Inligands most cases,was AR-specific fluorine substitution because it was reduced effectively binding blocked affinity by a fewsimultaneo fold, andus sometimesadministration to a of greater a saturating extent; dose of unlabeled DHT, but was unaffected by triamcinolone acetonide, which has high affinity for both PgR and GR [113]. When tested in the DES-pretreated adult rat model, most of our high affinity radiofluorinated compounds also showed marked uptake by the prostate that was effectively blocked by the AR ligand DHT, but not by triamcinolone acetonide [111–114]. At 2 h postinjection, prostate/muscle ratios for most compounds were typically 6–14, and were reduced by 60–80% after blocking with DHT; prostate/muscle ratios rose to as high as 25 at 4 h [17]. As typical for steroids, there was considerable activity in organs involved in metabolism and excretion (liver, kidney, and intestine) that was not blockable. In contrast to the behavior of the F-18 labeled estrogens, we found that some radiofluorinated compounds underwent rapid defluorination, evident by activity uptake in bone that was not reduced by a blocking dose of unlabeled DHT (Figure 16) [111–114]. Defluorination was greatest with 16β- fluorine substitution, presumably due to an active 16α-hydroxylase activity. Fortunately, we soon found that this potentially undesirable pathway was not observed in biodistribution studies in nonhuman primates [17], and this was later confirmed in FDHT imaging studies in humans [3,115]. Cancers 2020, 12, 2020 14 of 32 nevertheless, several of the fluorine-substituted analogs still had subnanomolar binding affinities (RBA >20), and together with their low NSB values, their BSI values were quite reasonable. Before studying the biodistribution of all of these analogs in a preclinical model, we carefully analyzed the level of AR in the prostate of adult male rats and the degree to which it was occupied by the normally high circulating levels of testosterone. We did this by comparing the effect of reducing endogenous testosterone levels by castration vs. by feedback inhibition from pretreatment with high doses of an estrogen (diethylstilbesterol, DES) on the biodistribution of [3H]R1881, [3H]testosterone, [3H]19-nortestosterone, [3H]DHT, and [3H]mibolerone. In all cases, we found robust and AR-selective uptake in the prostate, and although castration more markedly reduced AR occupancy by endogenous androgens, high-dose estrogen pretreatment was nearly as effective and proved much more convenient [113]. While some of these AR ligands (notably, R1881 and mibolerone) also have some affinity for PgR, the prostate uptake of these five tritiated test ligands was AR-specific because it was effectively blocked by simultaneous administration of a saturating dose of unlabeled DHT, but was unaffected by triamcinolone acetonide, which has high affinity for both PgR and GR [113]. When tested in the DES-pretreated adult rat model, most of our high affinity radiofluorinated compounds also showed marked uptake by the prostate that was effectively blocked by the AR ligand DHT, but not by triamcinolone acetonide [111–114]. At 2 h postinjection, prostate/muscle ratios for most compounds were typically 6–14, and were reduced by 60–80% after blocking with DHT; prostate/muscle ratios rose to as high as 25 at 4 h [17]. As typical for steroids, there was considerable activity in organs involved in metabolism and excretion (liver, kidney, and intestine) that was not blockable. In contrast to the behavior of the F-18 labeled estrogens, we found that some radiofluorinated compounds underwent rapid defluorination, evident by activity uptake in bone that was not reduced by a blocking dose of unlabeled DHT (Figure 16)[111–114]. Defluorination was greatest with 16β-fluorine substitution, presumably due to an active 16α-hydroxylase activity. Fortunately, we soon found that this potentially undesirable pathway was not observed in biodistribution studies in nonhuman primatesCancers 2020 [,17 12],, x and this was later confirmed in FDHT imaging studies in humans [3,115]. 14 of 32

AB

%ID/g Bone

βFDHT

C

Figure 16. MetabolicMetabolic defluorination defluorination of of various various fluorine-18 fluorine-18 labeled labeled AR AR ligands ligands in in rodent rodent models. Numbers are the %ID%ID/g/g bone activity. ( A, coral) ThoseThose havinghaving an exposedexposed 1616α-hydrogen-hydrogen undergo extensive defluorinationdefluorination andand have high bone uptake. (B, teal ) Those lacking a 16 α-hydrogen have less defluorination.defluorination. ( C) Proposed 16 α-hydroxylation results in loss ofof F-18.F-18.

18 We studied three of our promising compounds, 16β-[18F]fluoro-5F]fluoro-5αα-dihydrotestosterone-dihydrotestosterone (FDHT), (FDHT), 18 18 16β-[18F]fluoro-mibolerone,F]fluoro-mibolerone, and and 21-[ 21-[18F]fluoro-miboleroneF]fluoro-mibolerone (encircled (encircled in in Figure Figure 14),14), in a nonhuman primate model, a baboon [[17].17]. Al Alll three compoundscompounds had shownshown essentiallyessentially equivalent AR binding affinity and AR-specific uptake in the prostate of androgen-suppressed adult rats [111,114], but they had very different affinities for PgR (20-F-Mib was very high) and for the serum steroid carrier protein, sex steroid binding globulin (SHBG, FDHT was very high) (Figure 17). Quite to our surprise, FDHT gave clearly better and more selective uptake than the other two in baboon prostate [17]. We believe that this difference is due to the greater affinity that FDHT has for the blood steroid carrier protein, SHBG, than the mibolerone analogs [17,109,111]. SHBG is found in primates but not rodents. As we had noted during our development of PET probes for ER, some level of binding of FES to SHBG also seemed important for effective ER tracer distribution in vivo [18]. In addition, unlike the high bone activity found in rats due to metabolic defluorination (Figure 15), minimal bone activity was found in baboons [17]. Hence, FDHT was chosen as a compound for further studies in humans. The first FDHT-PET imaging of AR in men with advanced prostate cancer was published together with my colleagues at Washington University Medical School [3] and nearly simultaneously by investigators at Sloan Kettering Cancer Center [115]. In both studies, clear FDHT uptake was evident in numerous lesions, with uptake being blocked following treatment with AR antagonists. Further discussion of FDHT-PET imaging in prostate cancer and its potential utility in management of this disease are presented in Section 5.3.

OH F OH OH CH3 H H H F CH3 F

O O O CH3 CH3 H 20-Fluoromibolerone 16 -Fluoromibolerone 16 -Fluoro-5 -dihydrotestosterone (20-F-Mib) (16 -F-Mib) (FDHT) %ID/g (VP) = 0.97 %ID/g (VP) = 0.67 %ID/g (VP) = 0.57 VP/Muscle = 5.5 VP/Muscle = 3.8 VP/Muscle = 5.4 RBA (AR) = 53 RBA (AR) = 31 RBA (AR) = 43 RBA (PgR) = 44 RBA (PgR) = 3.0 RBA (PgR) = 0.12 RBA (SHBG) = 4 RBA (SHBG) = 1.3 RBA (SHBG) = 385

Figure 17. Three AR ligands evaluated in nonhuman primates. In androgen-suppressed adult male rats, %ID/g is at 1 h and ventral prostate (VP)/muscle ratio is at 4 h. RBA are relative to standards: for AR, R1881 = 100; for PgR, R5020 = 100; and for sex hormone binding globulin (SHBG), estradiol = 1. Cancers 2020, 12, x 14 of 32

AB

%ID/g Bone

βFDHT

C

Figure 16. Metabolic defluorination of various fluorine-18 labeled AR ligands in rodent models. Numbers are the %ID/g bone activity. (A, coral) Those having an exposed 16α-hydrogen undergo extensive defluorination and have high bone uptake. (B, teal) Those lacking a 16α-hydrogen have less defluorination. (C) Proposed 16α-hydroxylation results in loss of F-18.

18 CancersWe2020 studied, 12, 2020 three of our promising compounds, 16β-[ F]fluoro-5α-dihydrotestosterone (FDHT),15 of 32 16β-[18F]fluoro-mibolerone, and 21-[18F]fluoro-mibolerone (encircled in Figure 14), in a nonhuman primate model, a baboon [17]. All three compounds had shown essentially equivalent AR binding aaffinityffinity and AR-specificAR-specific uptake in the prostate of androgen-suppressed adult rats [[111,114],111,114], but they had veryvery didifferentfferent a ffiaffinitiesnities for for PgR PgR (20-F-Mib (20-F-Mib was was very very high) high) and forand the for serum the steroidserum steroid carrier protein,carrier sexprotein, steroid sex binding steroid binding globulin globul (SHBG,in (SHBG, FDHT was FDHT very was high) very (Figure high) (Figure17). Quite 17). to Quite our surprise,to our surprise, FDHT gaveFDHT clearly gave clearly better andbetter more and selective more selective uptake uptake than the than other the two other in baboontwo in baboon prostate prostate [17]. We [17]. believe We thatbelieve this that diff erencethis difference is due to is the due greater to the agreaterffinity thataffinity FDHT that has FDHT for thehas blood for the steroid blood carriersteroid protein, carrier SHBG,protein, than SHBG, the than mibolerone the mibolerone analogs analogs [17,109,111 [17,109,]. SHBG111]. is SHBG found is in found primates in primates but not but rodents. not rodents. As we hadAs we noted had during noted ourduring development our development of PET probes of PET for probes ER, some for ER, level some of binding level of of binding FES to SHBG of FES also to seemedSHBG also important seemed forimportant effective for ER effective tracer distribution ER tracer distributionin vivo [18 ].in In vivo addition, [18]. In unlike addition, the highunlike bone the activityhigh bone found activity in rats found due in to rats metabolic due to defluorinationmetabolic defluorination (Figure 15), (Figure minimal 15), bone minimal activity bone was activity found inwas baboons found in [17 baboons]. Hence, [17]. FDHT Hence, was FDHT chosen was as achosen compound as a compound for further for studies further in studies humans. in humans. The first FDHT-PETThe first FDHT-PET imaging ofimaging AR in menof AR with in men advanced with ad prostatevanced cancerprostate was cancer published was published together withtogether my colleagueswith my colleagues at Washington at Washington University MedicalUniversity School Medical [3] and School nearly [3] simultaneously and nearly simultaneously by investigators by at Sloaninvestigators Kettering at Sloan Cancer Kettering Center [Canc115].er In Center both studies, [115]. In clear both FDHTstudies, uptake clear FDHT was evident uptake in was numerous evident lesions,in numerous with lesions, uptake beingwith uptake blocked being following blocked treatment following with treatment AR antagonists. with AR antagonists. Further discussion Further ofdiscussion FDHT-PET of FDHT-PET imaging in imaging prostate cancerin prostate and cancer its potential and its utility potential in management utility in management of this disease of this are presenteddisease are in presented Section 5.3 in. Section 5.3.

OH F OH OH CH3 H H H F CH3 F

O O O CH3 CH3 H 20-Fluoromibolerone 16 -Fluoromibolerone 16 -Fluoro-5 -dihydrotestosterone (20-F-Mib) (16 -F-Mib) (FDHT) %ID/g (VP) = 0.97 %ID/g (VP) = 0.67 %ID/g (VP) = 0.57 VP/Muscle = 5.5 VP/Muscle = 3.8 VP/Muscle = 5.4 RBA (AR) = 53 RBA (AR) = 31 RBA (AR) = 43 RBA (PgR) = 44 RBA (PgR) = 3.0 RBA (PgR) = 0.12 RBA (SHBG) = 4 RBA (SHBG) = 1.3 RBA (SHBG) = 385

Figure 17.17. Three ARAR ligandsligands evaluatedevaluated inin nonhumannonhuman primates.primates. In androgen-suppressedandrogen-suppressed adult male rats, %ID%ID/g/g is at 1 h and ventral prostate (VP)(VP)/muscle/muscle ratioratio isis atat 44 h.h. RBA are relativerelative to standards: for AR, R1881 = 100; for PgR, R5020 = 100;100; and and for for sex sex hormone hormone binding binding globulin globulin (SHBG), (SHBG), estradiol estradiol = 1.1.

4.2. Opportunities and Efforts to Improve the Imaging Characteristics of FDHT in Prostate Cancer In early FDHT-PET imaging studies, a circulating lipophilic but somewhat more polar radiolabeled metabolite was noted [115], and although it did not bind to AR and was not taken up by prostate cancer lesions, it made accurate measurement of blood levels of FDHT for pharmacokinetic modeling more difficult [116]. Its almost immediate formation in blood suggests that rapid metabolism may also be limiting AR-mediated uptake of FDHT. Thus, it led us to consider further potential refinements of PET imaging agents for AR. A-ring saturated androgens like FDHT are known to undergo rapid reduction by 3α-hydroxysteroid dehydrogenases (3α-HSDs), converting the 3-ketone to the 3α-hydroxy analog, which has very poor AR binding affinity [6,69,108]. Though it has not been unequivocally verified, the labeled metabolite noted in the Sloan-Kettering studies [115,116] was likely the 3α-hydroxy analog of FDHT, 16β-[18F]fluoroandrostane-3α,17β-diol. The known routes of metabolism of different androgens by 3α-HSD and 5α-reductase (5α-R) are displayed in Figure 18, and this scheme illustrates for different steroids, the contrasting preference or resistance for these two pathways [69]. The 4,5-double bond in both testosterone and 19-nortestosterone makes them poor substrates for the 3α-HSD, which would make them relatively resistant to inactivation by this pathway. Testosterone, however, is a good substrate for 5α-R, which would eliminate the protecting 4,5-double bond. This would generate the higher affinity for 5α-dihydrotestosterone, which would be rapidly metabolized by 3α-HSDs; by contrast, 19-nortestosterones are poor 5α-R substrates. The boldness of arrows in the scheme illustrates the relative rates of these two potential routes of Cancers 2020, 12, x 15 of 32

4.2. Opportunities and Efforts to Improve the Imaging Characteristics of FDHT in Prostate Cancer In early FDHT-PET imaging studies, a circulating lipophilic but somewhat more polar radiolabeled metabolite was noted [115], and although it did not bind to AR and was not taken up by prostate cancer lesions, it made accurate measurement of blood levels of FDHT for pharmacokinetic modeling more difficult [116]. Its almost immediate formation in blood suggests that rapid metabolism may also be limiting AR-mediated uptake of FDHT. Thus, it led us to consider further potential refinements of PET imaging agents for AR. A-ring saturated androgens like FDHT are known to undergo rapid reduction by 3α- hydroxysteroid dehydrogenases (3α-HSDs), converting the 3-ketone to the 3α-hydroxy analog, which has very poor AR binding affinity [6,69,108]. Though it has not been unequivocally verified, the labeled metabolite noted in the Sloan-Kettering studies [115,116] was likely the 3α-hydroxy analog of FDHT, 16β-[18F]fluoroandrostane-3α,17β-diol. The known routes of metabolism of different androgens by 3α-HSD and 5α-reductase (5α-R) are displayed in Figure 18, and this scheme illustrates for different steroids, the contrasting preference or resistance for these two pathways [69]. The 4,5-double bond in both testosterone and 19- nortestosterone makes them poor substrates for the 3α-HSD, which would make them relatively resistant to inactivation by this pathway. Testosterone, however, is a good substrate for 5α-R, which would eliminate the protecting 4,5-double bond. This would generate the higher affinity for 5α- dihydrotestosterone,Cancers 2020, 12, 2020 which would be rapidly metabolized by 3α-HSDs; by contrast,16 of19- 32 nortestosterones are poor 5α-R substrates. The boldness of arrows in the scheme illustrates the relative rates of these two potential routes of metabolism, and one can see that FDHT, as a dihydrotestosteronemetabolism, and one cananalog, see that is FDHT,at risk as of a dihydrotestosteronerapid inactivation analog, by 3α is-HSD at risk activity, of rapid inactivationwhereas a correspondingby 3α-HSD activity, 19-nortestosterone whereas a corresponding tracer, which 19-nortestosterone is a poor substrate tracer, for which both 3 isα a-HSD poor and substrate 5α-R, for is likelyboth 3 toα-HSD have andmore 5α extended-R, is likely blood to have residence more extended and higher blood AR-mediated residence and target higher uptake. AR-mediated Thus, more target in- depthuptake. investigations Thus, more in-depth of F-18 investigationslabeled 19-nortestosterone of F-18 labeled and 19-nortestosterone methyltrienolone and analogs methyltrienolone might be productive.analogs might Inhibitors be productive. of both Inhibitors 3α-dehydrogenases of both 3α-dehydrogenases (flufenamic acid (flufenamic and others) acid and [117] others) and [1175α-] reductasesand 5α-reductases (dutosteride) (dutosteride) [118] are [118 available,] are available, and pretreatment and pretreatment with these with thesedrugs drugs might might slow slow the clearancethe clearance of these of these types types of tracers of tracers and enhance and enhance their theiruptake uptake efficiency efficiency and selectivity and selectivity and reduce and reduce non- ARnon-AR binding binding circulating circulating metabolites. metabolites. These These ideas ideasfor enhanced for enhanced AR PET AR imaging PET imaging agent agentdesign design and drug and treatmentdrug treatment paradigms paradigms are open are opento future to future investigations. investigations.

F FNT X

Dutasteride X Dutasteride X

F X FDHT X Flufenamic Acid Flufenamic Acid

Figure 18. Design consideration to to increase metabolic stability stability and and extend circulation time of an AR PET probe. Boldness Boldness of of arrows arrows indi indicatescates relative rates. Red Red X’s X’s show show sites sites of of metabolism inhibition.

Another opportunity opportunity that that might might merit merit further further invest investigationigation would would be be to tobetter better define define the the role role of SHBGof SHBG in inthe the efficiency efficiency and and selectivity selectivity of ofimaging imaging AR AR in in prostate prostate cancers, cancers, which which proved proved to to be be a challenging problem problem to to address address in in preclinical preclinical models models [119]. [119 As]. As a prelude a prelude to such to such an investigation, an investigation, we preparedwe prepared two two AR AR ligands ligands that that were were closely closely matche matchedd structurally structurally and and had had very very similar similar AR AR binding affinities,aCancersffinities, 2020 but , 12, very x different different SHBG binding aaffinitiesffinities (Figure(Figure 1919).). As we had hoped, in rats, rats, which which16 lack lackof 32 SHBG, these two showed very similar uptake in the prostate. The definitive definitive experiments to test the role ofof SHBGSHBG would would involve involve comparing comparing these these two two in nonhumanin nonhuman primates, primates, as we as hadwe donehad done earlier earlier with ratherwith rather different different compounds compounds [17], or[17], better or be yet,tter in yet, men in withmen prostatewith prostate cancers. cancers.

Figure 19. Matched set of fluorine-18 fluorine-18 labeled AR PET prob probeses with equivalent AR binding and uptake eefficiencyfficiency and selectivity, butbut veryvery didifferentfferenta affinitiesffinities for for SHBG. SHBG.

There are also a wholewhole hosthost ofof nonsteroidalnonsteroidal ligandsligands forfor AR,AR, developeddeveloped forfor variousvarious endocrineendocrine therapeutic purposes, such as antiandrogensantiandrogens to treat prostate cancer (enzalutamide, apalutamide, darolutamide, and earlier flutamideflutamide and bicalutamide) and selective androgen receptor modulators (SARMs) for for treating treating men men with with hypogonadal hypogonadal conditions conditions (many (many compounds compounds from from GTx) GTx) [120]. [ 120The]. Thethiohydantoin thiohydantoin core core of most of most of these of these compounds compounds wa wass actually actually first first described described many many years years ago ago [121]. [121]. Because some of these have high AR binding affinity and are small, relatively polar compounds, we radiolabeled a few of them with fluorine-18 or bromine-76; however, none of them showed AR- selective prostate uptake in our preclinical model [122–124]. Note is made of another recent review covering the development of PET imaging agents for AR [59].

5. Considerations for How PET Imaging of Steroid Receptors Might Best be Used to Improve the Management of Breast and Prostate Cancers It might seem axiomatic that PET imaging of ER with FES and PgR with FFNP in breast cancers, and AR with FDHT in prostate cancers, would be of use in detecting and staging disease and as prognostic and predictive factors in guiding optimal treatment for women and men with these cancers. An evaluation of their clinical utility, however, requires deeper analysis of the distinct roles played by these three receptors in their associated cancers, as well as consideration of the extent to which the information they provide goes beyond that can be obtained from other more easily available imaging modalities and from biomarker analyses on tumor biopsy samples. These issues will be discussed in the three sections that follow. Irrespective of their use in precision medicine to guide optimal endocrine therapy for individual patients, however, PET imaging of ER [125–127] and AR [128,129] in tumors plays a key role in guiding optimal dose selection during the clinical development of new receptor antagonists through pharmacodynamic or target engagement studies.

5.1. Utility of FES-PET Imaging of ER in Breast Cancers for Guiding Endocrine Therapy Selection As early as the late 1890s, long before ER and its activating ligand estradiol were identified, some women with breast cancer were found to respond—often dramatically—to surgical removal of their ovaries. This discovery by Sir George Beatson in 1896 is considered the first example of targeted endocrine therapy for cancer [130,131]. It soon became apparent that surgical endocrine ablation was effective in only about one-third of breast cancer patients [132]. Thus, even a century ago, it was clear that a means for predicting response to endocrine therapy was needed to spare many women with breast cancer from the morbidity of a surgery that was likely to be ineffective. Of course, we now know that ER is a major driver of proliferation in many breast cancers and is the target for endocrine therapies, such as aromatase inhibitors and antiestrogens. In fact, about 70% of breast cancers have detectable levels of ER and are termed ER-positive, but only about half of ER- positive disease actually responds to endocrine therapy, with response rates being lower in more advanced or recurrent disease. Thus, irrespective of how ER assays are being done, positive Cancers 2020, 12, 2020 17 of 32

Because some of these have high AR binding affinity and are small, relatively polar compounds, we radiolabeled a few of them with fluorine-18 or bromine-76; however, none of them showed AR-selective prostate uptake in our preclinical model [122–124]. Note is made of another recent review covering the development of PET imaging agents for AR [59].

5. Considerations for How PET Imaging of Steroid Receptors Might Best be Used to Improve the Management of Breast and Prostate Cancers It might seem axiomatic that PET imaging of ER with FES and PgR with FFNP in breast cancers, and AR with FDHT in prostate cancers, would be of use in detecting and staging disease and as prognostic and predictive factors in guiding optimal treatment for women and men with these cancers. An evaluation of their clinical utility, however, requires deeper analysis of the distinct roles played by these three receptors in their associated cancers, as well as consideration of the extent to which the information they provide goes beyond that can be obtained from other more easily available imaging modalities and from biomarker analyses on tumor biopsy samples. These issues will be discussed in the three sections that follow. Irrespective of their use in precision medicine to guide optimal endocrine therapy for individual patients, however, PET imaging of ER [125–127] and AR [128,129] in tumors plays a key role in guiding optimal dose selection during the clinical development of new receptor antagonists through pharmacodynamic or target engagement studies.

5.1. Utility of FES-PET Imaging of ER in Breast Cancers for Guiding Endocrine Therapy Selection As early as the late 1890s, long before ER and its activating ligand estradiol were identified, some women with breast cancer were found to respond—often dramatically—to surgical removal of their ovaries. This discovery by Sir George Beatson in 1896 is considered the first example of targeted endocrine therapy for cancer [130,131]. It soon became apparent that surgical endocrine ablation was effective in only about one-third of breast cancer patients [132]. Thus, even a century ago, it was clear that a means for predicting response to endocrine therapy was needed to spare many women with breast cancer from the morbidity of a surgery that was likely to be ineffective. Of course, we now know that ER is a major driver of proliferation in many breast cancers and is the target for endocrine therapies, such as aromatase inhibitors and antiestrogens. In fact, about 70% of breast cancers have detectable levels of ER and are termed ER-positive, but only about half of ER-positive disease actually responds to endocrine therapy, with response rates being lower in more advanced or recurrent disease. Thus, irrespective of how ER assays are being done, positive predictive values (PPVs) are very low, only around 50%. (Negative predictive values (NPVs) are much higher.) Clearly, improved methods are still needed to predict whether ER-positive cancers will respond to endocrine therapies, which have relatively mild side effects, or whether more aggressive radiation or chemotherapies are warranted, despite their greater morbidities [133,134]. Earlier radiometric assays measured ER levels by the direct binding of [3H]estradiol, ex vivo on extracts of fresh or frozen tumor samples; current ER IHC assays measure an antigenic signal from ER on formalin-fixed, paraffin -imbedded (FFPE) sections from fine-needle biopsies and are graded on scales that consider staining intensity and distribution of stained cells [135,136]. By contrast, FES-PET imaging of breast tumors provides a more direct measurement of ER, based on its ligand-binding function that is assessed in vivo and in situ, and noninvasively (recently extensively reviewed [133,137]). There are sensitivity limits to FES-PET, and imaging of lesions in liver and near the colon is difficult due to high background from FES metabolism and hepatobiliary elimination, yet multiple tumor sites can be evaluated by FES-PET without patient discomfort, even those that are inaccessible (e.g., brain), are challenging to reach by needle biopsy, or require tissue processing (bone) [138–140]. ER levels in recurrent disease can also differ from those determined in the primary tumor at an earlier time [137,141], and FES-PET can provide a measurement at the time of treatment decision. PET also provides an image of the whole tumor and avoids the risk of sampling error from fine needle biopsies due to intratumoral heterogeneity in ER levels [142]. Cancers 2020, 12, 2020 18 of 32

In early FES-PET studies, we showed FES-specific uptake in primary breast tumors correlated quantitatively with ER levels determined subsequently by radioligand binding assays on biopsy samples [51]. FES uptake in metastatic disease also correlated with earlier radiometric ER determinations from the primary tumor [143]. Notably, there was no correlation between tumor uptake by FES-PET and [18F]fluorodeoxyglucose (FDG)-PET [144]; only FES-PET was predictive of benefit from endocrine therapy [145]. In advanced breast cancers, FES-PET provided more accurate prognoses and prediction of endocrine therapy success [146], adding value over that from ER IHC assays and FDG-PET images [147], particularly in patients otherwise difficult to evaluate [133,147–152]. A number of active clinical trials (some multicenter) continue to investigate the predictive value of FES-PET, often together with FDG-PET or other targeted imaging agents, or to assess its ability to identify metastases and intrapatient heterogeneity ((NCT01957332), (NCT02398773), (NCT03726931), and (EUDRACT 2013-000-287-29)). Suggestions for best practices in the clinical use of FES-PET have been published [153], as has a simulation of the cost effectiveness of FES-PET vs. conventional work-up of patients with metastatic breast cancer [154]. Zionexa, a French radiopharmaceutical company, has recently obtained approval for the clinical use of FES in France and the US. FES, termed “Cerianna” by Zionexa and registered as EstroTep®, is now available for clinical use in the US and should be widely available commercially through an agreement between Zionexa and PETNET. A very recent publication in The Oncologist provides a comprehensive analysis of the clinical value of FES-PET imaging in breast cancer [155]. Because of the high negative predictive value of FES-PET, it is likely to be particularly useful in predicting the absence of significant ER levels in metastatic disease, although its utility in providing more definitive prognostic and predictive information about other stages of disease are likely to be identified. The value of FES-PET imaging needs to be considered within the larger landscape of other approaches to predict benefit from endocrine therapy in breast cancer. Much current research seeks to predict endocrine therapy response in advanced breast cancer using various “-omics” (genomic, cistromic, transcriptomic, proteomic, and metabolomic) databases derived from analyses of tumor biopsy samples, as well as liquid biopsies of circulating tumor cells and circulating tumor DNA in metastatic breast cancer. “Signatures” derived statistically from these databases are used to define molecular pathological subtypes that are then correlated with disease prognosis and prediction of therapy response. A good example comes from a recent clinical trial (TAILORx) showing that a 21-gene profile (Oncotype-Dx) effectively identified patients with early stage ER-positive, node-negative disease who are very likely to benefit from endocrine therapies alone [156]. In advanced recurrent and metastatic ER-positive disease, endocrine therapy response rates are much lower and depend on prior therapies [157,158], and currently, there are no -omics assays that robustly predict benefit from endocrine-only therapy. It is in this context that the value of FES-PET is most likely to add predictive value to the benefit from continued endocrine therapies. In a yet larger context, PET imaging presents an interesting alternative approach of directly accessing whether ER in tumors remains functional through the use of “hormone challenge tests.” Such PET imaging-based challenge tests are discussed below in a broader context (Section 5.2.).

5.2. Clinical Utility of FFNP-PET Imaging of PgR in Breast Cancers: PET Imaging-Based Hormone-Challenge Tests Clinical assays of ER in breast cancers have good negative predictive values (NPVs), but much poorer positive predictive values (PPVs); it seems, in fact, that ER can be present, but somehow incapable of mediating endocrine therapy. Because the PgR gene is highly regulated by estrogen action through ER [159–161], as early as the 1970s, the measurement of breast tumor PgR levels was proposed as a way to improve the PPV of receptor assays for endocrine therapy benefit [162,163]: an ER-positive tumor that was also PgR-positive would indicate that ER was functional and would respond to endocrine therapy, whereas an ER-positive/PgR-negative tumor would not respond, because the ER, somehow, was nonfunctional. Consequently, breast cancer biopsies have been routinely assayed for Cancers 2020, 12, x 18 of 32

endocrine-only therapy. It is in this context that the value of FES-PET is most likely to add predictive value to the benefit from continued endocrine therapies. In a yet larger context, PET imaging presents an interesting alternative approach of directly accessing whether ER in tumors remains functional through the use of “hormone challenge tests.” Such PET imaging-based challenge tests are discussed below in a broader context (Section 5.2.).

5.2. Clinical Utility of FFNP-PET Imaging of PgR in Breast Cancers: PET Imaging-Based Hormone- Challenge Tests Clinical assays of ER in breast cancers have good negative predictive values (NPVs), but much poorer positive predictive values (PPVs); it seems, in fact, that ER can be present, but somehow incapable of mediating endocrine therapy. Because the PgR gene is highly regulated by estrogen action through ER [159–161], as early as the 1970s, the measurement of breast tumor PgR levels was proposed as a way to improve the PPV of receptor assays for endocrine therapy benefit [162,163]: an CancersER-positive2020, 12 tumor, 2020 that was also PgR-positive would indicate that ER was functional and would19 of 32 respond to endocrine therapy, whereas an ER-positive/PgR-negative tumor would not respond, because the ER, somehow, was nonfunctional. Consequently, breast cancer biopsies have been bothroutinely ER and assayed PgR for for many both years. ER Theand logicalPgR for and many attractive years. rationale The logical notwithstanding, and attractive however, rationale PgR assaysnotwithstanding, have not had however, a transformative PgR assays eff haveect on not the had predictive a transformative value of aneffect ER-positive on the predictive result, particular value of inan advancedER-positive disease. result, In particular fact, the valuein advanced of PgR diseas measurementse. In fact, remainsthe value an of area PgR of measurements active disagreement remains in recommendationsan area of active disagreement for best clinical in recommendations practice in breast for cancer best [clinical164–169 practice]. in breast cancer [164–169]. As most breast cancers are found in older, postmenopausal women, a reasonable but not widely considered explanation is it that PgR assays in these women are compromisedcompromised because their low endogenous levels of estradiol are insuinsufficientfficient to elevate PgR levels, even when ER isis active.active. In our initial clinical study of FFNP-PET in breast cancer patients (all of whomwhom werewere postmenopausal)postmenopausal) (NCT00968409),(NCT00968409), we found that FFNPFFNP uptakeuptake waswas greatergreater inin PgR-positivePgR-positive thanthan PgR-negativePgR-negative breast cancers (Figure(Figure 20 20)) and and the the average average SUV SUV values values were we separated;re separated; uptake uptake values, values, however, however, were neither were veryneither high very nor high markedly nor markedly separated sepa [4],rated very likely[4], very because likely of because the low of levels the oflow circulating levels of estradiolcirculating in theseestradiol women. in these (There women. are other (There ongoing are other clinical ongoing trials clinical using FFNP-PET trials using on FFNP-PET invasive breast on invasive cancer breast at the Universitycancer at the of University Wisconsin of (NCT03212170).) Wisconsin (NCT03212170).)

Figure 20. FFNP-PET sagittal image of breast cancer inin womanwoman withwith ERER+/PgR+/HER2-disease.+/PgR+/HER2-disease. (Image provided byby Drs.Drs. Farrokh Dehdashti and Barry Siegel, WashingtonWashington UniversityUniversity MedicalMedical School.)School.)

Considering these issues, we hypothesized that th thee value of PgR as a biomarkerbiomarker forfor functionalfunctional ER would be clearer if it were used as a dynamic biomarker, byby FFNP-PETFFNP-PET imaging of PgR levels in tumors before andand then shortly after a brief, stimulatingstimulating dose of estradiol: if tumor ER was was functional, functional, we expected to see an increase in FFNP-PET SUV, whereaswhereas thethe absenceabsence ofof an increase would suggest that ER was nonfunctional. We demonstrated the validityvalidity of such an PgRPgR basedbased “estrogen-challenge test” throughthrough proof-of-principleproof-of-principle preclinical studies in syngeneicsyngeneic murinemurine mammarymammary cancercancer modelsmodels derived from STAT1 knockoutknockout micemice [[170];170]; by serial Western blotblot analyses and FFNP-PET imaging, PgR levels decreased decreased with with estradiol estradiol deprivation deprivation and and increa increasedsed after after a 1-day a 1-day estradiol estradiol stimulation stimulation in the in ER- the ER-positivepositive and and responsive responsive SSM3 SSM3 tumors, tumors, whereas whereas ther theree were were no no apparent apparent changes changes in in the ER-positive nonresponsive SSM2 tumor system [84,85]. [84,85]. A A more recent study has tracked the time course of PgRPgR level increase in human breast cancercancer T47DT47D xenograftsxenografts inin micemice byby biodistribution,biodistribution, with levels continuing to riserise eveneven afterafter 2-days2-days ofof estrogenestrogen treatmenttreatment [[171].171]. WhileWhile treatingtreating womenwomen withwith ER-positiveER-positive breastbreast cancer with estradiol, an ER agonist, would seem counterintuitive, high-dose estrogen (originally diethylstilbestrol) was actually the first form of endocrine therapy [172,173]. Response rates are equivalent to those with antiestrogens, but because of side effects, high-dose estrogen therapy has been replaced by antiestrogens [174], although it is still used in certain advanced disease [175,176]. In a clinical study at Washington University that was just completed (NCT02455453), we used FFNP-PET to monitor the change in PgR after a 1-day estradiol challenge in 43 women with ER-positive advanced breast cancer that had failed on various prior endocrine therapies and chemotherapies, but were still candidates for treatment with the powerful antiestrogen fulvestrant or other endocrine therapies, often combined with CDK4,6 inhibitors. The value of the change in FFNP-PET SUV before and after the 1-day treatment with estradiol in predicting clinical benefit is being compared with baseline FFNP SUV and PgR by IHC. The results from this trial will be reported shortly. It is worth considering our current use of a hormone-challenge paradigm in PET imaging in breast cancer within the broader landscape of endocrinology (for a more extensive review, see reference [47]). Breast cancer is an endocrine-driven disease, and challenge tests of various types are commonly used Cancers 2020, 12, 2020 20 of 32 in endocrinology, examples being the glucose challenge test for diabetes and the dexamethasone suppression test to assess adrenal gland function. In this regard, I was intrigued by the clinical observation of a “tamoxifen flare” [177]. This was a transient exacerbation of clinical pain symptoms experienced in some breast cancer patients about 1–2 weeks after starting on tamoxifen therapy that would then subside as patients began to benefit from the continued endocrine therapy. This tamoxifen flare was ascribed to a transient agonist effect of tamoxifen early in the course of treatment as therapeutic levels were building up [178]. Later, it was demonstrated that very low levels of tamoxifen will actually stimulate MCF-7 breast cancer cell growth [179], although higher levels are inhibitory [180]. Clinical tamoxifen flare was not very useful as a predictor of benefit from tamoxifen, because responders without bone metastases rarely experienced a flare in pain, and some nonresponders experienced increasing pain due to disease progression. Thus, pain was not the proper parameter to use to monitor this transient stimulation of tumors in which ER was functional, a response that actually represented an early response to what would eventually be a beneficial outcome. FDG-PET imaging is widely used in cancer for assessing the extent of disease, but it is of less value in detecting “indolent disease,” because the Warburg effect develops when cancer cells are avidly proliferating. If tamoxifen flare was the result of a transient stimulation of the tumor by the agonistic effect of low-dose tamoxifen that cannot be reliably detected by pain symptoms, perhaps the flare could be more readily detected by changes in tumor metabolism monitored by two FDG-PET images, one at baseline and one after 1–2 weeks of tamoxifen treatment, the time of maximum clinical tamoxifen flare development. Thus, we added to our early clinical studies of FES-PET imaging in ER-positive breast cancer two FDG-PET images, i.e., women received both FES-PET and FDG-PET at baseline and then a second FDG-PET image at 7–10 days, after tamoxifen therapy begin. While the FES-PET image was more predictive of therapy benefit than the results of the clinical assay of ER, the change in FDG-PET image was even more predictive [144,181–183]. When aromatase inhibitors replaced tamoxifen as preferred endocrine therapy, we modified the hormone challenge paradigm to be just 1 day of estradiol prior to the onset of aromatase inhibitor treatment, because the therapy agent would no longer cause transient stimulation of responsive tumors [184]. Again, the change in FDG-PET SUV from this estradiol challenge test of metabolic flare proved to be very predictive, even in women with advanced breast cancers, many of whom had received prior therapies of various kinds. Though predictive, the changes we saw in FDG-PET were relatively modest, with many values clustering around the positive–negative cut off value. Our recent clinical study of the changes in PgR measured by FFNP-PET in advanced breast cancer, uses the same 1-day estradiol challenge paradigm (NCT02455453). By monitoring PgR level rather than FDG uptake rate, we are hoping to find a larger separation between responders and nonresponders in the change of FFNP SUV than what we had found with FDG-SUV. In the aforementioned preclinical trial with FFNP in a mouse mammary tumor model, we had found the changes in FFNP-SUV to be much larger than the changes in FDG-SUV [84,85].

5.3. Clinical Studies of FDHT-PET in Prostate Cancer PET imaging of AR in men with advanced prostate cancer was published in 2004 and 2005 by me and my colleagues at Washington University Medical School in St. Louis [3] and by investigators at Memorial Sloan Kettering Cancer Center in New York (Figure 21)[115]. Clear FDHT uptake was evident in numerous lesions, with uptake being blocked following treatment with AR antagonists. Subsequent studies made comparisons between FDHT and FDG images in the same patient to characterize lesion heterogeneity, and comparative images in a series of patients showed both concordance (i.e., high or low uptake of both PET tracers) and discordance (FDHT high, FDG low, and the reverse) [185]. Some prognostic trends were correlated with these four distinct patterns of tracer uptake, with survival being greatest when FDG and FDHT images were concordant and lowest when FDHT uptake by lesions was very low. Survival was also inversely related to total lesion count and intrapatient heterogeneity; notably, PSA levels had little prognostic value [185]. Because often Cancers 2020, 12, x 20 of 32

5.3. Clinical Studies of FDHT-PET in Prostate Cancer PET imaging of AR in men with advanced prostate cancer was published in 2004 and 2005 by me and my colleagues at Washington University Medical School in St. Louis [3] and by investigators at Memorial Sloan Kettering Cancer Center in New York (Figure 21) [115]. Clear FDHT uptake was evident in numerous lesions, with uptake being blocked following treatment with AR antagonists. Subsequent studies made comparisons between FDHT and FDG images in the same patient to characterize lesion heterogeneity, and comparative images in a series of patients showed both concordance (i.e., high or low uptake of both PET tracers) and discordance (FDHT high, FDG low, and the reverse) [185]. Some prognostic trends were correlated with these four distinct patterns of tracer uptake, with survival being greatest when FDG and FDHT images were concordant and lowest Cancerswhen 2020FDHT, 12 ,uptake 2020 by lesions was very low. Survival was also inversely related to total lesion21 count of 32 and intrapatient heterogeneity; notably, PSA levels had little prognostic value [185]. Because often many moremore lesionslesions are are seen seen by by PET PET imaging imaging than than can can be biopsied,be biopsied, the imagingthe imaging results results can direct can tissuedirect samplingtissue sampling to the to sites the of sites possibly of possibly greatest greatest concern concern (e.g., high(e.g., metabolichigh metabolic activity activity (high (high FDG FDG PET) PET) and lowand ARlow responsivenessAR responsiveness (low (low FDHT FDHT PET)) PET)) [185 ].[185].

Figure 21.21. FDHT-PETFDHT-PET and and FDG-PET FDG-PET image image of prostateof prostate cancer cancer in an in individual an individual with FDHT-avid with FDHT-avid disease. (Imagesdisease. provided(Images provided by Dr. Steven by Dr. Larson, Steven MemorialLarson, Memorial Sloan Kettering Sloan Kette Cancerring Center.) Cancer Center.)

The predictive and and prognostic prognostic role role of of AR AR in in prostate prostate cancer cancer is israther rather different different than than that that of ofER ER in inbreast breast cancer. cancer. With With ER-positive ER-positive breast breast cancers, cancers, en endocrinedocrine therapies therapies often often become become less less effective effective in advanced disease;disease; soso PET PET imaging imaging of of ER ER with with FES FES and and ER ER functional functional status status by serialby serial FDG- FDG- or FFNP-PET or FFNP- withPET anwith estradiol an estradiol challenge challenge can provide can provide additional additional predictive predictive value for value benefit for from benefit continued from continued endocrine therapies.endocrine Bytherapies. contrast, By as contrast, long as AR as islong present as AR in prostateis present cancer, in prostate it typically cancer, remains it typically an effective remains target an foreffective hormone-directed target for hormone-dire therapies, evencted in advancedtherapies, stages even ofin castration-resistant advanced stages prostateof castration-resistant cancer (CRPC), althoughprostate cancer increasingly (CRPC), powerful although endocrine increasingly strategies powerful are endocrine needed [186 strategies]. The conversion are needed from [186]. CRPC The toconversion neuroendocrine-type from CRPC cancer to neuroendocrine-type typically involves loss cancer of both typically AR and involves responsiveness loss of toboth AR-targeted AR and therapiesresponsiveness [187]; so,to AR-targeted FDHT-PET might therapies prove [187]; useful so, inFDHT-PET identifying might this clinically prove useful significant in identifying progression this byclinically imaging. significant progression by imaging. More recently,recently, extensiveextensive developmentdevelopment of PETPET tracerstracers forfor prostate-specificprostate-specific membranemembrane antigenantigen (PSMA),(PSMA), which is widelywidely expressedexpressed in prostateprostate cancer,cancer, hashas producedproduced agentsagents ofof superbsuperb selectivityselectivity for detection of lesions; these are likely to play a majormajor role in PET imaging of prostate cancer [188,189]. [188,189]. Further investigations should help to clarify the relativerelative value of FDHT-PET vs.vs. PSMA-PET imaging in guiding therapies for prostate cancer, though it is likelylikely that both will have distinct roles to play, as will FDG-PET [[185,188,189].185,188,189]. PSMA PSMA also appears to be an ef efffectiveective target for radiotherapy agents that show eefficientfficient update and very prolongedprolonged retentionretention in lesions;lesions; localizationlocalization in cellcell nuclearnuclear structuresstructures makemake possible selective radioablation [190], [190], even even usin usingg Auger Auger electron-emitting electron-emitting radionuclides radionuclides [191]. [191 ]. Nevertheless, FDHT has proven to have enduring value in pharmacodynamicspharmacodynamics or receptor occupancy studiesstudies that that provide provide valuable valuable guidance guidance for dose for selection dose selection in clinical in trials clinical for the trials development for the of new antiandrogens. Such studies have been used for the development of enzalutamide (Xtandi) [129] and apalutamide (Erleada) [128].

6. Closing Comments The capability of PET as a modality for imaging function is unparalleled; this was articulated in an insightful comment made to me by Henry Wagner, a grandfather of the field, who said, “PET can image function better than any other mode of clinical imaging!” [192]. The prognostic and predictive value of PET imaging of ER by FES, PgR by FFNP, and AR by FDHT compare very favorably in many respects with the results of IHC assays. PET imaging, however, also provides several advantages by being noninvasive and assessing multiple sites and by measuring in vivo and in situ the capacity of the Cancers 2020, 12, 2020 22 of 32 receptor to bind its cognate ligand; this assessment of the binding function of these receptors, not an antigenic feature, is of immense value. Beyond imaging receptor-binding function, the use of serial PET imaging in a hormone-challenge paradigm, as we have done with FDG-PET to assess changes in tumor metabolism and FFNP-PET to assess changes in PgR gene expression, goes beyond hormone binding function; it rapidly evaluates the functional status of ER regulation of cellular activities in cancer that are associated with response to endocrine therapy. There is no reason that hormone challenge tests based on imaging other therapy-responsive functions in cancer could not also be developed, such as ones that would use PET probes for different cellular activities such as proliferation, hypoxia, and protein and DNA synthesis. Hence, the use of PET imaging of steroid receptors enables observation of the extent of cancer metastasis and predicts responsiveness to endocrine therapies. Its use in assessing new aspects of tumor clinical status noninvasively suggests a robust future for PET imaging in providing meaningful benefit for patients with breast and prostate cancers.

Funding: My research on the development of PET imaging agents for breast and prostate cancer has been funded over an extended period of time by grants from the National Institutes of Health (PHS 5R01CA015556) and the Department of Energy (DE FG02 86ER60401) Acknowledgments: I am grateful to the many graduate students and postdoctoral scientists in my laboratory who have contributed to the development of our ER, PR, and AR-targeted PET imaging agents. Although I am not able to name them all (see references from my laboratory cited in this review), I wish to acknowledge, as particularly central to this work, the contributions by Steven Senderoff, Scott Landvatter, Dale Kiesewetter, Kathryn Carlson, Martin Pomper, Aijun Liu, Henry VanBrocklin, Brad Buckman, and Sung Hoon Kim. At Washington University Medical School, the support of Mike Welch, as well as his friendship and that of his family, were immensely beneficial; critical as well was the help of many of his coworkers and colleagues, in particular, Karen McElvany, Carla, Mathias, Tim Tewson, Mike Kilbourn, and Dong Zhou, as well as Barry Siegel, Farrokh Dehdashti, Mark Mintun, Andrea McGuire, Joanne Mortimer, Ruby Chan, Amy Fowler, and Sally Schwarz. Support of my research over many years by grants from the National Institutes of Health, the Department of Energy, the American Cancer Society, and the Breast Cancer Research Foundation is gratefully acknowledged. I thank Farrokh Dehdashti and Barry Siegel (Washington University, St. Louis) and Steven Larson (Memorial Sloan Kettering Cancer Center, New York) for providing PET images used in this article. Conflicts of Interest: The author is a cofounder and stockholder of Radius Health, Inc. He is also a consultant of Celcuity, Inc.

References

1. Kiesewetter, D.O.; Katzenellenbogen, J.A.; Kilbourn, M.R.; Welch, M.J. Synthesis of 16-fluoroestrogens by unusually facile fluoride ion displacement reactions: Prospects for the preparation of fluorine-18 labeled estrogens. J. Org. Chem. 1984, 49, 4900–4905. [CrossRef] 2. Kiesewetter, D.O.; Kilbourn, M.R.; Landvatter, S.W.; Heiman, D.F.; Katzenellenbogen, J.A.; Welch, M.J. Preparation of four fluorine-18-labeled estrogens and their selective uptakes in target tissues of immature rats. J. Nucl. Med. 1984, 25, 1212–1221. [PubMed] 3. Dehdashti, F.; Picus, J.; Michalski, J.M.; Dence, C.S.; Siegel, B.A.; Katzenellenbogen, J.A.; Welch, M.J. Positron tomographic assessment of androgen receptors in prostatic carcinoma. Eur. J. Nucl. Med. Mol. Imaging 2005, 32, 344–350. [CrossRef][PubMed] 4. Dehdashti, F.; Laforest, R.; Gao, F.; Aft, R.L.; Dence, C.S.; Zhou, D.; Shoghi, K.I.; Siegel, B.A.; Katzenellenbogen, J.A.; Welch, M.J. Assessment of progesterone receptors in breast carcinoma by PET with 21-18F-fluoro-16alpha,17alpha-[(R)-(1’-alpha-furylmethylidene)dioxy]-19-norpregn- 4-ene-3,20-dione. J. Nucl. Med. 2012, 53, 363–370. [CrossRef] 5. Katzenellenbogen, J.A.; Johnson, H.J., Jr.; Myers, H.N. Photoaffinity labels for estrogen binding proteins of rat uterus. Biochemistry 1973, 12, 4085–4092. [CrossRef][PubMed] 6. Brandes, S.J.; Katzenellenbogen, J.A. Fluorinated androgens and progestins: Molecular probes for androgen and progesterone receptors with potential use in positron emission tomography. Mol. Pharmacol. 1987, 32, 391–403. 7. Kochanny, M.J.; VanBrocklin, H.F.; Kym, P.R.; Carlson, K.E.; O’Neil, J.P.; Bonasera, T.A.; Welch, M.J.; Katzenellenbogen, J.A. Fluorine-18-labeled progestin ketals: Synthesis and target tissue uptake selectivity of potential imaging agents for receptor-positive breast tumors. J. Med. Chem. 1993, 36, 1120–1127. [CrossRef] Cancers 2020, 12, 2020 23 of 32

8. Katzenellenbogen, J.A.; Carlson, K.E.; Heiman, D.F.; Lloyd, J.E. Receptor binding as a basis for radiopharmaceutical design. In Radiopharmaceuticals, Structure-Activity Relationships; Grune and Stratton: New York, NY, USA, 1981; pp. 23–86. 9. Katzenellenbogen, J.A.; Heiman, D.F.; Senderoff, S.G.; Landvatter, S.W.; Carlson, K.E.; Goswami, R.; Lloyd, J.E.; McElvany, K.D. Estrogen receptor-based agents for imaging breast tumors: Binding selectivity as a basis for design and optimization. In Applications of Nuclear and Radiochemistry; Pergamon: Elmsford, NY, USA, 1982; pp. 311–323. 10. Pomper, M.G.; VanBrocklin, H.; Thieme, A.M.; Thomas, R.D.; Kiesewetter, D.O.; Carlson, K.E.; Mathias, C.J.; Welch, M.J.; Katzenellenbogen, J.A. 11. beta.-methoxy-, 11. beta.-ethyl, and 17. alpha.-ethynyl-substituted 16. alpha.-fluoroestradiols: Receptor-based imaging agents with enhanced uptake efficiency and selectivity. J. Med. Chem. 1990, 33, 3143–3155. [CrossRef] 11. Katzenellenbogen, J.A.; Hsiung, H.M. Iodohexestrols. I. Synthesis and photoreactivity of iodinated hexestrol derivatives. Biochemistry 1975, 14, 1736–1741. [CrossRef] 12. Katzenellenbogen, J.A.; Hsiung, H.M.; Carlson, K.E.; McGuire, W.L.; Kraay, R.J.; Katzenellenbogen, B.S. Iodohexestrols. II. Characterization of the binding and estrogenic activity of iodinated hexestrol derivatives, in vitro and in vivo. Biochemistry 1975, 14, 1742–1750. [CrossRef] 13. Jensen, E.; Jacobson, H. Recent Progr. Hormone Res. 1962, 18, 387. 14. Jensen, E.; Jacobson, H.; Flesher, J.; Saha, N.; Gupta, G.; Smith, S.; Colucci, V.; Shiplacoff, D.; Neumann, H.; DeSombre, E. Estrogen receptors in target tissues. Steroid Dyn. 1966, 133–157. 15. McElvany, K.D.; Carlson, K.E.; Katzenellenbogen, J.A.; Welch, M.J. Factors affecting the target site uptake selectivity of estrogen radiopharmaceuticals: Serum binding and endogenous estrogens. J. Steroid Biochem. 1983, 18, 635–641. [CrossRef] 16. Peterson, L.M.; Kurland, B.F.; Link, J.M.; Schubert, E.K.; Stekhova, S.; Linden, H.M.; Mankoff, D.A. Factors influencing the uptake of 18F-fluoroestradiol in patients with estrogen receptor positive breast cancer. Nucl. Med. Biol. 2011, 38, 969–978. [CrossRef][PubMed] 17. Bonasera, T.A.; O’Neil, J.P.; Xu, M.; Dobkin, J.A.; Cutler, P.D.; Lich, L.L.; Choe, Y.S.; Katzenellenbogen, J.A.; Welch, M.J. Preclinical evaluation of fluorine-18-labeled androgen receptor ligands in baboons. J. Nucl. Med. 1996, 37, 1009–1015. [PubMed] 18. Jonson, S.D.; Bonasera, T.A.; Dehdashti, F.; Cristel, M.E.; Katzenellenbogen, J.A.; Welch, M.J. Comparative breast tumor imaging and comparative in vitro metabolism of 16alpha-[18F]fluoroestradiol-17beta and 16beta-[18F]fluoromoxestrol in isolated hepatocytes. Nucl. Med. Biol. 1999, 26, 123–130. [CrossRef] 19. Hochberg, R.B.; Rosner, W. Interaction of 16 alpha-[125I] iodo-estradiol with estrogen receptor and other steroid-binding proteins. Proc. Natl. Acad. Sci. USA 1980, 77, 328. [CrossRef] 20. McManaway, M.; Jagoda, E.; Eckelman, W.; Larson, S.; Francis, B.; Gibson, R.; Reba, R.; Lippman, M.E. Binding characteristics and biological activity of 17α-[125I] iodovinyl-11β-methoxyestradiol, an estrogen receptor-binding radiopharmaceutical, in human breast cancer cells (MCF-7). Cancer Res. 1986, 46, 2386–2389. 21. Kabalka, G.W.; Gooch, E.E.; Hsu, H.C.; Washburn, L.C.; Sun, T.T.; Hayes, R.L. Rapid and mild syntheses of radioiodinated estrogen derivatives via organoborane technology. In Applications of Nuclear and Radiochemistry; Pergamon: Elmsford, NY, USA, 1982; pp. 197–203. 22. Hanson, R.N.; Seitz, D.E.; Botarro, J.C. E-17-[125 I] Iodovinylestradiol: An estrogen-receptor-seeking radiopharmaceutical. J. Nucl. Med. 1982, 23, 436. 23. DeSombre, E.R.; Mease, R.C.; Sanghavl, J.; Singh, T.; Seevers, R.H.; Hughes, A. Estrogen receptor binding affinity and uterotrophic activity of triphenylhaloethylenes. J. Steroid Biochem. 1988, 29, 583–590. [CrossRef] 24. Ali, H.; Rousseau, J.; Van Lier, J. Synthesis of (17α, 20E/Z) iodovinyl testosterone and 19-nortestosterone derivatives as potential radioligands for androgen and progesterone receptors. J. Steroid Biochem. Mol. Biol. 1994, 49, 15–29. [CrossRef] 25. Rijks, L.J.; Bakker, P.J.; van Tienhoven, G.; Noorduyn, L.A.; Boer, G.J.; Rietbroek, R.C.; Taat, C.W.; Janssen, A.G.; Veenhof, C.H.; van Royen, E.A. Imaging of estrogen receptors in primary and metastatic breast cancer patients with iodine-123-labeled Z-MIVE. J. Clin. Oncol. 1997, 15, 2536–2545. [CrossRef][PubMed] 26. Heiman, D.F.; Senderoff, S.G.; Katzenellenbogen, J.A.; Neeley, R.J. Estrogen receptor-based imaging agents. 1. Synthesis and receptor binding affinity of some aromatic and D-ring halogenated estrogens. J. Med. Chem. 1980, 23, 994–1002. [CrossRef][PubMed] Cancers 2020, 12, 2020 24 of 32

27. Katzenellenbogen, J.A.; Senderoff, S.G.; McElvany, K.D.; O’Brien, H.; Welch, M.J. 16α-[77Br] bromoestradiol-17β: A high specific-activity, gamma-emitting tracer with uptake in rat uterus and induced mammary tumors. J. Nucl. Med. 1981, 22, 42–47. [CrossRef] 28. Katzenellenbogen, J.A.; McElvany, K.D.; Senderoff, S.G.; Carlson, K.E.; Landvatter, S.W.; Welch, M.J. 16 alpha-[77Br]bromo-11 beta-methoxyestradiol-17 beta: A gamma-emitting estrogen imaging agent with high uptake and retention by target organs. J. Nucl. Med. 1982, 23, 411–419. 29. Senderoff, S.G.; McElvany, K.D.; Carlson, K.E.; Heiman, D.F.; Katzenellenbogen, J.A.; Welch, M.J. Methodology for the synthesis and specific activity determination of 16α-[77Br]-Bromoestradiol-17 β and 16α-[77Br]-11 β-methoxyestradiol-17 β, Two estrogen receptor-binding radiopharmaceuticals. Int. J. Appl. Radiat. Isot. 1982, 33, 545–551. [CrossRef] 30. Katzenellenbogen, J.A. Estrogen and progestin radiopharmaceuticals for imaging breast cancer. In Estrogens, Progestins, and Their Antagonists; Pavlik, E.J., Ed.; Birkhäuser: Boston, MS, USA, 1997; Volume 2, pp. 197–242. 31. Krohn, K.A.; Vera, D.R. Concepts for design and analysis of receptor radiopharmaceuticals: The Receptor-Binding Radiotracers series of meetings provided the foundation. Nucl. Med. Biol. 2020. [CrossRef] 32. McElvany, K.D.; Katzenellenbogen, J.A.; Shafer, K.E.; Siegel, B.A.; Senderoff, S.G.; Welch, M.J. 16α-[77Br] bromoestradiol: Dosimetry and preliminary clinical studies. J. Nucl. Med. 1982, 23, 425–430. 33. Tewson, T.J.; Welch, M.J.; Raichle, M.E. [18F]-labeled 3-deoxy-3-fluoro-D-glucose: Synthesis and preliminary biodistribution data. J. Nucl. Med. 1978, 19, 1339–1345. [CrossRef] 34. Welch, M.J.; Kilbourn, M.R.; Mathias, C.J.; Mintun, M.A.; Raichle, M.E. Comparison in animal models of 18F-spiroperidol and 18F-haloperidol: Potential agents for imaging the dopamine receptor. Life Sci. 1983, 33, 1687–1693. [CrossRef] 35. Berridge, M.; Apana, S.; Hersh, J. Teflon radiolysis as the major source of carrier in fluorine-18. J. Labelled Compd. Radiopharm. 2009, 52, 543–548. [CrossRef] 36. Bergmann, K.E.; Landvatter, S.W.; Rocque, P.G.; Carlson, K.E.; Welch, M.J.; Katzenellenbogen, J.A. Oxohexestrol derivatives labeled with fluorine-18. Synthesis, receptor binding and in vivo distribution of two non-steroidal estrogens as potential breast tumor imaging agents. Nucl. Med. Biol. 1994, 21, 25–39. [CrossRef] 37. Landvatter, S.W.; Katzenellenbogen, J.A. Nonsteroidal estrogens: Synthesis and estrogen receptor binding affinity of derivatives of (3R*, 4S*)-3, 4-bis (4-hydroxyphenyl) hexane (hexestrol) and (2R*, 3S*)-2, 3-bis (4-hydroxyphenyl) pentane (norhexestrol) functionalized on the side chain. J. Med. Chem. 1982, 25, 1300–1307. [CrossRef][PubMed] 38. Landvatter, S.W.; Kiesewetter, D.O.; Kilbourn, M.R.; Katzenellenbogen, J.A.; Welch, M.J. (2R , ∗ 3S )-1-[18fluoro-2, 3-bis (4-hydroxyphenyl) pentane ([18F] fluoronorhexestrol), A positron-emitting estrogen ∗ that shows highly-selective, receptor-mediated uptake by target tissues in vivo. Life Sci. 1983, 33, 1933–1938. [CrossRef] 39. Katzenellenbogen, J.A. The pharmacology of steroid radiopharmaceuticals: Specific and non-specific binding and uptake selectivity. In Radiopharmaceuticals: Chemistry and Pharmacology; Nunn, A.D., Ed.; M Dekker: New York, NY, USA, 1992; pp. 297–331. 40. Katzenellenbogen, J.; Heiman, D.; Carlson, K.; Lloyd, J. In vivo and in vitro steroid receptor assays in the design of estrogen radiopharmaceuticals. In Receptor-Binding Radiotracers I; Eckelman, W.C., Ed.; CRC Press: Boca Raton, FL, USA, 1982; Volume I, pp. 93–126. 41. Lim, J.L.; Zheng, L.; Berridge, M.S.; Tewson, T.J. The use of 3-methoxymethyl-16 beta, 17 beta-epiestriol-O-cyclic sulfone as the precursor in the synthesis of F-18 16 alpha-fluoroestradiol. Nucl. Med. Biol. 1996, 23, 911–915. [CrossRef] 42. Kil, H.S.; Cho, H.Y.; Lee, S.J.; Oh, S.J.; Chi, D.Y. Alternative synthesis for the preparation of 16alpha-[(18) F]fluoroestradiol. J. Label. Comp. Radiopharm 2013, 56, 619–626. [CrossRef] 43. Zhou, D.; Lin, M.; Yasui, N.; Al-Qahtani, M.H.; Dence, C.S.; Schwarz, S.; Katzenellenbogen, J.A. Optimization of the preparation of fluorine-18-labeled steroid receptor ligands 16alpha-[18F] fluoroestradiol (FES),[18F] fluoro furanyl norprogesterone (FFNP), and 16beta-[18F] fluoro-5alpha-dihydrotestosterone (FDHT) as radiopharmaceuticals. J. Label. Compd. Radiopharm. 2014, 57, 371–377. [CrossRef] 44. Fedorova, O.; Nikolaeva, V.; Krasikova, R. Automated SPE-based synthesis of 16alpha-[(18)F]fluoroestradiol without HPLC purification step. Appl. Radiat. Isot. 2018, 141, 57–63. [CrossRef] Cancers 2020, 12, 2020 25 of 32

45. Shi, J.; Afari, G.; Bhattacharyya, S. Rapid synthesis of [18F]fluoroestradiol: Remarkable advantage of microwaving over conventional heating. J. Label. Comp. Radiopharm 2014, 57, 730–736. [CrossRef] 46. Liang, S.; Lan, X.; Zhang, Y.; Xu, X.; Li, B. Fully automatic synthesis of [(1)(8)F]FES for reporter gene hERL expression imaging. Nucl. Med. Commun. 2012, 33, 29–33. [CrossRef] 47. Katzenellenbogen, J.A. The quest for improving the management of breast cancer by functional imaging: The discovery and development of 16α-[18F] fluoroestradiol (FES), a PET radiotracer for the estrogen receptor, a historical review. Nucl. Med. Biol. 2020.[CrossRef][PubMed] 48. Paquette, M.; Phoenix, S.; Ouellet, R.; Langlois, R.; van Lier, J.E.; Turcotte, E.E.; Benard, F.; Lecomte, R. Assessment of the novel estrogen receptor PET tracer 4-fluoro-11beta-methoxy-16alpha-[(18)F]fluoroestradiol (4FMFES) by PET imaging in a breast cancer murine model. Mol. Imaging Biol. 2013, 15, 625–632. [CrossRef] [PubMed] 49. Mathias, C.J.; Welch, M.J.; Katzenellenbogen, J.A.; Brodack, J.W.; Kilbourn, M.R.; Carlson, K.E.; Kiesewetter, D.O. Characterization of the uptake of 16 alpha-([18F]fluoro)-17 beta-estradiol in DMBA-induced mammary tumors. Int. J. Rad. Appl. Instrum. B 1987, 14, 15–25. [CrossRef] 50. Katzenellenbogen, J.A.; Mathias, C.J.; VanBrocklin, H.F.; Brodack, J.W.; Welch, M.J. Titration of the in vivo uptake of 16 alpha-[18F]fluoroestradiol by target tissues in the rat: Competition by tamoxifen, and implications for quantitating estrogen receptors in vivo and the use of animal models in receptor-binding radiopharmaceutical development. Nucl. Med. Biol. 1993, 20, 735–745. [CrossRef][PubMed] 51. Mintun, M.; Welch, M.; Siegel, B.; Mathias, C.; Brodack, J.; McGuire, A.; Katzenellenbogen, J. Breast cancer: PET imaging of estrogen receptors. Radiology 1988, 169, 45–48. [CrossRef][PubMed] 52. Wagner, H.N., Jr. SNM Highlights as History: 1987. J. Nucl. Med. 2004, 45, N27–N33. 53. Bennink, R.J.; Rijks, L.J.; van Tienhoven, G.; Noorduyn, L.A.; Janssen, A.G.; Sloof, G.W.Estrogen receptor status in primary breast cancer: Iodine 123-labeled cis-11beta-methoxy-17alpha-iodovinyl estradiol scintigraphy. Radiology 2001, 220, 774–779. [CrossRef] 54. Bennink, R.J.; van Tienhoven, G.; Rijks, L.J.; Noorduyn, A.L.; Janssen, A.G.; Sloof, G.W. In vivo prediction of response to antiestrogen treatment in estrogen receptor-positive breast cancer. J. Nucl. Med. 2004, 45, 1–7. 55. Bénard, F.; Turcotte, É. Imaging in breast cancer: Single-photon computed tomography and positron-emission tomography. Breast Cancer Res. 2005, 7, 153. [CrossRef] 56. Desombre, E.R.; Pribish, J.; Hughes, A. Comparison of the distribution of radioiodinated di-and tri-hydroxyphenylethylene estrogens in the immature female rat. Nucl. Med. Biol. 1995, 22, 679–687. [CrossRef] 57. Yasui, L.S.; Hughes, A.; Desombre, E.R. DNA damage induction by 125I-estrogen. Acta Oncol. 1996, 35, 841–847. [CrossRef] 58. Hanson, R.N.; Tongcharoensirikul, P.;Barnsley,K.; Ondrechen, M.J.; Hughes, A.; DeSombre, E.R. Synthesis and evaluation of 2-halogenated-1, 1-bis (4-hydroxyphenyl)-2-(3-hydroxyphenyl)-ethylenes as potential estrogen receptor-targeted radiodiagnostic and radiotherapeutic agents. Steroids 2015, 96, 50–62. [CrossRef][PubMed] 59. Allott, L.; Smith, G.; Aboagye, E.O.; Carroll, L. PET imaging of steroid hormone receptor expression. Mol. Imaging 2015, 14.[CrossRef][PubMed] 60. Spitznagle, L.; Kasina, S.; Marino, C. Structure-distribution studies with fluorine-18 labeled pregnenolones: Effect of structure on adrenal uptake and elimination. J. Label. Compd. Radiopharm. 1982, 19, 1512–1513. 61. Spitznagle, L.; Marino, C.; Kasina, S. Design of radiopharmaceuticals for structure distribution studies: Some fluorine-18 labeled pregnane derivatives. Nucl. Med. Biol. 1981, 182, 28. 62. Spitznagle, L.A.; Marino, C.A. Synthesis of fluorine-18 labeled 21-fluoroprogesterone. Steroids 1977, 30, 435–438. [CrossRef] 63. Grill, H.; Heubner, A.; Pollow, K.; Manz, B.; Schulze, P.; Hofmeister, H.; Laurent, H.; Wiechert, R.; Elger, W. (Z)-17 beta-hydroxy-17 alpha-(2-[125I] iodovinyl)-4-estren-3-one: A new specific gamma-emitting ligand for determination of progesterone receptor. J. Clin. Chem. Clin. Biochem. Z. Fur Klin. Chem. Und Klin. Biochem. 1987, 25, 107–112. 64. Hoyte, R.; Rosner, W.; Johnson, I.; Zielinski, J.; Hochberg, R. Synthesis and evaluation of potential radioligands for the progesterone receptor. J. Med. Chem. 1985, 28, 1695–1699. [CrossRef] 65. Hochberg, R.B.; Hoyte, R.M.; Rosner, W. E-17 alpha-(2-[125I]iodovinyl)-19-nortestosterone: The synthesis of a gamma-emitting ligand for the progesterone receptor. Endocrinology 1985, 117, 2550–2552. [CrossRef] Cancers 2020, 12, 2020 26 of 32

66. Salman, M.; Chamness, G.C. A potential radioiodinated ligand for androgen receptor: 7. alpha.-methyl-17. alpha.-[2’-(E)-iodovinyl]-19-nortestosterone. J. Med. Chem. 1991, 34, 1019–1024. [CrossRef] 67. Van den Bos, J.C.; Rijks, L.J.; van Doremalen, P.A.; de Bruin, K.; Janssen, A.G.; van Royen, E.A. New iodinated progestins as potential ligands for progesterone receptor imaging in breast cancer. Part 1: Synthesis and in vitro pharmacological characterization. Nucl. Med. Biol. 1998, 25, 781–789. [CrossRef] 68. Rijks, L.J.; van den Bos, J.C.; van Doremalen, P.A.; Boer, G.J.; de Bruin, K.; Janssen, A.G.; van Royen, E.A. New iodinated progestins as potential ligands for progesterone receptor imaging in breast cancer. Part 2: In vivo pharmacological characterization. Nucl. Med. Biol. 1998, 25, 791–798. [CrossRef] 69. Brandes, S.J.; Katzenellenbogen, J.A. Fundamental considerations in the design of fluorine-18 labeled progestins and androgens as imaging agents for receptor-positive tumors of the breast and prostate. Int. J. Rad. Appl. Instrum. B 1988, 15, 53–67. [CrossRef] 70. Ojasoo, T.; Raynaud, J.-P. Unique steroid congeners for receptor studies. Cancer Res. 1978, 38, 4186–4198. [PubMed] 71. Zeelen, F.; van den Broek, A. Synthesis of 16α-ethyl-21-hydroxy-19-norpregn-4-ene-3, 20-dione (Org 2058). Recl. Des Trav. Chim. Des Pays Bas 1985, 104, 239–242. [CrossRef] 72. Carlson, K.E.; Brandes, S.J.; Pomper, M.G.; Katzenellenbogen, J.A. Uptake of three [3H] progestins by target tissues in vivo: Implications for the design of diagnostic imaging agents. Int. J. Radiat. Appl. Instrum. Part B Nucl. Med. Biol. 1988, 15, 403–408. [CrossRef] 73. Kontula, K.; Jänne, O.; Vihko, R.; de Jager, E.; de Visser, J.; Zeelen, F. Progesterone-binding proteins: In vitro binding and biological activity of different steroidal ligands. Eur. J. Endocrinol. 1975, 78, 574–592. [CrossRef] 74. Pomper, M.G.; Katzenellenbogen, J.A.; Welch, M.J.; Brodack, J.W.; Mathias, C.J. 21-[18F]fluoro-16 alpha-ethyl-19-norprogesterone: Synthesis and target tissue selective uptake of a progestin receptor based radiotracer for positron emission tomography. J. Med. Chem. 1988, 31, 1360–1363. [CrossRef] 75. Pomper, M.G.; Pinney, K.G.; Carlson, K.E.; Van Brocklin, H.; Mathias, C.J.; Welch, M.J.; Katzenellenbogen, J.A. Target tissue uptake selectivity of three fluorine-substituted progestins: Potential imaging agents for receptor-positive breast tumors. Int. J. Radiat. Appl. Instrum. Part B Nucl. Med. Biol. 1990, 17, 309–319. [CrossRef] 76. Dehdashti, F.; McGuire, A.H.; Van Brocklin, H.F.; Siegel, B.A.; Andriole, D.P.; Griffeth, L.K.; Pomper, M.G.; Katzenellenbogen, J.A.; Welch, M.J. Assessment of 21-[18F]fluoro-16 alpha-ethyl-19-norprogesterone as a positron-emitting radiopharmaceutical for the detection of progestin receptors in human breast carcinomas. J. Nucl. Med. 1991, 32, 1532–1537. 77. Verhagen, A.; Elsinga, P.H.; de Groot, T.J.; Paans, A.M.; de Goeij, C.C.; Sluyser, M.; Vaalburg, W. A fluorine-18 labeled progestin as an imaging agent for progestin receptor positive tumors with positron emission tomography. Cancer Res. 1991, 51, 1930–1933. [PubMed] 78. Verhagen, A.; Luurtsema, G.; Pesser, J.; De Groot, T.; Wouda, S.; Oosterhuis, J.; Vaalburg, W. Preclinical evaluation of a positron emitting progestin ([18F] fluoro-16α-methyl-19-norprogesterone) for imaging progesterone receptor positive tumours with positron emission tomography. Cancer Lett. 1991, 59, 125–132. [CrossRef] 79. Verhagen, A.; Studeny, M.; Luurtsema, G.; Visser, G.M.; De Goeij, C.C.; Sluyser, M.; Nieweg, O.E.; Van der Ploeg, E.; Go, K.G.; Vaalburg, W. Metabolism of a [18F]fluorine labeled progestin (21-[18F]fluoro-16 alpha-ethyl-19-norprogesterone) in humans: A clue for future investigations. Nucl. Med. Biol. 1994, 21, 941–952. [CrossRef] 80. Buckman, B.O.; Bonasera, T.A.; Kirschbaum, K.S.; Welch, M.J.; Katzenellenbogen, J.A. Fluorine-18-labeled progestin 16 alpha, 17 alpha-dioxolanes: Development of high-affinity ligands for the progesterone receptor with high in vivo target site selectivity. J. Med. Chem. 1995, 38, 328–337. [CrossRef] 81. Fried, J.; Sabo, E.; Grabowich, P.; Lerner, L.; Kessler, W.; Brennan, D.; Borman, A. Progestationally Active Acetals and Ketals of 16-Alpha,17-Alpha-Dihydroprogesterone. Chem. Ind. London 1961, 15, 465–466. 82. Lerner, L.J.; Brennan, D.M.; Borman, A. Biological activities of 16α, 17α dihydroxyprogesterone derivatives. Proc. Soc. Exp. Biol. Med. 1961, 106, 231–234. [CrossRef] 83. Kym, P.R.; Carlson, K.E.; Katzenellenbogen, J.A. Progestin 16. alpha., 17. alpha.-dioxolane ketals as molecular probes for the progesterone receptor: Synthesis, binding affinity, and photochemical evaluation. J. Med. Chem. 1993, 36, 1111–1119. [CrossRef] Cancers 2020, 12, 2020 27 of 32

84. Chan, S.R.; Fowler, A.M.; Allen, J.A.; Zhou, D.; Dence, C.S.; Sharp, T.L.; Fettig, N.M.; Dehdashti, F.; Katzenellenbogen, J.A. Longitudinal noninvasive imaging of progesterone receptor as a predictive biomarker of tumor responsiveness to estrogen deprivation therapy. Clin. Cancer Res. 2015, 21, 1063–1070. [CrossRef] 85. Fowler, A.M.; Chan, S.R.; Sharp, T.L.; Fettig, N.M.; Zhou, D.; Dence, C.S.; Carlson, K.E.; Jeyakumar, M.; Katzenellenbogen, J.A.; Schreiber, R.D.; et al. Small-animal PET of steroid hormone receptors predicts tumor response to endocrine therapy using a preclinical model of breast cancer. J. Nucl. Med. 2012, 53, 1119–1126. [CrossRef] 86. Basuli, F.; Zhang, X.; Blackman, B.; White, M.E.; Jagoda, E.M.; Choyke, P.L.; Swenson, R.E. Fluorine-18 Labeled Fluorofuranylnorprogesterone ([18F] FFNP) and Dihydrotestosterone ([18F] FDHT) Prepared by “Fluorination on Sep-Pak” Method. Molecules 2019, 24, 2389. [CrossRef] 87. Lee, J.H.; Zhou, H.-b.; Dence, C.S.; Carlson, K.E.; Welch, M.J.; Katzenellenbogen, J.A. Development of [F-18] fluorine-substituted Tanaproget as a progesterone receptor imaging agent for positron emission tomography. Bioconjug. Chem. 2010, 21, 1096–1104. [CrossRef][PubMed] 88. Pomper, M.G.; Kochanny, M.J.; Thieme, A.M.; Carlson, K.E.; Vanbrocklin, H.F.; Mathias, C.J.; Welch, M.J.; Katzenellenbogen, J.A. Fluorine-substituted corticosteroids: Synthesis and evaluation as potential receptor-based imaging agents for positron emission tomography of the brain. Int. J. Radiat. Appl. Instrum. Part B Nucl. Med. Biol. 1992, 19, 461–480. [CrossRef] 89. Volden, P.A.; Conzen, S.D. The influence of glucocorticoid signaling on tumor progression. Brain. Behav. Immun. 2013, 30, S26–S31. [CrossRef][PubMed] 90. Fensome, A.; Bender, R.; Chopra, R.; Cohen, J.; Collins, M.A.; Hudak, V.; Malakian, K.; Lockhead, S.; Olland, A.; Svenson, K. Synthesis and Structure Activity Relationship of Novel 6-Aryl-1, 4-dihydrobenzo [d][1,3] − oxazine-2-thiones as Progesterone Receptor Modulators Leading to the Potent and Selective Nonsteroidal Progesterone Receptor Agonist Tanaproget. J. Med. Chem. 2005, 48, 5092–5095. [CrossRef] 91. Zhou, H.-B.; Lee, J.H.; Mayne, C.G.; Carlson, K.E.; Katzenellenbogen, J.A. Imaging progesterone receptor in breast tumors: Synthesis and receptor binding affinity of fluoroalkyl-substituted analogues of tanaproget. J. Med. Chem. 2010, 53, 3349–3360. [CrossRef] 92. Allott, L.; Miranda, C.; Hayes, A.; Raynaud, F.; Cawthorne, C.; Smith, G. Synthesis of a benzoxazinthione derivative of tanaproget and pharmacological evaluation for PET imaging of PR expression. EJNMMI Radiopharm. Chem. 2019, 4, 1. [CrossRef] 93. Merchant, S.; Allott, L.; Carroll, L.; Tittrea, V.; Kealey, S.; Witney, T.H.; Miller, P.W.; Smith, G.; Aboagye, E.O. Synthesis and pre-clinical evaluation of a [18 F] fluoromethyl-tanaproget derivative for imaging of progesterone receptor expression. RSC Adv. 2016, 6, 57569–57579. [CrossRef] 94. Zhou, D.; Carlson, K.E.; Katzenellenbogen, J.A.; Welch, M.J. Bromine-and iodine-substituted 16α, 17α-dioxolane progestins for breast tumor imaging and radiotherapy: Synthesis and receptor binding affinity. J. Med. Chem. 2006, 49, 4737–4744. [CrossRef] 95. Auwerx, J.; Baulieu, E.; Beato, M.; Becker-Andre, M.; Burbach, P.; Camerino, G.; Chambon, P.; Cooney, A.; Dejean, A.; Dreyer, C. A unified nomenclature system for the nuclear receptor superfamily. Cell 1999, 97, 161–163. 96. Leo, J.C.; Guo, C.; Woon, C.T.; Aw, S.E.; Lin, V.C. Glucocorticoid and mineralocorticoid cross-talk with progesterone receptor to induce focal adhesion and growth inhibition in breast cancer cells. Endocrinology 2004, 145, 1314–1321. [CrossRef] 97. Tonsing-Carter, E.; Hernandez, K.M.; Kim, C.R.; Harkless, R.V.; Oh, A.; Bowie, K.R.; West-Szymanski, D.C.; Betancourt-Ponce, M.A.; Green, B.D.; Lastra, R.R. Glucocorticoid receptor modulation decreases ER-positive breast cancer cell proliferation and suppresses wild-type and mutant ER chromatin association. Breast Cancer Res. 2019, 21, 82. [CrossRef][PubMed] 98. Steiniger, B.; Kniess, T.; Bergmann, R.; Pietzsch, J.; Wuest, F.R. Radiolabeled glucocorticoids as molecular probes for imaging brain glucocorticoid receptors by means of positron emission tomography (PET). Mini Rev. Med. Chem. 2008, 8, 728–739. [CrossRef][PubMed] 99. Wuest, F.; Kniess, T.; Henry, B.; Peeters, B.W.; Wiegerinck, P.H.; Pietzsch, J.; Bergmann, R. Radiosynthesis and radiopharmacological evaluation of [N-methyl-11C] Org 34850 as a glucocorticoid receptor (GR)-binding radiotracer. Appl. Radiat. Isot. 2009, 67, 308–312. [CrossRef][PubMed]

100. Hoyte, R.M.; Labaree, D.C.; Fede, J.-M.; Harris, C.; Hochberg, R.B. Iodinated and fluorinated steroid 20-aryl-[3, 2-c] pyrazoles as potential glucocorticoid receptor imaging agents. Steroids 1998, 63, 595–602. [CrossRef] Cancers 2020, 12, 2020 28 of 32

101. Wuest, F.; Carlson, K.E.; Katzenellenbogen, J.A. Expeditious synthesis of steroids containing a 2-methylsulfanyl-acetyl side chain as potential glucocorticoid receptor imaging agents. Steroids 2008, 73, 69–76. [CrossRef] 102. Eakins, M.; Waters, S. The synthesis of 77Br-labelled 5α-dihydrotestosterone and a comparison of its distribution in rats with 77Br-bromide. Int. J. Appl. Radiat. Isot. 1979, 30, 701–703. [CrossRef] 103. Ghanadian, R.; Waters, S.; Chisholm, G. Investigations into the use of 77 Br labelled 5α-dihydrotestosterone for scanning the prostate. Eur. J. Nucl. Med. 1977, 2, 155–157. [CrossRef] 104. Hoyte, R.; Rosner, W.; Hochberg, R. Synthesis of 16α-[125I] iodo-5α-dihydrotestosterone and evaluation of its affinity for the androgen receptor. J. Steroid Biochem. 1982, 16, 621–628. [CrossRef] 105. Tarle, M.; Padovan, R.; Spaventi, Š. The uptake of radioiodinated 5α-dihydrotestosterone by the prostate of intact and castrated rats. Eur. J. Nucl. Med. 1981, 6, 79–83. [CrossRef] 106. Hoyte, R.M.; MacLusky, N.J.; Hochberg, R.B. The synthesis and testing of E-17α-(2-iodovinyl)-5α-dihydrotestosterone and Z-17α-(2-iodovinyl)-5α-dihydrotestosterone as γ-emitting ligands for the androgen receptor. J. Steroid Biochem. 1990, 36, 125–132. [CrossRef] 107. Ali, H.; Rousseau, J.; Ahmed, N.; Guertin, V.; Hochberg, R.B.; van Lier, J.E. Synthesis of the 7α-cyano-(17α, 20E/Z)-[125I] iodovinyl-19-nortestosterones: Potential radioligands for androgen and progesterone receptors. Steroids 2003, 68, 1163–1171. [CrossRef][PubMed] 108. Schänzer, W. Metabolism of anabolic androgenic steroids. Clin. Chem. 1996, 42, 1001–1020. [CrossRef] [PubMed] 109. Liu, A.; Carlson, K.E.; Katzenellenbogen, J.A. Synthesis of high-affinity fluorine-substituted ligands for the androgen receptor. Potential agents for imaging prostatic cancer by positron emission tomography. J. Med. Chem. 1992, 35, 2113–2129. [CrossRef][PubMed] 110. Choe, Y.S.; Katzenellenbogen, J.A. Synthesis of C-6 fluoroandrogens: Evaluation of ligands for tumor receptor imaging. Steroids 1995, 60, 414–422. [CrossRef] 111. Liu, A.; Katzenellenbogen, J.A.; VanBrocklin, H.F.; Mathias, C.J.; Welch, M.J. 20-[18F] fluoromibolerone, a positron-emitting radiotracer for androgen receptors: Synthesis and tissue distribution studies. J. Nucl. Med. Off. Publ. Soc. Nucl. Med. 1991, 32, 81–88. 112. Choe, Y.S.; Lidstroem, P.J.; Chi, D.Y.; Bonasera, T.A.; Welch, M.J.; Katzenellenbogen, J.A. Synthesis of 11. beta.-[18F] Fluoro-5. alpha.-dihydrotestosterone and 11. beta.-[18F] Fluoro-19-nor-5. alpha.-dihydrotestosterone: Preparation via halofluorination-reduction, receptor binding, and tissue distribution. J. Med. Chem. 1995, 38, 816–825. [CrossRef] 113. Carlson, K.E.; Katzenellenbogen, J.A. A comparative study of the selectivity and efficiency of target tissue uptake of five tritium-labeled androgens in the rat. J. Steroid Biochem. 1990, 36, 549–561. [CrossRef] 114. Liu, A.; Dence, C.S.; Welch, M.J.; Katzenellenbogen, J.A. Fluorine-18-labeled androgens: Radiochemical synthesis and tissue distribution studies on six fluorine-substituted androgens, potential imaging agents for prostatic cancer. J. Nucl. Med. 1992, 33, 724–734. 115. Larson, S.M.; Morris, M.; Gunther, I.; Beattie, B.; Humm, J.L.; Akhurst, T.A.; Finn, R.D.; Erdi, Y.; Pentlow, K.; Dyke, J. Tumor localization of 16β-18F-fluoro-5α-dihydrotestosterone versus 18F-FDG in patients with progressive, metastatic prostate cancer. J. Nucl. Med. 2004, 45, 366–373. 116. Beattie, B.J.; Smith-Jones, P.M.; Jhanwar, Y.S.; Schöder, H.; Schmidtlein, C.R.; Morris, M.J.; Zanzonico, P.; Squire, O.; Meirelles, G.S.; Finn, R. Pharmacokinetic assessment of the uptake of 16β-18F-fluoro-5α-dihydrotestosterone (FDHT) in prostate tumors as measured by PET. J. Nucl. Med. 2010, 51, 183–192. [CrossRef] 117. Pippione, A.C.; Carnovale, I.M.; Bonanni, D.; Sini, M.; Goyal, P.; Marini, E.; Pors, K.; Adinolfi, S.; Zonari, D.; Festuccia, C. Potent and selective aldo-keto reductase 1C3 (AKR1C3) inhibitors based on the benzoisoxazole moiety: Application of a bioisosteric scaffold hopping approach to flufenamic acid. Eur. J. Med. Chem. 2018, 150, 930–945. [CrossRef][PubMed] 118. Naslund, M.J.; Miner, M. A review of the clinical efficacy and safety of 5α-reductase inhibitors for the enlarged prostate. Clin. Ther. 2007, 29, 17–25. [CrossRef][PubMed] 119. Downer, J.B.; Jones, L.A.; Engelbach, J.A.; Lich, L.L.; Mao, W.; Carlson, K.E.; Katzenellenbogen, J.A.; Welch, M.J. Comparison of animal models for the evaluation of radiolabeled androgens. Nucl. Med. Biol. 2001, 28, 613–626. [CrossRef] Cancers 2020, 12, 2020 29 of 32

120. Nevedomskaya, E.; Baumgart, S.J.; Haendler, B. Recent Advances in Prostate Cancer Treatment and Drug Discovery. Int. J. Mol. Sci. 2018, 19, 1359. [CrossRef] 121. Teutsch, G.; Goubet, F.; Battmann, T.; Bonfils, A.; Bouchoux, F.; Cerede, E.; Gofflo, D.; Gaillard-Kelly, M.; Philibert, D. Non-steroidal antiandrogens: Synthesis and biological profile of high-affinity ligands for the androgen receptor. J. Steroid Biochem. Mol. Biol. 1994, 48, 111–119. [CrossRef] 122. Parent, E.E.; Dence, C.S.; Jenks, C.; Sharp, T.L.; Welch, M.J.; Katzenellenbogen, J.A. Synthesis and biological evaluation of [18F] bicalutamide, 4-[76Br] bromobicalutamide, and 4-[76Br] bromo-thiobicalutamide as non-steroidal androgens for prostate cancer imaging. J. Med. Chem. 2007, 50, 1028–1040. [CrossRef] 123. Parent, E.E.; Jenks, C.; Sharp, T.; Welch, M.J.; Katzenellenbogen, J.A. Synthesis and biological evaluation of a nonsteroidal bromine-76-labeled androgen receptor ligand 3-[76Br] bromo-hydroxyflutamide. Nucl. Med. Biol. 2006, 33, 705–713. [CrossRef] 124. Parent, E.E.; Dence, C.S.; Sharp, T.L.; Welch, M.J.; Katzenellenbogen, J.A. Synthesis and biological evaluation of a fluorine-18-labeled nonsteroidal androgen receptor antagonist, N-(3-[18F] fluoro-4-nitronaphthyl)-cis-5-norbornene-endo-2, 3-dicarboxylic imide. Nucl. Med. Biol. 2006, 33, 615–624. [CrossRef] 125. Linden, H.M.; Kurland, B.F.; Peterson, L.M.; Schubert, E.K.; Gralow, J.R.; Specht, J.M.; Ellis, G.K.; Lawton, T.J.; Livingston, R.B.; Petra, P.H.; et al. Fluoroestradiol positron emission tomography reveals differences in pharmacodynamics of aromatase inhibitors, tamoxifen, and fulvestrant in patients with metastatic breast cancer. Clin. Cancer Res. 2011, 17, 4799–4805. [CrossRef] 126. van Kruchten, M.; de Vries, E.G.; Glaudemans, A.W.; van Lanschot, M.C.; van Faassen, M.; Kema, I.P.; Brown, M.; Schröder, C.P.; de Vries, E.F.; Hospers, G.A. Measuring residual estrogen receptor availability during fulvestrant therapy in patients with metastatic breast cancer. Cancer Discov. 2015, 5, 72–81. [CrossRef] 127. Heidari, P.; Deng, F.; Esfahani, S.A.; Leece, A.K.; Shoup, T.M.; Vasdev, N.; Mahmood, U. Pharmacodynamic imaging guides dosing of a selective estrogen receptor degrader. Clin. Cancer Res. 2015, 21, 1340–1347. [CrossRef][PubMed] 128. Rathkopf, D.E.; Morris, M.J.; Fox, J.J.; Danila, D.C.; Slovin, S.F.; Hager, J.H.; Rix, P.J.; Maneval, E.C.; Chen, I.; Gönen, M. Phase I study of ARN-509, a novel antiandrogen, in the treatment of castration-resistant prostate cancer. J. Clin. Oncol. 2013, 31, 3525. [CrossRef][PubMed] 129. Scher, H.I.; Beer, T.M.; Higano, C.S.; Anand, A.; Taplin, M.-E.; Efstathiou, E.; Rathkopf, D.; Shelkey, J.; Evan, Y.Y.; Alumkal, J. Antitumour activity of MDV3100 in castration-resistant prostate cancer: A phase 1–2 study. Lancet 2010, 375, 1437–1446. [CrossRef] 130. Beatson, G.T. Meeting IX.—May 20, 1896: On the Treatment of Inoperable Cases of Carcinoma of the Mamma: Suggestions for a New Method of Treatment, with Illustrative Cases. Trans. Med. Chir. Soc. Edinb. 1896, 15, 153. 131. Clarke, M.J. Ovarian ablation in breast cancer, 1896 to 1998: Milestones along hierarchy of evidence from case report to Cochrane review. BMJ 1998, 317, 1246–1248. [CrossRef] 132. Kaufmann, M. A review of endocrine options for the treatment of advanced breast cancer. Oncology 1997, 54, 2–5. [CrossRef] 133. Kumar, M.; Salem, K.; Tevaarwerk, A.J.; Strigel, R.M.; Fowler, A.M. Recent Advances in Imaging Steroid Hormone Receptors in Breast Cancer. J. Nucl. Med. 2020, 61, 172–176. [CrossRef] 134. van Kruchten, M.; de Vries, E.G.; Brown, M.; de Vries, E.F.; Glaudemans, A.W.; Dierckx, R.A.; Schröder, C.P.; Hospers, G.A. PET imaging of oestrogen receptors in patients with breast cancer. Lancet Oncol. 2013, 14, e465–e475. [CrossRef] 135. Harvey, J.M.; Clark, G.M.; Osborne, C.K.; Allred, D.C. Estrogen receptor status by immunohistochemistry is superior to the ligand-binding assay for predicting response to adjuvant endocrine therapy in breast cancer. J. Clin. Oncol. 1999, 17, 1474–1481. [CrossRef] 136. King, W.J.; DeSombre, E.R.; Jensen, E.V.; Greene, G.L. Comparison of immunocytochemical and steroid-binding assays for estrogen receptor in human breast tumors. Cancer Res. 1985, 45, 293–304. 137. Fowler, A.M.; Clark, A.S.; Katzenellenbogen, J.A.; Linden, H.M.; Dehdashti, F. Imaging Diagnostic and Therapeutic Targets-Steroid Receptors in Breast Cancer. J. Nucl. Med. 2016, 57, 75S. [CrossRef][PubMed] 138. Gertych, A.; Mohan, S.; Maclary, S.; Mohanty, S.; Wawrowsky, K.; Mirocha, J.; Balzer, B.; Knudsen, B.S. Effects of tissue decalcification on the quantification of breast cancer biomarkers by digital image analysis. Diagn. Pathol. 2014, 9, 213. [CrossRef][PubMed] Cancers 2020, 12, 2020 30 of 32

139. Kurland, B.F.; Peterson, L.M.; Lee, J.H.; Linden, H.M.; Schubert, E.K.; Dunnwald, L.K.; Link, J.M.; Krohn, K.A.; Mankoff, D.A. Between-patient and within-patient (site-to-site) variability in estrogen receptor binding, measured in vivo by 18F-fluoroestradiol PET. J. Nucl. Med. 2011, 52, 1541–1549. [CrossRef][PubMed] 140. Nienhuis, H.H.; van Kruchten, M.; Elias, S.G.; Glaudemans, A.; de Vries, E.F.J.; Bongaerts, A.H.H.; Schroder, C.P.; de Vries, E.G.E.; Hospers, G.A.P. (18)F-Fluoroestradiol Tumor Uptake Is Heterogeneous and Influenced by Site of Metastasis in Breast Cancer Patients. J. Nucl. Med. 2018, 59, 1212–1218. [CrossRef] 141. Amir, E.; Miller, N.; Geddie, W.; Freedman, O.; Kassam, F.; Simmons, C.; Oldfield, M.; Dranitsaris, G.; Tomlinson, G.; Laupacis, A. Prospective study evaluating the impact of tissue confirmation of metastatic disease in patients with breast cancer. J. Clin. Oncol. 2012, 30, 587. [CrossRef] 142. Gerlinger, M.; Rowan, A.J.; Horswell, S.; Larkin, J.; Endesfelder, D.; Gronroos, E.; Martinez, P.; Matthews, N.; Stewart, A.; Tarpey, P. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N. Engl. J. Med. 2012, 366, 883–892. [CrossRef] 143. McGuire, A.H.; Dehdashti, F.; Siegel, B.A.; Lyss, A.P.; Brodack, J.W.; Mathias, C.J.; Mintun, M.A.; Katzenellenbogen, J.A.; Welch, M.J. Positron tomographic assessment of 16 alpha-[18F] fluoro-17 beta-estradiol uptake in metastatic breast carcinoma. J. Nucl. Med. 1991, 32, 1526–1531. 144. Dehdashti, F.; Mortimer, J.E.; Siegel, B.A.; Griffeth, L.K.; Bonasera, T.J.; Fusselman, M.J.; Detert, D.D.; Cutler, P.D.; Katzenellenbogen, J.A.; Welch, M.J. Positron tomographic assessment of estrogen receptors in breast cancer: Comparison with FDG-PET and in vitro receptor assays. J. Nucl. Med. 1995, 36, 1766–1774. 145. Mortimer, J.E.; Dehdashti, F.; Siegel, B.A.; Katzenellenbogen, J.A.; Fracasso, P.; Welch, M.J. Positron emission tomography with 2-[18F]Fluoro-2-deoxy-D-glucose and 16alpha-[18F]fluoro-17beta-estradiol in breast cancer: Correlation with estrogen receptor status and response to systemic therapy. Clin. Cancer Res. 1996, 2, 933–939. 146. Park, J.H.; Kang, M.J.; Ahn, J.-H.; Kim, J.E.; Jung, K.H.; Gong, G.; Lee, H.J.; Son, B.-H.; Ahn, S.-H.; Kim, H.-H. Phase II trial of neoadjuvant letrozole and lapatinib in Asian postmenopausal women with estrogen receptor (ER) and human epidermal growth factor receptor 2 (HER2)-positive breast cancer [Neo-ALL-IN]: Highlighting the TILs, ER expressional change after neoadjuvant treatment, and FES-PET as potential significant biomarkers. Cancer Chemother. Pharmacol. 2016, 78, 685–695. 147. Kurland, B.F.; Peterson, L.M.; Lee, J.H.; Schubert, E.K.; Currin, E.R.; Link, J.M.; Krohn, K.A.; Mankoff, D.A.; Linden, H.M. Estrogen Receptor Binding (18F-FES PET) and Glycolytic Activity (18F-FDG PET) Predict Progression-Free Survival on Endocrine Therapy in Patients with ER+ Breast Cancer. Clin. Cancer Res. 2017, 23, 407–415. [CrossRef][PubMed] 148. van Kruchten, M.; Glaudemans, A.W.; de Vries, E.F.; Beets-Tan, R.G.; Schroder, C.P.; Dierckx, R.A.; de Vries, E.G.; Hospers, G.A. PET imaging of estrogen receptors as a diagnostic tool for breast cancer patients presenting with a clinical dilemma. J. Nucl. Med. 2012, 53, 182–190. [CrossRef] 149. Linden, H.M.; Peterson, L.M.; Fowler, A.M. Clinical Potential of Estrogen and Progesterone Receptor Imaging. PET Clin. 2018, 13, 415–422. [CrossRef][PubMed] 150. Liao, G.J.; Clark, A.S.; Schubert, E.K.; Mankoff, D.A. 18F-Fluoroestradiol PET: Current Status and Potential Future Clinical Applications. J. Nucl. Med. 2016, 57, 1269–1275. [CrossRef][PubMed] 151. Evangelista, L.; Guarneri, V.; Conte, P.F. 18F-Fluoroestradiol Positron Emission Tomography in Breast Cancer Patients: Systematic Review of the Literature & Meta-Analysis. Curr. Radiopharm. 2016, 9, 244–257. [CrossRef][PubMed] 152. Chae, S.Y.; Kim, S.-B.; Ahn, S.H.; Kim, H.O.; Yoon, D.H.; Ahn, J.-H.; Jung, K.H.; Han, S.; Oh, S.J.; Lee, S.J. A randomized feasibility study of 18F-Fluoroestradiol PET to predict pathologic response to neoadjuvant therapy in estrogen receptor–rich postmenopausal breast cancer. J. Nucl. Med. 2017, 58, 563–568. [CrossRef] 153. Venema, C.M.; Apollonio, G.; Hospers, G.A.; Schröder, C.P.; Dierckx, R.A.; de Vries, E.F.; Glaudemans, A.W. Recommendations and technical aspects of 16α-[18F] fluoro-17β-estradiol PET to image the estrogen receptor in vivo: The Groningen experience. Clin. Nucl. Med. 2016, 41, 844–851. [CrossRef] 154. Koleva-Kolarova, R.; Greuter, M.; Van Kruchten, M.; Vermeulen, K.; Feenstra, T.; Buskens, E.; Glaudemans, A.; De Vries, E.; De Vries, E.; Hospers, G. The value of PET/CT with FES or FDG tracers in metastatic breast cancer: A computer simulation study in ER-positive patients. Br. J. Cancer 2015, 112, 1617–1625. [CrossRef] 155. Kurland, B.F.; Wiggins, J.R.; Coche, A.; Fontan, C.; Bouvet, Y.; Webner, P.; Divgi, C.; Linden, H.M. Whole-Body Characterization of Estrogen Receptor Status in Metastatic Breast Cancer with 16α-18F-Fluoro-17β-Estradiol Positron Emission Tomography: Meta-Analysis and Recommendations for Integration into Clinical Applications. Oncologist 2020, 25, 1–10. [CrossRef] Cancers 2020, 12, 2020 31 of 32

156. Sparano, J.A.; Gray, R.J.; Makower, D.F.; Albain, K.S.; Saphner, T.J.; Badve, S.S.; Wagner, L.I.; Kaklamani, V.G.; Keane, M.M.; Gomez, H.L.; et al. Clinical Outcomes in Early Breast Cancer With a High 21-Gene Recurrence Score of 26 to 100 Assigned to Adjuvant Chemotherapy Plus Endocrine Therapy: A Secondary Analysis of the TAILORx Randomized Clinical Trial. JAMA Oncol. 2019.[CrossRef] 157. Mouridsen, H.; Palshof, T.; Patterson, J.; Battersby, L. Tamoxifen in advanced breast cancer. Cancer Treat. Rev. 1978, 5, 131–141. [CrossRef] 158. Wittliff, J.L. Steroid-hormone receptors in breast cancer. Cancer 1984, 53, 630–643. [CrossRef] 159. Kraus, W.L.; Katzenellenbogen, B.S. Regulation of progesterone receptor gene expression and growth in the rat uterus: Modulation of estrogen actions by progesterone and sex steroid hormone antagonists. Endocrinology 1993, 132, 2371–2379. [CrossRef][PubMed] 160. Eckert, R.L.; Katzenellenbogen, B.S. Effects of estrogens and antiestrogens on estrogen receptor dynamics and the induction of progesterone receptor in MCF-7 human breast cancer cells. Cancer Res. 1982, 42, 139–144. [PubMed] 161. May, F.E.; Johnson, M.D.; Wiseman, L.R.; Wakeling, A.E.; Kastner, P.; Westley, B.R. Regulation of progesterone receptor mRNA by oestradiol and antioestrogens in breast cancer cell lines. J. Steroid Biochem. 1989, 33, 1035–1041. [CrossRef] 162. Horwitz, K.B.; McGuire, W.L. Predicting response to endocrine therapy in human breast cancer: A hypothesis. Science 1975, 189, 726–727. [CrossRef] 163. McGuire, W.L.; Horwitz, K.B.; Pearson, O.H.; Segaloff, A. Current status of estrogen and progesterone receptors in breast cancer. Cancer 1977, 39, 2934–2947. [CrossRef] 164. Boland, M.; Ryan, É.; Dunne, E.; Aherne, T.; Bhatt, N.; Lowery, A. Meta-analysis of the impact of progesterone receptor status on oncological outcomes in oestrogen receptor-positive breast cancer. Br. J. Surg. 2020, 107, 33–43. [CrossRef] 165. Purdie, C.; Quinlan, P.; Jordan, L.; Ashfield, A.; Ogston, S.; Dewar, J.; Thompson, A. Progesterone receptor expression is an independent prognostic variable in early breast cancer: A population-based study. Br. J. Cancer 2014, 110, 565. [CrossRef] 166. Inda, M.A.; Blok, E.J.; Kuppen, P.J.;Charehbili, A.; den Biezen-Timmermans, E.C.; van Brussel, A.; Fruytier, S.E.; Kranenbarg, E.M.-K.; Kloet, S.L.; van der Burg, B. Estrogen Receptor pathway activity score to predict clinical response or resistance to neo-adjuvant endocrine therapy in primary breast cancer. Mol. Cancer Ther. 2019. 167. Wu, J.-R.; Zhao, Y.; Zhou, X.-P.; Qin, X. Estrogen receptor 1 and progesterone receptor are distinct biomarkers and prognostic factors in estrogen receptor-positive breast cancer: Evidence from a bioinformatic analysis. Biomed. Pharmacother. 2020, 121, 109647. [CrossRef][PubMed] 168. Li, Y.; Yang, D.; Yin, X.; Zhang, X.; Huang, J.; Wu, Y.; Wang, M.; Yi, Z.; Li, H.; Li, H. Clinicopathological Characteristics and Breast Cancer–Specific Survival of Patients With Single Hormone Receptor–Positive Breast Cancer. JAMA Netw. Open 2020, 3, e1918160. [CrossRef][PubMed] 169. Allison, K.H.; Hammond, M.E.H.; Dowsett, M.; McKernin, S.E.; Carey, L.A.; Fitzgibbons, P.L.; Hayes, D.F.; Lakhani, S.R.; Chavez-MacGregor, M.; Perlmutter, J. Estrogen and Progesterone Receptor Testing in Breast Cancer: American Society of Clinical Oncology/College of American Pathologists Guideline Update. Arch. Pathol. Lab. Med. 2020, 144, 545–563. [CrossRef][PubMed] 170. Chan, S.R.; Vermi, W.; Luo, J.; Lucini, L.; Rickert, C.; Fowler, A.M.; Lonardi, S.; Arthur, C.; Young, L.J.; Levy, D.E. STAT1-deficient mice spontaneously develop estrogen receptor α-positive luminal mammary carcinomas. Breast Cancer Res. 2012, 14, R16. [CrossRef][PubMed] 171. Salem, K.; Kumar, M.; Yan, Y.; Jeffery, J.J.; Kloepping, K.C.; Michel, C.J.; Powers, G.L.; Mahajan, A.M.; Fowler, A.M. Sensitivity and isoform specificity of 18F-fluorofuranylnorprogesterone for measuring progesterone receptor protein response to estradiol challenge in breast cancer. J. Nucl. Med. 2019, 60, 220–226. [CrossRef] 172. Haddow, A.; Watkinson, J.M.; Paterson, E.; Koller, P.C. Influence of Synthetic Oestrogens on Advanced Malignant Disease. Br. Med. J. 1944, 2, 393–398. [CrossRef] 173. Carter, A.C.; Sedransk, N.; Kelley, R.M.; Ansfield, F.J.; Ravdin, R.G.; Talley, R.W.; Potter, N.R. Diethylstilbestrol: Recommended dosages for different categories of breast cancer patients: Report of the Cooperative Breast Cancer Group. JAMA 1977, 237, 2079–2085. [CrossRef] 174. Ingle, J.N.; Ahmann, D.L.; Green, S.J.; Edmonson, J.H.; Bisel, H.F.; Kvols, L.K.; Nichols, W.C.; Creagan, E.T.; Hahn, R.G.; Rubin, J. Randomized clinical trial of diethylstilbestrol versus tamoxifen in postmenopausal women with advanced breast cancer. N. Engl. J. Med. 1981, 304, 16–21. [CrossRef] Cancers 2020, 12, 2020 32 of 32

175. Lønning, P.E.; Taylor, P.D.; Anker, G.; Iddon, J.; Wie, L.; Jørgensen, L.-M.; Mella, O.; Howell, A. High-dose estrogen treatment in postmenopausal breast cancer patients heavily exposed to endocrine therapy. Breast Cancer Res. Treat. 2001, 67, 111–116. [CrossRef] 176. Ellis, M.J.; Gao, F.; Dehdashti, F.; Jeffe, D.B.; Marcom, P.K.; Carey, L.A.; Dickler, M.N.; Silverman, P.; Fleming, G.F.; Kommareddy,A. Lower-dose vs. high-dose oral estradiol therapy of hormone receptor–positive, aromatase inhibitor–resistant advanced breast cancer: A phase 2 randomized study. JAMA 2009, 302, 774–780. [CrossRef] 177. Plotkin, D.; Lechner, J.J.; Jung, W.E.; Rosen, P.J. Tamoxifen flare in advanced breast cancer. JAMA 1978, 240, 2644–2646. [CrossRef] 178. Reddel, R.R.; Sutherland, R.L. Tamoxifen stimulation of human breast cancer cell proliferation in vitro:A possible model for tamoxifen tumour flare. Eur. J. Cancer Clin. Oncol. 1984, 20, 1419–1424. [CrossRef] 179. Katzenellenbogen, B.S.; Kendra, K.L.; Norman, M.J.; Berthois, Y. Proliferation, hormonal responsiveness, and estrogen receptor content of MCF-7 human breast cancer cells grown in the short-term and long-term absence of estrogens. Cancer Res. 1987, 47, 4355–4360. [PubMed] 180. Fabian, C.; Sternson, L.; El-Serafi, M.; Cain, L.; Hearne, E. Clinical pharmacology of tamoxifen in patients with breast cancer: Correlation with clinical data. Cancer 1981, 48, 876–882. [CrossRef] 181. Katzenellenbogen, J.A.; Coleman, R.E.; Hawkins, R.A.; Krohn, K.A.; Larson, S.M.; Mendelsohn, J.; Osborne, C.K.; Piwnica-Worms, D.; Reba, R.C.; Siegel, B.A. Tumor receptor imaging: Proceedings of the National Cancer Institute workshop, review of current work, and prospective for further investigations. Clin. Cancer Res. 1995, 1, 921–932. [PubMed] 182. Mortimer, J.E.; Dehdashti, F.; Siegel, B.A.; Trinkaus, K.; Katzenellenbogen, J.A.; Welch, M.J. Metabolic flare: Indicator of hormone responsiveness in advanced breast cancer. J. Clin. Oncol. 2001, 19, 2797–2803. [CrossRef] 183. Dehdashti, F.; Flanagan, F.L.; Mortimer, J.E.; Katzenellenbogen, J.A.; Welch, M.J.; Siegel, B.A. Positron emission tomographic assessment of “metabolic flare” to predict response of metastatic breast cancer to antiestrogen therapy. Eur. J. Nucl. Med. 1999, 26, 51–56. [CrossRef] 184. Dehdashti, F.; Mortimer, J.E.; Trinkaus, K.; Naughton, M.J.; Ellis, M.; Katzenellenbogen, J.A.; Welch, M.J.; Siegel, B.A. PET-based estradiol challenge as a predictive biomarker of response to endocrine therapy in women with estrogen-receptor-positive breast cancer. Breast Cancer Res. Treat. 2009, 113, 509–517. [CrossRef] 185. Fox, J.J.; Gavane, S.C.; Blanc-Autran, E.; Nehmeh, S.; Gönen, M.; Beattie, B.; Vargas, H.A.; Schöder, H.; Humm, J.L.; Fine, S.W. Positron emission tomography/computed tomography–based assessments of androgen receptor expression and glycolytic activity as a prognostic biomarker for metastatic castration-resistant prostate cancer. JAMA Oncol. 2018, 4, 217–224. [CrossRef] 186. Chaturvedi, A.P.; Dehm, S.M. Androgen Receptor Dependence. Adv. Exp. Med. Biol. 2019, 1210, 333–350. [CrossRef] 187. Montironi, R.; Cimadamore, A.; Lopez-Beltran, A.; Scarpelli, M.; Aurilio, G.; Santoni, M.; Massari, F.; Cheng, L. Morphologic, Molecular and Clinical Features of Aggressive Variant Prostate Cancer. Cells 2020, 9, 1073. [CrossRef][PubMed] 188. Sheikhbahaei, S.; Afshar-Oromieh, A.; Eiber, M.; Solnes, L.B.; Javadi, M.S.; Ross, A.E.; Pienta, K.J.; Allaf, M.E.; Haberkorn, U.; Pomper, M.G. Pearls and pitfalls in clinical interpretation of prostate-specific membrane antigen (PSMA)-targeted PET imaging. Eur. J. Nucl. Med. Mol. Imaging 2017, 44, 2117–2136. [CrossRef] [PubMed] 189. Han, S.; Woo, S.; Kim, Y.J.; Suh, C.H. Impact of 68Ga-PSMA PET on the management of patients with prostate cancer: A systematic review and meta-analysis. Eur. Urol. 2018, 74, 179–190. [CrossRef][PubMed] 190. Zang, J.; Liu, Q.; Sui, H.; Wang, R.; Jacobson, O.; Fan, X.; Zhu, Z.; Chen, X. (177)Lu-EB-PSMA radioligand therapy with escalating doses in patients with metastatic castration-resistant prostate cancer. J. Nucl. Med. 2020.[CrossRef] 191. Shen, C.J.; Minn, I.; Hobbs, R.F.; Chen, Y.; Josefsson, A.; Brummet, M.; Banerjee, S.R.; Brayton, C.F.; Mease, R.C.; Pomper, M.G.; et al. Auger radiopharmaceutical therapy targeting prostate-specific membrane antigen in a micrometastatic model of prostate cancer. Theranostics 2020, 10, 2888–2896. [CrossRef] 192. Wagner, H.N., Jr.; Conti, P.S. Advances in medical imaging for cancer diagnosis and treatment. Cancer 1991, 67, 1121–1128. [CrossRef]

© 2020 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).