I'm asking Denmeade about this. Too early to rely on it. I haven't seen even low-quality studies.
Doxazosin and Prazosin and BAT synergy
Theoretical Basis for Synergy
Doxazosin, a quinazoline α₁-adrenergic antagonist, has demonstrated pro-apoptotic effects in prostate cancer cells by directly damaging DNA and disrupting cell-cycle progression. At the molecular level, doxazosin can intercalate into DNA and inhibit topoisomerase function, leading to accumulation of DNA double-strand breaks (DSBs)
mdpi.com
. As a result, cells treated with doxazosin exhibit activation of DNA damage response pathways (e.g. ATM/ATR) and arrest in the G₂/M phase of the cell cycle
pubmed.ncbi.nlm.nih.gov
. Notably, doxazosin (and similar α₁-blockers like prazosin) causes inactivation of Cdk1 via enhanced phosphorylation and nuclear exclusion of the Cdc25c phosphatase, enforcing the G₂ checkpoint and preventing mitotic entry
pubmed.ncbi.nlm.nih.gov
. This G₂ arrest gives the cell time to attempt DNA repair; if repair fails, apoptotic cascades (mitochondrial caspase pathways) are triggered
pubmed.ncbi.nlm.nih.gov
. In parallel, bipolar androgen therapy (BAT) involves exposing castration-resistant prostate cancer (CRPC) to cyclical surges of supraphysiological testosterone. Paradoxically, acute high-androgen exposure has a cytotoxic effect on prostate cancer cells despite the hormone’s role in driving growth at normal levels
jci.org
. Mechanistically, high-dose androgen stimulation heavily engages the androgen receptor (AR) and its coregulators, which can induce DNA damage and replication stress. AR activation is known to recruit DNA topoisomerase IIβ (TOP2B) to androgen response elements on chromatin, causing transient DSBs that relieve transcriptional torsional stress
pmc.ncbi.nlm.nih.gov
. Under BAT’s extreme testosterone levels, this AR-dependent DNA damage is magnified beyond the cell’s repair capacity
jci.org
mdpi.com
. Chatterjee et al. (2019) showed that supraphysiological androgen exposure leads to a dose-dependent increase in DSBs, accompanied by AR-driven transcriptional repression of DNA repair genes and accumulation of cells in G₀/G₁ arrest or senescence
jci.org
. Thus, both doxazosin and BAT converge on inducing DNA double-strand breaks and enforcing cell-cycle checkpoints. Doxazosin’s ability to down-regulate key DNA repair factors (it was shown to suppress expression of PRKDC (DNA-PKcs) and XRCC5 (Ku80), crucial for DSB repair)
pubmed.ncbi.nlm.nih.gov
means that a tumor cell exposed to doxazosin is less capable of fixing DNA breaks. BAT creates a burst of AR-mediated DNA lesions at peak testosterone, so administering doxazosin during these peaks could, in theory, synergize by compounding DNA damage while simultaneously blunting the cell’s repair and mitotic recovery mechanisms. In essence, high testosterone “loads the gun” by introducing DNA breaks, and doxazosin “removes the safety net” by abrogating repair and pushing the cell past its ability to survive the damage. Additionally, there may be phase-specific complementarity: BAT’s cytotoxic effect is associated with a blockade in G₁ and entry into a quiescent/senescent state
jci.org
, whereas doxazosin traps cells that do enter S/G₂ in a checkpoint arrest
mdpi.com
. This dual assault on the cell cycle could ensure that whether a cell attempts to cycle or not, it cannot easily escape damage. Taken together, the theoretical synergy is that doxazosin amplifies and locks in the DNA-damaging stress imposed by BAT’s testosterone surges, leading to enhanced tumor cell death during peak testosterone periods.
Preclinical Evidence Supporting or Contradicting Synergy
Doxazosin and DNA Damage: Multiple preclinical studies have documented the DNA-damaging and anti-tumor effects of doxazosin (and related α₁-blockers) in prostate cancer models. In vitro, doxazosin induces apoptosis in both androgen-dependent (LNCaP) and androgen-independent (PC-3, DU-145) prostate cancer cell lines via an AR-independent mechanism involving oxidative DNA damage and chromatin binding
pubmed.ncbi.nlm.nih.gov
pubmed.ncbi.nlm.nih.gov
. Gene expression profiling of LNCaP cells treated with doxazosin found upregulation of stress response genes like GADD45A and suppression of DNA repair genes (mentioned PRKDC and XRCC5), consistent with the accumulation of DNA damage and impaired DSB repair
pubmed.ncbi.nlm.nih.gov
. Lin et al. (2007) directly showed that prazosin (a close analogue of doxazosin) causes an increase in DNA strand breakage in prostate cancer cells, evidenced by comet assays and γH2AX foci, which in turn activates the ATM/ATR-mediated checkpoint signaling
pubmed.ncbi.nlm.nih.gov
. Treated cells exhibited Cdc25c sequestration and Cdk1 (CDK1) inhibition, confirming a G₂ checkpoint arrest prior to apoptosis
pubmed.ncbi.nlm.nih.gov
. Importantly, this mechanism was observed even in p53-null, AR-null PC-3 cells, indicating that α₁-antagonist-induced DNA damage is sufficient to engage the checkpoint and kill cells independent of p53 or AR status
pubmed.ncbi.nlm.nih.gov
. In vivo evidence echoes these findings: prazosin given orally to mice bearing PC-3 xenograft tumors significantly reduced tumor mass over several weeks
pubmed.ncbi.nlm.nih.gov
. Tumors from prazosin-treated mice showed increased apoptosis and reduced proliferation, consistent with the drug’s DNA damage stress effect leading to cell death in vivo
pubmed.ncbi.nlm.nih.gov
. While doxazosin itself has been less reported in animal tumor models, it belongs to the same class and was noted to cause DNA fragmentation in treated cancer cells
mdpi.com
. Doxazosin has even been proposed to act as a topoisomerase I inhibitor, which would create DNA strand breaks and was shown to produce synergistic cytotoxicity when combined with established DSB-inducing chemotherapeutics like etoposide (topo II poison) or adriamycin (doxorubicin)
mdpi.com
. This last point is especially relevant, as it demonstrates doxazosin’s ability to augment DNA damage from other sources – an analogy for its potential to synergize with BAT-induced damage.
BAT and DNA Damage: The concept of BAT’s antitumor activity has its roots in decades-old observations of “androgen overdose” causing tumor regression after a period of deprivation
jci.org
. Preclinical models have consistently shown a biphasic growth response to androgens: prostate cancer cell proliferation increases as androgen rises from low to normal ranges, but beyond a threshold of supraphysiologic androgen concentration, growth is stunted or reversed
jci.org
. For instance, in androgen-sensitive cell lines adapted to low androgen, re-addition of high-dose dihydrotestosterone triggers widespread DNA damage within hours. One mechanistic explanation came from Haffner et al. (2010), who discovered that androgen-bound AR can physically collaborate with TOP2B to induce site-specific DSBs in chromosomal DNA
jci.org
. These breaks are normally transient and part of gene regulation, but excessive AR signaling or impaired repair can convert them into lethal lesions. Chatterjee et al. provided direct evidence of this in CRPC models: treating AR-overexpressing prostate cancer cells with supraphysiological androgen led to a surge in DSB markers and a global downregulation of DNA repair genes (including genes in the homologous recombination pathway)
mdpi.com
. Cells exposed to high testosterone showed cell-cycle arrest (accumulating in G₀/G₁) and often entered senescence or apoptosis due to unresolved DNA damage
jci.org
. In the same study, xenograft tumors grown in castrated mice responded to testosterone supplementation with reduced growth and increased DNA damage signaling, confirming the in vivo relevance
jci.org
jci.org
. Notably, tumors or cell lines with defects in DNA repair (e.g. BRCA2 mutations) were especially vulnerable to BAT: high-androgen treatment caused more extensive DSB accumulation in these backgrounds and led to greater cell death
jci.org
. This synergizes with the idea that hampering repair magnifies BAT’s cytotoxicity. In fact, combining supraphysiologic androgen with a PARP1 inhibitor or a DNA-PKcs inhibitor in vitro resulted in significantly greater suppression of prostate cancer cell growth than high androgen alone
jci.org
. This is a strong analog to what doxazosin could do, since doxazosin functionally inhibits DNA-PKcs by down-regulating its gene expression
pubmed.ncbi.nlm.nih.gov
. There is no published preclinical study explicitly testing doxazosin together with high testosterone exposure; however, the separate evidence outlined above strongly supports the concept of synergy. Doxazosin’s pro-apoptotic, DNA damage–inducing action in prostate cancer cells has been demonstrated across multiple models, and BAT’s DSB-inducing, repair-suppressing effect is well documented in cell lines and animal studies. The overlap of these effects suggests that co-treating CRPC cells with doxazosin during BAT could amplify DNA damage beyond what either alone achieves. Moreover, since doxazosin can kill a portion of cells even in AR-negative conditions
pubmed.ncbi.nlm.nih.gov
, it might help eradicate subpopulations that BAT cannot (e.g. cells with low AR activity), addressing a potential resistance niche. There is no strong preclinical evidence contradicting this proposed synergy, but it should be noted that optimal timing and dosing would need to be refined – e.g. ensuring doxazosin concentrations are sufficient during the peak testosterone window to have its effect, and that the sequence allows AR signaling to generate DSBs for doxazosin to exploit.
Potential Drawbacks and Limitations
While the theoretical and preclinical rationale for combining doxazosin with BAT is compelling, several potential drawbacks must be considered. Toxicity to normal tissues is a primary concern when enhancing DNA damage. Doxazosin is a systemic medication that, at higher-than-standard doses, could affect non-cancerous cells. Research has shown, for example, that doxazosin can induce apoptosis in cultured cardiomyocytes via an off-target mechanism (unrelated to α₁ blockade)
pmc.ncbi.nlm.nih.gov
. This hints that normal cells with high sensitivity (like cardiac muscle or endothelial cells) might undergo unintended apoptosis if exposed to high concentrations of doxazosin. In patients, doxazosin’s well-known side effects include hypotension, dizziness, and fatigue due to its vasodilatory action; used in combination with BAT, which can itself cause fluid retention and cardiovascular strain, there is a risk of compounding cardiovascular side effects. Clinical BAT studies to date report mostly mild to moderate toxicities – common effects include hot flashes, musculoskeletal aches, edema, and breast tenderness
pmc.ncbi.nlm.nih.gov
– and these occurred in the setting of ongoing ADT (to suppress endogenous testosterone). Introducing doxazosin could add orthostatic hypotension or syncope to that profile, potentially limiting the doses of doxazosin that can be safely given alongside high-dose testosterone. Careful patient selection (e.g. excluding those with baseline low blood pressure or cardiovascular instability) and monitoring would be necessary.
Another limitation is that the efficacy of BAT (and thus the synergy) is contingent on AR activity. Tumors that have lost dependence on the androgen receptor – through AR gene loss, lineage plasticity to neuroendocrine prostate cancer, or the expression of AR splice variants that lack the ligand-binding domain – may not experience the DNA-damaging effects of high testosterone
mdpi.com
. In such cases, BAT would not strongly induce DSBs, and doxazosin would be acting more like a lone agent. Doxazosin alone can still induce some apoptosis, but the synergy would be blunted. There’s also the converse: highly AR-driven tumors could, in the short term, respond to testosterone by proliferating (a transient “flare” of tumor activity) before the cytotoxic effects kick in. If doxazosin is introduced at that moment, it’s unclear whether it might mitigate a flare or whether any initial burst of proliferation could facilitate more DNA damage. Managing the timing (perhaps giving doxazosin slightly before the testosterone peak to pre-empt DNA repair) might be needed but has not been tested.
Resistance mechanisms are an important consideration for any combination therapy. One concern is that repeated cycles of BAT + doxazosin could select for tumor cell clones that adapt by improving their DNA damage tolerance. For instance, cancer cells might up-regulate alternative DNA repair pathways (such as switching to error-prone but quick repair mechanisms), or enhance checkpoint kinase signaling to better arrest and repair damage, thereby diminishing the combination’s efficacy over time. It’s also conceivable that tumor cells surviving multiple BAT cycles undergo adaptive changes – e.g. mutations in AR that reduce its activity or changes in cell-cycle control – making them less susceptible to the therapy. Clinical BAT trials have shown that while many patients respond with tumor regression, the disease eventually progresses again (all patients in one pilot progressed by ~1 year, despite some deep initial responses)
pubmed.ncbi.nlm.nih.gov
. This indicates that even with intense androgen cycling, resistant populations emerge. Doxazosin might delay this by killing more cells per cycle, but it may not prevent it entirely.
Another potential drawback is the risk of excessive DNA damage and the consequences of how cells respond. If the combination induces a very high burden of DSBs, cells may die via apoptosis – which is desired – but there is also the possibility of some cells undergoing catastrophic mitoses or necrosis. Extensive DNA damage in tumors could provoke inflammation in the tumor microenvironment; in some contexts this could be beneficial (immune activation), but it could also lead to edema or pain flares in sites of disease (for example, inflammatory reactions in bone metastases). Additionally, cells that do not die might enter a state of senescence. Senescent cells stop proliferating (which is good) but can secrete pro-tumorigenic cytokines (the SASP, or senescence-associated secretory phenotype) that could alter the tumor milieu negatively. High-androgen treatment is known to cause senescence in a fraction of prostate tumor cells
jci.org
, and combining it with a DNA-damaging agent might increase the senescent fraction if apoptosis is not complete. There is also a theoretical concern that inducing lots of DNA breaks in cancer cells could lead to genomic instability in any cells that manage to recover. Mis-repaired DSBs can result in chromosomal aberrations – for example, androgen-induced TOP2B breaks have been implicated in driving the TMPRSS2-ERG gene fusion in prostate cancer when repairs go awry
jci.org
. While the goal of therapy is to kill cells, any that survive with such aberrations might behave more aggressively. Thus, overshooting with DNA damage could, in rare scenarios, promote the evolution of treatment-resistant, more malignant clones. Similarly, genotoxic stress in normal stem cell compartments raises a long-term worry about secondary malignancies (as is seen with some chemotherapies years later). To summarize, the potential drawbacks of a doxazosin + BAT strategy include increased normal tissue toxicity (especially cardiovascular), dependence on an AR-active tumor phenotype, the emergence of resistance through cellular adaptation, and the possibility of unwanted biological outcomes like senescence or genomic instability if the induced DNA damage is not lethal to all tumor cells. These concerns underscore the need for cautious, stepwise investigation – perhaps starting with preclinical models and low-dose combinations – before this strategy could be applied clinically.
Comparison with Other DNA-Damaging Agents in BAT Context
The concept of pairing BAT with a DNA-damaging agent has already been explored with more conventional therapies, yielding insights that are relevant to combining BAT with doxazosin. In the initial pilot study that established BAT’s clinical feasibility, BAT was combined with etoposide, a topoisomerase II poison
pubmed.ncbi.nlm.nih.gov
. The scientific rationale came from preclinical observations that AR-mediated transcriptional events create TOP2B-linked DNA breaks, and that etoposide can “trap” TOP2B on DNA, converting transient breaks into permanent DSBs
pmc.ncbi.nlm.nih.gov
. In vitro experiments confirmed that high-dose androgen plus etoposide leads to more DSBs (and cancer cell death) than either alone
pmc.ncbi.nlm.nih.gov
. Clinically, the BAT+etoposide trial (Schweizer et al., 2015) reported that 7 of 14 evaluable CRPC patients achieved a ≥50% decline in PSA (some with near complete responses), and about half had radiographic tumor regressions
pubmed.ncbi.nlm.nih.gov
. This was notable given that etoposide alone has modest activity in CRPC. The success of that combination suggests that BAT effectively sensitized tumors to a DNA-damaging chemotherapeutic, supporting the synergy hypothesis. It is an analog for doxazosin, which, like etoposide, will add to the DNA damage burden during high androgen exposure (albeit via a different mechanism).
Another pertinent comparison is the combination of BAT with PARP inhibitors, which target DNA repair. PARP inhibitors (like olaparib) by themselves cause lethal DNA damage in cells that have homologous recombination deficits. Since BAT can phenocopy some aspects of homologous recombination deficiency (by downregulating repair genes), researchers posited that BAT and PARP inhibition would work well together
pmc.ncbi.nlm.nih.gov
. This was tested in a recent phase II trial of BAT plus olaparib. The results were encouraging: the combination had a 31% PSA₅₀ response rate at 12 weeks, and overall ~44% of patients experienced a significant PSA decline during therapy
pmc.ncbi.nlm.nih.gov
. Median progression-free survival exceeded a year, and intriguingly, responses were observed regardless of whether tumors had BRCA1/2 or other HR repair mutations
pmc.ncbi.nlm.nih.gov
. In other words, even patients with intact DNA repair genes benefited, implying that BAT had functionally compromised the cancer cells’ DNA repair to a degree that made the PARP inhibitor effective. This mirrors how doxazosin might work in tandem with BAT – essentially as a “pharmacologic DNA repair inhibitor” (through DNA-PKcs and Ku70/80 suppression and perhaps topoisomerase inhibition) combined with an androgen-mediated assault on the genome. The BAT+olaparib trial also demonstrated that adding a second agent (olaparib) did not ruin the tolerability of BAT; the combo was feasible with manageable side effects, an important proof-of-concept for multi-agent BAT strategies
pmc.ncbi.nlm.nih.gov
.
Beyond drugs, radiation therapy offers another DNA-damaging approach that could synergize with BAT. Radium-223, a bone-seeking alpha-particle emitter, induces DNA double-strand breaks specifically in bone metastases. It has been hypothesized that BAT could prime tumor cells in bone lesions to be more susceptible to radium-223–induced DNA damage, or conversely that radium-223 could add lethal damage to BAT-treated cells, similar to chemo
researchgate.net
. Anecdotally, case reports of patients with aggressive, DNA-repair-deficient prostate cancers have shown profound responses when BAT was combined with subsequent immunotherapy or when radium-223 was used after BAT
researchgate.net
. The rationale is that extensive DNA damage (from BAT or radium) can increase tumor mutational burden or cell death, potentially enhancing immune recognition – a concept under investigation. Traditional external beam radiation might also synergize with BAT, since AR signaling modulates radio-sensitivity (AR can promote DNA repair, so high androgen might, somewhat counterintuitively, hinder repair as discussed). However, no clinical trial results are yet available for BAT with radiation.
It’s also informative to consider other DNA-damaging drugs in the context of BAT. For example, platinum-based chemotherapies (like carboplatin or cisplatin) create interstrand DNA crosslinks that result in DSBs during replication. These have not been formally combined with BAT in trials, but given that platinum efficacy in prostate cancer is typically seen in tumors with DNA repair deficiencies (e.g. those with BRCA2 or ATM mutations), one could speculate BAT might broaden their utility by inducing a transient repair-deficient state in AR-positive tumors. Similarly, anthracyclines (like doxorubicin) or mitoxantrone (an older chemotherapy for prostate cancer) could see heightened activity with BAT – in fact, doxorubicin’s mechanism (TOP2 poison and DNA intercalation) is not far removed from the mechanistic effects attributed to doxazosin at high doses
mdpi.com
. The key theme across these examples is that BAT turns on a vulnerability in prostate cancer cells (DNA damage with reduced repair) that various DNA-damaging agents can exploit. Doxazosin falls into this category as a non-classical DNA-damaging agent. Although it’s officially an antihypertensive drug, at the molecular level doxazosin’s effects in prostate cancer (DNA breaks, G₂/M arrest, apoptosis) resemble those of certain chemotherapeutics
pubmed.ncbi.nlm.nih.gov
mdpi.com
. Therefore, its use during BAT can be conceptually likened to using a low-grade “chemotherapy” alongside the hormonal cycling. The experiences with etoposide and olaparib suggest that if doxazosin can be delivered at a sufficient concentration without intolerable toxicity, it has a good chance of enhancing BAT’s efficacy, much as those other agents did. One caveat from those comparisons: the timing of combination is crucial (etoposide was given during the first two weeks of each BAT cycle when testosterone was highest
pubmed.ncbi.nlm.nih.gov
pubmed.ncbi.nlm.nih.gov
, and presumably a similar timing would be ideal for doxazosin to coincide with peak testosterone-induced DNA stress).
Conclusion
In summary, the synergy between doxazosin and bipolar androgen therapy in prostate cancer is grounded in a convergence of mechanisms: both induce DNA double-strand breaks and enforce cell-cycle arrests that lead to cancer cell death. Doxazosin’s ability to damage DNA and disable repair pathways aligns remarkably well with BAT’s strategy of overwhelming prostate cancer cells with an androgen surge to create lethal DNA and replication stress. Preclinical studies have shown that doxazosin/prazosin can drive prostate cancer cells into G₂ arrest and apoptosis via DNA damage, and likewise that high-dose testosterone can trigger DNA breaks, senescence, and cell death in AR-driven prostate tumors. When used together (particularly timed so that doxazosin is present during BAT’s peak testosterone phase), these treatments could reinforce each other – doxazosin exacerbating and prolonging the DNA damage caused by BAT, and BAT providing a stream of targets (AR-initiated DNA lesions and an impaired repair environment) for doxazosin to act upon. This combination, if successful, might translate into deeper tumor responses in castration-resistant prostate cancer, potentially overcoming some forms of resistance to endocrine therapy.
However, this theoretical synergy comes with important considerations. Safety and specificity will be the biggest challenges – maximizing tumor DNA damage while minimizing harm to normal tissues. Unlike targeted PARP inhibitors or carefully delivered radiation, doxazosin will expose the entire body to its effects, so future studies must establish a dosing schedule that cancer cells can’t easily escape but normal cells can withstand. Additionally, patient selection will matter: tumors must have functioning AR signaling for BAT + doxazosin to work as intended. The combination might find its niche in AR-positive, heavily DNA-repair-deficient cancers (where it could be most effective), or it might be used earlier in CRPC to prevent or delay resistance to next-generation AR inhibitors (taking advantage of its capacity to resensitize tumors to hormone therapy as seen with BAT alone
pubmed.ncbi.nlm.nih.gov
). Further preclinical research is warranted to test BAT and doxazosin side by side in cell culture and animal models – this would clarify the extent of added benefit and map out any unexpected interactions between high androgen levels and quinazoline drugs. If those studies demonstrate a favorable therapeutic window, cautious clinical trials could be designed (perhaps starting with low-dose doxazosin added to BAT in a small cohort). The lessons learned from combinations like BAT+etoposide and BAT+olaparib provide a blueprint, but doxazosin’s unique profile (as an inexpensive, repurposable oral drug) makes it an attractive and novel addition if it can be made safe. In conclusion, doxazosin and BAT represent a rational pairing against prostate cancer, attacking the tumor’s DNA integrity from two angles. The strategy embodies the principle of synthetic lethality – AR over-stimulation creates DNA damage and a repair deficit, which doxazosin further exploits – but its ultimate success will depend on careful balancing of efficacy and toxicity. Continued investigation will determine whether this theoretical synergy can be translated into a practical therapy for patients with advanced prostate cancer.
References
1. Arencibia J.M. et al. (2005). Doxazosin induces apoptosis in LNCaP prostate cancer cells through DNA binding and DNA-PK down-regulation. Int. J. Oncol. 27(6): 1617–1623. PMID: 16273218 pubmed.ncbi.nlm.nih.gov
2. Lin S.C. et al. (2007). Prazosin displays anticancer activity against human prostate cancers: targeting DNA and cell cycle. Neoplasia. 9(10): 830–839. PMID: 17971903; PMCID: PMC2040210 pubmed.ncbi.nlm.nih.gov
3. Chatterjee P. et al. (2019). Supraphysiological androgens suppress prostate cancer growth through AR-mediated DNA damage. J. Clin. Invest. 129(10): 4245–4260. PMID: 31361603; PMCID: PMC6771231 jci.org
4. Schweizer M.T. et al. (2015). Effect of bipolar androgen therapy in asymptomatic CRPC: a pilot clinical study. Sci. Transl. Med. 7(269): 269ra2. PMID: 25568070 pubmed.ncbi.nlm.nih.gov
5. Simmonds J. et al. (2016). The role of α1 adrenoceptor antagonists in treatment of prostate and other cancers. Int. J. Mol. Sci. 17(8): 1339. DOI: 10.3390/ijms17081339 mdpi.com
6. Haffner M.C. et al. (2010). Androgen-induced TOP2B-mediated double-strand breaks and prostate cancer gene rearrangements. Nat. Genet. 42(8): 668–675. DOI: 10.1038/ng.613 jci.org
7. Yu E.Y. et al. (2023). Bipolar androgen therapy plus olaparib in men with metastatic castration-resistant prostate cancer. Prostate Cancer Prostatic Dis. 26(3): 475–482. PMID: 36564459; PMCID: PMC10286318 pmc.ncbi.nlm.nih.gov
8. Gongora A.B.L. et al. (2022). Extreme responses to a combination of DNA-damaging therapy (BAT or radium-223) and immunotherapy in CDK12-altered mCRPC. Clin. Genitourin. Cancer. 20(4): 330–334. DOI: 10.1016/j.clgc.2021.11.015 researchgate.net