In this article, the pathologic effects of hormonal therapy on benign prostatic tissue and prostate cancer are presented. Hormonal therapy is the main treatment for men with disseminated prostate cancer.  In addition, hormonal therapy, also known as endocrine- or androgen-deprivation therapy, is commonly used in locally advanced or high-grade, high-risk disease. (See the image below.)
Compared with conservative management, primary androgen-deprivation therapy is not associated with improved survival in the majority of men aged 66 years or older with clinically localized prostate cancer.  Neoadjuvant hormonal therapy before radical prostatectomy has not been shown to be effective.
Molecular genetic abnormalities are not currently used in the diagnosis of prostatic carcinoma with treatment effect. Deoxyribonucleic acid (DNA) ploidy and chromosomal abnormalities are similar in treated versus untreated carcinoma, [3, 4] although gene expression profiling shows that several hundred genes are differentially expressed in the comparison of untreated versus hormonally treated and androgen-independent carcinomas. 
Grading of hormonal treatment effects (regression grading) to predict clinical outcome has been attempted but has yielded variable results. This is not routinely performed, and special studies on prostatic carcinoma with treatment effect are not currently done.
For more information on prostate cancer, see the following:
Endocrine therapy for prostate cancer is viewed as palliative and not curative, although most patients do experience an objective response. Eventually, all patients with metastatic disease treated with hormonal therapy develop hormone-independent (hormone-refractory or hormone-resistant) prostate cancer, with a rising prostate-specific antigen (PSA) level or other evidence of progression in the presence of androgen blockade.
Methods of endocrine treatment include surgical castration—that is, orchiectomy (because testicular Leydig cells are the major source of androgens in males)—and medical castration. 
Medical castration can be achieved with synthetic estrogens (such as diethylstilbestrol), luteinizing hormone-releasing hormone (LHRH) analogues, inhibition of androgen synthesis (by aminoglutethimide and ketoconazole), and antiandrogens (of steroidal and nonsteroidal types).  These therapies can be used as monotherapy or together in combined androgen blockage, in which there is an attempt to block production of testicular androgens and the effect of adrenal androgens. However, the survival benefit of combined androgen blockade compared with standard androgen-deprivation therapy is minimal. 
Another type of androgen deprivation is via inhibition of 5-alpha-reductase by finasteride or dutasteride. These inhibitors have been used mainly to treat benign prostatic hypertrophy (BPH), but there is also interest in them for chemoprevention of prostate cancer. 
Androgen-deprivation therapy effects may be seen in prostatic tissue from needle biopsy, transurethral resection of the prostate (TURP), and radical prostatectomy. These effects can also be found in tissue samples of metastatic prostatic carcinoma.
Adverse effects of androgen-deprivation therapy
Possible side effects from androgen ablation therapy include hot flashes and decreased libido or potency (in most patients) with orchiectomy and LHRH agonists, and gynecomastia, nausea, diarrhea, hepatotoxicity, fluid retention, and thromboembolism with diethylstilbestrol and some of the antiandrogens. Longer-term adverse effects include decreased muscle mass, cognitive function decline, osteoporosis, and anemia. [7, 8, 9]
When evaluating the effects of androgen-deprivation therapy on prostatic tissue, also consider the possibility of benign conditions that may mimic carcinoma following such therapy, such as benign atrophy and histiocytic infiltration.
Gross and Microscopic Findings
The gross findings and microscopic effects of hormonal therapy on prostatic tissue and prostate cancer are briefly reviewed in this section.
Gross findings of hormonal effects on BPH and prostate cancer
Grossly, prostates from patients treated with luteinizing hormone-releasing hormone (LHRH) analogues or antiandrogens are small, and benign prostatic hypertrophy (BPH) and carcinoma can be more difficult to identify macroscopically, compared with untreated glands. Treated glands are shrunken and have a rubbery to soft consistency. [3, 10]
Microscopic findings of hormonal therapy effects on benign prostatic tissue
Microscopically, there is stromal dominance, with fewer glands than normal. The remaining glands are atrophic. Hyperplastic glands in BPH are simplified, with loss of branches and intraluminal undulations, such that they appear small, ovoid, round, or comma-shaped. The secretory cell lining layer shows loss of cytoplasmic volume (atrophy), nuclear pyknosis, and cytoplasmic clearing. The basal cell layer is prominent and can be hyperplastic. Glycogenated squamous metaplasia may be found in prostates treated with estrogens or diethylstilbestrol, whereas immature squamous metaplasia may be noted with other forms of androgen-deprivation therapy. Stromal changes are less conspicuous than epithelial changes, with scattered lymphocytes being detected. Similar changes have been observed with finasteride and dutasteride treatment.  (See the images below.)
Microscopic findings of hormonal therapy effects on prostatic intraepithelial neoplasia
Prostatic intraepithelial neoplasia (PIN) is reduced in prevalence and extent after androgen-deprivation therapy.  The treated PIN glands show reduced luminal complexity, with flattening of the luminal epithelium. The luminal epithelial cells display nuclear pyknosis, inconspicuous nucleoli, and cytoplasmic shedding, similar to adenocarcinoma cells. Basal cells are often prominent.  Apoptotic bodies are readily seen. 
Due to the luminal cell alterations, grading of PIN after hormonal therapy is often not feasible. High-grade PIN architecture and luminal cell cytology are altered by hormonal therapy, such that high-grade PIN may be difficult to distinguish from low-grade PIN. The effect of 5-alpha reductase inhibition on PIN is not clear. 
Microscopic findings of hormonal therapy effects on adenocarcinoma
There is a decrease in the tumor volume and number of malignant glands in adenocarcinoma. In a small percentage of patients treated with neoadjuvant hormonal therapy before radical prostatectomy, no residual tumor is identified. A decrease in size of the adenocarcinoma glands is characteristic, with small and atrophic glands being detected. Loss or collapse of luminal spaces is typical.
A particularly insidious pattern of growth that can be difficult to recognize as malignant is the presence of widely scattered single tumor cells embedded in a wide expanse of stroma. A branching pattern and mucinous spaces and clefts (so-called pseudomyxoma ovarii-like pattern) may be seen. Cytologically, the carcinoma cells classically show nuclear pyknosis, hyperchromasia, inconspicuous nucleoli, cytoplasmic clearing, and cytoplasmic vacuolization. These changes can impart a histiocytoid appearance to the tumor cells. The stroma may be inflamed and edematous, fibrotic, hyalinized, and sclerotic. There is no change in high–molecular-weight cytokeratin expression in basal cells, so this marker still has diagnostic value in difficult-to-diagnose cases following hormonal therapy.
Hormonal therapy has been associated, in rare cases, with the emergence (or selection and progression) of prostatic carcinoma variants, including small cell carcinoma, sarcomatoid carcinoma, and squamous cell carcinoma. Images of hormonal therapy effects on prostate cancer are provided below.
Microscopic findings of hormonal therapy effects on adenocarcinoma Gleason grade
It has been established that Gleason grading should not be performed after androgen-deprivation therapy. This is because androgen-deprivation therapy effects induce histomorphologic changes that simulate higher-grade disease, such as luminal space loss and single-cell infiltrates (see the image below). It should be noted, however, that finasteride has little, if any, effect on the Gleason grade of adenocarcinoma. [14, 15]
Tumor spread and staging
Diagnostic recognition of carcinoma cells with hormonal treatment effects that are outside the prostate and in metastatic sites may be difficult (see the image below). Immunohistochemical stains can be of utility in these settings, and they can facilitate a more accurate stage assignment.
Immunohistochemical stains for basal cells and alpha-methylacyl-coenzyme A (CoA) racemase (AMACR) can be helpful in the diagnosis of prostatic carcinoma after hormonal treatment. [16, 17, 18] There is a reduction in AMACR immunoreactivity following hormonal therapy, such that 45% to 71% of cases are AMACR positive. [16, 18] In these post–hormonal therapy, AMACR-negative cases, basal cell stains assume great importance.
In radical prostatectomy cases with neoadjuvant hormonal treatment, detection of single-cell carcinoma infiltrates (see the image below) and distinction of carcinoma cells with treatment effect from inflammatory cells such as histiocytes and lymphocytes may be facilitated by the use of pan-cytokeratin, prostate-specific antigen (PSA), prostate-specific acid phosphatase (PSAP), and prostate-specific membrane antigen (PSMA) immunostains. [11, 19] PSA and PSAP expression may be reduced after hormonal therapy, whereas cytokeratin expression is unaltered.
Unlike breast carcinoma in which immunostains for estrogen and progesterone receptors are useful in predicting response to hormonal therapy, androgen-receptor immunohistochemistry has no value in predicting response to hormonal therapy for patients with untreated prostatic adenocarcinoma.
Wissing et al conducted a study to assess the correlation between nuclear Eg5 expression, docetaxel response, and disease aggressiveness in prostate cancer.  Immunohistochemical staining for nuclear Eg5 was performed on 117 archival specimens from 110 patients with prostate cancer who were treated with docetaxel between 2004 and 2012. Analysis of samples taken before hormonal therapy showed that overall survival and time to docetaxel use were significantly decreased in patients with nuclear Eg5-expressing tumors. Eg5-positive nuclei were found more frequently in T4-staged tumors, Gleason 8-10 tumors, and metastasized tumors. Multivariate analyses indicated that nuclear Eg5 expression may be an independent parameter for tumor aggressiveness. The investigators concluded that nuclear Eg5 expression may be a predictive biomarker for docetaxel response in metastatic castration-resistant prostate cancer and a prognostic biomarker for hormone-naive patients with prostate cancer.