Permanent closure of trial to further accrual based on IDE disapproval

This is a phase II multi-center study to determine the clinical impact of using a patient-specific genomic expression signature of androgen receptor (AR) activity to determine therapy for patients with castration-resistant metastatic prostate cancer (CRPC). After patient eligibility is determined, the genomic signature will be applied to fresh frozen tissue harvested from a metastatic lesion during image-guided biopsy. After assessing for androgen receptor activity, the investigators will select patients for either continued androgen manipulation with nilutamide (high AR activity) or targeted therapy with dasatinib (low AR activity). Once patients develop a first progression on either arm, patients will receive combination therapy with dasatinib and nilutamide. The primary aim is to estimate the median progression free survival in men with CRPC treated according to tumor AR activity. The investigators hypothesize that by treating men based upon AR activity, median progression free survival (PFS) will improve from a historical median of 3.0 months to 6.0 months.

Prostate cancer is the most commonly diagnosed non-cutaneous malignancy of men in the United States and remains the second leading cause of cancer-related mortality among men. Novel approaches are necessary to personalize and improve treatment. The androgen receptor (AR) plays a critical role in the normal development and function of the prostate and promotes growth in most prostate cancers. Most patients with advanced prostate cancer have cancer that will initially respond to androgen-targeting therapy (focusing on decreasing the circulating levels of testosterone which is the primary source of ligand for the AR receptor). However, castrate resistance usually develops within 18 to 24 months and the median survival of men with castration resistant prostate cancer (CRPC) ranges between 12 to 18 months. Current options are limited including: secondary hormonal manipulations, radiopharmaceuticals, and chemotherapy and only docetaxel, a taxane that targets microtubules, has been proven to prolong survival.

Importantly, it has become clear that for some patients with CRPC, the prostate tumors remain dependent on AR activity. This has been supported by studies demonstrating different genetic and metabolic mechanisms by which prostate cancer cells maintain AR activity despite low levels of circulating androgen. An assay detecting AR activity that more comprehensively reflects the variety of mechanisms by which AR activity is preserved has the potential to accurately differentiate between men who have tumors still dependent on AR activity from those that are truly independent of AR activity. The identification of patients with continued AR activity has the potential to improve response to secondary hormonal manipulations; men with tumors having low levels of AR activity are likely to require alternative approaches.

We have developed a transcriptional "signature" for AR activity with the goal of identifying the true status of AR in tumors of men with CRPC. After validating the AR signature on in vitro and human prostate samples to ensure that it accurately and reproducibly detects AR activity, we applied the AR signature to several independent datasets to determine the distribution of CRPC tumors with preserved AR activity. Interestingly, there is consistent heterogeneity with respect to predicted AR activity. While overall AR activity decreases in CRPC, there is a subgroup of patients with persistently elevated AR activity. In tumors with low AR activity, we observed that the probability of AR activity was negatively correlated with predicted SRC activity in localized prostate cancer and CRPC.

Patients with persistently high AR activity will be treated with nilutamide, an approved oral agent used in metastatic CRPC to target the AR. Patients with tumors having low AR activity will be treated with dasatinib (an oral drug known to target the SRC family kinases). As there is compelling pre-clinical evidence of interactions between the SRC pathway and AR signaling, patients failing either single agent treatment will be treated with the combination of nilutamide and dasatinib and followed again for progression.

Inclusion Criteria:

1. Confirmed diagnosis of adenocarcinoma of the prostate.

2. Radiographic evidence of metastatic disease amenable to image-guided biopsy.

3. Testosterone <50ng/dL on androgen deprivation therapy (ADT). ADT must continue while on study.

4. The patient must have discontinued antiandrogens 30 days prior to baseline PSA unless the patient did not respond to anti-androgen therapy or experienced a decline in PSA lasting < 3 months after starting antiandrogen therapy.

5. Evidence of disease progression on ADT.

6. Patients must have adequate organ and marrow function as defined below:

   • Hemoglobin >9.0g/dL (without transfusion of PRBC)
   • ANC/AGC >1,500/μl
   • Platelets >75,000/μl
   • Total bilirubin < 2.0 times the institutional ULN
   • Creatinine <1.5 times the institutional ULN
   • PT or INR and aPTT < 1.5 times the institutional ULN
   • AST and ALT <2.5 x ULN

7. Age > 18 years

8. Ability to take oral medications (pills must be swallowed whole)

9. ECOG performance status 0-2

10. Concomitant Medications:

    • Patient agrees to discontinue and not to initiate taking St. Johns Wort while receiving dasatinib therapy (discontinue St. Johns Wort at least 5 days before starting dasatinib)
    • Patient agrees not to initiate IV bisphosphonates while on dasatinib. Patients on IV bisphosphonates for > 4 weeks prior to dasatinib will continue on therapy

11. Men of reproductive potential who have not had a radical prostatectomy must agree to use an effective contraceptive method. Patients who have had a prostatectomy are sterile and do not need to use contraception

12. Ability to understand and the willingness to sign a written informed consent document

Exclusion Criteria:

1. Patients who have received prior treatment with nilutamide or dasatinib

2. Patients who have not recovered to Grade 1 or Grade 0 from the toxic effects of prior investigational therapy, biologic therapy, hormonal therapy (other than ADT), immunotherapy, or chemotherapy

3. Medical contraindications to stopping aspirin or coumadin for 1 week prior to image-guided tumor biopsy AND while on dasatinib treatment.

4. History of the following cardiac related conditions:

   • Uncontrolled angina, congestive heart failure or MI within (6 months)
   • Diagnosed congenital long QT syndrome
   • Any history of clinically significant ventricular arrhythmias (such as ventricular tachycardia, ventricular fibrillation, or Torsades de pointes)
   • Prolonged QTc interval on pre-entry electrocardiogram (> 450 msec)
   • Subjects with hypokalemia or hypomagnesemia if it cannot be corrected prior to dasatinib administration

5. History of significant bleeding disorder unrelated to cancer.

6. Concomitant use of Category I drugs that are generally accepted to have a risk of causing Torsades de Pointes including: (These medications can be stopped while the patient is on the protocol and the patient needs to be off the drugs for at least 7 days prior to starting dasatinib)

7. Patients who have a history of amiodarone use.

8. Clinically significant pericardial or pleural effusion or severe respiratory insufficiency

9. Uncontrolled intercurrent illness including, but not limited to, ongoing or active infection (requiring antifungal, antibiotic or antiviral therapy), symptomatic congestive heart failure (NYHC II or greater, unstable angina pectoris, cardiac arrhythmia (uncontrolled SVT or any VT), or psychiatric illness/social situations that would limit compliance with study requirements.

10. Patients with a medical contraindication to image-guided biopsies

64 consented but 7 screen failures, so 57 enrolled. 2 never assigned to rx (withdrew consent, clinically unstable), 33 assigned by AR activity (15 high AR to nilutamide, 18 low AR to dasatinib) and 22 QC technical failures (received dasatinib). 14 of 15 high AR and 17 of 18 low AR began protocol treatment. Analyses include the 31 treated patients.