This investigator-initiated clinical trial (IIT) builds on and extends research on autophagy
      in prostate cancer with PAR-4 and HCQ conducted by a transdisciplinary investigative team,
      and is sponsored by funding from Markey Cancer Center.

      Prostate cancer is the most common cancer in men and the 2nd most common cause of cancer
      death in men. Secondary to the greying of the U.S. population and increasing life expectancy,
      metastatic prostate cancer (mPCa) rates are increasing. Androgen deprivation therapy (ADT)
      has been the primary treatment for metastatic prostate cancer. The survival benefit of ADT is
      juxtaposed against significant adverse effects including cardiovascular morbidity, skeletal
      fractures, diabetes, sexual dysfunction, and a decrease in cognitive function. Recent studies
      have indicated a potential benefit to treatment of metastatic lesions in men with limited
      metastatic disease and have identified a potential delay in initiation of ADT in men whose
      limited metastatic lesions were treated with stereotactic radiotherapy. Prostate apoptosis
      response-4 (PAR-4) is a tumor suppressor protein that facilitates apoptosis in numerous types
      of cancer cells. Hydroxychloroquine (HCQ) has been found to induce PAR-4 expression and
      subsequently promote apoptosis of cancer cells and inhibit metastatic progression. HCQ has
      also been reported to both inhibit and enhance immune responses via changes in Th1, Th2, Th17
      and Treg subsets and reduce inflammatory markers. It also acidifies lysosomes which
      potentially inhibits antigen presentation by antigen-presenting cells, and can be measured
      with LysoSensor Yellow/Blue DND-160 by flow cytometry and induces autophagy which can be
      detected by measuring up-regulation of the microtubule associated protein LC3B. Treatment of
      oligometastatic prostate cancer may be enhanced by the addition of Hydroxychloroquine to
      either surgical resection or radiation treatment of metastatic lesions. Potential benefits of
      Hydroxychloroquine include delayed disease progression and delayed initiation of ADT,
      lessening morbidity and increasing quality of life in men with oligometastatic prostate
      cancer.

      A single-arm, open-label phase II trial will be conducted in a population of men with
      recurrent oligometastatic prostate cancer following primary treatment of localized disease.
      The oligometastatic sites will be treated with either surgical resection or stereotactic
      radiation per standard of care, in addition to 400 mg of Hydroxychloroquine per day for a
      period of 3 months.

      Sample Size:

      Based on an ongoing Phase I trial of HCQ, the proportion of patients who will exhibit a 50%
      induction in PAR-4 from baseline levels is equal to 0.50, compared to a null hypothesis of
      0.20. A sample of 18 patients will provide 84% power to detect this hypothesized difference
      in proportion based on a two-sided test with 5% significance level. Prior studies at Markey
      Cancer Center in this population have demonstrated an attrition rate of less than a 10%.
      Thus, a total of 20 patients will be enrolled into the study.

      Specific and Secondary Aims:

      The primary objective is to assess the rate of attainment of a 50% increase in tumor
      suppressor PAR-4 levels from baseline in patients treated with 3-months of hydroxychloroquine
      (HCQ), in combination with radiation or surgery for recurrent, oligometastatic prostate
      cancer.

      Secondary aims of the trial include assessment of: median progression-free survival; 1- and
      3-year ADT-free survival; treatment toxicity and quality of life.

      Correlative studies of the trial comprise assessment of immunological effect of
      Hydroxychloroquine by analyzing peripheral blood mononuclear cells (PBMC's).

      Statistical Analytic Plan:

      Descriptive statistics will be calculated to summarize PAR-4 levels at each time point of
      follow-up. Percent change from baseline compared to each follow-up time point will likewise
      be calculated. The proportion of patients who exhibit a 50% induction in PAR-4 compared to
      baseline within the 3-month therapy period will be calculated and a one-sample test for
      proportion will be performed. Secondary analyses of the primary endpoint will also be
      performed. Continuous changes in PAR-4 levels will also be assessed using paired t-test or
      nonparametric analog. Linear mixed models will be employed to analyzed repeatedly measured
      levels of PAR-4 and association with clinical parameters as well as PSA levels.

        -  Other secondary analyses will include summary of the PAR-4 induction profile over the
           3-month treatment period and after end of therapy period. PSA levels will be assessed
           and PSA doubling time will be calculated. Association of PSA levels with PAR-4 levels
           will be determined using Spearman or Pearson's correlation coefficient. Quality of life
           as measured by the EORTC QLQ-C30 supplemented with QLQ-PR25 will be evaluated and
           descriptive statistics of QOL scores will be calculated. Association of QOL with PAR-4,
           PSA and other clinical endpoints will be determined in an exploratory manner using
           correlations and survival analysis models.

        -  All participants who received HCQ will be included in the safety analysis. Frequency and
           incidence tables of toxicity and AEs will be generated.

        -  Immune outcome endpoints including markers of immune function/activation and systemic
           inflammation will be summarized at each time point of follow-up. Linear mixed models
           will be utilized to model these repeatedly measured immune endpoints to determine
           changes over time. Associations between markers and PAR-4 levels will be assessed via
           calculation and comparison of correlation coefficients. Data processing, data analysis
           pipelines and bioinformatics methods for NGS data will be employed in collaboration with
           MCC Shared Resource Facilities. False discovery rate (FDR) q values will be calculated
           for multiple comparison adjustment, with threshold significance set to 0.05.