Circulating Inflammatory Cytokine Expression in Men With Prostate Cancer Undergoing Androgen Deprivation Therapy

Authors

  • Marcello Maggio,

    1. Longitudinal Studies Section, Clinical Research Branch and the Laboratory of Immunology, National Institutes of Health, National Institute on Aging, Baltimore, Maryland
    Search for more papers by this author
  • Amanda Blackford,

    1. the Department of Oncology, Division of Biostatistics; Johns Hopkins University School of Medicine, Baltimore, Maryland
    Search for more papers by this author
  • Dennis Taub,

    1. Longitudinal Studies Section, Clinical Research Branch and the Laboratory of Immunology, National Institutes of Health, National Institute on Aging, Baltimore, Maryland
    Search for more papers by this author
  • Michael Carducci,

    1. Department of Oncology, Prostate Cancer Research Program, Kimmel Cancer Center at Johns Hopkins, Baltimore, Maryland
    Search for more papers by this author
  • Alessandro Ble,

    1. Longitudinal Studies Section, Clinical Research Branch and the Laboratory of Immunology, National Institutes of Health, National Institute on Aging, Baltimore, Maryland
    Search for more papers by this author
  • E. Jeffrey Metter,

    1. Longitudinal Studies Section, Clinical Research Branch and the Laboratory of Immunology, National Institutes of Health, National Institute on Aging, Baltimore, Maryland
    Search for more papers by this author
  • Milena Braga-Basaria,

    1. Department of Medicine, Division of Endocrinology, Johns Hopkins University School of Medicine, Baltimore, Maryland
    Search for more papers by this author
  • Adrian Dobs,

    1. Department of Medicine, Division of Endocrinology, Johns Hopkins University School of Medicine, Baltimore, Maryland
    Search for more papers by this author
  • Shehzad Basaria

    Corresponding author
    1. Longitudinal Studies Section, Clinical Research Branch and the Laboratory of Immunology, National Institutes of Health, National Institute on Aging, Baltimore, Maryland
    2. Department of Medicine, Division of Endocrinology, Johns Hopkins University School of Medicine, Baltimore, Maryland
    Search for more papers by this author

  • Supported by the Johns Hopkins University School of Medicine General Clinical Research Center Grant M01RR00052.

Dr Shehzad Basaria, Department of Medicine, Division of Endocrinology and Metabolism, Johns Hopkins University School of Medicine, Bayview Medical Center, 4940 Eastern Avenue, Suite B-114, Baltimore, MD 21224 (e-mail: sbasari1@jhmi.edu).

Abstract

ABSTRACT: Prostate cancer (PCa) is one of the most common cancers in men. Androgen deprivation therapy (ADT) is employed in the treatment of patients with metastatic or recurrent PCa, resulting in castrate levels of testosterone. Recent studies have shown that male hypogonadism is associated with increased levels of proinflammatory and diminished concentrations of anti-inflammatory cytokines, which normalize upon testosterone treatment. Furthermore, an inflammatory state is associated with osteoporosis, sarcopenia and metabolic abnormalities. We examined 3 groups of men: 1) 20 men with PCa undergoing ADT for at least 12 months prior to the onset of the study (ADT group); 2) 18 age-matched men with non-metastatic PCa who had undergone local surgery and/or radiotherapy and had not yet received ADT and were eugonadal (non-ADT group); and 3) 20 age-matched healthy eugonadal men (control group). None of the subjects were suffering from any acute or chronic inflammatory conditions. Mean age was similar in the 3 groups (P = .41). Men in the ADT and non-ADT groups had higher BMI compared to the control group (P = .0005 and P = .01, respectively). Men in the ADT group had significantly lower mean serum total (P < .0001) and free (P < .0001) testosterone and estradiol (P < .0001) levels compared to the other 2 groups. No significant differences in serum levels of pro-inflammatory or anti-inflammatory cytokines were observed between the 3 groups. These data suggest that men with PCa undergoing long-term ADT do not have elevated levels of pro-inflammatory cytokines compared to age and disease matched controls. Prospective studies are needed to evaluate for any acute changes in these inflammatory markers that might occur after the initiation of ADT.

Prostate cancer (PCa) is one of the most common cancers in men, with an incidence second only to lung cancer (Landis et al, 1999). Radical prostatectomy and/or radiation therapy has been the mainstay of treatment in patients with locally confined PCa. However, for patients with recurrent or metastatic PCa, androgen deprivation therapy (ADT) is frequently used, which results in profound hypogonadism. Orchiectomy and GnRH agonist therapy are the 2 main modes of ADT employed in the treatment of these patients, the latter being the most common. Although the use of ADT has resulted in improved survival in some patients (Messing et al, 1999), its use is associated with a number of complications, including osteoporosis, sarcopenia, and metabolic derangements (Basaria et al, 2002; Smith et al, 2002).

Recently, there has been a growing interest in understanding the interaction between testosterone levels and inflammatory markers. Animal studies have shown that castration results in an inflammatory state (Spinedi et al, 1992). Observational studies have shown that women are more prone to develop autoimmune diseases than men; however, men with hypogonadism have a higher incidence and prevalence of autoimmune diseases compared to eugonadal men (Bizzarro et al, 1987). These data support the notion that testosterone may influence immune cell activation and prevent autoimmune disease development.

Few interventional studies have evaluated the relationship between male hypogonadism and serum levels of pro- and/or anti-inflammatory cytokines. In one study, acute induction of hypogonadism with GnRH agonists in healthy elderly men was followed by elevated serum levels of inflammatory cytokines TNF-alpha and IL-1 beta, suggesting an anti-inflammatory role for testosterone (Khosla et al, 2002). Indeed, testosterone replacement in hypogonadal men results in a significant decrease in pro-inflammatory cytokines TNF-α and IL-1 β and a significant increase in the anti-inflammatory cytokine IL-10 (Malkin et al, 2004). Recent population studies have further confirmed the inverse relationship between testosterone and inflammatory cytokines (Maggio et al, 2006).

Given that men undergoing ADT have profound hypogonadism and that the cytokine profile in patients on long-term ADT is not known, we hypothesized that men undergoing long-term ADT may demonstrate increased serum levels of pro-inflammatory and decreased levels of anti-inflammatory cytokines compared to age- and disease-matched controls. Since osteoporosis, sarcopenia, and metabolic derangements have been associated with an inflammatory state, the findings of this study may have important implications.

Materials and Methods

Study Design and Subjects

The study design was cross-sectional. We studied 3 groups of men: 1) 20 men with PCa who were undergoing ADT (for recurrent or metastatic disease) for at least 12 months prior to the onset of the study (ADT group), 2) 18 age-matched men with non-metastatic PCa who had undergone prostatectomy and/or radiotherapy and recently found to have a rising PSA but had not received ADT and were eugonadal (non-ADT group), and 3) 20 age-matched healthy eugonadal men with a normal PSA (control group). Normal serum testosterone level was defined as >280 ng/dL. The average duration of ADT (ADT group) was 45 months (range 12–101 months). Three patients in the ADT group had undergone orchiectomy, while the remaining 17 men were on GnRH agonists. None of the patients were on androgen receptor antagonists. All the participants in the ADT group were in remission and had a normal PSA (Basaria et al, 2002).

The men undergoing ADT provided an excellent clinical model (since they have profound hypogonadism) to examine the association between hypogonadism and inflammatory cytokines. The evaluation of the non-ADT group permits us to account for any influence of the disease (PCa) itself on cytokine expression, while the control group allows us to account for any influence that aging may have on these inflammatory markers. Men in the ADT and non-ADT groups were recruited from the Kimmel Cancer Center at Johns Hopkins. The age-matched control group was recruited from a database of eugonadal men at the Johns Hopkins Hospital Clinical Trials Unit. None of these subjects were suffering from any acute or chronic inflammatory conditions. The cytokine levels were measured in those subjects in the 3 groups for whom sufficient serum samples were available (Table). All subjects signed informed consent that was approved by the Institutional Review Board of the Johns Hopkins Medical Institutions (Basaria et al, 2002).

Exclusion Criteria

Men were excluded from the study if they had any of the following: liver function tests or serum creatinine >2 times the upper limit of normal, glucocorticoid use in the previous 3 months, history of thyroid disease, history of any form of hypogonadism prior to the diagnosis of PCa (both ADT and non-ADT groups), and past or present cytotoxic chemotherapy.

Hormonal Methods

Total and free testosterone, PSA, lipid profile, and glucose levels were measured commercially (Quest Diagnostics). The normal range for total testosterone was 241–827 ng/dL and for free testosterone 8–24 ng/dL. Serum concentrations of inflammatory [MIP-1b, TNF-, IL-7, IL-8, IL-12 (p70), IL-13] and anti-inflammatory (IL-10) cytokines were assessed using a Luminex soluble human multiplex cytokine assay according to the manufacturer's instructions (LINCO Research Inc, St Charles, Mo). Briefly, thawed sera were assayed for soluble cytokines using the Luminex fluorescent microsphere technology and compared to the fluorescence intensity values derived from standard curves for each cytokine. Samples were incubated with fluorescent microspheres coupled to antibodies to MIP-1b, TNF-, IL-7, IL-8, IL-10, IL-12 (p70), and IL-13 and were analyzed using a Luminex100 cytometer fitted with an XY Platform and Luminex Data Collector software (Luminex Corporation, Austin, Tex). Fluorescent beads (300 beads per analyte) that were bound by cytokine were differentiated from free beads through the use of a phycoerythrin-bound reporter antibody to the cytokine analyte. Based on the standard curves generated for each cytokine, concentrations were determined for each serum sample and are expressed as pg/mL. The intra-assay precision of these assays range from 4.4%–7.6% CV (average 5.6% CV). The sensitivity of this assay system for these various cytokines using patient sera ranged from 0.5–16 pg/mL.

Statistical Analysis

T tests were used to test for differences in mean age and BMI between the 3 two-sample combinations of the groups (ADT vs non-ADT, ADT vs controls, and non-ADT vs controls). The distributions of age and BMI were separately examined across the 3 groups using ANOVA. Nonparametric Mann-Whitney tests were used to test for differences in the medians of the cytokines, testosterone, and estradiol between the 3 two-sample combinations of the groups. The distributions of these variables across the 3 groups were examined separately using nonparametric Kruskal-Wallis tests. The distribution of race (white vs nonwhite) between the 3 two-sample combinations of groups and all 3 groups simultaneously was examined using Fisher's exact test.

Table 1. 
       P Value for Pairwise Group Comparisons†P Value for Between-Groups Comparison‡
 ADT GroupNon-ADT GroupControlsADT vs Non-ADTADT vs ControlsNon-ADT vs Controls
 Value*NValueNValueN
  1. a

    Demographics, hormonal and cytokine profile of the 3 groups

  2. *All values are the median, except for age and body mass index, which are mean (SD), and race.

  3. P values for t tests for difference in means for variables age and body mass index; P values for nonparametric Mann-Whitney tests for differences in medians between cytokine, testosterone, and estradiol variables; P value for Fisher's exact test for differences in distribution of race between groups.

  4. P values for ANOVA for differences in means across groups for variables age and body mass index; P values for nonparametric Kruskal-Wallis tests for differences in medians across groups for cytokine, testosterone, and estradiol variables; P value for Fisher's exact test for differences in distributions of race across groups.

Age (years)69.9 (7.8)2066.2 (10.0)1869.1 (8.2)20.22.77.34.41
Race, white (%)15 (75)2017 (94)1818 (90)20.18>.99.41.25
Body mass index (kg/m2)29.6 (4.8)2027.6 (3.6)1824.7 (3.0)20.16.0005.01.0009
Total testosterone (ng/dL)5202961850620<.0001<.0001.0002.0001
Free testosterone (ng/dL)0.352012.11813.220<.0001<.0001.24.0001
Estradiol (pmoL/L)02012181620<.0001<.0001.06.0001
MIP-1b (pg/mL)121.11775.91391.818.98.29.40.53
Interleukin-7 (pg/mL)11.2187.9179.319.18.13.61.24
Interleukin-8 (pg/mL)9.4118.4159.115.84.98.52.86
Interleukin-10 (pg/mL)1.6172.4152.317.35.18.82.40
Interleukin-12p70 (pg/mL)7.4166.9156.018.74.77.59.85
Interleukin-13 (pg/mL)4.2164.3176.318.99.48.44.68
TNF-α (pg/mL)4.3168.5156.818.11.88.21.27

Results

The demographic and laboratory data are summarized in the Table. There were no significant differences in distributions of age across the 3 groups (P for ANOVA = .41). The mean body mass index (BMI) in the ADT and non-ADT groups was significantly higher than the control group (P = .0005 and P = .01, respectively). There was no significant difference in mean BMI between the ADT and non-ADT groups. Mean PSA levels (SD) were 3.15 ng/mL (4.73), 9.55 ng/mL (17.0), and 2.24 ng/mL (1.43) in the ADT, non-ADT, and control groups, respectively (overall P = .06). Men in the ADT group had significantly lower serum total (P < .0001) and free (P < .0001) testosterone and estradiol (P < .0001) levels than the non-ADT and control groups. Men in the non-ADT and control groups were eugonadal, with total testosterone level higher in the former group. No significant differences in median serum levels of pro-inflammatory cytokines or anti-inflammatory cytokine IL-10 were observed between the 3 treatment groups (Table).

Discussion

In this study, we found that men with PCa undergoing long-term ADT did not demonstrate elevated levels of inflammatory cytokines compared to age- and disease-matched controls. These findings suggest that at least long-term androgen deprivation in this patient population is not associated with significant changes in selected cytokines.

Review of the literature suggests that testosterone does indeed exert immunosuppressive properties. This is supported by the observation that men with Klinefelter syndrome have a higher incidence of autoimmune diseases than their age-matched counterparts (Bizzarro et al, 1987). We recently showed in a cross-sectional study of elderly men that serum testosterone levels were inversely related to soluble IL-6 receptor (sIL-6r) (Maggio et al, 2006). Importantly, interventional studies with testosterone replacement have revealed similar information. A recent study of testosterone replacement in hypogonadal men showed a significant decrease in pro-inflammatory and a significant increase in anti-inflammatory cytokines (Malkin et al, 2004).

Based on this paradigm, we hypothesized that ADT in men with PCa (which results in profound hypogonadism) would be associated with elevated serum levels of pro-inflammatory cytokines. However, our results did not support this hypothesis. Our current findings are in contrast to those of Khosla et al (2002). In this study, Khosla and coworkers induced hypogonadism in healthy elderly men and found that short-term suppression of gonadal axis resulted in an increase in circulating pro-inflammatory cytokine (TNF-alpha, IL-6, and sIL-6r) expression. The discrepancy between our results and the Khosla study may be explained by the duration of hypogonadism. In the Khosla study, the duration of hypogonadism was only 4 weeks, while the minimal duration of ADT in our study was 12 months. Hence, it is conceivable that during the acute phase of hypogonadism there is an increase in inflammatory cytokines, but that over time there may be an adaptation of the immune system, resulting in normalization in circulating cytokines.

Few prior studies have measured inflammatory cytokines in men with PCa and attempted to correlate cytokine levels with the extent of their disease. One study reported elevated levels of inflammatory cytokines in cachectic men with advanced PCa (Pfitzenmaier et al, 2003). In our study, although some men on ADT had demonstrated metastatic disease before the initiation of ADT, their disease was in remission and their metastases had resolved prior to the initiation of the study. Similarly, men in the non-ADT group were also healthy and only showed evidence of biochemical recurrence (elevated PSA) at the time of recruitment (Basaria et al, 2002). Hence, absence of cancer cachexia in our study population may account for the difference in our results from the subjects studied by Pfitzenmaier et al. In another study, Wise and colleagues measured inflammatory cytokines in 3 groups of men, namely those with hormone sensitive PCa, hormone refractory PCa, and untreated benign prostatic hyperplasia (BPH) (Wise et al, 2000). The authors found increased levels of IL-6, IL-4, and IL-10 only in the hormone refractory group. In contrast, in our study all men in the ADT group had responded successfully to ADT and were in remission. Hence, the patient populations in these 2 studies are quite distinct.

Our current study possesses several strengths. Firstly, we examined men who had been on ADT for a long period of time (range 1–9 years). To the best of our knowledge, this is the first study that has examined the levels of inflammatory cytokines in patients on long-term androgen deprivation. Secondly, we included 2 different groups of men to compare to the ADT group. The evaluation of a non-ADT group permitted us to account for any influence of PCa itself on cytokine expression, and the control group provided us the opportunity to account for the influence of age on cytokine levels. Lastly, we evaluated a wide array of cytokines in this study. Despite these strengths, our study also has several limitations. First, this study was cross-sectional. Hence, prospective studies would be necessary to determine if changes in cytokine levels occur immediately after the initiation of ADT. Second, the majority of men in the 3 groups examined were Caucasians. Future studies should include subjects from diverse ethnic backgrounds. Finally, we did not measure IL-6, sIL-6r, C reactive protein, or TNF-alpha receptor levels in this study. Measurement of these markers could have strengthened our data.

In conclusion, our data suggest that long-term ADT in men with prostate cancer is not associated with a significant increase in pro-inflammatory or decrease in anti-inflammatory cytokines. These data should be confirmed in prospective studies.

Ancillary