| |
Serum prostate-specific antigen levels in older men with or at risk of HIV infection
|
| |
| |
HIV Medicine
Volume 7 Page 471 - October 2006
LE Vianna1, Y Lo2 and RS Klein1,2
1Division of Infectious Diseases, Department of Medicine and 2Division of AIDS Research, Department of Epidemiology & Population Health, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY, USA
"....Risk factors for prostate cancer include age, family history, race, and possibly diet. Prostate cancer mainly affects men older than 65 years. It is more common and is associated with increased mortality in African-American men compared with white men... Age-appropriate screening for prostatic cancer should be encouraged for all men, regardless of their HIV status....
...In this study of HIV-infected and -uninfected at-risk older men, only increasing age was associated with elevated PSA levels; there was no association with HIV status. Furthermore, levels >4.0 ng/mL were found in only 4% of men, suggesting that it is unlikely that HIV infection causes a substantially increased rate of false positive assays....Some studies suggest that a diet high in dairy products and animal fat or meat may increase the risk of prostate cancer, while fish fat consumption and diets rich in fruits, vegetables and soybean products may be associated with a lower risk.....A possible association of prostate cancer risk with immunosuppression is suggested by the occurrence of prostate cancer in patients who have received transplanted organs and immunosuppressive therapy; most reports of such patients have involved kidney transplantation...."
Abstract
Objectives
The aim of the study was to determine the rate of, and factors associated with, elevated prostate-specific antigen (PSA) levels in older men with or at risk of HIV infection.
Methods
Using a cross-sectional analysis, we interviewed 534 men ≥49 years old at risk for HIV infection on demographics, behaviours and medical history. Laboratory testing included serum PSA level and HIV serology, and T-cell subsets for those who were HIV seropositive. Elevated PSA level was defined as >4.0 ng/mL, and men with elevated PSA levels were referred for urological evaluation.
Results
Fifteen per cent of men were white, 55% black, and 23% Hispanic; median age was 53 years (range 49-80 years); 74% were sexually active; 65% currently smoked cigarettes; and 16% had taken androgens. Among 310 HIV-positive men, CD4 counts were >500 cells/L in 31%, 200-500 cells/L in 51%, and <200 cells/L in 19%. Twenty men (4%) had elevated PSA. On univariate analysis, only older age was significantly associated with elevated PSA, and there was no significant difference in the number of men with elevated PSA between HIV-positive and HIV-negative men (nine of 310 vs 11 of 224; P=0.28). On multivariate analysis, older age remained the only variable associated with elevated PSA level [reference group ≦50 years; adjusted odds ratio (ORadj) 1.0 for age 51-60 years; ORadj 5.9 (95% confidence interval 1.2-30.1) for age ≥61 years] adjusted for HIV status, family history of prostate cancer, and androgen use.
Conclusions
Among older men, PSA levels increased with age but did not differ by HIV status. The clinical use of PSA levels in older men currently do not need to be modified for those with HIV infection.
Introduction
Persons with HIV infection have increased rates or altered clinical presentations of several neoplastic diseases. Among these are lymphoma, Kaposi's sarcoma and anogenital neoplasia [1-3]. Other malignancies also occur in HIV-infected individuals, but their possible relationship with HIV infection is less well defined. Potential mechanisms for the association of neoplasms with HIV infection include a direct effect of viral proteins [4,5], decreased immune surveillance [6], cytokine dysregulation, or other immunological or viral cofactors [5].
With improved survival among persons with HIV infection with access and adherence to highly active antiretroviral therapy (HAART), persons with HIV infection are ageing [7,8]. With advancing age there is increased risk for certain cancers [9-11]. As a result, the implementation of age-appropriate screening for certain malignancies becomes an increasingly important consideration.
Although the use of HAART has led to decreased incidence and improved outcomes of Kaposi's sarcoma and non-Hodgkin's lymphoma [12-15], this has not been the case for certain other cancers, such as anal [16] and cervical cancer [17]. Even for neoplasms not shown to occur with increased incidence in ageing persons with HIV infection, risk is likely to be at least as high as in persons of similar age without HIV infection. Also, there might be altered presentation of these disorders in HIV-infected individuals. Therefore, it is important to assess whether HIV infection and its related immune suppression alter the incidence rate or natural history of other malignancies and to determine their possible effect on the predictive value of any available screening tests.
Prostate cancer is the second most common form of cancer and second only to lung cancer as a cause of cancer-related death among men in the USA [18]. Although there have been anecdotal reports of prostate cancer in men with HIV infection [19,20], large studies comparing the frequency of prostate cancer in men with HIV infection with that in men without HIV infection have failed to show a significantly increased risk with HIV infection [2,3,12,15,21,22].
There has been little published on the utility among older HIV-infected men of prostate-specific antigen (PSA) screening, which generates controversy even for use in men without immunosuppression. Determination of PSA level seems likely to become an increasingly important screening test among HIV-infected men as they age. The availability of a research cohort of older men with or at risk of HIV infection offered the opportunity to study whether HIV infection might affect PSA levels in older men.
Discussion
In this study of HIV-infected and -uninfected at-risk older men, only increasing age was associated with elevated PSA levels; there was no association with HIV status. Furthermore, levels >4.0 ng/mL were found in only 4% of men, suggesting that it is unlikely that HIV infection causes a substantially increased rate of false positive assays. Although only seven (35%) of the men with elevated PSA levels are known to have undergone urological evaluation, two of these (29%) were found to have prostate cancer.
Since approval of PSA testing in 1986 by the US Food and Drug Administration, its proper clinical use and value have remained controversial [23]. Current recommendations emphasize the importance of patient education and involvement in the decision to test, with a clear understanding of the benefits and limitations of testing being essential for patients. Most agree that serum PSA should be used in conjunction with annual digital rectal examination (DRE). The patient should also be aware of the options for further evaluation and treatment if an elevated level is detected [24,25].
After the high incidence of prostate cancer led to the widespread use of PSA screening in attempts to detect cancer at an early stage, a consensus developed that a PSA level >4.0 ng/mL had good predictive value for prostate cancer [26], and the rate of subsequent diagnosis of disease increased substantially [27]. It has been suggested that the level at which PSA should be considered abnormal should be modified by age and race [28]. However, controversy remains regarding the optimal PSA level for detecting clinically relevant disease. Recently, in a large cancer prevention trial, the prevalence of prostate cancer found was substantial even among the 2950 men in their seventh to tenth decades of life with lower PSA levels (<4.0 ng/mL). After 7 years, the proportions of men found to have prostate cancer ranged from 6.6% among those with PSA levels ≦0.5 ng/mL to 26.9% in those with PSA levels of 3.1-4.0 ng/mL [29].
Risk factors for prostate cancer include age, family history, race, and possibly diet. Prostate cancer mainly affects men older than 65 years. It is more common and is associated with increased mortality in African-American men compared with white men. Some studies suggest that a diet high in dairy products and animal fat or meat may increase the risk of prostate cancer, while fish fat consumption and diets rich in fruits, vegetables and soybean products may be associated with a lower risk [30,31].
A possible association of prostate cancer risk with immunosuppression is suggested by the occurrence of prostate cancer in patients who have received transplanted organs and immunosuppressive therapy; most reports of such patients have involved kidney transplantation [32-35]. In a population-based study in Norway, in which risk for various cancers was calculated using information on 5692 renal transplant recipients, national cancer registries and population registries, risk of prostate cancer was found to be significantly increased for men who had received a transplant [33]. Cases of prostate cancer after liver transplantation [36-38] and heart transplantation [38,39] have also been reported. Recent guidelines regarding the prevention and treatment of solid organ cancers recommend that male renal transplant recipients ≥50 years old have yearly PSA testing prior to regular DRE. However, the effects of a regimen of post-transplant immunosuppressive drugs are not identical to the immunosuppression of HIV, and persons undergoing transplant may differ with respect to baseline risk factors for prostate cancer compared with HIV-infected men, so that a difference in prostate cancer risk in the two groups would not be surprising. A recent review of the increased incidence of prostate cancer with age also suggested the importance of the host immune response in prostate cancer risk; the authors concluded that the prostate is an immune-privileged site and that local immune surveillance is progressively further affected with increasing age [6].
As survival with HIV infection improves, infected persons are ageing [7,8]. Despite little evidence, as yet, suggesting an association of HIV infection with prostate cancer risk, screening for prostate cancer will become increasingly important among HIV-infected men as they age. This fact prompted this investigation of PSA levels in older HIV-infected men. One prior study of risk for prostate cancer utilizing PSA levels in men with HIV infection found that seven (3.2%) of 216 men had elevated PSA levels; none had abnormal DREs, three were diagnosed with prostatitis, and five (2.3%) had had prior prostate neoplasia, but only a single participant (0.5%) was newly found to have prostatic intraepithelial neoplasia [40]. Including those with prior diagnoses of prostate cancer, four (36.4%) of 11 men ≥60 years old were found to have prior cancer; all had elevated PSA levels (range 8.0-15.9 ng/mL). The high rate of prostate cancer in older men with HIV infection found in that study suggests a substantial risk for this population. However, the number of older men included was small, so that the 95% confidence intervals were wide (10.9-69.2%). Furthermore, while prostate screening was offered to all HIV-infected men ≥35 years old who were attending an infectious disease clinic, there was no information provided as to the proportion of such men who participated or the characteristics of those who did not, nor was an HIV-uninfected comparison group included. It is not possible to assess whether there might have been a study bias, with men who had prior prostate cancer being more likely to participate in the evaluations.
In the present study, among older men with or at risk for HIV infection, only 4% had elevated PSA levels, with no significant difference by HIV status. Older age was the only factor associated with elevated PSA level. Only two men, representing 0.37% of the entire study sample, were found to have newly diagnosed prostate cancer. However, the majority of men with elevated PSA levels did not undergo recommended urological evaluation, highlighting the difficulty of delivering optimal care to this population. If all 13 men with elevated levels who did not complete urological evaluations had prostate cancer, which seems unlikely, the rate of prostate cancer would be 2.8% of the study sample.
Although this study does not provide new information on the sensitivity, specificity and predictive values of PSA screening, it does inform the use of PSA screening in older men at risk for HIV infection. The lack of an association of HIV status with rate of elevated PSA levels found in this study, combined with the accumulating data suggesting no increased rate of prostate cancer in men with HIV infection, suggests that at present the use of PSA screening and further evaluation of elevated PSA levels in older men currently need not be modified for those with HIV infection. Age-appropriate screening for prostatic cancer should be encouraged for all men, regardless of their HIV status.
Results
Of the 643 men in CHAMPS, 534 men (83%) who completed a second study visit and provided additional informed consent were included in this study. CHAMPS participants who agreed to PSA testing did not differ significantly from those not tested in HIV status, age, smoking, use of androgens, being sexually active, having been previously diagnosed with cancer, or reporting a father or brother having had prostate cancer. However, they were more likely to report ever having had sex with men [146 of 534 (27.3%) vs 17 of 102 (16.7%); P=0.02] and to be Hispanic [124 of 534 (23.2%) vs 42 of 109 (38.5%); P<0.001].
Of the 534 men included in this analysis, 224 (42%) were HIV-seropositive and 310 (58%) were seronegative. Fifteen per cent of the study sample were white, 55% black, and 23% Hispanic; median age was 53 years (range 49-80 years); 74% were sexually active; 27% reported ever having had sex with men; 65% currently smoked cigarettes and an additional 25% had smoked in the past; 16% had taken androgens at some time in the past, of whom 55% were currently taking androgens; 6% had previously been diagnosed with cancer, including six with prostate cancer; and 9% reported that a father or brother had been diagnosed with prostate cancer. Among HIV-positive men, CD4 counts were >500 cells/L in 31%, 200-500 cells/L in 51%, and <200 cells/L in 19%.
The baseline characteristics of the participants by HIV status are shown in Table 1. HIV-infected participants were significantly younger by a mean of 1 year and were more likely to have had sex with men, to have used androgens in the past or be using them at the time of the study, and to have been diagnosed previously with cancer. Among those who had prior diagnosis of cancer, two (28.6%) of seven HIV-seronegative men and four (16.7%) of 24 seropositive men had prior prostate cancer.
Twenty men (4%) had PSA levels >4.0 ng/mL. The associations of participant characteristics with PSA levels are shown in Table 2. On univariate analysis, older age and current smoking were significantly associated with PSA level. Older age was the only variable significantly associated with an increased prevalence of elevated PSA level. There was no significant difference between HIV-seropositive and HIV-seronegative men; nine (2.9%) of 310 HIV-seropositive men and 11 (4.9%) of 224 HIV-seronegative men had elevated PSA levels (P=0.28). None of the six men with a history of prior prostate cancer had elevated levels; exclusion of these men from the analysis had no effect on our findings. On multivariate analysis, older age remained the only variable significantly associated with elevated PSA levels [reference group ≦50 years; adjusted odds ratio (ORadj) 1.0 for age 51-60 years; ORadj 5.9 (95% confidence interval 1.2-30.1) for age ≥61 years], adjusted for HIV status, family history of prostate cancer, and androgen use.
All 20 men with elevated PSA levels were encouraged to obtain urological evaluation and were offered expedited referral to specialists. However, for only nine (45%) of these could we find records of further evaluation. One (9%) of 11 HIV-seronegative and one (11%) of nine HIV-seropositive men had diagnosis of prostate cancer confirmed histologically after biopsy, two other men had prostate biopsy with no malignancy detected, one other was diagnosed with chronic prostatitis, two had further evaluation postponed after repeat measurements of serum PSA levels by their care providers were reportedly within the normal range, and two had repeat PSA assays confirming elevated levels, but refused further work-up. No follow-up information was available on the remaining 11 men.
Methods
The Cohort of HIV at-risk Ageing Men's Prospective Study (CHAMPS) enrolled 643 men aged 49 years or older with or at risk of HIV infection. The primary aims of that study are to examine the possible association of HIV infection with atherosclerosis, bone loss and drug use among older men. Men were recruited from the community, via advertisements and by word of mouth, from drug treatment programmes and medical clinics, and from participants in a prior study of the natural history of HIV infection. Men were eligible if they were aged 49 years or older; were HIV-uninfected or -infected, as determined by enzyme immunoassay (EIA) for HIV antibody with western blot confirmation of positive results; were capable of providing informed consent; were able to interact and participate in the interview process; had a history of (1) drug use, (2) sex with a man, (3) sex with a woman known by the study participant to have HIV infection, (4) five or more sexual partners in the previous 5 years, or (5) having exchanged sex for money or drugs. Semi-annual visits with confidential interviews focusing on demographic information, medical and family history, drug use and sexual behaviours were conducted by trained research staff. Laboratory assessment included phlebotomy for HIV status and, if positive, CD4 cell count and HIV viral load testing.
At the second study visit, participants were asked to provide separate written informed consent for PSA testing, for the purposes of the present study. The Immunlite 2000 PSA analyser (Diagnostic Products Corporation, Los Angeles, CA), an immunometric assay, was used for the quantitative measurement of PSA on serum samples that had been obtained on participants' initial visits and stored at -70C. PSA levels >4.0 ng/mL were considered elevated. All men with elevated PSA levels were offered referral to designated urologists for evaluation.
Univariate analysis of baseline characteristics was performed using the chi square (2) test or Fisher's exact test for categorical variables, and Student's t-test, the Wilcoxon rank-sum test or the Kruskall-Wallis test for continuous variables. Logistic regression models were used to identify factors associated with elevations in PSA. Variables examined included age, race, cigarette smoking, use of androgens, having had sex with men, family history of prostate cancer, HIV status, CD4 cell count, and HIV viral load from the same (initial) study visit. Data were analysed using EPI INFO 2002 (Centers for Disease Control and Prevention, Atlanta, GA), and SAS for Windows version 8.1 (SAS Institute, Inc., Cary, NC, USA).
This study was approved by the Montefiore Medical Center Institutional Review Board for the Protection of Human Subjects.
|
|
| |
| |
|
|
|
