Registry of Hypogonadism in Men (RHYME): design of a multi-national longitudinal, observational registry of exogenous testosterone use in hypogonadal men

Abstract

Objective: Despite the prevalence of hypogonadism (HG) and widespread use of testosterone therapy, little is known about the safety/effectiveness of long-term testosterone use. The Registry of Hypogonadism in Men (RHYME) is a multi-national patient registry assessing prostate health and other outcomes associated with testosterone treatment in men.

Design: Observational patient disease registry.

Methods: RHYME is a non-interventional disease registry with longitudinal data collection on a large sample (N = 999) of well-characterized, hypogonadal men aged 18 years or older. The Registry will prospectively evaluate male patients diagnosed with HG, who have not previously been treated with testosterone therapy. Key design features include: (1) broad inclusion/exclusion criteria, (2) standardized central laboratory hormone assays, (3) independent adjudication of prostate biopsies and mortalities, (4) standard of care treatment, (5) comprehensive medical record and questionnaire data at six months and annually post-enrollment and (6) adequate statistical power for assessing prostate endpoints at 36 months.

Results: A total of 25 clinical sites in six European countries (Germany, Italy, the Netherlands, Spain, Sweden and the United Kingdom) have completed recruitment for the study. Recruitment was initiated in May 2009, and completed in December 2011. Data collection is ongoing with a minimum of two years of follow-up on all patients.

• Disease registry
• hypogonadism
• prostate cancer
• testosterone deficiency

Introduction

Hypogonadism (HG) in men is a clinical syndrome comprised of symptoms and signs of androgen deficiency confirmed by persistently low circulating testosterone levels. Primary or hypergonadotrophic HG results from a failure of testicular function (e.g. Klinefelter’s syndrome), while secondary or hypogonadotrophic HG is a consequence of reduced gonadotropin production due to pituitary and/or hypothalamic diseases or dysfunctions. In contrast to these cases of classical HG, decreased testosterone in middle-aged or older men, commonly referred to as late-onset hypogonadism (LOH) when associated with symptoms of androgen deficiency, generally occurs in the absence of recognizable pathologies in the testes, pituitary or hypothalamus. Recent data from the European Male Aging Study [1] indicate that the prevalence of LOH is 2.1% in middle-aged and elderly men, and increases progressively with age [2].

Symptomatic benefits of testosterone supplementation in men with classical HG have been demonstrated across many settings [3–6]; however, long-term safety data are limited, which is of particular concern in men with LOH being started on testosterone in late middle age when the background risks of cardiovascular disease and prostate cancer are increasing [7,8]. While the traditional view of endogenous or exogenous testosterone as a risk factor for prostate cancer or benign prostatic hyperplasia (BPH) in men has recently been challenged [9,10], there remain significant concerns regarding long-term safety of testosterone supplementation in men with LOH [11]. Meta-analyses of existing clinical trials [12–14] have generally failed to provide definitive outcome data regarding the potential risk of cardiovascular and prostate events. These gaps in evidence about long-term safety and potential adverse effects of treatment, as well as lack of familiarity with current treatment options may be major barriers to effective management of HG.

Previous large-scale observational studies have examined short-term outcomes of testosterone supplementation in product registries in the United States and other countries [15,16]. Major limitations of these studies include the use of a single product or treatment option for all patients (Testim®, Nebido®), limited follow-up periods (12 months or less) and lack of central laboratory measurements or independent adjudication of key safety outcomes.

The Registry of Hypogonadism in Men (RHYME) is designed to provide longitudinal data in an observational cohort design for assessing prostate safety and other long-term outcomes associated with the use of testosterone among men diagnosed with HG. The primary objective of RHYME is to monitor prostate health outcomes, including incident prostate cancer rates relative to the number of biopsies performed, in men diagnosed with HG who may or may not be receiving testosterone or other androgen-based treatments. The specific type of testosterone substitution and duration of use will be systematically recorded. Secondary outcomes include changes in BPH symptoms, cardiometabolic health and quality of life outcomes including sexual function. A uniquely comprehensive outcome assessment, rigorous registry design, central measurement of key laboratory parameters and external adjudication of all study endpoints are being performed.

Methods

Registry design

RHYME is an observational, non-interventional disease registry with longitudinal data collection on a large sample (N = 999) of well-characterized, hypogonadal men. The Registry is designed to enroll and prospectively evaluate adult men aged 18 years and older with diagnosed HG, with no history of treatment with androgen therapy. RHYME was designed in accordance with Agency for Healthcare Research and Quality guidance on registries for evaluating patient outcomes [17]. Clinical and laboratory measures are administered according to a protocol-defined schedule of assessments that was selected to correspond with the current standard of care for HG [11,18] (see Table 1). Additional diagnostic tests or procedures, including prostate biopsy or ultrasound testing, are determined by the treating physician in consultation with the patient. This clinical decision-making is intended to reflect actual clinical practice and increase generalizability of the RHYME findings, and is consistent with the non-interventional nature of the Registry design.

Table 1. Patient eligibility criteria.

Inclusion criteria

Exclusion criteria

• Male patients with a current diagnosis of HG • Aged 18 years and older (maximum: 150 men <50 years) • Informed consent

• Any previous treatment with testosterone therapy for any reason • History of certain cancers or high-grade prostatic intraepithelial neoplasia • Prior radical prostatectomy • Life expectancy shorter than 24 months • Current major psychiatric disorders or drug/alcohol abuse • Gender dysphoria or sexual reassignment (e.g. transsexualism) • Enrolled in any interventional clinical trial within the past 30 days • Planned relocation outside site region within 24 months

Inclusion and exclusion criteria

Inclusion and exclusion criteria are shown in Table 2. All eligible patients are enrolled consecutively, a key element of the design for reducing selection bias.

Table 2. Registry baseline and follow-up measures.

Symptoms of HG	Concomitant medications	Central laboratory results
ED	Androgen replacement	Total testosterone (LC-MS)
Decreased desire for sex	Other hormone replacement	SHBG
Decrease in spontaneous (morning) erections	ED drugs	LH*
Fatigue/muscle weakness/strength decrease	BPH/LUTS drugs	PSA
Inability to concentrate	Anti-hypertensive drugs	Local laboratory results
Depressed mood	Anti-anginal drugs	Hematocrit
Sleep disturbance	Lipid lowering drugs	Total testosterone
Infertility	Anti-diabetic drugs	Free testosterone
Hot flushes/flashes (sweats)	CNS drugs	SHBG
Breast enlargement and/or tenderness	Peptic ulcer drugs	LH
Loss of body hair/decreased shaving	Anti-coagulants	PSA
Osteoporosis or fracture	Anticonvulsants	Glucose
Poor sexual development	Analgesics	Total cholesterol
Other	Corticosteroids	High density lipoprotein
Medical history and comorbidities	Cancer chemotherapy	Triglycerides
Prostate biopsies	Anti-tuberculosis drugs	Creatinine
Prostate cancer†	Medications for HIV/AIDS	Blood urea nitrogen
Familial prostate cancer*	Testosterone†	Albumin
Cardiovascular disease	Demographics*	Bone density scan
Pulmonary disease	Age*	TRUS
Gastrointestinal disease	Education*	Physical examination
Hepatic disease	Race*	Height
Renal disease	Ethnicity*	Weight
Urologic disease	Marital status*	Body mass index
Endocrine disease	Health behaviors	Waist circumference
Hematologic disease	General health	Pulse
Musculoskeletal disease	Physical activity	Blood pressure
Neurological disease	Smoking	DRE
Psychiatric disease	Alcohol use	Gynecomastia
Infectious disease	Patient questionnaires	Tanner staging
Other cancers	AMS scale	Testicular size
Surgeries	IPSS	Prostate enlargement/abnormalities
Radiotherapy	IIEF

HG = Hypogonadism; ED = erectile dysfunction; BPH/LUTS = benign prostatic hypertrophy/lower urinary tract symptoms; CNS = central nervous system; AMS = Aging Male Symptoms; IPSS = International Prostate Symptom Scale; IIEF = International Index of Erectile Function; LC-MS = liquid chromatography tandem mass spectrometry; SHBG = sex hormone binding globulin; LH = luteinizing hormone; PSA = prostate-specific antigen; DRE = digital rectal exam; TRUS = transrectal ultrasound.

*Baseline visits only.

†Follow-up visits only.

Registry assessment schedule

Figure 1 shows the schedule of assessments for baseline through two years of follow-up data collection. At all visits, decisions to order relevant follow-up testing such as ultrasound, prostate biopsy or urology consult are carefully documented. The indications and occasions for referral are determined by treating physicians and patients in consultation, in keeping with the non-interventional registry design. Because positive prostate biopsy rate is the primary endpoint, clinical indications for prostate biopsy, such as an abnormal digital rectal exam (DRE) or rise in prostate-specific antigen (PSA), are documented. Important medical events that occur between visits, such as transrectal ultrasound (TRUS), urogenital surgery (e.g. treatment for acute urinary retention, transurethral resection of the prostate, other pelvic surgery), cardiac or cerebrovascular events (e.g. stroke, myocardial infarction) and cancer diagnosis (e.g. prostate, breast and other cancers), are routinely captured at all Registry visits.

Figure 1. Registry design and schedule of assessments.

Registry measures

Table 2 provides a detailed list of the registry measures at baseline and follow-up, including symptoms of HG, medical history and comorbidities, demographics, health behaviors, local laboratory results, physical examination and patient questionnaires. Careful consideration was given to the selection of an optimal set of self-report measures, including the International Index of Erectile Function [19], International Prostate Symptom Score (IPSS) [20] and the Aging Male Symptoms (AMS) Scale [21,22]. Additionally, the collection of several medical and physical health outcomes allows for the assessment of cardiovascular risk status according to algorithms developed and validated in the Framingham study [23,24].

Prostate cancer

All prostate biopsies and supporting clinical documents are subject to review by the Prostate Health and Clinical Endpoints Committee (PHCEC). The review includes access to all clinical data and pathology reports, but does not include access to tissue samples. The committee adjudicates the outcome as positive or negative for the presence of prostate cancer independent from their testosterone treatment status. Adjudication includes determination of the final Gleason score and proportion of positive biopsy cores. Other diagnoses (e.g. BPH, prostatitis) are also recorded.

Other prostate outcomes: PSA, DRE, TRUS and BPH

Local laboratory PSA measures are entered into the medical record database, and central laboratory PSA measures are also obtained. DRE and TRUS are performed only as medically indicated. All patients are required to complete a BPH symptom checklist (IPSS) at baseline and at each registry visit. In addition to these independent assessments, the clinician is asked to choose a clinical indication (rise in PSA or abnormal PSA, DRE or TRUS) for each prostate biopsy performed. Urogenital surgical procedures and cancer-related outcomes are monitored at every study visit.

Mortality

Mortality events are also reviewed by the PHCEC, with assignment of the primary cause of death, to evaluate whether or not cancer was involved, and to assess whether or not the patient’s death was due to cardiovascular or other causes.

Central laboratory rationale

A central laboratory (ICON Laboratories, Dublin, Ireland) was selected to provide standardization of key laboratory measurements across clinical sites. Serum testosterone is assessed by liquid chromatography tandem mass spectrometry (LC-MS). Other key laboratory parameters, PSA, sex hormone binding globulin (SHBG) and luteinizing hormone (LH) are assayed centrally. To prevent any interference with standard clinical care, individual central laboratory results are not made available to site investigators or patients.

Registry organization

RHYME is funded by an investigator-initiated grant from Bayer Pharma AG (Berlin, Germany) to New England Research Institutes, Inc. (NERI) (Watertown, MA), which serves as the registry sponsor and coordinating center (Figure 2). The Scientific Advisory Committee (SAC) meets annually to provide independent scientific review and oversight, and the PHCEC performs blinded adjudication of all prostate biopsies and other clinical outcomes. All decisions regarding the registry protocol and endpoints are made by NERI with oversight by the SAC. The grantor has no role in these decisions.

Figure 2. Registry organizational structure.

Data quality assurance

Registry data are monitored for completeness and accuracy according to a pre-specified data monitoring plan. This plan includes remote monitoring of each clinical site at least twice during registry participation, with source document verification performed for a randomly selected subset of patients, in addition to on-site monitoring of the top-enrolling site and two additional randomly selected clinical sites.

Statistical power and analysis plan

A unique feature of RHYME is the incorporation of precise sample size estimates based on review of relevant literature and population-based data. Statistical power estimates were developed prior to registry implementation. Calculations informed our plan to enroll 1000 European patients in six countries with a minimum of two years follow-up on all patients.

The primary endpoint for RHYME is the proportion of positive prostate biopsies observed after two years. Key assumptions regarding the evaluation of this endpoint are as follows:

• testosterone/hormonal therapy use in 90% of men;

• biopsy rate 3–6% [25,26];

• no more than 15% of sample <50 years of age;

• drop-out rate no more than 10% after first year of follow-up.

Based on these assumptions, we expect to observe between 60 and 118 prostate biopsies during the two-year follow-up period. For our primary statistical comparison, we further assumed that the expected positive biopsy rate under the null hypothesis of no effect of testosterone therapy would be 30% [15,25–29]. Thus, the null hypothesis is: H₀: p = 0.30 and the alternative is H_A: p ≠ 0.30, where p equals the proportion with prostate cancer, defined as a positive biopsy result. A two-sided test indicates that if 85–90 men undergo biopsy, there is approximately 80% power to detect a 50% relative increase in risk (15% absolute difference, i.e. 30% versus 45% positive biopsies for untreated and testosterone-treated men).

Results

Registry implementation

A total of 25 clinical sites in six countries (Italy, Germany, the Netherlands, Spain, Sweden and the United Kingdom) are currently participating in RHYME (see Figure 3). Following a feasibility assessment of the first 25 patients enrolled at five vanguard sites, site and patient enrollment was initiated at an additional 20 sites. The distribution of clinical sites and investigators is shown in Figure 3. Clinical sites were selected to represent investigators in endocrinology, urology, andrology, and primary care with familiarity and experience in the diagnosis and management of HG. Institutional Review Board approval was obtained at the registry coordinating center and local ethics committee approval and informed consent were obtained at each clinical site prior to initiation of registry activities.

Figure 3. Geographic distribution of RHYME clinical sites.

Discussion

Unlike existing clinical trials and observational studies of testosterone therapy, RHYME is the first multi-national disease registry for assessing prostate health and other clinically relevant outcomes in association with long-term use of testosterone therapy or untreated HG. The multi-national, broad patient population, non-interventional design and precise sample size calculations allow the registry to address the key prostate health outcomes in RHYME (proportion of positive prostate biopsy, changes in BPH symptoms), in addition to providing adequate sample size for assessing potential benefits of treatment.

Unlike other recent observational cohorts of hypogonadal subjects [15,29] RHYME is not restricted to one treatment product, but rather offers a robust disease registry, developed according to state-of-the-art registry guidelines. Of particular interest, RHYME will provide unique outcomes data regarding the long-term safety of different forms of testosterone therapy, including their effects on prostate health and cardiovascular outcomes, in middle-aged and older men at risk for developing prostate cancer or benign prostatic disease. RHYME will also serve as a rich source of data for the investigation of other relevant outcomes including changes in PSA, increases in voiding symptoms, urogenital complications and surgical interventions, cardiovascular events and other non-prostate cancers such as breast cancer. The registry will also provide outcomes data on other key study endpoints, including physical and mental health, sexual function and quality of life in a large group of hypogonadal men. These data will extend previous findings on the association between testosterone and mortality [30].

Several methodological issues were considered in the planning and design of RHYME. First, the potential role of ascertainment bias in monitoring prostate health outcomes was carefully considered. Ascertainment bias may present a risk for an increased prostate biopsy rate amongst patients on testosterone therapy, by virtue of the fact that testosterone treatment increases PSA and the safety concerns regarding testosterone therapy. One strategy to protect against this risk would be to perform “end of study” prostate biopsies in all enrolled patients, as done in the Prostate Cancer Prevention Trial [25]. Such procedures are not tenable or appropriate for an HG registry, as biopsies should only be performed as medically necessary in the standard of care for HG. As an alternative approach to mitigating this risk, we chose as our primary endpoint the rate of positive prostate biopsy. In doing so, the influence of factors known to increase the probability of biopsy and subsequent cancer detection is reduced. Other potential sources of bias were addressed by means of: (1) inclusion of a central laboratory for monitoring testosterone and PSA levels throughout; (2) independent adjudication of all biopsy reports and clinical data relevant to the primary and secondary endpoints and (3) independent scientific oversight by a high-level advisory committee. A plan is under consideration to compare registry prostate cancer outcomes with results from a matched sample of men in the European Randomized Study of Screening for Prostate Cancer [28,31]. Available data will allow for the control of factors that might increase the probability of prostate cancer detection (i.e. age, family history, geographic location, prior prostate biopsy, frequency of PSA determination and duration of observation).

Careful consideration was also given to ensuring that there is adequate statistical power to detect the primary outcomes of interest. Because the biopsy rate will vary according to participant age, no more than 150 patients <50 years old will be enrolled into the registry (15% of the overall registry). This restriction will allow us to achieve the desired/target prostate biopsy rate of 3–6%.

In summary, a large, multi-national, innovative disease registry for male HG has been designed and implemented. RHYME will provide well-characterized observational data for guiding clinical decision-making in this increasingly important area of men’s health. This registry is designed to address fundamental prostate health outcomes, in addition to symptom improvements, sexual function, mood and quality of life. The registry will provide novel data on the overall risks and benefits of testosterone treatment for male HG.

Declaration of interest

HMB, FHS, FSS, JFM and JDF have no conflicts of interest that could be perceived as prejudicing the impartiality of the research reported.

ABA has consulted with Lilly USA LLC (Indianapolis, IN) and has received research support from GlaxoSmithKline (Research Triangle Park, NC) and Abbott Laboratories (Abbott Park, IL).

RCR receives research support from and is a paid consultant for Lilly USA LLC (Indianapolis, IN), Auxilium Pharmaceuticals, Inc. (Malvern, PA) and Ferring Pharmaceuticals Inc. (Saint-Prex, Switzerland).

FCWW has consulted with Eli Lilly (Indianapolis, IN), Galapagos (Mechelen, Belgium), Ligand Pharmaceuticals Inc (San Diego, CA) and Novartis Pharmaceuticals (Basel, Switzerland) and has received lecture fees from Bayer-Schering Pharma and Eli Lilly.

CGR is a member of the RHYME PHCEC.

Acknowledgements

Funding support from Bayer Pharma AG is gratefully acknowledged. The participation and many contributions of the RHYME site investigators and clinical coordinators are also gratefully acknowledged.