Vitamin D Supplementation and Male Infertility: The CBG-study a Randomized Clinical Trial

Overview

Today, it is evident that vitamin D (VD) has more widespread effects than the classical actions related to bone mineralization and calcium homeostasis1. VD deficiency results in impaired reproductive performance in various species of animals, and recently the investigators have shown that the VD receptor (VDR), activating (CYP2R1, CYP27A1, CYP27B1) and inactivating (CYP24A1) enzymes are expressed in the human testis, epididymis, seminal vesicle, prostate and spermatozoa. Our following functional studies showed that VD increases intracellular calcium in mature spermatozoa, and hence may be important not only for spermatogenesis but also for sperm maturation. A new, and yet unpublished cross sectional study of 300 young healthy Danish men showed that men with lower levels of serum VD have significantly lower number of normally developed and motile spermatozoa. Hitherto, most cases of male infertility have been classified as "idiopathic", and infertile couples have been referred to symptomatic treatment at infertility clinics. These fertility treatments are often physically demanding for the female partner as well as expensive for the health care system. Any treatment that might improve semen quality of involuntary infertile men would be beneficial both for the infertile couples and the society in general. Our findings that VD may play a role for human semen quality have not yet been tested clinically. However, if VD supplementation proves efficient this opens for the first time for a causal, safe and cheap treatment of at least some cases of "idiopathic" impaired semen quality. The investigators believe our new human data supported by the results from the VD deficient and VDR KO animal studies and the high proportion of VD deficient Danish men provide sufficient evidence to initiate a randomized clinical trial of VD supplementation to infertile men. Infertile men have also have unfavorable altered levels of sex hormones and higher mortality than fertile men. Since VD deficiency is associated with increased mortality, regulation of aromatase, immune system, bone metabolism, glucose metabolism, cardiovascular system etc. our suggested clinical trial may also be able to evaluate several secondary endpoints in addition to the potential effect on semen quality.

Background In 2008, approximately 8% of all newborn children were conceived by assisted reproduction. Intracytoplasmatic sperm injection (ICSI) accounts for \~50% of all treatments and is mainly used in cases of severely reduced semen quality. This is in many cases most likely caused by prenatal factors adversely affecting the developing testicles, and it is therefore unlikely that any treatment of adult men will be able to completely normalize their semen quality. However, less would also be clinically relevant. If reduced semen quality could be improved prior to any assisted reproduction a less invasive treatment would be needed. Eg. classical in vitro fertilization (IVF) rather than ICSI, simple intrauterine insemination (IUI) rather than IVF or natural conception rather than IUI. Several endocrine factors have been implicated in sperm production and maturation, but little is known about the potential role of VD. VD is a key regulator of calcium homeostasis and bone mineralization, although expression of the vitamin D receptor (VDR) in various tissues has been related to several diverse actions. VD affects reproduction in several animal species, convincingly shown in rodents, where VD deficiency in male rats resulted in reduced sperm counts, and female rats inseminated with semen from VD deficient male rats had lower fertility rates. The impaired reproductive performance is reversible and can be corrected either by supplying VD or by normalizing calcium levels. Supported by VDR knockout mice, which showed decreased sperm counts, reduced sperm motility and histological abnormalities of the testis, which unlike the VD deficient male rats only partly can be restored by calcium supplements. The investigators have recently shown expression of VDR and the VD metabolizing enzymes in the human testis, ejaculatory tract and in mature spermatozoa. The investigators subsequently showed that VD in physiological concentrations increased intracellular calcium in spermatozoa. VD acts through a rapid non-genomic response and the VD induced increase in calcium may be crucial for the spermatozoa, because VD induced sperm motility and the acrosome reaction. Expression analysis of spermatozoa from fertile and infertile men showed that men with impaired semen quality have fewer that can metabolize VD than normal men(p \< 0.0005) making presence of one of the proteins a potential marker of semen quality. Furthermore, the investigators have just completed a cross sectional study of 300 young men from the general population and found that low serum VD was associated with reduced sperm motility and morphology. SETTING, SCIENTIFIC PLAN AND RECRUITMENT Participants will be included among men referred to the Department of Growth and Reproduction (dept. of GR), Rigshospitalet (RH) for evaluation of male infertility DESIGN This is a prospective, double blinded, two-arm randomized controlled trial Group of intervention: Each man will receive 300,000 IU (7500 ug) cholecalciferol (D3) orally once after blood- and semen sampling and performed DXA scan. Thereafter they will receive VD tablets of 1,400 IU (35 ug) + 500 mg calcium daily for 3 months. Telephone control after 4 weeks and at 3 months a clinical control and blood sampling will be performed, followed by continued daily intake of 1400 IU VD + 500 mg calcium for 2 additional months. At end of treatment at five months after inclusion the men deliver two semen samples, have a blood sample drawn and a DXA scan performed. Group receiving placebo: Will follow the same chain of events, although VD will be replaced with placebo. PARTICIPANTS Screening of \~700-800 infertile men will be performed. The investigators assume that \~340 men will be excluded and \~100 men do not wish to participate. Thus, 300 men will be included in the study and half will be randomized to active VD treatment. The investigators expect a small drop out (\< 20) because of high motivation and no adverse effects. ANALYSIS AND INTERVENTION Reproductive hormones and growth factors will be analyzed at dept. of GR, Rigshospitalet. An aliquot of the first blood sample will be analyzed before inclusion to the study and another kept to be analyzed at the same time as the second blood sample at the end of the trial to overcome the interassay variation. Other serum analyses will be analyzed at department of clinical biochemistry, Rigshospitalet. Semen analyses and DXA scans will be performed at dept. of GR. SAMPLE SIZE CALCULATION AND STATISTICS In our association study with 300 participants the investigators are able to detect significant dose response relationship between VD status and sperm motility and morphology, and found significant differences between men with low and high VD levels. Based on these results, the investigators assume to be able to increase serum levels of VD with 50 nM with our setup. Including 150 men in each group will enable us to detect changes in sperm motility and morphology of 15% and a 25% change in Inhibin B levels. Intention to treat principles will be used and co-variate analyzes will be used to analyze the potential effect of VD. BIOSTATISTICAL ANALYSIS All the analyses will be performed according to Good Clinical Practice guidelines and the primary analyses in the intention-to-treat population, which included all patients who underwent randomization and received the first dose of medicine on day 1. We will analyze the data in 3 ways. The primary analysis will proceed according to the randomized vitamin D and calcium dose group assignment. The secondary analysis will be based on stratifying the men according to subgroup analyses in relation to the predefined primary and secondary endpoints, while the last analysis will try to elucidate any dose-response relationship. An author (JHP) who was the statistician for the study and unaware of the study-group assignments will perform the primary analysis. Data analysis and quality The primary end points for this protocol will be changes in semen quality especially sperm motility followed by sperm concentration, morphology and semen volume. Multiple secondary endpoints exist but for the initial investigation focus will be on changes of the following secondary endpoints: Sperm DFI, pregnancy rate, Inhibin B, vitamin D and calcium homeostasis. Subjects who terminate participation after visit at 90 days but before visit day 150 will be included for data analysis up to day 90. Men that only deliver one semen sample or have missing data at any visit will still be included in the analysis. Men with fever up to 3 months prior to semen analysis will be considered as a potential confounder. Men that do not meet the criteria in the protocol will be excluded from the analysis. Those values will then be carried forward for analyses. A significance level of 5% is used. For the primary analyses Bonferronu-Holm p-value correction is calculated additionally. For the secondary analysis no multiple test correction are used. Instead results are discussed in view of the multiple testing situations. 1. Analyses between placebo versus active substance Between group analyses placebo versus vitamin D + calcium: The first step will be to compare the changes in primary outcomes across the two groups placebo versus vitamin D + calcium. This analysis will show if there is a significant difference between groups. For outcomes measured repeatedly, this will entail comparing the estimated slopes, or rates of change, of each outcome between the groups. Mixed models allow for the correlation between the repeated observations baseline-day 1-day 90- day 150 from each man to be suitably incorporated into parameter estimation. For all endpoints measured at baseline and day 150, paired t-tests will be used to assess there is a significant difference between the groups and determine whether the mean change within each group differs significantly from zero. In both cases, data will be transformed as necessary to meet model assumptions. Afterwards, the same analysis will be conducted by using multiple regression with relevant confounders such as season, BMI, smoking, duration of abstinence, time from ejaculation to motility assessment, fever etc. to see if this changes the results For outcomes measured that cannot be compared with t-test or other parametric tests at day 1, day 90 and day 150, groups will be compared using non-parametric tests such as Wilcoxon Mann-Whitney test. For Binary outcome the data will be compared between the two groups by means of conditional logistic-regression analysis with adjustment for relevant confounders (defined as being significantly p\<0.05 associated). . 2. Analyses after stratification into subgroups Subjects will be grouped according to their BMI, calcium, season, semen quality, bonefactors, BMD or vitamin D levels and the effect of placebo versus active substance will be evaluated at day 90 and day 150. The subgroup analyses will in accordance with normal clinical practice and stratification in appropriate groups according to the clinical (25OHD \< 25 nmol/l, BMI \<25, 25-30, \>30 etc.), tertiles/quartiles/quintiles or highest/lowest versus remaining at baseline. 3. Data analysis for comparison of changes in vitamin D and calcium We will also compare the dose-response relationships between changes in vitamin D and calcium and primary and secondary endpoints between placebo and active substance group and in all men. We will determine delta values for circulating vitamin D progenitors, calcium ion, total calcium and albumin corrected calcium from start to 90 days and 150 days and compare the difference with the observed changes in primary and secondary endpoints. We anticipate that there is no difference in shapes of the dose-response curve between the different endpoints i.e. they will be parallel and linear at least after transformation of the data. Afterwards, the same analysis will be conducted by using relevant confounders such as season, BMI, smoking, duration of abstinence, time from ejaculation to motility assessment etc. to see if this changes the results. SCREENING AND TIME COURSE Men, who are investigated at dept. of GR due to infertility will be screened for eligibility to the study, and those who meet the criteria for participation will be informed, and if they consent allocated to active treatment with VD or placebo. Allocation will be done by minimization using minim to avoid unbalanced grouping due to randomization failure. Following variables will be balanced: Sperm concentration, BMI, serum inhibin-B, and VD level. ETHICS AND SIDE EFFECTS All the patients will have fulfilled their investigation, before they are invited to the study. They will be informed of potential adverse effects, signs of intoxication and they can leave the trial at any point without any consequences. The trial will run in accordance with "good clinical practice". VD treatment gives virtually no side effects and the risk of intoxication is almost nonexisting in the suggested setting, with relatively low doses of VD and close monitoring of VD and calcium status. The participants will be exposed to oral VC/placebo, to 2 DXA scans, have 3 extra blood samples drawn and deliver 2 extra semen samples. All participants will be informed and counseling according to their VD status. PUBLICATION OF RESULTS All results, positive or negative will be submitted to peer reviewed scientific journals. Data will successively be obtained and transferred to a statistical database. PRACTICAL ISSUES Dept. of GR investigates annually more than 500 infertile men. Participants will be included from that group. The department have the clinical expertise, experience and capacity to perform all the investigations, except for the measurements of some of the blood samples, which will be analyzed by department of clinical biochemistry. VD and placebo oral suspensions are purchased from Glostrup apotek, while VD and placebo tablets are purchased from Ferrosan, MD Martin. The investigators are collaborating with associate professor in biostatistics J.H. Petersen, University of Copenhagen.