COMPENSATED HYPOGONADISM IN MEN WITH SICKLE CELL DISEASE

28 29

Sickle cell disease (SCD) is associated with the development of 30 hypogonadism, but there is still controversy regarding its etiology and clinical 31 implications. Objective: To evaluate the prevalence of hypogonadism in a population 32 of men with SCD and characterize its etiology. Methods: We performed a cross-33 sectional study of 34 men with SCD aged > 18 years. Sociodemographic and clinical 34 data, including anthropometric measurements (weight, height, and BMI), were 35 obtained. Early morning blood samples were collected and total testosterone (TT), 36 free testosterone (FT), luteinizing hormone (LH), follicle-stimulating hormone (FSH), 37 a complete blood count, and hemoglobin electrophoresis were measured. Results: 38 Median age was 33 [26-41] years, and SS genotype was the most frequent (73.5%). 39 The prevalence of eugonadism, compensated, and secondary hypogonadism was 40 67.5%, 26.4%, and 5.88%, respectively. No men with primary hypogonadism were 41 identified in our sample. Those with compensated hypogonadism had also higher 42 FSH levels than individuals with eugonadism; p < 0.001). Conclusion: In our study 43 population of men with SCD a high prevalence of compensated hypogonadism was 44 identified, which is a controversial and distinct clinical entity that warrants monitoring 45 and further research. 46 47

sickle cell disease, compensated hypogonadism, testosterone 48 49 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 4

50 51 Sickle cell disease (SCD) is a relatively common genetic disease and 52 comprises a group of disorders characterized by the presence of at least one 53 hemoglobin S (1). The sickle gene mutation is common in sub ‐ Saharan Africa and 54 other parts of the world and it is estimated that more than 300 000 children are born 55 each year with SCD, with millions of people currently affected across the globe (2,3), 56 about two‐ thirds of them in Africa (ref). SCD has been recognized as a public health 57 issue and a neglected problem by several key agencies, including the United Nations 58 (UN) and the World Health Organization (WHO)(4). The chronic morbidity associated 59 with SCD may lead to an increased socioeconomic burden and requires long-term 60 quality of care (5) SCD is commonly associated with the development of 61 hypogonadism, but there is controversy regarding its etiology, mechanisms, and 62 clinical implications (5). 63 Hypogonadism is characterized by impaired testicular function, which may 64 affect spermatogenesis and/or testosterone synthesis. Usually the diagnosis of 65 hypogonadism requires identification of low serum testosterone (T) levels. Individuals 66 with low T levels may be asymptomatic or present with a worse metabolic status, 67 reduced energy, diminished physical performance, fatigue, depression, reduced 68 motivation, poor concentration, infertility, reduced sex drive, erectile dysfunction and 69 increased rates of all-cause mortality (6–8). 70 Although the development of androgen deficiency is currently considered 71 multifactorial, male hypogonadism has been classically classified as 72 hypergonadotropic (primary) or hypogonadotropic (secondary) according to its 73 etiology (9). More recently a new clinical variant of hypogonadism defined as defined 74 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 5 as compensated hypogonadism has been proposed, which is characterized by 75 normal T and elevated LH levels. These authors suggested that compensated 76 hypogonadism represents a distinct clinical state, which warrants monitoring despite 77 its unknown etiology and consequences (8). Compensated hypogonadism has also 78 been reported in 3% of patients with male infertility, with similar outcomes of those 79 with primary hypogonadism (9). 80 Multiple theories have been proposed in attempts to explain the 81 hypogonadism related to SCD. Zinc deficiency, socioeconomic factors, constitutional 82 variables, and repetitive vaso-occlusive episodes in the testes and pituitary, are 83 possible explanations though the definite cause remains unknown. Vaso-occlusion is 84 the most widely accepted theory as recurrent microinfarctions in patients with SCD 85 are commonly found in other organs and systems (10). Few studies have been 86 designed to evaluate hormonal abnormalities in men with SCD and a more precise 87 understanding is yet to be accomplished (11). 88 As studies evaluating detailed hormonal profiles in men with SCD are lacking, 89 the aim of this study was to estimate the prevalence of hypogonadism in men with 90 SCD and to characterize its etiology using a more comprehensive classification. 91 92 93 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 6

94 95 Study design and population 96 97 This was a cross-sectional study involving 34 men with SCD aged 18 or older 98 who were followed up at a local SCD referral center between January and December 99 2019. Men with a history of cryptorchidism, testicular tumors, testicular or inguinal 100 surgery, or with acute onset of testicular at the time of the interview were excluded. A 101 structured questionnaire, including sociodemographic (age, sex, self-reported 102 race/color) and clinical variables (anthropometric measurements, type of 103 hemoglobinopathy, therapy with hydroxyurea or NSAIDs, and the occurrence of 104 priapism), was applied to participants. Anthropometric data (weight and height) were 105 measured and used to calculate the BMI as weight in kilograms divided by height in 106 meters squared (kg/m²). 107 108 Hormonal evaluation 109 Blood samples were collected between 7 and 9 a.m. for the determination of 110 total testosterone (TT), free testosterone (FT), LH, and FSH levels, as well as a 111 complete blood count and hemoglobin electrophoresis. The methods used for the 112 laboratory tests included: hydrodynamic focusing, flow cytometry, SLS-hemoglobin, 113 and Giemsa microscopy for the complete blood count; electrochemiluminescence 114 assays (Atellica IM® analyzer, Siemens Healthcare Diagnostics Inc., Tarrytown, NY, 115 USA) for the determination of T, LH, and FSH levels; an equation involving TT, sex 116 hormone-binding globulin (SHBG), and the association constant of albumin for T, 117 assuming a fixed albumin concentration of 4.3 g/dL, for the calculation of FT(12); and 118 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 7 high-performance liquid chromatography (HPLC) and capillary electrophoresis 119 (Capillarys Hemoglobine) for the quantitation of hemoglobin fractions. 120 Subjects were classified into four groups according to T and LH levels(13): 121 men were considered to have eugonadism if T ≥ 300 ng/dL and LH ≤ 9.4 mUI/mL; 122 primary hypogonadism was defined as T 9.4 mUI/mL; 123 secondary hypogonadism as T 9.4 mUI/mL. FSH levels above 7.8 ng/dL 125 were considered elevated (14). 126 All subjects were verbally and individually approached and provided informed 127 consent. This study was approved by our Research Ethics 128 129 Statistical analysis 130 131 Quantitative variables were presented as medians and interquartile ranges, 132 while nominal variables were expressed as absolute values, percentages, or 133 fractions. The Mann-Whitney U test was used to compare continuous variables, while 134 the Fisher’s test was used to compare categorical variables. A p < 0.05 was 135 considered statistically significant, and 95% confidence intervals were presented as a 136 measure of precision. GraphPad Prism, version 8.0.3, San Diego-CA, USA, was 137 used for data analysis. 138 139 140 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 8

141 142 Our sociodemographic data are detailed in Table 1. We assessed 34 men with 143 a median age of 33 years [26-41], most of whom had an SS genotype (73.5%) and 144 were black or brown (94.1%). Five (14.7%) of them were on continuous hydroxyurea 145 therapy and all had been medicated with NSAIDs, with ibuprofen being the most 146 commonly used drug. 147 Fifty percent of patients had LH levels of 5.92 mUI/mL [4.36 – 9.42]. Median 148 FSH levels were 6.01 mUI/mL [4.08 – 8.96], clinical data are detailed in Table 2. In 149 our sample, no men were diagnosed with primary hypogonadism, whereas 23 150 (67.6%) were classified as eugonadal and 2 (5.8%) as having secondary 151 hypogonadism. In addition, compensated hypogonadism was identified in 9 men 152 (26.4%) (Figure 1). 153 We found no differences regarding SCD genotype, anthropometric 154 measurements, or disease severity between eugonadal men and those with 155 compensated hypogonadism. Median FSH levels among men with compensated 156 hypogonadism were significantly higher than among eugonadal men. The proportion 157 of men with FSH levels above 7.8 mUI/mL was significantly higher among those with 158 compensated hypogonadism (p < 0.0001) (Table 2). 159 160

161 162 Gonadal dysfunction, disturbances of the hypothalamic-pituitary-testicular axis, 163 and the etiology of hypogonadism in patients with SCD are controversial issues, and 164 there are still many uncertainties about the pathophysiology of these conditions and 165 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 9 their clinical significance(11). We investigated the prevalence of different types of 166 hypogonadism in men with SCD according to the classification described by Tajar et 167 al (2010) in the EMAS. The prevalence of compensated hypogonadism in our sample 168 of men with SCD was 26.4%, a figure considerably higher than those reported in 169 similar studies, 9.5%(13) and 3%(14), respectively. 170 Tajar at al. (2010) were the first to explore the concept of compensated 171 hypogonadism in a study with 3369 community-dwelling men aged 40-79 years from 172 eight European centers. They also introduced the concept that different 173 hypogonadism categories may present with different clinical features. According to 174 this study, sexual complaints were more frequently reported in cases of primary 175 hypogonadism, whereas isolated physical symptoms such as inability to walk long 176 distances or perform brisk physical activity were more common in compensated 177 hypogonadism (8). 178 A retrospective study involving 4173 men with sexual dysfunction by Corona et 179 al., reported that 4.1% of individuals had compensated hypogonadism and were 180 more likely to present with both increased frequency of psychological symptoms, 181 such as anxiety, obsessive-compulsive symptoms, and depression; and increased 182 cardiovascular mortality in comparison to those with primary or secondary 183 hypogonadism. The authors hypothesized that compensated hypogonadism could be 184 a surrogate marker of an underlying disease, rather than a new clinical entity(15). 185 Compensated hypogonadism also appears to be relevant in the elderly. Ucak 186 et al. (2013) conducted a study with 250 men over 70 years of age with compensated 187 hypogonadism and found their T and LH levels to be independently associated with 188 worsening performance of activities of daily living, as well as deteriorating cognitive 189 function, nutritional status, and mood, when compared to healthy controls. The 190 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 10 authors concluded androgen and LH levels should be assessed in elderly men, and 191 those diagnosed with compensated hypogonadism should then undergo a physical 192 and neuropsychiatric evaluation (16). 193 The clinical significance of compensated hypogonadism is still poorly 194 understood, but this condition is known to be associated with both aging and an 195 increased frequency of physical symptoms related to testosterone deficiency, but not 196 with sexual complaints, as previously mentioned. These observations and its striking 197 prevalence of 9.5% as reported in the EMAS raise the question of whether this 198 subtype of hypogonadism warrants treatment. In, especially older men, may benefit 199 from the inclusion of LH levels in the initial screening for hypogonadism (17). 200 Studies about compensated hypogonadism and its etiology are controversial 201 and scarce in patients with SCD. Rhodes et al (2009)(18) did not found a statistically 202 significant difference in the testosterone levels of 19 boys with SCD when compared 203 to controls and so they were unable to detect primary hypogonadism. Özen et al. 204 (2013)(19) studied 50 Turkish children aged 4 to 18 years and found, among the 35 205 boys included in the sample, one with hypergonadotropic (primary) hypogonadism 206 and 3 with small testes and low testosterone, but with normal luteinizing hormone 207 (LH) levels, suggesting this condition can be either primary or secondary. Abbasi and 208 colleagues (1976)(20) analyzed hormone levels of 14 patients and found elevated 209 concentrations of LH and follicle-stimulating hormone (FSH), as well as low 210 testosterone (T), suggesting primary hypogonadism. Dada and Nduka (1980)(22) 211 demonstrated reduced LH, FSH, and testosterone levels in 19 men with SCD, 212 findings suggestive of hypothalamic axis dysfunction (secondary hypogonadism), 213 rather than gonadal failure (primary hypogonadism). Martins et al(23) studied some 214 aspects of compensated hypogonadism among 10 men with homozygous SCD. 215 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 11 Despite the limited number of patients, the authors referred to compensated 216 hypogonadism as a likely transient condition and considered it a potential state of 217 androgen resistance(23). 218 All subjects with compensated hypogonadism in our sample had FSH levels 219 above 7.8 mU/mL, a threshold previously reported as a predictor of impaired 220 spermatogenesis(24). It is likely that the exocrine compartment might be impaired in 221 while the endocrine compartment might still be enough to sustain appropriate T 222 production. A cross-sectional study with 786 Caucasian-European discussed the 223 concept of compensated hypogonadism among infertile subjects. The authors found 224 that this condition had a similar clinical characteristic to those with primary 225 hypogonadism, and both groups had the worst clinical outcomes in terms of impaired 226 fertility. While this classification was not designed for the setting of male infertility, it 227 could be useful in clinical practice to indicate impaired spermatogenesis(14). 228 It is important to highlight that all men in our sample had history of ibuprofen 229 intake for acute pain episodes. Previous published data indicate a strong association 230 between ibuprofen use and the elevation of LH levels, which might lead to 231 compensated hypogonadism. In a randomized controlled trial including 31 men aged 232 18-35 years who received 600 mg of ibuprofen twice daily for 6 weeks, the authors 233 showed ibuprofen use increased LH levels in 23% after 14 days and 33% after 44 234 days (p = 0.01). They also linked ibuprofen use with a reduction in anti-Müllerian 235 hormone levels and postulated this drug would affect steroidogenesis by inhibiting 236 the expression of related genes, thus resulting in HPT axis dysfunction (25). 237 However, this study evaluated solely the impact of short-term course of ibuprofen on 238 hormonal profile, and clinical significance of its effect in the long-term are probably 239 negligible. 240 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 12 Although our cohort had only 34 participants, it is one of the largest studies in 241 the scientific literature to evaluate hormonal profiles in men with SCD. We also 242 identified a high prevalence of an underreported and poorly understood condition, 243 which might be of clinical significance for monitoring and counseling patients. The 244 main limitation of this study is perhaps its cross-sectional design, as no definitive 245 explanation for this condition may be given. In addition, this sample came from a 246 specialized center for patients with SCD and may not represent the general 247 population of SCD. 248 249 250 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 13

251 252 A high prevalence of compensated hypogonadism was identified in our sample 253 of men with SCD, which may reflect early testicular injury and may progress to 254 permanent dysfunction. It seems to be more prevalent in young men with SCD than 255 in the general population of older men. Vaso-occlusive phenomena, testicular 256 microinfarctions, changes in the testicular microenvironment, long-term or recurrent 257 ibuprofen use, and genetic aspects might be associated with such findings. Although 258 this clinical entity is not completely understood, continuous monitoring might be 259 useful in preventing or anticipating additional clinical deterioration. 260 261 Disclosure statement 262 The authors report no conflicts of interest. 263 264 Funding 265 This study received no external funding. 266 267 Academic affiliation 268 This paper is part of the master’s dissertation of Anna Paloma Martins Rocha Ribeiro, 269 Postgraduate Program in Public Health, State University of Feira de Santana. 270 271 272 273 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 14

274 275 1. Bunn HF. Pathogenesis and treatment of sickle cell disease. N Engl J Med. 276 1997;337(9):762–9. 277 2. Wastnedge E, Patel S, Goh MY, Rudan I. The global burden of sickle cell 278 disease in children under five years of age: a systematic review and meta-279 analysis. J Glob Health. 2018;8(2):1–9. 280 3. Piel FB, Hay SI, Gupta S, Weatherall DJ, Williams TN. Global Burden of Sickle 281 Cell Anaemia in Children under Five, 2010 – 2050: Modelling Based on 282 Demographics, Excess Mortality, and Interventions. PLoS One. 2013;10(7):1–283 14. 284 4. Mburu J, Odame I. Sickle cell disease: reducing the global disease burden. Int 285 J Lab Hematol. 2019;41(February):82–8. 286 5. Pleasants S. A moving target: migration has led to an increase in the 287 occurrence of sickle-cell disease in countries with previously low incidence of 288 the disorder. Nature. 2014;515(11):1–3. 289 6. Mulhall JP, Trost LW, Brannigan RE, Kurtz EG, Bruce J, Chiles KA, et al. 290 Evaluation and Management of Testosterone Deficiency: AUA Guideline. J 291 Urol [Internet]. 2018; Available from: https://doi.org/10.1016/j.juro.2018.03.115 292 7. Livingston M, Kalansooriya A, Hartland AJ, Ramachandran S, Heald A. Serum 293 testosterone levels in male hypogonadism/i3 : Why and when to check — A 294 review. Int J Clin Pract. 2017;(August):1–9. 295 8. Peterson MD, Belakovskiy A, Mcgrath R, Yarrow JF. Testosterone Deficiency , 296 Weakness, and Multimorbidity in Men. Sci Rep [Internet]. 2018;8(5897):1–9. 297 Available from: http://dx.doi.org/10.1038/s41598-018-24347-6 298 9. Jungwirth A, Diemer T, Kopa Z, Krausz C, Minhas S, Tournaye H. EAU 299 Guidelines on Male Infertility. Vol. 1, European Association of Urology. 2018. 300 47 p. 301 10. Li M, Fogarty J, Whitney KD, Stone P. Repeated testicular infarction in a 302 patient with sickle cell disease: A possible mechanism for testicular failure. 303 Urology. 2003;62(3):551. 304 11. Huang AW, Muneyyirci-Delale O. Reproductive endocrine issues in men with 305 sickle cell anemia. Andrology. 2017;5(4):679–90. 306 12. Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods 307 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 15 for the estimation of free testosterone in serum. J Clin Endocrinol Metab. 308 1999;84(10):3666–72. 309 13. Tajar A, Forti G, O’Neill TW, Lee DM, Silman AJ, Finn JD, et al. Characteristics 310 of secondary, primary, and compensated hypogonadism in aging men: 311 Evidence from the European male ageing study. J Clin Endocrinol Metab. 312 2010;95(4):1810–8. 313 14. Ventimiglia E, Ippolito S, Capogrosso P, Pederzoli F, Cazzaniga W, Boeri L, et 314 al. Primary, secondary and compensated hypogonadism: a novel risk 315 stratification for infertile men. Andrology. 2017;5(3):505–10. 316 15. Corona G, Maseroli E, Rastrelli G, Sforza A, Forti G, Mannucci E, et al. 317 Characteristics of compensated hypogonadism in patients with sexual 318 dysfunction. J Sex Med. 2014;11(7):1823–34. 319 16. Ucak S, Basat O, Karatemiz G. Functional and nutritional state in elderly men 320 with compensated hypogonadism. J Am Med Dir Assoc. 2013;14(6):433–6. 321 17. Lee JA, Kuchakulla M, Arora H, Kulandavelu S, Gonzalez E, Masterson TA, et 322 al. Age induced nitroso-redox imbalance leads to subclinical hypogonadism in 323 Male mice. Front Endocrinol (Lausanne). 2019;10(MAR):1–6. 324 18. Rhodes M, Akohoue SA, Shankar SM, Fleming I, An A, Yu C, et al. Growth 325 Patterns in Children with Sickle Cell Anemia during Puberty. Pediatr Blood 326 Cancer. 2009;53(4):635–41. 327 19. Özen S, Ünal S, Erçetin N, Ta ş delen B. Frequency and Risk Factors of 328 Endocrine Complications in Turkish Children and Adolescents with Sickle Cell 329 Anemia. Turk J Hematol. 2013;30:25–31. 330 20. Abbasi AA, Prasad AS, Ortega J, Congco E, Oberleas D. Gonadal Function 331 Abnormalities in Sickle Cell Anemia Studies in Adult Male Patients. Ann Intern 332 Med. 1976;85:601–5. 333 21. Dada OA, Nduka EU. Endocrine function and haemoglobinopathies: Relation 334 between the sickle cell gene and circulating plasma levels of testosterone, 335 luteinising hormone (LH) and follicle stimulating hormone (FSH) in adult males. 336 Clin Chim Acta. 1980;105(2):269–73. 337 22. Dada O, Nduka E. Endocrine function and haemoglobinopathies: relation 338 between the sickle cell gene and circulating plasma levels of testosterone, 339 luteinising hormone (LH) and follicle stimulating hormone(FSH) in adult males. 340 Clin Chim Acta. 1980;105:269–73. 341 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 16 23. Martins PRJ, De Vito FB, Resende GAD, Kerbauy J, Pereira G de A, Moraes-342 Souza H, et al. Male sickle cell patients, compensated transpubertal 343 hypogonadism and normal final growth. Clin Endocrinol (Oxf). 2019;91(5):676–344 82. 345 24. Barbotin AL, Ballot C, Sigala J, Ramdane N, Duhamel A, Marcelli F, et al. The 346 serum inhibin B concentration and reference ranges in normozoospermia. Eur 347 J Endocrinol. 2015;172(6):669–76. 348 25. Kristensen DM, Desdoits-Lethimonier C, Mackey AL, Dalgaard MD, De Masi F, 349 Munkbøl CH, et al. Ibuprofen alters human testicular physiology to produce a 350 state of compensated hypogonadism. Proc Natl Acad Sci U S A. 351 2018;115(4):E715–24. 352 353 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 17 354 Variables n (34) % SCD genotype Homozygous HbSS 25 73.5 Heterozygous SC/SB-thal 9 26.5 Self-reported race/color Black 21 61.8 Brown 11 32.4 Yellow 1 2.9 Indigenous 1 2.9 Hydroxyurea therapy No 29 85.3 Yes 5 14.7 NSAIDs use No 0 0 Yes 34 100 Education level Primary education 7 20.6 Secondary education 20 58.8 Higher education 7 20.6 Occupation Activea 6 17.7 Inactiveb 28 82.3 History of priapism Yes 18 52.9 Table 1 – Sociodemographic and clinical data of men with SCD. 355 356 Table 2 – Anthropometric and laboratory data of men with SCD 357 Variables Median [p25-p75]* Weight (kg) 61.5 [56.3 – 67.3] Height (m) 1.7 [1.64 – 1.75] BMI (kg/m2) 21.6 [19.2 – 23.8] Hemoglobin (g/dL) 9.75 [8.05 – 12.0] Total testosterone (ng/dL) 582.9 [428.8 – 685.9] Free testosterone (ng/dL) 9.85 [8.07 – 11.37] LH (mUI/mL) 5.92 [4.36 – 9.42] FSH (mUI/mL) 6.01 [4.08 – 8.96] *Data are expressed as medians and interquartile ranges (25th and 75th percentiles) 358 359 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 18 Table 3 – Clinical, anthropometric, and laboratory data of eugonadal men with SCD in 360 comparison to those with compensated hypogonadism. 361 Variables Eugonadism (n = 23) Compensated hypogonadism (n = 9) p value Homozygous HbSS (%) 16 (69%) 7 (77%) 0.943 Age (years) 34 [27 – 41] 33 [22-41] 0.612 Hydroxyurea therapy (%) 6 (26%) 0 (0%) 0.153 History of priapism (%) 4 (44%) 12 (48%) 0.162 BMI (kg/m²) 21.6 [18.9-23.9] 21.7 [19.7-22.9] 0.991 Hemoglobin (g/dL) 9.8 [8.5-12.1] 9.3 [7.6-11.4] 0.331 FSH (mUI/mL) 5.37 [3.7-6.3] 9.5 [9.0-23.8] 7,8 (%) 5 (21.7%) 9 (100%) < 0.0001* *Statistically significant differences 362 BMI: body mass index; FSH: follicle-stimulating hormone; LH: luteinizing hormone. 363 364 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint 0 300 600 900 15 25 Testosterone (ng/dl) LH (mUI/ml)9.4 9(26%) 2(6%) 23(68%) All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted April 28, 2020. ; https://doi.org/10.1101/2020.04.21.20074666doi: medRxiv preprint