Testicular function after non‐cytotoxic and immunotherapy drug treatment

The effects of novel non‐cytotoxic and immunotherapy drugs for cancer treatment on human testicular function have not been studied systematically.


INTRODUCTION
Over decades, prognosis for long-term remission or cure of cancers has improved due to therapeutic advances in conventional treatment based on chemotherapy and/or radiotherapy. 1For the increasing numbers of long-term survivors, quality of life is a pivotal concern, with reproductive function and fertility of prominent concern for younger men.However, both conventional treatment modalities are gonadotoxic creating severe, and often irreversible, spermatogenic damage, leading to male infertility and/or hypogonadism. 2Gonadotoxic spermatogenic damage reflects the exquisite sensitivity of the intensely replicating germinal epithelium to cytotoxic drugs and irradiation, a sensitivity even greater than the often dose-limiting bone marrow toxicity of cancer treatments.[5][6] Hence, fertility preservation procedures aiming to preserve or protect reproductive potential arising from gonadotoxic therapy have developed as fertility insurance to minimize the impact of treatment on a patient's future fertility. 7,8The original, most widely available fertility preservation procedure is sperm cryostorage, a long standard clinical practice recommended by most international guidance documents on comprehensive cancer care, 9,10 and the forerunner of the new oncofertility discipline. 11[14] An important contemporary advance in cancer treatment has been the advent of non-cytotoxic regimens including immunotherapy using monoclonal antibody checkpoint inhibitors, 15 as well as small molecule enzyme inhibitors. 16Conventional cancer treatment aims to directly kill any replicating cells, albeit indiscriminately, leading to off-target effects damaging rapidly proliferating epithelia such as the germinal epithelium as well as the often dose-liming bone marrow toxicity and bystander damage to spermatogenesis.Such treatment regimens are provided in discrete treatment periods of limited duration, leaving residual gonadotoxic damage in its long-term wake.Administration identified no human studies of reproductive effects for any of these drugs. 18Furthermore, pre-clinical data were very limited, with nine drugs showing impaired male fertility and three showing impaired male and female fertility.As a result, it remains prudent for men to undertake sperm cryostorage prior to NCIT treatments.Nevertheless, as the impact of these newer NCIT drugs on male fertility remains largely unknown, we aimed to characterize the impact of novel NCIT drugs on male reproductive function in patients not previously treated with gonadotoxic drugs or irradiation (Figure 1).

Study design
This observational study had a mixed longitudinal and cross-sectional design with two parallel cohorts of men with cancer or non-cancer diseases who had planned, or had undergone, treatment with NCIT drugs.
Participants were recruited from two centers, the Andrology Department, Concord Hospital and the Royal Hospital for Women, both in Sydney, NSW, Australia.
Cohort 1 involved longitudinal investigation of men scheduled to be treated with NCIT drugs who had not been treated with cytotoxic drugs or radiotherapy.They were studied before and on long-term NCIT drugs at about 12 months after starting treatment.Participants were usually referred for sperm cryostorage as part of standard care.
Men who had stored spermatozoa before their treatment were already involved in routine follow-up to determine ongoing need for sperm cryostorage. 14  .The y axis is on a cube-root scale, which optimizes normalization of sperm distributions, and the x axis represents sequential visits before and after the start of NCIT treatments, with the start of treatment indicated as the 0 visit and uneven time intervals between visits.Prior to start of treatment, men were having spermatozoa cryostored, and the visits were usually within 1-2 weeks.After treatment, the men in Cohort 1 were scheduled to be studied at 12 months after start of treatment.However, the time on treatment for Cohort 2 was by definition longer and more variable, as they only entered the study after the start of NCIT treatment.For further details see text.

Study procedures
The inclusion criteria were (i) aged between 18 and 65 years, (ii) diag- At subsequent visits, the men again provided semen and blood samples.

Semen analyses and sperm cryostorage
Semen analysis was conducted according to the WHO Semen Analysis Manual 19 in a NATA-accredited Andrology laboratory, providing semen analysis and sperm cryostorage, and consistent with the journal's guidelines for semen analysis. 20In addition to routine semen analysis (abstinence interval, semen volume, sperm concentration and output, sperm motility and morphology), sperm DNA fragmentation was measured by Halosperm G2, an improved commercial version of the Halosperm sperm chromatin dispersion test. 21

Reproductive hormone assays
As far as possible, blood samples were obtained non-fasting in the mornings, but collections during sperm cryostorage were opportunistic and considered at random times.Serum luteinizing hormone (LH), follicle-stimulating hormone (FSH), testosterone, and sex hormonebinding globulin (SHBG) were measured as they were drawn in routine hospital pathology laboratory using commercial (Roche) immunoassays, subject to routine internal and external quality control programs.Sera were also stored frozen for analysis in a single batch by liquid chromatography-mass spectrometry (serum testosterone, dihydrotestosterone, dehydroepiandrosterone, estradiol, estrone, 3α androstanediol, and 3β androstanediol), 22 as well as serum inhibin B and AntiMullerian Hormone (AMH) by ELISA immunoassays (ANSH laboratories, Banksia Scientific) performed according to the manufacturer's instruction.

Data analysis
The primary study endpoints were the impact of NCIT drug treatment on sperm output and concentration in gonadotoxic-naïve men with

Patients (Table 1)
The

Sperm output and function
By linear mixed model for repeated measures, sperm output (p < 0.001) and concentration (p = 0.005) were significantly lower on treatment compared with pre-treatment when adjusted for semen volume (see also Table 2 and Figure 1).

Serum reproductive hormones
NCIT treatment was associated with increased serum FSH (6.5 vs. 2.6 IU/L, p = 0.005) and decreased serum Inhibin B (218 vs. 293 pg/mL, p = 0.020).Serum testosterone, LH, and SHBG did not show any difference according to NCIT treatment, with or without adjustment for all covariates (excluding sperm output or concentration, data not shown).

DISCUSSION
The present study demonstrates that among men not previously exposed to therapeutic gonadotoxicity, NCIT drug treatments reduce sperm output by about 50% after a median of over 30 months on treatment.The reduction in sperm production was associated with increased serum FSH and decreased serum inhibin B, expected effects of any drug having adverse impact on spermatogenesis. 24,25e man treated with both antibody and enzyme inhibitor drugs for cause was identified, but autoimmune orchitis triggered by checkpoint inhibitor treatment 26,27 was not excluded, although the lack of testicular pain and swelling made it unlikely.The changes in sperm production could not be attributed statistically to effects of antibody or enzyme inhibitor treatments alone, or their combination.The potential contribution from autoimmune hypophysitis 28 contributing to partial gonadotrophin deficiency, representing a potentially reversible mechanism for the reduction in sperm output, could not be excluded; however, none of the men in this study had clinical evidence of hypophysitis, which, while a well-recognized, is often temporary and/or reversible, complication of immunotherapy treatments, 29 remains rare with a prevalence of about 1.5%. 29,30The lack of significant change in serum testosterone, LH, and SHBG indicates that androgen deficiency is unlikely because of NCIT treatments.Overall, however, these effects on sperm production and its related hormones (FSH, Inhibin B) appear substantially less than expected for conventional gonadotoxic cancer treatments involving chemo-and/or radiotherapy.For men in whom NCIT drugs are therapeutically successful, it is likely they may continue to be used indefinitely, so further research into longer term effects is still needed.Conversely, despite the relatively mild testicular impact of NCIT drugs, if they are used in combined regimens with gonadotoxic therapies, the net effect on testicular function and fertility is likely to be more severe.
5][36] The median sperm output and concentration in men on NCIT treatment (89 million and 48 million/mL, respectively) are still well above lower 95% confidence limits (20 million and 9 million/mL, respectively) for unselected men in the community 37 or for young Australian men. 38 men treated with NCIT drugs.However, this is subject to the reservation that spermatogenic damage may occur on combination or more intense NCIT regimens, so that men embarking on NCIT drug treatments, when the ultimate treatment regimen(s) may not be predictable, should continue to be advised on both pre-treatment sperm cryostorage as well as adequate contraception, both options intended to ensure timely future paternity.
The present study extends previous reports on male reproductive function on NCIT drug treatments.A post-mortem study of men with metastatic melanoma reported that six of seven men treated with immunotherapy, as well as two of six men not treated with immunotherapy, had severe spermatogenic damage. 40The lack of statistically significant differences reflects both the small sample as well as prior knowledge that terminal illness and its treatments are major, non-specific determinants of post-mortem testicular damage. 41Consequently, post-mortem testicular damage may not reflect pre-mortem testicular function or sperm output.Checkpoint inhibitor treatment-related orchitis may also damage sperm output, 26,27 although we observed no history or evidence of orchitis in our study.A cross-sectional study of 25 men treated for a median 20 months with immune checkpoint inhibitors for melanoma reported that semen analysis was qualitatively normal in 18/22, including four of 22 on treatment, as well as one of five men having abnormal pre-treatment semen analysis. 33No quantitative data on sperm output or function nor any reproductive hormone analyses were reported.
More data are available for effects of imatinib on paternity 31,32 with limited congruent data on paternity data among men treated with second-generation tyrosine kinase inhibitors (nilotinib, dasatinib, bosutinib). 32The significant increase in serum FSH in this study is consistent with the moderate decline in sperm production, noting one man treated most intensively with both types of NCIT drugs, becoming azoospermic with a high serum FSH.Other than our finding of a significant increase in serum FSH, the present findings are consistent with a study of 48 Chinese men with CML in which serum testosterone, LH, FSH, and estradiol by immunoassay did not differ from 50 outpatient control or 10 infertile men 31 ; however, sperm output was not reported. 31Similarly, the inverse correlation between serum SHBG and BSA in this study is consistent with the known effects of larger body size in obesity with lower serum SHBG. 42trengths of this study were its longitudinal component with anal- Overall, we conclude that these non-cytotoxic and immunotherapy drugs mostly have little or no effects on testicular exocrine function and none on testicular endocrine function.Only in the most heavily treated individuals-also those with the most advanced underlying disease-the effects on sperm output may be due to the nonspecific effects of underlying disease and its severity rather than the non-cytotoxic and immunotherapy drug therapies.Further larger and longer studies including disease-specific measures of underlying disease severity would be valuable.

Currently, there are
limited clinical data on NCIT drug effects on human reproductive function.A comprehensive review of 32 new noncytotoxic oncology drugs marketed between 2014 and 2018 by the US Federal Drug Administration or the Australian Therapeutic Goods

F I G U R E 1
Plots of sperm output (million spermatozoa per ejaculate, left panel) and sperm concentration (million spermatozoa per milliliter ejaculate, right panel) of study participants who had multiple pre-treatment and on-treatment semen analyses due to their use of sperm cryostorage either coinciding with recruitment to the study for non-cytotoxic and immunotherapy (NCIT) treatment (Cohort 1) or having been completed before starting NCIT treatment (Cohort 2).The results for each man are plotted as connected symbols and lines with 15 men in Cohort 1 (filled circles, straight lines) and six men in Cohort 2 (open squares, dashed lines) nosis of cancer or non-cancer disease requiring treatment with NCIT drugs (immunotherapy checkpoint or enzyme inhibitors or analogous non-cytotoxic drugs), and (iii) provide written informed consent to participate in the research including undergoing additional tests and providing blood and semen samples.Exclusion criteria were (a) having received chemotherapy or pelvic radiotherapy prior to current treatment, and (b) major psychiatric disease or psychological condition that in the investigator's opinion, limited participant's understanding of, and compliance with, study requirements.After providing written informed consent, men completed a standardized medical and reproductive history, underwent physical examination (including testicular size), and provided blood and semen samples.At entry, men provided information on their reproductive, marital and fertility status, smoking, and alcohol consumption, as well as age, height, and weight.Their underlying diseases were classified as melanoma, hematological, or other, with treatments classified as antibody-based and/or enzyme inhibitor.
cancer.Secondary endpoint included semen volume, sperm motility and morphology, serum testosterone, LH, FSH, SHBG, and testicular volume.The impact of NCIT drugs on the primary outcomes (sperm output, concentration) was analyzed by mixed model analysis for repeated measures using an autoregressive covariance structure with Fisher scoring and Newton-Raphson maximization (NCSS 2022 Statistical Software, NCSS).This method fits an intercept for each individual with the slope of the regression based exclusively on the sperm parameters from that individual.This optimizes the integrated analysis of longitudinal and cross-sectional data with unequal numbers of samples and spacing of endpoints, and circumvents the bias or inefficiency of reducing the data to individuals or samples alone.Covariate adjustments were performed for continuous (age, height, weight, body mass index, body surface area) or categorical (type of NCIT drug [antibody, enzyme inhibitor], alcohol, smoking, fertility status) variables.Descriptive summary of variables was reported as mean and standard error of the mean (for gaussian data) or median with interquartile range (for non-gaussian data).Normality of data distribution was evaluated by the Shapiro-Wilks test together with the Box-Cox transformation to identify optimal power transformations to normalize nongaussian distributions prior to parametric analyses.Sperm concentration and output were right-skewed and cube-root transformed23 with serum testosterone, FSH, and DNA fragmentation log transformed and serum SHBG cube-root transformed prior to analysis.After analysis, transformed variables were backtransformed for ease of interpretation.Each semen analysis was classified as pre-treatment or on NCIT treatment.
ysis by a data modeling technique that focused on within-person time-course changes, notably before versus on treatment, that was robust to irregular sampling times and missing data.The relatively limited duration of observation on NCIT treatment (median 33.5 months) is long enough to observe gonadotoxic effects on spermatogenesis, which has a cycle time of 3 months.The main limitations of this study are its relatively small sample size and the lack of data on sexual function.Further prospective comprehensive studies of the systematic impact of non-cytotoxic drugs on male reproductive function are needed.
Baseline descriptive data of male participants.
18 months displayed severe reduction of sperm production, becoming azoospermic with highly elevated serum FSH and undetectable inhibin B consistent with severe spermatogenic damage.No specific TA B L E 1 Note: Data are expressed as mean ± standard error of the mean (SEM) and as median and interquartile range (IQR) [Q1-Q3].Abbreviations: AMH, AntiMullerian Hormone; DHEA, dehydroepiandrosterone; DHT, dihydrotestosterone; FSH, follicle-stimulating hormone; LH, luteinizing hormone; SHBD, sex hormone-binding globulin.a Measured by testosterone immunoassay as samples became available.b Measured batchwise at end of study by high-pressure liquid chromatography mass spectrometry.
Effects of non-cytotoxic and immunotherapy treatment.Descriptive data are tabulated as mean ± SEM if gaussian, or median and interquartile range (IQR) [Q1-Q3] if non-gaussian.p-Values from unpaired t-test or Mann-Whitney rank test (if non-gaussian) with Bold data to indicate statistically significant difference.Semen variables are adjusted by covariance analysis for semen volume as a surrogate for abstinence interval.
and no known defects in sperm function.These considerations support and extend previous findings on the high likelihood of paternity among TA B L E 2 Abbreviations: AMH, AntiMullerian Hormone; DHEA, dehydroepiandrosterone; DHT, dihydrotestosterone; FSH, follicle-stimulating hormone; LH, luteinizing hormone; SHBD, sex hormone-binding globulin.a Measured by testosterone immunoassay as samples became available.b Measured batchwise at end of study by high-pressure liquid chromatography mass spectrometry.