Sperm Cryopreservation Prior to Gonadotoxic Treatment

Experience of a Single Academic Centre Over 4 Decades

Nandini Shankara-Narayana; Irene Di Pierro; Carolyn Fennell; Lam P. Ly; Fay Bacha; Ljubica Vrga; Sasha Savkovic; Leo Turner; Veena Jayadev; Ann J. Conway; David J. Handelsman


Hum Reprod. 2019;34(5):795-803. 

In This Article


Anthropometric and Demographic Data

This study analysed data from 2717 men comprising 2085 men with cancer, 234 men with non-cancer medical disease and 398 controls who were healthy volunteers for sperm donation (Table I). Men seeking cryostorage (median 29 years, range 12.8–67 years) were similar in age to controls (median 30, range 18–58 years) with 136 (6.5%) under 18 years of age. There were small differences between some groups in height, weight, body mass index and body surface area (Table I). Most (73%) men seeking cryostorage had unknown or untested fertility status and nearly half (48%) had no regular female conjugal partner.

Residential postcodes were available for 1944 of the attendances with expected rates based on NSW Cancer Registry regional population data. Of them 1881 (96.8% observed versus 92.6% expected) were from Metropolitan Sydney and inner regional NSW, 57 (observed 2.9% versus 6.8% expected) were from outer regional NSW and 6 (observed 0.3% versus expected 0.6%) from remote NSW, suggesting only about half of the expected numbers of men from outer and remote areas compared with urban residents of New South Wales (NSW) attended cryostorage.

Over 40 years requests for cryostorage increased progressively more than 10-fold with a consistent distribution between haematological, testicular and other cancers (Figure 1) whereas there was a 3–5-fold contemporaneous increase in all cancers, haematological and testicular cancers in this the state cancer registry (Figure 2).

Figure 1.

Shows attendance for new sperm cryostorage by year from 1978 to 2017 together with the incidence of new cancers (all, haematological, testis) as shown in the state registry over the same period.

Figure 2.

Shows attendance for new sperm cryostorage by year from 1978 to 2017 for all requests as well as the breakdown according to the major cancer types (haematological, testis, other cancers) over the same period.

Feasibility of Cryostorage

Most men (89%) could collect sperm for storage. Failure to cryostore resulting from azoospermia or near-azoospermia (7%), inability to collect semen (4%) or to attend (1%), the latter usually due to being too sick or conflict with timing of urgent treatment. Most men cryostored sperm prior to start of gonadotoxic treatment but a few already received chemotherapy (3.5%), radiotherapy (1.6%) or both (0.7%). Most men provided three ejaculates (52%) creating a median of 23 straws per man. Most underage (<18 years) adolescents completed cryostorage although overall they were slightly less likely to complete cryostorage (80%) and produce three ejaculate (46%) (Table II).

Time to Disposal of Samples

For all banked samples, the overall median time in cryostorage was 8.5 years (Figure 3). By 5 years about a quarter (26%) had been removed and ultimately 62.6% of cryostored samples were disposed of. Of the removed samples, 36% were discarded due to recovery of fertility or sperm production (natural paternity or normal sperm output), 26% were discarded due to death and 7% were transferred for use with the remainder lost to follow-up (Figure 3) The median time to transfer for use was 2.5 years while median time for disposal due to death was 3.5 years, for return of fertility was 6.5 years and for lost to follow-up was 10.5 years.

Figure 3.

Kaplan–Meier plot of cumulative outcomes for disposal of cryostored sperm samples according to the reasons for disposal—patient died (deceased), sperm was no longer required (discontinued), patient was lost to follow up and transferred for use. Inset: Kaplan–Meier plot of cumulative outcomes for disposal of cryostored sperm normalized for completed outcomes.


Haematological malignancies accounted for nearly a third of the patients (Hodgkin's 12%, non-Hodgkin's lymphoma 10%, leukaemia 10%), followed by testicular cancers accounting for nearly a quarter (seminoma 14%, teratoma 10%) with other malignancies (17%) including sarcoma, gastrointestinal, glioma, head and neck, prostate, melanoma, lung and others. The non-cancer disease group (9.2%) was mostly comprised of patients with immunological, renal or genitourinary diseases and multiple sclerosis.

Sperm Output

Semen volume, sperm concentration and total sperm output were all significantly higher on the first compared with the median of all semen samples for that individual (by 0.15 ± 0.03 [SEM] ml, 7 ± 0.8 million per ml, and 32 ± 3 million, respectively, all P < 0.0001 by paired t-test). Nevertheless, the differences were small (median difference of 0 for each variable) and subsequent analyses used the median semen variables.

Sperm output for all cancers and non-cancer diseases was lower than healthy controls of similar age (Table I). Testicular cancers had the lowest total sperm output, a finding that was nullified by accounting for unilateral orchidectomy of tumour-bearing testis prior to cryostorage (present in 613/680, 90%) when total sperm output was adjusted for total testicular volume (data not shown).

Lower sperm output was associated with a history of cryptorchidism (26 versus 90 million, P < 0.001), of prior infertility (46 versus 89 million, P = 0.005) or no prior fertility (85 versus 99 million, P = 0.02) and known testicular pathology (54 versus 91 million, P < 0.0001), smoking (79 versus 95 million, P < 0.005) and living alone (single, divorced) compared with living with a partner (married, de facto) (89 versus 103 million, P = 0.013), but not significantly with alcohol intake (P = 0.55). Recent weight loss at the time of cryostorage was significantly associated with lower sperm output (78 versus 94 million, P = 0.005) but not recent fever (P = 0.11).

Sperm output was correlated positively with total testicular volume (r = 0.44, P < 0.0001) and negatively with serum FSH (r = −0.24, P < 0.0001) and LH (r = −0.12, P < 0.001) but not with serum testosterone (−0.008, P = 0.68). Serum FSH was increased in men with a single compared with two testes (median 6.35 versus 4.2 IU/l, P < 0.0001). Sperm motility was significantly lower in cancer and non-cancer disease compared to the controls but there was no significant difference between the disease groups.