Among the 31 men enrolled, 30 completed the course of treatment. One subject dropped out of the study due to side effects and is not included in the analysis. The mean age of study subjects was 37.9 years (± 4.7 years) and mean testicular volumes were 14.4 mL (± 4.8 mL) and 13.8 mL (± 3.9 mL), on the right and left sides, respectively. The laboratory evaluation of study subjects at baseline and after 4 weeks on treatment are outlined in Table 1. Laboratory assessment demonstrated only a significant increase in serum triglycerides (p = 0.04) with isotretinoin treatment; however, no subject had an elevation of serum triglycerides to above 500 mg/dl and no subject experienced pancreatitis. A mild significant increase in serum AST was also noted but was not clinically significant.

Among those completing treatment, 26 (87%) were azoospermic, and 4 (13%) were cryptozoospermic. Six men with azoospermia (20%) had a past history of cryptozoospermia. Among men with azoospermia or cryptozoospermia respectively, 24/26 (92%) and 3/4 (75%) had previously undergone testicular sperm retrieval or fine needle aspiration (FNA) mapping [14] procedures (Table 2). Regarding the etiology of infertility, 23/30 had idiopathic infertility; 3/30 had unilateral or bilateral undescended testes; 3/30 harbored a genetic diagnosis [46, XY,t(2;7;8)(q34;q31.2;q22.1); 46, XY,t(15:16)(q11.2;p13.1), and 46, XY 16qh +] and 1/30 had a history of crushed pelvis.

Among 30 treated subjects, 11 (37%) developed reliable, motile ejaculated sperm counts with isotretinoin treatment. Among azoospermic men, 7/26 (27%) responded. Among cryptozoospermic men, 4/4 (100%) responded (Table 3). Mean total motile sperm count (TMC) achieved among all responders was 48,000 sperm. The TMC response by individual in the eleven men who developed sperm in their ejaculates is illustrated in Fig. 2. There was no clear association between the magnitude of the TMC response and the cohort status (i.e., azoospermia or cryptozoospermia; prior TESE with or without sperm found). Median ejaculated TMCs in the two categories of subjects is outlined in Table 3. Response times to achieve motile ejaculated sperm counts are illustrated in Fig. 3. Most subjects who responded to therapy did so within 3 months of initiating treatment. All those who responded at month three, and who continued with therapy, maintained their response through 6 and 9 months of therapy.

Individual patient total motile sperm count (TMC) response to isotretinoin treatment. Values are means of the two highest measurements for each subject. Note y-axis is logarithmic and not linear in scale

Response times to achieve motile ejaculated sperm counts while on isotretinoin therapy in n = 11 responding patients. Only n = 4 patients remained on isotretinoin at 9 months. Note: mos = months

We also examined the response rate by testicular histology as assessed by testicular biopsy or FNA mapping prior to isotretinoin therapy (Table 4). Although not statistically significant (Fisher’s exact p-value = 0.275 for the difference of proportions between the four groups), there was a trend toward better response as the level of germ cell maturation increased. We did not see a responder among the few patients with Sertoli cell-only histology patterns. However, among 11 subjects with early or late maturation arrest phenotypes, 50% and 60% respectively, responded to therapy.

Among men with azoospermia who did not develop sperm in their ejaculates, and who then proceeded to repeat sperm microdissection TESE retrieval procedures on isotretinoin, 6/7 (86%) had sufficient sperm found to cover all eggs at IVF-ICSI. In 5 of 7 cases (71%), sperm retrieval also generated sufficient sperm to cryopreserve for future IVF cycles. In addition, when comparing surgical times of prior microTESE procedures performed prior to isotretinoin therapy (mean 105 min; [range 75–150 min]) to those while on isotretinoin treatment (mean 63 min; [range 45–90 min]), they decreased by 40%.

To date, 9 IVF-ICSI cycles have been performed using ejaculated sperm in 6 complete responders. Thirteen euploid (e.g., biopsy normal) embryos have been generated with 1 live birth to date. Among partial responders having repeat sperm retrievals, 6 IVF-ICSI cycles have been performed with 1 pregnancy obtained. No spontaneous pregnancies have been observed as yet from treatment.

All 30 participants (100%) reported dry skin and chapped lips. Additional side effects included rashes 4/30 (13%), irritability 14/30 (47%), and altered cholesterol levels 5/30 (17%) (Table 1). One study subject dropped out of the study due to anxiety early in the study and is not included in the overall analysis.