Pretreatment decrease of the anti-müllerian hormone level does not translate into the proportionally lower efficiency of ovarian stimulation in patients with haematological malignancies.

OBJECTIVE

To study whether a pretreatment decrease in ovarian reserve in oncohaematological patients determines the success of fertility preservation before gonadotoxic therapy.

DESIGN

Age-matched retrospective control study.

SUBJECTS: A total of 73 individuals after oocyte vitrification: 23 patients with haematological malignancies and 50 age-matched healthy oocyte donors.

EXPOSURE: Measurement of pretreatment anti-müllerian hormone level. Ovarian stimulation, oocyte retrieval and vitrification.

MAIN OUTCOME MEASURES: Number of retrieved and vitrified oocytes.

RESULTS

Oncohaematological patients have significantly lower anti-müllerian (AMH) hormone levels than healthy age-matched oocyte donors: median value of anti-müllerian hormone was 2,91 ng/mL IQR [1.6;4.7], 95% CI [1.74;4.36]) versus 4,74 ng/mL (IQR [3.2;6.8], 95%CI [3.84;6.32]). In contrast to the donor group there was no correlation between AMH level and female age. Controlled-ovarian stimulation parameters were virtually identical between groups: an antagonist protocol with a GnRH agonist trigger, a median initial rFSH dose of 200 IU [95% CI 175;200], a median total rFSH dose 2000 IU [95% CI 1800;2200], and a median stimulation length 10 days [95% CI 10;11]. The protocols differed in timing: a random-start protocol was used in most patents of onco-fertility preservation group (median cycle day of stimulation start - 14 [95% CI 3;21]), whereas the control group followed a conventional start (median cycle day of stimulation start - 2 [95% CI 2;2]). The number of MII and total retrieved oocyte did not differ between the patients of both groups: the median value of MII oocyte number was 12 [95% CI 8;16] during oncofertility preservation vs. 14 [95% CI 11;16] during voluntary oocyte donation, the median value of total retrieved oocyte number was 16 [95% CI 9;20] vs.17 [95% CI 14;20], respectively. The median value of %mature oocyte was also comparable (79% [95% CI 72.7;94.0] in oncohaematological patients vs. 81,81% [95% CI 78.95;88.89] in oocyte donors). No significant correlation was found between AMH level and the number of MII oocytes in oncohaematological patients (Spearman rank correlation, ρ = 0.2249, p = 0.3405) in contrast to the control group (Pearson correlation, r = 0.59, p = 0.000006527). Negative binomial regression revealed no significant main effect of AMH on MII oocyte yield in the oncofertility group (coefficient estimate -0.0106; 95% bootstrap CI [-0.0541;0.2204]) but a modest interaction with group (IRR 0.89 [95% CI 0.82;0.96]), indicating attenuated association between AMH and MII yield in oncology patients compared to oocyte donors.

CONCLUSION

Within this cohort and stimulation approach, reduced pre-treatment AMH levels did not translate into a substantially lower oocyte yield in patients with haematological malignancies. Therefore, lower AMH levels should not be the primary main reason for refusal in onco-fertility preservation protocols.