Between the 29th of November 2012 and the 1st of May 2020, 82 breast cancer patients underwent ovarian stimulation for fertility preservation. A total of 8 patients were excluded for premature LH surge/premature triggering (n = 4), non-compliance to the ovarian stimulation protocol (n = 2), fertility preservation for breast cancer relapse (n = 1), and expression of formal refusal communicated to the institution to use their clinical data for clinical trials (n = 1). Among the 74 patients included in the study, 14 patients were exposed to at least one imaging procedure involving ionizing radiation during ovarian stimulation (exposed group) and 60 patients had already had their staging and risk assessment before starting ovarian stimulation (non-exposed group).
Exposed and non-exposed patients had similar baseline and oncological characteristics, except for BMI at diagnosis (median BMI of 24.4 and 22.1 in exposed and non-exposed groups, respectively; P = 0.04) (Table 1 and Table S1). Mean age at diagnosis was 31.2 years in the exposed cohort and 32.6 years in the non-exposed cohort (P = 0.21). A total of 23 patients (31.1%) had children at the time of diagnosis. Ovarian reserve was similar in both cohorts, with a median AMH level of 2.5 µg/L (range: 0.2–13) in the exposed group, compared to 1.9 µg/L (range: 0.3–7.1) in the non-exposed group (P = 0.20) (Table 2).
Table 1
Breast cancer characteristics
| Ionizing radiation exposed cohort (n = 14) | Non-exposed cohort (n = 60) | P-value |
Mean age at diagnosis (SD) | 31.2 (3.4) | 32.6 (4.0) | 0.21 |
BRCA pathogenic variants – n (%) Of which: BRCA1 BRCA2 | 5 (35.7) 3 (21.4) 2 (14.3) | 9 (15.0) 5 (8.3) 4 (6.7) | 0.12 |
Histology - n (%) Ductal carcinoma Lobular carcinoma Other Unknown | 13 (92.9) 0 (0) 0 (0) 1 (7.1) | 54 (90.0) 1 (1.7) 1 (1.7) 2 (3.3) | 0.72 |
Tumor grade - n (%) 1–2 3 Unknown | 1 (7.1) 13 (92.9) 0 (0) | 22 (36.7) 36 (60.0) 2 (3.3) | 0.06 |
Tumor size - n (%) T1 T2 T3-T4 | 4 (28.6) 8 (57.1) 2 (14.3) | 22 (36.7) 30 (50) 8 (13.3) | 0.85 |
Nodal status - n (%) N0 N1-N3 Unknown | 10 (71.4) 4 (28.6) 0 (0) | 36 (60.0) 22 (36.7) 2 (3.3) | 0.63 |
Hormone receptor status - n (%) ER and/or PR positive ER and PR negative | 8 (57.1) 6 (42.9) | 38 (63.3) 22 (36.7) | 0.76 |
HER2 status - n (%) HER2 negative HER2 positive | 10 (71.4) 4 (28.6) | 40 (66.7) 20 (33.3) | 1.00 |
Abbreviations: HER2, human epidermal growth factor receptor 2; ER, estrogen receptor; PR, progesterone receptor. |
Table 2
Ovarian stimulation and oocyte retrieval
| Ionizing radiation exposed cohort (n = 14) | Non-exposed cohort (n = 60) | P-value |
Basal AMH (µg/L)- median (range) | 2.5 (0.2–13) | 1.9 (0.3–7.1) | 0.20 |
Number of cycles | 14 | 69 | |
Time between breast cancer diagnosis to day 1 of OS - in days- median (range) | 11.5 (5–33) | 28 (1-164) | < 0.01 |
Time between imaging and oocyte retrieval, in days - median (range) | 8.7 (4–12) | Not applicable | |
Type of ovarian stimulation cycle - n (%) Standard Random follicular Random luteal Unknown | 9 (64.3) 1 (7.1) 4 (28.6) 0 (0) | 39 (56.5) 5 (7.2) 24 (34.8) 1 (1.4) | 0.92 |
Gonadotropins Recombinant FSH- n (%) HMG- n (%) | 12 (85.7) 2 (14.3) | 53 (76.8) 16 (23.2) | 0.72 |
Total FSH dose (IU) - mean (SD) | 2794.6 (892.3) | 2524.1 (950.1) | 0.33 |
Stimulation, in days - median (range) | 11 (8–14) | 10 (3–16) | 0.40 |
Triggering method - n (%) hCG GnRH agonists | 3 (21.4) 11 (78.6) | 22 (31.9) 47 (68.1) | 0.54 |
Data at triggering - median (range) E2 (ng/l) Progesterone (µg/l) Number of follicles > 18 mm Number of follicles 15–18 mm Number of follicles < 15 mm | 319.8 (95-1345) 0.8 (0.4–2.4) 2.5 (1–11) 3 (1–17) 5 (1–15) | 317 (20-1024) 1 (0.2–5.7) 2 (0–7) 4 (0–20) 6 (0–24) | 0.44 0.53 0.76 0.82 0.89 |
OS outcomes Number of oocytes collected- median (range) Number of oocytes collected- mean (SD) Number of mature oocytes median (range) Number of mature oocytes collected- mean (SD) Maturation rate - % median (range) | 10.5 (1–21) 11.3 (7.4) 8 (1–20) 9.7 (6.9) 92.5 (46.2–100) | 7 (0–23) 8.1 (5.1) 6 (0–19) 6.7 (4.4) 85.7 (0-100) | 0.16 0.17 0.54 |
Fertilization outcomes Total number of oocytes fertilized Fertilization rate (%) Total number of frozen embryos | 45 62.2 22 | 132 65.4 89 | 0.70 |
Abbreviations: OS, ovarian stimulation; AMH, anti-Müllerian hormone; FSH, follicle-stimulating hormone; HMG, human menopausal gonadotropin; IU, international units; SD, standard deviation; hCG, Human chorionic gonadotropin; GnRH, Gonadotropin-releasing hormone. |
The majority of the patients had a stage 2 tumor (57.1% and 50% in the exposed and non-exposed groups, respectively), without nodal invasion (71.4% and 60% in the exposed and non-exposed group, respectively), with positive hormone receptors (57.1% and 63.3% in the exposed and non-exposed cohort, respectively), and HER2 negative status (71.4% and 66.7% in the exposed and non-exposed group, respectively) (Table 1).
Radiation exposure
Among the 14 exposed patients, 5 (35.7%) underwent a PET scan, 9 (64.3%) a bone scan, 2 (14.3%) a CT scan, and 1 patient underwent a MUGA scan (7.1%) during the ovarian stimulation cycle. Four patients underwent 2 imaging procedures during the ovarian stimulation cycle, and one patient underwent 3 imaging procedures (Table S2).
The mean time between the beginning of ovarian stimulation and the first ionizing radiation exposure was 3.9 days (range: 0–7). The mean time between first exposure and oocyte collection was 8.7 days (range: 4–12). Taking into account the highest estimated scattering dose according to the literature [18, 19], patients were exposed to a median pelvic radiation exposure of 0.7 mGy (range: 0.5–45.5).
Fertility preservation outcomes
All the patients exposed to ionizing radiation had one ovarian stimulation cycle while 9 out of 60 patients in the non-exposed group had two consecutive stimulation cycles. The median time between diagnosis and the beginning of the first ovarian stimulation cycle was shorter in the exposed cohort (11.5 days, range: 5–33) than in the non-exposed group (28 days range: 1-164) (P < 0.01). The characteristics of the ovarian stimulation cycles were similar in both groups (Table 2). hCG triggering was used at the beginning of the protocol and then replaced by GnRH analogues [17].
Median number of collected oocytes was similar in both groups (10.5 versus 7 in the non-exposed and exposed group, respectively; P = 0.16) as well as the maturation rate (92.5% versus 85.7% in the exposed and non-exposed groups, respectively; P = 0.54). Incidence rate ratios (IRR) of ionizing radiation exposure on the number of mature oocytes was 1.37 (IC: 0.94-2.0; P = 0.10) in the univariate model and 1.13 (IC: 0.77–1.65; P = 0.53) in the multivariate model.
Age (IRR = 0.95; IC: 0.91–0.99; P = 0.02) and AMH (IRR 1.18; IC: 1.09–1.28; P < 0.0001) were both significantly associated with the number of mature oocytes in the univariate model. AMH was still significantly associated with the number of mature oocytes collected in the multivariate model (IRR 1.16; IC: 1.07–1.27; P = 0.001), while age was not (IRR 0.98 IC: 0.94–1.02; P = 0.40). In contrast, the presence of a germline BRCA pathogenic variant was not significantly associated with the number of mature oocytes collected (IRR 0.92; IC: 0.62–1.36; P = 0.67 and IRR 1.06; C: 0.72–1.56; P = 0.75) in the univariate and multivariate models, respectively (Table 3).
Table 3. Incidence rate ratios
Negative Binomial
|
IRR (IC95%)
|
P-value
|
aIRR (IC95%)
|
P-value
|
Treatment
exposed vs non-exposed to ionizing radiation
|
1.37 (0.94-2.0)
|
0.10
|
1.13 (0.77-1.65)
|
0.53
|
Age
|
0.95 (0.91-0.99)
|
0.02
|
0.98 (0.94-1.02)
|
0.40
|
AMH
|
1.18 (1.09-1.28)
|
<0.0001
|
1.16 (1.07-1.27)
|
0.001
|
BRCA
BRCA PV vs BRCA VUS, negative or unknown
|
0.92 (0.62-1.36)
|
0.67
|
1.06 (0.72-1.56)
|
0.75
|
Abbreviations: IRR, incidence rate ratios; aIRR, adjusted incidence rate ratios; AMH, anti-Müllerian hormone; PV, pathogenic variant; VUS, variants of unknown significance.
A total of 45 and 132 mature oocytes were fertilized in the exposed and non-exposed groups, respectively. Fertilization rates were similar in both groups (Table 2).
Oncological and fertility outcomes
Patients had a median follow-up of 3.7 years from breast cancer diagnosis (range: 0.8–7.5). Twelve and 58 patients had at least one year of follow-up after treatment in the exposed and non-exposed groups, respectively. Three patients out of 14 in the exposed cohort experienced a relapse (21.4%) compared to 6 out of 60 (10%) in the non-exposed cohort (P = 0.4) (Table S2). No patients died in the exposed cohort, while there was one death in the non-exposed cohort (1.7%).
Among exposed patients, none returned to the clinic to recover cryopreserved material or had a pregnancy after their breast cancer. In the non-exposed cohort, 11 patients out of 60 (18.3%) used their frozen oocytes (n = 4), embryo (n = 6), or both (n = 1) to achieve pregnancy. The mean time between fertility preservation and oocyte/embryo thawing was 3.4 ± 1.5 years. Mean survival rates after thawing were 55.6% for the oocytes (10/18) and 84.6% for the embryos (11/13). Fifteen embryos were transferred into 10 patients and 10 pregnancies were obtained (implantation rate 66.7%) leading to 5 live births, 3 miscarriages, and 2 ongoing pregnancies. All patients who received an embryo transfer from cryopreserved oocytes/embryos had at least one positive hCG test (10/10). In addition, 7 patients had at least one spontaneous pregnancy and 2 others became pregnant using fresh oocytes retrieved in IVF/ICSI cycle.