If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Clinical Laboratory, Department of Biopathology, Henri Becquerel Centre, Rouen, FranceEquipe de Recherche en Oncologie (IRON), Inserm U1245, Rouen University Hospital, Rouen, France
Equipe de Recherche en Oncologie (IRON), Inserm U1245, Rouen University Hospital, Rouen, FranceDepartment of Oncology, Henri Becquerel Centre, Rouen, FranceDepartment of Gastroenterology, Rouen University Hospital, Rouen, France
Equipe de Recherche en Oncologie (IRON), Inserm U1245, Rouen University Hospital, Rouen, FranceDepartment of Oncology, Henri Becquerel Centre, Rouen, France
The impact of chemotherapy on fertility appears to be of essential importance for the youngest cancer survivors. The aim of this study was to assess plasma anti-Mullërian hormone (AMH) evolution, using an automated sensitive AMH immunoassay in women younger than 35 years old before and after treatment with adjuvant chemotherapy for early breast cancer. We selected 54 women aged less than 35 years old, at the time of breast cancer diagnosis, who received chemotherapy between 2008 and 2014, and with plasma samples collected from the diagnosis, to 1 year, 3 years and 5 years post-diagnosis. The median AMH decreased markedly in the year after the diagnosis compared with the pretreatment values (P < 0.0001), and slightly increased 2 years later (P = 0.007, comparing 1-year and 3-years post-diagnosis concentrations), without any additional AMH recovery 5 years after diagnosis. This recovery did not reach age-dependent AMH expected values (P < 0.0001, comparing AMH measured values to AMH expected values). Addition of taxanes to an anthracyclines + alkylating-based regimen was associated with a worse AMH decrease (P = 0.007). Ovarian tissue cryopreservation before treatment did not influence the AMH recovery. These results highlight the necessity of fertility counselling before treatment, especially in women wanting children.
Women younger than 35 years of age represent 2.4% of all cases of breast cancer and approximately one-third of all cancers diagnosed in women below 35 years are breast cancer (
Early Breast Cancer Trialists' Collaborative Group (EBCTCG) Comparisons between different polychemotherapy regimens for early breast cancer: meta-analyses of long-term outcome among 100,000 women in 123 randomised trials.
), the number of young patients treated with adjuvant chemotherapy for early breast cancer who have a desire for pregnancy after treatment is increasing. For these patients, the question of age becomes essential and is influenced by the necessary delay to start a pregnancy after a breast cancer diagnosis (
). In this setting, an assessment of the post-chemotherapy gonadotoxicity in the youngest patients appears to be crucial.
Since the 2000s, anti-Müllerian hormone (AMH) has become the major biochemical marker of ovarian reserve. Initially dedicated to the assisted reproductive technology context, AMH use has been progressively extended to explore ovarian reserve after chemotherapy (
Centers of FertiPROTEKT Network Reduced pretreatment ovarian reserve in premenopausal female patients with Hodgkin lymphoma or non-Hodgkin-lymphoma–evaluation by using antimüllerian hormone and retrieved oocytes.
). Indeed, AMH measurements appeared more sensitive than traditional biomarkers (inhibin B and FSH) and transvaginal ultrasound examination (the determination of ovarian volume and count of antral follicles) (
The aim of the present study was to evaluate the AMH evolution in women aged less than 35 years, who were treated with adjuvant chemotherapy for early breast cancer, from the diagnosis to 1 year, 3 years and 5 years after the diagnosis, to better understand the middle- to long-term effects of chemotherapy on the ovarian reserve in this specific population.
Materials and methods
Subjects
All women aged less than 35 years old who were treated in the Henri Becquerel Centre for early breast cancer from 2008 to 2014 were retrospectively included. The patients' characteristics (e.g. tumour characteristics, treatment and evolution) and fertility history (before cancer and after treatment) were collected. Because of the retrospective design of this study, no information about the menstrual status was analysed. Blood samples were collected for routine analyses related to the biological follow-up before and after treatment: around 1 year (between 6 months to 18 months) from diagnosis, around 3 years (between 2 years and 4 years) from diagnosis, and around 5 years. Remaining heparinized plasma samples were stored frozen (−20°C).
Due to the evolution of adjuvant chemotherapy recommendations during the study period, patients were treated with either six cycles of FEC (5 fluorouracil 500 mg/m2, epirubicin 100 mg/m2, cyclophosphamide 500 mg/m2) or three cycles of FEC followed by three cycles of taxanes (docetaxel [D] [100 mg/m2]).
Ethical approval
All patients signed a consent form allowing their biological samples to be conserved. The study was approved by the Institutional Review Board of our Centre on 14 March 2016 (registration number: 1602B).
AMH measurements
AMH measurements were performed from heparinized plasma samples using the fully automated Elecsys® AMH assay on the Cobas® e 601 instrument (Roche Diagnostics, Mannheim, Germany). According to the manufacturer's data sheet, the detection limit was 0.01 ng/ml, the quantification limit was 0.03 ng/ml, and the intra-assay imprecision coefficient of variation was 1.2% at 1.19 ng/ml, and 0.9% at 5.89 ng/ml. All measurements were performed in one run, excluding inter-assay variation.
Statistical methods
The primary objective was to evaluate the evolution of AMH values between pre- and post-treatment settings. For the secondary objectives, the AMH variations were analysed from baseline to 1, 3 and 5-year post-diagnosis in the available samples according to: the age group (above or below the median age); the type of chemotherapy regimen received; the exposure to adjuvant hormonotherapy; and the ovarian tissue preservation before the treatment. The impact of adjuvant hormonal therapy has been analysed at 3 years after the diagnosis considering the need of a minimal drug exposure. Comparisons of AMH evolution over time were restricted to the same patients, using paired tests (Wilcoxon). The tests were performed using the MedCalc® statistical software (MedCalc Software, Belgium). Calculation of the theoretical AMH decrease expected in the study population without treatment was performed according to the physiological published AMH decrease of 5.6% per year (
). The 95% confidence intervals (CI) were calculated based on a bootstrap method using 10,000 permutations and were performed by the R free software (https://cran.rstudio.com/). The data are presented as the median [min-max]. A P-value of <0.05 was considered statistically significant. In case of no significant differences between the study sub-groups of population, the conclusions must be moderate, considering a possible risk of underpowered comparisons.
Results
Patient and sample characteristics
Among the 5775 patients treated for breast cancer during the study period, 119 were aged less than 35 years old. Of these, 65 were excluded due to a lack of remaining plasma before the treatment (n = 31) or after the diagnosis (n = 16), metastatic disease (n = 10), or no chemotherapy after surgery (n = 8). Thus, 54 patients were finally included with at least one AMH measurement before and one measurement after diagnosis. AMH post-treatment measurements were evaluated at 1 year, 3 years and 5 years in 49, 32 and 11 patients, respectively. Due to missing samples, for each included patient, the median number of samples analysed was 3[2–4], with a median number of samples analysed after chemotherapy of 2[1–3]. The patients' characteristics are presented in Table 1.
3FEC-3D = Fluorouracil 500 mg/m2 + epirubicin 100 mg/m2 + cyclophosphamide 500 mg/m2, every 21 days for three cycles, followed by sequential docetaxel 100 mg/m2 every 21 days for three cycles.
a 3FEC-3D = Fluorouracil 500 mg/m2 + epirubicin 100 mg/m2 + cyclophosphamide 500 mg/m2, every 21 days for three cycles, followed by sequential docetaxel 100 mg/m2 every 21 days for three cycles.
b 6FEC = Fluorouracil 500 mg/m2 + epirubicin 100 mg/m2 + cyclophosphamide 500 mg/m2, every 21 days for six cycles.
For the 54 patients included, a total of 144 samples were analysed. The concentrations of AMH were non-detectable (<0.01 ng/ml) in 14 samples (none in the pretreatment analysis, 10 from 49 patients at the 1-year analysis, 2 from 32 patients at the 3-years analysis and 2 from 11 patients at the 5-years analysis) obtained from 11 patients. No ovarian suppression function was used in the cohort from this study.
Pre-treatment AMH concentrations
The median AMH at diagnosis was 2.12 ng/ml [0.33–17.49] with AMH values decreasing not significantly with age as follows: median AMH = 3.37 ng/ml [3.09–3.65] for women aged 20–24 years old (n = 2), median AMH = 2.29 ng/ml [1.18–17.49] for women aged 25–29 years old (n = 10) and median AMH = 1.95 ng/ml [0.34–7.2] for women aged 30–35 years old (n = 42).
AMH did not vary significantly according to smoking status (median AMH = 2.54 ng/ml [0.83–6.5] and 1.95 ng/ml [0.37–17.49] in smokers (n = 16) and non-smokers (n = 38), respectively), or BRCA mutations (median AMH = 2.13 ng/ml [0.33–6.50] and 2.12 ng/ml [0.57–17.49] in mutated [n = 14] or non-mutated [n = 40], respectively). A body mass index (BMI) > 25 kg/m2 was associated with a higher AMH concentration (median AMH = 3.29 ng/ml [0.78–17.49] for n = 14 patients with a BMI >25 kg/m2, versus 1.82 ng/ml [0.33–7.2] for n = 40 patients with a BMI <25 kg/m2, P = 0.046).
The median AMH was not different in women with a history of pregnancy before the treatment compared with women without any previous pregnancies (median AMH = 2.00 ng/ml [0.33–17.49] versus 2.25 ng/ml [0.56–5.13]). Of note, these two groups had comparable ages (median age = 32 years [23–35] and 30 years [22–34], respectively).
Post-treatment AMH concentrations
After a median delay of 379 [172–571] days from the diagnosis, the median AMH decreased to 0.13 ng/ml [0.01–6.11]; this decrease was significant compared with the pretreatment values (n = 49, P < 0.0001). There was a slight increase in the median AMH to 0.31 ng/ml [0.01–3.28] after two more years of survey (when comparing 1-year and 3-years post-diagnosis concentrations, n = 32, P = 0.007) (median delay from the diagnosis = 1128 [730–1490] days). No additional AMH recovery was observed 5 years after the diagnosis (median delay from the diagnosis = 1913 [1770–1913] days) compared with the 3-years post-diagnosis values (n = 9) (Figure 1).
Figure 1Evolution of anti-Müllerian hormone (AMH) concentrations from the time of diagnosis to 1, 3 and 5 years after diagnosis. Each coloured line represents a patient. The median AMH decreases drastically 1 year after diagnosis (n = 49, P < 0.0001, comparing pre-diagnosis and 1-year post-diagnosis concentrations) and increases slightly 3 years after (n = 32, P = 0.007, comparing 1-year and 3-years post-diagnosis concentrations), but it does not recover to pretreatment concentrations. No further improvement is observed 2 years later (n = 9, P = 1, comparing 3-year and 5-years post-diagnosis concentrations). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
The median AMH was not different in women younger or older than the median age (31.5 years old) at 1, 3 or 5 years after the diagnosis. To estimate the specific effect of breast cancer treatment on the AMH decrease, the study compared AMH values observed 1, 3 and 5 years after the diagnosis to theoretical AMH values that were calculated according to the physiological decline rate of 5.6% per year (
). The observed median AMH values were significantly lower than the theoretical expected medians at 1, 3 and 5 years after the diagnosis (P < 0.0001 for each point) (Figure 2).
Figure 2Evolution of median anti-Müllerian hormone (AMH) concentrations from the pretreatment to 1, 3 and 5 years after the diagnosis, compared with the theoretical physiological AMH decrease. The solid line represents the AMH decrease observed in this study population, women <35 years treated for a breast cancer at 1, 3 and 5 years after the diagnosis. The dashed line represents the theoretical AMH decrease that would have been expected in this population without treatment. This decrease was calculated from the median AMH concentration observed in the population before the diagnosis, and assuming the physiological rate of decline of 5.6% per year (
). The AMH concentrations of the patients were reported using the median value. The 95% confidence intervals were calculated based on a bootstrap method using 10,000 permutations and were performed by the R free software. (*) indicates a significant difference (P < 0.0001) between the AMH measurements in patients, represented with the full line, and the expected AMH values according to the theoretical physiological AMH decrease, represented with the dashed line.
At 1 year after the diagnosis, the median AMH was significantly lower after 3FEC-3D (median AMH = 0.09 ng/ml [0.01–6.11], n = 41) than after 6FEC (median AMH = 0.39 ng/ml [0.21–2.33], n = 8) (P = 0.007). The median age at diagnosis of these two groups was not significantly different. A comparison between 3FEC-3D versus 6FEC was not performed at 3- and 5-years post-diagnosis because the sub-groups were too small. Considering the hormonal therapy, every patient (except one, who was treated with letrozole) received tamoxifen. The hormonal therapy did not influence the 3-years post-diagnosis AMH concentrations (median AMH = 0.34 ng/ml [0.01–2.76] among the 15 patients with hormonal therapy versus 0.25 ng/ml [0.01–3.28] among the 17 patients without hormonal therapy). At 5-years post-diagnosis, the impact of hormonal therapy on AMH concentrations could not be analysed because of the lack of data.
The median AMH concentration at 1-year post-diagnosis was not decreased in patients who had benefitted from an ovarian tissue cryopreservation (median AMH = 0.138 ng/ml [0.01–0.52] for 11 patients) compared with those who had not (median AMH = 0.13 ng/ml [0.01–6.11] for 43 patients). Of note, the median ages were similar between these two groups (31 years [22–34] and 32 years [23–35], respectively).
Discussion
In this study, focused on the AMH evolution in patients younger than 35 years of age during and after adjuvant treatment for early breast cancer, a marked decrease in AMH concentrations after chemotherapy was observed, which remained below the age-expected values despite 3 to 5 years of follow-up, and was influenced by the chemotherapy regimen. AMH analysis was performed using the automated Elecsys AMH assay (Roche Diagnostics), which has been reported as sensitive (
). The improvement of the sensitivity enables us to minimize the proportion of undetectable specimens. In this study, only 20% of the specimens were undetectable 1 year after chemotherapy, which represents an improvement compared with the previously published ratios: using the MIS ELISA assay (Diagnostic Systems Laboratories),
) using the Gen II Enzyme-linked immunosorbent assay (ELISA) kit (Beckman Coulter).
Before the treatment, no impact was observed on AMH concentrations according to smoking status and a slight increase observed of AMH concentrations in the heaviest patients. These confounding factors have been previously discussed without any consensus (
), that is associated with high AMH concentrations. Regarding the breast cancer context, the impact of BRCA mutations has also been reported. Two recent studies concluded that there was a negative impact of BRCA1 mutations on AMH concentrations (
Kathleen Cuningham Consortium for Research into Familial Breast Cancer (kConFab) Anti-Müllerian hormone serum concentrations of women with germline BRCA1 or BRCA2 mutations.
). In the present study, this phenomenon was not observed; one explanation could be the increase in the impact of BRCA mutation on AMH concentrations after 35 years (
) demonstrated an improvement of AMH values, quantified as a recovery of AMH concentrations above the detection level for 13% of the patients, in the first case, or an increase of about +1.2% per month, in the second case. Data from this study, with a marked fall of AMH values at one year post-diagnosis and a slight increase 3 years after diagnosis, without any supplementary improvement 5 years after diagnosis, confirmed a trend for a partial recovery that was not enough to reach the age-dependent AMH expected values (
). The data obtained after a longer follow-up confirmed the negative and persistent impact on the ovarian reserve of women treated for breast cancer, compared with age-matched controls (
) have been demonstrated to be good prognostic factors for menses recovery. In terms of AMH, the studied populations were frequently older, with median ages ranging from 37 years to more than 43 years (
), which presented a cohort with a median age = 35.5 years and with 57 patients younger than 35 years, all previous studies included less than eight breast cancer patients younger than 35 years in their cohort (
Pretreatment antimüllerian hormone levels determine rate of posttherapy ovarian reserve recovery: acute changes in ovarian reserve during and after chemotherapy.
Pretreatment antimüllerian hormone levels determine rate of posttherapy ovarian reserve recovery: acute changes in ovarian reserve during and after chemotherapy.
). The essential role of the alkylating agents in breast cancer treatments makes indispensable specific studies on the ovarian reserve in young breast cancer patients.
This study confirmed a persistent negative impact of anthracyclines and alkylating agents on the ovarian reserve in women <35 years and demonstrated the supplementary toxicity of taxanes, added to the anthracyclines + alkylating combination. The effect of the association anthracyclines + alkylating + taxanes on ovarian function continues to be debated (
reported a significant decrease of AMH 6 months after an anthracyclines + alkylating + taxanes-based chemotherapy; however this has not been confirmed with a longer period of follow-up (
showed a lower probability of pregnancy after breast cancer chemotherapy containing anthracyclines + alkylating + taxanes, compared with an anthracyclines + alkylating only regimen. To our knowledge, only two studies analysed AMH following the use of hormonal therapy (tamoxifen) after chemotherapy for breast cancer, with contradictory results: whereas
did not observe any significant difference (n = 87). Independent of the impact of chemotherapy, the hormonal therapy has been associated with an AMH decrease (
). This question is crucial to correctly interpret the AMH values in breast cancer patients who have received chemotherapy as an adjuvant treatment and are under hormonal therapy to adapt fertility counselling.
Any supplementary negative impact of ovarian tissue cryopreservation on AMH values after treatment was not observed. These results are consistent with those obtained by
, who observed slightly lower concentrations of AMH immediately after chemotherapy, no matter the initial malignancy, in patients who had an ovary removed, but no difference after a median delay of 8 months after the last chemotherapy treatment.
In conclusion, this study shows that: (i) the use of an automated sensitive AMH immunoassay improves the biological assessment of the ovarian reserve after breast cancer treatment, reducing the number of non-detectable concentrations compared with the published data; (ii) a marked AMH decrease exists after adjuvant chemotherapy for breast cancer, despite the young age of the patients; (iii) a slight increase of AMH 3 years after diagnosis does not reach the normal age-expected values; (iv) an anthracyclines + alkylating + taxanes regimen has a worse effect on AMH concentrations compared with anthracyclines + alkylating alone; and (v) ovarian tissue cryopreservation does not affect post-treatment AMH values. Although the role of AMH values in predicting the chances of pregnancy after chemotherapy remains to be established (
), results from this study, which indicate a negative impact of breast cancer chemotherapy on the ovarian reserve in young patients, highlight the necessity of fertility counselling before treatment, even in women younger than 35 years, especially when they have not yet borne children.
References
Anckaert E.
Öktem M.
Thies A.
Cohen-Bacrie M.
Daan N.M.P.
Schiettecatte J.
Müller C.
Topcu D.
Gröning A.
Ternaux F.
Engel C.
Engelmann S.
Milczynski C.
Multicenter analytical performance evaluation of a fully automated anti-Müllerian hormone assay and reference interval determination.
Pretreatment antimüllerian hormone levels determine rate of posttherapy ovarian reserve recovery: acute changes in ovarian reserve during and after chemotherapy.
Early Breast Cancer Trialists' Collaborative Group (EBCTCG)
Comparisons between different polychemotherapy regimens for early breast cancer: meta-analyses of long-term outcome among 100,000 women in 123 randomised trials.
Reduced pretreatment ovarian reserve in premenopausal female patients with Hodgkin lymphoma or non-Hodgkin-lymphoma–evaluation by using antimüllerian hormone and retrieved oocytes.
Anne Perdrix, MD-PhD, is working as a medical biologist in a French Comprehensive Cancer Center, specialized in tumor markers. She has been working from 2006 to 2013 in the Reproductive Laboratory of Rouen University Hospital, with an activity in Assisted Reproductive Technology and fertility preservation.
Key message
Anti-Müllerian hormone in women younger than 35 years decreased markedly after treatment with adjuvant chemotherapy for early breast cancer and remained under the age-expected values at 3 and 5 years post-diagnosis.
Article info
Publication history
Published online: June 14, 2017
Accepted:
June 2,
2017
Received in revised form:
June 2,
2017
Received:
January 3,
2017
Declaration: The authors report no financial or commercial conflicts of interest.
30254-7&id=rbmo1757-fig-0001.jpg){kind=link}
30254-7&id=rbmo1757-fig-0002.jpg){kind=link}