Longitudinal follow-up in female Childhood Cancer Survivors: no signs of accelerated ovarian function loss.

STUDY QUESTION: Is the long-term decline of ovarian function, as reflected by a decrease in serum anti-Müllerian hormone (AMH) concentration, accelerated over time in female childhood cancer survivors (CCS) as compared to healthy women of the same age? SUMMARY ANSWER: The median decline of AMH levels in long-term female CCS is not accelerated and similar to that observed in healthy controls. WHAT IS KNOWN ALREADY: Gonadal function is compromised in female CCS treated with chemotherapy and/or radiation therapy. Ovarian function is most compromised in survivors treated with total body irradiation, abdominal or pelvic irradiation, stem cell transplantation or high doses of alkylating agents. STUDY DESIGN SIZE, DURATION: Longitudinal single-centre cohort study in 192 CCS in Rotterdam, The Netherlands, between 2001 and 2014. PARTICIPANTS/MATERIALS, SETTING, METHODS: Serum AMH levels of 192 adult female CCS were assessed, at least five years after cessation of treatment and at a follow-up visit with a median of 3.2 years (range: 2.1-6.0) later and were compared to the age-based P50 of AMH in healthy controls. MAIN RESULTS AND THE ROLE OF CHANCE: Median AMH levels were below the P50 at both visit 1 (-0.59 µg/L) and at visit 2 (-0.22 µg/L). In women with a sustained ovarian function (AMH > 1.0 µg/L), the decline in AMH is similar to that in the normal population (difference in decline per year: -0.07 µg/L (range: -2.86 to 4.92), P  = 0.75). None of the treatment modalities was correlated with a significant acceleration of decline of AMH per year. LIMITATIONS REASONS FOR CAUTION: We selected CCS that visited our late effect outpatient clinic and who had two AMH levels available. It is conceivable that women without any apparent late effects of treatment as well as women with extreme late effects, which might be the ones with the largest impact on ovarian function, could be more likely to be lost to follow-up. However, general characteristics did not differ between the included and excluded patients. WIDER IMPLICATIONS OF THE FINDINGS: While prospective longitudinal research is required to strengthen our findings, they may help physicians to counsel female CCS about their expected reproductive lifespan. STUDY FUNDING/COMPETING INTERESTS: A.L.F.v.d.K., M.M.v.d.H.-E. and S.M.F.P. are supported by FP7-PanCare LIFE. J.S.E.L. has received grants from the following companies (in alphabetical order): Ferring, Merck Serono, Merck Sharp and Dome, Organon, Serono, Shering Plough and Shering. The other authors have no conflicts of interest to declare.

Decreased serum anti-Müllerian hormone levels in girls with newly diagnosed cancer.

STUDY QUESTION: Are anti-Müllerian hormone (AMH) levels reduced in girls with newly diagnosed cancer before the start of treatment? SUMMARY ANSWER: AMH levels are already compromised in girls at the time of cancer diagnosis compared with healthy girls. WHAT IS KNOWN ALREADY: In women diagnosed with cancer, evidence of reduced ovarian function has been described even before treatment has started. In girls with newly diagnosed cancer, no data are available. STUDY DESIGN, SIZE, DURATION: We performed an age-matched case-control study in girls with newly diagnosed cancer. PARTICIPANTS/MATERIALS, SETTING, METHODS: We determined serum AMH levels in a cohort of 208 girls with newly diagnosed cancer, up to 18 years of age at diagnosis, and compared them with AMH levels of 250 age-matched healthy girls. The diagnoses included were acute lymphoblastic leukaemia, acute myeloid leukaemia, Hodgkin lymphoma, non-Hodgkin lymphoma, nephroblastoma, sarcoma and neuroblastoma. MAIN RESULTS AND THE ROLE OF CHANCE: The median age was 6.6 years (range 0.0-17.4), comparable with that in the control group (median 6.3 years, range 0.3-18.0). Girls with childhood cancer presented with significantly lower serum AMH levels compared with healthy age-matched controls (standard deviation scores (SDS) -0.8, P < 0.001). Median AMH level in patients was 1.4 µg/l (0.1-10.2) versus 3.0 µg/l (0.1-18.3) in controls. Specifically, 84% of all patients had AMH levels below the 50th percentile of normal AMH levels, and 19% below the 10th percentile. Surrogate markers of general health status (temperature, C-reactive protein and haemoglobin levels at diagnosis) were significantly correlated with AMH SDS. LIMITATIONS, REASONS FOR CAUTION: Some caution is warranted because AMH levels increase with age in healthy children but the cases and controls were age-matched in our study. Although our sample size was large, additional studies are still required in an independent cohort. WIDER IMPLICATIONS OF THE FINDINGS: Our study shows that AMH levels are reduced in girls with newly diagnosed cancer even before the cancer treatment has started. AMH levels correlate with impairment of general health status in girls. Therefore, besides (pre) antral follicle number, other factors may influence serum AMH levels. Longitudinal studies during and after childhood cancer are currently being performed in order to evaluate possible ovarian recovery after discontinuation of treatment. STUDY FUNDING/COMPETING INTEREST(S): W.v.D. is supported by the Paediatric Oncology Centre Society for Research (KOCR), Rotterdam, The Netherlands. J.S.E.L. has received grants from the following companies (in alphabetical order): Ferring, Genovum, Merck Serono, Merck Sharp and Dome, Organon, Serono, Shering Plough and Shering. All other authors have nothing to disclose.

Genetic variation may modify ovarian reserve in female childhood cancer survivors.

STUDY QUESTION: Are genetic polymorphisms, previously identified as being associated with age at menopause in the healthy population, associated with ovarian reserve and predicted age at menopause in adult long-term survivors of childhood cancer? SUMMARY ANSWER: The CT genotype of rs1172822 in the BRSK1 gene is associated with lower serum anti-Müllerian hormone (AMH) levels and a younger predicted age at menopause in adult survivors of childhood cancer. WHAT IS KNOWN ALREADY: Gonadotoxicity is a well-known late side effect of chemotherapy and radiotherapy in adult survivors of childhood cancer. In the healthy population, several genetic polymorphisms are associated with age at natural menopause. Currently, data on the impact of previously identified variants in gene loci associated with ovarian reserve in adult long-term survivors of childhood cancer are lacking. STUDY DESIGN, SIZE, DURATION: We performed a pilot study in a single-centre cohort of adult female Caucasian childhood cancer survivors (n = 176). PARTICIPANTS/MATERIALS, SETTING, METHODS: We determined serum AMH levels (a marker of ovarian reserve) in adult survivors of childhood cancer (n = 176) and studied single nucleotide polymorphisms (SNPs) previously reported to be associated with age at natural menopause: BRSK1 (rs1172822), ARHGEF7 (rs7333181), MCM8 (rs236114), PCSK1 (rs271924), IGF2R (rs9457827) and TNF (rs909253). Association analysis was performed using the additive genetic model. Linear regression was conducted to assess the effect of significant polymorphisms in two previously published menopause prediction models. MAIN RESULTS AND THE ROLE OF CHANCE: The CT genotype of rs1172822 in the BRSK1 (BR serine/threonine kinase 1) gene was negatively associated with serum AMH levels in our cohort (odds ratio: 3.15, 95% confidence interval: 1.35-7.32, P = 0.008) and significantly associated with the predicted age at menopause (P = 0.04). The other five SNPs were not associated with serum AMH levels. LIMITATIONS, REASONS FOR CAUTION: This is a pilot study showing preliminary data which must be confirmed. To confirm our findings and enlarge the project, a nationwide genome-wide association (GWA) project on the ovarian reserve in female survivors of childhood cancer should be performed, including a replication cohort. WIDER IMPLICATIONS OF THE FINDINGS: Our findings support the hypothesis that previously identified genetic polymorphisms associated with age at menopause in healthy women may have an effect on the onset of menopause in female survivors of childhood cancer. Our study highlights a new aspect of the influences on the ovarian reserve after childhood cancer, which should be investigated further in a nationwide GWA study. Eventually, this information can help us to improve counselling on fertility preservation prior to cancer treatment based on genetic factors in individual patients. STUDY FUNDING AND CONFLICT OF INTEREST: W.D. is supported by the Paediatric Oncology Centre Society for Research (KOCR), Rotterdam, The Netherlands. J.S.E.L. has received fees and grant support from the following companies (in alphabetic order): Ferring, Genovum, Merck-Serono, Organon, Schering Plough and Serono. All other authors have nothing to disclose.